Solar Irradiance and Pan Evaporation Estimation from Meteorological Satellite Data

  • Author(s): Ming-Ren Syu, Ping-Ho Lee, Tzay-Ming Leou, and Yuan Shen
  • DOI: 10.3319/TAO.2015.11.11.01(A)
  • Keywords: Geostationary satellite, Solar radiation, Evapotranspiration, Water Resources
  • Citation: Syu, M. R., P. H. Lee, T. M. Leou, and Y. Shen, 2016: Solar irradiance and pan evaporation estimation from meteorological satellite data. Terr. Atmos. Ocean. Sci., 27, 221-239, doi: 10.3319/TAO.2015.11.11.01(A)

Knowledge about spatial and temporal variations in surface global solar radiation (GSR) and evaporative water loss from the ground are important issues to many researches and applications. In this study empirical relationships suitable for Taiwan were established for GSR retrieval from geostationary satellite images using the Heliosat method for the period from 2011 - 2013. The derived GSR data has been used to generate consecutive maps of 10-day averaged pan evaporation (Epan) as the basis to produce regional ET estimation using a strategy that does not require remote sensed land surface temperatures (LST). The retrieved daily GSR and the derived 10-day averaged Epan were validated against pyranometer and class-A pan measurements at selected Central Weather Bureau (CWB) stations spread across various climatic regions in Taiwan. Compared with the CWB observed data the overall relative mean bias deviations (MBD%) and root mean square differences (RMSD%) in daily solar irradiance retrieval were about 5 and 15%, respectively. Seasonally, the largest MBD% and RMSD% of retrieved daily solar irradiance occur in spring (9.5 and 21.3% on average), while the least MBD% (-0.3% on average) and RMSD% (9.7% on average) occur in autumn and winter, respectively. For 10-day averaged Epan estimation, the mean MBD% and RMSD% for stations located in the coastal plain areas were 0.1 and 16.9%, respectively. However, in mountainous areas the mean MBD% and RMSD% increased to 30.2 and 34.5%, respectively. This overestimation was due mainly to the large differences in surrounding micro-environments between the mountainous and plain areas.

Read 331 times