Impact of lightning and convection on reactive nitrogen in the tropical free troposphere

Latitudinal distributions of NO, NO y , O 3 , CO, CH 3 I, and H 2 O mixing ratios at 8.9–12 km were obtained between 30°N and 10°S by DC‐8 aircraft measurements made in February 1994 during Pacific Exploratory Mission‐West B (PEM‐West B). Very low NO y mixing ratios with a median value of 51 parts per trillion by volume (pptv) were observed at 9.5–12 km at 1°N–14°N during two flights made within 3 days. A very low median O 3 mixing ratio of 19 parts per billion by volume (ppbv) and high mixing ratios of H 2 O and CH 3 I were simultaneously observed, suggesting that the low NO y values were probably due to the convective transport of air from the tropical marine boundary layer to this altitude. The median NO y /O 3 ratio being a factor of 2 smaller than in the air masses in the tropical marine boundary layer might suggest the possibility that the heterogeneous removal of HNO 3 during convective transport further reduced NO y levels. In addition to the measurements between 9.5 and 12 km, low values of NO y and O 3 were observed between 4 and 12 km at 1°N. Divergent wind fields at 200 and 1000 hPa and infrared (IR) cloud images show that there was large scale convection (>1000 km × 1000 km) in the northeast of New Guinea Island centered around 0°S and 150°E as part of systematic convective activity of the Intertropical Convergence Zone (ITCZ) and the South Pacific Convergence Zone (SPCZ). This type of large scale convection could have transported air with low levels of NO y and O 3 to the middle and upper troposphere over a wide area in the tropics. On the other hand, NO mixing ratios of 50–200 pptv and high NO x /NO y ratios of 0.4–0.6 were observed at 9.5 km between 4°S and 10°S. High H 2 O mixing ratios of 600–1200 parts per million by volume (ppmv) and low CO mixing ratios of 65 ppbv observed in the air mass indicated that the high NO values were probably due to NO production by lightning. Satellite observations showed relatively frequent lightning flashes over the New Guinea Island for 3 days prior to the aircraft measurements. These results are considered to be consistent with the idea that, in general, marine convection is not accompanied by lightning activity, whereas convection over land is. Because of the large areal extent of the influences from these processes, the convective transport of low NO y air and NO production by lightning should play critical roles in controlling the abundance of reactive nitrogen in the equatorial region.