Review of the LimnoTech Report, "Comparison of Load Estimates for Cultivated Cropland in the Chesapeake Bay Watershed"

Abstract

Executive Summary The LimnoTech/ANPC report Comparison of load estimates for cultivated cropland in the Chesapeake Bay watershed analyzed the results of two Chesapeake watershed modeling efforts. The models were the Chesapeake Bay Program's watershed model (the CBP model, which was developed to evaluate actions needed to meet TMDL requirements) and a recently published USDA-NRCS model (CB-CEAP model) developed to quantify the effects of conservation practices applied to cultivated cropland in the Chesapeake Bay watershed. LimnoTech is the consulting firm that prepared the report for its client, the Agricultural Nutrient Policy Council (ANPC), an interest group representing several agricultural trade organizations. LimnoTech reported differences between the CBP and CB-CEAP models and their results, and then recommended suspending implementation of the Chesapeake Bay TMDL until the differences were resolved. The Chesapeake Bay Partnership asked the Scientific and Technical Advisory Committee (STAC, an advisory board for the Chesapeake Bay Partnership) to convene an independent, expert panel to review the LimnoTech report and to make recommendations concerning the application of multiple models in environmental management of the Chesapeake Bay. This report presents the findings of the review committee. The committee concludes that the LimnoTech analyses have poor scientific merit and promote a false set of criteria by which to judge the suitability of the CBP watershed model for use in the TMDL implementation process. LimnoTech based its recommendations on unrealistic criteria for watershed model performance, inappropriate expectations for agreement between watershed models developed for different objectives, selective interpretation of the findings of the CB-CEAP report, and errors in the interpretation of the models and their results. LimnoTech failed to acknowledge that fundamental differences in models (such as the input data, assumptions, and process representations) are unavoidable because of the different objectives of the models and differences in the data and resources available to support each effort. LimnoTech's analysis also ignores the appreciable differences between the models in purpose, history, extent of calibration, extent of validation with independent data, level of spatial discretization, and degree of stakeholder involvement in model scenario development--differences that favor the continued use of the CBP model to inform and guide the implementation of management actions to meet TMDL requirements. When LimnoTech's errors in interpretation of model results are corrected, the results of the two models are more similar to each other than reported by LimnoTech. The corrected results indicate that the model predictions of loads are in approximate agreement despite the differences in model objectives, assumptions, input data, model frameworks, and spatial and temporal details. More importantly, the results of the two models are similar in their assessment of the need for implementing more management practices on cropland. The CB-CEAP model and its supporting data provide new knowledge and approaches that can inform and improve the CBP model and its application to watershed management planning. The review committee commends the ongoing efforts between the CBP and USDA to compare and integrate their data and analyses, and the committee recommends many other activities that could enhance the application of multiple models in managing nutrient and sediment pollution of the Chesapeake Bay (see section on Recommendations for Integrating Models). In summary, the review committee finds that LimnoTech's comparison of the CBP and CB-CEAP models is flawed and does not provide sufficient evidence to suspend implementation of the Chesapeake Bay TMDL.