Background Agricultural soils contribute towards the emission of CH4 (mainly from paddy fields) and N2O (from N-fertilizer application), the two important greenhouse gases causing global warming. Most studies had developed the inventories of CH4 and N2O emission at the country level (larger scale) for India, but not many studies are available at the local scale (e.g. district level) on these greenhouse gases (GHGs). Assam is an important state in the North Eastern region of India. In addition to being the regional economic hub for the entire region, agriculture is the major contributor to the state’s gross domestic product. In Assam about three-fourths of the area is under paddy cultivation and rice is the staple food. With this background, a district wise inventory of CH4 and N2O emission in the North Eastern state of Assam, India was carried out using different emission factors, viz., IPCC, Indian factors and others, to highlight the discrepancies that arose in the emission estimation of these important GHGs while used at the smaller scale i.e. district level. This study emphasizes the need for better methodologies at the local level for GHGs inventories. This study also reiterates the fact that no emission factor is universally applicable across all regions. The GHGs like CH4 and N2O are highly site and crop specific and the factors required for their inventory are driven by cultural practices, agronomic management, soil resources and socio-economic drivers. Material and methods In this study, Intergovernmental Panel on Climate Change (IPCC) methodology was used for the estimation of CH4 and N2O emission. In case of N2O emission, both direct and indirect emission from agricultural soil was estimated for the various districts of Assam. Results The CH4 (base year 2000–2001) and N2O (base year 2001–2002) emission was estimated to be 121 Gg and 1.36 Gg from rice paddy and agricultural fields of Assam state respectively. Conclusions This study is the first report on the estimation of the GHG emission at the district level from the entire state of Assam, agriculturally one very important state of North Eastern India. This state is also considered as remote due to its geographical location. The study clearly elucidates that there is large variation in the emission inventory of CH4 and N2O at the district level (local scale) when different emission factors are used. This calls for detailed and comprehensive data collection and mapping at the micro level for accurate inventory of greenhouse gases in future from agriculture fields.