In rice production greenhouse gas emission (GHG) reduction is an important task for many countries, Tanzania included. Of global agricultural GHG emitted from rice fields, about 30 and 11% are represented by CH4 and N2O, respectively. For successful climate smart rice cultivation, rice management practices, including nitrogen fertilization are two key crucial components that need evaluation. The objective of this study was to evaluate the crop management practices and N fertilization on yield and greenhouse gases emission in paddy rice production, Experiments were designed in split-plot randomized complete block and replicated three times. Two rice management practices namely conventional practice (CP) and system of rice intensification (SRI) and six rates of nitrogen fertilizer (absolute control, 0, 60, 90,120 and 150 kg N ha−1) were applied in two consecutive seasons. The Source-selective and Emission-adjusted GHG CalculaTOR for Cropland (SECTOR) was used to calculate the GHG emission. Methane emission was in the range of 88.7–220.6 kg ha−1season−1, where higher emission was recorded in CP treatments (ABC, CP 0 and CP 120N) compared to SRI treatments. SRI reduced methane and carbon dioxide emission by 59.8% and 20.1% over CP, respectively. Seasonal nitrous oxide emissions was in the range of no detected amount to 0.0002 kgN2O ha−1 where SRI treatments recorded up to 0.0002 kgN2O ha−1 emissions while in CP treatment no amount of N2O was detected. The interaction of system of rice intensification and 90 kg N ha−1 (SRI90N) treatment recorded higher grains yield (8.1, 7.7 t ha−1) with low seasonal global warming potential (GWP) (3,478 and 3,517 kg CO2e ha−1) and low greenhouse gas intensity (0.42, 0.45 kg CO2e per kg paddy) compared to other treatments in wet and dry season, respectively. Therefore, SRI with 90 kg N was the treatment with mitigation potential and reduced GWP without compromising rice yield.