Balance characteristics of multivariate background error covariances and their impact on analyses and forecasts in tropical and Arctic regions

Abstract

For variational data assimilation, the background error covariance matrix plays a crucial role because it is strongly linked with the local meteorological features, and is especially dominated by error correlations between different analysis variables. Multivariate background error (MBE) statistics have been generated for two regions, namely the Tropics (covering Indonesia and its neighborhood) and the Arctic (covering high latitudes). Detailed investigation has been carried out for these MBE statistics to understand the physical processes leading to the balance (defined by the forecasts error correlations) characteristics between mass and wind fields for the low and high latitudes represented by these two regions. It is found that in tropical regions, the unbalanced (full balanced) part of the velocity potential (divergent part of wind) contributes more to the balanced part of the temperature, relative humidity, and surface pressure fields as compared with the stream function (rotational part of wind). However, the exact opposite happens in the Arctic. For both regions, the unbalanced part of the temperature field is the main contributor to the balanced part of the relative humidity field. Results of single observation tests and six-hourly data assimilation cycling experiments are consistent with the respective balance part contributions of different fields in the two regions. This study provides an understanding of the contrasting dynamical balance relationship that exists between the mass and wind fields in high-and low-latitude regions. The study also examines the impact of MBE on Weather Research and Forecasting model forecasts for the two regions.

Background error statistics for aerosol variables from WRF/Chem predictions in Southern California

“…The balance constraints between different analysis variables may also be obtained by using some statistical regression methods. By subtracting the balanced parts, the residual unbalanced parts are considered to be independent analysis variables (Chen et al, 2013b). In future studies, we will explore to develop statistical regression relations and/or formulate dynamic balance constraints to account for cross correlations.…”

Impact of variational assimilation using multivariate background error covariances on the simulation of monsoon depressions over India

“…Also, the location of the maximum impact of the observation is shifted in the cv6 run as compared with that in the cv5 run. Since the covariance matrix is symmetric, assimilation of a single temperature observation will result in a relative rotation of increment pattern in the u field (Chen et al, 2013). Hence, the single-observation experiments reveal that the difference in formulation of control variables has resulted in different analysis increment structures in the cv6 option as compared to that in the cv5 option.…”

“…Hence, temperature and surface pressure are influenced by the divergent component of wind in the cv6 option unlike in the cv5 option. Similarly, additional correlations defined in the moisture variable make the moisture analysis multivariate in nature in the cv6 option (Chen et al, 2013). In both the cv5 and cv6 option, the estimation of BECs proceed in the following five stages:…”

Application of Feature Calibration and Alignment to High-Resolution Analysis: Examples Using Observations Sensitive to Cloud and Water Vapor

“…7. Length scales are calculated following the procedure for the Weather Research and Forecasting Model's community variational/ ensemble data assimilation system (WRFDA) error covariance computation described in Chen et al (2013). One should note that the length scales for rain (,1 km) are much smaller than the grid mesh (3 km) and should not be interpreted physically.…”