National Academies Press: OpenBook
« Previous: 10 Supplementary Note on Rock Mechanics Problems at Active Margins
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 113
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 114
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 115
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 116
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 117
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 118
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 119
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 120
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 121
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 122
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 123
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 124
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 125
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 126
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 127
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 128
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 129
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 130
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 131
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 132
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 133
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 134
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 135
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 136
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 137
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 138
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 139
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 140
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 141
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 142
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 143
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 144
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 145
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 146
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 147
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 148
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 149
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 150
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 151
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 152
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 153
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 154
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 155
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 156
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 157
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 158
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 159
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 160
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 161
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 162
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 163
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 164
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 165
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 166
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 167
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 168
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 169
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 170
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 171
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 172
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 173
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 174
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 175
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 176
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 177
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 178
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 179
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 180
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 181
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 182
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 183
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 184
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 185
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 186
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 187
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 188
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 189
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 190
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 191
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 192
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 193
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 194
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 195
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 196
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 197
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 198
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 199
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 200
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 201
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 202
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 203
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 204
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 205
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 206
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 207
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 208
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 209
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 210
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 211
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 212
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 213
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 214
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 215
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 216
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 217
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 218
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 219
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 220
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 221
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 222
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 223
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 224
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 225
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 226
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 227
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 228
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 229
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 230
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 231
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 232
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 233
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 234
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 235
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 236
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 237
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 238
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 239
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 240
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 241
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 242
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 243
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 244
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 245
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 246
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 247
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 248
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 249
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 250
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 251
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 252
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 253
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 254
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 255
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 256
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 257
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 258
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 259
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 260
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 261
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 262
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 263
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 264
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 265
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 266
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 267
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 268
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 269
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 270
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 271
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 272
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 273
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 274
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 275
Suggested Citation:"Part III: Background Papers." National Research Council. 1989. Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence. Washington, DC: The National Academies Press. doi: 10.17226/1500.
×
Page 276

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

PART III BACKGROUND PAPERS These papers express the views of the authors, not necessarily those of the National Research Council. As is customary with reports of this kind, the background papers are reproduced here, for the reader's convenience, as they were received from the authors without the NRC review and editorial attention given to the preceding sections of this report.

CONTENTS Section 1: Active Continental Margins I. Mechanics of P1 ate Motions A. Mechanics, Kinematics, and Deformation Modes at Convergent Margins; Mark Te Brandon and Dan M. Davis. ~ ~ ~ e ~ ~ ~ ~ e B. Driving Forces: Slab Subduction and Mantle Convection; Bradford Hager . . . . . ~ ~ . e II. Geologic Evolution of Active Continental Margins A. Initiation of Subduction; Dan Karig. e ~ ~ ~ ~ ~ 14 6 B. Intraoceanic Convergent Margins; Brian Taylor. . 160 C. Collision of Seamounts, Ridges, and Continental Fragments: Their Effects on Convergent Margins; Roland van Huene. . . . . . ~ . . . . . ~ ~ . . . 169 . . 117 . . 131 TIT. Mass and Chemical Transfer A. Mass Flux and Crustal Evolution at Convergent Margins; R.W. Kay and S. Mahiburg Kay . . . e Section 2: Passive Continental Margins I. Mechanics of Rifting of the Lithosphere 181 A. Comments on Rifting and Passive Margin Evolution in Light of Some Recent Studies; John Mutter and Brian Wernicke. . . . . . . e ~ ~ 208 B. Igneous Processes and the Evolution of Rifted Continental Margins; Jeffrey Parson and Carolyn Zehnder e ~ e ~ e ~ ~ ~ ~ ~ e e e e e 230 II. Rift and Passive Margin Basins -- The Sedimentary Record A. Investigating the Sedimentary Record: Sequence Stratigraphy--the Record of Tectonism and the Global Ocean Environment; Joel S. Watkins. . . . 247 B. Post Rifting Evolution of Passive Margin Basins; Dale S. Sawyer. . . . . . . . . . . . . . . . . . 269 115

Dative ~i~ttal Mars ME=P,2~CI; OF HE Gallon 116

MECHANICS, KINEMATICS, AND DEFORMATION MODES AT CONVERGENT MARGINS Mark T. Brandon and Dan M. Davis INTRODUCTI ON Much of the deformation that affects the crustal portion of the lithosphere occurs at convergent plate boundaries, which in the broadest sense encompass oceanic subduction zones, oniand thrust belts, and everything in between. Horizontal convergence at these boundaries is accommodated by a combination of two processes. The first is pervasive erogenic shortening within the leading edge of the overriding plate, as exemplified by accretionary wedges and collisional mountain belts such as the Himalayan or Taiwan thrust belts. In this case, excess mass is accommodated by continental growth and by uplift and erosion. The second is wholesale plate subduction, as illustrated by the non-accreting or eroding subduction zones such as the Mariana and southern Middle America trenches. In this case, excess mass is removed by assimilation into the asthenosphere. Modern and ancient convergent boundaries populate the spectrum between these two end- members. As a result, convergent boundary processes can have important and variable effects on the growth of continents and the chemical evolution -of the mantle. Moreover, convergence-related deformation can also give rise to a variety of important tectonic processes, such as regional metamorphism resulting from a perturbed thermal regime, and frequent large earthquakes due to episodic slip on a decollement tenet. We have not sought to provide an exhaustive review of the literature on convergent margins; such reviews exist (e.g., van Huene, 1984; Jarrard, 1986; Kanamon, 1986; Moore and Silver; 1987~. Instead, we hope to raise what we believe to be some of the most pressing current issues related to convergent margin deformation at all scales, up to and including those observed geodetically and seismically. Our understanding of Reformational processes at convergent boundaries has advanced considerably in the last ten to fifteen years. This improvement is due to a number of factors, both observational and theoretical. Increasingly sophisticated models have been developed to explain the geometry, kinematics 117

and mechanics of thrust belts. These models are broadly encompassed by the concept of an erogenic wedges as originally proposed by Elliot ( 1976) and Chapple (1978~. The leading edge of the overriding plate at a convergent boundary deforms into a wedge-shaped profile. The base of the wedge. is bound by an active sole thrust or decollement which accommodates most of the horizontal convergence. Mechanical models dictate that the wedge must maintain a critical taper angle for decollement slip to occur (Davis et ale, 19831. This critical taper is apparently maintained by deformation within the wedge and accretion at the base and front of the wedge. Several aspects of these models for the mechanics of erogenic wedges remain unproven and are still being critically assessed. Most of the questions concern Reformational processes and decollement structure within the deeper, more internal part of the wedge (e.g., PavIis and Brown, 1983; Platt, 1986; lamieson and Beaumont' 1988~. Unfortunately, the internal regions of most oniand thrust belts are commonly obscured by post-orogenic metamorphism (due to thermal relaxation) and by younger superimposed erogenic events. Furthennore, because the base of the wedge commonly dips more steeply than the erosional section, the deeper part of the decollement is commonly not exposed. As a result, our understanding of the large-scale structure of these boundaries remains incomplete. We contend that modern subduction zones (Figure ~ ), where the downgoing plate is oceanic, provide a unique natural laboratory for the study of erogenic deformation at a variety of scales. At the largest scale, we can examine the relations between plate motions and intra-plate deformation. At some convergent margins, the ovembing plate acts as a stress guide, resulting in large-scale contractional deformation many hundreds of kilometers landward of the trench (examples fabled "C" in Figure ~ ). The Andean thrust belt (Iordan et al, 1983; Suirez et al., 1983) is a prime -example. At other margins, such as the Marianas (Mrozowski and Hayes, ~980; Hussong and Uyeda, ~ 982), structural features suggest that extensional deformation may dominate the entire margin, virtually all of the way to the trench (examples fabled "X" in Figure ~ ). At intermediate scales, we can study tectonic processes associated with the development of an erogenic wedge. Subduction complexes can be used to critically test and to extend existing models for erogenic wedges. 118

At the small scale, we can examine the effects of a variety of low temperature Reformational processes. Rocks and sediments in this tectonic setting are subjected to a wide range of stresses, strain rates and confining pressures while under all degrees of lithification (Figure 2~. Modern subduction complexes provide well-controlled "laboratories" for the study of erogenic deformation at both intermediate and small scales: a) The Reformational environment is relatively steady and uniform over long periods of time (up to, or in some cases exceeding 10 million years). b) Temperatures are generally sufficiently low that only a limited number of microscale Reformational mechanisms are active. c) Growth of the wedge is not complicated by subaerial erosion. d) Deformation involves a relatively simple range of lithologies. e) The amount and rate of overall convergence is generally well known. f) The position of the downgoing plate is usually easily resolved by Benioff zone seismicity. g) The elastic properties of the downgoing plate are well understood. The more than 20 modern subduction complexes that presently populate the surface of the Earth (Figure 1 ) show a marked range in tectonic setting and defonnational behavior (e.g., Uyeda and Kanamori, 1979; Uyeda, 1982; larrard, 1986~. Convergence rates range from 10 to 1 00 mm/yr, and sedimentary cover on the downgoing plate from 0.2 to 10 km thick. Subduction complexes vary widely in their accretionary behavior: some grow by accretion of sediment and possibly oceanic crustal rocks, whereas others show evidence for long periods of non-accreting, possibly accompanied by loss of material (subduction erosion). The seismogenic character of the subduction thrust also shows considerable variability between margins and along strike within a single margin (e.g., Kanamori, 1971; Uyeda and Kanamori, 1979; Ruff and Kanamori, ~ 980~. In some cases fault slip occurs in an "aseismic" manner, progressing at a relatively steady rate without a significant rupture events, whereas in others berg., Sykes and Quit~meyer) slip occurs in an episodic or. seismogenic fashion, marked by large thrust earthquakes (Mb greater than 7) and long repeat times (greater than 50 years). Aseismic and seismic slip may occur at different levels in the same subduction zone (Figure 2), with seismic slip limited to the region between the shallow sediment-rich and the deeper high-temperature portions of the boundary. 119

The main limitation on research at modern subduction complexes is that their submarine setting makes them relatively inaccessible. This places restrictions on the type of research problems that can be addressed and the methods that can be used to address them. In the summary below' we highlight some important problems and research opportunities. SELECTED PROBLEMS: (1) What factors control the accretionary behavior of subduction complexes? Why do some subduction complexes grow by frontal accretion (e.g., Barbados, Cascadia, Nankai), whereas others show no evidence' of frontal accretion (e.g., southern Middle America)? Subduction erosion of the overriding plate has been proposed at a few margins (e.g., Mariana, Peru-Chiley, but this potentially important process remains poorly documented. How does such erosion take place? Does accretion occur at a deeper level at "nonaccreting" and "eroding" margins? Is accretionary behavior directly influenced by such factors as the internal structure and lithological composition of the overriding plate or the subduction geometry of the downgoing plate? (2) What are the dominant microscale Reformational mechanisms and the associated structural responses for sedimentary materials within the forearc wedge? Our understanding of microscale Reformational processes under low temperatures and variable pore fluid pressures is relatively limited compared with that for high temperature metamorphic settings. In particular, there is a wide range of opinions on the rheological behavior of unlithified sediments under high pore fluid pressures: Do these sediments deform as a low viscosity fluid (e.g., Cloos, 1982) or do they fail in a brittle fashion? What is the role of solution mass transfer (stress solution)? The temperature range and deviatoric stress conditions under which this mechanism dominates are very poorly resolved. A better understanding of this mechanism would greatly improve our ability to interpret structures and Reformational histories at ancient, uplifted subduction complexes. The deep limit of interplate thrust earthquakes is apparently controlled by temperature. However, the factors controlling the trenchward limit of such earthquakes are less well understood. Wedges at seismogenic subduction zones may show a mix of Reformational mechanisms, 320

associated with alternating high and low strain-rate regimes (co-seismic versus inter-seismic deformation). (3) How does an accreting wedge maintain a critical wedge t a p e r ? Accretion at the front of a thrust wedge must be balanced by thickening at the rear of the wedge. Does thickening occur by basal accretion (e.g., thrust imbrication) or by ductile flow within the wedge? What fraction of shortening occurs in the immediate "toe" area of the wedge, as opposed to farther upslope? How does the wedge respond to and recover from the passage of a bathymetric feature such as a seamount, transform fault, or ridge? Also important is the potential presence of a backstop within the wedge, which would appear as a distinct kinematic boundary marking the transition from an actively deforming portion of the wedge seaward of the boundary to a more stagnant region landward of it. It is uncertain how such a boundary might relate to major structural boundaries or lithologic transitions within the wedge, or to the pattern of seismicity on the master thrust beneath the wedge. (4) What factors control the localization of the subduction zone decollement and how deeply does this decollement incise into the downgoing plate? Reflection seismic profiles have demonstrated the presence of well developed decollement horizons at the front of several subduction complexes. Thrust seismicity shows that the subduction thrust remains a localized and nearly planar feature to depths on the order of 50 km. The geometry and position of this master thrust probably exerts an important control on accretionary processes and ultimately determine the flux of crustal and sedimentary materials in and out of the wedge. (5 ~ What is the nature of heat and fluid flow within subduction complexes? These fluxes can strongly influence diagenetic and metamorphic processes, and can also greatly accelerate ductile defo'~ation due to solution mass transfer (e.g., Ethendge et al., 1983; Shi and Wang, 1984, Reck, 1987~. Some important questions are: What are the major sources of fluid within the accretionary wedge: sediment compaction, dewatering of ocean crust, or gravity-driven flow? How do faults, stratigraphic units, and surface recharge affect the pattern of fluid flow? What geological conditions are ~21

required to produce and maintain high excess fluid pressure within the wedge? (6) What determines the Reformational response of the overriding plate at convergent margins? Can the development of onland erogenic zoned' far from the trench, be related to plate boundary interactions, such as changes in the rate or type of sediment accreted to the wedge, or in the geometry or rate of plate subduction? Empirical studies (e.g., Ruff and Kanamor~' ~ 980; larrard, ~ 986) have shown that several factors, including speed, dip, and age of the subducting plate are strongly correlated with the Reformational style of the overriding plate, whether contractional or extensional. Several explanations have been posed for this correlation, but we are still far short of resolving the mechanics of this interaction. Furthennore, some margins show evidence of substantial Reformational gradients, changing from active accretion at the front of the wedge to within-arc or back-arc extension (e.g., Ryukyus). Margins showing this behavior indicate that several competing processes may be involved in determining the state of stress and defonnational response in the upper plate. (7 ~ What factors govern seismic slip behavior at subduction b o u n ~ a r i e s ? It has been proposed that slip behavior, whether convergence occurs by steady slip or seismic rupture, is controlled by large-scale attributes of the plate boundary, such as the age of the downgoing lithosphere and the rate of subduction (e.g., Ruff and Kanamori, 1980; Ruff,- 1983; Jarrard, 1986~. Rock mechanicians consider large shallow earthquakes to - be caused not by rupture of new rock, but rather by a velocity weakening instability that is dependent upon rock type (e.g., Dieterich, 1978; Stuart and Mavko, 1979; Rice, 1980; Tse and Rice, 1986). Are these two points of view compatible? Does the presence or absence of subducted sediment influence the slip behavior of the subduction thrust (Byrne et al., ~ 988~? What is the mechanical behavior of rocks associated with deep subduction thrust earthquakes (at depths of about 50 km or greater, where the expected temperatures may suggest ductile flow)? How much "aseismic" slip occurs at seismogenic subduction zones? What is the physical significance associate with spatial variations in moment release during subduction thrust earthquakes (i.e., asperities)? Can these be related to geometric factors such as ramps or bends in the master thrust fault, or 122

basement structures, such as seamounts? Finally, can we achieve reliable assessment of the seismic risk at those margins that have not had instrumentally observed great earthquakes? ~ ~ ~ How is the seismogenic character of a subduction zone related to geodetically measured strain? To the first order, the elastic dislocation model can account for the temporal changes in geodetically measured strain at seismogenic convergent margins (e.g., Savage, 1983; Thatcher and Rundle, 1984~. However, there remain important gaps in our understanding of the relationship between seismicity (paleoseismic, historic, and instrumentally observed) and the results of geodetic studies. There appears to be no strain accumulating in the Shumagin Gap (Alaska Peninsula - Figure ~ ~ where there is good evidence that great earthquakes have occured. In contrast, geodetic data from the Cascades subduction zone suggest ~ buildup of strain (preseismic ?), but there is no historic evidence of large thrust earthquakes. Further complicating this problem is the presence of tectonic processes unrelated to the seismic cycle that can produce long-term secular strain. It has been proposed that short-term geodetic behavior of subduction zones is related to their seismogenic character. How are these two Reformational processes related to each other, and what are the implications of this relationship for the tectonic development of convergent margins? Finally, does the seismic character of a margin have any clear relation to the geodetically measurable secular strain? FUTURE RESEARCH DIRECTIONS Many of the problems above are being actively pursued using conventional research methods, such as field studies of uplifted subduction complexes (structure, metamorphism and fluid flow history)' routine marine surveys of modern subduction complexes (multichannel seismic profiles and swath mapping methods), and analysis of local and global seismic data associated with subduction thrust earthquakes (precise event locations, source mechanisms and rupture histories). These studies have contributed to the bulk of our present understanding of subduction complexes and will undoubtedly continue to do so. However, the direction these studies take and the rate at which they advance will be strongly affected by future "high tech" studies. For instance, results from the relatively limited suite of deep ocean drilling 123

sites at modern subduction complexes have provided an important impetus for the development of new concepts about the subduction/accretion process. Thus, we suggest several "h~gh-tech't directions with the hope of kindling more general discussion about new research directions in the study of Reformational processes at subduction complexes. (1) More emphasis and support should be directed toward improving MCS capabilities for the study of structurally complicated regions. Multichannel seismic reflection (MCS) profiles will probably remain the main technique for studying the medium and large scale structure of modern subduction complexes. Unfortunately, MCS surveys have had only limited success at imaging the deep structure of convergent margins. Part of the problem is due to the rough seafloor and relatively steep slopes that characterize this tectonic setting. Possible directions for future improvements might be the development of deep-towed source and receivers and more sophisticated processing methods to account for non-horizontal and rough bathymetry. Drill holes are a scarce resource, so it is important to advance as far as possible in using remote sensing techniques, such as seismic reflection methods. (2) Deep drilling in -the frontal region of several modern clastic-rich subduction complexes to examine the mechanical state and structural response of unlithified sediments. These sediments are just entering the upstream end of the subduction/accretion process. A thorough knowledge of their mechanical state at this stage will provide important constraints for Reformational processes operating downstream along the subduction thrust. The success of this type of drilling will depend on (a) improved abilities to control hole stability under conditions of moderate to high pore fluid pressure and (b) the development of a suite of in-situ tools to measure physical properties, stress and pore fluid pressure in the undisturbed region ahead of the drill string. (3) Deformation testing of water saturated sediments and sedimentary rocks under moderate stresses and high fluid pressures. There has been little research conducted on the Reformational behavior and structural response of sediments under conditions that typify the 124

subduction setting. At present, we can only interpolate between the low stress experiments of the soil mechanicians and the moderate stress experiments of the rock mechanicians. One critical question is: How do porosity, fluid pressure and stress path affect the bnttle/ductile behavior of sediments? Also, how are the Reformational properties of sediments affected by the long load times that typify geologic processes? (4) Long-term deployment of onshore-offshore arrays of geodetic instruments and seismometers across some well characterized subduction zones. Because geodetic studies are presently restncted to oniand traverses, little information exists (save for the odd island) about the short-term pattern of uplift and subsidence of the subduction complex and the downgoing plate. It is welt known that onland geodetic records include both a cyclic earthquake-related signal and a secular signal, due, at least in part, to steady deformation within the wedge (i.e. non- recoverable strain). The capability to monitor uplift over the entire width of the margin would allow these two signals to be separated. The ultimate goal would be to correlate the geodetic data with local structure and seismicity. The seismographic portion of this array is essential for the precise location of seismic events. Successful deployment of these arrays would help answer some major outstanding questions. For example, how is deformation partitioned across a subduction complex? What are the relative contributions of seismic versus aseismic slip, and how is the variation of moment release during a large thrust earthquake related to local structure along the subduction thrust and to the pattern of coseismic uplift? As

REFERENCES Byrne, D., D.M. Davis, and L.R. Sykes, Mechanics of the Shallow Regions of Subduction Zones and the Loci and Maximum Size of Thrust Earthquakes, Tectonics, 7, 833- 857, 1987. Chapple, W.M., Mechanics of Thin-Skinned Fold-and-Thrust Belts, Geol. Soc. Am. Bull., 89, 1189-1198, 1978. Cloos, M., Flow Melanges: Numerical modeling and geologic constraints on their origin in the Franciscan Subduction Complex, California, Geol. Soc. Am. Bull., 93, 330-345, 1982. Davis, D., J. Suppe, and F.A. Dahlen, Mechanics of fold-and-thrust belts and accretionary wedges, ]. Geophys. Res 8S, 1153-1172, 1983. Dieterich, J.H., Time Dependent Friction and the Mechanics of Stick-Slip, Pure Appl. Geophys., 116, 790-806, 1978. Elliot, D., The Motion of Thrust Sheets, I. Geophys. Res., file 949-963, 1976. Etheridge, M.A., V. J. Wall, and R.H. Vernon, The Role of the Fluid Phase During Regional Metamorphism and- Deformation, I. Metamorph. Geol., 1, 205-226' 1983. Hussong, D.M., and S. Uyeda, Tectonic Processes and the History of the Mariana Arc A Synthesis of the Results of Deep Sea Dnlling Project Leg 60, Init. Repts. DSDP, 60, 909-929, 1982. Jamieson, R.A., and Beaumont' C., Orogrny and Metamorphism; A Model for Deformation and Pressure-Temperature-Time Paths With Applications to the Central and Southern Appalachians, Tectonics, 7' 417-445, 1988. larrard, I., Relations Among Subduction Parameters, Revs. of Geophys., 24,2 17-2 84, 1986. Jordan, T E., B.~. Isacks, R.w. -Alimendinger, I.A. Brewer, V.A. Ramos, and C.~. Ando, Andean Tectonics Related to the Geometry of Subducted Nazca Plate, Geol. Soc Am. Bull., 94, 341-361, 1983. Kanamori, H., Great Earthquakes at Island Arcs and the Lithosphere, Tectonophys-ics, 12, 187-198, 1971. Kanamori, H., Rupture Processes of Subduction Zone Earthquakes, Ann Revs. Earth Planet. Sci.' 14, 293-322, 1986. Moore, I.C., and E.A. Silver, Submar~ne Accretionary Prisms, ReYs. of Geophys., 25, 1305-1312, 1987. Mrozowski, C., and D.E. Hayes, A Seismic Refiection Study of Faulting in the - Marianas Fore Arc, in Hayes, D.E. (Ed.), The Tectonic and- Geologic Evolution of Southeast Asian Seas and Islands: Washington (American Geophysical Union), pp. 223-234, 1980. PavIis, T.~., and Bruhn, R.~., Deep-seated Flow as a Mechanism for the Uplift of Broad Forearc Ridges and its Role in the Exposure of High PIT Metamorphic Terranes., Tectonics, 2, 473497, 1983. Platt, J.P., Dynamics of Orogenic Wedges and the Uplift of High-Pressure Metamorphic Rocks, Geol. Soc. Am. BITIl., 97, 1037-1053, 1986. Reck, B.H., Implications of Measured Thermal Gradients for Water Movement Through the Northeast lapan Accretionary Pnsm, I. Geophys. Res., 92, 3683-3690, 1987. Rice, I. R., The mechanics of earthquake rupture in Physics of the Earth's Interior, ed. A. M. Dziewonski, and E. Boschi, 555-649, Italian Physical Society Bologna, 1980. Ruff, L., and H. Kanamon, Seismicity and the Subduction Process, Phys. Earth Planet. Inter., 23, 240-252, 1980. Ruff, L.~., Fault Asperities Inferred From Seismic Body Waves, in Kanamori, H. and E. Boschi, Earthquakes: Observation, Theory and Interpretation, North-HolIand, Amsterdam, pp. 25 ~ -276, 1983. 126

Savage, JeC.? A Dislocation Mode] of Strain Accumulation and Release at a Subduction Zone, ]. Geophys. Res., 88, 4984-4996, 1983. Stuart, W. D, and G. M. Mavko, Earthquake Geophys. Res., 84, 2153-2160? 1979. i nstability on a strike-slip fault, J. Suarez$ G., P. Moinar' and BeCe Burchfiel, Seismicity, Fault Plane Solutions, Depth of Faulting, and Active Tectonics of the Andes of Peru, Ecuador, and Southern Chile, ]. Geophys. Res., 88, 10403-10428, 1983. Sykes, L.R., and R.C. Quittmeyer, Repeat Times of Great Earthquakes Alone Simple Plate Boundanes, in Earthquake Prediction: An International Monagr. Ser., vol. 4, edited by D.W Simpson and P.G. Richards, Washington, D.C., lggl. Review, Geophys. pp. 217-247, AGU, Thatcher, W., Rundle, J.B., A Viscoelastic Coupling Mode] for the Cyclic Deformation Due to Periodically Repeated Earthquakes at Subduction Zones, J. Geophys. Res., 89, 763 I-7640, 1984. Tse, S.T., and I.R. Rice, Crustal Earthquake Instability in Relation to the Depth Vanation of Frictional Slip Properties, IT. Geophys. Res., 91, 9452-9472' 1986. Uyeda, S., Subduction Zones: An Introduction to Comparative Subductology, Tectonophysics, 81, 133-159, 1982. Uyeda, S., and H. Kanamon, Back-arc opening and the mode of subduction, ]. Geophys. Res., 84, ]049-106l? 1979. von Huene, R. Tectonic Processes Along the Front of Modern Convergent Margins Research of the Past Decade, Ann. Revs. Earth Planet. Sci., 12,-359-381, 1984 Wang, C.Y., and Yip. Shi, On the Thermal Structure of Subduction Complexes: A Preliminary Study, ]. Geophys. Res., X9, 7709-7718, 1984. 127

FIGURE CAPTIONS FIGURE- I) Locations of major subduction zones (after Jarrard, 1986~. The letters located on the overlying plate near each subduction zone indicate the rough classification of that zone on a scale ranging from predominantly extensional (X ), thru intermediate strain character Ail, to generally contractive (C ~ These correspond, respectively to strain classes (as defined by JalTard, 1986) 1 and 2 (X), 3 thru 5 (I), and 6 and 7 (C). section thou the shallow part of a The tectonic behavior of a specific ' ' -'- structure. FIGURE 2) One schematic cross subduction zone (after Byrne et al., 1988~. convergent margin is controlled to varying degrees by the Ethology, strain, seismicity, and rheology of the subduction zone. These features vary widely between various convergent margins, offering a variety of well-controlled natural laboratories. 1 128

FIGURE 1 ..~ .~, W: 'I . , - . ~ of. ..,~. a_ all,', ~ .~ P:~ <O ~ ~ ~ $,0 .~ ~ Z Vet , o: of 1 1` ; A: W~ 1 ~~ art : , ~ o A- ~ ,= ( ~ A A J was r -~ w ~ C) z 0 or ED i.. N... _. ~:- ~. ~- ., . ri. . - I. ha: . o MINX :C ,9w, ~ ~ ·.~ · . .~ .. - .P < 129

PI GORE 2 1 ~ c:) C, fir 141 C=~ 1 trap ~= o fir lo: lo I ~ _ ~ a ~ - ~# o ~ ~ Z Vat Let fir IL fir ~ Q A Z ~. IS 1 In k: Z Q W L O In c: ~S :r Z: ~n 1 1 ~n ~ Z - Z a~ W C O 2 Cl ~ J J 1 1 Z Z O C) Z Z m m 1 o L: ~ ~ ~ . 1 ~ * - V ~ ~ o ~ V V Y 7 ; 4e - Ct \ O \> ~ ~ _ 130

Driving Forces: Slab Subduction and Mantle Convection Bradford. H. Hager Seismological Laboratory, 252-21 Division of Geological and Planetary Sciences California Institute of Technology Pasadena, CA 91125 Introduction - c, - Mande convection is He mechanism uldm~tely responsible for most geological amity at Earth's surface. To zeroth order, the lithosphere is Be cold outer thermal boundary layer of the convecting mande* Subduction of cold dense, li~osphem provides He major source Of negative buoyancy dying mande convection and, hence, surface tectonics berg., Forsythe and Uyeda, 1975; Richter and McKenzie, 1978; Hager and O'Connell, 19811. There are, however, important differences between plate tectonics and the more familiar convecting systems observed In He Oratory. Most important, He temperature dependence of He effective viscosity of mande rocks makes He thermal boundary layer mechanically sing, leading ~ nearly nod plates. This strength stabilizes He cold boundary layer against small a~plinlde pem~anons arid allows it to store substantial gTavitanona1 potential energy. Paradoxically, though going faults at subdu~cm zones make He lithosphere there locally weak, allowing rapid convergence, unlike what is observed In laboratory experiments using fluids wad tempemwe dependent v~scos~hes. This b~moda] strength distnbunon of He lithosphere distinguishes plam ~n~cs mom simple convection expedients. In addinon, Earn has a buoyant, relatively weak layer (tile crust) occupying die upper part of die thermal boundary layer. Phase changes lead to extra sources of heat and buoyancy (e.g., Schubert et al, 1975; Anderson, 1987~. These phenomena led to He observed richness of behavior of He plate tectonic style of mande convection. In dais note, ~ summarize tile current paradigms, Den state my New of He key questions that need to be addressed' as well as techniques for addressing ~em. Ibis review Is nev~cably biased towards He research In which ~ have been most heavily involved State of Current Knowledge Empirical Much of He current state of understanding of He subduction process is based on empirical associations. For example, plates with subducting slabs attached move faster Tan plates without sub~uc~ang slabs, consistent wad Me negative buoyancy associated wad slabs being a dominant Hiving force (Forsythe and Uyeda, 1975~. The maximum size of earthquakes at subduction zones is dimly proportional to convergence velocity and inversely proportional to We age, and hence negative buoyancy, of He subducting plate (Ruff and Kan~nor~, 198O). The presence of back-arc spreading is associated with steep subduction of old lithosphere, while back-arc spreading is absent where tile dip of He subducting slab is shallow and where rapid convergence is occuITing (e.g., Uyeda and 131

ganamO~7 1979~. file at show wavelengths? amp "a wenches have grid lows over them, at wavelengths of 4,000-10~000 km, Mere is a spectacular association of geoid highs with subduction zones (Chase, 1979; Hager, 1984) as can be seen in Figure I. L`ong-wavelength geoid highs are also associated with hotspot provinces (e.g., ADica and the central Pacific, Cough and Jurdy, 1980; Richards and Hager, 1988~. Inversions of lower mande structure using seismic tomography have revealed Hat these hotspot provinces at the surface am associated with anomalously slow, presumably hot, regions in the lower mantle, suggesting a lemmas link between He surface and He deepest mantle (Hager et al, 1985; Hager and Clayton, 1988). Seisrrucity paces the location of and state of stress within subducted slabs down to the base of He upper mande at the 670 km discontinuity. Slabs are gene~aBy ~ extension above 300 km depth berg., Isacks and MoInar, 1971) and In down-dip compression below 300 km depth. Se}s~c activity is high at the surface, decreases exponentially untie depth to about 350 lcm depth, dlen increases exponentially m~ Kept to 670 km, where it abruptly ceases (e.g., Vassiliou et al, 1984~. Observations Several more specific obser~ra~ons seem important In understanding He dynamics of mantle convection and sub~uchon. The fate of subducted slabs when they reach He 670 km discon~u~ s a first Her question, related to the geoche~cal evolution of Earth. Important observations relevant to this question include He topography and sharpness of this discon~liity (Hager and Clayton, 1988; Hager and Richards' 1988~. At present, Here is no observational evidence for any substantial topography (i.e., greater than 20 km) on this discontinuity. Reflection and conversion of seismic phases suggests Hat this boundary is, at least locally, very sharp (e.g., Bock and Ha, 1984~. Analysis of travel time residuals mom deep seismic events indicates Cat subduct material extends over significantly greater volumes Han has seismic acuity. Long-waveleng~ versions of Gavel time anomies projected onto He focal spheres of deep earthquakes have been interpreted as showing slabs extending deep into He lower mande, often with a kink at the 670 km discontinuity (e.g., Creager and Jordan, 1984, 1986~. Local tomographic analyses of these Aver time anomalies reveal significant thickening of subducted slabs in He transition zone, consistent wad He state of stress inferred from focal mechanisms (Zhou et al, 1987~. Regional tomographic studies of die lower mande beneath Norm America reveal high velvety anomalies Hat have been tnterpre~i as He fossil remains of the Farallon Plate (Grand 1987~. While subducts slabs are Fought to have high density and pavane a major source of the body forces Hiving global plate modons, in two locators In Soup America, He subduct slab seems to be moving subhonzontally, rather Han sinking into He mantle (e.g., Isacks and Barazan~, 1977~. This subhonzontal subduction has been proposed for North America clueing He I~ramide orogeny (e.g., Bird, 1988) The state of stress In subduction zones is minutely relay to He dynamic processes occulting. Recent obsessions of changes in stress state associated m~ major earthquakes at converging plate boundaries seem potentially important in illu~na~g He absolute level of stress In these regions. Before these major events focal mechanisms show compression Pip of the events and tension downdip, while this simanon seems to be reversed afterward (Asuz and Kanamo¢i, 1983; Dmowska and Rice, 1988; Christensen and Ruff, 1983~. Since stress drops associated with these events are 100 bars or less, this change In sign of the apparent stress state is suggestive of a low overall stress level. 332

Local tomographic studies of Me upper mande beneath southern C~~fo~ have revealed a curtain of high vel~ty matenal extending deco a Reps of 250 km beneath We Transverse Ranges in He Big Bend region of He San Andreas fault (Figure 2, after Humphreys et al, 19841. While there are no deep earthquakes associated why this feared, and hence it is not typical subduction, it has been mt~prete~ as Me convective downwe~ing of He cold, dense base of He Demean lithosphere. We basal tractions from this convection cell have been proposed as the dynamic exp~anahon for tile maintenance of the Big Bend (Humphreys, 19851, although kinematic models have also been proposed to explain He Big Bend as He result of He effects of melange monon between plates (Bi~ and-Rosenstock, 1984~. Models Many of these obse~va~aons have been interpreted quandmively In teams of numerical models. The geoid obsenranons have been interpreted In terms of fluid mechanical models Hat include He effects on He geoid of He rnn.cs anomalies introduced by dynamically maintained topography (Richards and Hager, 1984~. The geoid can be explained by two families of models (Hager and Clayton, 1988; Hager and Richanls, 1988~. The Best allows mande-w~ flow and requires a substance Unease In viscosity amass He 670 km disconunwty. The second class of models has a mantle which is chemically steamed; it Squires dial subducted slabs have very high density and predicts many hundreds of lan of dynamically maintained topography on He 670 On discontinuity. Fluid mechanical models of subduction zones based on He first mode! of mande structure show a variety of features, including kinking at tile 670 km discon~ui~ (Audis and Hager, 1988~. Such ~ model is also consistent wad He state of Stress In subducted slabs and He infers advec~ve thickening of slabs at the base of He upper mande. Simple v~scoelasac models of subduction zones have been proposed to address the Observer! change In stress sme assocmeti with great subduction zone ~hquakes (Dmowska and Rice? 19881. Key Questions There are a number of in~ponant questions suggest by these and over obsewadons. One of He most general is He relationship between He observed kinematics of subduction and the dynarrncs of He process. What driving forces are transmitted over great distances through He strong prams and what are gentled by local sources of buoyancy? Related issues are the stress level and amount of dissipation occ~g locally ~ subduction zones since the Wiving fomes from density congests am evennuaDy balances! by dissipative resisting forces, the distribution of this dissipation is a crucial question (e.g. Christensen, 1985~. - A relatec! quesdon is the-amoa~nt of neganve buoyancy assoc~ated m~ subduchng slabs. Is this mainly He result of simple Vernal expansion, or are He effects of phase changes dominant (Anderson, 1987~? Knowing He phase diagram of subducting slabs Is important for understanding He driving forces, as weD as detem~ng whether He slab penetrates He 670 Ian discontinue. Detemun~ng He fate of subducting matena] at 670 ~ mportant issue for much of Earth Science. What happens to tile crust? Is it shred back into He depleted lithosphere or does it separate? Does subducted material unx into He lower mantle? Does it penetrate 133

briefly only to be regurgitated when reheal? Is it stopped at 670 km Ape? How are subducted slabs distorted in this radon (s=, e.g., Silver et al, 1988~? On a more regional scale, important questions include Me mechanics of flat su~uchon and the dynamics of back arc basins. What forces are responsible for sliding a subducting plate for ~l,000 km beneath an overriding plate? Once back arc Spreading is Equated, how is He back arc spreading center shut off? The variation of He dip of subducted slabs fiom place to place has not yet been explained in a comprehensive model. What are tile competing effects of slab buoyancy, mange viscosity, and global flow (e.g., Hager et al, 1983~? How are die dynamics of slab dip and back arc spreading related? While tile empirical association of maximum earthquake she win convergence velocity and plate age has intuitive appeal, it is genmally recognized dial eardlquake size is controlled by the distribution of asperities on He fault plane (ego Kanamon, 1986). How do tile empirical variables relate to the physical state of the fade plane? On a local scale, -~e process of Snap scale colon such as is seen nomographically beneath southern California raises ~ number of questions. How are these mande mohons linked to deformation in He upper crust? How Is He convective umescale linked to Be nmescale assocm~ wad earthquakes? What is He dis~bunon of crush theologies? Why does the lithosphere go unstable ~ some places, but not elsewhere? The latter question is closely relay to He prowess of Craig stable craton~c nucia. The premier question associated with subduction zones is what causes the Non of a new subduction zone? How is tibe sing, cold lithosphere inimically Niche ~ form a weak plate boundary? The subdllcdon process is extremely important ~ regulatung He Vernal balance of Earn; understanding the ~tiadon of subduction ~ al In understanding die dynamic evolution of our planes. New Observations, Experiments, and Models Understanding He process of subduction win tequila aches In a number of areas spanning a range of geosciences. Given die limited resolution of most techniques, these advises win be most p~ucave if Hey are carried out In such a way as ~ answer specific questions and test specific, relevant hypotheses. Suggested acovides are grouped by discipline, roughly in arder of Knotty within each group. Seismology The fate of subducted slabs when Hey reach He 670 On discontinmtr is a first carder question that can be addressed by se~smolo~ts. The topography and sharpness of the 670 km discontinuity are two features Hat can disseminate between mantle-w~de and chemically stratified convection scenarios. imaging of the 670 An disconomiity in He vicinity of subducted slabs is of highest pewter. Determination of He seismic velocity Euchre in tile vicinity of slabs by tomographic means is also a high priority. DeteImining the shape of subducting slabs places strong constraints on dynamic models. InvesugaDon of locations in He deep n~nrltle beneath fossil subduction zones are important to ~nacase He temporal coverage of the subduction process. 134

Regional tomographic studies of He upper mande In both Stoically above areas and cratons would help ~ understand Be dis~bunon of "lithosphenc Lips" such as have been observed beneath soughed California, as wed as the processes of craton~zanon. Further ~nvesuganon of temporal and spatial vans In focal mechanisms associated wide large earthquakes will help to constrain the absolute level of stress ~ subduction zones, as well as the mechanical properties of the Biosphere enosphe~e system. Given the importance of tile concept of asperities, direct imaging of fault asperities by reflection seismology would be an important accomplishment. Reflection seismology and Over seismological techniques should also be used to image Me deep crust to cons Be structures and material propemes involved In Be coupling between mande convection and crustal deformation. Determination of whether over phase boundaries within Be slabs are elevate ~ Pressed is of high pounce, bearing on questions of tile mineralogy of Be mantle, the thee state of Be slab, and Be magnitude of body force bring subduction. A regional tomographic study of Be upper nuance in tile vicinity of the flat-lying subducted slabs beneath South America would place implant constraints on Be dynamics of flat subduction. Such a study would abbess whether nature hete~geneitr outside die slab important in driving die system. Geodesy The newly developed, highly accurate space-based geodetic techniques (e.g., GPS) make it possible to obey frugal obsenradons at ~eLai~~rely Ji~e cost* Surveys should be catTied out in regions such as sounded C:alifom~a where good tomographic images of mantle stmc~re exist in order ~ better consuain the coupling of mande convection to surface tectonics. These measurements should ~ made frequently enough ~ Be coupling of forces Cam convective tunescales to the omescales of Be seismic cycle can be addressed Since tile basic mmpmal spectrum of regiorm1 crustal defoImaaon is as yet unconstrained by obsen~a~ons, permanent, congruously mo'iitared regional swain networks should be installed in a few active regiOlISe Transfer of stress and swain after large subduction zone events should be monitored to address Be questions of stress level anti coupling of subduction zones. This requites Anal epoch surreys. Development of high precision underwater geodetic controls is also very Important Obsenrmons of spatial var~abons In gravity have—ven useful In discriminating among geodynamic models. Gathering of data sew spanning the con~ent-ocean t~nsinons ~ active margin amas would be very valuable to incIeas~g our tmde~tanding of dynamic processes associated wad subduction Mineral physics Determination of He state and physical properties of n~at~ials under ambient conditions is crucial for making further progress in understanding Earn dynamics. Determining phase diagrams for subduct slabs is Important for coast nag Be body forces associated wad subducted slabs. The p~dichons of these phase diagrams must be tested using seismological observations to dimnminate among different mnHelc of mantle In temperature, and composition. _~ _^ ,~,,~,,,._ v ~ ~~? 135

Be=r constructs on crustal and mantle rheology are also inane Rheolog~cal desc~pnons are needed on aU dmescales, fun betide fails ~ mscoelasuc defamation · ~ S. OWE creeping convection. Numerical Modeling Prowess in geodynamics requires quanutanve testing of hypotheses against observations. Numencal modeling helps to provide He Unison needed ~ formulate hypotheses to be tested, as well as providing quanutaDve predictions ~ be test win Increasing computations power available, numencal models wig continue to become more realistic. An important improvement win be the ability to amass Be effects of Wee dimensions and rime dependence. Advances ~ computational geophysics will be most rapid if trained nllmencal analysts work closely m~ geophysicists. For convection modeling, important problems to address include He effects of phase changes and vanadons in composition and rheology on flow. Specifically, He mteraciion of subducted slabs wad He 670 km discontinue must be addressed for a wide range of models of upper and lower mande composition. In these models, it wiD be important to consider realistic geometries for subduction lie., asymme~ca] convergence and Wee dimensions), as weD as He effects of global flow. The models must be suff~aentiv wed ., resolved to address He amount of endowment of layers wad differing compositions. The process of stabilization of subcraton~c lithosphere is another problem involving var~anons in composition and phase. We problem of the mechanics of subhonzon~ suWucdon Is anodes important problem Hat requires a fully dynamic Laurent ~ is ~mpormnt Mat Lose models be guided by He observanons of mande structure and rheologr discussed above. Transmission of stress and strain Hugh v~scoelashc effects should be addressed. These models should ~ncIude Wee dimensional effccm, as wed as malishc paTamelenzahons of He rhealog~cal vananons within He crust and mande. Dynamic models of flow in He back arc region' including dynarrncally determined flab dips, should be posed to amass the questions of ~tianon and cessation of back are spreading. 136

RIBLIOGRAPHY Anderson D.L., 1987. Thermally induced phase changes, lateral heterogeneity of He mantle, continental roots and deep slab anomalies, J. Geophys. Res., 92, 10341-10349 Astiz, L. an] Kanamor~, H., 1986. Inte~plate coupling and temporal Deacon of mechanisms of ~ntermediate-~p~ earthquakes In Chile, BSSA' 76, 1614-1622. Bird, P., 1988, Form anon of He Rocky Mountains, Western United States: A continuum com- puter motiel, Science, ES9, 1501-1507. Bird, P. and Rosenstock, R.W., 1984. Kinematics of present crust and mantle flow in Southern California, Geol. Soc. Amer. Bull., 95, 94~957. Bock, G. and Ha, J., 1984. Short period S-P conversion at a depth near 700 km, Geophys. J.R. astir. Soc., 77, 593-614. Christensen, U.R., 1985. Thermal evolution models for the Earth, J. Geophys. Res., 90, 2995-3007. Christensen, C. and Ruff, L., 1983. Outer-nse earthquakes and seismic coupling, Geophys. Res. I`ett., 10, 697-700. Creager, K.C. and Jordan, T.H*, 1984. Slab pene~ai~on into the lower mande, J. Geophys. Res., 89, 3031-3049. Creager, K.C. and Jordan, T.H., 1986. Slab penetration into the lower mande beneath the Manna and other island arcs in the Northwest Pacific, -1. Geophys. Res., 91, 3573-3589. Trough, S.T. and Jurdy,~D*M., 1980. Subducted lithosphere, hotspots and He geoid, Earth Planet. Sci. Lett., US, 15-22. Dmowska, R., Rice, I., Lovison, Lo., and Josell, D., 1988. Stress transfer and seismic phenomena in coupled subduction zones during the earthquake cycle, J. Geophys. Res., US, 7869-7884. Forsyth, D.W. and Uyeda, S., 1975. On the relative importance of the driving forces of plate motions, Geophys. ]. R. astr. Soc., IS, 163-200. 137

Grand, S.- P., 1987, Tomographic inversion for shear velocity beneath the North American plate, d. Geophys. Res., 92 14065-14090. Hager, B.H., 1984. Subducted slabs and the gcoid: Constraints on mande rheology and flow, J. Geophys. Res., 89, 6003-6015. Hager, B.H. and Clayton, R.W., 1988. Constraints on the structure of mande convection using seismic observations, flow models, and the geoid, in W.R. Peltier, ed. Mantle Convection, Gordon and Breach, New York, in press. Hager, B.H., Clayton, W.R., Richards, M.A., Comer, R.P., and Dziewonski, M., 1985. Lower mantle heterogeneity, dynamic topography and the geoid, Nature, S1S' 541-545. Hager, B.H. and O'Connell, R.J., 1981. A simple global model of plate dynamics and mantle convection, J. Geophys. Rese, 86) 4843~867. Hager, B*H., O'Connell, R.J., and Raefsky, A., 1983. Subduction, back-arc spreading and global mantle flow, Tectonophysics' 99, 165-189. Hager, B.H. and Richards, M.A., 1988. Long-waveleng~ variations in Earth's geoid: Physical models and dynamical implications, submitter! to Phil. Trans. Roy. Soc. Lond. A. Humphreys, Eugene D., 1985. Studies of the crust-mantle system beneath Southern California, Ph.D. Thesis, California Institute of Technology, 189 pp. Humphreys, E., Clayton, R.W., and Hager, B.H., 1984. A tomographic image of mantle structure beneath southern California, Geophys. Res. 1,ett., 11, 625-627. Isacks, B.L. and Barazangi, M., 1977. Geometry of Benioff zones: Lateral segmentation and downwards bending of He subducted lithosphere, in Island Arcs, Deep Sea Trenches, and Back-Arc Basins, ed by M. Tal~vani and W.C. Pitman, m, American Geophysical Union, Washington, D.C., 480 pp. Isacks, B. and MoInar, P., 1971. Distribution of stresses in the descending lithosphere from a global survey of focal mechanisms solutions of mande earthquakes, Rev. Geophys. Sp. Phys., 9, 103-174. Kanamor~, H., 1986. Rupture process of subduci~on-wne earthquakes, Ann. Rev. Earth Planet. Sci., 14, 293-322. Lerch, F.~., Klosko, S.M., and Patch, G.B., 1983. A refined gravity model from LAGEOS (GEM-L2), NASA Tech. Memo.' 81986. 138

Richards, M.A. and Hager, B.H., 1984. Geoid anomalies in a dynamic Earn, J. Geophys. Res., 89, 5987-6002. Richards, M.A. and Hager, B.H., 1988. The Earth's geoid and the large-scale structure of mantle convection, The Physics of the Planets, ed. by S. K. Runcorn, John Wiley & Sons Ltd., 247-272. Richter, F.M. and McKenzie, D.P., 1978. Simple plate models of mantle convection, J. Gco- phys.' 4¢, 441-471. Ruff, L. and Kanamori, H., 1980. Seismicity and the subduction process, Phys. Earth Planet. Inter., 28, 240-252. Schubert, G., Yuen, D.A., and Turcotte, D.L., 1975. Role of phase transitions in a dynamic mantle, Geophys. J. R. astr. Soc.' 42, 705-735. Silver, P. G., Carlson, R. W., and Olson, P., 1988, Deep slabs, geochemical heterogeneity, and the large-scale structure of mantle convection, Ann. Rev. Earth Planet. Sci., 16, 477-541. Uyeda, S. and Kanamori, H., 1979. Back-arc opening and the mode of subduction, J. Geophys. Res., 84, lW9-1061. Vassiliou, M.S., Hager, B.H., and Raefsky, A., 1984. The distribution of earthquakes with depth and stress in subducting slabs, A. Geodyn., 1, ~ I-28. Zhou, H.-W. and Clayton, R.W., 1987. Abstract, EOS, Trans. Am. Geophys. U., 6S, 1379. i39

Figure Captions Figure la) The observed long-wavelength geoid Perch et al, 1983) referred to the hydrostatic figure of Me earth ( f = 1/299.63), Ads plate boundaries and hotspots indicated. The contour interval is 20m and geoid lows am shaded. Cylindrical equidistant projection. Figure Ib) The observed geoid, filtered to include spherical harmonic degrees 4-9 to emphasize We association win subduction zones. Figure Ic) A mode} geoid calculated from a fluid dynamical mode} of mantle flow Liven by density contrasts infested for subducted slabs (Hager, 1984~. Figure 2) Tomographic reconstruction of He mantle structure beneath southern California In tile upper-left parle} a map view of the velocity structure at a depUl of 100 lan is superimposed on a location map. Also shown are He locations of He cross sections shown ~ He over three panels. These sections extend Cam He surface to 500 km Reps, with no vertical exaggeration. The contour interval Is 1.5% relative velocity v~ai~ons, with regions faster Can I.5% dotted and regions slower than -~.5% hatched The major feature is a slaWlike high velocity anomaly penetaung the uppermost mantle beneath He Transverse Ranges (After Humphreys et al.). 140

- ~ - 1 ma so o c: v: o 141 Lo Ct c - Or - L

9 in l Don an; o c-' I= o 142 i a, CO - A - C~ UD - teD Lr) CU g C ~7 G

l to o C:5 = - ~- ~ :: Hi: !:' tin, ,i,ii!,t...._ _ lily CO -

,]4' i—' ' ; ~ ~ ,` rat ran .,. 1~1~' I ~'1'-.. it I 1 =N .-~ I ~ y ~ : N`, ! ~ 1 Ned, _ _ At--— ~ 1 '':1i. 1 - ~b by din _ _. 144 e 11, _ I ! ~ ~ . ,_, ~ ma: ' .^i '- . - ,." 1 · · . . .' : ?~'' .\ . ,,.- ~ §'""f""'"""'`., -, ". ."' "' ,,r, ~~ ~ .,1 t " ,.' I \\ TV __ ~;~CtY ~ -4

@ @ ma -ma 145

INITIATION OF SUBDUCTION Dan Gary Introduction One aspect of plate tectonics that has received remarkably little attention to date is the initiation of the activity that develops the wide range of responses now observed at convergent plate boundaries. The state of an incipient convergent plate boundary may be impossible to recognize except in retrospect, but there have been few if any detailed studies even of young arc margins, particularly studies focused toward their origin. Understanding how convergent plate margins initiate will not merely fil 1 a gap in our understanding of the mechanics of these featurese. It will also elucidate the signify cance of metamorphic aureoles beneath ophiolites, the origin of boninities, the enigmatic distribution of forearc igneous rocks having arc provenance, and in general, the nature of forearm basements. Clearly the initiation of convergent margins must be better understood if we are to have unified understanding of how plate boundaries evolve. State of Knowledge The character and location of the initial phases of plate convergence are still very poorly constrained. Such ideas as do exist are most often based on sparse data and inferences from highly evolved, active plate boundaries. Remarkably few data have been derived from very young boundaries. Because convergent Pilate boundaries lie both along continental margins, as marginal arc systems, and within ocean basins, as island arc systems, there is still a question as to which are preferred settings and why. There is a general sense that most convergent boundaries form near a pre-existing interface, but largely this ~ s an exercise in inductive logic . Thus, in cases 146

where arc systems are presumed to have formed within ocean basins, and to have trapped a back-arc basin, they are suggested to have initiated along fracture/transform zones or spreading ridges. This idea is re-inforced by the existence of ophiolitic slabs having oceanic crustal affinities that now lie above continental margins. Such ophiolitics must have occupied the forearcs ot arc systems, clearly faced the continent, and are often backed against an existing ocean basin (e.g. the Semail, Papuan ophiolites). In a few cases the orientation of spreading structures and fossil transform zones within an ophiolite suggest that the ophiolitic-bearing arc formed along a transform boundary (e.g. Casey et al., 1983), but in other examples the internal geometry of the ophiolite is incompatible with this explanation (e.g. Karig, 1983). Many or most convergent boundaries are assumed to have formed along continental margins, primarily because of the many arc systems now in that setting or separated from continental margins by extensional back arc basins. Ironically there is no example in which a convergent margin is observed to be initiating or to have recently initiated along a passive continental margin. There are, however, examples in which arcs are or have recently formed in back arc basins, along the rear margins of island arcs (e.g. Western Honshu, New Hebrides, New Guinea). Although there is little argument that arcs do form, generally, along such crustal interfaces, there is much less concensus as to the degree of coincidence of the interface and the incipient arc. Many models infer that the initial break lies well into the oceanic crust, and that the resultant forearc is underlain by oceanic crust (e.g. Hamilton, 1978; Fig. 1). The more explicit model of Dickinson and Seeley (1979) assumes that the initial rupture has a regular, large curvature and cannot follow the irregular outline of a 147

continental margin (Fig. 2). In this model oceanic crust would be trapped in sections of forearcs representing marginal embayments. Karig (1982) argues from both mechanical and observational bases that the locus of initiation should more closely follow the interface between continental and oceanic crust. The nature of the forearc basement is a critical component of arc initiation, not only for the understanding of arc evolution but also for interpretation of ophiolites. There is much less information on this topic than is generally realized. Most is derived from three sources: 1. geologic relations in presumed examples of old and emerged forearc basins. 2. Seismic velocity data from forearcs. 3. Dredge results from oceanic forearcs. Of emergent forearc basins, the Great Valley of California has probably influenced our thinking the most, as it is located near a concentration of geologists who shaped early ideas about convergent margins. Subsequent information from this area suggests' however, that the Coast Range ophiolite, which underlies the western flank of this basin, is not autochthonous with respect to the Sierran arc. A very similar situation was presented by the Zambales ophiolite of Central Valley, Luzon, which also initially was identified as a forearc basin floored by oceanic crust. In this case the ophiolite proved to have been sutured to the arc, so that the basin would more correctly be termed a successor basin. Seismic velocity data from forearms is still relatives y rare and ambivalent. Some forearc basins are clearly underlain by an old marginal slope (Fig. 3; Westbrook, et al., 1988), others in part by probable-deformed sediments. In both cases velocity structures are intermediate between oceanic and continental. No oceanic-floored forearc can be defended strongly on the basis of seismic data. 148

Mafic and ultr~mafic rocks are commonly dredged from, and have been drilled in a number of oceanic forearcs. These rocks, for the most part, have been shown to have strong affinities with arc magmatism, and not to be oceanic crust. This observation is enigmatic in itself, but offers no support for the initiation of arc systems within ocean crustal plates. Initiation of arc systems has also been addressed, both explicitly and implicitly, by students of ophiolite complexes. For the most part ideas of arc origin from this perspective are based on observations from the ophiolite and its surroundings, and seldom from the settings in which arcs might initiate. Unfortunately, observation of these subsequently collided, uplifted, and dissected masses permit a wide latitude of interpretations. The high temperature basal metamorphic aureole, for instance, has led to ideas that.some arcs initiate along spreading r' dges . Probably the most successful use of ophiolite data to guide arc initiation has been by Casey and Dewey (1984), but the oceanographic data even- in their treatment are scant, biased, and in some cases erroneous. The problem of lack of integration of land and oceanographic approaches to the origin of ophiolites, and indirectly to arc initiation, was illustrated by the near absence of an oceanographic input at a major conference on "Ophiolites and Oceanic Lithosphere" (lamentation of Gass, et al ., 1984, p . 1) . Ideas dealing with processes that trigger initiation of convergence are equally unconstrained. The appeal to a "Wilson" cycle implies a progressive chain of events and inevitability that, as yet, has no justification. Moreover, it merely substitutes a name for processes that are only vaguely understood. Collisions are commonly assumed to cause the shift in location of plate convergence. However, not only is collision a more stable situation than 149

was intially assumed, but also the location at which new convergence initiates remains to be defined. Instead of approaching the problem of arc initiation in a two-dimensional sense, several suggestions have been made that propagation is affected or depends on processes that migrate along the plate boundary. Dewey (1975) has noted that migrating poles of plate rotation can change transform or even spreading boundaries into convergent ones. Conversion of long transforms to convergent boundaries appears to explain recent subduction in places like the southern New Hebrides (Matthew-Hunter island section). Conversion of spreading ridges to subduction zones by pole shifts either requires very large shifts or assumes that very slow convergence rates are adequate to generate typical subduction characteristics. Convergence directly from a typical spreading ridge to a recognizable convergent margin is very speculative and unlikely. Migration of a new convergent margin along a crustal interface, analagous to the propagation of a crack tip, had been suggested on-the basis of the apparent behavior at both ends of the Philippine Trench. Such crack tips might propagate across areas of oceanic crust and trap it in a forearm position (Fig. 4). Stress concentrations at these tips could alter calculations of stress- strength relations along continental margins, such as those of Cloetingh et al., 1983 e - Perhaps because of the lack Of concensus concerning the formative processes for convergent margins, very little effort has been expended on the characterization of these zones. Almost as afterthoug~chs of studies of arc magmatism, chemical compositions of initial magmas have been suggested to be highly anomalous, and the initial magmatic arc has been suggested to be spread 150

out over a broad zone so as to generate a forearm ophiolite of arc provenance (e.g. Hawkins et al., 1984). There is a similar dearth of information about crustal displacements preceding and accompanying initiation of convergence. The curvation of the downgoing plate in the vertical plane is much greater in very young arcs than in other ones, probably because of initially very narrow forearms. There should be very large precursor vertical displacements adjacent to the interface that will become a subduction zone, but the only information available that might support this is a now mislaid report of large uplift along the northeast coast of Luzon~preceding northward migration of the offshore trench. Outstanding Ouestions With so little known about the initiation of convergence, almost every aspect of the process is a question. Certainly there are outstanding questions related to the-geometries,-states of stress, magmatic character, geophysical signature, and structure of the initial response to convergence. These are tabulated as follows: 1. Where, relative to~previously existing features and boundaries, do arcs initiate? -What implications does this geometry have for the basement of forearms and the development of-ophiolites? If convergent margins originate at different settings, how are these identified and discriminated in the geologic record, and what is the distribution of the various initial settings ? ~ ~ - 2. What is the state of stress at the settings destined to become convergent plate boundaries? What is the distribution or nature of crustal strength that governs initiation? 151

3. What are the precursor and initial displacements and geometries of convergent margins? What is the initial geometry of the downgoing slab and what is the force balance on it? 4. What are the characteristics of magmatism at the initiation of a convergent margin? What is the lag time before volcanism appears? What is the initial position, width, and structure of this magmatism? What is the chemistry, petrology, and petrogenesis of initial magmation? 5. What is the mode of rupture when a convergent margin initiates? Does this rupute propagate along strike and if so, at what speeds? Targets for Future Research Research targets for the understanding of the initiation of convergence can be divided into groups Chat are defined on the mode of attack. Many questions concerning very young and incipient arcs can be addressed with conventional oceanographic techiques, directed toward appropriate examples. On-land studies, although earlier criticized, are still very important, particularly if integrated and iterated with data from the oceanic realm. More ambitious approaches can be envisaged as the data base develops, and would involve more complex, expensive, and/or Futuristic technologies. 1. Conventional Oceanographic Studies The most pressing task that would employ conventional oceanographic techniques in the characterization of very young or incipient convergent margins. Examples that come to mind are the north and south ends of the Philippine Arc (east side) , the eastern end of the New Hebrides Arc, Gorringe Banlc(?), ache Mussau Trough' the northwestern margin of Honshu, and the northern margin of Papua/~ew Guinea. 152

The objectives and techniques are almost self-evident. High frequency acoustiCs, in particular swath mapping and side scan methods, are needed to develop the details of morphology and structure and to identify areas of active volcanism. Seismic methods should include refraction profiles across the incipient arc' particularly across the forearc. Gravity and heat flow measurement would provide needed information on the initial temperature and mass distributions Dredging, guided by the swath mapping and side scan data, is a simple but necessary technique to obtain a preliminary understanding of early arc magmatism. 2. On-land studies of ophiolites and other emergent forearcs. The danger in misinterpreting or missing the earliest phases of arc activity in an emergent convergent margin are offset by the exposure and ability to conduct studies of details and of deeper structures. Regional relationships are often better expressed, with the constraints of stratigraphic control. Crustal deflections and displacements associated with initiation of convergence can be measured using geomorphic and geodetic techniques. 3 . Special Proj ects Following seismic studies, the structure of forearcs in incipient arcs should be probed with drill holes. Such holes will not only establish the structural and petrologic nature of the forearc but will help determine the location of the initial rupture with respect to the earlier geologic framework. Such holes should also penetrate the young plate interface to determine shallow stresses, and structural and thermal responses. An array of shallow holes around the propagating 153

tip of a new trench could be used to determine the stress in that area. Seismologic studies, perhaps with arrays of OBS's could be used to evaluate the geometry and Precocity structure of the descending oceanic lithosphere . 154

REFERENCES Biju-Duval, B., Le Quellec, P., Mascle, A., Renard, A., and Valery, P., 1982, Multibeam bathymetric survey and high resolution seismic investigation on the Barbados Ridge Complex (Eastern Caribbean): a key to the knowledge and interpretation of an accretionary wedge: Tectonophysics, fir. 80, p. 275- 304 . Casey, J . F., Karson, J .A., El~chon, D., Rosencrantz , E., and Titus , M., 1983 , Reconstruction of the geometry of Accretion during formation of the Bay of Islands Ophiolite Complex: Tectonics, v. 2, p . 509 - 528 . Casey, J . F., and Dewey, J e F. ~ 1984 ~ In' tiation of subduction zones along transform and accreting plate boundaries, triple junction evolution, and forearc spreading centires- - implications for ophiolitic geology and abduction ~ n Ophiolites and Oceanic Lithosphere ~ I . G . Gass, S . J . Lippard, and A.W. Shelton, eds.) Geol. Soc. Lond. Spec. Publ. 13, p. 269-291. Cloetingh, S.A.P.L., Uortel, M.J.R., and Vlaar, N.J., 1983, State of stress at passive margins and initiation of subduction zones-AAPG Mem. 34, p. 717- 723. Dewey, J . F., 1975 , Finite plate evolution: imp' ication for the evolution of transforms, triple junctions, and orogenic belts. Am. J. Sci*, v. 275-A, p . 260- 284. Di ckinson , W . R ., and Seeley , D . R ., 1979 , Structure and stratigraphy of forearc regions: Amer. Assoc. Petroleum Geol. Bull; v. 63, p. 2-31. (:ass, I.G., Lippard, S.G., and Shelton, A.W., (eds), 1984, Ophiol ites and oceanic lithosphere : Geol . Soc . Lond., Sp . Pub1 . 13 , p . 1- 6 . Hamil ton, to., 1979, Tectonics of the Indonesian region: U. S. Geol. Survey, Prof. Paper, 1078, 345p. Hawkins, Jim., Bloomer, S.H., Evans, C.A., and Melchior, J.T., 1984, Evolution of intra-oceanic arc Trench systems: Tectonophysics, v. 102, p. 175-205. Karig, D. E ., 1982 , Initiation of subduction zones : implication for arc evolution and ophiolite development , J . Geol . Soc . Lond., Sp . Publ ., 10 , p . 563 - 576 . Karig, D.E., 1983, Accreted terranes in the northern part of the Philippine Archipelago; Tectonics, v. 2, p. 211-236 ~ Westbrook, G.K., Ladd, J.~., Buh1, P., Bangs, N., and Tiley, G.J., 1988, Cross section of an accretionary wedge: Barbados Ridge complex: Geology, v. 16, p. 631-635. 155

TRENCH OUTER-ARC BASIN OUlER-ARC RIDGED Pr~subduction strata, beneath basin sediments, lie on oceanic crust 0 ._ - 0 o In ~ ' ' Melange wedge2~2-2.4 _ ~ Phse~transition ~ SUBDUCTING PLATE Mantle 3.4 o 50 100 1 50 KILOMETERS Fig. 1 Section of the Sunda forearc from Ilamilton (1978) showing common interpretation of oceanic crust ~ in black) under the forearc (outer-arc) teas in. 156

A' ' - - ~ MARGINALLY C0~INENT ~ -TRAM t — ~—iFIG.4Bl~)- ~ Baja Ft~ \\ .. TRaPPED DREG ~~NG RIMED OC—IC CONT1~lAL MARO:N CRUST OCEAN 8. r.ONTiNENTAt MARGIN ARC-TRENCi] SYSTEM iFlAN VIEW it; |-J. CE)NT1~JEN7~AL NlAF.;~'i {,Hr ti'Fl\'~H ~YtiTrM !R'(ICI; )!(M) Fig. 2 Initiation model of Dickinson & Seeley (1978) showing zones of trapped oceanic crust in reentrance of the continental margin. 157

HW SE~I W ~ CENTRAL ATLANTIC VOLCANIC I TOBAGO TROU6HtBARBADOs CREST BTRBADOSI BARBADOS RIDGE RECENT ABYSSAL Scc t!.~ ~_ 6 8 1 0 0 50k'~i ~ _ . NW SE | W E VOLCANIC ~ BARBADOs BARBADOS CENTRAL RECENT ATLANTIC ARC TOBAGO TROUGH - Ct.E-ST ' TROUGH BARBADOS RIDGE ACCRETION t- ARYSSAL Fig 18 Fig t7 ~ 1 Fig 9 figil F;~7Fig 13 ~ ·~_ a_ _~ r~. _ ~ .... ~ n c~ctaccous corihbcon crust Wotcr _ occonic ~ helow thc crust Icssc, Ant'IIcs 0 50km sedimantc~ry - basemcnt (Pa Icocenc to tniddle miocene) Lowc r moS ~ B p I ioconc hori z on Fig. 3. Line drawings across the I~c~r Antilles active margin from multichannel high resolution sasmic data (CEPM An~illes III survey). Crustal sections are from Westbrooic ( 19?5). Location on Fig. 2. 158 Q 4 6 _ ~ 1 0

~ Mariana Type Tonga- New Hebrides Type /: A. B. / O~` N.; .%= /; C- 1. i// / ,~u;`,); ~~ ' ~ C-2. Gulf of Colifornia Type FIG. ~ Several examples of plate geometries along convergent margins in which transverse upper plate spreading zones could lead to creation of ophiolite sheets in a forearc setting. (A) In this case, modelled after the southern end of the Manana arc, forearc rifting results from the extension of a back-arc spreading zone to the trench. Cessation of back-arc spreading could lead to the creation of a volcanic chain across very young 'oceanic' crust. (B) In a related case, drawn from the southern New Hebrides arc, upper plate rifting separates trench from transform segments of plate boundary. Initiation of subduction and of a volcanic arc would follow the north-easterly migration of the triple junction. (C) In a different set of circumstances, as could evolve in the Gulf of California, a main ocean basin spreading zone could rupture a pre-existing arc system, creating a large area of oceanic crust within the arc or continental crust ( 1~. Migration of poles of relative motion might subsequently reinitiate subduction along the margin, trapping oceanic crust in and behind the forearc area (2~. ~ from Karig, 1982 ~ 159

INTRAOCEANIC CONVERGENT MARGINS Brian Taylor Hawaii Institute of Geophysics University of Hawaii Honolulu, Hawaii 96822 Introduction Intraoceanic convergent margins comprise three principal tectonic components (Fig. 1): forearc, volcanic arc, and backarc, the relative position of which may change through time. They may be defined as those convergent margins in which both the subducting and backarc regions are oceanic. They include the W. Aleutian, Kuril, Bonin, Mariana, New Britain, Solomon, New Hebrides, Tonga, Kermadec, Scotia, and Antilles arcs. Some arcs known to be built on continental fragments and composite terranes, such as Japan, Ryukyus' Philippines' Banda, S. Shetlands, Middle America, and Calabria, may also be included in a broad view, especially given some uncertainty that the former group of arcs were definitely built only on oceanic crust. We shall consider the processes occurring at these margins under the following three headings: magmatism; structure and stratigraphy; fluids and thermal regime. In each cases several important questions that need to be addressed are summarized. Magmatism 1) What controls the temporal and spatial distribution of magmatism? a) The depth to the Benioff zone beneath the volcanic front is commonly about 100km, but there are significant exceptions such as the New Hebrides where it is 170km. Most island arcs are festoon shaped in plan view (Fig. 2), but some 'arcs' such as the Bonins and Tonga are straight. The volcanic front of an island arc is typically segmented, and the distance between volcanoes varies from 10 to 100's of kilometers. Not all subduction zones have volcanoes. Individual volcanoes as well as chains occur behind the volcanic front for tens to over 200km (e.g. Willaumez Peninsula, New Britain: Zenisu Ridge, SW of O Shima). What controls these phenomena: the stress and structures of the upper plate, the structures and geometry of the subducting plate, andJor the pattern of mantle convection? b) Several special events may punctuate the magmatic evolution of an arc. These include: the initial subduction beneath the arc; arc rifting and backarc spreading; subduction of spreading centers, seamounts and ridges- plateaux; and arc reversal. The magmatic effects of these events are controversial. Basement samples from the Bonin-Mariana and Tonga-Kermadec forearcs are composed of middle Eocene to early Oligocene arc tholeiites and boninites (Fig. 3). Were these 150-200km wide fore arc s covered by unusually extensive arc volcanism following the initial stages of subduction? Did subsequent magmatism occur in the forearc? Anomalous tholeiitic or granitic magmatism in the forearc has been proposed to accompany subduction of spreading ridges, and both associated increases and decreases in arc magmatism have been postulated. In the present subduction of the Woodlark spreading system under the Solomon Islands, arc magmas are being erupted on both sides of the trench. This phenomenon 160

requires lateral migration of arc magmas and/or lateral convection of previously enriched mantle sources. Most observers regard magmatism as episodic and some suggest circum-Pacific synchroneity of magmatic pulses, possibly related to plate kinematic effects. In the western Pacific these variations have been correlated with periods of backarc spreading, but authors are divided as to whether volcanic maxima (e.g. present Mar~ana-Bonins) or minima (e.g. present Ryukyus) are associated with the initiation of backarc spreading. A hiatus in volcanism may be associated with aseismic ridge subduction (e.g.. Yap) and arc reversal e g. mid-Miocene Solomons-New Hebrides) but not necessarily (e.g. Id. Ryukyus and Philippines respectively). Arc reversal commonly results in a new volcanic front, and an in~a- arc basin typically forms between the two volcanic lines. A new volcanic chain commonly forms following arc rifting also. The active volcanoes of the Montana, Ton ,a and Scotia arcs have been erupted through, and built on, oceanic crust formed at their backarc spreading centers. c) What is the scale of tectonic segmentation of backarc spreading centers and how does such segmentation relate to melt production in the mantle en c] magmanc activity in the crust (Fig. 4~? All the types of edge-axis discon~anuii~es common to major ocean basins have been surveyed in backarc basins, including overlapping spreading centers, propagating rifts, transforms, triple junctions, hot spots and m~croplates. Some unique plate boundaries such as the lOOkm long Extensional Transform Zone in the Manus Basin have also been discovered. However the systematic surveying and sampling of backarc spreading centers has just begun, and the variations in crustal s~atigraphy resulting from these discontinuities is virtually unknown. Furthermore, the presence of the volcanic arc may influence the location of backarc magmatism, for spreading centers such as Valu Pa in ache Lau Basin are anomalously close to the arc. Off-axis volcanism forming seamounts and laccoliths also occurs but its distribution is poorly known. 2) What controls the temporal and spatial variation of magma composition? a) What are the relative roles of manacle, subducted sediments and oceanic crust in the sources of magma? To constrain this problem requires better data on the geological input to subduction zones and the subsequent addition of subducted materials. to the fore are. b) How are these source compositions modified on their ascent to the surface? - by processes such as partial melting, crustal assimilation, magma mixing, and crystal tracuonanon. c) What are the relationships between the tectonically controlled distnbui~on of magmausm and the geochemistry of the magmas? For example, to fonn boninites requires large degrees of partial melting of a refractory source under hydrous conditions, and to avoid amphibole crystallization the wet magmas must be brought to shallow depth while at high temperatures. Is boninite genesis lifted to the unique conditions prevailing during initial suWuci~on? Alternatively, are bononites, alkali basalts and/or rhyo-dacites characteristic of periods of arc Fifing? How does the geochemistry of backarc basin crust vary during the opening of the basin? Structure and Stratigraphy I) What role do the processes of tectonic accretion, underplating andlor erosion play in the evolution of the forearc? a) What controls which of these processes will occur and when?: the material properties/thickness of sediments, the geomorphology and the velocity/obliquity of the plate 161

being subducted; the slopes, Euchres and fluid pressures of the upper plate? Is subduction erosion the result of extensional/gravitai~onal collapse of the upper plate and/or interaction with subducting horsts/graben or larger features such as aseis~c ridges? by What conditions, together with subduction obliquity, result in block rotations anion s~ke-slip monon in the arc-forearc? Paleomagnetic data indicate rotations of more than 60° and large latitudinal shifts in rocks of some fore arc s. How Ease motions occur and their relationship to the overall structural evolution of arc-forearcs are not understood. 2) What roles do serpeni~nite diapinsm and block faulting play in foreaTc tectonism? a) Diapirs resulting from subduction-related dewatenng causing serpentinisation of ul~amafic rocks may produce a large region of foreare remobilisa~aon. Are serpen~ite diapers present, locally beneath sedimentary cover, in most forearcs? by What condors the spatial and temporal dis~bui~on of serpentinite diapir~sm and block faulting? In the Bonins, domes of serpenunised/chioni~sed mafics/ultramafics form a linear chain along a lower-slope terrace only 5-10 km above the subducted slab. In the M~anas, serpeni~nite diapers form a semi- random pattern on and landward of the trench- sIope break, 15-30km above the subducted plate. Large collapse graben commonly occur nearby. Are the differences between the Mmanas and the Bonins mainly due to He greater fracturing of the Mariana forearc due to backarc spreading and the disrupnon of subducting seamounts? Do the collapse features also result from this and/or are they caused by withdrawn of material into the diapers? 3) What is the uplift/subsidence history of the moron? a) Using backslapping techniques on cored/logged driliholes and seismic straugraphic analysis of interconnecting seis~c-profiles, we need to determine, for example, whether the frontal arc and outer-arc high develop by igneous intrusion or differential uplift, whether the upper-sIope basin between them is due to forearc spreading or differential subsidence, and whether flexural loading by either arc volcanoes or by coupling ~~ He subducting plate is an important process. Furthermore, how do the processes in I) and 2) above affect the vertical history of the forearc? b) What features of the upper and/or subducting plate condom the along- stoke variation in this history? c) What are the effects of collision events? d) What are the effects of arc Offing? e) Why do some backarc basins have positive (e.g. Manana, Parece Vela) and others negative (e.g. Lau, Manus) depth anomalies? 4) How does arc lithosphere rift? a) What initiates, and what stops, backarc basin opening? - some combination of kinematic boundary conditions (convergence rate/obliquity, slab age-and dip, absolute morons) together with geodynamic forces (due to aesthenospher~c convection and trench suction)? b) Is the rifting process one of simple shear (detachment), pure shear (stretching and thinning), or both? c) What effect does the presence of the line of active volcanoes have on the processes? Are there significant differences to continental Offing? What controls whether the rifts split the arc, or occur on the forearc angler backarc side? 162

5) How do the stress and strain fields vary across and along the margin? As a function of the varying physical properties, strain rate, pore pressure etc. how does the deformation style change temporally and spatially? e.g. from normal faulting vs. thrusting of the subducting plate, to tectonic accretion vs. erosion of the inner wall, to . ... . . . . ~ . ~ . . upllit/diapinsm vs. subsidence/block faulting at the outer-arc high, to upper-slope extension vs. forearc backthrusting, to serike-slip faulting vs. rifting of the arc, lo backarc thrusting vs. spreading. What are the forces which vary the stress/strain field such that compression in one pare of the margin may occur at the same time as extension in another? 6) What is the crustal s~angraphy and structure of the backarc-a~c-forearc (Fig. 3~? a) Our knowledge of the basement beneath intraoceanic convergent margins is almost totally limited to the extrusive carapace. We have almost no detailed crustal stratigraphy or seismic sections to compare with oceanic crust or with ophiolites (many of which, geoche~ca] studies suggest, formed In near-arc settings). b) Am the initial arc-forearcs built on oceanic crust or on older te~Tanes? c) Are all intra-arc basins the product of arc reversal forming a second volcanic chain or were some formed by in~a-arc spreading? 7) What is the sedimentary response (both deposition and erosion) to all the tectonic and magmaiic processes? Fluids and Thermal-Regime I) What is the distr~bunon of pressure and temperature across intraoceanic margins? Magmatic and hydrothermal heat sources generate metamorphism and metasomatism; the metamorphic facies distribution is determined by pressure- temperature conditions. However He thermodynamics of intraocean~c convergent margins is almost totally unknown. All quanuta~ave models of the forearc temperature field relate to the effect of the subducting slab and do not consider the influence of the the volcanic arc, the age of and fluid flow through the fore arcs and the thermal blanketting effect of sediments. No detailed. Quantitative thermodynamic models of the arc-backarc have been attempted. Displaced or non-eguilibrium metamorphic facies (e.ge blueschist) may reveal mass movements of rock or changes In the PIT regime of the margin if a reference facies distribution were known. _ ~ —1 — 2) What is the nature of the hydrothermal systems? Magmas beneath arc volcanoes and backarc spreading centers, and probably upwelling serpentinite diapirs in the forearc, maintain steep geothermal gradients in the surrounding crust which drive vigorous forced convection. Low temperature, free convective circulation occurs through the backarc and outer fore arc crust, where not blanketed by voicanicIastic aprons. The flow regimes and chemical flux of all these hydrothermal systems is just beginning to be studied. To define the temperature, pressure, and flow rates of these systems, it will be critical to determine He in situ physical properties of the host materials; their geometry and structural associations, the chemistry of intershtia} fluids and gasses; the petrology and alteration state of host rocks; and the tectonic and regional associations of the hydrological system. 163

3) What processes control the formation of ore deposits? A large proportion of the worlds economic metal deposits were formed at convergent margins. These include Kuroko-type (volcano hosted) and Besshi- type (sediment hosted) massive sulphide:deposits, porphyry copper and epithermal gold deposits, as well as sulphide, nickel and other deposits associated with ophiolites. All 1987 ALVIN dive programs in the Mariana region found sulphide deposits associated with active hydrothermal systems. These include cold seeps from the fore arc diapirs, high and low temperature venting from the Dack-arc spreading center, low temperature venting off-axis, and low temperature venting on the volcanoes of the northern volcanic cross chain. ALVIN dives in the Bonin rifts found silicate and Mn-hydroxide chimneys on a rhyolite volcano surrouded by basalts in a setting directly comparable to the mid-Miocene Kuroko deposits. Hydrothermal venting and associated metalliferous deposits have also been discovered in the Lau, N. Fiji, Okinawa, and Woodlark Basins, as well as on several submarine arc calderas. Understanding the hydrothermal systems and host rock interactions which precipitated these deposits will directly improve our understanding of ore formation processes in an environment more akin than mid-ocean ridges to that of land deposits. Methods for Studying These Margins The full suite of geological and geophysical techniques must be employed in a multidisciplinary approach if the questions discussed above are to be answered. 1) Seafloor swathmapping using multibeam bathymetry and side scan tools will be necessary to provide the base maps for detailed sampling and structural interpretation. 2) Deep surveying (towed or untethered) using photography, video, sidescan; magnetics, gravity, seismics; and geochemical and physical oceanographic detectors, will be needed for higher resolution. 3) Sampling using navigated dredges, cores, ROY's, submersibles, and ocean drilling will be necessary to provide data on the age, physical properties, and chemical composition of sediments, rocks, and fluids. 4) Crustal irna~in~ using magnetic, gravity and E-M, but particularly seismic, methods will be _ _ structure, high resolution digital single channel is excellent, but for processes occurring beneath thick sedimentary sections or in basement, MCS, township ESP and OBS techniques are required. 5) Bottom instrumentation and moored arrays will be important for navigation, sampling, and long-term observation/experiments (e.g. flow meter, tiltmeter, seismometer). 6) Island mapping & drilling: intensive outcrop studies of structure, petrology, paleo- magnetics, paleodepth and environment are needed, but the greatest opportunity is for land- based drilling platforms to Will deep (several km) holes into the crust. , ~ needed to see the third dimension. For shallow Selected References Arculus, R.J., 1985, Arc magmatism—an unresolved problem of sources, material fluxes, tectonic evolution and thermochemical regions of subduction zones, in Nasu, N. et al., eds., Formation of Active Ocean Margins: Tokyo, Terra Scientific Publishing Co., p. 367-397. Hawkins, J.W., Bloomer, S.H., Evans, C.A., and Melchior, J.T., 1984, Evolution of intra- oceanic arc-trench systems: Tectonophysics, v. 102, p. 175-205. 164

Hussong, D.M. and Fryer, P., 1985, Fore-arc tectonics in the northern Manana arc, in Nasu, N. et al., eds., Formation of Active Ocean Margins: Tokyo, Terra Scientific Publishing Co., p. 273-290. Hussong, D.M* and Uyeda, S., 1981, Tectonic processes and the history of the Mariana Arc- a synthesis of the results of Deep Sea Drilling Project Leg 60, in Hussong' D.M., Uyeda, S., et al., Initial Reports of the Dep Sea Dnlling Project, v. 60, p. 909-929. Taylo, B. and Karner, G.~., 1983, On the evolution of marginal basins: Reviews of Geophysics and Space Physics, v. 2l, p. 1727-1741. 365

Or In. a. b. 'A ~ R ~ A f - Fig. 1. Schematic representation of cxicnsion and sha;tening of li~bosphcrc at conver~cnt plate margins. R—ren~nant island arc, B—back-arc basin, ~ ~ active volea~uc island arc, F ~ [ore-arc. Vertical din~en- sions not to scale. a. Arrows show relative convergence bc~ween points behind remnant arc and on scoward side of trench. Relative motion of the subducted plate is shown by arrows. b. Extension of lithosphere in back-arc and tore-arc rc~,ions is required by I'cncralion of resew crust, and by removal of fore-arc and trench slope material. c. Evidence [or excision includes normal faults in back-arc and fore-arc, new crust in back-erc basin exposure of arc material on trench slope. Binge line of flexed plate "rolls back" toward ocean basin accommodating extension of lithosphere in the ale system (Elsasscr, 1971; Kanan~ori, 1917). (from Hawkins et al. ~ 1984) 166

~ ~ ~ ~ --- ~ 1. ~L:~ l\~ 1''~: -'at I ~> W~ W~2~-~-= = - - ,-——~~q~D ~ . t! => ~ : ~ ' ~ ~ ~ ('~W _ i ~ ~ ~ ~~ ~ ~ 0 ~ ~ 4~ | |_ J ~ ·'— ; ~ · · ~ t~ _ _ ~ ~ ~ I o ~, O ~31 ' ~ ~ ~ ~ ~~ ~ ; ~~~ >; O me, Aim ;t,~eJ Oc`< I ~ ~ ~: ~.~: ~ ---- -~3 ~ ~ _~ 7. /~>Jj] , ~ . } ~ ' ! ~: Figure 2 (f ram Taylor and Karner, 1983 ) 167

Jo KM Figure 3 to (from Hawkins et al., 1984) AL !~ O Back-Arc Crust ~3 Sed~rner't ~3 Arc Crust 1 =Magma 0 Ate Em] SerpenI'ruto ~ Oceans Crust =^rC Crust 20Manl~ fit . C:ross-secti~~n through arc-back-arc basin system such as the Lau-Tonga and tv1ariana arc region. A sprawling h;~ck-arc basin separates a remn.`nt arc from an active volcanic arc which has been superposed on part of the back-arc crust. Tectonic erosion has removed fore-arc m;~tcri;`l exposing Ants of alder arc c`,mp<,nents on the trench slope. Dilation of tore-arc formed sediment trap [or arc-`lerived cl;~stic rocks. Periods of plate-coupling have forced up serpentinite diapirs ore the fore-arc and have caused accretion of scomount [ragmcnts to the trench slope. Rising partial melts of the m;~ntic thici;cn the arc-root and also generate new back-arc basin crust. Bock Arc Foreorc ~ - ~ ARC Ocron~c Lii~osphet, _ -\ f ~ .. 7/ / it / / Iv// Figure 4 (f ram Taylor and Karner, 1983) / If/// ~ Lo~'rorO' 168 - Hypothetica1 models of back arc basin formation. (a) Active diapirism resul Ling f rom heat and/or water generated along the Benioff zone lKerig, 1971. 1974J. (b) Convective flow induces in the mantle by the subduct ing 1 ithosphere [sleep and Tokso~, 197 I; Andrewe and Steen' 1974; Toksoz and Bird, 197 7; Toksoz and Heui, 19781. (c) Byproduct of ma jor plate interactions on ~ global scale 1e.8., Chase, 1978a; Dewey, 1980~. OF denotes overriding plate ant RA? denotes remnant arc plate.

COLLISION OF SEAMOUNIS, RIDGES, AND CONTINENTAL FRAGMENTS: THEIR EFFECIS ON CONVERGENT MARGINS Roland van Huene The consequences of subducting positive features such as seamounts and ridges have been a fascinating topic of speculation for geoscientists, perhaps because of the catastrophic image it inspires. That image is enhanced by the vertical exaggeration introduced into bathymetric records by limitations of echo-sounding paper recorders, and from the exaggeration used for emphasis in physiographic diagrams of the oceans. Vertical exaggeration makes it appear impossible for the slope to accommodate colliding seamounts, and lends support to geologic interpretations Hat seamounts are sheared from the subducting ocean crust and were accreted in ancient subduction zones now exposed on land. However, surveys of He intersection of edges with trenches using conventional echo-sounding profiles showed little chaotic local topography (Nazca Ridge in the Peru Trench, for instance). The discovery of a seamount that was "crashing" into the "trench wall" occurred in the late 70's when Magi proposed the subduction and normal faulting of Kashima Seamount (Figure 1) along the Japan Trench (Mogi and Nishizawa, 1980~. Subsequent SEABEAM maps of Kashima Seamount (Oshima et al., 1985; Kobayashi et al., 1987) defined a normal fault that splits the seamount in half as well as numerous small faults, and a large bulge on the adjacent lend ward slope of the trench caused by the subducting seamount. The reprocessing of a seisz~uc-reflection record! provided an image of the leading flank of Kashima Seamount in the subduction zone and the thrust-faulted strata above it (Lallemand et al., 1988 in press). SEABEAM surveys of the New Hebrides and Middle America trenches also revealed subducted seamounts (Daniel et al., 1986), and in the Peru and Tonga trenches, SEABEAM provided detailed morphology at subducting ridges (Bourgois et al., 1988). From such data it was clear that the morphology of subducting positive features, and the subsurface image of the causative feature, could be resolved with modern geophysical instrument Systems. 169

The term "collision" is used when a positive topographic feature Impinges against the landward slope of a trench. This figurative term is not sharply definecl in a geological sense. Implied is the displacement of material as one body intrudes the space of another. If material ndes over a subducting topographic feature and is deformed but stays essentially in tact, over Repave teens might be used. There is a continuous progr~mon from an encounter ~~ little disturbance to one which leaves a lame tectonic scar or suture. The effects of collision depend on the scale of me anpinging feature. Horst and graben of He seaward slope that riffle or jostle the front of a sub~iucHon zone leave few features which are resolved by the SEABEAM instrument. As graben are subducted, they uncloub~y cause some tectonic erosion if Heir relief has not been smoothed by sediment ponded In the trench axis. Subducting seamounts and ridges produce effects which are resolved by most swath- mapping techniques, and He subsurface structure is Raged by Rodeo seismic instants This involves a meso-scale~ relief and is commonly erosive since the relief of seamounts is not obliterated by smIimentabor~ At the macro scale is the collision of continental and island-arc fragments which generally results in terrain accretion. The collision of the Yakutat terrane with North America is assomateci win some of the highest mountains along the cent Gulf of Alaska. Such collision commonly involves large scale strik~slip movement of crustal slices parallel to the margin. SOME TECTONIC PROCESSES ACCOMPANYING COLLISION Examining the consequences of collision along convergent margins Is somewhat analogous to testing the physical state of materials in He laboratory. Just as He response to punching a sample unto a probe reveals its hardiness, so He damaged zone at a collision reveals the strength of He crust that forms a margin and indicates some of the tectonic processes that Operated (see for instance Dubois et al., 1988). SEABEAM surreys during He French-Japanese KAIKO project revealed seamoun~ in the Japan Trench at venous stages of subduction. Using observations from Rose seamounts, Lallemand and Le PYchon (1987) developer! a model of 170

seamount subduction. The flaw of Kashima Seamount dip about 15° and wedge into the seduction zone rather than tear up the front of the slope. A more general moclel, developed from that of Lallernancl and Le Pichon., illustrates some of the questions facing researchers (Figure 2). As me leading flank of a seamount wedges into a subduction zone, the lmndward slope of the trench ndes up the flank (Figure 2A). The added horizontal compression from uphill travel on the seamount flank thickens the slope tectonically by tiamst faulting, and a thickened welt forms ahead of the subducted seamount. Since most landward slopes of trenches dip at a cnidcal angle' the oversteepening of the slope causes failure and mass wasting (Figure 2B). Once the crest of tile seamount has bed subducted and the tong flank begins to move beneath Me landward slope, the thickened welt is on a descending rather than ascending ramp (Figure 2C). The slope oversteepens and produces slides that travel down the flank to the trench Failure is facilitated by the weakling of these rocks dunng the preceding uplift and deformation about the seamount's leading flare The resulting debris avalanches and slumps pond In the trench and are swept down the subduction zone because they do not seem to build at the base of the slope and stabilize it. As the trailing flank of the seamount is subducted, sufficient material is removed from the lanclward slope to leave an indentation or scar. Plate convergence- compresses secInnent and debns on the floor of We inclentabon which awn builds an imbricate wedge (Figure 2D). Such accretion rapidly fills the ~ndentabon. The scar where the seamount was suWuctecI is healed and it-contains a hiatus between a young accretiona~y complex and~an older failed slope, as is observed along convergent margins, where a accre~donary complex is stacked against a "'backstop." This example of a mesmscale collision shows how a scar is former! in a convergent margin without "chain-saw" abrasion, the forcible shouldering asicle, or long distance displacement of large volumes of matenal. If me slope of the impinging feature were steep and the roclcs of the landwarcl slope were strong, a gash might be carved or forcibly abraded into the slope. Once the leading edge of the impinging feature is buried in the margin, it probably 171

takes on a shape of least resistance just as a flat-nosed projectile entering the earth builds ~ cone of least resistance during the first part of its subsurface trajectory. The same processes might occur on a muc h smaller scale as the scarps of horst and gTaben are subtluctecl, because along some margins the scarps leave no indentation once they meet the landward slope of the trench. This temptation of the horst and graben morphology indicates that the front of the slope rides over morphology of this scale rather than being significantly disrupted by it. When continental fragments and continents collicle, the pushe~l-up and disturbed material is unable to escape from the collision zone and it forms thrust belts. In the Gulf of Alaska, the Yakutat terrane is such a fragment with an eastern Continental pant and a western oceanic part. Where the continental part collides, the mountain ranges have developed 19,000 ft-high peaks, and metamorphic rocks that formed 10 and 15 km below the surface are expo - . The ocean margin part, on the other hand, is subducting beneath western Alaska without pushing up high mountains. However, adjacent mountain ranges are found along the non- collision parts of We continental margin, and so the effects of collision alone are not obvious. The effect of oblique subduction causes larg~scale lateral transport of continental margin or island-arc fragments. For instance, where the Indian Ocean is subducting obliquely under Southeast Asia at the Suda subduction zone, the entire Sumatra forearc region is being translated northward along the Sumantran Fault zone (Moore et al., 1980~. Fragments of island-arc terrains and ophiolites have been transport northward along strike~lip faults in the Philippines as well (ICarig et al., 1986~. Such oblique transport of "allochthonous terrains" such as forearcs, volcanic arcs, and ophiolites may have been very important in the development of the North American Cordillera. Oblique transport may also be an important mechanism in "tectonic erosion" of forearcs. 172

QUESTIONS Collisions provide the abnormal or non-steady state conditions which reveal charactenstics not otherwise obvious. Many of the questions to be answered by the study of collisions can be grouped under two In headings: A. What tectonic processes shape the convergent continental margins? 1. Do the subducting meso-scale topographic features form the asperities of the seismologist that produce centers of earthquake generation? to 2. How much material is eroded by the subduction of micro-scale features such as the horst and graben commonly found on the seaward slope of a trend h? -What are the conditions under which such topography is erosional, and are modern non-accretionary margins shaped by such a mechanism? 3. What is the critical size of a subducting feature that causes displacement (lumping) of a subduction zone? 4. Under what conditions does oblique transport of allochthonous terrains take place? B. What is the physical state of the upper 15 km of the continental crust? 1. Are subducted topographic features the cause of changes in volcanism as well as changes in the configuration of the Benioff Zone? 2. Can the-subduction of topographic features cause sufficient slope instability to produce the huge slides which generate destructive tsunamis? 3. Under what conditions are seamounts sheared from the subclucHng plate? 4. What differences in the physical state of convergent margins causes subducting edges to disrupt large parts of some whereas others show little effect? 173

REFERENCES Bourgois, J., Pautot, G., Bandy, W., Boinet, T., Chotin, P., Huchon, P., Mercier de Lepinay, B., Mange, F., Monlau, J., Pelletier, B., Sosson, M., and van Huene, R., 1988, SEABEAM and seismic-reflection imaging of the tectonic regime of the Andean continental margin off Peru (4°S to 10°S): Earth and Planetary Science Letters, v. 87, p. 111-126. Daniel, J., Collot, J Y., Monzier, M., Pelletier, B., Butscher, Je' Dupllls, C., Dubois, J., Gerard, M., Maillet, P., Monjaret, M.C., Recy, J., Renard, V., Rigolot, P., and Temakon, S.J., 1986, Subduction et collisions le long de l'arc des Nouvelles Hebrides (Vanuatu): resultats preliminaires de la campagne SEAPSO (Let 1), C.R. Acad. Sci. Paris, 303, S6r. II: 805-810. Dubois, ]., Deplus, C., Diament, M., Daniel, ]., and Collot, l.-Y.' 1988, Subduction of the Bougainville seamount (Vanuatu): Mechanical and geodynamic implications: Tectonophysics, v. 149, p. 111-119. Kang, D.E., Sarewitz, D.R., and Haeck, G.D., 1986, Role of strik~slip faulting in the evolution of allochthonous terrains in the Philippines: Geology, v. 10, p. 817-896. Kobayashi, K.' Cadet, J.P., Aubouin, J., Boulegue, J., Dubois, J. et al., 1987, Normal faulting of the Daiichi-Kashima Seamount in the Japan Trench revealed by the Kailco I Cruise, Leg 3: Earth ancl Planetary Science Letters, v. 83, p. 257-266. Lallemancl, S., Culotta, R., and von Huene, R, 1988 in press, Subduchon of the Daiichi- Kashima Seamount in tbe Japan Trench: Tectonophysics. Mogi, A., and Nishizawa, K., 1980, Breakdown of a seamount on ~e slope of the lapan Trench: Proc. lapan Acad., v. 56, p. 257-259. Moore, G.F., Curray, ].R., Moore, D.G., and Karig, DE., 1980, Variations in geologic structure along the Sunda forearc, Northeastern Indian Ocean: American Geophysical Union Monograph 23, p. 14~160. 174

Oshirna, S., Ogimo' T., Katsura, K., Ikada, M., Uchida, M. et al.' 1985, Subduction of the Daiichi-Kashima Seamount into the landward slope of the Japan Trench: Rep. ~Iydrogn. Res., v. 20, p. 25~6. 175

FIGURE CAPIION Figure 1. Perspective diagram of Daiichi-Kashima Seamount (foreground) and Katori Seamount (background) subducting in the Japan Trench These SEABEAM data are generally displayed at vertical exaggeration of 5 (see Kobayashi et al., 1987) but are shown here at 1.5 to illustrate that accommodation of the seamount does not appear impossible. Daiichi-Kashima (-3500 m high) is partially subducted whereas Katori is just entering the trench axis. Note the disrupted drainages on the slope and the uplift as the Seamount wedges into the subduction zone. Figure 2. Diagram of four stages during the subduction of a Kashima-like Seamount (A) Seamount has entered trench axis and begins to break along normal faults that develop as the ocean crust is fixed downward into the subduction zone. (B) Leading flank of Seamount is wedged under me landward slope of the trench causing uplifted bulge in the overlying plate. The slope is oversteepened and slumps to retain a critical angle. (C) Trailing flank of Seamount is subducted and uplifted bulge of upper plate fails sending debris avalanches and blocks into the trench axis. A debris cone builds into the trench axis and its height is moderated by the efficiency of sediment subcluction. (D) The scar left by subduction of the Seamount is healed by accretion of the debris cone and trench sediment that form an accretionary ridge behind which slope sediment ponds. c 176

- -. ~ =— : ~7 ,=_ _ , .. . . \ ~ ~ .~ O ~0 \ o \ C \ / I_~ _. O \C G // O / ;_ To / / - - LL

`;~- Hi `co' B Figure 2 Ha. act 6 _--~ a - `.— - i78 . ,

Figure 2 :~ Hi,.. ~ . D ~~ . , ~— 6~5

Active Conti~t=1 M—ins MASS AND C~CAL ~RP115E~ 180

Mass Flux and Crus tal Evolution at Convergent Margins R . W . Kay and S ~ Mahlburg Kay For Active Margi n Processes National Academy of Sciences January 1989 Introduction: The Present Crust-Mantle System Over earth history, convergent plate margin processes have created or modified most of the chemical heterogeneities in the crust-mantle system. A simplified depiction (box model) of the present-day system (Figure 1) illustrates that the formation of relatively durable continental crust accompanies subduction of relatively ephemeral oceanic crust at convergent plate margins. Fundamental to understanding the significance of subduction for both crust and mantle is the use of compositional tracers to calculate mass flux in the system. In principal, the goal is to apply the kinetic treatment of geochemical cycles (e.g., Lasaga, 1980) to the crust-mantle problem. The applicability of this method to crustal growth is apparent from some preliminary observations. It is easy to see that the composition of subducted oceanic crus is not exactly that of the fresh mid- oceanic ridge basalt: elements from the continental crust and the hydrosphere are added to the oceanic crustal column during its short (<200my) residence time on the sea floor, and at least some of these elements are subducted. It is equally easy to see that not all the mass that is subducted is returned to the mantle: at increasing depths tectonic off-scraping, fluid transfer and melt transfer return some fraction of the subducted mass to the crust. These various return mechanisms are grouped under the term crustal recycling. Two important correlaries to the concept of crustal recycling are that material subducted at the trench is compositionally dissimilar to unrecycled material returned to the 181

o to = e o o - = : ~ / ......... ~ . · · . ~ . I 1 nag ~ ]- 182 # 1 # T , i ;1,) . _ o o

mantle, and that net crustal growth can only be calculated when recycled crust is subtracted from gross crustal addition. The diversity of plate margin processes in space and time is ignored in the above treatment. The entire question of crustal growth, for instance, is reduced to calculation of the long timescale crust (c) -mantle (I) flux difference (FmC ~ FCm). But the box model can also be taken to represent a specific convergent margin, in which case we are drawn to the necessity of arriving at an input - output inventory or element-by-element accounting of mass at that margin. There are benefits from looking at indivdual arcs with a this systems-analysis type of approach. Past experience has shown that identification of discrepancies in input - output budgets, and of inconsistencies in perturbation response times are some of the best ways to discover unanticipated processes: an example is the failure of budgets for Mg in sea water to balance until the hydrothermal uptake of Mg by the oceanic crust was discovered. It may seem that emphasis on mass flux detracts from a physical process oriented discussion. But focussing on the control of physical processes on mass flux within the system has many benefits. In any model, the rate constants (or functions) of the processes of mass transfer in the system -- from tectonic off- scraping to magma migration to convection in mantle over the subducting plate -- must be physically realistic. But, some of the more significant systems-oriented questions center around definition of boundary conditions in space and time for physical models. Mass tracers used in an interactive way are essential to substantiate choices of alternative physical processes and to define rates. A third, and perhaps more valid point is that the boxes in the system are inadequately subdivided. An analogous problem was solved by Berner (1987) in his models for atmospheric oxygen by subdividing crustal carbon and sulphur 183

reservoirs into rapidly and slowly recycling subreservoirs. Sundquist (1985) illustrates, for models of the carbon cycle, how choice of reservoirs depends upon the time scale considered. In the present case, it is quite obvious that the upper and lower crust, for instance, are affected by quite different mass transfer processes (e.g. erosion vs partial melting) at quite different time scales. As a result, it is unlikely that crustal compositional heterogeneity formed at convergent plate margins is preserved intact in the terrane collage that constitutes mature continental crust. As shown below, mass budgets point toward delamination of mafic lower crust and its sinking into the mantle as a reasonable process to operate when crustal terranes collide -- which represents the last event in the history of many crustal terranes created at convergent plate margins. II. Mass Flux at Convergent Margins a) Global Overview Igneous crust that forms the top of oceanic lithospheric plates is very efficiently returned to the asthenospheric mantle at subduction zones. This conclusion follows directly from the exponentially decreasing total - mass of preserved oceanic crust (including ophiolites) with its age, and from the low ratio of the total mass production rate of arc crust to the total mass subduction rate of oceanic crust. Referring to igneous rocks only, we define the ratio of returned to subducted mass--either for total mass or for individual elements--as the efficiency ratio: ER. On a global basis, by mass, the ER is about 1/20 (the ratio of magma production rates). For many major elements (O. Mg, Fe, Ca) the ER that we calculate is almost certainly higher than the actual mass transfer efficiency from subducted oceanic crust to the magmatic arc: it is widely held, with ample experimental and observational evidence (e.g. Crawford, et al 1987), 184

that the source of arc magmas is the mantle wedge over the subducting plate (see Figure 2), and that this mantle wedge furnishes most of the major elements in arc magmas. Appealing as it is, simple melting of the subducted oceanic crust is not an adequate source for arc magmas. However, sedimentary part of the subducted oceanic crust is an important source for many trace elements--especially those that are enriched in arc basalts compared to altered Mid-Oceanic Ridge Basalt (MORB). For these elements the ER can be considerably higher than for the major elements as can be seen from the simple relationship: ARC OC r Mass Mass Ci ~ ERi time time Cite ; Cite concentration of element i ARC OC For instance, for the element Rb, the ratio C b /C is about 15, and ERRb is nearly 1. The class of elements that, like Rb, are highly enriched in arc basalts relative to unaltered MORB have been called "excess" elements, and it must be emphasized that while there are significant differences in the "excess" element content of magmas from various arcs, in all cases the subduction-related magmas have higher concentrations than do unaltered MORB. A present day global input- output is summarized in Table 1 for 4 "excess" elements. For some elements (e.g. H2O, K) the hydrothermal uptake by oceanic crust is sufficient to raise MORB concentrations to levels sufficient to account for the "excess" elements. For some other "excessn elements, in particular those not soluble in ocean water and therefore not enriched in hydrothermally altered MORB, the flux from subducted MORB alone is insufficient to account for the "excess" elements. Two cases are particularly obvious: Th and 10Be. These two elements are extremely insoluble in sea water, and in the case of 10Be (1.S m.y. half life cosmic ray-produced 185

TRENCH 10- 20- 30 - , 50- CENTRAL ALEUTIAN I SLAND ARC VOLCA N I C ARC fin a< . I= ~~ _~ ?~7 w~ V.E. 1:1 1 ~ ~ =- _ Crust sediments @~ upper > arc-derived lower ~3 basalt >oceanic ICY sediment Mantle cumulole ultramof~c \ E] deformed peridotiti\ with interstitial basalt 2: it at;'/;' ~Oo°C .' , Wo 341~ _ -10 20 -30 -50 100 Figure 2 - Cross-section (no vertical exaggeration) showing crustal formation in the central Aleutian arc (modified after Kay and Kay, 1989 ~ . Configuration of subducted plate, crustal thickness and forearc li tholog,ies are cc: ns trained by geophysical and seismological data. Vola~ciles and melt released front the subducted plate cause melting in peridotite of the overlying wedge of mantle, malting it buoyant . Con~recti~re ri se of the buoyant peridotite causes further melting (by decompression) accompanied by percolation of the light melt through the denser peridotite. Segregate on of mafic minerals from ponded oliv~ne tholeiite forms of cu~n~nulate ultramafic rocks that occur at the crust-mantle boundary. A basic lower crust consists of unassimilated residues of crystalline oceanic crust and intrusive hi-A1 basalt and its mafic crystalline fractionates. Intermediate composition upper crust consists of igneous intrusions and extrusions of magic to intermediate composition. The four lettered regions are places where subducted crustal material is added to the mantle ~ see text ~ . 186

TABLE 1A INPUT Hydrothermal Flux: Hid Ocean Rld~,e Tota1 ~ . ~ % Hydrother~1 Fixed Concentration In~rease Ctrculation ~ 6 ~ crus ~ ) H2Q 4.4 X lol8 g~ 0 .021 1.51 14001 yr (~/~- 100)} ~+ L. 8 x 1015 — 3%2 0.11 200' yr Rb~ 0.5 x 10 Yt 20, 1. 5ppm 400% C 1 x 1014 g~ 209e 0.03' 100% Yt tWater to Rock Ratio 2Sufficient to balance Ri~rer ~Ja~er Flux TABLE 1B OUTPUT Volcanic Arc Me~a Flux tC}obe} Productlon Rate at Voleenic Ares: 30 x 1014 - /yr) - Efficleney Rat~o: ~Excesc. ~ of al~ered of aleacnt MORB required ln arc to fu~lsh exces. element basalts laventory H2O - 21 - 8t K 11 65\ Bb 20 ppm ~ 100% C 1t >> 100' Othcr ~Excess~clea~nts in arc besalts are not in sea water, and are not ·dded to oceenic crust during, hydrother~1 clteration: Th, Pb, 1OB. TABLE 2 Sourees of ~Excess ~ Element~ in Volcanic Arc l5~-lts Top Scd~nt - - Carbonate X C H2O K U Th Pb 1OBe C1~y -- X X X X X Altered Oceanic Crust (low T) X X X X Alterct Occanic Crus t (hlSh T) Arc besalts with Excess C, U, Th, Pb, 1OB- require mixing of all sed"entasy coaponente. DATA FROM VARIOUS SOU RCES 187

nuclide, formed in the atmosphere), any 10Be originally present in the igneous part of the oceanic crust would have decayed prior to subduction. A more systematic catalog of the sources of "excess" elements is attempted in Table 2. For those who would regard such a table as a sufficient explanation for the origin of "excess" elements by mixing of these compositionally extreme components, there are some disquieting observations. For instance, the concentrations of excess elements like U and Th are extremely different in potential sources (various sediment types, altered oceanic crust, etc.), yet it is remarkable that ratios of Th to U arc volcanic rocks don't show more variability (e.g. Kay and Kay, 1988). The homogenization of U-rich and Th-rlch regions within of a large volumes of mantle that is the source region for arc magmas seems to be required. Isotopic tracers (Pb, Sr, Nd, Hf) are important in identifying continental crustal components in arc magmas, as first emphasized by Armstrong (1971). Over the past two decades, these recycled crustal tracers have been detected in many convergent margin magmas. Many workers have also emphasized the small percentage of continental crustal components involved in the peridotite source of arc magmas. However, mass flux values, derived from mass balance of tracer input and output, as indicated in the next section, have been calculated only infrequently b) Input-Output Analysis of Subduction process (S-Process) Crustal Recycling A single example of a crustal recycling calculation will be given: Karig and Kay's (1981) model for the Mariana arc (Figure 3). The following description of the model is paraphrased from Karig and Kayos (1981) paper. More than 500 km3 of igneous oceanic crust and perhaps 50 km3 of sediment is being transported to the Mariana arc per million years. Both of these figures 188

OU 1 I'U I (|x ~ l;ilo'',' Ire of arc) 1~:~r2lsizr 1~) kIli'/~n XiO ', 5 ~ ";~; Ii.(), ll. ~ 4',, ( 1.!~; x Ilk' ~ /hIa) l'`lrwire 1(} k',~hl.t 5'i0~, ~0 i,i,; ~ () = £. ~ 4$o (:~.!J 2 x lV't g/~) 4C) mm /a ~ NL\1'ERI;\L B.\L.~NCE 1! :lS ",, of ~hc i; O ilk the miasmatic output is Doris cd from sc%dimc'~t arid li(~"O is decried from Mitered oct.~1ic C~St. tom DI no of Edit and ·0~) m of al~cred oc~:~:~ cn'`t · ill account for thc to`al ];.(~. I:;t~liT (~`r i;ilomc~rc of arcj 'k'',~,/>la i;. () (~,l,~ r ~..' `;n~?. f). ~ "~, (u 2- g1( n~ hIa) i;~0 (1~t r 4.;~ I;~), dJ.I "o ( ~*~/(m ~I; S`di~r'~nt I., k~/~1 };. (). '.., "~ (~.. g/(n~ >~) Figure 3 - (Kasig and Ksy, 1971), Flow diagram of the Hariana arc, illustrating the mass and of K2O involved during processes of plate convergence and arc magmatism. Use of realistic element ratios in possible sources res~slts i~ teh conclusion that no more than a few tens of meters of the basal pelagic sediments are in`,o1`red in magma production. S act(N) La`N, 3 2 1 o /\ / GS / \ MOF / (/ OK ~voicanics - ~/1 / - · · i>J I ,'~ `.# o ~o o / ~ ° . // `` Martinioue ~MORB ~ ~.~—1 . I . I 0 1 2 3 ' n LatN)!sm(N) Figure 4.- Ba/La versus L~/Sm for volcanic rocics from Mar~cinique (Da~ridson, 1986) and from three Aleutian volcanoes (indicated as fields): OK (Eastern arc, Okmok), GS (Western arc, Great Sitkin), and bSOF (Western Arc, Moffett). Compositions are normalized to chondritic ~ralues and the field outlines for oceanic ~rolcan~cs is from Kay (1980~. Samples from the old Mart~nique arc fall outside the Martinique field and may reflect enrichment of Ba in the man~cle source. The majori~y of Martinique volcanics are, however, indistinguishable from intra-oceanic lavas in tctms of Ba/La, in contrast to the Aleutian lavas. The large range in Ba/La ratios cannot be explained by fractional crystallization alone, s~nce both elements have similar (low) bulic distribution coefficients in silicate melts. 189

far exceed the total mass of volcanic rocks erupted over a million years, or even the total igneous mass added to the arc in this time. The convective flux of mantle overlying the down-going slab (also a possible source for arc magmas) is also large compared to the igneous output of arc. It appears that even inefficient (<101) extraction of melt from any or all of these sources can account for the total mass of igneous rock added to the crust. Thus, total mass calculations alone provide no constraints for issues like the origin of the arc magmas or for the fate of the sediment column. The contrasting compositions of input and output to the arc implies that some elements, such as potassium, are in much shorter supply than others. Even though K2O content of the outputs is higher than that of the hydrothermally altered igneous oceanic crust, that source as well as the sediment or mantle wedge above the descending plate could supply sufficient K2O to account for the K2O in Mariana arc lavas (e.g., the ER for K is much less than 1~. Examination of Ratios of elements that are thought to behave similarly during partial melting, and of radiogenic to non-radiogenic isotope ratios, are Two ways to discriminate between element sources. To be successful, the element and isotope ratios chosen should be different in the variety of element sources (e.g.' see Table 2). As mentioned above, Kay (1980), following Armstrong (1971), has developed a melting-mixing model for magma genesis in arcs that predicts the proportions of rare-earth elements, Ba, Rb, Pb and K from sediment, seawater alteration of oceanic crust, and igneous sources (residual mantle or oceanic crust). Ratios of elements like K, Ba and Pb to elements like La, Sm. (summarized in Table 3) and Sr are quite different for the various possible sources, as are the isotope ratios of Sr, Pb, and Nd. Ratios of K and Ba to La in Mariana lavas (Dixon & Batiza, 1979) indicate that K derived from sedimentary sources contributes about 190

TABLE 3 Percenta%e of Elements Derivcd tron~ Oitltr.:nt Sources in 1~1~1ti-Sousc~ .\ludel ~ _ r_ _ Rceyclcd hom Conunents Setisaacnt Sea W:~era Occ:~n C:rust (~ciogite) o~ Unulepleted .51;`nt1e Dcplet':d hl;~ntlc Isl~r~d Arc Thc~leute: Tafua 17 (l or`~) . La 87 0 0 13 Nd 37 0 0 66 Yb 5 0 0 95 Ba 99 1 0 0 Rb 99 1 0 0 ~ 36 64 - 0 ~ Sr 40 0 0 60 ( ClIUIt = ~ 5.6, Sr/ Sr = 0.?05 1 )b Sliusht,~it~: ,\0l,~ 5 1 6 (New l lebridcs) La. 1S ~ 8? 3 Nd 10 0 67 ~3 Vb 4 0 7 89 133 5' 0 48 0 Rb 43 AS 13 0 —1 ?6 5 ? 1 Sr 10 , 0 7~ 18 (,C:lI\JK = ~ 8.7, "Sr/8'Sr = 703~)b IligI~ ,Xlu',~ina B:~salts: As-3 (Zil;`s~n;~) La 3a' 0 60 6 Nd 14 0 53 33 Yb 4 0 5 91 B;~ 90 0 10 0 Rb S 1 .51 8 t3 ~ 38 40 20 2 Sr ~3 0 35 4 ? (`CIlVK = + 8.2, "Sri" Sr = .7040)b I ligla `~1UTnin] 13as;lits: t1515 (,Xlcuti;3ns) L;l 30 0 65 5 Nd 17 0 59 79 Yb 4 0 6 90 Ba 88 0 1~ 0 Rb 63 ~' 15 0 ~ 38 35 26 1 St 21 0 41 38 (.C! tUR = ~ 8.4, " Srl'. Sr = .7039)b ]1igh Alumin: llas;~lts: 1376-9-' (Philippines) ~- Le 21 0 75 4 Nd 9 0 70 2 I Yb 4 0 9 87 ll;, 82 0 1 ~ 0 Rb 75 0 2S 0 l; 26 44 29 1 Sr 17 0 S~ 31 - (4CIlllR = + 8.9,"Sr/"Sr = .7036)b ~CeIcubtions ~umc that IC ~ unaccounted tor oftcr contributions ~om depleted mantle and scdi- ment, and eclogite comes from ses watcr metasomatism of oceanic crust. i3Value of tCHUR =—3 in sediment, ~ 10 in depleted mantle snd oceanic crust (McCulloch and Wasserbur' 1918 DePeolo and Wasscrburg 1977). ~ Sr/" Sr values assun~ed to bc .709 in sculiment .702 5 in deoleted mantic and oceanic crust. Ali welues listed are calculated from the model. Measured values of 87S!/"Sr are .7034 (Jlu—3, O=on and Bat~ze 1979), .7034 (UMS, Kay et a1. 1978), .tO37 (P76 - 9 - 2, this paper). KAY: 1 980) 191

38% of the total K in those lavas. The remaining 62% is derived from seawater alteration of oceanic crust (40~) and from the K contributed by the unaltered basaltic oceanic crust, and perhaps from overlying mantle peridotite (22)--The last two sources can't be distinguished by trace element or isotopic contrasts. The proportional contributions to the total K flux can then be used to determine the required thickness of pelagic sediment and basalts involved in the Mariana magmas (Figure 3). About 50 m of sediment and 700 m of altered oceanic crust are sufficient to furnish all the required K. Although details of the Mariana calculation have been the subject of debate, we claim that the methodology is sound, and reveals the data that one must have to construct a coherent mass flux model. Also note that physical models must be able to reproduce the mass flux in the system, which includes contemporaneous arc and backarc magmatism with "excess" element signatures. c) Regional Variab~lirv Mass input and output correlate at the earth's 40,000 km of convergent plate margin; extreme values (units: mass divided by time per km of plate margin) differ by at least a factor of 10 (see Karig and Kay, 1981). Among oceanic arcs, where fractionation-independent chemical and isotopic differences in arc magmas are thought to reflect mantle composition and processes, three end-member magma types are distinctive: The low K, low La/yb tholeiitic magmas (nIsland Arc Tholeiitesn) of the western Pacific, the widespread intermediate K and La/yb tholeiitic and calc-alkaline magmas that are the common magma type of the main volcanic lines associated with many subducting zones, and the high K and La/yb shoshonites of back-arc regions. All three types have high ratios of Ba and alkali metals to La, and low ratios of Ti, Zr, Hf. Ta and Nb to La, although these well-recognized arc signatures are less pronounced in some of the more 192

alkalic magmas (e.g. shoshonites). The association of the low K and the boninite magma series with extensional arcs, (those with back-arc basins) is noteworthy. Many of the chemical contrasts between the three types are attributed to differences in their mantle per~dotite sources. In general, the currently popular source models (of the past decade) call for linear mixtures of melt- depleted peridotite' undepleted peridotite' and subducted oceanic crust (including sediment and hydrothermal alteration components). Kay (1980) calculated proportions of these components, excluding undepleted peridotite, (see Table 3) in the various arc magma types. Subsequent discussions by Morris and Hart (1986), Perfit and Kay (1986) and Ellam and Hawkesworth (1988a) illustrate the course of current debate over arc components. Most recently, the presence of OBe, a tracer of young sediment (Table 2), in arc magmas has infused new life into the debate (see Tera et al., 1986 and Monaghan et al., 1988). If sedimentary components are subducted into the mantle and mixed with peridotite to form "modified mantle" that is then a source for arc basalt, chemical and isotopic variability in arc magmas should mimic that in adjacent oceanic sediments. Armstrong (1971) pointed out that differences in Pb isotope composition of magmas of the Lesser Antilles and Japan correlate with regional differences in sediment (~chat mainly-reflect age differences Of sediment sources). Similarly, the high Ba content (B a/La ratio) in Aleutian magmas relative to those from the Lesser Antilles (see Figure 4) correlates with the abundance of Ba-rich Siliceous sediments entering the Aleutian trench, and their paucity in the Lesser Antilles trench. The probable regional variability of other components that form the modified mantle must always be kept in mind. For instances Arculus et al (1986) have pointed out that the depleted mantle component may vary from harzburgite (Marianas) to lherzolite (Aleutians), and an 193

a oceanic island basalt (OIB)-like mantle component is probably higher in 207Pb in the southern than in the northern Hemisphere (Perfit and Kay, 1986~. On a finer scale, the regional variations of Ba/La ratio in the oceanic part of the Aleutian arc (Figure 4) may reflect variability in the proportion of subducted biogenic siliceous ooze (high Ba/La ratio) along the arc. There are a number of studies that show the sediment-arc mimicking of various sediment-sensitive elements (e. g. Ba, Pb) and isotopes (e.g.Sr, Pb). III. Crustal Evolution a) Coupling of Mantle-Crust Evolution Continental crust is derived from mantle; the two are complementary mass reservoirs. As shown by many investigators, the mean age of crustal and mantle reservoirs can be calculated using Nd, Sr, and Pb isotopes (among others). For simple one-way transfer of mass from mantle to crust, one can also-use the evolution of the isotope ratios in the mantle to calculate crustal growth rates. If crust is returned to mantle for long time scales (e.g., it escapes immediate crustal recycling) then isotopic evolution of the mantle depends on both crustal growth rate and crust-mantle mass return rate. It is essential to obtain independent values for crust to mantle mass flux in order to use any isotopic data from the mantle to calculate crucial growth rate. Current estimates for crust to mantle flux and alternative functions for Nd isotopic evolution of the mantle point to crustal growth rate values well below the mean value required to grow crust over 4.5 by. A decreasing growth rate is implied--perhaps an exponentially decreasing rate, but this is not required by the data. Given the probable long-term return of crust to mantle, some types of mantle heterogeneity must be closely related to the Subduction Process (S-Process). Any inefficiency in the immediate recycling (from mantle to crust) of crustal 194

components of The oceanic lithosphere will leave crustal components in the mantle. These inefficiencies probably exist in four regions labelled in Figure 2: Three in the mantle over subducting plates, as well as in the subducting plate itself. The compositional relationship between the subducted crustal components and those returned to the mantle is not obvious, although it seems probable that intraelement fractionation that occurs would return K, Rb, Sr, and Ba to the crust in higher concentrations than those of the subducted material. In arcs with associated back-arc basins, some of the crustal elements appear to be extracted from the mantle in the back-arc basin basalts. b) Crustal Additions, Crustal Composition, and Lower Crustal Delamination In magmatic arcs, the common basalt and andesite volcanic rocks (and intrusive equivalents) that reach the shallowest arc levels do not originate directly from the mantle (e.~., they are not primary compositions). In the Aleutian arc, Kay and Kay (1985) have identified olivine tholeiite as a primary arc lava and have proposed that the early fractionated phases (olivine and clinopyroxene) have accumulated at Moho depth and represent newly formed upper mantle. Shallow-level silicic volcanic and plutonic rocks can be derived by intracrustal compositional differentiation from fractionation of a high-A1 basalt, assimilation of a low-melting fraction from the arc crust, crustal melting, or from a combination of these processes (see Crawford et al, 1987, for a recent review). The actual crustal section of a particular arc depends on the preexisting crust (oceanic and cratonic are end-members) at the arc magma tic axis, and on the structural responses of the arc. But in all cases, the composition of new crustal addition, and in particular the newly added lower crust, is basaltic. 195

This creates a problem (Kay and Kay, 1985, 1986, Ellam and Hawkesworth, 1988b), for if the bulk composition of the Earth's continental crust is andesitic, the crustal formation process is not duplicated in present-day arcs. We are left with two choices. First, the mean composition of new crust (non-recycled additions from the mantle to the continental crust) was more Andes in the past, in particular at the time around 2.5 Ga. when large amounts of crust (See Figure 5) formed' Second, basaltic continental crust, once formed, is transitory, and returns to the mantle by crustal delamination together with some mantle lithosphere at continental convergent margins or at continent-continent collision zones, where crust is unusually thick. At the base of these thick crustal sections, crust recrystallizes to a dense garnet-bearing mineral assemblage (garnet granulite, then eclogite). The likely sites of this thickening are in terrane-suturing or collisional zones (e.g. Himalayas) and collapsed active margins (e.g. Puna-Altiplano-zone of the Andes). After delamination, the crust that remains has andesite composition (Figure 6). Direct evidence for dense, garnet-bearing lower crustal sources of some crustal melts is provided by the rare-earth elements (REE). Figure 7 shows that some silicic melts have very low heavy REE concentrations and high ~a/Yb ratios.~ These REE characteristics can easily be explained by melting of gabbroic composition crustal rocks at relatively high pressure, where garnet is stable in equilibrium with silicic melt. In the Puna-Altiplano region of the Andes, these high La/yb melts develop coincident with crustal thickening (Kay et al. 1987). With lower crustal delamination, the mean composition of crust returned to the mantle is a weighted sum of crustal components that survive recycling in the S-process (presumably, relatively silicic) and mafia to ultramafic lower crust. The weighted sum may be compositionally similar, or even a little less silicic, 196

CRUSTAL RECYCLING AT CONSTANT M,~ I A_ 0 _ ~ o By, -D cry -. Allegre ( 8\ car o 06 / my. lo' 0.2 ~ 7 Allegre (82 Em(t) I CRUSTAL GROWTH Ad, i' a/ CRUSTAL ? GROWTH De Paolo f 83) Empty · , , 1 0 1 2 3 4 Figure 5 - (Kay, 198S) ~ Crucial recycling rate corresponding to a cons Cant volume crust with constant Sm and Ed concentrations . The `~d ~ t) curves of DePaolo (1983) and Allegre (1982) together with DePaolo' s (1983) equation for End (t) (Table 1) were used to derive the two recycling curves plotted. The ratio of Nd mass in the crust to that in the mantle is assumed to be 0. 6S (subscript 1) ~ fSm/Nd~m equals 0. 2S and other prancers as given by DePaolo ( 1981 ) . At some time in the past, the constant volume crust assumption is clearly violated. The effort of net crustal growth in the calculation is shown schematically by the dashed lines Of course, the `~d curves could be matched just as well using, crustal growth alone, w, th no crustal recycling. ROCK TYPES: COAST -PLUTONIC COMPLEX (BRITISH COLUMBIA 20- . 30~ 4~ KM JURASSIC-EARLY TERTIARY ARC- RELATED a Gronodiorite UPPER ~ UGobb.o CRUST Sac AND Pal FOZOIC Rif ~ - Refined (Triassic ) 3 Nor Rif ~ -Related --- o:A ~ _ = ~ | I I I ~ ~ Line of Tronsect A-A' -- on Continent-Oceon 0 50 KM Tronseet 82 OWER ~ SLICES OF OCEAN CRUST CRUST ~ MATRIX OF ACCRETIONARY PRISM MATERIAL A—~ A' \~ 300 KM Insular On)eneco Crys1~1l~ne Belt Belt Coost Plutonic Compicx Figure 6 - (Kay and Kay, 1988), Lithologic cross section of Coast Plueonic Complex, British Columbia (after Longer et al., 198S) showing granodiorite-rich upper crust and lower crust comprised of slices of oceanic crust in a matrix of accretionary prism material (greywacke). The proportion of gabbro related to Mesozoic arc magmatism is 5% compared to 67% in Aleutian crust (see Figure 2~. Thrusting indicated in figure, has juxtaposed several oceanic terranes. 197

- - I=-r 60. 30. 10. O 6. 3. v t.O 0.6 0.3 0.1 . t · 1 ~~ 4 ~~ lo ~1 CF1USTAl ME' TING: RE E REMOTE MELT G~-Fr" O D Ooci's G~~el-Ele<~] ~ C3090~ D - to As~ - il. `~ am~ ~~ ~~e ~ - Fi~s~re 7 - (Kay and Kay, 1986), The rare earth content, depicted as multiples of the concentrations in a chondritic meteorite (reedy), of silicic magmas derived by mel tiny of continental crust. The light rare earth to heavy rare earth ratio, of which the La/Yb ratio is good index, reflects source REE content, mineralogy of residual minerals and percentage of melting. A shift in restite mineralogy from two-pyroxene granulite to ~,arnet-bearing or ~ to a lesser extent) horr~blede - beari ng rock, will cause partial melts to have higher L~/YL ratios, everything else held constant. Higher water content will cause melting at lower temperatures, favoring garnet and hornblende-bearing restite mineralogies. Presence of feldspar in the restite causes the melt to have ~ negative Eu - anomaly. Several igneous rocks (granites, dacites, quartz diorite) have been segregated from garnet- free and garnet-bearing crustal sources, as shown, Data are from Arch and Barker (1976): dacite porphyry from the Raton-Clayton volcanic field, New Mexico; Arth and Hanson (1972): Sagan~ga quartz diorite, nothern Minnesota; O'Brien (1983) granite xenolith, Buell Park diatreme, Arizona; and Conrad (1983) dacite from HeDermitt Caldera, Nevada. The amphibolite (A) and two-pyroxene granulite (2PO) are xenoliths from Buell Park tiatreme that may represent protolith amphibolite, and resite granulate locality (O'Brien, }983). CENTRAL. .OREGON - - - CENTRAL AMERICA .o ~ i, ~ V ~ 4_ .& E co ~— Ci _ ~ . d~ ~ 11 - - _ ~ — - - c. ~ ~ \c ~ ~ ~ ~ f ~ O2 ~ i1 ' '' 1 AL _ , _ _ _ _ ~ Figure 8 - (Kennett et al., 1977), Comparison of histograms of all available Middle and Late Cenozoic K-Ar dates with relative volumetric estimates of basaltic-andesitic and rhyolitic ignimbrites for Oregon and Central America specific volcanic rock formations are indicated. Relative parolees of igneous rocks are estimates based on observed abundances in the field and are not on an abso lute scale . 198

than the basaltic component added to the crust from the mantle at convergent margins . Outstanding Questions The outstanding questions about mass flux at convergent margins fall inot three categories: those related to the processes of mass transfer, especially definition of the rate constants (or functions) in box models like Figure 1; those related to the history of mass flux over specified time spans (nonlinearities and singularities in the mass flux are a matter of the geological record); and those related to the relationship of subduction processed to long- term mantle heterogeneities. a) Processes of Mass Transfer Fundamental theoretical and experimental advances in rock mechanics, experimental petrology, and fluid and magma migration by porous flow, diapirism, and fracturing mechanisms are the foundation for advances in definition of the functional form of the rate constraints that describe mass transfer at convergent plate margins. The role of volatiles H2O and CO2 is particularly important, for the rates of many mass transfer processes are coupled to the rates of volatile transfer. Water reduces melting temperatures, decreases melt viscosity, and localizes deformation by enhancing deformation mechanisms. Often, models of deformation at convergent margins are dominated by the thermal structure alone, but it is increasingly apparent that tectonics may be more fundamentally controlled by chemical heterogeneities that control the rheology--e."., Hollister and Crawford's (1986) concept of tectonic surges that correlate with plutonic episodes, and Kay et al's (1986) concept of localization of deformation by chemical heterogeneity ("chemotecton~cs"~. 199

- - ~ Chemical and isotopic tracers introduced by subducting plates are valuable, really indispensable, tracers for defining mantle convection above the plates. The progressive tracer changes in main arc, back arc, and back-arc basin magmas at one time provide a primary constraint on the interaction of mass reservoirs in the subduction system. b) Mass Flux and the Geological Record The geological record reveals that the rate of mass transfer at convergent margins is nonlinear on a wide range of timescales. It may well be that "limit cycle'' or "chaotic" behavior, characteristic of turbulent dissipative system, is being exhibited by the mantle wedge. The nature of the episodicity must be gleaned from the record itself: it is unlikely that operation of the system will be deduced from first principles. Mass output is episodic, as emphasized by Figure 8 Kennett et al. (1977), and Verplanck and Duncan (1987) at times when subduction rate has been constant or smoothly varying. Definition of episodicity requires a major effort that integrates land and oceanic studies. Most of the record of volcanic episodes at convergent margins lies in sediment shed from the arc, and some mass tracers (especially isotopic, e.g., Nd and Pb isotopes) are well-suited to correlating land and oceanic records. Ideally, one must identify and characterize the entire mass transferred intracrustally by erosion in order to calculate volcanic output, and thereby provide constraints on physical processes. As an illustration of our general lack of knowledge of processes that may operate over a 5 to 50 my time span, consider the following. The subducted plate contains far more water than is released in volcanic rocks erupted in the arc. There is evidence for release of some of this water from the plate at crustal depths: water rises along the thrust zone at the top of the subducted plate 200

(e. g. Fyfe and Kerrick, 1985) . But it is possible that water released all along the thrust zone loads the hanging wall peridotite, ultimately triggering a tectonic and magmatic episode (including back-arc basin formation)? To study such a process, one must sacrifice the detail of observation available to study of, say, an active volcano, but gain understanding of what, for the earth's systems, may represent a far more important process. Related to the 5 to 50 my time scale are several other questions: i) , Does a volcanic culmination in the late Cretaceous occur in both plate margin and intraplate localities ? Is the volcanism, by release of CO2, sufficient to cause modification of climate? Are there smaller episodes of this sort (e . g. in the Eocene? ~ ii) Is episodic intracrustal differentiation by pluton formation and structural thickening coincident with formation of (or lack of) subduction- related mineral deposits? Do the higher erosion rates influence climate or increase oceanic productive ty? c) Relation of Mantle Heterogeneity to Plate Convergence Mantle heterogeneities have a profound influence on the crust: the solution deco many regional tectonic probe ems lies in the mechanical properties of the mantle. Intraplate thermal and density anomalies in the mantle Elect chemical heterogeneities. It is well known that intraplate basalts in both continental and oceanic regions have melted out of chemically abnormal mantle regions. Formation of these mantle heterogeneities may be reasonably linked to the subduction process--where inefficiency of crustal recycling leaves elements from the upper continental crust in the mantle, or continental collisional processes, where lower continental crustal rocks may be mixed into the mantle. At present, the subduction process is specially concentrated. Sixty percent of the subducted 201

plate area-descends into a region of 9% of the global area centered at New Guinea in the western Pacific (Parsons, 1980~. Compositionally, the sediment carried by descending plates is Tonally distinct, resulting in potential ly distinct and spatially limited mantle heterogeneities. A convincing compositional match between observed heterogeneities in the subducting plate and those in the oceanic mantle could put a modern twist into Barth's (1968) prescient observation that "The diversification of igneous rocks is caused by sedimentary processes." 202

REFERENCES Allegro C.J., 1982, Chemical geodynamics, Tectonophysics, v. 81, p. 109-132. Arcu] us , R.J ., Kushiro ~ I ., Gill , J . B., Aoki , K. 1986 . Crystallization sequence in arc basalts and their relationship to mantle wedge T-X structure. EOS v. 67, 405. A, Armstrong, R., 1971 , Isotopic and chemical constraint on models of Vagina genes is in volcanic arcs, Earth Planet. Sci. Letters, v. 12, p. 137-142. Arth, J G , and Barker, R ., 1976 . Rare - earth partitioning between hornblende and dac;tie liquid and implications for the genes is of trondhj emitic - tonal~t' c magmas, Geology, v. 4, p . 534- 536 . Arth ~ J . G ., and Hanson, G . N ., 1972, Quartz diorites derived by partial my ng o eclogite or amphibolite at mantle depths, Contrib. Mineral. Petrol., v. 37, p . 161 - 174. Barth, T. Fir., 1968 , The geochemi&al evolution of continental rocks, a model , in Origin and Distribution of the Elements, Le H. Ahrens (ed), Pergamon Press ~ p . 587 - 597 . Berner, R.A-., 1 987, Models for carbon and sulfer cycles and atmospheric oxygen: Application to Paleozoic geologi c history, Am . J . Sci ., v. 287 , p . 177 - 196 . Crawford, A. J ., Falloon, T ~ J e ~ and Eggins , S ., The origin of island arc high- altlm~na basalts, Contributions to Mineralogy and Petrology, 1987, v. 97, p. 417-430. Conrad, W. K., 1983, Petrology and geochemistry of igneous rocks from the McDermitt Caldera complex, Ne~rada-Oregon, and Adak Island, Alaska: Evidence for crustal development, Cornell Un',rers' ty PhD thesis, 325 p. Conrad W.K., and Kay, R.W., 1984, Ultramafic and mafic inclusions from Adak Island: crystallization history and implications for the nature of primary magmas and crus tal evolution in the Aleutian arc, J our . Pet- rolo~Y, v. 25, p. 88-125. Davidson, J . P ., 1986, Isotopic and trace element constraints on the petrogenesis of subduction-related lacunas from Martinique, Lesser Ant' lies , J . Geophys . Res ., v. 91 , p . 5943 - 5962 . DeBari , S ., Mahlburg Kay , S . and Kay, R.W., 1987, Xenolithic evidence for the nature of the crus t - -mantle boundary and primitive magmas in the Aleutian arc, J . Geol ., v. 95, p . 329- 341. DePaolo, D.J ., 1981 , Reply to comment on nCrustal growth and mantle evolution: inferences from models of element transport and Nd and Sr isotopes". Geochim . Cosmochim . Ac ta, ~ . 4S, p . 12S1 . 203

DePeolo, D.J., 1983, The mean life of continents: estimates of continent recycling rates from Nd and Hf isotopic data and implications for mantle structure . Geoph~s . Res . Lett ., v. 10 , p . 705 - 708 . Dixon, T.H. and Batiza, R., 1979 , Petrology and chemistry of recent lavas in the northern Marianas: implications for the origin of island arc basalts. Contrib . Mineral . Petrol ., v. 70 , p . 167 - 181 . Ellam, R.M. and Hawkesworth, C.J., 1988a, Elemental and isotopic variations in subduction related basalts: evidence for a three component model, Contr~b . Mineral . Petrol ., v. 98, p . 72 - 80 . Ellam, R.~. and Hawkesworth, C.J ., ~ 988b, Is average continental crust generated at subduction zones? Geology, v. 16, p . 314- 317 . Engdahl , E. R., and Billington, S ., 1986 , Focal depth determination of central Aleutian earthquakes, Bull. of Seismological Soc. of Amer., v. 76, p. 77 -93 . Fyfe, W.S., and Kerrich, R., 1985, Fluids and thrusting, in Y. Kitano (Guest- Editor), Water-Rock Interaction, Chem . Geol ., fir. 49, p . 3S3 - 362 . Grow, J.A., 1973, Crustal and upper mantle structure of the central Aleutian arc, Geol . Soc . Amer . Bul l ., v. 84 , p . 2169 - 2192 . Hollister, L. S ., and Crawford, ILL., 1986, Melt-enhanced deformation: A major tectonic process, Geology, v. 14, p . 558 - 561. Karig, D., and Kay, R.W., 1 981, Fate of sediments on the descending plate at convergent place margins, Phil. Trans. Rov. Soc. Lond.' v. A301, p. 233 - 251. Kay, R.W., 1980 , Volcanic arc magma genesis : Implications for element recycling in the crust-upper mantle system, J. Geology, v. 88, p. 497-522. Kay, R.11., 1985, Island arc processes -relevant deco crustal and mantle evolution, TectonolphYs. fir. 112 p. 1-16.- ~ Kay, R.W., and Kay' S.~., 1986, Petrology and geochemistry of the lower continental crust: an Overview, Geol. Soc. Sc. Publ., v. 24,~p. 147- 159 . Kay, R.W., and Kay, S.Mah1-burg, 1988, Crustral Recycling and the Aleutian Arc. Geochem. Cosmochim. Acta. fir. 52, p. 1351-1361 . Kay , R. W., Rubenstone , J . L, and Kay , S . Mahlburg , 1986 , Aleutian terranes from Nd isotopes, Nature, v. 322, p . 605 - 609 . Kay, S. Mahlburg and Kay, R.W., 1989, Aleutian magmas in space and time; in Geology of Alaska The Geology of North America, feds. G. Plafker et al. Geol . Soc . Am., Boulder , CO ., ~ in press ~ . 204

Kay S . Mahlburg, and Kay, R.W. ~ 1985 ~ Aleutian tholeiitic and calc-allcaline Magma Series I : the mafia phnnocrysts, Contrib . Mineral ~ Petrol ., v. 90, p . 276- 290. Kay, S. Mahlburg, and Kay, Rig., 1985, Role of crystal cumulates and the oceanic crust in the formation of the lower crust of the Aleutian arc, Geology v. 13, p. 461-464. Kay, S. Mahlburg, Maksaev, V., Mpodozis, C., Moscoso, R., and Nasi, C., 1987 , Probing the evolving Andean lithosphere: Mid- late Tertiary magmatism i n Chile ~ 29 - 30 . 5 ° S ~ over the zone of subhor~zontal subduction, J . Geophys Res ., v. 92 , p . 6113 - 6189 . Kennett, J . P ., McBirney, A. R., and Thunell, R. C ., 1977, Episodes of Cenozoic volcanism in the Circum-Pacific Region, J. Ntolcanol. Geothermal Res., v. 2, p. 145-163. Lasaga, A C ., 1980, The kenetic treatment of geochemical cycles, Geochim . Cosmochim, Acta., v. 44 , p . 815 - 828 . Monger , J . W. H ., Clowes , R. M ., Price , R. A ., S imony , P . S ., Riddihough , R . P , and Woodsworth, G . J ., 1985, Cont, Dental Continent-Ocean Transect 7 B- 2 Juan de Fuca Plate deco Alberta Plains. Geol. Sac. Amer. Monaghan., Klein, J., and Measures, C., 1988, The origin of lobe in island-arc volcanic rocks, Earth and Planet Sci. Lett. ~ v. 89, p. 288-298. Morris J .D., and Hart, S .R., 1986 , Isotopic and incompatib1 e element constraints on the genesis of island arc volcanics from Cold Bay and Amak Island, Aleutians, and implications for mantle structure: Reply to a critical comment by M.R. Perfit and R.W. Kay, Geochem. Cosmochim. Acta., v. 50, p. 483-489. O' Brien, T. F., 1983 , Evidence for the nature of the lower crust beneath the central Colorado Plateau as derived from xenoliths in the Buell Park- Green Knobs diatremes, Ph.D. Thesis, Cornell University, 250 p. Parsons , B., 1981, The rates of plate creation and consumption, Geophys . Jour. Royal Astron* Sac., v. 67, p. 437-448. Perfit, M.R., and Kay, R.W., 1986, Comment on "Isotopic and incompatible element constraints on the genesis of island arc volcanics from Cold Bay and Amak Island, Aleutians, and implications for mantle structure" by J . D . Morris and S .R. Hart, Geochem Cosmochim. Acta., v. 50 , p . 477-481 . Sundquis~c, E., 1985 , Geological perspectives on carbon dioxide and the carbon cycle, in The Carbon Cycle and Atmospheric CO2: Natural Variations Archean to Present, E.T. Sundquist and U.S. Broecker, eds., pp. 5-58, American Geophysical Union, Geophys. Mono. 32, Washing~con, DC. 205

Tera, F., Brown, L., Morris, J., Sacks, I.S., Klein, J., and Middle ton, R., 1986. Sediment incorporation in island-arc Magyar: Inferences from 1OBe Geochim Cosmochim. Acta. v. 50 p. 535-550 Verplanck, E.P., and Duncan, R.A., Temporal variation in plate convergence and eruption rates in the Western Cascades, Oregon, Tectonics' v. 6, n. 2 p. 197-209, 1987. 206

Passive Cont~n~1 trains ME~NIC~; OF RIFTrNG OF WE LrI~EKE: 207

CONHENTS ON RIFTING AND PASSIVE MARGIN EVOLUTION IN LIGHT OP SOME RECENT STUDIES John C. Mutter Lamont-Doherty Geological Observatory of Columbia University ·& Brian Uernicke Harvard University It may be that we give insufficient recognition to the fact that much of the conceptual framework for quantitative models of extensional systems owe their origin to studies of rifting made several decades ago. The notion, for instance, that rifts evolve by progressive widening of symmetric grabens, culminating in the development of an ocean basin (Gulf of Suez to Red Sea to Gulf of Aden to Atiantic-type ocean) as described, for instance, by Dewey and Bird (1970) is at least fifty years old--DuToit (1937) considered oceans to be "merely rift-valleys of unusual width". The pre-plate tectonic literature is replete with examples of analyses of continental rift systems and sedimentary basins from which deductions were made that hold very nearly true today; even deductions made about passive margin evolution that were little more than inferences and speculations based upon a tiny amount of marine data, are astonishingly close to the mark when viewed today (see Bond, 1988 for an extensive review). In broad concept much of what we know about the tectonic evolution of extensional systems had been described in some detail by the early 1960's. An interesting comparison is provided by contrasting the continental rift-to-ocean basin cartoon of Heezen (1960) with that of Hellinger and Sclater (1983) in Figure I. Although more than two decades had passed between the construction of these stretches they are clearly very similar. Essential geodynam~c components Ouch as rift broadening and margin subsidence are 208

illustrated in both models. The older version is particularly remarkable in that it was used in support of Heezen's belief, at the time, in an expanding Earth model of global tectonics, and because it seems to include Gastric normal faults and even detachment surfaces that Heezen could only have imagined to be present. What is the essential difference between these sketches? The answer is that the modern sketch includes a description of the behavior of the subcrustal lithosphere in its analysis of margin evolution, Heezen's mode] treats everything beneath the crust as essentially uniform. It is this appreciation of the role of the lithosphere that marks a major point of departure between pre- and post-plate-tectonic descriptions of margins and enabled the first steps toward a quantitative analysis of margin evolution. Sleep (1971) first addressed the problem of passive margins forming by thermal contraction of a thin lithosphere, but was puzzled by the mechanism of both crustal and lithospher~c thinning. NcKenzie's (1978) analysis of sedimentary basin evolution described, in quantitative terms, the thermal and mechanical effect on crust and subcrustal lithosphere caused by instantaneous extension with mass conservation. Watts and Steckler (1979), Steckler and Watts (1978), Sawyer et al. (1982), LePichon and Sibuet (1981) soon after made direct applications of the "uniform extensions model to passive margin evolution. A particular focus of many of these efforts was to develop an understanding of the degree to which extension of-the continental crust takes place during margin evolution. McKenzie's mode] provided a basis upon which observations of margin morphology, basement structure, gravity pattern, heat flow and particularly the history of vertical motion could be "inverted" to derive fundamental thermo-mechanical properties of the lithosphere under extension, including the extension factor, 6. This quantity describes the degree by which crust is extended (stretched), and/or the amount by which the continental crust once

is thinned during extension. It could be recovered from observation of margin subsidence; a phenomenon that had been recognized at least as early as the turn of the century. Subsidence information was derived from the stratigraphic record in exploration wells after suitable corrections were made for sediment loading and compaction, and suitable account taken of the mechanism of isostatic compensation (Watts and Ryan, 1976; Steckler and Watts, 1978). The methodology proved extremely successful, and the late 1970's and early to middle 80's virtually became a subsidence "era". Little attention, if any, was paid to examining the actual mechanism by which the crust and lithosphere yielded under extension. In-NcKenzie's (1978) analysis of uniform extension there is an exact coupling between horizontal extension and vertical thinning of crust and underlying lithosphere. The extension factor, 6, that was obtained from subsidence analysis, therefore describes the extensional strain in a system in which deformation is achieved by a pure shear mechanism. The depth-dependence of the crust's mechanical properties can be incorporated into a pure shear model by considering the crust to be theologically layered in horizontal strata (Royden and Keen, 1980; Vierbuchen et al., 1982~. Other-refinements to McKenzie's original model include consideration of finite periods of extension (Javis and McKenzie, 1980)' lateral heat =transport-(Cochran, 1983)' and lack of conservation of crustal mass due to igneous intrusion (Royden et al., 1980~. Lithospheric thinning and the consequent mantle upwelling associated with extension in a pure shear system causes partial-melting-of the mantle immediately beneath the region of extension. Assuming adiabatic pressure-release melting and complete melt extraction, the amount of magma produced in this way has been calculated (Foucher et al., 1982; Keen, 1987; White et al., 1987~. Buck (1986) has modeled small scale convection in the 210

upper mantle induced by pure shear extension, and Mutter et al. (1988) have shown how this convection will enhance melt production during rifting and the earliest phases of seafloor spreading. Extension-related magma tism may be expressed as layers platonically emplaced (underplated) beneath the extended crust (Beaumont et al., 1982; McKenzie, 1984; LASE Study Group, 1986; Furlong and Fountain, 1986, Cans, 1987), and include eruptive units (Hinz et al., 1987; Mutter and Zehnder, 1988) if magma production rates are sufficiently great. Given the brief historic outline sketched above, it is pertinent to ask what the relevance and purpose of a margins workshop is at this time. One could argue that great progress has been made, that directions are fairly well set, and that our description of margin evolution is quite complete. Certainly there are details to be attended to, but the major task of quantifying what was conceptually known a century ago has been achieved. So why a workshop now, and what is its purpose? There are several reasons. First, along with the gains made by geophysicists during the subsidence era, many investigators began to look carefully at extensional structures in the upper crust, and found that the amount of extension implied by fault geometries was not necessarily consistent with that deduced from subsidence. Closer attention was paid to the actual mechanism by which the crust reacted to extensional stresses, and the importance of listric and low-angle, normal faulting, including subhorizontal detachment faulting, emerged. Studies on continental extensional systems contributed in a fundamental way. There is now an impressive data set from the Basin and Range province of the western U.S. that argues unequivocally for the importance of crustal detachment faulting during Tertiary continental extension. By contrast, however, in some modern extending terranes arrays of deeply biting normal faults characterize crustal deformation. In these areas evidence for low-angle normal faults of large 211

areal extent, and active in the brittle crust is lacking, and seismological observations show that some major steeply-dipping normal faults may remain planar to 16 km depth (Jackson and McKenzie, 1983; Eyidogan and Jackson, 1984; Nabelek and Eyidogan, in press). These large normal faults do not appear to have any listric character; if they flatten with depth, they must do so beneath the depths at which earthquakes nucleate. It is extremely difficult to imagine simultaneous operation of deeply-biting, planar normal faults and detachments faults which remain shallow dipping over large distances. The detachment faults would be cut by these major normal faults and rendered inactive. Both types of faulting are clearly observed in-extensional terraces, yet the existence of one seemingly precludes activity of the other. We are currently faced with a considerable paradox in understanding the relative importance of various extensional mechanisms. It is not too much of an exaggeration to say that we do not really know the mechanism by which the earth's crust responds to extensional stress. Do planar normal faults accommodate extension for small extension factors and detachments at large extension factors? If so one would expect to see some evidence for detachment faulting on the outer part of virtually all passive margins and planar normal faults on the inner margins. Do we have the data base to assess the problem? Does planar normal faulting, listric faulting and detachment faulting occur in a temporal progression? Again, is there a data base to allow us to make a judgment on this? - - The deliberations of a margins workshop should provide directions for research into crustal deformation mechanism aimed at obtaining a clear understanding of the mechanisms involved in extensional tectonics. Second, the mechanism by which the lithosphere as a whole fails under extension has been examined also, and this had led to one of the first major 212

breaks from the past in the way we view extensional tectonics. Much of the conceptual foundation that was built up during the late 1$00's and the first half of the twentieth century drew heavily on an extrapolation from lessons learned in investigations of the African rift system. It is perhaps appropriate, then, that the pure shear Reformational mechanism upon which virtually all quantitative analyses of extension made in the 1970s and early 1980s were based, has been called into question by study of a different region of the Earth that has recently undergone active extension; the Basin and Range province of the western U.S. (e.g., Uernicke, 1981, 1985; Lister et al., 1986). Here, detachment faults that are well known features of the upper crust have been proposed to be deep-rooted structures generated when extensional stresses are accommodated by simple shear deformation..of the entire crust and lithosphere (Vernicke, 1981, 1985~. Figure 2 cartoons a comparison that is now commonly made between simple and pure shear extensional deformation of the lithosphere. In particular, simple shear deformation contrasts in generating strong asymmetry in the extensional system? in decoupling upper and lower crustal sections so that upper crustal extension factors may be markedly different from those in the lower crust, and in causing a spatial separation between crustal thinning and lithospheric thinning. ~ One consequence of the mode] is that pressure-release melting may generate magma beneath relatively-unthinned continental crust. Buck et al. (1988) have modeled the evolution of the thermal structure of the rift during simple and pure shear deformation and predicted substantial differences in uplift/subsidence and patterns of heat flow. Kusznir and Park (1982, 1984, 1987) and England (1983) have described the fundamental effects that variations in geothermal gradient, crustal rheology, extensional strain rates and crustal 213

thickness have on the mechanism of extensional failure. An attractive feature of the simple shear mechanism is that it provides a potential model to explain the non-erosional exhumation of lower crustal rocks in the enigmatic structures of many extensional terranes known as metamorphic core complexes (Lister and Davis, in press). Continued simple shear deformation could essentially unroof the lower crust or even mantle as it is dragged to the surface beneath detachment faults (Figure 3~. Exposure of the lower crust by pure shear deformation is considerably more difficult to model. Recently, models have been advanced that incorporate simple shear tectonics that can incorporate the simultaneous activity of both low angle and steeply dipping faults. Figure 4, from Wernicke and Axen (1988), shows a conceptual model of a migrating zone of isostatic uplift during extension. Note that although detachment surface is originally subhorizontal, its ascent through the brittle crust occurs along a relatively steep zone. Note also scale of asymmetric half-graben in comparison to magnitude of crustal pull-apart. Restoration assuming stratigraphic cutoffs at the bottom and upper right corners at the basin fil ~ may lead to under estimation of extensions by an order of magni tude or more ~ The recent li terature has seen models of extension proposed in which detachment tectonics by simple shear deformation plays an essential role in passive margin evolution and formation of intra-continental rift systems (representative studies include LePichon and Barbier, 1987; Boillot et al., 1987; Lister et al. ~ 1986; Wernicke and Burchfiel, 1982; Ussami et al ., 1986; Mutter et al., submitted; Beach, 1986; Alimendinger et al., 1987; Beach et al., 1987; Gibbs, 1984, 1987; Davis, 1987; Sonder et al., 1987; John, 1987). Simple shear tectonics is in the process of rapidly gaining ground as a principal component of passive margin models. Boillot et al. (1987) have advanced a 214

wholly simple shear interpretation of the evolution of the Galicia margin based on seismic reflection data and results from ODE Leg 104 (Figure 5), which includes exhumation of upper mantle periodotites by a process analogous to that of a formation of metamorphic core complexes in the Basin and Range Province. Karson et al. (1987) invoke simple shear detachment tectonics to raise deep oceanic crust and upper mantle to surface exposure on the slow spreading mid-Atiant~c Ridge. While these studies represent a clear departure point in research into extensional tectonics, it has by no means been demonstrated that simple shear is indeed the dominant mechanism for -the extensional deformation that culminates in seafloor spreading. On some passive margins mid-crustal seismic reflecting horizons have been recognized that may be detachment surfaces (Boillot et al., 1987; Wernicke and Burchfiel, 1981; Lister et al, 1986, in press; LePichon and Barbier, 1987; Mutter et al., in press), but these observations of themselves do not signify that simple shear deformation of the entire crust and lithosphere has occurred; detachment surfaces can also form in the mid crust between regions of distinctly different crustal rheology during pure shear deformation (Barr, 1987; Kusznir and Park, 1987~. Buck et al. (submitted) argue that the pattern of heat flow in the Red Sea is, in fact, consistent with pure shear extension.- Hutter et al. (in press) propose that the margin off Nib. Australia evolved by an interplay in space and time between both simple and pure shear deformation and have suggested that the location of final break-up is probably controlled by pure shear deformation (Figure 6~. Allmendinger et al. (1987) cite extensive COCORP seismic reflection data in the Basin and Range in asserting that "no one model of intra-continental extension is applicable to the entire province". Beach et al. (1987) come to a similar conclusion using deep seismic data from the Northern North Sea e Clearly, the 215

un' form stretching model has been successful in describing the overall pattern of extension and subsidence. Deducing inhomogeneities in the stretched lithosphere from deviations in the uniform stretching model may reveal systematic patterns in the rift architecture we have only begun to contemplate. The fundamental mechanism and the governing constraints on continental extension and passive margin formation are in the process of evaluation. Definitive conclusions regarding the role of pure and/or simple shear deformation of the lithosphere are not yet available, and it is specious to suggest that one or the other is the "true" mechanism. Similarly, it seems imprudent to construct a general model of passive margin evolution at this time, from the example-specific foundation of the incompletely understood Basin and Range province. The definition of research directions for study of lithospheric scale deformation should be an important objective of this workshop. Third, we believe that it is a particularly auspicious time to establish strong intellectual ties between on-land research programs into the nature of extensional tectonism and offshore studies. All marine geoscientists studying passive margins readily acknowledge that recent fundamental insights into extensional tectonics have derived at least~as much from continental studies as from marine studies. Knowledge of the geometry and timing of extensional structures -- developed across a variety of pre-extensional terrane -- is only now being gained. These studies are revealing important similarities and - differences in styles of upper and middle crustal~extension between areas of differing pre-extensional geology. While the majority of seismic reflection interpretation of passive margins focuses on large half-grabens developed under thick post-rift cover, work in the Basin and Range is now showing that the structural expression of large-scale extension may be only subtly expressed in 216

reflection seismograms, and is often not associated with large, asymmetrical sedimentary basins. In any event, deducing the extension factor, 6, on passive margins by restoring asymmetric half-graben must be done with thorough knowledge of the range of structural styles expressed on land. It is apparent that there is a wide gulf in approach and understanding between onshore and offshore geologists. To gauge the severity of this gulf, and the opportunities that exist to understand rift architecture, Basin and Range geologists are skeptical that any study of a zone of offshore extension has measured the extension factor, $, within a reasonably bounded uncertainty. Local estimates of extension in the Basin and Range are quite precise, but regionally a data base of mapping geochronology and subsurface data is not yet available to reasonably constrain ~ for the whole province. The potential of doing 50, however, is quite high, but we must first understand the fault geometries in detail. A major potential pitfall in understanding half-graben is the role of very short wavelength (< 10 km) high amplitude (> 20 km) isostatic rebound in the footballs of normal faults (e.g., Wernicke and Axen, 1988; Figure 4) which, in general, causes underestimate of the amount of extension in half-grabens (Wernicke, 1988~. ~ Our inability thus far to have reasonably constrained 0, anywhere, is a severe limitation on assessing the validity of pure vs. simple shear, and the role of magma tic underplating in the evolution of passive margins. A national program of research that strongly interfaces onshore and offshore research is needed. This problem is particularly acute for extended continental margins relative to compressional ones because the end product of rifting is so difficult to access and the active examples so very few. It is of fundamental importance that we focus passive margin research 217

over the next several years into process-oriented investigations aimed at resolving the fundamental mechanism(s) controlling extensional deformation. We need to establish strong ties to studies presently under way that focus on active or recent continental extension, both in the Basin and Range Province and other extensional terraces. There is a need to quantify the relevance of various concepts concerning the role and nature of detachment faults, other styles of extension, and the timing and distribution of extension-related magmatism. To do this, it is vital that we obtain data that allow us to trace, where possible, the geometry of faults from surface outcrop to deep crust, and define the seismic structure of the lower crust and upper mantle. We should employ acquisition methodologies that provide data allowing direct comparison between active rifts and passive margins, from one margin to another, of different style, age, or volcanic history; from one margin segment to another, and across conjugate margins. Much of the focus should be on the deep crust and upper mantle because the interpretation of detachment faults, the inferred role of pure vs. so mple shear extensional mechanism, and the importance of magmatism during extension depend heavily upon establishing the nature of the lower crust and the manner in which it deformed. The Continental Margins Vorkshop creates a timely opportunity to define the key questions that we must answer in order to advance our understanding of the processes underlying the formation of these fundamental Earth structures, and set forth plans for long-term global investigations that can answer these questions. 218

REFERENCES Allmendingex' R., J.W. Sharp, D. Von Tish, L. Serpa, L. Brown, S. Kaufman, J. Oliver and R.B. Smith, 1983. Cenozoic and Mesozoic Structure of the Eastern Basin and Range Province, Utah, from COCORP Se~smic-reflection Data; Geology, Vol. 11, p. 532-536. Alimendinger, R., T.A. Hauge, E.C. Hauser, C.J. Potter and J. Oliver, 1987. Tectonic heredity and the layered lower crust in the Basin and Range Province, western United States, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics' Geol. Soc. Special Publication No. 28' p. 223-246. Barr, D. , 1987. L~thospheric Stretching, detached normal faulting and footwall uplift, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc. Special Publication No. 28 ? P - 75-94. Beach, A., 1986. A deep Seismic Reflection Profile Across the Northern North Sea; Nature, Vol. 323, p. 53-55. Beach, A. , T. Bird and Ae Gibbs, 1987. Extensional tectonics and crustal structure: Deep seismic reflection data from the northern North Sea Viking Graben, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc. Special Publication No. 28, p. 467-476. Beaumont, C., CeEe Keen and R. Boutillier, 1982. On the Evolution of Rifted Continental Margins: Comparison of Models and Observations for the Nova Scotia Margin, Geophys. J.R. astr. Soc.; Vol. 70, p. 667-715. Boillot, G., M. Recq, E.L. Winterer, A.W. Meyer, J. Applegate, M. Baltuck, J.A. Bergen, M.S. Comas, T.A. Davies, K. Dunham, C.A. Evans, J. Girardeauy, G. Goldberg, J. Haggarty' -L.~. Jansa, J.A. Johnson, J. Kasahara, J.P. Loreau, E. Luna-Sierra' H. lloullade, J. Ogg, M. Sarti, J. Thurow, and M. Williamson, 1987. Tectonic denudation of the upper mantle along passive margins: a model based on drilling results COOP leg 103, western Galic~a margin, Spain), Tectonophysics 132, p. 335-342 . Bond, G. C., 1988. Evolution of Thought on Passive Margins Prom the Origin of Geosynclinal Theory (1960) to the Present. Bull. Geological Society America, in press . Buck, U.R., 1986. Small-scale convection induced by passive rif ting: the cause for uplift of rift shoulders, Earth and Planet. Letters, 77, p. 362-372. Buck, W. R., 1988, Flexural rotation of high-angle normal faults to produce abandoned horizontal detachments: EOS, v. 69, p. 465. Buck W. R., F. Martinez, M. S. Steckler and J.R. Cochran, 1988. Thermal consequences of lithospheric extension: pure and simple, Tectonics, 7 (2) p. 212-234. Cochran, J.R., 1983. Effects of Finite Rifting Times on the Development of Sedimentary Basins ; Earth and Planet . Sci . Let t ., Vol . 66 , p ~ 289-302 . 219

Davis, G.H., 1987. A shear-zone model for the structural evolution of metamorphic core complexes in southeastern Arizona, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc Special Publication No. 2S, p. 247-266. Dewey, J.F. and J.M. Bird, 1970. Mountain Belts and the New Global Tectonics; J . Geophys. Res, 75, 2625-2647 . DuToit, A.L., 1937. Our wandering Continents; Edinburgh, Oliver and Boyd, 366 p. England, P.C., 1983. Constraints on extension of continental lithosphere, J. Geophys. Res. 88 p. 1145-1152. Foucher, J.P., X. LePichon and J.C. Sibuet, 1982. The Ocean Continent Transition in the Uniform Stretching Model: Role of Partial Melting in the Mantle ; Phil . Trans ~ R. Soc. Lond ., Vo] ~ A305 , p . 27-43 . Furlong, K.P. and Fountain, D.tI., 1986. Continental crustal underplating: thermal consequences and seism~c-petrologic consequences ; J. Geophys ~ Res e, 91, p. 8285-8294. Gans, P.B., 1987. An Open System, Two-Layer Crustal Stretching Model for Eastern Great Basin; Tectonics, 6~), pe 1-12 ~ Gibbs, A.D. ~ 1984. Structural evolution of extensional basin margins, Jour. Geol . Soc. London, 141, p . 609-620. Gibbs, A.D., 1987. Development of extension and mixed-mode sedimentary bas~ns, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc. Special Publication No. 28, p. 19-34. Heezen, B. Ce ~ 1960. The R~ft-in the Ocean Floor, Sct. Amer., 203, p. 98-110. Hellinger, S. J., and Sciater, J. G., 1983. : Some Comments on Two-Layer Extensional Models for the Evolution of Sedimentary Basins. J. Geophysics, Per., BB, p. fi25~1-8269. Hinz, K., J.C. Mutter, C.M. Zehnder and the NGT Study Group, 1987. Symmetric conjugation of continent-ocean boundary structures along the Norwegian and East Greeniand margins, Marine and petrol. Geol., 4, p. 165-~87. Jackson, J.A., 1987. Active normal faulting and crustal extension, In: M.P. Coward, J.F. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc. Special Publication No. 28, p. 3-17. Jackson, J.A., G. King and C. Vita-Firizi, 1982. The [leotectonics of the Aegean: An Alternate View; Earth and Planet. Sci. Lett., Vol. 62, p. 303-318. Jackson, J. and D. McKenzie, 1983. The Geometrical Evolution of Normal Fault Systems; J. Struct. Geol., Vol. 5, p. 471-482. Jarvis, G.T. and D.P. McKenzie, 1980. Sedimentary Basin Formation With Finite 220

Extension Rates ; Earth Planet . Sci . Let t ., Vol ~ 48 , p. 42-52 . John, B.E., 1987. Geometry and evolution of a mid-crustal extensional fault system: Chemehuivi Mountains, southeastern California, In: M. P. Coward, J. F. Dewey and P. L. Hancock, eds ., Continental Extensional Tectonics , Geol. Soc. Special Pub' ication No. 2S, p. 313-336. Karson, J. A., G ~ Thompson, S ~ E. Humphris , J ~ M e Edmonel , W. ~ ~ Bryan, A. T . Winters, R. A ~ Pockainy, J . F. Lasey ~ A. C. Campbell, G . Klinikhamer, M. R . Palmer, R.J. Kinzler, and M.M. Sulanowska, 1987. Along-axis variations in seafloor spreading in the MARK area; Nature, 328, 681-685. Keen, C.E., 1987. Some important consequences of lithospheric extension, In: M. P. Coward , J . P. Dewey and P. L. Hancock, eds ., Continental Extensional- Tectonics, Geol ~ Soc. Special Publ ication No. 28 p. 67-74. Kusznir, N. J ., 1982 ~ Li thosphere response to externally and internally derived stresses: a viscoelastic stress guide w'th amplification, Geophys. J.R. astr. Soc., 70, p. 399-414. Kusznir, N.J. and R.Ge Park, 1984. Intraplate lithosphere deformation and the strength of the lithosphere, Geophys J.R. astr. Soc. 79, p. 513-538. Kusznir, N.J. and R.G. Park, 1987. The Extensional strength of the continental lithosphere: its dependence on geothermal gradient, and crustal composition and thickness, In: M.P. Coward, J.F. Dewey and P.L. Hancock, ease, Continental Extensional Tectonics, Geol. Soc. Special Publication No. 28, p. 35-52 . LASE Study Group, 1986. Deep Structure of the U. S. East Coast Margin From Large Aperture Sei smic Experiment (LASE) ; Marine and Petrol ~ Geol., Vol. 3~3), p. 234-242. LePichon, X. and F. Barbier, 1987. Passive margin formation by low-angle faulting within the upper crust: the Northern Bay of Biscay margin, Tectonics, 6, p. 133-150. - LePichon, X. and J.-C. Sibuet, 1981. Passive Margins: A Model of Formation; J. Geophys. Res., Vol. 86, p. 3708-3720. Lister, GeS. ~ M.A. Etheridge and P.A. Symonds, 1986. Application of the Detachment Fault Model to the Formation of Passive Continental Margins; Geology, Vol. 14, p. 246-250. Lister, G.S. and G.A+--Davis, submitted. A mode] for the formation of metamorphic core complexes and mylonitic detachment terranes. Lister, G.S., M.A. Etheridge, and P.A. Symonds, subm~t tea. Detachment models for the formation of passive continental margins, ~n review. McKenzie, D.P., 1978. Some Remarks on the Development of Sedimentary Basins; Earth and Planet. Sci. Lett., Vol. 40, p. 25-32. McKenzie, D.P., 1984. A possible mechanism for epeirogenic uplift, Nature, 221

307, p. 616-618. Mutter, J.C., W.R. Buck and C.M. Zehnder, 1988. Convective Partial Melting I: A Model For Formation of Thick Igneous Sequences During the Initiation of Spreading; J. Geophys. Res., Vol. 93, No. B2, p. 1031-1048. Mutter, J.~. and C.M. Zehnder, 1988. Deep crustal Structure and Magmatic Processes: The inception of seafloor spreading in the Norwegian-Greenland Sea, Geol. Soc. London Special Publication: Early Tertiary Volcanism and the opening of the NE Atlantic, No. 39, p. 35-48. Mutter, J.C., R.L. Larson and the N.W. Australia Study Group, 1988. Extension of the Exmouth Plateau: Deep Seismic Reflection/Refraction Evidence for Simple and Pure shear mechanisms, GEOLOGY, in press. Nabelek, J. 1985. Geometry and mechanism of faulting of the 1980 El Asnam, Algeria, earthquake from inversion of teleseism~c body waves and comparison with field observations. J. Geophys. Res. Royden, L. and C.E. Keen, 1980. Rifting Processes and Thermal Evolution of the Continental Margin of Eastern Canada Determined From Subsidence Curves; Earth and planet. Sci. Lett., Vo1. 5l, p. 343-361. Royden, L., J.G. Sclater and R.P. von Herzen, 1980. Continental Margin Subsidence and Heat F~ ow: Important Parameters in -Formation of Petroleum Hydrocarbons ; AAPG Bul1., tTol . 64, p. 173-~87 . Sawyer, D.S., B.A. Swift, J.G. Sclater and H.N. Toksoz, 1982. Extensional model for the subsidence of the northern Uni ted States Atiantic continental margin, Geology, 10, p . ~ 34-140 . Sclater, J.G. and P.A. F. Christie, 1980. Continental Stretching: An Explanation of the Post and Mid-Cretaceous Subsidence of the Central North Sea Basin; J. Geophys. Res., Vol. 85, p. 3711-3739. Sonder, L.J., P.L. England, B.P. Wernicke, and R.L. Christiansen, 1987. a physical model for Cenozoic extension of western North America, In: H.P. Coward, J.~. Dewey and P.L. Hancock, eds., Continental Extensional Tectonics, Geol. Soc. Spec. Publ. No. 28, p.--~-87-201. S1eep, N. H., 1971, Thermal effects of-the formation of Atiantic continental margins by continental breakup: Geophysics Jour. R. Astr. Soc., v. 24, p. 325-350. Steckler, M.S. and A.B. Watts, 1978. Subsidence of the Atiantic-type continental margins off New York, Earth Planet, Sci. Lett., 4l, p I-13. Ussami, N., G.D. Karner and M.H.P. Bott, 1986. Crustal Detachment South Atiantic Rifting and Formation of Tuxano-Gabor Basin System; Nature, Vol. 322~6080), p. 629-632. Vierbuchen, R.C., R.P. George and P.R. Vail, 1982. A Thermal-Hechanical Model of Rifting With Implications for Outer Highs on Passive Continental Margins, In: J.S. Watkins and C.J. Drake, eds., Studies in Continental Margin 222

Geology; Amer. Assoc. Petrol. Geol. ~ Memoir 34, p. 765-778. Wat ts, A. B., and Ryan, W. B. F., 1986. FTexure of the Lo thosphere and Continental Margin Basins, Tectonophysics, 36, p. 25-44. Watts, A.B. and M.S. Steckler, 1979. Subsidence and Eustacy at the Continental margin of Eastern North America, Maurice Ewing Symp. Series 3, AGO, Washington, D.C., p. 218-234. Wernicke, B., 1981. Insights From Basin and Range Surface Geology for the Process of Large-Scale Divergence of Continental Li thosphere (Abs tract ), In : Papers Presented to the Conference on Processes of Planetary Rif ting, Lunar and Planetary Ins t ~ tote, Hous ton, p . 90-92 . Wernicke' -Be and B.C. Burchfiel, 1982. Modes of Extensional Tectonics; 3. Struct. Geol., Vol. 4, p. 105-115. Wernicke, B., 1985. Uniform-Sense Normal Simple Shear of the Continental Li thosphere ; Can. J ~ Earth Sci ., Vol. 22 , p . 108-126 . Wernicke? B. and G.V. Axen, 1988. On the role of isostasy in the evolution of normal fault systems Geo] ogy, v. 16, p. 848-851 . Wernicke' B., 1988. On determining modes, magnitudes, and rates of continental extensions surface geology and reflection seismograms; Geological Association of Canada Program wi th Abstracts , v. 13, p. A133. White, R.S. ~ GeDe Spence, S.R. Fowler, D.P. McKenzie, G.K. Westbrook and A.N. Bowen ~ 1987. Magmatism at rif ted continental margins, Nature, 330, p. 439-444. 223

E E _ _ ~ _ O ~ ~ O ~ O '.'.''.'.,— ";2,',,'.';,,2 ,2 :~.''2'. .::'," '.'-.' i: - · "_'"''.. ~ ....... : . ,.: . . ~ ...-. ~ ... O I` i`- - .: . . ...... ............ j:~ :.. ,. .. .- - :~-:': .' At . ;.-.',: 2 -. . ,,, 4,, . GINO,.. Z ~ 2-h' o .... .," .:.%,. . ..... . -. ;.:.', ;; ,, .... .' . o - 0 o "S o — o . ; - _ :: ....,, Z :: .... ~ .~,- X ~n .... , ~ .,..,. ~ ~ . . ~ ~ :: :: :: ~n ~ .- ~ o c. . - I_ ~ ~ er _ 0 ~ ec r O _ I,L O . Z Z _. S ~ ~ ~ ID - .< ~ . .. , ~ ~ _ ,., _ ., _ , . ~ _ , .. _ .. . - eL ~ , O "r '....,-. !: 't . . . ~ ,; . ~ 1~- . - . ' -~;# _ O ~ O ~ ~ ~ .... ~1 1 1 \ 1 ~ 1 1, ! 1, 1 I ! , ~ ~ F IGURE 1 i 1 ! i , ~ ,, ~ j ! ~ ! ~ ~ ~ , 224 , _ z C ~ z o o ~ ............. -. ............... . - ~ c ,; ·.

or '/~-~/] '\~\ ~"~"1 C ~ 111~ '-/'~ ~l'\ :~ i ~ . ~ ~ .~.-.~ :-: - . . ~ . .~- o o ED (my) Hldsa o cYl fir i In O ~2:~ In ,, ~ Car CO ~ ~ - c~0 ~ O O d" 0 (W1) HldJO FIGURE 2 225 -

km a BEFORE EXTENSION 40 242 50 40 ~0 10 it' SCALE km 0 10 20 , ~ _ , ,, _ b BEFORE ISOSTATIC REBOUND c AFTER REBOUND 422 km I. ~ D*-,, PA ::~ d WIDENING ZONE OF UPLIFT ~ 65 km _,, .. lo. FIGURE 3 me zone of permanent subvertical simple shear strain 226

1~ ~ \ \ \ o \ \ ' \ - - 1 a At onset of debn~inatior, ~ $,N, b Low-angb normal faults "fire'' from disseminating Bayer MuNipb d~eh~ tam ,9 Mylon~t¢tront C Lower pi - e bow upward - ~\ actn,. insctrve mylonites _~_ mvionites __ d l~letemorphic core complex in lower pl - e culmination FIGURE 4 227 Mylonitised granite Jn~cinre ~r~==

R3 'iai~iii~ = ~ ~ ~ ~ ~ 5_. \ 5.. ~1 1 ~ 1 ~m ~ r L`,. 6.~: , [Mimi! ;Ij ill!.,,! ~ |,,,, !!i . R2 , I=- . 1 ~ W~ _ ' ' 1 ' I'l' 1' FIGURE 5 228 s a a. so" CONTIN£NTAL CAT lo J

f -- ~ ~ EARLY JURASSIC ~ ~~ TIC 56~ - . Or (lo) MIDDLE TO LATE JURASSIC I.. ~t~lbot~d.~ . .'. ' . :. TRYST . - . . 1 1 LATE JURASSI EARLY CR£TACOUS 1, ~ i~100 km FIGURE 6 229

IGNEOUS PROCESSES AND THE EVOLUTION OF RIFTED CONTINENTAL MARGINS Jeffrey A. Karson ~ and Carolyn M. Zehnder ~ ~ TDepar~nent of Geology, Duke University, Durham, NC 27706 · .s. Lamont-Doherty Geological Observatory, Palisades NY, 10964-0190; also Department of Geological Sciences, Columbia University, New York, NY. Rifted or "passive" continental margins result when continental rifting has proceeded to the in~aanon of seafloor spreading and the birth of a new ocean basin. Whether rifting and breakup are achieved by a wholly pure shear, simple shear, or some combination of these mechanisms, the process must involve thinning of the lithosphere and upwelling of the asthenospheric mantle which, in turn, must give rise to decompressional melting. This is so regardless of whether rifting is a passive, stress-induced process or an active process driven by a hot upwelling mande plume. In The latter case, additional melt would result from the thermal anomaly itself. Magmatism is therefore intimately tied to extensional tectonics so that an understanding of extension is incomplete without a full description of the magmanc processes. Geophysical surveys and deep sea drilling of many passive continental margins have confined Cat volcanic and intrusive bodies are generally to be included as components of the rifted margin architecture, but their relative volumes and distributions are highly variable. It has been suggested that passive margins can be usefully classed as volcanic or non-volcanic (Figure 1; COSOD II Working Group on Stress An Deformation of the Lithosphere). Rift magmatism has been suggested by several investigators to result 230

\~/OLCANIC MARGIN ~~0—ISC}km -l ~ -50~ 300+km ~ ~ , . ~ 2 2 ~~ — NONVOLCANIC MARGIN 1~ -- 100—300+km - - ~ ~ Of*: ~~~ ~+ ~ ~ . . ~ ~ ~ ~ ~ . ~ ~ ~ ~ ~ ~ * ~ * . ~ * * * ~ Me {30 Figure 1. Comparison of the typical structure elements of volcanic and nonvolcaruc margins. The n~nbers refer to the following structural elemenm: 1, the nonnal thickness oceanic crust; 2, the Wick volcanic succession associated with the initiation of seafloor spreading of volcanic margins of which the sowed dipping Its form the upper sequence; 3, a s~ctura1 high in continental crust that often occurs adjacent to the thick volcanic succession; 4, Tinned, subsided continental crust; 5, unstretched con~ntal crush The dot dash line mark Me s=ugraphic level Of blowup. Parallel ruled regions indicate sediments. The random dot pattern indicates igneous units, the crosses show continental crust From Mutter et al., 1987. 231

from; a) hot upwelling mantle plumes that drive the rifting activity (e.g. Figure 2; Bonatt' and Seyler, 1988; Karson and Curtis, in press), b) the initiation of partial melting processes as the mantle passively upwells through its solidus during rifting (e.g. Figure 3; Mutter et al., 1988; Perry et al., 1987), or c) some combination of these processes (e.g. White and McKenzie, in press). Magmatic activity plays an important role in the structural development of the crust, the thermal and mechanical evolution of the lithosphere, and subsidence history of rifts and continental margins. The Continental Margin Workshop provides an opportunity to evaluate the present state of knowledge of magmatism and rift margin development and to consider future directions that might more clearly elucidate magmatic and tectonic processes that form rifted margins. Ma~,mahsm associated with the formation of rifted margins may be a fundamental and unique indicator of the nature of processes that operate in the mantle beneath continents, recording the history of transition from initial leaks in a new lithospheric crack to a mature steady-state mid-ocean ridge spreading center. Because magma transport represents an extremely efficient means of removing heat from stretched and thinned lithosphere, intrusive and extrusive rocks in these environments also have important consequences for the thermal evolution of the lithosphere of rifted margins (Royden et al., 1980; Keen, 1987). During the tectonic extension of rifts the intrusion of magmas may also play a role in the partitioning of strain through the lithosphere as it fills "potential volumes" created by stretching (Figure 4, Karson and Curtis, in press, Bohannon, 1986). Magmansm may also provide fundamental insights into Me extensional mechanisms). For instance, volcanic active n the Basin and Range began in Me Oligocene and has occurred sporadically since, with major eruptions temporaBy and spatially related to extensional events. Thompson and McCarthy (1986), Klemperer et al. (1986) and Wernicke (1986)have suggested that magmatism during extension could invade and unde~plate the extended crust, preserving Moho depth (Figure 4). The common observation of a highly reflective lower continental crust (e.g. Brown et al., 1986) could be 232

DETACHMENT (FAU Ll + + + ~ + ~ + + + ~ + ~ ~ ~ + + ~ ~ ~ + + ~ ,.CONTI HI ENTA t ~~;~+~ " ++ +~ ~ +++ ~ a+ MOHO CRUST + ~ + ~ ~ ` - + + + + + + GARNET GRANULITE/ ~1GH-P GABBRO 1 ST4GE I E. AFRIcaN R IFT STAGE 2 NORTH ERN RED SEA dikes AE)B ROS . ~ STAGE 3 SOVTHERN RED SEA ~/~///////^ OC:EANIC ~_, (;CRUST < + ~ + + ~+ + + + + ~ + + ~ + +~. - - ~+ ~ + + + GABBROS LOW-P ~ f tilGH-P FiEure 2. Schemadc mode1 showing different stages in the evolution of the crust in the Red Sea rift. Veriical and ho~ntal scales are arbitra~y. The possibili~ of as mmelnc development of rifiing and extension along low-angle detachment faults is indicated sche~n~tically. Underplating of ~e continental crust before a~d d~g rif~ng by gabbroic intrusions is emphasized in ~is carmon. Stage 1: A th~mal anomaly develops in the mande in the earliest stages Of rifi~ng, causing the ~ of ~e as~enosphere and thinning of ~e subcrusta1 continental lid~osphere and the emplacement of mah~c magmas close to the base of the continental crust. Stage 2: Progressive s~retching and ~inning of ~e con~ntal crust results in ever decreasing depths of mafic intrusion a~ad an incteased frequency of d~use basaltic in - tions tow~ds ~e incipient rift axis Stage 3: Basalt injections gradually become res~icted to ~e incipient rift axis as seaf]~ s~e~ing ini~s wi~ ~e development of seafloor magnenc anomaly stripes. Seafloor ~preading =dally develops in equidistant axial "hot pO~ltS", which are probably reLited to upwelling a~henospheric diapirs ~igg~d by a R~leighl Taylor-~e ins~bility in the upper mantle. These "hot points" act as nuclei f~ a~cial propagation of the oceanic rift, evenmaBy resu~ng in a more or less continuous ems of pp~ing. From Bonatn and Seyler, 1987. 233

a —3L— ~ 5L ~ :T ~ _ ,~. : ~ 3L— I SIMPLE SHEAR U2o~ PURE SHEAR . . ~ CRUST _ _. V ! MANTLE : ,l0— ASTttENOSPHER£ ~~ I_ ·, ~~ 1 —, pej I L I THOSPHERE ~ , ~ E inure 3. Schematic model showing how the fonnanon of Volcanic and nonvolca~uc margins might reflect differences in rifting scenarios; a: preexisting s~ruc~al grains within the old Continental coon may effect the width of the rifting region and may vary along strike of Me incipient rift. In As carton, rifting occurs over width 3L above, and width L below. Extension of mddl 2L results in different seething factors for to rifting CmSL b: The widening of the rift zone from 3L to 5L results ~ a sketch factor of 1.7, as shown in We upper panel, whereas widening of the rift zone from L to 3L results in a stretch factor of 3, as shown in Me lower panel. c: Mantle upwelling results in the juxtaposition of hot asthenosphere and cool lithosphere. The h~ontal temperatme differences result in variable densities that are acted upon by gravitational forces and induce a small scale convection in the mantle. The magnitude of the mnperamre gradients, and the vigor of the convection, depend on the width of Me zone of extension and the rate at which extension occurs. The convective circulation in the upper panel is thereby less vigorous than that in the lows panel. As Me ppwelling mantle crosses the solidus, partial melting occurs by adiabatic decompression. The slob scale convection can result in enhanced malt production as mantle is copulated Trough the soughs. d: The extension of Me lithosphere could occur by simple shear, pure shear, or some combination of these mechanisms. e: lbe different histories of nfung can result in the juxtaposition of margins Mat are largely arnagmaiic, producing nonnal Sickness oceanic crust at the onset of seafloor spreading, and margins Hat indicate a prolific magmaiic episode that accompanied the onset of seafloor spreading. This cartoon depicts ~ juxtaposition of the nonvolcanic Exmouth margin (background) and t}= Volcanic Cllvier margin (foreground) off He coast of NW Auk (Mutter et al., unpublished survey data),which appear to result from rift-induced processes alone. 234

a. b. ~ _7m Meson_ ~~ - -I r - :~ An= M O H 0~ `~0 I'm o f i c _ I,— ... ultramof~c crust - . . . :;: mantle Emigre 4. The deem of Moho may not always be a good indication of the magnitude of extension. a. Dunng amagmanc pure-shear extension of He lithosphere, He depth to Moho may provide good estimate of overall coastal extension factors. b. Tectonic extension results in He eventual production of manic melt as the mantle upwells and decompresses, so Hat rift-induced magmauc activity could intrude or unde~plate He stretched cm~ntal crust and "preserve" Moho depths (Wernicke, 1986). Rabat is, We intrusion of magmas during extension may play a role in the partitioning of sow through the li~osph=E as it fills "potential vol~nes" cream by stretching (Larson and Curtis, in press). From Parson and Curtis, in press. - - 235

explained by He effects of layered igneous un~rpla~ng and sill Infusion (Hauser et al., 1986; Cheadle et al., 1987). UndeIplated igneous sequences have been suggested to have formed a well-developed, deep, high-velocity crustal layer in the UeSe East Coast (LASE Study Group, -1986) and in the Rockall Plateau margin (White -et al., 1987) during rifting that culminated in seafloor spreading. Unusually thick igneous crust formed during initial seafloor spreading in the conjugate seaward dipping reflector margins off Norway and East Greenland (Hinz et al., 1987) has been interpreted to result from the enhanced melt production by convective partial melting due to rapid, abrupt rifting (Figure 3; Mutter et al., 1988; Mutter and Zehnder, 1988; Z~hnder et al., submitted). The spatial and temporal development of magmatism is therefore a primary guide to the history of deformation of Be lithosphere. While crustal deflation is well-recorded in the struck development of extensional basins and passive margins, and subsidence can be deduced from the sedimentary record, the history of lithospheric deformation is much more difficult to recover from the structural and strangraphic record Magmai~sm is the direct response of mantle uplift induced by extension or rifting and is reflected In the volcanic rocks at the surface and In seismic images of intrusive complexes. Hence it is a . . . . . . primary guide to lithospheric processes that occurred during rifting. , , . Clearly the formation of continental margins by rifling and magmausm is one of die primary processes that has shaped the Earth, and we need to~develop a strategy: to study in detail a process that operated on such a large scale and Hat is often not directly accessible to us. While it is certainly very unportant to continue to investigate mature continental margins directly with seismic techniques and sampling, it is suggested that a major, long- tenn commitment be made to He study of active rifts and juvenile continental margins where subaerial exposures and relatively shallow driDing might permit an evaluation of the along-strike and across-strike variability of magmatism and its temporal evolution. The Gulf of Aden -- Red Sea -- East African Rift System is one logical study area because of the clear progression from continental rifting to early seafloor spreading. Another study 236

area should address the effects of variable rates and mechanisms of extension prior to Me initiation of seafloor spreading For example, the largely arnagmanc Bay of Biscay (Montadert et al., 1979; LePichon and Sibuet, 1981) army have formed by extensional mechanisms much like those presently recognized in the Basin and Range Province of the western U.S. The role of magmausm in the Basin and Range (Morgan and Golombek, 1985; Wernicke et al., 1987) and margins that may have evolved similarly (Lister et al., 1986) should also be investigated. A comprehensive, interdisciplinary study of large areas of the Earth's surface is extremely ambitious but may well be necessary if we are to better understand the magi, Nazism associated with the ruin, process. An this regard, the evolution of thinking on magmausm at m-ocean edges provides important lessons in We invesuganon of similar global-scale tectonic features. Only recently has Be patchwork of diverse studies including nearly thee decades of geophysical surveys and basalt geocheniical analyses become interpretable in terries of a mode] for structural and magmatic segmentation of m~- ocean ridge spreading centers (e.g. Langmuir et al., 198~6~. This is Be result of systematic surveys conducted at appropriate scales on Be East Pacific Rise. Different scales and styles of segmentation are observed with wavelengths as large as lOOO's of kilometers to as small as lO's of kilometers. While many questions concerning spreading center processes stall remain, the concentrated effort along Be East Pacific Rise has yielded results that now form the framework for future studies. Recent studies of the East African Rift System have also begun to document se~,rnentanon on similar scales (Figure 5; Rosendahi, 1987; Dunkelman et al., 1988; Karson ant! Curtis, in press). Interdisciplinary studies have recently resulted in a global co:Tela~on of ocean ridge basalt cheesy why axial depth and coastal thickness resulting i rom temperature variations in the subsolidus mande (Klein and Langm~ir, 19879. Similarly, Be effects of vananons in the mantle geothenn should be apparent in Be amount and composition of voicanics emplaced dunng rifting (White, 1988; McKenzie and Bicker, in press) and early seafloor 237

! STRUCTURAL ARCHITECTURE OF THE - - ~ ! b EAST AFRICAN RIFT ZONES l a MAIN ELEMENTS OF THE =1 '/ '`` l EAST AFRICAN RIFT SYSTEM I ~ ^~ — ~ I l E l ! ; x't~'cTc~ ~ ~ L ~ . ~ | ~ ~DE. I ,~E. =,e,,, C ~ 1~— 1 O ME—Weft- OFF GIG C ~ le ~ ~ ~ / · c,7 twit—Ca73 r ~ ~ ~ ~ ~ ~ l ~ ~ ~.csr~ ~ PA Oo ~1 ~ I ~ . /. _ ~ . | ~ BOXY D ED ;~C ~ ~~ ,~ Ooze ~t \t ~ ! aft Ads 1 ~ "hi, _ ~ 34\ `4~'t.~uc~ ~,'.5 f ~ oot ~ ~1m o ~~1~ - , "" ~ r o ~ ~ ~ ~ 1 ~ ~ i 4/ ~ ~ t~J - 4~4 ~ ~ ~ ~'~ ~ ~ k: ; ji I /~ n - CCEsiAN ~h,E_S 1l I f ° . ~\ ~ ~? ~ :~ I _ ,~,47 ^ - 1 Figure 5. Stn~ctural segmentaiion of ~e East AfYican Rift System occurs at a range of scales. a Segmentadon occ~g at the 1000 km scale compnses m~ajor d~xnal ~lifts and the segments between domes. The few 100 km scale includes rift zones and disconiimiides (~ple junciions, pplay ~ult terminations, strik~sl~ faults, and pre ems~g struc~es). b. At the few 101an scale ~re rift basins (full-gr~en basins and half-g~aben basins) "d lir~ing s~c~res (tsnsfonn faults lin~g othea di~rem riR segments, o~er s~e-slip faults d~at may be related to pre- exisong faults, complex zones of obliq~e-slip faulting often refe~d to as accommod~on zones, and magmadc constructional zones such as dike swa~~ns, plueons, etc.). At ~e few hn scale (not shown) are indi~ndual fault s~nds a~ad volcanos, and secdons of the rift floor and walls, etc. Prom Rosen~ahl, 1987. 238

spreading (McKenzie and Bickle, in press; Zehnder en al., submitted) and should alit: w ar evaluation of nft-:nduced ma ,mnansm, prox~nity of hot spots, and the evolution of subcontinental lithospher~c mantle (Perry et al.? 19871. Using the success of m-ocean Edge studies as an analogy' a systematic approach to the study of exposed, or pliably exposed rifts seems well justified. It is emphasized Hat such a program can succeed only with a commitment to developing muli~disciplin~tv geochem~cal, geological and geophysical surveys that are aimed at detemun~ng the nature and scale of major processes in these environments. Studies made to Ate bring to mind several questions involving crust-mantIe interactions during rifting that should be accessed by future programs. These questions fan under one or more of the following general headings: A. Mechanisms and rates of crustal extension, associated mantle upwelling and the effects of mantle temperature variations on nfung processes I. Is the vananon In magmauc activity associated with conunentai nfung and passive margin formation due to the rate and degree of extension, proximity lo a hot spot, or a combination of these and other as yet undetermined factors? 2. what is the role of magmanc cons~ucuon at depth (underplaying, intrusion beneath voIcanos, lateral dike injection' etc.) ? Can extension at the Earths surface be accommodated at Kept by magmauc intrusion with little piashc deformation of the pre-exist~ng crust and upper mantle county rocks (~;igv~e 2; Karson and Curtis, in press) ? 3. What defines the Moho; how do we measure crustal thicknesses in presently extending rifts and quiescent passive margins, and how can we distinguish deformed continental crust from uplifted highly defonnec} pre-r~ft mantle, residual lower crystal rocks depleted by partial melting, layered mafic and ul~amaf~c cumulates, or some comb~nai~on of these (Af~mendinger et al., 1987; Cans, 1987) ? A. How are the Mung of subsidence and magmatism related in magmancally active areas (Baker et al., 1972; White et al., 1987) ? As a spreading plate boundary becomes well established, at what point do intrusive and extrusive activity cease to involve the rifted mares, n? Does magmai~c activity help to increase the density of extended crust during rifting and therefore increase the potential for eruption of denser ma~,rnas with time (Cox, 1980) ? This has been suggested on theoretical grounds and is supported for some areas of the Basin and Range Province, but may be best tested where deeper levels of the crust can be examined. 239

6. How does back-arc extension and related ma~,mauc acuity compare with m~d-conunent rifting and passive margin development., Can models acr~vou from back-arc basins be applied to passive margin formation in general? What dis~anct geocheTn~smes am evident In ma~=,rnai~sm due to back arc extension, and what infom~anon does this reveal for He upper mantle in these regions? Since many immature ocean basins are of the back-arc ~e, an assessment of the similani~es and differences between Weir development and the development of passive margins In general would be a useful constraint in process evaluation for both systems. · ~ A A 1 _ I _ _ ~ ~ 7. To what regional extent cart asymmetric lithosphenc extension occur? For example, the Exmouth Plateau appears to have undergone asymmetric extension with upwell~ng of the mantle offset nearly 600 hen from the initial region of extension (Mutter et al., in press). Could the volcanism evident in the slighUy-r~fted Eastern Branch of the East African Rift System result from a similar asymmetric extension involving the western amagma~c branch (Figure 3) ? Would voicanics Mom asymmetric rift systems be expec~d to have distinct geochemical signatures Tom symmetric rift voicanics? 8. To what extent can nft-related magmausm be explained by passive extensional processes before invoking active mantle upwelling? 9. What factors In addition to partial melt layer thickness (such as lithosphere thickness and relative viscosity (Whitehead et al., 1984), amount of extension, rate of extension, previous history of the lithosphere, and position relative to hot spots) condor the mande upweD~ng geometry? Regional geochemical studies In conjunction with surface geology arid geophysical surveys may show important correlations. B. The relationship between tectonic segmentation and magrnai~c segmentation . How are magmatic and tectonic segmentation related, if at all, at various scales? 2. At what scale are individual magmanc segments bed to discrete mantle sources? Individual voIcanos? Clusters of nearby volcanic centers? Entire rift zones? 3. In the earliest stages of rifting and In rifts wad only small amounts of volcanism, are voicariic eruptions controlled by major extensional andlor strike-slip faults or are major extensional structures controlled by magma~ac infusions into the crust (Karson and Curtis, in press) ? 4. Are there "~ansfonn-edge effects" (fracuonanon or melting) at some structural disconunuities and not others, and can this be related to the geometry of rifting? C. The progressive depletion of subconunental lithosphenc mande I. Is there a discrete mantle Tesenoir blat is tapped doing early rifting, and if so, why? How and at what stage of evolution is the primary oceanic mande source 0'REMA) tapped? 240

2. How does rift volcano chemistry vary wad time? Is this related to the distnbunon of magma~asm and structural segmentation along the nits? Can a progressive trend in magma composition be associated ~~ continued extension in a rift system? How are highly fractionated ~,oicanics formed, and what volume of residual cumulate matenal is required In the crust beneath rifts? SOME SUGGESTIONS FOR FUTURE RESEARCH Future research that could provide important insights into these problems would involve multidisciplinary studies carried out both along and across presently extending rifts as well as.passive continental margins In apparently different Pragmatic and tectonic settings. In order to better understand magma~ac passive margins, amagmaiic margins will also need to be considered The type localities for these kinds of studies should be chosen to address the research questions of greatest importance to the continental margin workshop members and their colleagues. High pnonty should be given to He development of a detailed sampling of Quaternary basalts along the axis of He East African Rift -- Red Sea Rift System. About half of this area is exposed subaerally in Kenya and Ethiopia. Sampling along several axial segments of the Red Sea axis could be accomplished by a combination of dredging and drilling. The basalts from both localizes should be analyzed for major, trace, and rare each elements as wed as various isotopic systems to fully evaluate the mantle and magmauc evolution of this region as well as to define magmatic cells and Weir relation to structural rift basins. This sampling should be accompanied by detailed geologic mapping and geophysical surveying of the axial portions of rift traverses. Surveys should be designed to define the current locus of rifting as well as He rift segmentation that has developed during the past few minion years. The internal structure of the crystalline crust is of particular interest because of the potential importance of intrusive bodies. Surveys should also be carried out along strike of the rift system to evaluate the temporal evolution and variability in basaltic volcanism by extensive radiome~c age dating in andiron to He previously mentioned techniques. 241

similar smbies should be made In the Basin and Range Province of the western U.S., and should include studies of metamorphic core complexes. In addition, an "underplated" igneous body such as those identified In multichannel seismic reflection lines (Alimendinger et al., 1987) should be drilled. Also, the relationship between Recent magmausm and extensional structures needs to be dete~Truned. In parallel with the above studies, drilling Reverses along a flow line of oceanic crust immediately adjacent to and seaward of contend margins should be castled out proximal to magmanc, amagmaiic, and back-arc margins to determine the mantle evolution associated with the development of an ocean basin in different tectonic settings. These should be accompanied by multichannel refiecuon and refraction, geopotennal, and heat flow surveys that extend landward into Me rifted margins. Studies of rift geometry and melt emplacement should allow for an evaluation of crusts de ~nteracuons resulting In We production of a particular volume of mande partial melts of a paIiicular geochemistry. In this way, the prevailing tempera Ore structure and elemental abundances for the mantle underlying a rifting margin will also be revealed. Along strike geophysical transects should also be done to determine any segmentation patterns Rat are known from both the lift stage blat these areas evolved through and the mad spreading systems that formed from ~em. 242

REE;ERE~TCES Allmendinger, R., Hauge, T. A., Hallser, E. C., Potter, C. J. and Oliver J., 1987. Tectonic heredity and He layered lower crust in the Basin and Range Province, western United States. In: Coward, M. S., Dewey, J. F. and Hancock, P. L., eds., Continental Extensional Tectonics, Geol. Soc. Lond., Spec. Publ. 2g, 223-246. Baker, B. H., 1987. Outline of the penology of the Kenya Rift ALkaline Province. In: Fitton, J. G. and Upton, B. G. J., eds., Alkaline Igneous Rocks, Geol. Soc. Lond., Spec. Publ. 30, 293-3 ~ 1. Baker, B. H., Mohr, P. A. and Williams, Lo. A. J., 1972. Geology of the Eastern Rift System of Africa. Geol. Soc. Amer., Spec. Paper 136, 67 p. Bohannon, R. G., 1986. Tectonic configuration of the western Arabian continental margin, southern Red Sea. Tectonics 5, 477-500. Donate, E. and Seyler, M., 1987. Crustal Underplating and Evolution In He Red Sea Rift: Uplifted Gabbro/Gneiss Crustal Complexes on Zabargad and Brothers Islands. ]. Geophys. Res. 92, 12803-12822. Brooks, C. K. and Nielsen, T. F. D., 1982. -~e E. Greenland continental margin: a ~ansii~on between oceanic and continental magmatism. ]. Geol. Soc. Lond. 139, 2 6 5 27 5 e Brown, :~., Gephan, J., Hauser, E., Latham, T., Nelson, D. and Potter, C.' 1986. Seismic layering in Continental Crust; EOS Trans. 67, 1095 Buck, W. R., 1986. Small-scale convection induced by passive rifting: The cause for uplift of rift shoulders. Earth Planet. Sci. Lett. 77, 362-372. Cheadle, J. J., McGeary, S., Warner' M. R. and Matthews, D. H., 1987. Extensional structures on He western UK contend shelf: a review of ewdence from deep seismic profiling. In: Coward' M. P., Dewey, J. F. and Hancock, P. L., eds., Continental Extensional Tectonics, Geol. Soc. Lond. Spec. Pub. 2S, 445-466. Cox, K. G., 1980. A mode} for flood basalt-volcanism. ]. Petrol. 2l, 629-650. Dunkelmann, T. J., Karson, J. A. and RosendahI, B. R., 1988. Structural style of the Turkana Rift, Kenya. Geology 16, 258-261. Cans, P. B., 1987. An open system, two-layer crustal stretching mode] for the eastern Great Basin. Tectonics 6, I-12. Hart, W. K., W.-Gabr~el, G., Aronson, J. L., Walter, R. C., Mertzman, S. A. and Westgate, J. A., 1987. Evolution of magic volcanism in He Main Ethiopian Rift (MER). Geol. Soc. Amer., Abstracts with Programs 19, 695. Hauser, E. C., Barnes, A., Gephart, I.' Latham, T., L`undy, T., Brown, L. and Oliver, J., 1986. COCORP Deep Reflection Transect in Arizona: Across the Transition Zone from Colorado Plateau to Core Compleses. EOS Trans. 67, 1096. Hinz, K., Mutter, J. C., Zehnder, C. M. and NOT Study Group, 1987. Symmetric conjugation of continent-ocean boundary structures along the Norwegian and East Greenland margins. Mar. Petrol. Geol. 4, 165-~87. Karson, J. A. and Curtis, P. C., in press. Teconic and magmatic processes in the Eastern Branch of He East African Rift and unplicanons for magrnancally active continental nits. In: Rogers, J. J. W., RosendahI, B. R. and Rach, N. M., eds., Rifting in Africa-Karoo to Recent, Jour Africal Earth Sci., Spec. Vol. Keen, C. E., 1987. Some important consequences of lithosphenc extension. In: Coward, M. P., Dewey, J. F. and Hancock, P. L., eds., Continental Extensional Tectonics, Geol. Soc. Lond., Spec. Publ. 2S, 67-74. Klein, E. M. and Langmuir, C. H., 1987. Global Correlations of Ocean Ridge Basalt Chemistry with Axial Depth and Crustal Thickness* ]. Geophys Res. 92, 8089-~l 15. 243

Klemperer, S. L., Hauge, T. A., Hauser, E. C., Oliver, J. E. and Potter, C. J., 1986. The Moho in the northern Basin and Range province, Nevada, along the COCORP 40 N seism~c-refleci:on transect. Geol. Soc. Am. Bull. 97, 603-61g. Lan~mu~r' C. H., Bender, J. F. arid Balboa' R., 1986. Serological and tectonic segmentation of the East Pacific Rise, 5°30'-14°30'N. Nature 332, 422-429. LASE Study Group, 1986. Deep Structure of the U.S. East Coast margin from large aperture seismic experiments MASER. Mar. Petrol. Geol. 8, 234-242. LePichon, X. and Sibuet, J. C., 1981. Passive margins: a model of formation. ]. Geopl2ys. Res. X6, 3708-3720. Lippard, S. J., 1973. The penology of phonolites from the Kenya Rift. Lithos 6, 217-234. Lister, G. S., Ether~dge, M. A. and Symonds, P. A., 1986. Application of the detachment model to the fo~Tnai~on of passive continental margins. Geology 14, 246-250. Macconald, R., 1987 QuatemaIy peralkaline silicic rocks and caldera volcanoes of Kenya. In: Fitton, J. G. and Upton, B. G. J., eds., Alkaline Igneous Rocks, Geol. Soc. Lond.' Spec. Publ. 30, 313-333. McKenzie, D. and Bickle, M. J*, in press. The volume and composition of melt generated by extension of the lithosphere. ]. Petrol. Mohn, P. A. and Wood, C. A., 1976. Volcano spacings and lithospher~c attenuation in the Eastern Rift of Africa. Earth Plasnet. Sci. Lett. 33, 126-144. Montadert, L., Roberts, D. G., Decharpal, O! and Guennoc, P., 1979. Rifi~ng and subsidence of the northern continental m~gm of He Bay of Biscay. In: Animal Reports of the Deep Sea Drilling Project Letg 4S, 1025-1060' U. S. Gov. Printing Off., Washington, DC. Morgan, P. and Golombek, M. P., 1984. Factors controlling the phases and styles of extension in the northern Rio Grande Rift. In: Baldridge, W. S., Dickerson, P. W., Riecker, R. E. and Zidek, J., eds., Rio G ran de Rift: Northern New Mexico. New Mexico Geol. Soc., 13-20. Muner, I. C., Buck, W. R. and Zehnder, C. M., 1988. Convective partial melting I: A mode] for the development of thick igneous cmst dming the initiation of spreading. ]. Geophys. Res. 93, 1031-1048. Mutter, J. C., Larson, R. L. and IOWA Study Group, in press. Extension of the Exmouth Plateau, offshore northwestern Austria: Deep seismic reflection/refraction evidence for simple and pure shear mechanisms. Geology. Mutter, J. C., Talwan:, M. and Stoffa, P. L., 1984. Evidence for thick oceanic crust adjacent to the Norwegian margin. ]. Geophys. Res. X9, 483-502. Mutter, J. C. and Zehnder, C. M., 1988. Deep crustal structure- magmai~c processes: the inception of seafloor spreading in the Norwegian-Greeniand Sea. In: Morton, A. C. and Parsons, L. M., eds., Early Tertiary Volcanism and the Opening of the NE Atlantic. Geol. Soc. Lond. Spec. Publ.. 39, 35-48. Perry, F. V., Baldridge, W. S. and DePaolo, D. J., 1987. Role of Asthenosphere and Lithosphere in the Genesis of Late Cenozoic Basaltic Rocks From the Rio Grande Rift and adjacent Regions of the Southwestern United States. .7. Geophys. Res. 92, 9193-9214. RosendahI, B. R., 1987. Architecture of continental rifts with special refernce to East Africa. Ann. Rev. Earth Planet. Sci. 15, 445-503. Royden, L., Sciater, J. C. and von Herzen, R. P., 1980. Continental margin subsidence and heat flow; important parameters in formation of petroleum hydrocarbons. Am. Assn. Petrol Geol. Bull. 64, 173-~87. Sengor, A. M. C. and Burke, K., 1978. Relative timing of rifting and volcanism on Earth and its tectonic implications. Geophys. Res. Lett. 5, 419-421. Thompson, G. and McCarthy, J., 1986. Geophysical Evidence for Igneous Inflation of the Crust in Highly Extended Terranes. EOS Trans. 67, ~ I S4 244

Wernicke, B. P., 1986. The Basin and Range Moho, 5-10 sec. reflectivity, and a simple physical model for continental rift magma~asm. EOS Trans. 67, ~ ~ 84 Wernicke, B. P., Christiansen' R. Lo., England, P. c and Sonder, L. J., 1987. Tectonoma~,manc evolution of Cenozoic extension In the North American Cordillera. hi: Coward, M. P., Dewey, J. F. and Hancock, P. L., eds., Continental Extensional Tectonics. Geol. Soc. Fond. Spec. Publ. IS, 203-222. White, R. S., 1988. A hot-spot for early Tertiary volcanism in the N Atian~c. In: Morton, A. C. and Parson, L. M., eds., Early Tertiary Volcanism and the Opening of the NE Atlantic, Geol. Soc. Lond. Spec. Pub. 39, 3-13. White, R. S., Spence, G. D., Fowler, S. R., McKenzie, D P., Westbrook, G. K. and Bowen, A. N., 1987. Magmatism at rifted continental margins. Nature 330, 439-444. White, R. S. and McKenzie, D. P., In press. Magmatism at Rift Zoners: The Generation of Volcanic Continental Margins and Flood Basalts. ]. Geophys Res. Whitehead, J. A., Ir., Dick, H. J. B. and Schouten, H., 1984. A mechanism for magmatic accretion under spreading centers. Nature 312, 146-148. Williams, L. A. J., Macdonald, R. and Chapman, G. R., 1984. Quarternary caldera volcanoes of the Kenya Rift Valley. ]. Geophys Res. X9, 8553-~570. Zehnder, C. M., Mutter, I. C., and BuhI, P., submitted. Deep seismic and geochemical constraints on the nature of rift-induced magmatism during breakup of the N Atlantic. Tectonophysics, Spec. Pub. . 245

Passive Continental Groins Rip AND PASSIVE MARGIN USE THE SEOI~ WARD 246

INVESTIGATING THE SEDIMENTARY RECORD: - Sequence Stratigraphy The record! of tectonism and the global ocean environment Joe} S. Watkins Departments of Geophysics and Oceanography Texas A&M University College Station, TX 77843 INTRODUCTION - The confirmation of gIoba~y synchronous sea le~rel changes has been one of the most exciting scientific events of the past decade. The effects of eustatic change have revolution- ized interpretation of marine seismic data from sedimentary environments. The pervasive and widespread! interaction of global sea level changes with other marine and atmospheric phenomena is unique. Sea level changes have been shown to correlate with changes in the CCD (Berger' 1970), oxygen isotopic composition (Miller et al., 1987), carbon isotopic composition (Woodruff and Savin, 1985), fauna] productivity and distribution (Berggren and HoDister7 1974), silica diagenesis (Berger, 19741` global climate {Fischer and Arthur. ~ ~ __\ ~ . ~ ~ . ~ TO q' ~ ~ ~ 7 =- lStt'9, deep ocean circulation Keener aunt warren, 1983), seismic reflectors in carbonate sediments (Mayer et al., 1985), and preservation of organic carbon (Arthur et al., 1984). It is often not clear whether these processes are the direct result of sea level changes or are caused by phenomena that also affect sea level, but it is clear that we have an opportu- nity to study an important scientific problem and a family of associated problems of great . · ~ slgulucance. Continental and insular margins form the principal library of data relating to past rises and falls of sea leered. Outer shelves ~d upper slop es, both present and past, are of particular significance On the inner shed, deep erosion accompanying sea level low- stands can erase the record of past sea level cycles, and Merle and lower slope deposition is somewhat insensitive to perturbations of sea level. Deposition on the outer sheE and upper slope, on the other hand, is not only sensitive to sea level changes, but depositiona] sequences tend to be thicker and contain more information about nondepositional environ- mental changes associated with sea level changes. Other marine libraries, particularly the creep sea basin sediments, also contain important sea level related data, as, for example, those associated with chemical and temperature changes, which the oxidizing environment and stronger currents on the outer shelf and upper slope tend to destroy. Resolution of several fundamental problems wiD greatly enhance our understanding of sea level changes. For example, (~) global correlations of several major bounding un- conformities are not convincing, (A amplitudes and precise timing of globe] rises and fads in sea level are poorly determined, (3) systems tracts have not been described for many 247

depositional environments, (4) investigators have questioned the attribution of important unconformities to sea level changes, noting that currents and changes in rates of subsi- dence and sediment flux can also create unconformitites, (5) slope and basinal systems tract characteristics are poorly determined, (6) interrelationships between global sea level change? climate, oceanic circulation, ocean chemistry, and sediment flux are poorly known, and (7) the underlying mechanism responsible for sea level cycles with periods of 5-50 m.y. is unknown. Sequence stratigraphy is the method of choice in many investigations of glob e] sea level change. Sequence stratigraphy records a multidimensional matrix with dimensions of Ethology, climate, tectonic regime, sediment flux, and other parameters which are useful in the study of global sea level changes and associated phenomena. Sequence stratigraphy is a relatively recent development, deriving from post-1970 seismic stratigraphic concepts. The study of stratigraphy dates from William Smith's publication of geologic maps in 1815* but stratigraphy remained landlocked for a century and a half because the quality of seismic reflection data was not adequate for stratigraphic interpretation. Digital acquisition, digital processing and better seismic sources developed during the 1960s and 1970s provided the tools for the development of sequence stratigraphy. Improved data, the Exxon database, and the inspired work of a few Exxon researchers led to publication of a comprehensive review of E~xonts work on seismic stratigraphy in the American Association of Petroleum Geologists Memoir 26 (Vail and others, 1977). Concepts described in Memoir 26 by Exxon and other geologists gave rise to a remarkable flowering of subsequent studies and reports. Stratigraphic sequences are unconformity-bounded depositional sequences or cycles in which certain facies or systems tracts (Brown and Fisher, 1977) can be identified. In the marine environment, they are identified through the study of seismic reflection data, well logs? paleontologic data, and, rarely, cores. System tracts are functions of water depth, a factor that makes them especially useful for the study of changes in sea level. Seismic reflection data, well logs, cores, and paleontologic data each contribute uniquely to marine sequence stratigraphic studies, but seismic stratigraphy dominates. This is partly due to a large global data base, but is also due to the method's power in defining stratigraphic sequences. In the jargon of marine geophysics, seismic stratigraphy and sequence stratigraphy are often considered a single entity. The following paragraphs review the sequence stratigraphic matrix and suggest a strategy for determining values in the empty portions of the matrix that will contribute to the resolution of the broader problems of global sea level change and associated phenomena. 248

MA:IOR PROBLEMS IN SEQUENCE STRATIGRAPHY Problems in sequence stratigraphy fall into three interrelated and overlapping cate- gories: i. Definition of stratigraphic sequences 2. Global (and non-giob~) sea level changes, and 3. Related environmental changes (e.g., climate, ocean circulation, etc.~. The following reviews the main elements of these three categories. Definition of Stratigraphic Sequences Figure ~ (Vail, 1987, van Wagoner, 1987) shows a hypothetical composite depositional sequence formed during a single sea level highstand-lowstand-highstand cycle. The figure also shows associated Systems tracts" or major sea level-related facies. From bottom to top, the Lowstand Systems Tract (LST) consists of fan sequences (sf and bf) and a lowstand wedge complex superimposed on an extensive erosional unconformity (SBl) and its correlative non-depositional surface on the slope and basin floor. The LST Is overlain by the Transgressi~re Systems Tract (TST) deposited during a relative rise in sea level, and a Highstand Systems Tract (MST) deposited during a relative sea level highstand. The SheK Margin Systems Tract (SMST) resembles the LST but differs in that it was deposited during a slow rise in relative sea level or subsequent stiDstand whereas the LST was deposited dunug fading sea level or subsequent stilistand. SMSTs and LSTs usually consist in part of sediments eroded from shoreward unconformities. SMSTs and LSTs are similar in many respects because of similar depositional histories. The similarities in SMSTs and LSTs illustrate two major problems in sequence stratig- raphy. First, sea level changes observed in seismic data are relative and not absolute. Vail et al., (1977) strongly emphasized this point, but it appears little considered in many inves- ligations. Second, systems tracts are a function of both sea level Egg and sediment flux rates (Pitma~ and Golovchenko, 1983, SchIager and Camber, 1986~. These characteristics create arnbigmties in seismic interpretation of sequences. Heavy reliance on seismic reflection data for the investigation of sequence stratigraphy creates a further problem in that seismic data acquired in the same location but with different acquisition parameters or processed ~ different ways may differ significantly in detail. This is particularly serious when data acquired ten or more years ago are compared with contemporary data. Legitimate differences in interpretation of systems tracts may result. The problem is most-acute in data from continental margins. Avmiability of data Tom outcrops reduces the problem to some extent in lade] data. There is no absolute right and no wrong in selection of acquisition and processing parameters. Seismic imaging remains part art and part science. However, a few mutually agreed-upon standards in acquisition and processing wiD eliminate much of the problem without sacrificing image resolution. 249

Figure 2 shows the hypothetical section shown in Figure 1, but with the vertical axis converted from depth to geologic time. Figure 2 implies a continuity Or deposition in the sequence. That is, sediments eroded in one location are deposited in another at any given time. The locus of redeposition, however, may be far from the locus of erosion. For example, Poag and Ward (1987, Figure 3) show that turbidity (?) currents have extensively eroded the slopes offshore New Jersey at DSDP site 612 and offshore Ireland at DSDP site 548 with the result that lengthy periods are missing from the record. The locus of redeposition during hiatuses is probably out on the aby-ssal plain. My work on the upper slope of the northcentral Gulf of Mexico suggests extensive turbidity current erosion during periods of falling sea level. Considerable work will be required to bring models illustrated in Figures ~ and 2 into agreement with reality. It is usually assumed that erosional unconformities indicate sea level lowstands, but this is not necessarily true. Popenoe et al., (1987) have show-e that the Guh Stream flowed across northern Florida during several Tertiary highstands creating an extensive erosional unconformity. We do not know hour many other unconformities result from shelf currents. Identification and dating of slope nonconformities is a non-trivial matter requiring extensive site surveying as wed as driving. Paleontologic dating of shelf and slope sediments may require riser driDing insofar as these zones are prime sites for the accumulation of hydrocarbons. In summary, present depositional sequence models are generalized and do not ad- equately portray stratigraphic signatures expected in many environments. We need to know how- sequences formed on rapidly subsiding margins differ from those formed on slowly subsiding margins. Rapid subsidence is often associated with young margins and high sedimentation rates, whereas slow subsidence is associated with mature margins and slower sedimentation rates. Sequences and systems tracts are likely to differ significantly from one of the above depositions environments to the other Sequences developed in basins underpinned by subsiding continental crust, e.g., the North Sea and East Texas Salt Basin, appear to differ from those underpinned by subsid- ing oceanic or greatly thinned continental crust. Stratigraphic signatures in these environ- ments are poorly investigated. Sequences and systems tracts in carbonate and siliciclastic regimes are known to differ in a number of important Revs. Aspects of sea level rise and fall may be much better preserved in carbonate than in siliciclastic rocks. but as yet stratigraphic signatures in these environments are poorly documented. Several environments. notably- those associated with Tertiary deltas and those con- taining extensive mobile salt deposits? appear to have distinctive sequence and system tract signatures. The importance of these environments to the oil industry adds urgency to their investigation. 250

The effect of sediment-flux rate on the creation of unconformitites and the extent of current-clerived, non-Iou~stand unconformities need further investigation. Widespread unconformities resulting from flux rate variations or highstand current scour could alter our perception of sequence stratigraphy. Standardization of acquisition and processing procedures for seismic reflection data is highly desirable. Comparative sequence stratigraphy can be quite difficult using data sets collected and processed with different acquisition equipment and processing parameters. Fin any, comple~nentary outcrop studies are valuable in the investigation of offshore sequence stratigraphy. Global Sea Level Changes The concept of global or eustatic sea level changes is a powerful tool, central to sequence stratigraphy. Sequences bounded by global unconformities can be correlated with coeval sequences elsewhere and provide an overall view of a regional or global depositional environment at an instant in geologic time. A series of global "snapshots,' can provide insight into global tectonics and evolution during the period covered. Four main problems with our knowledge of global sea level changes need to be ad- dressed. They are: I. Global nature of 26- and 3d~rder cycles, 2. Cycle timing and duration, 3. Cycle amplitudes, and 4. Eustatic cycle mechanisms. Global nature of 26- and 3d~rder cycles. - A large body of evidence, too abundant to reference here, indicates that first~rder (T > 50 may.' T period) cycles, some second- order (5 > T > 50m.y.) cycles, and third~rder (T c 5 m.y.) cycles of glacial origin are real. Pitman (1978) showed that changes in the volume of the ocean basins during the opening of the South Atlantic were consistent in amplitude and timing with a first order highstand in sea level that peaked during the Cretaceous. Similarly? Awarder glacial cycles are consistent in amplitude and timing with independent estimates of ice volume. Miller et al., (1987), Keigwin et al., (1986) and Shackleton et al., (1984) find oxygen isotopes in forami~fera indicative of ic~growth events at 35, 3l, 25, 13-15, 10, 5, and 2.4 Ma. These dates are consistent with sea level drops deduced by Haq and others (1987) from coastal onlap studies. The giobad nature of other 3d~rder and many 2d~rder cycles remains to be demonstrated. For example? Figure 4 compares paleodepth data Tom the Arabian Peninsula (Harris et al., 1984) with data Tom the Western Interior of the U.S. (from Kauffman, 1977) and with the globe] synthesis of Vail and others (1977~. Figure 5 shows a similar time frame Tom the latest Vail curve (Haq et al.? 1987~. Sharp sea level drops in the Turonian-Coniacian and in the Aptian correlate reasonably wed on Al curves. A major fall in the Maastrichtian is evident in Harris et al.'s (1984) sea level curves and Haq et al.'s global curves but not in the Western U.S. and Vail global curves. A major 251

Valanginian sea level fall appears in the Vail and Haq curves but not in the Kauffman and Harris curves. The Vail-Haq curves are difficult to evaluate because they were developed using pro- prietary Exxon data which has not yet been made public. This is a serious problem because many Haq global correlations appear to be at or below the threshold of biostratigraphic resolution, and it is impossible to confirm the validity of their correlations. The North Atlantic bias of the \7ail-Haq curves may account for some discrepancies in correlation with other data (see flail and others. 1977? for their data distribution). Hubbard (1988) examined this issue in his study of data from the Santos Basin in the western South Atlantic, the Grand Banks of southeastern Canada, and the Beaulort Sea in the Arctic Ocean. His results are shown in Figure 6. Hubbard concludes that his data do not support synchronous worldwide sequence boundaries resulting from periodic, short-term eustatic falls in sea level, and that global similarity in ages of sea level fans may be an illusion created by similarities in ages of basins studied. Hubbard's work emphasizes the need for careful studies in widely geographically sep- arated parts of the globe, a major recommendation of COSOD-I! (JOlDES, 1987~. Cycle timing and duration. - Many of the difficulties in correlation mentioned in the preceding section result from insufficient precision in dating. Hubbard (1988) gives age uncertainties of +0.5 m.y. to +8 m.y., with an average uncertainty of +2.5 m.y. Typical late Tertiary age uncertainties were ~ 1 m.y., whereas typical Jurassic age uncertainties were 05 m.v. Haq et al. (1987) identified 26 eustatic sea level falls in the Paleogene, or an average of one per 1.65 m.y., a value at or below the threshold of resolution in much biostratigraphic data. More accurate age dates are obtainable in some time periods. Recent advances in strontium 87/86 dating in Tertiary carbonates suggests accuracies ranging from 0.3 to 3.0 m.y. Aubry (198S) has used magnetostratigraphy to document the lengths of hia- tuses in northwestern Europe with accuracies of 0.1 m.y. Miller et al. (1985) have com- bined biostratigraphy, magnetostratigraphy and isotopic stratigraphy to refine the dating of Oligocene and Miocene sediments in the western North Atlantic. Overall, accuracies of 0.2 m.y. or better should be obtainable in many Tertiary sections. Further improvements in accuracy may be possible by Milankovich cycle correlation. Heckel (1986) has suggested a correlation between Milankovich cycles, cyclothems and eustatic cycles in Pennsylvanian sediments from the mid-continent of the U.S. and Fischer (quoted in JOIDES, 1987) has correlated Coniacian-Santonian cycles between Wyoming and Germany with an apparent resolution of 20-40 k.y. If Milankovich cycles are found to be widely applicable to eustatic cycle dating, ages accurate to less than 100 k.y. can be attained. The best opportunities for adequate precision in dating seem to be in the Tertiary where biostratigraphic, magnetostratigraphic and Milankovich cyclography are applicable. 252

The problem is more difficult in Cretaceous and Jurassic sediments where biostratigraphic intervals are less precise and where magnetostratigraphy and isotope stratigraphy are of limited value. Cycle amplitudes. - Measurement of cycle amplitudes during the past decade has relied mainly on coastal onlap measurements from seismic and wed log data. Resolution of seismic data is usually in the range of 15 to 20 m. Resolution of well bore data is variably better or worse depending on the density of weds. A second source of error is the interpreter's ability to pick the exact time of the highstand andior lowstand. And, oniap must be corrected for subsidence, basin tilt, compaction, and converted to sea level elevation. These factors combine to create significant errors in cycle amplitude measurements. Recent amplitude data obtained Tom subsiding atolls in the Pacific (Halley and Lud- wig, 1987, Lincoln and Schlanger, 1987, I,udwig et al., 1988, and Major and Matthews, 1983) indicate that minimum amplitude estimates can be determined from atoll carbon- ates. Figure 7 (JOlDES, 1987, data from Haq et al., 1987, and lIalley and Ludwig, 1987) compares cycle amplitudes from atoD measurements with coast oniap estimates. The length of each arrow represents the thickness of a limestone unit deposited during rising sea level. Dashed lines represent the long term subsidence of the atoll and distances be- tween arrows represent hiatuses due to subaerial exposure of the carbonates (from Ludwig and Halley, 1987). The solid line is the eustatic sea level curve of Haq et ~1., (1987). Uncertainties in purr plitude and dating appear to be significantly -less in these data. If this is true, coastal onlap estimates of sea level elevations are significantly in error. Timing agreement between curves is consistent within expected errors in biostratigraphic dating. G. Baum (pers. comm., 1988) reports that coring in continental margin carbonates has produced results similar to those from atolls. - - In summary, the tools appear to be at hand to solve the amplitude problem in the equatorial carbonate margins. Dating of siiicicIastic sediments remains more difficult with comensurately greater uncertainties. Eustat~c ~cle mechanisms. - Lone period chances in eustatic sea level are general thought to be caused by changes In the volume of the ocean basins as in the case of the South Atlantic described by Pitman (1978~. Subduction is also capable of producing Ist- order eustatic changes, although of smaller amplitudes. Calculations show that subduction in the Greater Himalayas should have lowered sea level approximately 20 m during the past 40 Ma. Inclusion of Anatolia, the Alps, and the Andes would lower sea level approx- imately 32 m. Hag et al., (1987) estimate from coastal onIap studies that sea leYe] feD approximately 100 m during this time. We have no way of knowing whether the difference results from errors in the Haq curve or from the contribution of unrecognized mechanisms. Although a~nplitucles associated with subduction and oro~env are comparable to those predicted bv coastal onian {Han et al.. 19871. c, ,, ~ ¢__~__ ___ _- ~ _ ______¢ ~_~~ _~ 7 ~ 77 and the onset in some cases is roughly equivalent to that of inferred eustatic cycles, rates of fad are too small and periods too long to account for inferred 26- and 36-order eustatic cycles. 253

The eustatic curves in Figure 5 show, marked differences between sea level rises and sea level fads. The sea leered fads tend to be sharp, higher ~.mp~itude and shorter in period. These characteristics, if not an artifact of coastal oniap measuring technique, place strong constraints on the driving mechanism. Confirmation of the reality of these differences between rising and falling sea level is urgently required. _ _ , Polar ice is the principal mechanism capable of changing the volume of water in the oceans over relatively short periods of time (Pitman and Golovchenko, 1983~. While changes in ice volume are demonstrably capable of producing 26- and Awarder eustatic cycyles, 26- and Awarder cycles have been inferred during periods when the earth appears to have been ice free. Cloetingh et al., (1985) and Cloetingh (1986) have presented a mode} suggesting that intraplate stresses during plate collison, fragmentation or reorganization at convergent plate boundaries are sufficient to cause sea level changes of 100 m at rates of ~ cm/k.y. There is some evidence to support this model. The initiation of subduction of the Hi- malayas began approximately 40 m.y. ago (Mattaur? 1986), a time which correlates with the onset of rising sea level on the Haq et al. (1987) long-per~od curve. A decrease in the spreading rate in the Indian Ocean from IS cm/yr to 5 cm/yr accompanied the plate reorganization (Pierce, 1978~. A period ofincreased convergence also beganin the Alps 40 m.?. ago and lasted until about 12 m.y. ago (Hsu, 1979, Milnes, 1978). Approximately 12 m.y. ago, the Arabian subcontinent coDided with Anatolia to form the Zagros Moun- tains (Dewey et al., 1986~. A change in the long period sea level curve (Haq et al., 1987) coincides with the collision but the signal is mixed with that of the Neogene glaciation. Watkins et al. (1987) reported possible correlations between terrane collisions and sea leered excursions between 235-200 Ma, 140-125 Ma, and Il2-89 Ma. Discovering the mechanism responsible for non - laced 26- and Awarder sea level changes is probably the most-important issue in the a~ea-of sequence stratigraphy. Con- vincirlg models with predictive capabilities of global amplitudes en cl durations of sea level changes would provide a foundation for future investigations of sea level change and se- quence stratigraphy. Related Environmental Changes The multidimensional aspects of sequence strati~raphy challenge the investigator while .. . ... -—~ ~ ~ ~ ~ simultaneously prowling a means of independent testing and confirming hypotheses. A sedimentary sequence records the effects of sea level change (both global and local), cli- mate, sediment flux, ocean circulation in the region?-ocean chemistry, planetary wobbles, margin tectonics, global tectonics and other parameters. For example, Mayer et al., (1985) correlated a single reflector in a Pacific basinal seismic sequence with a carbonate rnin- imum? a widespread hiatus, a major sea level fluctuation, and a significant increase in Pacific silica deposition. 254

Other reflectors at the site were correlated with hiatuses, sea level drops, faunal changes, isotopic shifts? carbon depletion, and a subsea erosional event. Arthur and Jenkyus (1981) correlated phosphorite genesis with sea level changes and climate, Arthur and Dean (1986) correlated vanations in the sediment carbon budget with sea level changes, and Prakes and Bolton (1984) correlated sea level regression on the Australian margin with giant manganese deposits. The panorama of sea level-related changes is sometimes ciaunt- ing, but always exciting. A special note should be directed toward the investigation of Milankovich cycles. Milankovich cycles cannot be resolved by present-day standard multichannel redirection seismology. The technology is available to construct acquisition equipment capable of re- solving Milanko~rich reflections where depositiona] rates are sufficiently great. Penetration is limited to the uppermost few hundred meters but data from this zone could be quite useful when used in conjunction-w~th hydraulic piston coring. Subsidence rates are a major source of uncertainty in both sequence analysis and eu- stasy. Contemporary subsidence models are largely one-climensionaI; that is, they predict the subsidence at a single site or weld The lack of a good tw~dimensional forward model results in part from imperfect understanding of the rifting process. The recent discovery of cletachment surfaces beneath some rifts add associated asymmetric faulting on the sicles of rifts may provide a basis for better models. The extent and diversity of phenomena affected by and aBecting sea level changes is too large to discuss in detail. But clearly, the opportunities in this area are maIiifoicI. 255

SUGGESTIONS FOR FUTURE INVESTIGATIONS The number of problems related to sequence stratigraphy that need attention greatly exceeds the scope of this paper. Some of these have been mentioner! above. The following is a bme£ fist of a few- broader operational en c] scientific steps that need to be taken. I. Selection of geographically distributed "field laboratories" for comparative studies has the highest priority. Three or four locales in equatorial or Tow temperate latitucles could best serve as initial study areas. These areas should be selected to pronde as nearly complete Tertiary carbonate sequences as possible and good transects across margins to stucly facies variation. As knowledge of sequence stratigraphy increases, higher-latitude, silicicIastic field laboratories can be added. 2. A first~rder objective of sequence stratigraphic investigations should be the confir- mation of global sea level events. Uncertainties in the area of global vs. regional unconformities is a serious problem. Equally important is the determination of precise amplitude, age and duration of global and major regional changes in sea levels. It is also important to determine if the absence of highstand "spikes" is real. Objective testing of mechanistic models responsible for 28- and 36-order sea level changes are not possible until these matters are resolved. 3. A high-priority objective is the clarification of the systems tracts concept. The concept needs to be rigorously tested, applied and described for a number of clifferent litholog~c and tectonic environments. Utilization of industry and government expertise along with academic expertise is highly desirable. Costs of driLing and biostratigraphic work can be reduced through in- dustry participation. Industry will benefit equally from the joint effort. 4. With respect to the management of the program, it is suggested that a mllIti-national, multi-institutional, multi-disciplinary committee (including of] industry members) be organized to coordinate the investigations. Such an organization can bring combined expertise to bear on a wide range of problems and in a wide range of field! laboratories. 256

REFERENCES CITED Arthur, M. A., Dean, W. E., and Stone, D. A. V., 1984, Models for the deposition of Mesozoic-Cenozaic fine grainecI orgamc-Carbon-rich sediment in the deep sea; in Stone, D. A. V. and Pipers, D. J. W., errs., Fine-grained sediments: Deep water processes and Facies: Geol. Soc. London Sp Publ. 15, p. 527-559. Arthur,M. A., and Dean,VV'. E.? 1986, Cretaceouspaleoceanographyof the western North Atlantic Ocean; in Vogt, P. R., and Tucholke, B. E., errs., The Geology of North America, Volume M, the Western North Atlantic Region: Geol. Soc. Amer. Arthur, M. A. and Jenkyns, H. C., 1981, Phosphorites and paleoceanography: Oceanol. Acta, Proce 26th Intl. Geol. Congress, Geology of Oceans Symp., p. 83-96. Aubry, M.-P., 1985, Northwestern Europe Paleogene magnetostratigraphy, Calcareous nan- nofossil evidence: Geology' v. 13, p. 198-202. Berger, W. H., 1970, Planktonic foraminifera in selective solution and the lysocline: Mar. Geol., v. 8, p. 111-183. Berger, W. H., 1974, Deep sea sedimentation; in Burk, C. A. and Drake, C. L., eds., The geology of continental magrins: Spunger-Verlag, New York, p. 213-214. Berggren, W. A. and Hollister, D. C., 1974, Paleogeography, paleabiogeography and the history of circulation in the Atlantic Ocean; in Hay, W. W., ea.? Studies in Pale~ Oceanography: SEPM Sp. Publ. 20, p. 12~186~ Brown, L. F. Jr., and Fisher, W. L., 1977, Seismic-stratigraphic interpretation of depo- -sitiona] systems: Examples from Brazilian r~ft-and pull-apart basins; ~n Payton, C. E., ea., Seismic stratigraphy - Appli;cation to hydrocarbon exploration: Am. Assoc. Petr. Geol. Mem. 26, p. 214-248. Cloetingh, S., McQueen, H., and Lambeck, K., 1985, On a tectonic mechanism for regional sealevel variations: Earth and Planet. Sci. Ltrs., V. 75, p. 157-166. Cloetingh, S., 1986, Intraplate stresses: A new tectonic mechanism for fluctuations of relative sea level: Geology, v. 14, p. 617-620. Dewey, J. F., Hempton, M. R., Kidd, W. S. F., Saroglu, F., and Sengor, A. M. C., 1986, Shorten~ng of the lithosphere: the neotectonics of eastern AnatoHa - a young coDision zone, in Coward, M. P., and Ries, A. C., eds., Collision Tectonics: Am. Geol. Soc. Amer. Sp. Pub., no. 19, p. 3-36. 257

Fischer, A. G. and Arthur, M. A., 1977, Secular variations in the pelagic realm: SEPM Sp. Publ. 25, p. 19-20. Fisher, A. G., 1986, Climatic rhythms recorded in strata: Ann. Rev. Earth and Plan. Sol., v. 147 p. 315-376. Frakes, L. A., and Bolton, B. R., 1984, Origin of manganese giants: Sea level change and anoxic~xic history: Geology, v. 12, p. 83-86. Halley, R. B. and Ludwig, K. R., 1987, Disconformities and Srisotope stratigraphy reveal a Neogene sea level history from Enewetak Atoll, Marshall Islands, Central Pacific: Geol. Soc. Amer. Absts. with Frog., v. 19, p. 691. Haq, B. U., Hardenbol, J., and Vail, P. R., 1987, Chronology of fluctuating sea levels since the Triassic Science, v. 235, p. 1156-1167. Harris, P. M., Frost., S. H., Seiglie, G. A., and Schneidermann, N., 1984, Regional un- conformities and depositional cycles, Cretaceous of the Arabian Peninsula; In Schiee, J. S., ea., Interregional unconformities and hydrocarbon accumulation: Am. Assoc. Petrol. Geol. Mem. 36, p. 67-80. Heckel, P. H., 1986, Sea-level curve for Pennsylvanian eustatic marine transgressive- regressive depositions cycles along m~dcontinent outcrop belt, North America: Geol- ogy, v. 14, p. 330-334. Hsu, K., 1979, Thin-skinned plate tectonics during neo-Alpine orogenies: Am. Jour. Sci., v. 279, p. 353-366. Hubbard, R. J., 1988, Age and significance of sequence boundaries on Jurassic and Early Cretaceous rifted continental margins: Am. Assoc. Petrol. Geol. Bulk, v. 72, p. 49-72. JOIDES, 1987, Scientific goals of an Ocean Drilling Program designed to investigate changes in the globe environment: Report of the Second Conference on Scientific Ocean Drilling, COSOD-II: p. 15-46. Kauffman, E. G., 1977, Geological and biological overview, Western Interior Cretaceous Basin: Mountain Geologist, v. 14, p. 75-99. Keigwin, L. D., Aubry, M. P. and Kent, D. V., 1986, North Atlantic late Miocene stable- isotope stratigraphy, biostratigraphy and magnetostratigraphy: Init. Repts. Deep Sea DriLing Project, or. 94, 935-963. 258

Keller, G. and Barron, J. A., 1983, Paleoceanographic implications of Miocene deep-sea hiatuses: Geol. Soc. Amer. Bull.. fir 94, p 590-613. Lincoln, J. M., and Schlanger, S. O., 1987. Miocene sea level falls related to the geologic history of Midway AtoD: Geology, v. 15, p. 454-457. Ludwig, K. R., Halley. R. :B., Simmons, K. R. and Peterman, Z. E., 1988, Strontium- isotope stratigraphy of Enewetak Atoll: Geology' v. 16, p. 173-177 Majors R. P. and Matthews, R. K ~ 1983, Isotopic composition of bank margin carbonates on Alidway Atoll: Amplitude constraints on post~arly Miocene eustasy: Geology, v. 11, p. 335-338. Mattaur, M., 1986, Intracontinental subduction, crust-mantle decoDement en c] crustal wedge stacking in the Himalayas and other collision belts; in Coward, M. P. and Ries, A C., eds., Collision Tectonics: Geol. Soc. Amer. Sp. Publ., no 19, p. 37-50. Mayer, L. A., Shipley, T. H., Theyer, F., Wilkens, R. H., and Winterer, E. L., 1985, Seismic modeling and paleoceanography at Deep Sea Drilling Project site 574; in Mayer, 1~. A. and Theyer, F., eds., Initial Reports of the Deep Sea Drilling Project: U. S. Gout. Pnuting Off., v. 85' p.. 947-969. Miller, K. G., Aubry, M.-P., Khan, M. J., Miller, A. J., Kent, D. V., and Berggren, W. A.7 19857 Oligocene-Miocene biostratigraphy, magnetostratigraphy, and isotope stratigraphy of the western l\i-orth Atlantic: Geology, v. 13, p. 257-261. Miller, K. G., Fairbanks, R. G., and Mountain, G. S., 1987, Tertiary oxygen isotope synthesis, sea level history, and continental margin erosion: Paleoceanography, v. 2, no. 1, p. 1-9. Miines, A. G., 1978, Structural zones and continental collisions, Central Alps: Tectono- physics, v. 47, p. 369-392. Pierce, J. W., 1978, The northward motion of India since the Late Cretaceous: Geophys. Jour. Roy. Soc. Astr. Soc., v. 52, p. 277-311. Pitman, W. C., III, 1978, Relationship between eustacy and stratigraphic sequences of passive margins: Geol. Soc. Amer. BuB., v. 89, p. 1389-1403. Pitman, W. C., III, and Golovchenko, X., 1983, The effect of sea level change on the shelledge and slope of passive margins: SEPM Sp. Publ. 33, p. 41-58. 259

Popenoe,-P., Henry, V. P., and Idris, F. M., 1987, Gulf Trough- The Atlantic Connection: Geology, v. 15, p. 327-332. Poag, C. W., and Ward, L. W., 1987, Cenozoic unconformities and depositional superse- quences of North Atlantic continental margins: Testing the Vail model: Geology, v. 15. p. 159-162. Schlager, W. and Camber, O., 1986, Submarine slope angles, drowning unconformities, and self erosion of limestone escarpments: Geology, v. 14' p. 762-765. Shackleton, N. J., EIall, M. A., and Boersma, A., 1984, Oxygen and carbon isotope data Tom Leg 74 forami~fers: Init. Repts. Deep Sea Drilling Project, v. 74' p. 599-612. van Wagoner, J. C., Mitchum, R. M. Jr., Posamentier, H. W., and Vail, P. R., 1987, Key definitions of seismic stratigraphy; in Bally, A. W., ea., Atlas of Seismic Stratigraphy: Am.- Assoc. Petrol. Geol. Studies in Geol. #27, v. 1, p. 11-14. Vail, P. R., 1987, Seismic stratigraphy interpretation using sequence stratigraphy; in Badly, A. W., ea., Atias of Seismic Stratigraphy: Am. Assoc. Petrol. Geol. Studies in Geol. #27, a. 1, p. 1-10. Vail, P. R., Mitchum, R. M. Jr., Widmier, J. M., Thompson, S., Sangree, J. B., Bubb, J. N. and Hatlelid, W. G., 1977, Seismic stratigraphy and glob e] changes of sea level; In PaytoIl, C. E., ea., Seismic stratigraphy - Application to hydrocarbon exploration: Am. Assoc. Petr. Geol. Mem. 26, p. 49-212. Watkins, J. S., Wilson, K. M., and Hay, W. W., 1987, Does eustatic sea level reflect terra ne histories? Geol. Soc. Amer. Absts. with Prog. Woodruff, F. and Savin, S. M., 1985, Delta C-13 Rues of Miocene Pacific benthic foraminifera: Correlations with sea level and biological productivity: Geology, v. 13, p. 119-122. 260

ILLUSTRATIONS Figure 1. The systems tracts model (Vail, 1987). The sequence comprises Al sediments deposited between the lower highstand unconformity (SB 1) and the upper high- stand unconformity. The sequence includes a] sediments deposited during a complete highstand-highstand sea level cycle. Figure 2. The depositional sequence model from Figure 1, replot ted as a function of geologic time rather than dentin (Vail. 19871. Figure 3. Slope hiatuses at DSDP Site 612 offshore New Jersey and at DSDP Site offshore Ireland (from Poag and Ward, 1987~. Figure 4. Comparison of paleodepth data from Arabia, the U. S. Western Interior, and glob e] sea level curves. Prom Harris et al. (1984~. Figure 5. Cretaceous global sea level changes. From Eaq et al. (1987~. Figure 6. Correlation of major unconformities in the Beaufort Sea (Arctic Ocean), Grand Banks (Northwestern Atlantic Ocean) and the Santos Basin (Western South Atlantic). From Hubbard (1988). Figure 7. Comparison of sea level curves from coastal onlap (Haq et al., 1987) with sea level data from Enewetak (Ludwig and Halley, 1987~. 261

1- Hlasa- ~ 1 I'm - ~ 11 ~ it At ARC a_ LL — In _ z r r \ _ In X a) Q E o C) At C, J (' ~ C7 ~ Z - = ~ Hi ~ ~ ~ ' 0 ~ I c 0 3 0 3 ~~~ en ~~ ,, C I ~ .~ O .= ,l o O ~ _ it 11 11 3 Al ll 1~ he—., In In In IL O ~ ,, 11 3 a _—C', '0 m m `,, In ~ a) a) Cot Cot CO . ~ ~ 0 0 cn cn c' ~ o o [D {~) ~ o x Q Q ~ ° E o o ~ _ 11 11 11 ~ ~ X __ ~ _ ~ ._ IL

ir o ~ al 1:5 z ~ ° m 0 ~ { LL a) ~ U) o z o LL in LL en z LL if o . ; 9 1 ~ I E 3 Wll DIOG1039 ~ ~ 1: m *! 1 zo 1 Z ~ . . 263 \ cn1 ~1 _ 1 ~ l CO ma' in_ in m ~n cn I ~: ~: LL m CO CE o 11 o C: z ll 1 -~ o lL z o . U] i ~ 1 1 1 ~ · CC o - Ct { o m a' ~r cn ea E ~' J.LI ~— 1.O ~ ~_ 1~: ° 1 - o 1 cn C3 - m C~ ~_ lL

DSDP 612 _ USGS ISEAcH COST COST MA AGE No.1 S2H2-tlL B-2 B O PLEIST ; 10~ 1 1 I?? of =~ ~ ~ -20 30 U1 1 1 . -40 ~ ~ ~ 501z~ ~~ L ~ /?~? - ~ 60 1 0 ~~tT —70 MAEST. ~2~ ~ ~ ~~: 80 SANT. ~ ~ ~ ~ ? MVAAJoR DSDP UNCOHF. _ ~ ~ ~ww~ ,`~~_~ w`~,w. ~,~~ ~ `.,w~ wit Devonian Basement Fig. 3 264

~ 1 In ` A o, 2 _ at: ~ [Ui_ 0 #,C Carl IIJ t- Z Z ~ Us C] ~ IS O ~ W ~Y, C, Z Z== _ Cat ~ Mama AS—lU ·~m HI — Cal CO ~,, l v /v~ ~'~zZO O O r _ I i, 1 _ Z Z zO~ C ~ ~ - > Z Z Z. ~ NYIW0003N 31~1 ..31GGlW,. A18~3 snow 1 1 1 ~ I 1 ~ I I ~ ~ l I ~ I (D O O O O ' O O O O I ~ ~ 1 1 ~ I I ~ 1 1 1 1 1 1 1 265 - . AL

70 75 80 85 90 95 100 105 1 1 0 30 33 ~4 C30 C31 C32 a: LL AL Q en o LL 111 66.5 MAASTRICHTIAN A: z o z 111 An - 74 CAMPANIAN 84 89 SANTONIAN 88 CONIACIAN - 92 - TU8ONIAN CENOMANIAN ALBIAN 106 APTIAN _ _ LONG TERM ~—SHORT TERM 70 75 80 85 90 95 100 105 110 Fig. ~ 266

SANTOS BASIN ~ ]U At= -70- . {D E —80— ~ ' _90_ :~~ -110 =, -120 _ ~ a -l60- cO =~ =6 -1BO  o 1 90 _= 200 _ ~ e 65 A 21 0- - —7" a: ·_ En .> co co m Q _ .= co 'co co co Q ;~ co _ not present .= _ co co .L C: - Fig. 6 267 GRAND BAN KS BEAUFORT SEA _ _ .~ CO CO Cal Q _ r .= co .m ecu .= _ not mapped 52 ~ tl9__ Ol °2 _ .Q, _ _ ~~ 5~: ~ 128 _ _ ~ 220/ - . ~

- ~ - cn ~ - ~ At: a) I co a) , A o a) Q In ~ _ I ~ fit: — ~7 o o o o · Call ~ o In a' . Ct _. / ~ . I ~—/ 1 ./ C W; ~ //; A' lo Cal _ ~ - '\ ~ in , i' , o id J U] . ~ LL a: LL In O O . O O O CD Cal Beaus 268 O O at Cal 1 1

-POST RIFTING EVOLIJ~IION OF PASSIVE MARGIN BASINS Dale S. Sawyer Department of Geology and Geophysics Rice University P.O. Box 1892 Houston,TX 77251 (713)285-5106 Introduction Rifted or Impassive" continent margins are sites-of tremendous sediment accumulation. After their deposition, sediments are far from passive and undergo a broad spectrum of physical and chemical changes. The chemical changes include diagenesis, metamorphism, hydrocarbon generation, and interaction with or contributions to the seawater system. In addition to the obvious economic importance of these phenomena, the influence of passive margins on the global geochemical balance is perhaps large, but largely unknown. The physical environment of the sediments, including pressure and temperature, along with the mechanics of the sediments, control physical properties changes and sediment tectonics. Crustal processes such as subsidence rate and isostatic response to sediment loading also affect the conditions in passive margin sediments. The proposed Continental Margin Workshop is an opportunity to examine the interactions among these diverse phenomena and devise strategies for studying ~em. Physical Processes Fluid flow in rocks affects diagenesis, mineralization, metamorphism and tectonics in basins. While this is well known, studies of fluid circulation have been largely confined to looking at the effects of fluid circulation rather than directly observing it. Fluid flow is 269

probably the primary means of redistributing elements in passive margin basins. Mass balance calculations in the Gulf of Mexico Coast region pertaining to such diverse sedimentary phenomena as sandstone cementation Wand and Dungeon, 1979), formation of sediment hosted mineral deposits (Price and others, 1983), development of secondary porosity (Schmidt and McDonald, 1979) and carbonate loss from muds tone ~undegard and Land, 1986), an require large amounts of matenal and energy transfer up through the sediment column. Fluid flow basins produces chemical fluxes into the oceans* Although not as intense as the fluxes at m~- ocean ridges, those at margins may be greater. The character of fluids entering the sea water from passive margin basins is likely to be quite anomalous and variable. Biological systems, including those associated with hydrocarbon seeps and cold saline seeps, can be affected or even owe Weir existence to fluid c~ulanon (Bright et ale 1980; Kle~nschmidt and Tschauder, 1986). Mechanisms of fluid flow in passive margin basins include compaction, differential loading, hydrothermal convection, and gravitational flow of meteoric and saline water. Compaction produces an upward flow of water when permeabilities permit, but leads to the formation of ove~pressure when water camlot escape. The thickness and physical properties of sediments deposited in passive margin basins are often laterally heterogeneous leading to significant lateral fluid flow. Under appropriate circumstances hydrothermal circulation may serve to move large quantities of water and heat through sediments. This may pareiaDy explain puzzling results from diagenetic studies that require tremendous amounts of water, much more than can be attributed to compaction dewatering, to deposit or remove dissolved elements. Water flow in passive margin basins also results from density differences between pore fluids such as meteonc and saline water. The migration arid accumulation of oil arid natural gas from source regions to resenro~s, or probably more frequently to escape into the ocean or atmosphere, are poorly understood. Migration and accumulation are probably at least partly a result of water circulation Trough sediments of varying porosity and permeability. Migration can be constrained using 270

geochernical signatures that in some cases allow sources to be identified and pathways determined. A better understanding of hydrocarbon migration win have significant economic value as wed as improve our knowledge of the hydrogeology of margin basins. A massive database for addressing these problems lies in the oil industry and is currently mostly unavailable to academic researchers. Isostanc response of the margin lithosphere affects the s~augraphy md configuration of sediments. To study other processes, including subsidence and sea level variation, it is important to be able to remove the effects of the isostatic response. The isostatic response at a margin is a manifestation of the rheology of continental, extended continental, and ocean crust. Of those 3 types of crust, it is only in oceanic crust that we have much understanding of the isostatic response. Each type of crust or lithosphere flexes when loaded. The wavelength and amplitude of Me flexurE vanes spatially In a basin and wad time. Subsidence In a marginal basin is the result of tectonic processes initiated when the margin formed and We isostatic response to sediment loading. We are usually interested in separating the sediment loading induced subsidence from the tectonic subsidence by observing the sediment loading history and predicting its isostatic response. The sediments on margins have served as recorders of the subsidence history. Tectonic subsidence history may be used to constrain me mechanisms Mat formed the margin. Two end-member models of continental rifting and subsequent passive continental margin formation are currently popular. In the first, the pure shear model (McKenzie, 1978 for example), conjugate passive margin subsidence is predicted to be symmetrical (although I suspect that this is false; Dunbar and Sawyer, 1988). The second, the simple shear model (Wernicke, 1981, 1985; Lister et al., 1986), predicts that subsidence on conjugate margins will be highly asymmetrical. Subsidence studies, along with seismic reflection smbies, will play a role in distinguishing between these models. Temperature is an important control on the rates of many chemical reactions in margin basin sediments ~opatin, 1971). Cooling of the lithosphere causes its contraction which is a principal cause of tectonic subsidence in margin basins. Rock physical properties are sensitive 271

to temperature. Present temperature can be measured by drilling. Present surface heat flow can be measured. Paleotemperatures can be estimated by observing the progress of temperature sensitive chemical reactions. Most often these give us values of the integral of a function of time and temperature rather than temperature directly. Temperature distribution within passive margin sediments is largely a function of thermal conductivity, permeability, porosity, and supply of heat to the bottom of the basin. In basins with low permeability sediments the temperature distribution may be controlled by heat conduction alone. When fluids circulate freely, however, the temperature distribution may be completely controlled by thermal convection. This or other mechanisms are required in many basins to move energy upward faster than conduction should allow. The physical properties of margin basin sediments and the lithosphere below influence literally all of the geophysical observations: gravity, geoid height, seismology, magnetics, well logs, and physical processes we seek to understand: fluid flow, heat flow, deformation, fracture, sediment compaction, and etc. The physical properties of sediments change, in some cases dramatically, as they are compacted and/or chemically altered during burial and aging. Knowledge of the history of these changes is critical to understanding of every other physical process we discuss here. Physical properties are observed or inferred from studies of surface samples, well samples, well logs, seismic experiments, and potential field observations. Distinct physical properties are then linked by, often empirical, mathematical relations. Often porosity and Ethology are used as variable parameters in these relations. Then porosity can be linked though Ethology to depth of burial and then incorporated into geodynamic models to make a suite of testable or useful predictions about processes. Sediments are often deformed or faulted after deposition. Further, they develop cracks on a variety of scales that influence fluid flow. Growth faulting due to differential loading is an important class of fault. These are common where sediment deposition is locally rapid. The dynamics of growth fault formation are poorly understood. The role of fluids in lubricating the fault plane and the use of fault planes as conduits are also unknown. Salt mobility is common 272

In margin basins. It takes several forms including the formation of pillows, diapirs, walls and sills. The intricacies of diapir growth are becoming better known due to physical and numerical model studies and observation. The stress environment in margin basins is unknown although methods now exist to make measurements In wells. Chemical Processes The generation of oil and natural gas in margin basins is an economically important process. Hydrocarbons are generated by heating kerogens, biological products deposited sufficiently quickly or in anoxic seas. Within bounds, time and temperature can be interchanged to achieve a particular level of kerogen maturation (Lopatin, 1971; Wapples, 1980). Important questions remain about the affect of other chemicals on hydrocarbon generation, the types of kerogen source beds and the hydrocarbons they may produce, the relations between time, temperature and the many ways to measure hydrocarbon source maturity. We must understand the mineralogical, chemical and textural changes that sediments undergo with increasing burial, fluid pressure and temperature. This will improve prediction of the distribution of hydrocarbon reservoir rocks. It win allow us to build quantitative models of diagenes~s. The magnitude of inorganic chemical fluxes into the ocean through passive margin basins are largely unknown. The weathering cycle is particularly important in establishing the chemistry of the oceans. Since the bulb of the solid products of weathering end up in margin basins, it is possible that significant interactions exists Some Key Questions What mechanisms condor the hydrogeology of passive margin basins? What are the nature and relative importance of proposed mechanisms of sediment ove~pressuring? 273

How does the lithosphere at passive continent margins respond to sediment loading? What are He dynamics and kinematics of salt diap~nsm? What are the present temperatures in passive continental margin basins and how important is hydrothermal circulation in influencing He temperature field? How are sediments of different Ethology modified by burial and chemical interaction in passive condiment margin basins? What are the chemical fluxes into the global seawater system at passive continental margin basins? How, when, and where do hydrocarbons mature in passive continental margin basins? How and when do hydrocarbons migrate in passive continental margin basins? What are He mass and energy balances In evolving passive continental margin basins? Suggestions for Future Research The types of methods that will be required to approach these problems are as diverse as the problems themselves. A combination of seismic reflection methods and driving will be required to establish the structural and s~atigraphic Stonework of particular marginal basins. In the case of some marginal basins, such as the Gulf of Mexico or North Sea basins, large quantities of geophysical data are available. Data are quite sparse in most other areas. Although large numbers of exploratory wells have been drilled, because of the methods used to drill them, relatively few high quality scientific studies were, or can be, performed in them or on samples from them. Deepening or reusing exploratory wells, although it sounds economical, is rarely a viable approach because He bottom hole diameter is usually too small to allow further casing and drilling. Most of the questions of sediment chemist and chemical flux can only be addressed using uncontaminated samples of rock and pore fluid. This is not usually possible if conventional industry drilling practice has been employed. In some cases these problems will require the installation of long term downhole systems to monitor temperature, pressure and allow fluid sampling. Methods of studying margin hydrogeology 274

include determination of patterns of fluid circulation, diagenetic patterns of continental margin sediments, and nature of deposits formed by sea floor seeps. These observations should be incorporated into hydrogeological and geochemical models of greater sophistication than are available today. It is likely that such work will be most productive if pursued in a few basins where ~e" most data am available, the Gulf of Mexico and Norm Sea. Smaller study areas within the basins should be the subject of new data acquisition aimed at determining the seafloor fluxes of fluids and chemicals. Drilling will eventually be required but is not useful until more survey work is complete. The principal means of study of the isostatic response of the lithosphere under passive continental margin basins involve comparing observations of gravity, geoid height, topography, sediment density and distribution, crust thickness, density and li~ology' and subsidence history' using geodynamic models. lleferences Because He range of subjects to be addressed In this document is so great, ~ drew heavily on ideas presented ~ previous workshop documents. ~ particular ~ used the report of a DOSECC sponsored! workshop on Ul~adeep Scientific Drilling In He Texas Gulf of Mexico Coast and He report of He Second Conference on Scientific Ocean Drilling. Bnght, Tot., Larock, P.A., Issuer, R.D., and Brooks, I'M., 1980, A brine seep at the East Flower Garden Bank' northwestern Gulf of Mexico, Int. Revue gesamt. Hydrobiol. v. 65, p. 535-549. Dunbar, J. A. and Sawyer, D. S., 1988, Continental rifting at preexisting lithospheric weaknesses, Nature, v. 333, p. 450-452. Kle~nschm~dt, M., and Tschauder, R., 1986, Shallow-water hydrothermal vent systems off the Palos Verdes Peninsula, T-os Angeles County, California, Biol. Soc. Wash. Bull., v. 6, p. 485-488. Land, L.S. and Dutton, S.P., 1979, Cementation of Sandstones: reply, J. Sedimentary Petrology' v. 49, p. 13S9-1361. Lister, G.S., Etheridge, M.A., and Symonds, P.A., 1986, Application of the detachment fault model to He formation of passive continental margins. Geology' v. 14' p. 24~250. Lopatin, N.V., 1971, Temperature and geologic time as factors in coaliBlcation (in Russian), Akad. Nauk SSSR {zv. Ser. Geol., no. 3, p. 95-106. 275

L4undegard, P.D., and Land, L.S., 1986, Carbon Dioxide and organic acids: Weir role in porosity enhancement and cementation, Paleogene of the Texas Gulf Coast, SOce of Econ. Paleontologists and Mineralogists Spec. Publication No* 38, p. 129-146. McKenzie, D.P., 1978, Some remarks on the development of sedimentary basins, Earn and Planetary Science Letters, v. 40, p. 25-32. Pnce, P.E., Kyle, I.R., and Wessel, G.R., 1983' Salt dome related zinc deposits, in Kisvarsanyi, G., Grant, S*K., Pratt, W.P., and Koenig, J.W., (eds.~' Int. Conf. Mississippi Valley type lead-zinc deposits, Proc. Vol. Umv. Missoun Rolla' p 558-577. Schmidt, V., and McDonald, D.A., 1979, The role of secondary porosity in the course of sandstone diagenesis, Soc. of Econ. Paleontologists and Mineralogists Spec. Publication No. 26, p. 175-207. Wapples, D W., 1980' Time and temperature in petroleum formation: Application of Lopadnts method to petroleum exploration, American Association of Petroleum Geologists Bulletin, v. 64, p. 916-926. Wernicke, B., 1981, Insights from Basin and Range surface geology for the process of large- scale divergence of continental lithosphere (abstracts, ~ Papers Presented to the Conference on Processes of Planetary Rifting, Lunar and Planetary Institute, Houston, p. 90-92* Wernicke, B., 1985, Uniform-sense normal simple shear of the continental lithosphere, Canadian Journal of Earth Sciences, v. 22, p. 108-125. 276

Next: Appendix: List of Participants »
Margins: A Research Initiative for Interdisciplinary Studies of the Processes Attending Lithospheric Extension and Convergence Get This Book
×
Buy Paperback | $55.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Approximately 70 percent of the world's population is concentrated in the coastal borderlands, which geologists recognize to be the present continental margins. This new book on these continental margins provides a detailed account of a meeting which brought together specialists in marine and terrestrial geology, geochemistry, and geophysics. The workshop garnered widespread support and enthusiasm for a new direction in margins research focused on interdisciplinary studies of the fundamental processes of continental margin evolution. Scientific problems and solutions were identified for both divergent and convergent margins. Results of the workshop show that many of the fundamental plate interaction processes are common to all margins, whether formed by extension, contraction, or translation. This conclusion suggests a unified approach to margins research. A margins initiative has been proposed to follow up on the workshop results by developing science programs aimed at understanding the processes that control the initiation and evolution of continental margins.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!