Data processing pipeline for managing enteric methane emissions from sniffer: Application on dairy farming systems

Accurate quantification of enteric methane (CH 4 ) emissions from ruminants is essential to support global strategies aimed at mitigating greenhouse gas outputs from livestock sector. Among the available in vivo techniques, sniffer method offers a non-invasive and cost-effective solution for large-scale monitoring, but lacks standardized procedures for data processing and validation. This study presents a pipeline for processing sniffer-derived raw data to estimate individual CH 4 and carbon dioxide (CO 2 ) emissions from dairy animals. The procedure was tested on 3 commercial farms in central Italy, comprising 141 monitored animals and 47,018 milkings, each rearing a single breed: Holstein–Friesian, Brown Swiss cows, or Mediterranean buffaloes. The pipeline integrates measurements from sniffers with automatic milking system data and includes procedures for time alignment, data cleaning, background gas correction, and CH 4 peaks detection. Individual daily CH 4 production was estimated considering the amount of CO 2 predicted and the CH 4 :CO 2 ratio method. Methane emissions averaged 416.7 ± 74.9 g day −1 for Holstein-Friesian and 502.1 ± 158.4 g day −1 for Brown Swiss cows, differing by approximately 10% from the Intergovernmental Panel on Climate Change (IPCC) Tier II model estimates. For buffaloes, average CH 4 emissions were 279.0 ± 97.5 g day −1 , which were consistent with the emission factor estimated by Tier II of the IPCC. The proposed pipeline provides a harmonized and replicable framework for processing sniffer data, improving the reliability and comparability of enteric CH 4 and CO 2 emission measurements across ruminant species and farming systems.