Factors impacting the variability of effectiveness of agricultural best management practices (BMPs) in Minnesota

In the development of nutrient reduction strategies for agricultural watersheds in the TMDL process, effectiveness values are frequently used from the Agricultural BMP Handbook for Minnesota. However, BMP effectiveness is highly variable due to differences in hydrologic pathways, biogeochemical cycles, landscape position and management. In the updated manual the objective was to describe the primary causes and relative amounts of variability for managers so more realistic TMDLs goals are set. A database of BMP effectiveness studies was used to compare variability of effectiveness values by BMP type, landscape position and target nutrient, nitrogen, phosphorus and sediment, with; n = 17-100. The coefficient of variation (c.v.) was used to compare between studies. There were no significant differences in variability by landscape position or flow path. However there was a significant difference in the c.v of dissolved vs. particulate pollutant efficiency. Dissolved phosphorus had the highest variability with a c.v. of 1.98 for 31 studies while sediment and particulate phosphorus had c.v. values <0.32. The dissolved phosphorus cycle is more complex and it occurs in very low concentrations which creates measurement variability. Also some trapping practices such a wetlands and buffers may release phosphorus under anaerobic conditions. This is problematic for multi-purpose BMPs designed to remove nitrogen and phosphorus. Most agricultural BMPs in the U.S. were developed in response to the soil erosion crisis in the 1930s. Therefore the removal efficiency for many of the newer BMPs involving dissolved nutrients have greater variability particularly for dissolved phosphorus.