Multimodel Analysis of Energy and Water Fluxes: Intercomparisons between Operational Analyses, a Land Surface Model, and Remote Sensing

Abstract

Using data from seven global model operational analyses (OA), one land surface model, and various remote sensing retrievals, the energy and water fluxes over global land areas are intercompared for 2003/04. Remote sensing estimates of evapotranspiration (ET) are obtained from three process-based models that use input forcings from multisensor satellites. An ensemble mean (linear average) of the seven operational (mean-OA) models is used primarily to intercompare the fluxes with comparisons performed at both global and basin scales. At the global scale, it is found that all components of the energy budget represented by the ensemble mean of the OA models have a significant bias. Net radiation estimates had a positive bias (global mean) of 234 MJ m−2 yr−1 (7.4 W m−2) as compared to the remote sensing estimates, with the latent and sensible heat fluxes biased by 470 MJ m−2 yr−1 (13.3 W m−2) and −367 MJ m−2 yr−1 (11.7 W m−2), respectively. The bias in the latent heat flux is affected by the bias in the net radiation, which is primarily due to the biases in the incoming shortwave and outgoing longwave radiation and to the nudging process of the operational models. The OA models also suffer from improper partitioning of the surface heat fluxes. Comparison of precipitation (P) analyses from the various OA models, gauge analysis, and remote sensing retrievals showed better agreement than the energy fluxes. Basin-scale comparisons were consistent with the global-scale results, with the results for the Amazon in particular showing disparities between OA and remote sensing estimates of energy fluxes. The biases in the fluxes are attributable to a combination of errors in the forcing from the OA atmospheric models and the flux calculation methods in their land surface schemes. The atmospheric forcing errors are mainly attributable to high shortwave radiation likely due to the underestimation of clouds, but also precipitation errors, especially in water-limited regions.

The GEWEX LandFlux project: evaluation of model evaporation using tower-based and globally gridded forcing data

“…These data have provided a wealth of spatially and temporally varying information across a range of Earth system processes, including soil moisture (Liu et al, 2011a), vegetation change (Tucker et al, 2005;Liu et al, 2011bLiu et al, , 2013, groundwater (Famiglietti et al, 2011;Richey et al, 2015) and precipitation (Huffman et al, 1995;Nesbitt et al, 2004), enabling a capacity to enhance our understanding and description of regional-and global-scale water cycles and their spatial and temporal variability. While evaporation represents the key process returning the Earth's surface water to the overlying atmosphere and provides the linking mechanism between the water and energy cycles, it is only in relatively recent times that effort has been directed towards the development of global products (Mu et al, 2007;Fisher et al, 2008;Vinukollu et al, 2011a).…”

“…Uncertainties emerging from such aggregations can often reduce the confidence in any subsequent model performance ranking. One initial effort addressing this was the study of Vinukollu et al (2011a), which used the Surface Energy Balance System (SEBS) model (SEBS; Su, 2002), a twosource Penman-Monteith scheme by Mu et al (2007) and a three-source model based on parameterising the PriestleyTaylor model (PT-JPL) (Fisher et al, 2008) to estimate global evaporation for the period 2003-2004. The Vinukollu et al (2011a analysis revealed that the modelled instantaneous evaporation (coinciding with the time of satellite overpass) was in reasonable agreement with locally observed evaporation at 12 eddy-covariance towers across the United States, with correlations ranging from 0.43 to 0.54.…”

Examining the relationship between intermediate-scale soil moisture and terrestrial evaporation within a semi-arid grassland

“…The model has been used to estimate actual evapotranspiration at local (Ershadi et al, 2014) and global scales in various studies (e.g. Badgley et al, 2015;Ershadi et al, 2014;Vinukollu et al, 2011), including the recent LandFlux and WACMOS-ET efforts. Detailed descriptions of the model are provided in those references.…”

Three perceptions of the evapotranspiration landscape: comparing spatial patterns from a distributed hydrological model, remotely sensed surface temperatures, and sub-basin water balances

“…Recent studies that also make use of thermal and optical sensors range from "classical" rainfall-runoff modelling with remotely sensed pattern comparison (like our contribution) to integrated data assimilation systems. Examples of the former are from Boegh et al (2004) for 10 km 2 of agricultural landscape in Denmark and Vinukollu et al (2012) with a global ET pattern comparison; as well as a substantial contribution from Schuurmans et al (2011) who first compare and then assimilate the modelled and remotely sensed actual ET patterns of an area of 70 km 2 in the middle of the Netherlands; however, observed differences between the two data sources remain partly unexplained.…”