The impact of assimilation of satellite derived wind observations for the prediction of a monsoon depression over India using a mesoscale model

Abstract

Copyright 2008 Elsevier B.V., All rights reserved.The Penn State/NCAR mesoscale model (MM5) has been used in this study to ingest and assimilate the INSAT-CMV (Indian National Satellite System-Cloud Motion Vector) wind observations using analysis nudging (four-dimensional data assimilation, FDDA) to improve the prediction of a monsoon depression which occurred over the Bay of Bengal, India during 28 July 2005 to 31 July 2005. To determine the impact of assimilation of INSAT-CMV winds on the prediction of a monsoon depression, three sets of numerical experiments (NOFDDA, FDDA and FDDA CMV) were designed. While the FDDA CMV run assimilated satellite derived winds only, the FDDA run assimilated both satellite and conventional observations. The NOFDDA run used neither satellite nor conventional observations. The results of the study indicate that the simulated sea level pressure field from the FDDA run is more consistent with the sea level pressure field from NCEP-FNL compared to the FDDA CMV and NOFDDA runs. The highest correlation and lowest rms error of the sea level pressure field are associated with the FDDA run, and this provides a quantitative verification of the improvement due to the assimilation of satellite derived winds and the conventional upper air observations for the prediction of monsoon depression. All the three model simulated winds are in good agreement with the analysis winds at 850 hPa, 500 hPa and 200 hPa levels. The simulated structure of the spatial precipitation pattern for the assimilation experiments (FDDA and FDDA CMV) are closer to the TRMM observations with more rainfall simulated over the east coast regions in the assimilation experiments. The rms errors of the wind speed for the FDDA run show lower values at 500 hPa for all the three model runs, with a reduction in all three levels of up to 0.8-1.4 m s for the FDDA run and 0.5-1.9 m s for the FDDA CMV run with respect to the NOFDDA run. The statistical significance of the sea level pressure and the precipitation differences between the FDDA and the NOFDDA as well as the differences between the FDDA CMV and the NOFDDA have been calculated using the two-tailed Student's t-test and were found to be statistically significant. The influence of varying the nudging coefficients in the FDDA experiment has been studied

Impact of meso-net observations on short-term prediction of intense weather systems during PRWONAM: Part I—On wind variations

“…For locations all over India, attempts have been made systematically over the decades to improve the atmospheric modeling for short and medium range weather prediction of high impact weather systems, mesoscale phenomena, and the localized specific weather phenomena (Tyagi and Pattaniak, 2008;Roy Bhowmik et al, 2009, 2003Roy Bhowmik and Prasad, 2001;NCMRWF, 2006aNCMRWF, , 2006bNCMRWF, , 2008Potty et al, 2000;Mandal et al, 2004;Xavier et al, 2008;Vinod and Chandrasekar, 2006). Several field programmes, namely, MONTBLEX-90, LASPEX-97, BOBMEX, and ARMEX have been carried out in the country to improve the physics parameterization schemes and to understand the air-sea interactions (Rao Kusuma et al, 1995;Rao Kusuma, 1996;Rao Kusuma et al, 1996a, 1996bRao Kusuma, 2004a, 2004bRao Kusuma and Narasimha, 2006;Rajagopal, 2001;Rao Kusuma et al, 2001;Basu, 2001;Bhat, 2002Bhat, , 2006, which are crucial for improving weather prediction beyond 36 h. In view of the land-surface heterogeneity across the country, land-sea contrast, and due to our limited ability to make the measurements of required description, short term weather prediction has been a problem of concern even now.…”

Comparison of MMCFS and SINTEX‐F2 for seasonal prediction of Indian summer monsoon rainfall

“…Continuous developments in various aspects of the model, viz. (a) resolution (Oouchi et al ., 2006; Watanabe, 2008; Alexander et al ., 2010; T. Sakamoto et al ., 2012), (b) parameterization schemes (Rajendran et al ., 2002; Simmons et al ., 2004; Ratnam and Krishna, 2005; Park and Bretherton, 2009; Jung et al ., 2010; Fox‐Kemper et al ., 2011; Geller et al ., 2011; Hazra et al ., 2015; Ganai et al ., 2016; Sujith et al ., 2019), (c) data assimilation (Alves et al ., 2004; Xavier et al ., 2008; Balmaseda and Anderson, 2009; Koster et al ., 2010; Zhao et al ., 2014; Senan et al ., 2016; Koul et al ., 2018; Lin et al ., 2020), and (d) representation of monsoon teleconnections (Turner et al ., 2005; Wu and Kirtman, 2005; Bracco et al ., 2007) has resulted in significant improvement in prediction skill of ISMR from ~0.25 (Preethi et al ., 2010) to ~0.65 (Rao et al ., 2019a). Still, it has remained lower than the potential limit of predictability of around 0.7 (Krishna Kumar et al ., 2005).…”