Validation of the MODIS snow product and cloud mask using student and NWS cooperative station observations in the Lower Great Lakes Region

被引:111
作者
Ault, Timothy W.
Czajkowski, Kevin P.
Benko, Teresa
Coss, James
Struble, Janet
Spongberg, Alison
Templin, Mark
Gross, Christopher
机构
[1] Univ Toledo, Dept Geog & Planning, Toledo, OH 43606 USA
[2] Univ Toledo, Dept Earth Ecol & Environm Sci, Toledo, OH 43606 USA
[3] Univ Toledo, Coll Educ, Toledo, OH 43606 USA
[4] W Chester Univ, W Chester, PA 19383 USA
基金
美国国家航空航天局; 美国国家科学基金会;
关键词
GLOBE; SATELLITES; Great Lakes; snow; cloud; student; MODIS; validation; MOD10; MOD10_L2; NASA; Liberal Cloud Mask; LCM;
D O I
10.1016/j.rse.2006.07.004
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
NASA's Moderate Resolution Imaging Spectroradiometer (MODIS) snow product (MOD10) creates automated daily, 8-day composite and monthly regional and global snow cover maps. In this study, the MOD10 daily swath imagery (MOD10-L2) and the MODIS cloud mask (MOD35) were validated in the Lower Great Lakes Region, specifically the area to the east of Lake Michigan. Validation of the MOD10_L2 snow product, MOD35 cloud mask and the MOD10_L2 Liberal Cloud Mask was performed using field observations from K-12 student GLOBE (Global Learning and Observations to Benefit the Environment) and SATELLITES (Students And Teachers Evaluating Local Landscapes to Interpret The Earth from Space) programs. Student data consisted of field observations of snow depth, snow water equivalency, cloud type, and total cloud cover. In addition, observations from the National Weather Service (NWS) Cooperative Observing Stations were used. Student observations were taken during field campaigns in the winter of 2001-2002, a winter with very little snow in the Great Lakes region, and the winters of 2000-2001 and 2002-2003, which had significant snow cover. Validation of the MOD10_L2 version 4 snow product with student observations produced an accuracy of 92% while comparison with the NWS stations produced an accuracy of 86%. The higher NWS error appears to come from forested areas. Twenty-five and fifty percent of the errors observed by the students and NWS stations, respectively, occurred when there was only a trace of snow. In addition, 82% of the MODIS cloud masked pixels were identified as either overcast or broken by the student observers while 74% of the pixels the MODIS cloud mask identified as cloudless were identified as clear, isolated or scattered cloud cover by the student observers. The experimental Liberal Cloud Mask eliminated some common errors associated with the MOD35 cloud mask, however, it was found to omit significant cloud cover. (c) 2006 Elsevier Inc. All rights reserved.
引用
收藏
页码:341 / 353
页数:13
相关论文
共 45 条
[1]  
Ackerman S, 1997, Discriminating Clear-Sky From Cloud With MODIS Algorithm Theoretical Basis Document (MOD35)
[2]  
[Anonymous], ATLAS SATELLITE OBSE
[3]  
Brown RD, 2000, J CLIMATE, V13, P2339, DOI 10.1175/1520-0442(2000)013<2339:NHSCVA>2.0.CO
[4]  
2
[5]  
CARROLL T, 2001, 69 ANN M W SNOW C 20
[6]   SATELLITE DISCRIMINATION OF SNOW CLOUD SURFACES [J].
CRANE, RG ;
ANDERSON, MR .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 1984, 5 (01) :213-223
[7]  
CROOK AG, 1977, J SOIL WATER CONSERV, V32, P294
[8]  
Doesken N.J., 1997, A Snow Booklet, A Guide to the Science, Climatology, and Measurement of Snow in the United States
[9]  
DOESKEN NJ, 2000, WEATHERWISE, V53, P30
[10]   SPECTRAL SIGNATURE OF ALPINE SNOW COVER FROM THE LANDSAT THEMATIC MAPPER [J].
DOZIER, J .
REMOTE SENSING OF ENVIRONMENT, 1989, 28 :9-&