“…The detailed methodology of LMDI can be found in Ang (). According to previous modeling studies (Guan et al, ), we decomposed cropland‐N 2 O emissions into a combination of different drivers: total sowing area ( A k , ha), the share of nine different crops to total sowing area ( m jk , %) also known as crop mix, N application rate ( R jk , kg N ha −1 year −1 ), and the emission intensity ( e jk , %) per crop type and region:where region k = 1–8 correspond to China, the Northwest, the Northeast Plain, the North China Plain, the lower reach of Yangtze River basin, the Southwest, the Northwest, and Qinghai‐Tibet Plateau; A jk is the sowing area of crop j in cropping region k ; N jk and E jk are N‐fertilizer application amount and croplands‐N 2 O emission of crop j in cropping region k , respectively. It should be noted that e jk is defined as cropland‐N 2 O emission per unit of N jk , which is different from the EF defined in the 2006 IPCC Guidelines, and represents the gross emission intensity at a given N application level.…”