Adjusting ISCCP cloud detection to increase consistency of cloud amount and reduce artifacts

View More View Less
  • 1 NOAA/National Center for Environmental Information, Asheville, NC, 28801
  • 2 NOAA/National Center for Environmental Information, Boulder, CO, 80305
  • 3 NOAA/National Center for Environmental Information, Asheville, NC, 28801
  • 4 North Carolina Institute for Climate and Satellites, Asheville, NC, 28801
  • 5 Riverside Technology, Inc., Asheville, NC, 28801
© Get Permissions
Restricted access

Abstract

The International Satellite Cloud Climatology Project (ISCCP) began collecting data in the 1980s to help understand the distribution of clouds. Since then, it has provided important information on clouds in time and space and their radiative characteristics. However, it was apparent from some long term time series of the data that there are some latent artifacts related to the changing satellite coverage over the more than thirty years of the record. Changes in satellite coverage effectively create secular changes in the time series of view zenith angle (VZA) for a given location. There is an inconsistency in the current ISCCP cloud detection algorithm related to VZA: two satellites viewing the same location from different VZAs can produce vastly different estimates of cloud amount. Research is presented which shows that a simple change to the cloud detection algorithm can vastly increase the consistency. This is accomplished by making the cloud/no cloud threshold VZA dependent. The resulting cloud amounts are more consistent between different satellites and the distributions are shown to be more spatially homogenous. Likewise, the more consistent spatial data leads to more consistent temporal statistics.

Corresponding Author: Ken Knapp, 151 Patton Ave., Asheville, NC, Ken.Knapp@noaa.gov, 828-271-4339

Now at NOAA/Center for Satellite Applications and Research, College Park, MD, 20740.

Abstract

The International Satellite Cloud Climatology Project (ISCCP) began collecting data in the 1980s to help understand the distribution of clouds. Since then, it has provided important information on clouds in time and space and their radiative characteristics. However, it was apparent from some long term time series of the data that there are some latent artifacts related to the changing satellite coverage over the more than thirty years of the record. Changes in satellite coverage effectively create secular changes in the time series of view zenith angle (VZA) for a given location. There is an inconsistency in the current ISCCP cloud detection algorithm related to VZA: two satellites viewing the same location from different VZAs can produce vastly different estimates of cloud amount. Research is presented which shows that a simple change to the cloud detection algorithm can vastly increase the consistency. This is accomplished by making the cloud/no cloud threshold VZA dependent. The resulting cloud amounts are more consistent between different satellites and the distributions are shown to be more spatially homogenous. Likewise, the more consistent spatial data leads to more consistent temporal statistics.

Corresponding Author: Ken Knapp, 151 Patton Ave., Asheville, NC, Ken.Knapp@noaa.gov, 828-271-4339

Now at NOAA/Center for Satellite Applications and Research, College Park, MD, 20740.

Save