Advanced

Estimation of diurnal air temperature using MSG SEVIRI data in West Africa

Stisen, Simon; Sandholt, Inge; Norgaard, A; Fensholt, Rasmus and Eklundh, Lars LU (2007) In Remote Sensing of Environment 110. p.262-274
Abstract
Spatially distributed air temperature data with high temporal resolution are desired for several modeling applications. By exploiting the thermal split window channels in combination with the red and near infrared channels of the geostationary MSG SEVIRI sensor, multiple daily air temperature estimates can be achieved using the contextual temperature-vegetation index method. Air temperature was estimated for 436 image acquisitions during the 2005 rainy season over West Africa and evaluated against in situ data from a field test site in Dahra, Northern Senegal. The methodology was adjusted using data from the test site resulting in RMSE=2.55 K, MBE=-0.30 K and R-2=0.63 for the estimated versus observed air temperatures. A spatial validation... (More)
Spatially distributed air temperature data with high temporal resolution are desired for several modeling applications. By exploiting the thermal split window channels in combination with the red and near infrared channels of the geostationary MSG SEVIRI sensor, multiple daily air temperature estimates can be achieved using the contextual temperature-vegetation index method. Air temperature was estimated for 436 image acquisitions during the 2005 rainy season over West Africa and evaluated against in situ data from a field test site in Dahra, Northern Senegal. The methodology was adjusted using data from the test site resulting in RMSE=2.55 K, MBE=-0.30 K and R-2=0.63 for the estimated versus observed air temperatures. A spatial validation of the method using 12 synoptic weather stations from Senegal and Mali within the Senegal River basin resulted in overall values of RMSE=2.96 K, MBE=-1.11 K and R-2=0.68. The daytime temperature curve is interpolated using a sine function based on the multiple daily air temperature estimates from the SEVIRI data. These estimates (covering the 8:00-20:00 UCT time window) were in good agreement with observed values with RMSE=2.99 K, MBE=-0.70 K and R-2=0.64. The temperature-vegetation index method was applied as a moving window technique to produce distributed maps of air temperature with 15 min intervals and 3 km spatial resolution for application in a distributed hydrological model. (c) 2007 Elsevier Inc. All rights reserved. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
meteosat, TIMESAT, satellite
in
Remote Sensing of Environment
volume
110
pages
262 - 274
publisher
Elsevier
external identifiers
  • wos:000249161200011
  • scopus:34547416619
ISSN
0034-4257
DOI
10.1016/j.rse.2007.02.025
language
English
LU publication?
yes
id
bc9c5e6c-6d4b-4ef5-811a-bf88376b93e7 (old id 694657)
date added to LUP
2007-12-17 13:21:35
date last changed
2017-11-05 04:41:19
@article{bc9c5e6c-6d4b-4ef5-811a-bf88376b93e7,
  abstract     = {Spatially distributed air temperature data with high temporal resolution are desired for several modeling applications. By exploiting the thermal split window channels in combination with the red and near infrared channels of the geostationary MSG SEVIRI sensor, multiple daily air temperature estimates can be achieved using the contextual temperature-vegetation index method. Air temperature was estimated for 436 image acquisitions during the 2005 rainy season over West Africa and evaluated against in situ data from a field test site in Dahra, Northern Senegal. The methodology was adjusted using data from the test site resulting in RMSE=2.55 K, MBE=-0.30 K and R-2=0.63 for the estimated versus observed air temperatures. A spatial validation of the method using 12 synoptic weather stations from Senegal and Mali within the Senegal River basin resulted in overall values of RMSE=2.96 K, MBE=-1.11 K and R-2=0.68. The daytime temperature curve is interpolated using a sine function based on the multiple daily air temperature estimates from the SEVIRI data. These estimates (covering the 8:00-20:00 UCT time window) were in good agreement with observed values with RMSE=2.99 K, MBE=-0.70 K and R-2=0.64. The temperature-vegetation index method was applied as a moving window technique to produce distributed maps of air temperature with 15 min intervals and 3 km spatial resolution for application in a distributed hydrological model. (c) 2007 Elsevier Inc. All rights reserved.},
  author       = {Stisen, Simon and Sandholt, Inge and Norgaard, A and Fensholt, Rasmus and Eklundh, Lars},
  issn         = {0034-4257},
  keyword      = {meteosat,TIMESAT,satellite},
  language     = {eng},
  pages        = {262--274},
  publisher    = {Elsevier},
  series       = {Remote Sensing of Environment},
  title        = {Estimation of diurnal air temperature using MSG SEVIRI data in West Africa},
  url          = {http://dx.doi.org/10.1016/j.rse.2007.02.025},
  volume       = {110},
  year         = {2007},
}