Use of continuous and grab sample data for calculating total maximum daily load (TMDL) in agricultural watersheds.

Measuring the discharge of diffuse pollution from agricultural watersheds presents unique challenges. Flows in agricultural watersheds, particularly in Mediterranean climates, can be predominately irrigation runoff and exhibit large diurnal fluctuation in both volume and concentration. Flow and pollutant concentrations in these smaller watersheds dominated by human activity do not conform to a normal distribution and it is not clear if parametric methods are appropriate or accurate for load calculations. The objective of this study was to compare the accuracy of five load estimation methods to calculate pollutant loads from agricultural watersheds. Calculation of loads using results from discrete (grab) samples was compared with the true-load computed using in situ continuous monitoring measurements. A new method is introduced that uses a non-parametric measure of central tendency (the median) to calculate loads (median-load). The median-load method was compared to more commonly used parametric estimation methods which rely on using the mean as a measure of central tendency (mean-load and daily-load), a method that utilizes the total flow volume (volume-load), and a method that uses measure of flow at the time of sampling (instantaneous-load). Using measurements from ten watersheds in the San Joaquin Valley of California, the average percent error compared to the true-load for total dissolved solids (TDS) was 7.3% for the median-load, 6.9% for the mean-load, 6.9% for the volume-load, 16.9% for the instantaneous-load, and 18.7% for the daily-load methods of calculation. The results of this study show that parametric methods are surprisingly accurate, even for data that have starkly non-normal distributions and are highly skewed.

Impact of co-digestion on existing salt and nutrient mass balances for a full-scale dairy energy project.

Anaerobic digestion of manure and other agricultural waste streams with subsequent energy production can result in more sustainable dairy operations; however, importation of digester feedstocks onto dairy farms alters previously established carbon, nutrient, and salinity mass balances. Salt and nutrient mass balance must be maintained to avoid groundwater contamination and salination. To better understand salt and nutrient contributions of imported methane-producing substrates, a mass balance for a full-scale dairy biomass energy project was developed for solids, carbon, nitrogen, sulfur, phosphorus, chloride, and potassium. Digester feedstocks, consisting of thickened manure flush-water slurry, screened manure solids, sudan grass silage, and feed-waste, were tracked separately in the mass balance. The error in mass balance closure for most elements was less than 5%. Manure contributed 69.2% of influent dry matter while contributing 77.7% of nitrogen, 90.9% of sulfur, and 73.4% of phosphorus. Sudan grass silage contributed high quantities of chloride and potassium, 33.3% and 43.4%, respectively, relative to the dry matter contribution of 22.3%. Five potential off-site co-digestates (egg waste, grape pomace, milk waste, pasta waste, whey wastewater) were evaluated for anaerobic digestion based on salt and nutrient content in addition to bio-methane potential. Egg waste and wine grape pomace appeared the most promising co-digestates due to their high methane potentials relative to bulk volume. Increasing power production from the current rate of 369 kW to the design value of 710 kW would require co-digestion with either 26800 L d(-1) egg waste or 60900 kg d(-1) grape pomace. However, importation of egg waste would more than double nitrogen loading, resulting in an increase of 172% above the baseline while co-digestion with grape pomace would increase potassium by 279%. Careful selection of imported co-digestates and management of digester effluent is required to manage salt and nutrient mass loadings and reduce groundwater impacts.