Estimating the Carbon Footprint of Milk in Dual-Purpose Dairy Production System: Influence of Animal Productivity, Functional Unit, and Emission Allocation and Quantification Methods
Description
This dataset contains five years (2018–2023) of comprehensive herd, feed, manure, and farm operation data collected from a dual-purpose dairy production system in Bangladesh, used to estimate the carbon footprint of milk following IPCC Tier 1 and Tier 2 methodologies. It includes annual herd structure and Animal Units, breed composition, body weight measurements, milk yield and FPCM production, and full milk composition profiles. Detailed ration formulations for all animal categories are provided for each year, along with nutrient composition, feed production inputs (fertilizers, pesticides, diesel, irrigation, machinery), and transportation details. The dataset also includes IPCC equations for enteric methane, manure methane, manure nitrous oxide, nitrogen excretion, and feed-related emissions, as well as farm energy use (electricity and diesel). Annual greenhouse gas emissions by source (enteric fermentation, manure management, feed production, and farm energy) are presented alongside carbon footprint results under three allocation approaches (no allocation, biophysical, and economic) and three functional units (kg FPCM, 100-kcal energy, and 100-g protein). Uncertainty analysis results based on 1,000 Monte Carlo simulations per year are included, providing mean values, standard deviations, and 68% and 95% confidence intervals. The complete field questionnaire used for data collection—covering farm characteristics, feeding, reproduction, manure management, feed production, water and energy use, and historical system changes—is also provided, along with supplementary figures depicting emission flows and Monte Carlo distributions.
Files
Steps to reproduce
To reproduce the results, first download all dataset files and organize them by year (2018–2023). Using the provided tables, calculate annual FPCM for the herd using the IDF (2015) equation. Next, estimate enteric methane emissions by applying the IPCC Tier 2 equations for gross energy intake and the methane conversion factor. Manure-related CH₄ and N₂O emissions can be reproduced by using the Tier 2 equations for volatile solids, nitrogen excretion, and waste management system fractions included in the dataset. Feed production emissions should be calculated by multiplying daily ingredient intake by the corresponding fertilizer, diesel, electricity, transport, and field-operation emission factors. Farm energy emissions can be reproduced by applying the electricity and diesel emission factors provided. Summing all emission sources yields the total annual GHG emissions. The carbon footprint can then be calculated using the linear additive LCA model under no allocation, biophysical allocation, and economic allocation, and normalized using the three functional units: kg CO₂-eq/kg FPCM, kg CO₂-eq/100 kcal, and kg CO₂-eq/100 g protein. Finally, uncertainty analysis can be reproduced by conducting 1,000-iteration Monte Carlo simulations using the parameter ranges supplied, and the resulting distributions, confidence intervals, and summary tables can be generated following the examples in the supplementary materials.
Institutions
- University of Connecticut
- Bangladesh Agricultural University Faculty of Animal Husbandry