Comparison of the Number of Thunderstorm Days from Lightning Location System and Artificial Observations

Using the lightning location system (LLS) monitoring data of Hubei Province from 2007 to 2012 and 1983-2012 artificial observations, 25 meteorological stations with theoretical detection efficiency above 95% are selected to make a relative analysis on the number of thunderstorm days monitored by LLS with different monitoring radius (r) and artificial observations. Results show that annual mean thunderstorm days of artificial observations and LLS monitoring data agree mostly in the radius which range from 6.4 km to 10.2 km. In the circular area when r=7, 8, 9 km, the difference is minimum, with the average difference of about 19%. When r≤7 km, the annual mean thunderstorm days from LLS monitoring data is less than that of artificial observation, while it is more than artificial observation when r≥8 km, and it can replace the number of maximum annual mean thunderstorm days of artificial observations when r=22 km. The ground flash density do not change significantly with the monitoring radius in the range of 2-40 km, the annual mean flash density is 3.9-4.1 times/(km² ·a) and the average value is 4 times/(km² ·a) for each monitoring radius. According to data of LLS, 3 methods are suggested to calculate the number of annual mean thunderstorm days of artificial observations. The first method is using the number of annual mean thunderstorm days of LLS monitoring data when r=7 km to represent the artificial observations directly, called direct substitution method. The second method is to calculate the number of annual mean thunderstorm days of artificial observations by the equation with one unknown quantity on the basis of the annual mean flash density data when r=8 km, called ground flash density method. The third method is using the binary equation to calculate the number of annual mean thunderstorm days of artificial observations on the basis of the number of annual mean thunderstorm days and the annual mean flash density data when r=8 km, called binary method. The examination shows that the binary method is the best, followed by the ground flash density method and the direct substitution method. The number of annual mean thunderstorm days of 25 stations calculated by binary method from 2007 to 2012 are equal to that of artificial observations, and the average difference is 7.4%. In 2013, the gap of the number of annual mean thunderstorm days between binary method and artificial observation is 0.8 d, and the average difference is 12.3%.