XV International Conference on Atmospheric Electricity, 15-20 June 2014, Norman, Oklahoma, U.S.A. 1 B. Kochtubajda 1 , W.R. Burrows 1,2 , J.C. Brimelow 1 1. Environment Canada, Edmonton, Alberta, Canada 2. Environment Canada, Downsview, Ontario, Canada ABSTRACT: Thunderstorms are an important component of the water cycle on the Canadian Prairies, because they represent one of the mechanisms responsible for cycling moisture in the warm season between May and September. The Prairie landscape is extensive and diverse, and is comprised of six eco-climatic regions. A broad swath of the Boreal and Taiga regions, where station data are sparse, lie outside radar coverage but within coverage of the Canadian Lightning Detection Network (CLDN). Can lightning data be used to estimate seasonal thunderstorm rainfall over these data sparse regions? This work estimates warm season thunderstorm rainfall using two methods. The first method derives thunderstorm rainfall from the gridded Canadian Precipitation Analysis (CaPA) product (6-hr temporal resolution and interpolated to a 0.2° x 0.2° grid) and concurrent cloud-to-ground lightning data from the CLDN. The second method derives rainfall amounts from previously developed rain-yield relationships constructed from a database of coincident 6-hrly rain gauge rainfall and cloud-to-ground lightning observations. Spatial patterns of thunderstorm rainfall from the two prediction schemes are assessed for three summers from 2009 to 2011. Additionally, several verification statistics over each eco-climatic region are calculated to assess the accuracy of the two approaches INTRODUCTION Thunderstorms over the Canadian Prairie Provinces are an important component of the hydrological cycle and forest fire activity. Varying amounts of convective rainfall and lightning can trigger flash floods (Soula et al. 1998) or ignite wildfires (Rorig and Ferguson, 1999). Numerous studies have been conducted to determine lightning-rainfall relationships over a wide variety of spatial and temporal scales using various measurement platforms. A summary of these findings can be found in Soula (2009). In Canada, Kochtubajda et al. (2013) conducted an exploratory study to examine lightning-rainfall relationships during warm seasons (April to October) from 1999-2003 across Canada’s ecozones using a database of coincident 6-hourly rain-gauge measurements and lightning data. Rain yields, determined for this period, ranged between 1.05x10 8 kg fl -1 and 41.5 x10 8 kg fl -1 over western ecozones. This wide range of rain yields, both within and between ecozones, poses a challenge for deciding on an appropriate rain yield relation for estimating thunderstorm rainfall. In this study we compare two approaches to derive thunderstorm rainfall. The first approach derives thunderstorm rainfall from the Warm season thunderstorm rainfall estimation on the Canadian Prairies using lightning and gridded model data