structure; maximum rooting depth, salinity, and acidity/alkalinity (National Research Council, 1993a, p. 208). The report further recommends the development of "pedotransfer functions" that could be used to link quantitatively the measurement of soil quality to the functions soils perform. Efforts are under way to develop such an index, but these will entail a significant amount of research. National assessments of soil resources are conducted currently, but the kind of data collected and the approaches used to analyze the data (which focus largely on descriptions of soil types and on assessments of soil loss) do not facilitate a comprehensive assessment of soil quality.
The difficulties involved in developing performance standards for water quality are even more daunting than those related to standards for soil quality. Numerical criteria for water quality are set by regulatory agencies primarily to protect human health. Numerical criteria or other indicators related to the function of aquatic ecosystems, however, are not as well developed. Another recent NRC report, Restoration of Aquatic Ecosystems (1993b), stresses the need to develop both structural and functional criteria to assess the success of aquatic restoration projects. The absence of criteria for aquatic ecosystems makes it difficult to develop performance standards for farming and ranching. This problem is compounded by the difficulty in linking pollutants leaving a particular farm or ranch to their effect on the environment. For example, what level of total dissolved nitrate entering a tributary to the Susquehanna River from an adjacent dairy farm threatens water quality in the Chesapeake Bay?
The difficulty of quantifying these linkages has led researchers to propose qualitative standards that could be applied in more systematic ways to farming and ranching systems. The NRC soil and water quality report proposed two such criteria: the efficiency with which pesticides, nutrients, and irrigation water are used in farming systems; and the degree to which farming systems resist erosion and runoff. The efficiency criterion addresses the inputs to the farming system, whereas the resistance criterion addresses outputs. Quantitative measures of input efficiency and of erosion resistance are available and have been incorporated in models that predict the delivery of agricultural chemicals and sediment to groundwater or surface water. These measures are useful for indicating progress toward alleviating environmental degradation by farming, but the question of how much improvement has occurred remains unresolved. Agreement on nonpoint-source control provisions in the reauthorization of the Clean Water Act has been hampered by such uncertainty.
Soil and Water Quality recognizes that the inability to relate changes in farming practices to changes in soil and water quality will ultimately hamper attainment of environmental goals. The lack of quantifiable performance standards