Removal of ship mast contribution from MAGIC TSI fractional sky cover retrievals

 
Poster PDF

Authors

Victor R. Morris — Pacific Northwest National Laboratory
Jessica M Kleiss — Lewis and Clark College
Makayla Keydel — Lewis and Clark College

Category

General topics – Clouds

Description

(a) A sample TSI image with the ship mast clearly in the upper-right hand region. (b) The processed image containing the pixel-based cloud identification of this image. Blue indicates clear, grey indicates thin cloud, and white indicates opaque cloud.
The Marine ARM GPCI Investigation of Clouds (MAGIC) aimed to improve observations of the stratocumulus to-shallow-trade-wind-cumulus transition. These subtropical marine boundary-layer clouds play an important role in the earth’s radiative balance, albedo, and surface evaporation, yet the cloud-climate feedbacks are not yet well understood. The second ARM Mobile Facility (AMF2) was deployed aboard the cargo ship Horizon Spirit to gather atmospheric observations on a transit route between Los Angeles, California and Honolulu, Hawaii between October, 2012 and September, 2013. Among the suite of instruments deployed, the Total Sky Imager (TSI) captured hemispheric images of the sky during ship transit times, and calculated the image-based fractional sky cover. Unfortunately, the ship’s mast was included in the images and contributed to the sky cover retrievals. In this work, the location of the mast was determined automatically by constructing an image heat map of pixels identified as opaque clouds on otherwise clear days. The mast location was then removed from the images and the resulting sky cover was computed without the contribution of the ship’s mast. We present the net result of the mast removal on the fractional sky cover estimates for thin and opaque clouds and a discussion of additional possible errors. The corrected data set will be made available to the ARM Data Archive.