Uncertainties in cloud phase and optical thickness retrievals from the Earth Polychromatic Imaging Camera (EPIC)
This paper presents an investigation of the expected uncertainties of a single-channel cloud optical thickness (COT) retrieval technique, as well as a simple cloud-temperature-threshold-based thermodynamic phase approach, in support of the Deep Space Climate Observatory (DSCOVR) mission. DSCOVR clou...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Copernicus Publications
2016-04-01
|
Series: | Atmospheric Measurement Techniques |
Online Access: | http://www.atmos-meas-tech.net/9/1785/2016/amt-9-1785-2016.html |
Summary: | This paper presents an investigation of the expected uncertainties of a single-channel
cloud optical thickness (COT) retrieval technique, as well as a simple
cloud-temperature-threshold-based thermodynamic phase approach, in support of the
Deep Space Climate Observatory (DSCOVR) mission. DSCOVR cloud products will
be derived from Earth Polychromatic Imaging Camera (EPIC) observations in the
ultraviolet and visible spectra. Since EPIC is not equipped with a spectral
channel in the shortwave or mid-wave infrared that is sensitive to cloud
effective radius (CER), COT will be inferred from a single visible channel
with the assumption of appropriate CER values for liquid and ice phase
clouds. One month of Aqua MODerate-resolution Imaging Spectroradiometer
(MODIS) daytime granules from April 2005 is selected
for investigating cloud phase sensitivity, and a subset of these granules
that has similar EPIC Sun-view geometry is selected for investigating COT
uncertainties. EPIC COT retrievals are simulated with the same algorithm as
the operational MODIS cloud products (MOD06), except using fixed
phase-dependent CER values. Uncertainty estimates are derived by comparing
the single-channel COT retrievals with the baseline bi-spectral MODIS
retrievals. Results show that a single-channel COT retrieval is feasible for
EPIC. For ice clouds, single-channel retrieval errors are minimal (< 2 %) due to the particle size insensitivity of the assumed ice crystal
(i.e., severely roughened aggregate of hexagonal columns) scattering
properties at visible wavelengths, while for liquid clouds the error is
mostly limited to within 10 %, although for thin clouds (COT < 2)
the error can be higher. Potential uncertainties in EPIC cloud masking and
cloud temperature retrievals are not considered in this study. |
---|---|
ISSN: | 1867-1381 1867-8548 |