Cloud top height (CTH) plays an important role in the Earth’s radiation budget and thus in climate change.
However, accurate CTH retrieval based on satellite data remains a challenge when using currently widespread
methods based on passive remote sensing satellite data, particularly over the Arctic Ocean. It is well known that
using geometric correlation between CTH and cloud-shadow distance along the normal direction of sunlight,
cloud-edge top height (CETH) can be determined. However, this cloud-shadow method (CSM) exhibited larger
CTH errors due to low-resolution weather satellite data of the past. The aim of the paper is to analyze the effect of
higher-resolution images (MODIS with 250-m resolution) on CSM accuracy. We present a method in which CETH
is first retrieved using the MODIS data, and to calculate CTH in the center of cloud region by combining the
calculated CETH and cloud top brightness temperature using an environmental temperature lapse rate method.
We validated 14 cases over the Arctic Ocean by the CALIPSO (Cloud-Aerosol Lidar and Infrared Pathfinder
Satellite Observations) CTH lidar product and another two cases by radiosonde data. CSM revealed better per-
formance (mean difference is 0.45 km and standard deviation is 0.92 km) compared to the operational MODIS
CTH product (mean difference is 0.78 km and standard deviation is 1.78 km), when validated by the CALIPSO
product. When validated by sounding data, CSM also performed better than the MODIS CTH product: the CSM
CTH errors are 0.41 and 0.52 km in two sounding cases, while the corresponding MODIS CTH errors are 1.10
and 1.07 km. The CSM technique especially showed better performance for double-deck cloud systems. We
conclude that further improvements in CTH accuracy based on CSM can be achieved by mainly using higher-
resolution satellite imagery.