Data: Farm level factors influencing enteric methane emissions from dairy cattle in smallholder farming systems of Malawi

The data describes smallholder dairy systems in Malawi. This includes the characteristics of smallholder dairy systems and enteric methane concentrations from dairy cattle. The study area included five districts of Malawi: Lilongwe, Mchinji, Dowa, Mzimba and Dedza, and was conducted between May and June 2024, in the cool dry season. The research aimed to achieve the following at smallholder dairy level: to quantify milk production and actual enteric methane values from dairy cattle, to investigate the farm-level variables influencing enteric methane and to determine the relationship between enteric methane and dairy productivity.

Seventy-two smallholder dairy farmers were selected via random sampling and surveyed. Farmers met the following criteria: members of a milk bulking group, within the 5 selected districts and partnered with Lilongwe University in the FoodMa project (https://luanar.ac.mw/pco2/programmes.php). Sixty-eight farms comprising 100 dairy cattle (heifers and cows) were included for analysis.

Farm level data (farm and farmer characteristics) were collected via intra-household interview, guided with an "intra-household questionnaire". The KoboToolbox software was used to design and create the questionnaire, as well as store the farm level data. KoboCollect smartphone app was used to facilitate the interview. Enteric methane concentrations were measured directly from each animal in all the selected farms using the Laser Methane Detector (LMD), which was directed between the nostrils of the animal. Methane was measured per animal in the following procedure: every 6-minutes with 1-minute rest intervals, for 1 hour.

Data was categorized into two: Methane concentration data and Farm level data.
Methane concentration data (ppm) included "raw methane activity data" and "average methane data". Raw methane data was cleaned and analyzed to determine the enteric methane concentrations from the animal which was recorded in the dataset "average methane data".
Farm level data comprised the "raw farm data" and "processed farm data". The raw farm data was data recorded during the interviews which was processed by combining similar variables, to create the "processed farm data" . For analysis, both processed datasets were combined into the "farm and methane data".
The combined dataset was analyzed using a linear mixed model (LME) to determine farm-level factors influencing enteric methane emissions from dairy cattle in smallholder dairy systems and the relationship between enteric methane and dairy productivity.