Poster
Exploring the relationship between methane intensity and linear score in Holstein dairy cattle
| Date/Time: |
9/11/2025
|
| Author: |
Amy K Vasquez |
| Clinic: |
Cornell University |
| City, State, ZIP: |
Ithaca, NY 14850 |
A.K. Vasquez, DVM, PhD
1
;
F.A. Gutierrez-Oviedo, DVM
2
;
M.J. Farricker, BS
2
;
V. Ramos, BS
2
;
M. You, BS
2
;
V.R. Basnayake, BS
2
;
N.D Seneviratne, PhD
2
;
J.W. McFadden, PhD
2
;
1Department of Population Medicine and Diagnostic Sciences, Cornell College of Veterinary Medicine, Ithaca, NY 14853
2Department of Animal Science, Cornell University, Ithaca, NY 14853
Introduction:
Livestock systems are quoted to contribute 40% of total agrifood emissions. On dairy farms, the primary influences are enteric methane production and manure handling processes. Health events in ruminants are an economic concern as they often result in costly treatments and loss of productivity. Reductions in efficiencies in these cases can also be attributed to lower reproductive performance, higher mortality rates and culling, and increased replacement rates. Together, all have the potential to increase emissions intensity, and several have been modeled for their greenhouse gas (GHG) contributions. The direct impacts of bovine clinical or subclinical mastitis on GHG production have not been evaluated in an on-farm research setting. Our objective was to use data from a recent feed trial to determine if there are any direct impacts of animal health (specifically linear somatic cell score) on methane intensity of Holstein dairy cattle.
Materials and methods:
Multiparous Holstein cows (n=36; 138 ± 13 DIM) from one NY farm were enrolled in a split-plot clinical feed trial. For the purpose of the outlined objective, the approach is considered a retrospective observational cohort study. Each feed treatment (n=4 periods) lasted 21 days. Representative milk samples were retrieved from each of 3 milking sessions on days 19-21 of each period and submitted for somatic cell count (SCC) measurement via flow cytometry. Values per 24-hour period were averaged and subsequently transformed into linear scores (LS) using the following equation: Log2(SCC/100,000)+3. Additional data retrieved specific to this aim were: methane production (g/cow/d) provided by technician-led visits to Greenfeed units (C-lock, Inc., Rapid City, SD) 3 times/day/cow, milk production (kg/cow/d) using on-farm meters, and milk fat (kg/d) and milk protein (kg/d) calculated from Fourier Transform Infrared Spectroscopy percentages. Energy corrected milk (ECM; kg/cow/d) was calculated as (0.327 × kg of milk) + (12.95 × kg of milk fat) + (7.65 × kg of milk protein). This value was used to determine methane intensity (g/kg ECM) per cow per day. Methane intensity was used as an outcome variable within the model: data were analyzed using SAS version 9.4 using a mixed model with the fixed effects of day, LS, and day x LS, and accounting for within-subject factors using the REPEATED statement for day and cow as the SUBJECT. Stepwise removal of individual effects within the model was explored. An autoregressive covariance structure was applied. As we were only interested in exploring outcomes for cows on the same diet, we performed analysis by feed period. Significance was declared at P<0.05.
Results:
Mean and (standard deviation) of all ECM, methane, and SCC measurements were 48 kg/d (7), 455 g/d (91), and 76,000 cells/mL (230,000), respectively. No statistically significant findings were reported for the effect of day, LS, or the interaction during any period. However, evaluating the LSMEANS for the estimates of LS by day using graphical methods indicates a numerical increase in methane intensity for each one-point increase in LS. The pattern is likely driven by the significant negative linear association found between LS and daily ECM production (P<0.01 for entire trial and for individual periods). We suspect that lack of inclusion of cows with poor udder health (n=3 with LS>4.0) is the main contributor to our non-significant findings.
Significance:
Though no statistically significant associations between LS and methane intensity were highlighted, numerical increases in methane intensity for increasing incremental linear scores demonstrate a proof-of-principle concept that requires further exploration.
