With growing demand for energy and fossil fuels, biomass and biochar are gaining attention due to their abundance and sustainability. However, there is a crucial need for optimizing production conditions and minimizing environmental risks associated with biochar production. Machine learning is an emerging strategy for predicting biochar yield and composition, optimizing production conditions and minimizing environmental risks. This paper presents a ResNet-based autoencoder model that utilizes biomass properties and pyrolysis conditions to more accurately and robustly predict biochar yield and composition. The developed model has the advantage of addressing the common data uncertainty concerns in training data. Our model outperforms commonly used baseline methods, including MLP-NN (mean R² = 0.907), Random Forest (RF, mean R² = 0.798), XGBoost (XGB, mean R² = 0.826), and Gaussian Process (GP, mean R² = 0.786), by achieving a mean R² of 0.974. The performance of the model was further improved by incorporating previously discarded data with high missing rates, achieving an average R² of 0.983. The addition of the three newly collected covariates resulted in an average R² of 0.985. Additionally, robust sensitivity analyses of the input covariates revealed the impact of data uncertainty on the performance of the model, emphasizing the robustness of the model. In advancing the application of machine learning in biochar research, this study provides a reliable method to determine optimal production conditions. Graphical Abstract

Manure management costs are a substantial component of overall costs on a modern dairy farm. Due to the slim margins of contemporary milk production operations, dairies are under constant pressure to increase milking herd sizes to take advantage of size economies that enable them to compete in the global marketplace. This study provides a current assessment of manure management and overall costs and returns on four standard sizes of dairies typical of the southern Great Plains, particularly north central Texas. The study is necessitated by the fact that the changing economic landscape has resulted in substantial changes in manure management practices. This study also forms the basis for additional analyses that will explore alternative value-added options for dairy manure management. We utilize the Farm-level Economic Model to holistically simulate the costs and returns of four representative dairy herd sizes—small (300 cows), medium (720 cows), large (1500 cows), and very large (5000 cows). Based on prevailing assumptions about land areas farmed and farm management practices, we find that dairy farms require between 0.18 and 0.4 ha/cow to manage manure based on crop nitrogen uptake rates, versus 0.67 to 0.95 ha/cow for crop phosphorus uptake rates. Manure application costs alone range from USD 55/cow (USD 225/ha) to USD 115/cow (USD 300/ha) depending on dairy size, but some of these costs are offset by fertilizer cost savings. Proportion of manure hauled offsite ranges from 9% to 67% for phosphorus-based applications, depending on herd size, and net incomes per cow are reduced by USD 60 to USD 100 (USD 4.33 to USD 8.27 per Mg of milk) if manure is applied based on phosphorus uptake rates as compared to nitrogen uptake rates of receiving crops. Generating a broad array of economically viable value-added product options from dairy manure would enable farmers to be more competitive in a market characterized by thin margins.

This study presents virtual reality (VR) videos as an effective technique for delivering extension educational materials on manure and mortality management. Virtual reality is a technology that can enable immersive distance learning. Using VR tours can enable the public to learn manure and mortality management technologies without in-person farm visits, thus saving traveling time and money and avoiding exposure to pathogens and noxious odors in farms. As the first part of this project, 360° VR tours were created on multiple aspects of manure management, including composting, anaerobic digestion, and preliminary wastewater treatment. These videos were presented to a diverse audience of agricultural producers, extension specialists, and researchers through in-person and online extension events and the Texas Manure YouTube channel. The effectiveness of videos was evaluated using surveys, which measured the audience's awareness of the topics before and after watching VR videos. Statistical analysis revealed that the increase in awareness after watching the videos was statistically significant, with a p-value of less than 0.0001. Statistical analysis also revealed that job classification and professional experience significantly impacted the change in awareness. Researchers, extension specialists/agents, and participants with less experience in manure and mortality management experienced a greater change in awareness than agricultural producers and participants with more experience. Interest and impression of the audience on the VR videos was also evaluated. According to survey data, the majority of the audience rated these videos as an effective technique for educating the public about manure and mortality management, found the VR videos interesting, were willing to learn more content through VR videos, and were likely to recommend them to their colleagues. A statistically significant correlation could be observed between the awareness change and the effectiveness rating given to the videos by the participant.