3-D Lightning Location Based On Low Frequency Magnetic Field And Preliminary Detection Performance Assessment Of Lightning Mapping Imager On FY-4A Geostationary Meteorological Satellite

The development of the satellite-and ground-based lightning detection system has greatly enriched the ways of lightning research.It is necessary to evaluate the detection performance of Lightning Mapping Imager(LMI)on Fengyun-4A(FY-4A)with ground-based synchronous high precision lightning data during its on-orbit operation.Low-Frequency Magnetic Field Lightning Detection and Imaging System(MFLDIS)has been established by Nanjing University of Information Science & Technology(NUIST)in Nanjing and Kunming in 2018.On the basis of accurate three-dimension(3-D)lightning sources location algorithm,LMI detection performance was preliminarily evaluated with synchronized lightning network observation data from MFLDIS.A more efficient and general LMI data pre-processing method was also provided.The main results are as following.In terms of 3-D lightning location algorithm,time of arrival(TOA)method via crosscorrelation was applied to image entire lightning flash using observation data from 7 stations of MFLDIS in Nanjing.Monte Carlo theory simulation and other synchronized data from Doppler radar and temperature of brightness blackbody(TBB)data were used respectively to assess the performance of MFLDIS.The horizontal location error in the detection network is less than 150 m under the standard of MFLDIS timing error.The plane locations of lightning sources show great consistence with lower cloud-top temperature region and intense radar echo zone.It is also found that MFLDIS has a high analytical capability for the lightning 3-D development channel and discharge process from the location results of a cloud flash and cloudground flash.When comparing the results of different TOA calculation methods,the 360 cases of evaluated return strokes' locations associated with cross-correlation method and time of the peak field derivation are found to be closest,about 300 m or less in the detection network.The time differences of different methods appear to increase the uncertainty of location results.Considering the data pre-processing methods of LMI,the fact that there are many different kinds of false signals in the luminous events detected by LMI can't be ignored.In addition,different kinds of filtering algorithms had been established particularly to filter out corresponding false signals originating from different sources.A more common algorithm using Bayesian inference was proposed to filter out different kinds of false signals detected by LMI at the same time.The results show that the false signals account for about 50% of all signals.The true signals' locations show great correspondence with these of World-wide Lightning Location Network(WWLLN)strokes and lower cloud-top temperature region.The false alarm rate meets the technical target of less than 10%.After filtering out the false signals,detection performance of LMI in Kunming and Nanjing was evaluated during thunderstorms in the corresponding region on June 2 and August 3,2018 based on synchronized lightning data from LMI and MFLDIS.LMI detects less groups and flashes in both regions.The number of groups and flashes detected by LMI are four times higher than these of MFLDIS in Kunming,and the detection efficiency of return strokes by LMI is less than 30%.LMI monitors 3 flashes in an hour in Nanjing,and the detection efficiency is less than 10%.The distance deviation between the LMI and MFLDIS is found to be 10-12 km,and about 1s for time difference.LMI mainly detects the discharge process in the cloud after the return stroke as to cloud-ground flash.Lightning intensity itself plays a decisive role in the size of cloud-top radiation area and radiation energy density which can be detected by LMI.The cloud-top radiation characteristics are modulated by factors such as the position of lightning in clouds,microphysics characteristics of thunderstorms and observation zenith angles of LMI in different areas,which then affect the detection efficiency of lightning by LMI.Cloud height and changes of satellite attitude angles may give rise to LMI location uncertainty,which increases gradually along latitudes and shows greater location error in northwest and northeast directions,even scales out spatial resolution of LMI pixels.