loading page

An Accuracy Assessment of the Surface Reflectance Product from the EMIT Imaging Spectrometer
  • +11
  • Red Willow Coleman,
  • David R Thompson,
  • Philip G Brodrick,
  • Eyal Ben-Dor,
  • Evan Cox,
  • Carlos Pérez García-Pando,
  • Todd Hoefen,
  • Raymond Kokaly,
  • John M Meyer,
  • Francisco Ochoa,
  • Gregory S Okin,
  • Daniela Heller Pearlshtien,
  • Gregg Swayze,
  • Robert O Green
Red Willow Coleman
Jet Propulsion Laboratory, California Institute of Technology

Corresponding Author:[email protected]

Author Profile
David R Thompson
Jet Propulsion Laboratory, California Institute of Technology
Philip G Brodrick
Jet Propulsion Laboratory, California Institute of Technology
Eyal Ben-Dor
Tel Aviv University
Evan Cox
United States Geological Survey
Carlos Pérez García-Pando
Catalan Institution for Research and Advanced Studies, Barcelona Supercomputing Center
Todd Hoefen
United States Geological Survey
Raymond Kokaly
United States Geological Survey
John M Meyer
United States Geological Survey
Francisco Ochoa
University of California Los Angeles
Gregory S Okin
University of California Los Angeles
Daniela Heller Pearlshtien
Tel Aviv University
Gregg Swayze
United States Geological Survey
Robert O Green
Jet Propulsion Laboratory, California Institute of Technology

Abstract

The Earth surface Mineral dust source InvesTigation (EMIT) is an imaging spectrometer launched to the International Space Station in July 2022 to measure the mineral composition of Earth’s dust-producing regions. We present a systematic accuracy assessment of the EMIT surface reflectance product in two parts. First, we characterize the surface reflectance product’s overall performance using multiple independent vicarious calibration field experiments with hand-held and automated field spectrometers. We find that the EMIT surface reflectance product has a standard error of ±1.0% in absolute reflectance units for temporally coincident observations. Discrepancies rise to ±2.7 % for spectra acquired at different dates and times of day, which we attribute mainly to changes in solar geometry. Second, we develop an error budget that explains the differences between EMIT and in-situ field spectrometer data. We find that uncertainties in spatial footprints, field spectroscopy, and the EMIT-reported measurement were sufficient to explain discrepancies in most cases. Our approach did not detect any systematic calibration or reflectance errors in the timespan considered. Together, these findings demonstrate that a space-based imaging spectrometer can acquire high-quality spectra across a wide range of observational and atmospheric conditions.
16 Apr 2024Submitted to ESS Open Archive
16 Apr 2024Published in ESS Open Archive