Earth and Atmospheric Sciences, Department of


First Advisor

Mark R. Anderson

Second Advisor

Clinton M. Rowe

Third Advisor

Steve Hu

Date of this Version


Document Type



A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfilment of Requirements For the Degree of Master of Science, Major: Earth and Atmospheric Sciences, Under the Supervision of Professor Mark R. Anderson. Lincoln, Nebraska: July, 2019

Copyright © 2019 Ambrish Sharma


Gas flaring is a commonly used practice for disposing of waste gases emerging from industrial oil drilling and production processes. It is a serious environmental and economic hazard with adverse impacts on air quality, climate, and the public health. Accurate determination of flare locations and estimation of associated emissions are therefore of prime importance. Recently developed Visible Infrared Imaging Radiometer Suite (VIIRS) Nightfire product (VNF) has shown remarkable efficiency in detecting gas flares globally, owing primarily to its use of Shortwave Infrared (SWIR) band in its detection algorithm. This study compares and contrast nocturnal hot source detection by VNF to detections by other established fire detection products (i.e., Moderate Resolution Imaging Spectroradiometer (MODIS) Terra Thermal Anomalies product (MOD14), MODIS Aqua Thermal Anomalies product (MYD14) and VIIRS Applications Related Active Fire Product (VAFP)) over an extensive gas flaring region in Russia ‑ Khanty Mansiysk Autonomous Okrug, for the time period of April - August 2013. The surface hotspots detected by VNF were found to be much higher in magnitude than detected by other products. An attempt to replicate VNF algorithm locally for better comprehension, revealed threshold related discrepancies in VNF V1.0 in multiple spectral bands. Case studies for reconciliation between VNF‑R (VNF replicated product) and VAFP hotspots showed that convergence in hotspot detection between two products is possible by scaling up VNF-R thresholds, and, VAFP can detect large flares having strong spectral signature in SWIR bands. The efficacy of VNF hotspot detection was evaluated for 10 previously identified flare locations with varying hot source sizes over the period of April‑August 2013. VNF was able to detect all the test sites with frequency of detection varying between 20% to 42% of the days tested. Mean areas of tested gas flares estimated by VNF showed good agreement with areas of flares computed using Google Earth with a linear correlation of 0.91; however, VNF estimated areas were found to be somewhat underestimated. Overall the results indicate significant potential of VNF in characterizing gas flaring from space.

Advisor: Mark R. Anderson