Because of the potentially serious effects of lightning on operations of all kinds, Kennedy Space Center (KSC) and Cape Canaveral Air Force Station (CCAFS) jointly use a network of 31 surface-based electric-field mills distributed across the two installations for evaluation of Lightning Launch Commit Criteria and as part of the lightning warning decision process. In an effort to elicit patterns in the temporal and spatial evolution of the contours of surface electric field, we have begun to examine data from the unique archived sets of data from the network. To identify cases of interest, we use lightning ground-strike data from the Cloud-to-Ground Lightning Surveillance System, maps of in-cloud lightning discharges produced by the Lightning Detection and Ranging system, rainfall data from a network of rain gauges co-located with the field mills, and radar data from the WSR-88D at Melbourne, FL (KMLB). In particular, we focus on two critical problems of considerable importance at KSC/CCAFS, which are also important at any location where outdoor operations and activities take place during thunderstorm seasons. The problems are estimating when and where the first flashes (both in-cloud (IC) and cloud-to-ground (CG)) in a storm might occur and assessing the potential for flashes (both IC and CG) to occur at the end of storms after flash rates have decreased drastically.

Typical lightning hazard-warning guidelines are based on the consolidated wisdom of the lightning research community and operational meteorologists, derived from decades of experience. However, to the best of our knowledge, no current lightning hazard-warning criteria incorporate objective application and interpretation of the temporal and spatial evolution of contours of electric field at the surface before, during, and after active lightning periods in storms. Our long-term goal is to understand the evolution of surface contours of electric field for periods of 30 minutes or more before the first flash of any kind and 30 minutes or more before and after the last flash of any kind. For practical reasons, we are reporting here on analysis of data for periods of 30 minutes before the first CG flash and 30 minutes after the last CG flash. The period of 30 minutes is chosen to match the desired lead-time for lightning advisories at KSC/CCAFS.

We have begun by studying isolated air-mass convective storms that developed over the KSC/CCAFS from early May through late September, 2004-2006. There are several reasons for starting with the simplest class of thunderstorm. It is easier to interpret the surface fields of isolated thunderstorms, which can be expected to have much simpler and less extensive charge accumulations. In general, isolated thunderstorms are much easier to analyze and study than frontal storms and mesoscale convective systems, and much more likely to produce a lightning flash at a given location with little or no warning. Recent research has shown that a majority of lightning casualties are a result of the first or one of the first few CG flashes in a storm or flashes at the end of storms. This makes lightning safety guidance under some circumstances, e.g., air mass storms, problematic. Once long-lived moving storms have begun to make lightning, it is by comparison relatively easy to project where they will go and what they will do on the basis of observations of where they are and what they are doing now. It is much more difficult to determine when and where a growing storm will make the first lightning flash and whether a dissipating storm has produced its last flash.

To select isolated air-mass, or pulse thunderstorms for analysis, we examined KSC rainfall and CG lightning data to find days when there was either lightning observed somewhere within the network (not just over KSC) or rainfall over KSC, or in many cases, both. The rainfall and CG data for each day were used to determine cases for which we produced animations of radar data in order to determine the characteristics of the storms. For each day that rainfall and/or CG data were recorded, an animation of archived KMLB WSR 88D base reflectivity at tilt one (0.5°) was produced for the period of interest. These animations made it possible to determine immediately the manner in which a given thunderstorm formed. Isolated air-mass, or pulse, thunderstorms were selected for further analysis.

For the selected storm periods, we performed a two-pass Barnes objective analysis on the electric-field data. A first pass was computed and a bilinear interpolation performed to estimate the first pass error, and then the second pass using an updated weight parameter, or scale factor, was computed, taking the estimated error into account. The result was then used as a first-guess for a statistical objective analysis, which serves as the final analysis that is to be plotted. Each analysis cycle results in a contour plot of 20-second averaged data, superimposed on a map of the sites of the operational field mills, as well as the CG flash of interest. The result is a set of 90 jpeg images of the field contours during the period of 30 minutes up to the time of the first CG flash or during the period of 30 minutes after the last CG flash. These are then used to produce an animated sequence for each case.

We have generated 120 such animations, 58 for first cloud-to-ground flashes, and 62 for last cloud-to-ground flashes. Preliminary results suggest that the electric-field contours during the 30 minutes before first CG flashes generally show a smooth transition from weak to strong gradient, often with maxima displaced by several kilometers from the ground-strike point. Further, it appears that the field often remains elevated for perhaps tens of minutes after the last CG flash. We present the results of the analysis of all 120 cases.