| Popis: |
Lightning ignition is the major cause of natural wildfires in several regions worldwide. Determining if wildfires in remote uncontrolled areas result from natural lightning as opposed to anthropic action is a relevant and yet-unsolved challenge for large regions of the planet, with scientific and management implications ranging from environmental conservation to mitigation of climate-related emissions of gases and aerosols. Brazil is the country with one of the highest occurrences of lightning (50 to 100 million/year) and which is also subject to numerous and vast wildfires (up to ∼600 × 103 km2/year) affecting all its biomes. To quantify natural fires we combined cloud-to-ground (CG) lightning and CG dry-lightning (CGDL) detected by a ground network, with fire pixels mapped by satellite remote sensing (AQUA, S-NPP and NOAA-20) over ∼1,8 × 106 km2 in Central Brazil, between 2015 to 2019. Lightning ignition candidates were selected based on the distance between fires and lightning in time and space. The selected cases were investigated according to annual and monthly distributions in space and time, to local weather at the time of occurrence and, electrical characteristics related to ignition. Space-time distributions of CG lightning, CGDL and of active fires were also analyzed. Results showed that the CGDLs pattern is not different from that of the overall CG lightning, with both presenting similar kernel density, polarity and peak current. The lightning candidates indicated predominance of negative polarity and peak current frequency below 20 kA. In this range, average values for weather conditions for CG lightning matched to fires (CGDL matched to fires) had: precipitation 6 mm ( |
Full Text from ScienceDirect