Bacterial contamination of tile drainage water and shallow groundwater under different application methods of liquid swine manure.

A 2 year field experiment evaluated liquid manure application methods on the movement of manure-borne pathogens (Salmonella sp.) and indicator bacteria (Escherichia coli and Clostridium perfringens) to subsurface water. A combination of application methods including surface application, pre-application tillage, and post-application incorporation were applied in a randomized complete block design on an instrumented field site in spring 2007 and 2008. Tile and shallow groundwater were sampled immediately after manure application and after rainfall events. Bacterial enumeration from water samples showed that the surface-applied manure resulted in the highest concentration of E. coli in tile drainage water. Pre-tillage significantly (p < 0.05) reduced the movement of manure-based E. coli and C. perfringens to tile water and to shallow groundwater within 3 days after manure application (DAM) in 2008 and within 10 DAM in 2007. Pre-tillage also decreased the occurrence of Salmonella sp. in tile water samples. Indicator bacteria and pathogens reached nondetectable levels within 50 DAM. The results suggest that tillage before application of liquid swine manure can minimize the movement of bacteria to tile and groundwater, but is effective only for the drainage events immediately after manure application or initial rainfall-associated drainage flows. Furthermore, the study highlights the strong association between bacterial concentrations in subsurface waters and rainfall timing and volume after manure application.

Estimating dissolved reactive phosphorus concentration in surface runoff water from major Ontario soils.

Phosphorus (P) loss from agricultural land in surface runoff can contribute to eutrophication of surface water. This study was conducted to evaluate a range of environmental and agronomic soil P tests as indicators of potential soil surface runoff dissolved reactive P (DRP) losses from Ontario soils. The soil samples (0- to 20-cm depth) were collected from six soil series in Ontario, with 10 sites each to provide a wide range of soil test P (STP) values. Rainfall simulation studies were conducted following the USEPA National P Research Project protocol. The average DRP concentration (DRP30) in runoff water collected over 30 min after the start of runoff increased (p < 0.001) in either a linear or curvilinear manner with increases in levels of various STPs and estimates of degree of soil P saturation (DPS). Among the 16 measurements of STPs and DPSs assessed, DPS(M3) 2 (Mehlich-3 P/[Mehlich-3 Al + Fe]) (r2 = 0.90), DPS(M3)-3 (Mehlich-3 P/Mehlich-3 Al) (r2 = 0.89), and water-extractable P (WEP) (r2 = 0.89) had the strongest overall relationship with runoff DRP30 across all six soil series. The DPS(M3)-2 and DPS(M3)-3 were equally accurate in predicting runoff DRP30 loss. However, DPS(M3)-3 was preferred as its prediction of DRP30 was soil pH insensitive and simpler in analytical procedure, ifa DPS approach is adopted.

Risk assessment of unsuitable winter conditions for manure and nutrient application across Ontario.

Practical guidelines addressing the timing of manure and nutrient application must consider the concerns of the farm operators while ensuring the protection of the environment. An approach was developed and analyzed through case studies to determine the first recommended day in the spring, and the last in the fall, for manure and nutrient application based on probability analysis. Since most manure and nutrient application guidelines recommend avoiding adverse conditions, the three criteria established to perform a risk assessment were: (i) a frost depth greater than 0.05 m; (ii) a snow accumulation of greater than 0.05 m; and (iii) a soil volumetric water content greater than or equal to that of the plastic limit for the soil. Climatic data and typical soil information for seven locations in Ontario were used to model volumetric soil water contents, frost depths, and snow accumulation from the simultaneous heat and water (SHAW) model for a 48-yr period (1954-2001). Applying the three criteria to the modeled output, the average range between the least limiting probability (0.1, or one in ten year occurrence) and the greatest limiting probability (0.001, or one in one thousand year occurrence) analyzed among the locations was 16 d in the spring as compared to 29 d in the fall. Although geographical location affected the predicted spring start and fall end recommended manure and nutrient application dates, local climate and soil hydraulic properties also played an important part in the determination of these days. Overall the prediction method developed performed reasonably well and provided insight into the environmental factors influencing manure and nutrient application timing.