Evaluation of land surface temperature parameterization approaches using surface- layer observations
Surface temperature (Ts) is vital to the study of land-atmosphere interactions and climate variabilities. However, observed Ts data are still very scarce in humid tropical region. There is therefore a need to parameterize and improve the representation of Ts in Global Climate Models using available meteorological data. Six land surface temperature parameterization approaches (Force restored, Liebethal, Holtslag, Equilibrium Gradient, Tracy and Gottsche approaches) were validated with actual measurements using the Nigeria Micrometeorological Experiment (NIMEX) surface layer observations. The Liebethal approach showed the best agreement with the measured data with average coefficient of determination, mean bias error and root mean square error of 0.96 ± 0.01, 0.08 ± 0.04oC and 0.85 ± 0.14oC, respectively, in simulated Ts. The results also showed that the Force restored and Tracy approaches are applicable for land surface temperature parameterization in this region.
Keywords: Surface Temperature, Climate Change, Humid, Parameterization, Mean Bias Error