Using the Q10 model to simulate E. coli survival in cowpats on grazing lands

Microbiological quality of surface waters can be affected by microbial load in runoff from grazing lands. This effect, with other factors, depends on the survival of microorganisms in animal waste deposited on pastures. Since temperature is a leading environmental parameter affecting survival, it indirectly impacts water microbial quality. The Q₁₀ model is widely used to predict the effect of temperature on rates of biological processes, including survival. Objectives of this work were to (i) evaluate the applicability of the Q₁₀ model to Escherichia coli inactivation in bovine manure deposited on grazing land (i.e., cowpats) and (ii) identify explanatory variables for the previously reported E. coli survival dynamics in cowpats. Data utilized in this study include published results on E. coli concentrations in natural and repacked cowpats from research conducted the U.S. (Virginia and Maryland), New Zealand, and the United Kingdom. Inspection of the datasets led to conceptualizing E. coli survival (in cowpats) as a two-stage process, in which the initial stage was due to growth, inactivation or stationary state of the population and the second stage was the approximately first-order inactivation. Applying the Q₁₀ model to these datasets showed a remarkable similarity in inactivation rates, using the thermal time. The reference inactivation rate constant of 0.042 (thermal days)^− 1 at 20 °C gave a good approximation (R² = 0.88) of all inactivation stage data with Q₁₀ = 1.48. The reference inactivation rate constants in individual studies were no different from the one obtained by pooling all data (P < 0.05). The rate of logarithm of the E. coli concentration change during the first stage depended on temperature. Duration of the first stage, prior to the first-order inactivation stage and the initial concentration of E. coli in cowpats, could not be predicted from available data. Diet and age are probable factors affecting these two parameters however, until their environmental and management predictors are known, microbial water quality modeling must treat them as a stochastic source of uncertainty in simulation results.