High-Resolution Image

Summertime lightning activity monitored by the National Lightning Detection Network (NLDN) over the inland (non-coastal) southeast U.S. seems to prefer work days to weekends for the eleven years from 1998-2008. This behavior is consistent with the changes with the day of the week in average rainfall seen by the Tropical Rainfall Measuring Mission (TRMM) over the same area (Bell et al. 2008). The "clock plot" shows how strong (compared to random variability) the changes in average lightning strike rates with the day of the week are, based on fitting each summer of data to a smooth sinusoidal curve that repeats once a week. The day of the week of the maximum of the fit is shown by which sector the colored balloon falls in, with the last two digits of each year inscribed within the balloons. The distance of the balloon from the center indicates how strong the changes with the day of the week are.

What could cause this? We believe that it is because of the variations in particulate pollution that occur with the day of the week over much of the U.S. Pollution levels tend to peak during the work week. The extra pollution can cause more numerous but smaller cloud droplets to form as a storm grows. The smaller, lighter droplets formed in "dirty air" are carried much higher as big storms evolve than they would be in storms forming in "clean air". The droplets can reach altitudes where they freeze and release the heat energy needed to keep water liquid instead of solid. This causes the storms to grow even taller and bigger, pulling in extra moisture (in the moist, summertime SE U.S.), supplying the storm with still more "fuel" to grow even more. Satellite rainfall and other data have already provided evidence (Bell et al. 2008) of these changes, but increases in lightning are also expected because lightning is generated by the ice-forming processes in storms, and, indeed, the expected weekly changes in lightning are visible in the data. These changes are only seen in the summertime SE U.S., where the tendency for big storms to form is high and humidity is high, and are particularly strong when only afternoon data are used, as is the case here. Weekly changes are not evident over the western U.S. The lightning behavior we see tends to confirm both the "reality" of the weekly cycle seen in the satellite data and the theory proposed to explain the weekly cycles.

The precise area over which the data are averaged is depicted in an earlier Image of the Week. There one can also see "clock plots" based on TRMM estimates of rainfall over the area. The changes with the day of the week in lightning activity tend to track those of rainfall fairly well from year to year, though the weekly cycle of lightning data seems to be stronger than that of rainfall, possibly because of the excellent density of observations available for lightning activity, or because the changes in lightning activity are more sensitive to the effects of pollution than rainfall is.

NLDN data were provided by the NASA Lightning Imaging Sensor (LIS) instrument team and the LIS data center via the Global Hydrology Resource Center (GHRC) located at the Global Hydrology and Climate Center (GHCC), Huntsville, Alabama through a license agreement with Global Atmospherics, Inc (GAI). (The data available from the GHRC are restricted to LIS science team collaborators and to NASA EOS and TRMM investigators.). A paper co-authored with Daniel Rosenfeld is being prepared describing these results.

(Submitted by T. L. Bell)