Fire Detections and Fire Radiative Power Intercomparison Using Multiple Sensor Products over a Predominantly Gas Flaring Region

Friday, 19 December 2014
Ambrish Sharma and Jun Wang, University of Nebraska Lincoln, Lincoln, NE, United States
Gas flaring is a global environmental hazard severely impacting climate, economy and public health. The associated emissions are frequently unreported and have large uncertainties. Prior studies have established a direct relationship between radiative energy released from fires and the biomass burned, making fire radiative power (FRP), i.e., the rate of radiative energy release, an important proxy to characterize emissions. In this study fire properties from four different satellite products were obtained over a 10⁰ x 10⁰ gas flaring region in Russia for all days of May 2013. The target area is part of Russia’s biggest gas flaring region, Khanty-Mansiysk autonomous okrug. The objective of the study is to investigate the consistency of fire detections, FRP retrievals and effects of gridding FRP data from the region on a uniform grid. The four products used were: MODIS Terra level2 thermal anomalies (MOD14), MODIS Aqua level2 thermal anomalies (MYD14), VIIRS Active fire product and a recent NOAA Nightfire product. 1 km nominal resolution FRP from MOD14 AND MYD14, subpixel radiant heat (RH) from NOAA Nightfire product and fire detections from all four products were recorded on a 0.25⁰ x 0.25⁰ grid on a daily basis. Results revealed the Nightfire product had maximum detections, almost six times the number of detections by other products, mainly because of the use of M10 (1.6 µm) band as their primary detection band. The M10 band is highly efficient in identifying radiant emissions from hot sources during night-time. The correlation (after omitting outliers) between gridded NOAA Nightfire RH and corresponding MOD14 FRP and MYD14 FRP gave a moderate regression value, with MODIS FRP being mostly higher than RH. As an extension to this work, a comprehensive study for a larger temporal domain also incorporating viewing geometries and cloud cover would advance our understanding of flare detections and associated FRP retrievals not just for the target region but also gas flaring regions globally.