Project I

Integrating Data and Models in the Catskills to Assess Effectiveness of Phosphorus BMP's in the Cannonsville Watershed

Justification:

Best management practices (BMPs) have proven to be a cost effective means of reducing non-point phosphorus (P) loading to surface waters at the field and farm scale. However, very little research has explored BMPs efficiency at the watershed or basin scale. We are currently evaluating the impact of P BMPs in the Cannonsville Reservoir Watershed, New York. The Cannonsville is the largest of the reservoirs that supplies drinking water to New York City, and is at risk of eutrophication due to P loading from the watershed. In many rural watersheds, agriculture runoff is generally acknowledged to be the major source of P inputs to surface water bodies. BMPs are a commonly accepted management tool to reduce P loading to surface water bodies. In the Cannonsville many BMPs have been and are being implemented during the last twenty years. BMPs being implemented on a whole farm basis include crop rotation and tillage, barnyard improvements, manure management and export, nutrient management, fencing, and filter/buffer strips. Since BMPs were implemented there has been a reduction in the P load observed in the Cannonsville reservoir.

Objectives:

Our objective is to determine if implementing BMPs is truly responsible for the observed P load reduction through the use in situ data collection and a suite of water quality models. Field data collection includes groundwater monitoring of P levels in agricultural fields, water table and saturation dynamics in varying topological settings, and various geotechnical methods to delineate runoff contributing areas, such as ground penetrating radar and geophones. The models in use range from the fully distributed, process based model, the Soil Moisture Distribution and Routing model, to lumped models such as the Soil Water Assessment Tool (SWAT) and the Variable Source Loading Function model (modified from the GWLF model to simulate the spatial distribution of runoff producing areas).

Progress to Date:

To date we have collected extensive data on the distribution of saturated and runoff producing areas, the processes controlling terrestrial N and P dynamics, as well as data on the contribution of agricultural areas to ground water P migration. We have also extensively modified/re-conceptualized SWAT and GWLF to correctly capture and model the spatial distribution of runoff and P source areas, allowing us to analyzing the effectiveness of BMPs on the landscape. Results indicate that BMPs, implemented in concert, that protect riparian areas and streams from direct pollutant loading provide the most substantial water quality protection per land taken out of production.

Impact:

The results are expected to provide a more cost-effective means of determining the relative risk of nutrient transport and non-point source pollution for a given management or development scenario. These results will ultimately influence agricultural zoning regulations and public policy.

The Cannonsville watershed. Locations of hydrometeorological stations and special research sites in the basin.