During the July 2002 Cirrus Regional Study of Tropical Anvils and Cirrus Layers-Florida Area Cirrus Experiment (CRYSTAL-FACE), flights of a WB-57F aircraft revealed mixing ratios of nitric oxide 10--50 times background over distances of 25--175 km in the anvils of thunderstorms and in clear air downwind of storm systems due to lightning activity and possible transport from the boundary layer. Estimates of the total mass of NOx injected into the middle and upper troposphere differed considerably for a moderately versus highly electrically active storm system as expected. However, assuming that the total mass is dominated by lightning production, rough estimates of the production per average lightning flash for a moderately and a highly active storm also yielded quite different ranges of (0.33--0.66) ¿ 1026 and (1.7--2.3) ¿ 1026 molecules NO/flash, respectively. If the common assumption is made that intracloud flashes have 1/10th the NO production efficiency of cloud-to-ground (CG) flashes, the ranges of production for the moderately and highly active storms were (0.88--1.8) ¿ 1026 and (4.5--6.1) ¿ 1026 molecules NO/CG flash, respectively. The observed CG flash accumulations and NOx mass production estimate for the month of July 2002 over the Florida area are compared with results from the MOZART-2 global chemistry-transport model that uses a common lightning flash parameterization. Reasonable agreement was found after a correction to the lightning parameterization was made. Finally, broad-scale median mixing ratios of NO within anvils over Florida were significantly larger than found in storms previously investigated over Colorado and New Mexico.