Using local monitoring results to inform the Chesapeake Bay Program’s Watershed Model

Abstract

The Chesapeake Bay Program’s Watershed Model (CBWM) has been used as an accounting tool for the Chesapeake Bay Total Maximum Daily Load (TMDL). However, some of the fundamental parameters that underpin the watershed model may not represent local watershed characteristics at all scales. Significant investments have been made by state and local governments, and other local stakeholders, who are interested in validating loads and progress in implementing measures to achieve the pollutant reductions called for in the TMDL through local monitoring data. For the purposes of this STAC workshop, local monitoring is considered any relevant data collected by a local, regional, state, or federal organization that has not been used previously in the development, calibration, or validation of the CBWM. Some of these local monitoring efforts have been collecting data over the past 5-10 years, with some datasets extending back over more than two decades. However, the data and the CBWM are often not directly comparable due to differences in temporal and spatial scales or because the water quality parameters being monitored are not those estimated by the model. Therefore, a Scientific and Technical Advisory Committee (STAC) workshop was convened to bring together Chesapeake Bay Program (CBP) modelers, local and state government stakeholders, and scientists who are monitoring and analyzing local water quality data to recommend ways in which local monitoring data can be used to inform the CBWM, identify gaps between modeled and monitored data, and validate model predictions at the local scale. The workshop, “Using Local Monitoring Results to Inform the Chesapeake Bay Program’s Watershed Model”, was held in March 2023 to provide insight on the scope of local water quality monitoring efforts within and outside of the Bay watershed that could be used to inform the CBWM. Scientists and managers developed recommendations that could be used by modelers for either calibration or knowledge generation to inform the Phase 7 version of the CBWM currently under development for a 2027 decision by the CBP, recommendations for how local monitoring efforts could be designed or altered to better inform the CBWM, and recommendations for how monitored trends could be used in management. The preliminary presentations for the workshop provided essential background information on the CBWM and data used to parameterize it. This information was the foundation for discussions on existing data gaps, the importance of current local monitoring networks, and best practices for developing future monitoring networks. More information on this STAC-funded effort including workshop presentation slides and recordings can be accessed on the workshop webpage. Confidence in the loading estimates of the CBWM is critical because of its role as the accounting mechanism for measuring progress toward the Bay TMDL’s nutrient and sediment reduction goals. Those who are being asked or required to pay for these reductions, from state and local government managers to farmers, property owners and developers, must have confidence in the scientific validity of the CBWM’s loading estimates or trust in the restoration effort will dissipate. Toward that end, several local entities have invested in extensive urban, suburban, and agricultural monitoring programs to characterize nutrient and sediment loading (among other water quality parameters) at a relatively fine scale (from a few acres to 5 square miles). Monitoring networks outside of the Bay watershed were also included as their relevance and similarities to Bay watershed landscapes, hydrology, and climate conditions can help build the body of knowledge necessary for better parameterization of the CBWM. Local monitoring results could be analyzed for loads and trends for calibration of Phase 7, comparison against trends, informing the structure and parameterization of the model, and potentially in policy evaluation. The effectiveness of management practices at the small watershed scale is a primary question of watershed managers that could be addressed by local monitoring, but to do so study design and statistical techniques may need to be altered if these datasets are intended to inform parameterization of the Bay modeling tools. The partnership would benefit from the redesign of some existing monitoring programs so that they are hypothesis-driven, with fully described inputs, outputs, and practices. New statistical tools could be applied to evaluate the relative importance of various drivers affecting water quality and influenced by hydrogeologic setting and watershed condition.