Skin Temperature Analysis and Bias Correction in a Coupled Land-Atmosphere Data Assimilation System

Abstract

In an initial investigation, remotely sensed surface temperature is assimilated into a coupled atmosphere/land global data assimilation system, with explicit accounting for biases in the model state. In this scheme, a incremental bias correction term is introduced in the model's surface energy budget. In its simplest form, the algorithm estimates and corrects a constant time mean bias for each gridpoint; additional benefits are attained with a refined version of the algorithm which allows for a correction of the mean diurnal cycle. The method is validated against the assimilated observations, as well as independent near-surface air temperature observations. In many regions, not accounting for the diurnal cycle of bias caused degradation of the diurnal amplitude of background model air temperature. Energy fluxes collected through the Coordinated Enhanced Observing Period (CEOP) are used to more closely inspect the surface energy budget. In general, sensible heat flux is improved with the surface temperature assimilation, and two stations show a reduction of bias by as much as 30 W m À2 . At the Rondonia station in Amazonia, the Bowen ratio changes direction in an improvement related to the temperature assimilation. However, at many stations the monthly latent heat flux bias is slightly increased. These results show the impact of univariate assimilation of surface temperature observations on the surface energy budget, and suggest the need for multivariate land data assimilation. The results also show the need for independent validation data, especially flux stations in varied climate regimes.

A Modeling and Observational Framework for Diagnosing Local Land–Atmosphere Coupling on Diurnal Time Scales

“…There have been numerous efforts to assimilate screen-level observations in offline (Rhodin et al 1999;Hess 2001), single-column (Hacker and Snyder 2005;Hacker and Rostkier-Edelstein 2007), and fully coupled (Seuffert et al 2004) models. However, a great deal of testing of land data assimilation of soil moisture, surface temperature, and snow has been performed for offline models (Rodell and Houser 2004;Reichle et al 2007;Bosilovich et al 2007), which lack L-A interactions and feedbacks that otherwise would impact the assimilation results in coupled mode. As efforts to assimilate new remote sensing data increase along with the complexity of assimilation techniques, the manner in which the land and atmosphere are coupled as well as the strength of feedbacks becomes critical to the process.…”

Assimilation of GRACE Terrestrial Water Storage Data into a Land Surface Model: Results for the Mississippi River Basin

“…First, the GRACE observations themselves, though coarse, yield reasonably reliable estimates of TWS anomalies . Assimilating these data into an LSM, therefore, has the potential to improve the accuracy of TWS in LSM simulations, much as assimilation of remotely sensed snow cover Rodell and Houser, 2004), snow water equivalent , soil moisture (Margulis et al, 2002;Crow and Wood, 2003;Reichle and Koster, 2005), and skin temperature (Bosilovich et al, 2007) have had a positive impact on LSM simulations. Second, our understanding of hydrological processes, as captured by the model, are used to enhance the satellite observations, providing downscaling and quality control of GRACE observations while enabling synthesis of data from multiple observing systems in a physically consistent manner.…”

Simultaneous estimation of model state variables and observation and forecast biases using a two-stage hybrid Kalman filter

“…In order to demonstrate the benefit of the estimation of both forecast and observation biases, as opposed to the estimation of forecast biases alone (Dee and Da Silva, 1998;Dee and Todling, 2000;Drécourt et al, 2006;De Lannoy et al, 2007;Bosilovich et al, 2007;Reichle et al, 2010), the experiments described above were repeated, but the estimation of the observation bias was turned off. Under these conditions, the state and forecast bias estimation reduces to the methodology described in De Lannoy et al (2007).…”

“…In order to bypass this inconsistency, a number of studies have focused on the removal of systematic differences between the assimilated data and the model through rescaling the data to the model climatology (Reichle and Koster, 2004;Slater and Clark, 2006;De Lannoy et al, 2012). Other studies have focused on the estimation of the forecast bias in addition to the model state variables, using the discrete (Kalman, 1960) and the ensemble Kalman filter for both linear and nonlinear systems, in a wide range of applications ranging from groundwater modeling to soil moisture and temperature assimilation (Dee and Da Silva, 1998;Dee and Todling, 2000;De Lannoy et al, 2007;Drécourt et al, 2006;Bosilovich et al, 2007;Reichle et al, 2010). Dee (2005) further explains how forecast bias can be taken into account in a data assimilation system using the Kalman filter or variational assimilation as assimilation algorithm.…”