204-8Comparison of NET Radiation Between Ground Measurement and SEBAL Estimate in Arid Riparian Area.
See more from this Division: ASA Section: Climatology & ModelingSee more from this Session: Symposium--Evapotranspiration: Monitoring, Modeling and Mapping At Point, Field, and Regional Scales: I
Tuesday, October 23, 2012: 10:35 AM
Duke Energy Convention Center, Room 234, Level 2
Ground measured net radiation (Rn) is normally obtained with a net radiometer at a height of about 2 – 3 m above the canopy that covers typically a ground area on the order of 10 m2. The SEBAL estimated Rn with Landsat ETM+ imagery at the time of satellite overpass is calculated from reflectances in the visible, near-infrared, and mid-infrared bands from a 900 m2 pixel and the emittance in the thermal band from a 3600 m2 pixel. Although the Rn ground observation is based on a measurement area at least two orders of magnitude smaller than the SEBAL Rn, this difference will not matter much for homogeneous areas; however, for heterogeneous areas it may cause serious bias. In heterogeneous arid riparian areas, since radiometers are typically placed over the canopy of interest, it may cause under-representation of surrounding bare soil or ground cover in the angle of view. Therefore, ground measured Rn is expected to be biased towards the Rn of the vegetation of interest in heterogeneous arid riparian areas. The objective of this study is to compare the SEBAL Rn derived from Landsat to net radiometer measurements in NM riparian areas. Results of this study show that the difference between ground measured instantaneous Rn and the one determined by SEBAL is quite large. The instantaneous Rn mean relative difference (MRD) between ground measurement and SEBAL estimate is 14.6%. In heterogeneous pixels, the SEBAL Rn seems more reliable than the ground measured one due to its more representative footprint.
See more from this Division: ASA Section: Climatology & ModelingSee more from this Session: Symposium--Evapotranspiration: Monitoring, Modeling and Mapping At Point, Field, and Regional Scales: I