Remotely　sensed　nighttime　light　(NL)　data　collected　by　the　Suomi　National　Polar-orbiting　Partnership　Satellite　equipped　with　the　Visible　Infrared　Imaging　Radiometer　Suite　(NPP-VIIRS)　sensor　have　proven　to　be　effective　for　evaluating　fossil　fuel　combustion　carbon　emissions　(CEs).　However,　few　studies　have　analyzed　the　relationships　between　NL　and　CE　originating　from　different　sectors.　The　effects　of　impact　factors　on　the　NL-CE　relationship　have　not　been　thoroughly　examined　and　compared.　Utilizing　the　corrected　annual　composite　average　of　NPP-VIIRS　data　(NTL),　this　letter　individually　investigated　the　relationships　between　the　NTL　and　CE　from　all　types　of　fossil　fuels　total　CE　(TCE);　CE　from　gasoline,　diesel　oil,　natural　gas,　and　cement　urban　carbon　emission　(UC);　and　CE　from　raw　coal,　cleaned　coal,　other　washed　coal,　briquette,　and　coke　industrial　carbon　emission　(IC)　in　China　at　the　provincial　level.　The　impact　factors　governing　the　NTL-CE　relationship　were　also　examined.　The　results　showed　that　total　NLs　(TNLs)　may　be　a　more　effective　means　for　estimating　UC　than　other　types　of　CE　but　may　not　be　a　good　proxy　for　IC　due　to　the　mismatch　between　their　amounts　and　brightness.　The　R2　values　from　TNL　and　TCE　analyses　were　higher　than　those　of　TNL　and　IC　within　the　eastern,　central,　and　western　regions.　Meanwhile,　we　found　that　NTL　could　more　accurately　evaluate　CE　in　urban　areas　with　a　large　population　size　and　a　relatively　developed　social　economy.　Although　the　urbanization　rate　was　the　most　important　factor　in　the　assessment　of　CE　from　NTL,　China＇s　urbanization　rate　presented　an　inverted　U-shaped　impact　on　the　NTL-CE　relationship　in　the　long　run.　©　2004-2012　IEEE.