The Surface Energy Balance Algorithm for Land (SEBAL) framework in Graphics Processing Units (GPU) using Cuda and OpenCL

Evapotranspiration (ET) is the second largest component of terrestrial hydrological cycle on a global scale. The estimation ET at regional scale using satellite imagery and algorithms allow conversion instantaneous measures in daily totals which represents major contribution to study energy exchange between biosphere atmosphere. One most important remote sensing used estimate spatial partitioning SEBAL - Surface Energy Balance Algorithm for Land. time processing this algorithm can be very high depending size image. Thus, objective was propose framework use graphic units (GPU) run increase performance reduce required it. For implementation SEBAL’s local data collected micrometeorological tower installed an area located Northern Pantanal (Brazil) Landsat 5 – TM We Model Maker function ERDAS 9.2 Software algorithm, its Java, CUDA OpenCL languages. A estimated by JAVA, languages were similar value performed software. computed software 10% lower JAVA language, 9% than obtained Bowen ratio method. results showed that better due Speed Up up 35,000 times. Keywords: Wetland, evapotranspiration estimation, hyper seasonal savanna .