Previous | View all abstracts | Next

A VALIDATION OF THE MODIS CLOUD MASK OVER GREENLAND

STARKWEATHER, SANDY  CIRES, Univ. of Colorado.
Steffen, Konrad  CIRES, Univ. of Colorado.

The results of a validation study of the Moderate-resolution Imaging Spectroradiometer (MODIS) Cloud Mask (MOD_35) data product will be presented. The goal of this work is to determine the range of errors to be expected in using this data set to develop a cloud cover climatology over Greenland’s high plateau. It is expected that there will be shortcomings in the MOD_35 product because of the inherent difficulties of high latitude cloud detection. This validation will investigate systematic errors in order to understand biases in MOD_35 as well as to provide feedback to the authors of the algorithm. MOD_35 is used as an input for a suite of other data products in the NASA’s Earth Observing System (EOS) family. It is important to understand how errors in the MOD_35 will propagate through other data sets. Results indicate that MOD_35 performs very well during daylight hours and reasonably well during nighttime hours. There is a clear sky detection bias in data from all months. With the exception of midwinter cloud detection, most shortcomings can be improved with careful filtering. This validation is a single point analysis that will be extrapolated over Greenland’s dry snow region. This analysis is a first step towards achieving a robust cloud climatology over all of Greenland with MODIS.

Previous | View all abstracts | Next