lower-resolution) proxy records of past climate change, and synthesize these data into spatially and temporally resolved reconstructions of climate change in past centuries to millennia.

Develop a best-estimate climate forcing history for the past century to millennium.

Using an ensemble of climate models, simulate the regional and global climate response to the best-estimate forcings and compare to the observed climate record.

Conduct Accurate Long-Term Monitoring of Radiative Forcing Variables

Geophysical quantities relevant to climate forcing should be known with a level of accuracy that is significantly smaller than the expected changes. The current approach relies primarily on measurement repeatability (precision), using overlapping successive measurements to cross-calibrate their absolute uncertainties. The principles described in Karl et al. (1995) (endorsed in Adequacy of Climate Observing Systems; NRC, 1999) provide a suitable framework for guiding collection of observations of radiative forcing and other climate variables. These principles have been updated in the Strategic Plan for the U.S. Climate Change Science Program (USCCSP, 2003).

Ultimately, the measurement accuracies of the geophysical parameters must be tied to irrefutable absolute standards and be tested and validated in perpetuity. Such benchmark measurements of radiative and other climate forcings and climate variables are needed immediately. Because the radiative forcings and the climate responses are highly dependent on wavelength, space-based observations with high spectral resolution are needed to isolate the signatures of the relevant radiative processes and components. Because the forcings and responses that determine any one particular climate state involve a distribution about a mean, the ensemble must be properly characterized and quantified so that changes in the mean can be reliably identified. Ultimately, the specification of the forcings and responses must be integrated to test climate forecast models.

Observational networks for the detection of long-term changes in climate variables must be improved. For example, local land-use changes and vegetation dynamics (i.e., microclimate effects) have been shown for some long-term climate monitoring sites to result in surface air temperature trends that are not spatially representative. Photographic and other documentation of monitoring sites and the surrounding landscape is needed to document the integrity of the sites over time.

Surface and tropospheric heat content changes may provide in the future a robust evaluation of climate changes. Long-term, globally averaged changes in the heat content of the oceans permit the calculation of the



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement