Time filter

Source Type

Shukla B.P.,Atmosphere and Oceanic science Group | Pal P.K.,Atmosphere and Oceanic science Group
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing | Year: 2012

A new algorithm is developed, based on Source Apportionment (SA) technique for tracking and nowcasting of Mesoscale Convective Systems (MCS) using satellite image sequences. The algorithm does not use the conventional threshold method for identifying convective regions, instead it introduces neighbourhood search criteria to select contiguous pixels. This offers better opportunity for data mining, and inherently provides an automatic method for tracking. The nowcasting scheme is based on growth curve model fitting and extrapolation using time varying coefficients. The methodology proposed in the paper is demonstrated for the life cycle of a MCS. The algorithm is also tested on an ensemble set comprising of thermal infrared image sequences acquired from Kalpana-1 satellite. Error statistics and skill score analysis indicate the operational feasibility of the proposed algorithm. © 2011 IEEE.

Das S.K.,Atmosphere and Oceanic science Group | Deb S.K.,Atmosphere and Oceanic science Group | Kishtawal C.M.,Atmosphere and Oceanic science Group | Pal P.K.,Space Applications Center
Pure and Applied Geophysics | Year: 2015

The experimental seasonal forecast of Indian summer monsoon (ISM) rainfall during June through September using Community Atmosphere Model (CAM) version 3 has been carried out at the Space Applications Centre Ahmedabad since 2009. The forecasts, based on a number of ensemble members (ten minimum) of CAM, are generated in several phases and updated on regular basis. On completion of 5 years of experimental seasonal forecasts in operational mode, it is required that the overall validation or correctness of the forecast system is quantified and that the scope is assessed for further improvements of the forecast over time, if any. The ensemble model climatology generated by a set of 20 identical CAM simulations is considered as the model control simulation. The performance of the forecast has been evaluated by assuming the control simulation as the model reference. The forecast improvement factor shows positive improvements, with higher values for the recent forecasted years as compared to the control experiment over the Indian landmass. The Taylor diagram representation of the Pearson correlation coefficient (PCC), standard deviation and centered root mean square difference has been used to demonstrate the best PCC, in the order of 0.74–0.79, recorded for the seasonal forecast made during 2013. Further, the bias score of different phases of experiment revealed the fact that the ISM rainfall forecast is affected by overestimation in predicting the low rain-rate (less than 7 mm/day), but by underestimation in the medium and high rain-rate (higher than 11 mm/day). Overall, the analysis shows significant improvement of the ISM forecast over the last 5 years, viz. 2009–2013, due to several important modifications that have been implemented in the forecast system. The validation exercise has also pointed out a number of shortcomings in the forecast system; these will be addressed in the upcoming years of experiments to improve the quality of the ISM prediction. © 2015, Springer Basel.

Das S.K.,Atmosphere and Oceanic science Group | Deb S.K.,Atmosphere and Oceanic science Group | Kishtawal C.M.,Atmosphere and Oceanic science Group | Pal P.K.,Space Applications Center
Pure and Applied Geophysics | Year: 2015

The diurnal cycle of different surface parameters, viz. surface air temperature, surface pressure, and rain intensities, simulated by the Community Atmosphere Model (CAM) in the operational seasonal forecast of ISM-2012 using initial conditions (ICs) taken at synoptic hours of the day has been examined and compared with observations. Four members were simulated with ICs at 0000, 0600, 1200, and 1800 UTC on 1 August 2012. The impact of the initial conditions at the synoptic hours of the day was more visible over the landmass compared with the oceanic regions. The diurnal variation of the surface temperature in the model simulation showed the major features when compared with the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis except for the warm pool of northwest India and the Tibetan region. The surface pressure in the ECMWF reanalysis showed the semidiurnal cycle with two peaks at 0600 UTC and 1800 UTC; however, the range of the cycle was underestimated by the model simulation, showing only one peak at 0600 UTC. Significant variations in the diurnal cycle of rain intensities were seen among the different members. The model captured the diurnal cycle as the positive and negative peaks at 1200 and 0000 UTC with intensities at the peaks ~0.5 mm high and low, respectively, in the model simulation when compared with the observations. Presently, the seasonal forecast of ISM is generated through ensemble CAM experiments using different ICs taken from different dates but all at 0000 UTC. Consideration of ICs at different times of the day will add different ranges of diurnal variations in all the surface parameters within the family of ensemble members and also increase the number of members in the family. Indeed, these improve the ensemble processes in generating the seasonal forecast of ISM. © 2014, Springer Basel.

Kaur I.,Atmosphere and Oceanic science Group | Kishtawal C.M.,Atmosphere and Oceanic science Group | Deb S.K.,Atmosphere and Oceanic science Group | Kumar R.,Atmosphere and Oceanic science Group
IEEE Geoscience and Remote Sensing Letters | Year: 2012

The autocorrelation function of Meteosat-7-derived atmospheric motion vectors (AMVs) has been calculated over the Indian Ocean region for the Asian summer and winter monsoon seasons. The time where the autocorrelation function dropped to 0.5 was defined as the decorrelation time. It was observed that seasonal forcing caused the circulation patterns to be highly stable with decorrelation timescales on the order of 24 h or more at some regions. The analysis was done for the vector wind ($u + iv$), the zonal component $u$, and the meridional component $v$. The nature of the autocorrelation function clearly followed the changing wind circulations with season and pressure levels. © 2006 IEEE.

Discover hidden collaborations