Results from a three-year study of CG lightning activity over the CONUS. Provided by Bard Zajac and Steve Rutledge (CSU Dept. Atmos. Sci.).



Positive CG flash density: This map shows a strong west-to-east gradient in positive CG flash density from southeastern NM to the NE panhandle, most pronounced over western KS and the TX panhandle. This gradient appears to be associated with a similar west-to-east gradient in severe weather frequency over the High Plains as well as the location of the dryline.

Percentage of positive CG flashes, averaged over Jan-Dec 1995-97: This map also shows a strong west-to-east gradient over the High Plains with the strongest gradient over western KS. This gradient also appears to be associated with severe weather, specifically with the tendency for severe weather to occur during the upscale development of MCSs. The maximum seen over the north-central U.S. has been observed since the inception of the NLDN (Orville 1994; Orville and Silver 1997; Orville and Huffines 1999).

Positive CG mean peak current, averaged over Jan-Dec 1995-97: A maximum in positive CG mean peak current is collocated with a maximum in percent positive CG lightning. This correlation is common in positive strike dominated (PSD) storms (see Seimon 1993, Plainfield tornadic storm).

Phase difference in the diurnal cycles of positive and negative CG lightning frequency from June-August 1994-96 (positive values indicate +CGs lag -CGs): While +CGs tend to lag -CGs over most of the CONUS, +CG activity precedes -CG activity up to several hours over the region associated with maxima in percent positive CG lightning and positive CG peak current. This behavior appears to be associated with the production of +CGs [-CGs] by severe weather during the early evening [MCS during the nocturnal period]. This diurnal behavior -- as well as the correlation between percent positive CG lightning and positive peak current -- is detailed in the next three images over the Sioux Falls (FSD) and Goodland (GLD) areas.

Normalized hourly positive and negative CG flash density over the FSD area (200-km radius area) from June-August 1994-96: This image shows that the signal over the north-central U.S. in "4_diurnal_phase_diff.gif" is due to +CG production during the early evening and -CG production during the nocturnal period.

Hourly CG lightning statistics over KFSD from June-August 1994-96: This image shows the correlation between percent positive CG lightning and positive CG mean peak current in detail -- as well as an anti-correlation with negative CG mean multiplicity. The behavior of these time series around 1600 LMT (local) is representative of the general CG lightning behavior of PSD storms (Seimon 1993)

Normalized hourly positive and negative CG flash density over the GLD area (100 x 100 km area) from June-August 1994-96: The phase difference between positive and negative CG lightning is not as pronounced as was seen over FSD, but the separation between +CGs and -CGs is still clear. Again, this separation appears to be caused by the distinctly different CG lightning production by severe storms and MCSs.

Percent hail and tornado reports from 1994-95 warm seasons accompanied by greater than 50% positive CG lightning (Larry Carey): The collocation of the signals in images 2, 3, 4, and 8 is uncanny. The signals point to unique behavior of convection during the early evening over large portions of the north-central U.S. Radar-lightning case studies over FSD during 1996 show that these signals can be attributed to PSD storm (as well as their nocturnal MCS brethren).