The northern part of the Tomsk region is a typical area of the boreal forest zone. Large forest resources occupy this territory. The forest fund is situated on the 90.5% of its entire area. There are 17 million hectares of forested area, including 9.9 million hectares covered by coniferous. 20-30% of forest fires in the Tomsk region caused by the lightning activity. Analysis of the thunderstorm activity in the context of fires is possible using Tomsk region as sample region. World Wide Lightning Location Network (WWLLN) data for 2010-2015 years are used in paper. This network registers lightning discharges throughout the year. The majority of thunderstorms occur over Tomsk region during the summer time. Data on lightning discharges during the period from May to September was selected for investigation. Data on spherics recorded by WWLLN contain the points: date, time, latitude, longitude, error and number of stations in which the electromagnetic pulse was recorded. Lightning discharge density maps were performed for cells 10×10 km.
The aim of the work is to develop a comprehensive method for assessing thunderstorm activity using WWLLN and RS data. It is necessary to group lightning discharges to solve practical problems of lightning protection and lightningcaused forest fire danger, as well as climatology problems using information on the spatial and temporal characteristics of thunderstorms. For grouping lightning discharges, it is proposed to use clustering algorithms. The region covering Timiryazevskiy forestry (Tomsk region, borders (55.93 - 56.86)x(83.94 - 85.07)) was selected for the computational experiment. We used the data on lightning discharges registered by the WWLLN network in this region on July 23, 2014. 273 lightning discharges were sampling. A relatively small number of discharges allowed us a visual analysis of solutions obtained during clustering.
The article presents a forecast of dangerous convective phenomena using one of the indices of instability (Total Totals index). The index values were calculated by Semi-Lagrangian Absolute Vorticity Numerical weather prediction (SL-AV NWP) model for the south of Western Siberia on 13 July 2014. The simulation result was compared with the Total Totals index values retrieved from product MOD07_L2 of spectroradiometer MODIS (the space platform "Terra"). Synoptic conditions and dangerous weather phenomena weather report for the selected date are also given in this study.
The present article describes a new concept of lightning-caused forest fire danger using a probabilistic criterion. The assessment of forest fire danger is made on the basis of the algorithm that classifies the forest territory by vegetation conditions. Lightning activity is processed by the MODIS spectroradiometer according to the World Wide Lightning Location Network data and remote sensing data for the Timiryazevskiy forestry in the Tomsk Region. The cluster analysis of the WWLLN and MOD06_L2 product data are used in the present paper.