Air pollutant emissions from open biomass burning (OBB) in Yangtze River Delta (YRD) were estimated for 2005–2015 using three (traditional bottom-up, fire radiative power (FRP)-based, and constraining) approaches, and the differences between those methods and their underlying reasons were analyzed. The species included PM 10 , PM 2.5 , organic carbon (OC), black carbon (BC), CH 4 , non-methane volatile organic compounds (NMVOCs), CO, CO 2 , NO X , SO 2 and NH 3 . The inter-annual trends in emissions with FRP-based and constraining methods were similar with the fire counts in 2005–2012, while that with traditional method was not. For most years, emissions of all species estimated with constraining method were smaller than those with traditional method except for NMVOCs, while they were larger than those with FRP-based except for EC, CH 4 and NH 3 . Such discrepancies result mainly from different masses of crop residues burned in the field (CRBF) estimated in the three methods. Chemistry transport modeling (CTM) was applied to test the three OBB inventories. The simulated PM 10 concentrations with the constrained emissions were closest to available observations, implying that the constraining method provided the best emission estimates. To further evaluate the effects of method and data on OBB emission estimation, CO emissions in this study were compared with other national and global inventories. In general, inventories of FRP/BA-based method might underestimate the emissions, attributed to the detection limit on small fires. In contrast, the method based on the assumed/surveyed fraction of burned biomass could often overestimate OBB emissions and could hardly track their inter-annual trends. In particular, the constrained emissions in this work were close to GFEDv4.1s that contained emissions from small fires. The contributions of OBB to two particulate pollution events in 2010 and 2012 were analyzed with brute-force method. Attributed to varied OBB emissions and meteorology, the average contribution of OBB to PM 10 concentrations in June 8–14 2012 was estimated at 37.6 % (56.7 μg/m 3 ), larger than that in June 17–24, 2010 at 21.8 % (24.0 μg/m 3 ). Influences of diurnal curves of OBB emissions and meteorology on air pollution caused by OBB were evaluated by designing simulation scenarios, and the results suggested that air pollution caused by OBB would become heavier if the meteorological conditions were unfavorable, and that more attention should be paid to the OBB control at night. Quantified with Monte-Carlo simulation, the uncertainty of traditional bottom-up inventory was smaller than that of FRP-based one. The percentages of CRBF and emission factors were the main source of uncertainty for the two approaches, respectively. Further improvement on CTM for OBB events would help better constraining OBB emissions.