Premium
Estimating errors in cloud amount and cloud optical thickness due to limited spatial sampling using a satellite imager as a proxy for nadir‐view sensors
Author(s) -
Liu Yinghui
Publication year - 2015
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1002/2015jd023507
Subject(s) - nadir , environmental science , remote sensing , moderate resolution imaging spectroradiometer , satellite , cloud computing , cloud height , latitude , cloud fraction , ice cloud , radiance , sampling (signal processing) , cloud top , cloud cover , meteorology , atmospheric sciences , geodesy , geology , geography , computer science , optics , physics , detector , operating system , astronomy
Cloud climatologies from space‐based active sensors have been used in climate and other studies without their uncertainties specified. This study quantifies the errors in monthly mean cloud amount and optical thickness due to the limited spatial sampling of space‐based active sensors. Nadir‐view observations from a satellite imager, the Moderate Resolution Imaging Spectroradiometer (MODIS), serve as a proxy for those active sensors and observations within 10° of the sensor's nadir view serve as truth for data from 2003 to 2013 in the Arctic. June–July monthly mean cloud amount and liquid water and ice cloud optical thickness from MODIS for both observations are calculated and compared. Results show that errors increase with decreasing sample numbers for monthly means in cloud amount and cloud optical thickness. The root‐mean‐square error of monthly mean cloud amount from nadir‐view observations increases with lower latitudes, with 0.7% (1.4%) at 80°N and 4.2% (11.2%) at 60°N using data from 2003 to 2013 (from 2012). For a 100 km resolution Equal‐Area Scalable Earth Grid (EASE‐Grid) cell of 1000 sample numbers, the absolute differences in these two monthly mean cloud amounts are less than 6.5% (9.0%, 11.5%) with an 80 (90, 95)%chance; such differences decrease to 4.0% (5.0%, 6.5%) with 5000 sample numbers. For a 100 km resolution EASE‐Grid of 1000 sample numbers, the absolute differences in these two monthly mean cloud optical thicknesses are less than 2.7 (3.8) with a 90% chance for liquid water cloud (ice cloud); such differences decrease to 1.3 (1.0) for 5000 sample numbers. The uncertainties in monthly mean cloud amount and optical thickness estimated in this study may provide useful information for applying cloud climatologies from active sensors in climate studies and suggest the need for future spaceborne active sensors with a wide swath.