Subnational climate actions, such as the decarbonization of cities, require timely and spatially granular greenhouse gas (GHG) information. However, commonly used emissions datasets often do not capture subnational variations because such differences are masked by national-level aggregation. Here we present the National Emissions Modeling System for GHG emissions in Japan (NEMS-GHG), a framework for modeling monthly emissions at 1 1 km spatial resolution. The system currently estimates fossil-fuel CO ₂ (FFCO ₂) emissions for eight sectors, with sector-specific spatiotemporal variations derived from publicly accessible socioeconomic statistics. This paper outlines the methodology and the initial version (1. 0) of NEMS-GHG, focusing on estimates for 2015, and compares the resulting dataset with four established emissions inventories. Annual FFCO ₂ emissions from NEMS-GHG closely match those reported in Japan’s National Greenhouse Gas Inventory. At the subnational scale, NEMS-GHG reproduces regional variations reasonably well, although noticeable discrepancies appear in electricity generation and industry and commerce when compared with a widely used global dataset. Comparisons of 1 km emissions distributions show differences exceeding 100% for about 4–8% of grid cells with emissions 10 Gg C yr ^-1. While further refinement is needed to improve sector-specific spatial distributions, NEMS-GHG provides a transparent and detailed bottom-up framework capable of estimating recent emissions from local socioeconomic sources at municipality and prefecture scales. The system can support the development and evaluation of subnational mitigation strategies by identifying locally relevant emission sources.
Saito et al. (Wed,) studied this question.