Intercomparison of MODIS and VIIRS Fire Products in Khanty-Mansiysk Russia: Implications for Characterizing Gas Flaring from Space

Authors

Ambrish Sharma, University of Nebraska-LincolnFollow
Jun Wang, University of IowaFollow
Elizabeth M. Lennartson, University of Iowa

Document Type

Article

Date of this Version

2017

Citation

Sharma, Wang, & Lennartson in MDPI Atmosphere (2017) 8, 95.

doi:10.3390/atmos8060095

Comments

Copyright © 2017, the authors. Licensee MDPI, Basel, Switzerland. Open access, Creative Commons Attribution license 4.0.

Abstract

Gas flaring is commonly used by industrial plants for processing oil and natural gases in the atmosphere, and hence is an important anthropogenic source for various pollutants including CO₂, CO, and aerosols. This study evaluates the feasibility of using satellite data to characterize gas flaring from space by focusing on the Khanty-Mansiysk Autonomous Okrug in Russia, a region that is well known for its dominant gas flaring activities. Multiple satellite-based thermal anomaly data products at night are intercompared and analyzed, including MODIS (Moderate Resolution Imaging Spectroradiometer) Terra level 2 Thermal Anomalies product (MOD14), MODIS Aqua level 2 Thermal Anomalies product (MYD14), VIIRS (Visible Infrared Imaging Radiometer Suite) Active Fires Applications Related Product (VAFP), and VIIRS level 2 Nightfire product (VNF). The analysis compares and contrasts the efficacy of these sensor products in detecting small, hot sources like flares on the ground in extremely cold environments such as Russia. We found that the VNF algorithm recently launched by the National Oceanic and Atmospheric Administration (NOAA) has the unprecedented accuracy and efficiency in characterizing gas flares in the region owing primarily to the use of Shortwave Infrared (SWIR) bands. Reconciliation of VNF’s differences and similarities with other nighttime fire products is also conducted, indicating that MOD14/MYD14 and VAFP data are only effective in detecting those gas flaring pixels that are among the hottest in the region; incorporation of shortwave infrared (1.6 _m) band used in VNF may improve the detection of relatively cooler gas flares. The gas flaring locations from the VNF product are validated using Google Earth images. It is shown that VNF’s estimates of the area of gas flaring agree well with the Google image counterparts with a linear correlation of 0.91, highlighting its potential use for routinely monitoring emissions of gas flaring from space.