Modeling fate and transport of indicator microorganisms in small rural watersheds

Modeling is an efficient approach for predicting microbial water quality and suggesting related management practices. Escherichia coli or enterococci concentrations are commonly used to indicate microbial contamination and characterize microbial water quality. Small watersheds provide drainage into first- or second-order creeks, exhibit significant variation in land use, management, and conservation practices. Modeling microbial water quality in the small watersheds can help account for and mitigate the heterogeneity within larger hydrologic response units. A model for microbial water quality should incorporate key hydrologic components such as runoff, in-stream water fluxes, and meteorological inputs such as precipitation, air temperature, and solar radiation. Additionally, animal waste management, including the quantity and application schedule, are also important for microbial water quality simulations. The Agricultural Policy Environmental eXtender is a useful tool for hydrological, meteorological, and management drivers of microbial water quality, as it has been developed for small watersheds. Major microbial fate and transport processes include animal waste deposition, degradation, erosion, survival on soil, release from waste and transport by rainfall or irrigation, and microbial survival and resuspension in water or sediment. These processes can be simplified, for instance, by modeling proportional release of the indicators and animal waste during erosion. We can also use a two-phase survival model for manure and temperature-dependent rate of microbial survival in surface waters. Animal waste aging should also be considered in the microbial model, as daily bacterial survival and erodibility are influenced by it. The microbial module in APEX was used to the headwater watershed of Conococheague Creek in Pennsylvania, USA. The total watershed area is 34321.6 ha, with 15 subareas and the dominant land use is deciduous forest. Three years of hourly stage observations with rating curves and weekly E. coli concentrations at the outlet were available. The primary source of E. coli was animal waste from white-tailed deer, with an average density of 19 deer per square kilometers. Deer population dynamics reflect seasonal changes including fawn births, predation, pre-hunting, and post-hunting population phases. E. coli concentrations at the watershed outlet varied seasonally, ranging from 5 to 500 CFU (100 mL)^-1. The model reasonably captured the temporal fluctuations in E. coli concentrations at the outlet. Ongoing improvements to the model include incorporating deer behavior patterns, animal waste preservation in snow, and runoff during snowmelt.