Estimation of daily evapotranspiration over Africa using MODIS/Terra and SEVIRI/MSG data Zhigang Sun a , Mekonnen Gebremichael a, ⁎, Jonas Ardö b , Alecia Nickless c , Blandine Caquet d, e , Lutz Merboldh f , Werner Kutschi g a Civil & Environmental Engineering, University of Connecticut, 261 Glenbrook Road, Storrs, CT 06269‐2037, USA b Department of Earth and Ecosystem Sciences, Lund University, Lund, Sweden c Natural Resources and the Environment, CSIR, Pretoria, PO Box 395, 0001, South Africa d CIRAD, Persyst, UPR80, TA10/D, 34398 Montpellier Cedex 5, France e CRDPI, BP 1291, Pointe-Noire, People's Republic of Congo f ETH Zurich, Institute of Plant-, Animal- and Agroecosystem Sciences, Switzerland g Institute of Agricultural Climate Research, Braunschweig, Germany article info abstract Article history: Received 13 May 2011 Received in revised form 1 November 2011 Accepted 9 April 2012 Most existing remote sensing-based evapotranspiration (ET) algorithms rely exclusively on polar-orbiting satellites with thermal infrared sensors, and therefore the resulting ET values represent only “instantaneous or snapshot” values. However, daily ET is more meaningful and useful in applications. In this study, daily ET estimates are obtained by combining data from the MODIS sensor aboard the polar-orbiting Terra satellite and the SEVIRI sensor aboard the geostationary-orbiting MSG satellite. The procedure consists of estimating the instantaneous evaporative fraction (EF) based on the MODIS/Terra land data products, and estimating the daily net radiation and daily available energy based on the 30-min SEVIRI/MSG data products. Assuming constant EF during the daytime, daily ET is estimated as the product of the SEVIRI/ MSG-based daily available energy and MODIS/Terra-based instantaneous EF. The daily ET estimates are evaluated against flux tower measurements at four validation sites in Africa. Results indicate that the synergistic use of SEVIRI/MSG and MODIS/Terra has the potential to provide reliable estimates of daily ET during wet periods when daily ET exceeds 1 mm/day. The satellite-based daily ET estimates however tend to underestimate ET by 13% to 35%. The daily ET estimation algorithm can further be improved by incorporating a temporal data-filling interpolation technique to estimate the unavailable net radiation information during cloudy sky conditions, and by improving the accuracy of the instantaneous EF. The assumption of constant evaporative fraction during the daytime is reasonable, and does not result in substantial errors in the daily ET estimates. © 2012 Elsevier B.V. All rights reserved. Keywords: Daily evapotranspiration MODIS Terra SEVIRI MSG 1. Introduction Regional evapotranspiration (ET) from land surface is essential to understand water cycle, and to estimate surface runoff and groundwater (Idso et al., 1975; Shukla and Mintz, 1982). Routine high quality data from satellite remote sensing has allowed capturing continuous ET maps on a wide range of spatio-temporal scales (e.g., Nagler et al., 2005; Mu et al., 2007; Leuning et al., 2008). Most existing remote sensing-based ET algorithms rely exclusively on polar- orbiting satellites with thermal infrared sensors, and there- fore the resulting ET values represent only “instantaneous or snapshot” values (e.g., Wang et al., 2007; Venturim et al., 2008). However, daily ET is more meaningful and useful in applications. Atmospheric Research 112 (2012) 35–44 ⁎ Corresponding author at: Civil & Environmental Engineering, University of Connecticut, 261 Glenbrook Road, Unit 2037, Storrs, CT 06269‐2037, USA. Tel.: +1 860 486 2771. E-mail address: mekonnen@engr.uconn.edu (M. Gebremichael). 0169-8095/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.atmosres.2012.04.005 Contents lists available at SciVerse ScienceDirect Atmospheric Research journal homepage: www.elsevier.com/locate/atmos