Artificial Intelligence Approach for Estimating Dairy MethaneEmissions

California's dairy sector accounts for similar to 50% of anthropogenic CH4emissions in the state's greenhouse gas (GHG) emission inventory. Although Californiadairy facilities'location and herd size vary over time, atmospheric inverse modelingstudies rely on decade-old facility-scale geospatial information. For thefirst time, weapply artificial intelligence (AI) to aerial imagery to estimate dairy CH4emissions fromCalifornia's San Joaquin Valley (SJV), a region with similar to 90% of the state's dairypopulation. Using an AI method, we process 316,882 images to estimate the facility-scale herd size across the SJV. The AI approach predicts herd size that strongly (>95%)correlates with that made by human visual inspection, providing a low-cost alternativeto the labor-intensive inventory development process. We estimate SJV's dairy entericand manure CH4emissions for 2018 to be 496-763 Gg/yr (mean = 624; 95%confidence) using the predicted herd size. We also apply our AI approach to estimateCH4emission reduction from anaerobic digester deployment. We identify 162 large(90th percentile) farms and estimate a CH4reduction potential of 83 Gg CH4/yr forthese large facilities from anaerobic digester adoption. The results indicate that our AI approach can be applied to characterize themanure system (e.g., use of an anaerobic lagoon) and estimate GHG emissions for other sectors

Automated summary

This study utilizes artificial intelligence and aerial imagery to estimate dairy methane emissions in California's San Joaquin Valley. The AI method provides a low-cost alternative to the labor-intensive inventory development process and accurately predicts herd size. The estimated methane emissions for the region in 2018 are between 496-763 Gg/yr, and the adoption of anaerobic digesters could potentially reduce methane emissions by 83 Gg CH4/yr for large facilities.