The Biomass Burning and Lightning Experiment phase A (BIBLE A) aircraft campaign was carried out over the tropical western Pacific in September and October 1998. During this period, biomass burning activity in Indonesia was quite weak. Mixing ratios of NOx and NOy in air masses that had crossed over the Indonesian islands within 3 days prior to the measurement (Indonesian air masses) were systematically higher than those in air masses originating from the central Pacific (tropical air masses). Sixty percent of the Indonesian air masses at 9--13 km (upper troposphere, UT) originated from the central Pacific. The differences in NOy mixing ratio between these two types of air masses were likely due to processes that occurred while air masses were over the Islands. Evidence presented in this paper suggests convection carries material from the surface, and NO is produced from lightning. At altitudes below 3 km (lower troposphere, LT), typical gradient of NOx and NOy to CO (dNOy/dCO and dNOx/dCO) was smaller than that in the biomass burning plumes and in urban areas, suggesting that neither source has a dominant influence. When the CO-NOx and CO-NOy relationships in the UT are compared to the reference relationships chosen for the LT, the NOx and NOy values are higher by 40--60 pptv (80% of NOx) and 70--100 pptv (50% of NOy). This difference is attributed to in situ production of NO by lightning. Analyses using air mass trajectories and geostationary meteorological satellite (GMS) derived cloud height data show that convection over land, which could be accompanied by lightning activity, increases the NOx values, while convection over the ocean generally lowers the NOx level. These processes are found to have a significant impact on the O3 production rate over the tropical western Pacific.