Sniffer position in an automated milking system affects the variability, repeatability, and consistent ranking of enteric methane emission measurements of dairy cows
Open Access
- 1 March 2026
- journal article
- Published by American Dairy Science Association in Journal of Dairy Science
- Vol. 109 (3) , 2669-2687
Abstract
Methane (CH4) sniffers integrated into automated milking systems (AMS) are increasingly deployed to collect high-resolution, large-scale CH4 emission data from dairy cows. The objective of this study was to assess how sniffer nozzle placement in AMS affects enteric CH4 measurements in dairy cows. Nine CH4 sniffer units (SimpleScan; C-Lock Inc.) were installed inside the AMS (DeLaval VMS) to measure enteric CH4 concentration (ppm, equivalent to microliters per liter), with intake nozzles arranged in a 3 × 3, 15-cm grid above the feed bin. They were labeled from the top left to the bottom right position: upper left (UL), upper center (UC), upper right (UR), middle left (ML), middle center (MC), middle right (MR), lower left (LL), lower center (LC), and lower right (LR). Concurrently, a GreenFeed system (GF; C-Lock Inc.) was placed inside the freestall barn to measure enteric CH4 production (g/d) from the same animals. Data from 107 cows were collected for 154 d. Visit-level data were aggregated by calculating mean values for each cow at the daily, weekly, and biweekly levels. Considerable variation in sniffer-based CH4 concentration measurements was observed at the visit level, with mean (±SD) values ranging from 214 ± 102.5 ppm to 673 ± 239.9 ppm. The CV among sniffers ranged from 35.7% (LC) to 53.1% (MR) at the visit level and from 24.0% (UR) to 35.5% (LR) at the biweekly level. The Pearson correlation between each sniffer and the GF were moderate (rp ranged from 0.25 [MR] to 0.49 [MC and LC]), with the central positions showing the strongest correlation. Stronger correlations (rp >0.80) among sniffers were predominantly observed between sniffers positioned vertically. Visit-level CH4 emission data were analyzed using generalized additive mixed-effects models. Aggregated data at daily, weekly, and biweekly levels of each cow were analyzed using a linear mixed-effects model. The variance components from these models were extracted to calculate repeatability. Repeatability estimates based on visit-level data were 0.19 for GF, with the values for individual sniffers ranging from 0.16 (UR) to 0.40 (LC). For both GF and sniffers, repeatability improved substantially when averaging weekly, with LC and MC achieving high repeatability (>0.70). To assess the potential of sniffers to rank cows as high or low emitters, estimated random cow effects of each sniffer position and the GF were used to calculate Spearman's rank correlation coefficients. Low to moderate correlations were observed (rs ranged from 0.31 [MR] to 0.56 [MC]). The MC location showed the greatest potential to rank animals. Finally, the model-estimated systematic effects on CH4 emission varied by sniffer locations, with LC and MC being able to consistently capture similar diurnal variations, as well as breed- and parity-specific effects on CH4 emissions, comparable to those detected by the GF. In summary, the reliability, between-animal variability, and ranking correlations of CH4 measurements vary between sniffer locations in AMS, with the sniffer positioned near the cow's muzzle being the most repeatable. Benchmarking on GreenFeed, sniffers exhibited differential uncertainty in ranking lactating dairy cows based on their CH4 emissions, and require further investigations.Keywords
This publication has 23 references indexed in Scilit:
- A novel approach for anomaly detection in dairy cow gas emission recordsComputers and Electronics in Agriculture, 2023
- Quantification of methane emitted by ruminants: a review of methodsJournal of Animal Science, 2022
- Review: Genetic and genomic selection as a methane mitigation strategy in dairy cattleAnimal, 2020
- Enteric methane emission can be reliably measured by the GreenFeed monitoring unitLivestock Science, 2019
- Interchangeability between methane measurements in dairy cows assessed by comparing precision and agreement of two non-invasive infrared methodsComputers and Electronics in Agriculture, 2016
- Use of short-term breath measures to estimate daily methane production by cattleAnimal, 2016
- Methane production and diurnal variation measured in dairy cows and predicted from fermentation pattern and nutrient or carbon flowAnimal, 2015
- Animal board invited review: genetic possibilities to reduce enteric methane emissions from ruminantsAnimal, 2015
- Variation in enteric methane emissions among cows on commercial dairy farmsAnimal, 2014
- Reliability, repeatability and reproducibility: analysis of measurement errors in continuous variablesUltrasound in Obstetrics & Gynecology, 2008