Document Type

Dissertation - Open Access

Award Date


Degree Name

Doctor of Philosophy (PhD)


Geospatial Science and Engineering

First Advisor

Xiaoyang Zhang


biomass burning, CONUS, FRP, GOES, MODIS, VIIRS


Biomass burning is an important source of atmospheric greenhouse gases and aerosol emissions that significantly influence climate and air quality. Estimation of biomassburning emissions (BBE) has been limited to the conventional method in which parameters (i.e., burned area and fuel load) can be challenging to quantify accurately. Recent studies have demonstrated that the rate of biomass combustion is a linear function of fire radiative power (FRP), the instantaneous radiative energy released from actively burning fires, which provides a novel pathway to estimate BBE. To obtain accurate and timely BBE estimates for near real-time applications (i.e., air quality forecast), the satellite FRP-based method first requires a reliable biomass combustion coefficient that converts fire radiative energy (FRE), the temporal integration of FRP, to biomass consumption. The combustion coefficient is often derived in controlled small-scale fire experiments and is assumed a constant, whereas the coefficient based on satellite retrievals of FRP and atmospheric optical depth is suggested varying in a wide range. Undoubtedly, highly variable combustion coefficient results in large uncertainty of BBE estimates. Further, the FRP-based method also depends on high-spatiotemporalresolution FRP retrievals that, however, are not available in any active fire products from current polar-orbiting and geostationary satellites due to their sampling limitations. To address these challenges, this study first investigates the combustion coefficient for landscape-scale wildfires in the Conterminous United States (CONUS) by comparing FRE from the polar-orbiting Moderate Resolution Imaging Spectroradiometer (MODIS) and the Geostationary Operational Environmental Satellite system (GOES) with the Landsat-based biomass consumption. The results confirms that biomass consumption is a linear function of FRE for wildfires. The derived combustion coefficient is 0.374 kg · MJ- 1 for GOES FRE, 0.266 kg · MJ-1 for MODIS FRE, and 0.320 kg · MJ-1 considering both GOES and MODIS FRE in the CONUS. Limited sensitivity analyses indicate that the combustion coefficient varies from 0.301 to 0.458 kg · MJ-1, which is similar to the reported values in small fire experiments. Then, this study reconstructs diurnal FRP cycle to derive high-spatiotemporal-resolution FRP by fusing MODIS and GOES FRP retrievals and estimates hourly BBE at a 0.25°×0.3125° grid across the CONUS. The results indicate that the reconstructed diurnal FRP cycle varies significantly in magnitude and shape among 45 CONUS ecosystems. In the CONUS, the biomass burning annually releases approximately 690 Gg particulate matter (smaller than 2.5 μm in diameter, PM2.5). The diurnal-FRP-cycle-based BBE estimates compare well with BBE derived from Landsat burned areas in the western CONUS and with the hourly carbon monoxide emissions simulated using a biogeochemical model over the Rim Fire in California. Moreover, the BBE estimates show a similar seasonal variation to six existing BBE inventories but with variable magnitude. Finally, this study examines potential improvements in fires characterization capability of the Visible Infrared Imaging Radiometer Suite (VIIRS), which is the follow-on sensor of the MODIS sensor, for integrating VIIRS FRP retrievals into the FRP-based method for BBE estimation in future work. The results indicate that the VIIRS fire characterization capability is similar across swath, whereas MODIS is strongly dependent on satellite view zenith angle. VIIRS FRP is generally comparable with contemporaneous MODIS FRP at continental scales and in most fire clusters. At 1-degree grid cells, the FRP difference between the two sensors is, on average, approximately 20% in fire-prone regions but varies significantly in fire-limited regions. In summary, this study attempts to enhance the capability of the FRP-based method by addressing challenges in its two parameters (combustion coefficient and FRP), which should help to improve estimation of BBE and advance our understanding of the effects of BBE on climate and air quality. This research has resulted in two published papers and one paper to be submitted to a peer-reviewed journal so far.


Includes bibliographical references



Number of Pages



South Dakota State University


In Copyright - Educational Use Permitted

Available for download on Saturday, August 10, 2019