A physics-informed long short-term memory (LSTM) model for estimating ammonia emissions from dairy manure during storage.

Manure management on dairy farms impacts how farmers maximize its value as fertilizer, reduce operating costs, and minimize environmental pollution potential. A persistent challenge on many farms is minimizing ammonia losses through volatilization during storage to maintain manure nitrogen content. Knowing the quantities of emitted pollutants is at the core of designing and improving mitigation strategies for livestock operations. Although process-based models have improved the accuracy of estimating ammonia emissions, complex systems such as manure storage still need to be solved because some underlying science still needs work. This study presents a novel physics-informed long short-term memory (PI-LSTM) modeling approach combining traditional process-based with recurrent neural networks to estimate ammonia loss from dairy manure during storage. The method entails inverse modeling to optimize hyperparameters to improve the accuracy of estimating physicochemical properties pertinent to ammonia's transport and surface emissions. The study used open data sets from two on-farm studies on liquid dairy manure storage in Switzerland and Indiana, U.S.A. The root mean square errors were 1.51 g m-2 h-1 for the PI-LSTM model, 3.01 g m-2 h-1 for the base compartmental process-based (Base-CPBM) model, and 2.17 g m-2 h-1 for the hyperparameter-tuned compartmental process-based (HT-CPBM) model. In addition, the PI-LSTM model outperformed the Base-CPBM and the HT-CPBM models by 20 to 80 % during summer and spring, when most annual ammonia emissions occur. The study demonstrated that incorporating physical knowledge into machine learning models improves generalization accuracy. The outcomes of this study provide the scientific basis to improve policymaking decisions and the design of suitable on-farm strategies to minimize manure nutrient losses on dairy farms during storage periods.

Nitrogen transformation processes catalyzed by manure microbiomes in earthen pit and concrete storages on commercial dairy farms.

Storing manure is an essential aspect of nutrient management on dairy farms. It presents the opportunity to use manure efficiently as a fertilizer in crop and pasture production. Typically, the manure storages are constructed as earthen, concrete, or steel-based structures. However, storing manure can potentially emit aerial pollutants to the atmosphere, including nitrogen and greenhouse gases, through microbial and physicochemical processes. We have characterized the composition of the microbiome in two manure storage structures, a clay-lined earthen pit and an aboveground concrete storage tank, on commercial dairy farms, to discern the nitrogen transformation processes, and thereby, inform the development of mitigation practices to preserve the value of manure. First, we analyzed the 16S rRNA-V4 amplicons generated from manure samples collected from several locations and depths (0.3, 1.2, and 2.1-2.75 m below the surface) of the storages, identifying a set of Amplicon Sequence Variant (ASVs) and quantifying their abundances. Then, we inferred the respective metabolic capabilities. These results showed that the manure microbiome composition was more complex and exhibited more location-to-location variation in the earthen pit than in the concrete tank. Further, the inlet and a location with hard surface crust in the earthen pit had unique consortia. The microbiomes in both storages had the potential to generate ammonia but lacked the organisms for oxidizing it to gaseous compounds. However, the microbial conversion of nitrate to gaseous N2, NO, and N2O via denitrification and to stable ammonia via dissimilatory nitrite reduction seemed possible; minor quantities of nitrate was present in manure, potentially originating from oxidative processes occurring on the barn floor. The nitrate-transformation linked ASVs were more prevalent at the near-surface locations and all depths of the inlet. Anammox bacteria and archaeal or bacterial autotrophic nitrifiers were not detected in either storage. Hydrogenotrophic Methanocorpusculum species were the primary methanogens or methane producers, exhibiting higher abundance in the earthen pit. These findings suggested that microbial activities were not the main drivers for nitrogen loss from manure storage, and commonly reported losses are associated with the physicochemical processes. Finally, the microbiomes of stored manure had the potential to emit greenhouse gases such as NO, N2O, and methane.

Compositional variability of food wastes and its effects on acetone-butanol-ethanol fermentation.

Converting food waste into butanol via acetone, butanol, and ethanol (ABE) fermentation provides the potential to recover energy and value-added chemicals from food waste. However, the high variability of food waste compositions has hindered the consistency and predictability of butanol production, impeding the development of a robust industrial fermentation process. This study characterized the compositional variation of collected food wastes and determined correlations between food waste compositional attributes and butanol yields for a better prediction of food waste fermentation with Clostridium. The total sugar, starch, fiber, crude protein, fat and ash contents (on dry basis) in the food waste samples were in a range of 0.5-53.5%, 0-25.2%, 0.6-26.9%, 5.5-21.5%, 0.1-37.9%, and 1.4-13.7%, respectively. The high variability of food waste composition resulted in a wide range (3.5-11.5 g/L) of butanol concentrations with an average of 8.2 g/L. Pearson's correlation analysis revealed that the butanol concentrations were strongly and positively correlated with equivalent glucose and starch contents in food waste, strongly and negatively correlated with fiber content, and weakly correlated with total sugar, protein, fat, and ash contents. The regression models constructed based on equivalent glucose and fiber contents reasonably predicted the butanol concentration, with the R2 of 0.80. Our study investigated the variability of food waste composition and, for the first time, unveiled relationships between food waste compositional attributes and fermentation yields, contributing to a greater understanding of food waste fermentation, which, in turn, assists in developing new strategies for increased consistency and predictability of food waste fermentation.

Multiple Applications of Sodium Bisulfate to Broiler Litter Affect Ammonia Release and Litter Properties.

Ammonia (NH) emissions from animal manures can cause air and water quality problems. Poultry litter treatment (PLT, sodium bisulfate; Jones-Hamilton Co.) is an acidic amendment that is applied to litter in poultry houses to decrease NH emissions, but currently it can only be applied once before birds are placed in the houses. This project analyzed the effect of multiple PLT applications on litter properties and NH release. Volatility chambers were used to compare multiple, single, and no application of PLT to poultry litter, all with and without fresh manure applications. A field component consisted of two commercial broiler houses: one had a single, preflock PLT application, while the other received PLT reapplications during the flock using an overhead application system. In the volatility chambers, single and reapplied PLT caused greater litter moisture and lower litter pH and , relative to no PLT. After 14 d, NH released from litter treated with reapplied PLT was significantly less than litter with both single and no applications. Furthermore, total N in litter was greatest in litter treated with reapplied PLT, increasing its fertilizer value. In the commercial poultry houses, PLT reapplication led to a temporary decrease in litter pH and , but these effects did not last because of continued bird excretion. Although one preflock PLT application is currently used as a successful strategy to control NH during early flock growth, repeat PLT application using the overhead reapplication system was not successful because of problems with the reapplication system and litter moisture concerns.

Treating separated liquid dairy manure derived from mesophilic anaerobic digester effluent to reduce indicator pathogens and Salmonella concentrations for use as organic fertilizer.

Dairy manure has much potential for use as an organic fertilizer in the United States. However, the levels of indicator organisms and pathogens in dairy manure can be ten times higher than stipulated use guidelines by the National Organic Standards Board (NOSB) even after undergoing anaerobic digestion at mesophilic temperatures. The objective of this study was to identify pasteurization temperatures and treatment durations to reduce fecal coliforms, E. coli, and Salmonella concentrations in separated liquid dairy manure (SLDM) of a mesophilic anaerobic digester effluent to levels sufficient for use as an organic fertilizer. Samples of SLDM were pasteurized at 70, 75, and 80°C for durations of 0 to 120 min. Fecal coliforms, E. coli, and Salmonella concentrations were assessed via culture-based techniques. All of the tested pasteurization temperatures and duration combinations reduced microbial concentrations to levels below the NOSB guidelines. The fecal coliforms and E. coli reductions ranged 2from 0.76 to 1.34 logs, while Salmonella concentrations were reduced by more than 99% at all the pasteurization temperatures and active treatment durations.