A Comparison of MODIS-Derived Cloud Fraction with Surface Observations at Five SURFRAD Sites

Abstract

25Clouds determine the amount of solar radiation incident to the surface. Accurately 26 quantifying cloud fraction is of great importance, but difficult to accomplish. Satellite 27 and surface cloud observations have different fields of view (FOVs); the lack of 28 conformity of different FOVs may cause large discrepancies when comparing 29 satellite-and surface-derived cloud fractions. From the viewpoint of surface-incident 30 solar radiation, this paper compares Moderate Resolution Imaging Spectroradiometer 31 (MODIS) Level-2 cloud-fraction data with three surface cloud-fraction datasets at five 32 surface-radiation budget network (SURFRAD) sites. The correlation coefficients 33 between MODIS and the surface cloud fractions are in the 0.80-0.91 range and vary at 34 different SURFRAD sites. In a number of cases, MODIS observations show a large 35 cloud-fraction bias compared with surface data. The variances between MODIS and 36 the surface cloud-fraction datasets are more apparent when small convective or 37 broken clouds exist in the FOVs. The magnitude of the discrepancy between MODIS 38 and surface-derived cloud fractions depends on the satellite view zenith angle (VZA) 39 of the satellite. On the average, compared to surface cloud-fraction data, MODIS 40 observes a larger cloud fraction at VZA >40 º and a smaller cloud fraction at VZA 41 <20 º . When comparing long-term MODIS averages with surface datasets, 42 Aqua/MODIS observes a higher annual mean cloud fraction, likely because 43 convective clouds are better developed in the afternoon when Aqua is observing.44 45 46 3

Can MODIS cloud fraction fully represent the diurnal and seasonal variations at DOE ARM SGP and Manus sites?

“…In order to be comparable among the MODIS, ARSCL, and TSC cloud fractions, we need to figure out the optimal matches in both time length and space width. When compared with the TSI or TSC ground‐based FOV CFs (half an hour or 1 h averaged CFs), MODIS CFs averaged in 30 km radius circle were used in many previous studies [ Li et al ., ; Kotarba , ; An and Wang , ], as we have mentioned before in section 2.…”

“…CFs (half an hour or 1 h averaged CFs), MODIS CFs averaged in 30 km radius circle were used in many previous studies [Li et al, 2004;Kotarba, 2009;An and Wang, 2015], as we have mentioned before in section 2.…”

“…They are often evaluated with surface observations, while the mismatch of spatial coverage could introduce extra errors in addition to the instrument errors. The issues associated with the spatial coverage have been studied by several studies [ Xi et al ., ; Ma et al ., ; Wu et al ., ; An and Wang , ], but few studies have discussed the errors in the cloud fraction caused by the time representation of satellite observations.…”

“…They are often evaluated with surface observations, while the mismatch of spatial coverage could introduce extra errors in addition to the instrument errors. The issues associated with the spatial coverage have been studied by several studies [Xi et al, 2010;Ma et al, 2014;Wu et al, 2014;An and Wang, 2015], but few studies have discussed the errors in the cloud fraction caused by the time representation of satellite observations. MODIS on board the Terra and Aqua satellites provides global coverage four times per day with local equatorial crossing time at 10:30 A.M. and 10:30 P.M. for Terra, and 01:30 A.M. and 01:30 P.M. for Aqua.…”

Long‐term change of total cloud cover and its possible reason over South China during 1960–2012

“…In recent decades, ground‐based observation and satellite remote sensing are the two most normally used cloud observation methods and have been widely used in many fields for spatial distribution (Evan et al, ; Maddux et al, ), change trends (An et al, ; Kazantzidis et al, ), cloud optical thickness (Hong et al, ) and aerosol effects (Forster et al, ; Li et al, ), and the results reflect the different characteristics of climate change over some regions. Among them, a decreasing trend of the TCC has been revealed over much of China during the past several decades (Xia, ; Zong et al, ), and the reason can be attributed to the changes in atmospheric circulation conditions (including surface solar radiation, temperature, precipitation, wind fields, and stratification conditions) caused by climate warming (Xia, ).…”