Comparison of Fire Radiative Power Estimates From VIIRS and MODIS Observations

Abstract

Satellite‐based active fire data are a viable tool to understand the role of global fires in the biosphere and atmosphere. The Moderate Resolution Imaging Spectroradiometer (MODIS) sensors on Aqua and Terra satellites are nearing the end of their lives. The Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on the Suomi National Polar‐orbiting Partnership satellite and the subsequent Joint Polar Satellite System series is expected to extend the MODIS active fire record. Thus, understanding the similarities of and discrepancies between the two data sets during their overlap period is important for existing applications. This study investigated the dependence of the MODIS and VIIRS fire characterization capabilities on satellite view zenith angle and the relationship between the two sensors' fire radiative power (FRP) from individual fire clusters to fire data on continental and global scales. The results indicate that the VIIRS fire characterization capability is similar across swath, whereas MODIS is strongly dependent on view zenith angle. Statistical analyses reveal that the VIIRS and MODIS FRP relationship varies between different spatial scales. In fire clusters, MODIS and VIIRS FRP estimates are very comparable, except for large boreal forest fires where VIIRS FRP is approximately 47% smaller. At the continental scale, the contemporaneous FRP retrievals from MODIS and VIIRS are generally comparable and strongly correlated, but VIIRS FRP is slightly larger and their differences vary across seasons. At global 1° × 1° grids, the FRP difference between the two sensors is, on average, approximately 20% in fire‐prone regions but varies significantly in fire‐limited regions.

Investigating Smoke Aerosol Emission Coefficients Using MODIS Active Fire and Aerosol Products: A Case Study in the CONUS and Indonesia

“…Actively burning fires are detected by exploiting the strong emission of middle and thermal-infrared radiation from fires at 4-and 11-μm bands (Giglio et al, 2003;Wooster et al, 2003). The MODIS C6 active fire detection algorithm can detect active fires covering as little as 10 −3 to 10 −4 of a fire pixel (1 km × 1 km) when fire temperature is 1,000 k or higher (Giglio et al, 2003(Giglio et al, , 2016, and capture FRP as low as 2.5 MW at the nadir (Li, Zhang, Kondragunta, & Csiszar, 2018). The products also include a fire mask that classifies each pixel of a 5-min MODIS granule into one of eight categories: not processed, nonfire water, cloud, nonfire land, unknown, low confidence fire (0%-30%), nominal confidence fire (30%-80%), and high confidence fire (80%-100%; Giglio et al, 2016).…”

“…In addition, due to the MODIS "bow-tie effect" (adjacent scans overlap each other at off-nadir; Wolfe et al, 2002), the same fire in the offnadir region may be observed twice or even more times (Freeborn et al, 2014;Peterson et al, 2013). Therefore, the interscan duplicated fire detections were corrected following the method proposed in a previous study (Li, Zhang, Kondragunta, & Csiszar, 2018). The corrected MODIS daytime FRP is simply referred to as the FRP hereafter.…”

Missing emissions from post-monsoon agricultural fires in northwestern India: regional limitations of MODIS burned area and active fire products

“…Even so, VIIRS would not be able detect fires obscured by haze and clouds and those outside of its overpass time. Li et al (2018) showed that even slight differences in VIIRS and MODIS/Aqua overpasses of ∼15 min can lead to large discrepancies in active fire detections over Punjab. In addition, cloud cover and increasing haziness, indicated by AOD, can limit retrieved scenes that are usable and block active fires from satellite detection .…”

Progress and Challenges in Quantifying Wildfire Smoke Emissions, Their Properties, Transport, and Atmospheric Impacts

“…Our goal has been to provide an introduction to this special issue and to provide a review of current fire‐smoke emissions characterization, data, and modeling. There are several dominating themes that permeate this special issue and review that include the linkages and dependency of outcomes on ecosystem fuels (Desservettaz et al, ; Gomez et al, ; H Lee et al, ; X Liu et al, ; Petrenko et al, , all this issue) and the interdependencies between the terrestrial, atmosphere, and climate domains (e.g., fuels, fire weather, detrainment and transport, chemistry, deposition of smoke, impacts on radiation, and precipitation; Antokhin et al, ; Bluvshtein et al, ; Kalashnikova et al, ; F. Li, et al, ; Lu & Sokolik, ; Souri et al, ; Wang et al, ; Zhu et al, , all this issue). The ultimate purpose of understanding fire‐smoke emissions, their properties, transport, and atmospheric impacts is twofold: to clarify chemistry and transport to improve air quality and health and to accurately integrate and model feedback within climatic systems.…”