station for the highest positive-wind interval was for a sample
size of 25.)
As expected from the distributions in Fig. 3 there was little
change in median field values as the wind changes from upwind
to downwind. Only field values above the 75th percentile in a
wind interval showed significant elevation above the no-wind
and upwind values at the North Site. There was no significant
elevation of even the 95th percentile levels at the South Site.
4) Temporal Variability: Fig. 5 demonstrates the temporal
variability of electric-field measurements during periods of sustained
perpendicular wind direction at the North Site. The twohour
periods in Fig. 5 were selected because they contained
valid data and the most consistent perpendicular winds in their
respective directions.
Table V contains statistical descriptors for the one-minute average
electric field and wind data used to construct Fig. 5. For
these example two-hour periods, the increased downwind electric
field relative to upwind field was apparent for both wind
directions.
One measure of the variability is the difference between successive
one-minute averages (first difference). During the period
of west-to-east winds in Fig. 5, the mean absolute first difference
for the downwind (NE) field was 51 V/m. During the
east-to-west wind period the mean absolute first difference for