Improving rainfall retrieval algorithms over mountainous regions using multi-sources remotely sensed and lightning data

The objective of the present study is to improve satellite-based rainfall retrieval algorithms for estimating high-resolution (up to hourly 8km x 8km) precipitation from the combination of satellite-based infrared and lightning information for thunderstorm events, over the mountainous regions. High-resolution cloud-top infrared brightness temperature (IR-T_b) from geostationary satellites (GOES) and cloud-to-ground lightning (CGL) from The National Lightning Detection Network (NLDN) are used for rainfall estimation using an artificial neural networks-based algorithm, in this study. Rainy cloud with colder top temperature (T_b) and stronger lightning activity is assumed to produce heavier rainfall. CGL-rainfall studies have demonstrated that lightning (L) is correlated with rainfall more than remote sensing cloud information, particularly for thunderstorm events. In addition, some researchers have found that there is a high correlation between increasing terrain elevation and CG lightning activity. The selected study area is located at coordinates of latitude 32^oN – 38^oN and longitude 104^oW – 112^oW over a mountainous region. Preliminary results for a few hour storms demonstrate that using lightning in addition to cloud top T_b increases the accuracy of rainfall estimates, especially if a cloud classification schemes based on CGL is used (cc = 0.74).