Pravat Rabi Naskar’s research while affiliated with India Meteorological Department and other places

The importance of air-sea heat fluxes (ASFs) on the global climate systems, ocean–atmosphere interactions and the associated processes has prompted us to undertake this study. In this study, the spatiotemporal variations of the Surface Latent Heat Flux (SLHF) and the Surface Sensible Heat Flux (SSHF) over the North Indian Ocean (NIO) from 1980 to 2020 have been investigated. It is observed that the Arabian Sea (AS) and the Bay of Bengal (BoB) have the highest SLHF values in winter which can be related to the high air-sea specific humidity difference during this season. It is also noticed that in the winter season, the maximum SLHF is located near the Somalia and Gujarat-Maharashtra coast rather than in the middle of the Arabian Sea, possibly due to the high wind speed and high air-sea specific humidity difference, respectively. The trend in the SLHF is inextricably linked to the trend in the Sea Surface Temperature (SST). Over the BoB and the AS both the SLHF and the SST are increasing. Similar to the SST, the SLHF increases in the northernmost parts of the AS and decreases around the North Bay. The strong monsoon winds may be responsible for the highest SSHF in the AS during the monsoon. Variations of the SSHF with bulk variables in the AS and the BoB indicate that the SSHF is decreasing due to a reduction in air-sea temperature difference. In contrast, the SLHF is increasing during the study period.

In this study, we have tried to examine the sea surface temperature (SST) changes over the North Indian Ocean (NIO) under Shared Socioeconomic Pathway 2 (SSP2) and Shared Socioeconomic Pathway 5 (SSP5) in the near future (2020–2059) and in the far future (2060–2099). For this, we have selected 12 important CMIP6 GCMs. From the analysis, it is clearly seen that the entire NIO is getting hotter in the future under all selected emission scenarios. This increase in the SST over the NIO is more under SSP5 than under SSP2 and the difference is almost 3°C. It is also observed that under the considered emission scenarios, the NIO is projected to be warmer in the far future than in the near future. It is also found that under SSP5 in the far future, the AS will be 0.5°C warmer than the BoB. Analysing the external factors influencing the SST, it is observed that except for the wind speed (WS), the projected changes in other factors, such as the net surface heat flux (NSHF) and the cloud cover (CC), are positive and not favourable for the SST rise over the NIO.

In this study, we have tried to find out how closely air-sea heat fluxes (Surface Latent Heat Flux (SLHF) and Surface Sensible Heat Flux (SSHF)) are associated with the intensity, energy and destructiveness of the Bay of Bengal (BoB) Intense Tropical Cyclones (ITCs). For this, we have considered the 32 ITCs which originated over the BoB during 1990–2020. We have taken the vital parameters data from IMD best track data and the high resolution (0.25°×0.25°) flux data from the European Centre for Medium-Range Weather Forecast’s (ECMWF) ERA5. We have calculated the ACE and PDI from the vital parameters of each ITC and correlated them with accumulated SLHF and accumulated SSHF. We have found that ACE and PDI depict a strong correlation with accumulated SLHF (R²slhf_ACE=0.82, R²slhf_PDI=0.49) and accumulated SSHF (R²sshf_ACE=0.79, R²sshf_PDI=0.56). It is also observed that ITCs’ intensity is also strongly correlated with max SLHF (R²slhf_CP=-0.56, R²slhf_WIND=0.57) whereas it is weakly correlated with max SSHF (R²sshf_CP=-0.24, R²sshf_WIND=0.23).

This study has been undertaken to predict the Surface Latent Heat Flux (SLHF) over the North Indian Ocean (NIO) in the future period (2020–2099) under different emission scenarios. For this 12 Coupled Model Inter-comparison Project 6 (CMIP6), Global Climate Models (GCMs) SLHF and ERA5 SLHF data have been used. It is observed that the SLHF is going to increase in the far future (2060–2099) under all emission scenarios probably due to higher warming. In the near future (2020–2059) particularly in the first half (2020–2039) a clear fall in the SLHF is noticed. It is also observed that the SLHF rise under the Shared Socioeconomic Pathways 5 (SSP5) is more than that under the SSP2 in the far future due to higher warming. Seasonal variation of the SLHF depicts that under SSP5 in the far future, the rise in the SLHF is the maximum in MAM. The rise of the SLHF in the far future can be attributed to the rise in the SST over the NIO but the reason for the fall of the SLHF in the first half of the near future is not clear.

This study has been undertaken to Bnd out the spatiotemporal variations of rainfall and droughts over India in two future periods (2020-2059 and 2060-2099) under two SSP (SSP2-4.5 and SSP5-8.5) scenarios. Here we have selected the four most suitable Coupled Model Inter-comparison Project Phase 6 (CMIP6) Global Climate Models (GCMs) to project rainfall and drought over India. We have found that rainfall over India is on the rise under considered periods and emission scenarios with few spatial departures. The rainfall is higher under higher warming. Central India, particularly Maharashtra and Madhya Pradesh, shows an increase in rainfall in all emission scenarios. This region also shows a higher increase in rainfall variability. The rainfall variability is found to be the maximum in the northeast Himalayan region. In the future, drought frequency does not show any appreciable change but drought duration changes and shows spatial variations. An increase in drought duration is observed over northwestern parts of India, particularly over Gujarat, Rajasthan, parts of MP, UP and Haryana.

In this study, we have tried to find out the influence of air-sea heat fluxes (particularly the Surface Latent Heat Flux (SLHF) and the Surface Sensible Heat Flux (SSHF)) on the intensity of Intense Tropical Cyclones’ (ITCs’). We have analysed 32 ITCs which originated in the Bay of Bengal (BoB) during 1990–2019. We have used IMD best track data for track and vital parameters of ITCs. We have also used high resolution (0.25°×0.25°) ERA5 SLHF, SSHF and SST data for their variations during the lifetime of ITCs. It is observed that the SLHFmax during the whole lifetime and the study period is highly correlated with ITCs’ intensity (i.e. with the central pressure (CP) and the maximum sustained wind speed (MSW)) whereas the SSHFmax shows weak correlations with ITCs’ intensity. This suggests the strong association between the SLHFmax and ITCs intensity and strong latent heat flux exchange between the ocean and atmosphere during the whole lifetime and the study period. Similar results are observed in the pre-monsoon and the post-monsoon season. In the pre-monsoon season the association of SLHFmax with the ITCs intensity is stronger than the post-monsoon season due to strong background conditions, pointing towards the strong air-sea interaction. The SLHFmax in the growing and the decaying stage are also highly correlated with ITCs’ intensity but correlation coefficients of ITCs’ intensity with the SLHFmax in the decaying stage are slightly higher than those of in the growing stage. It is also found that the SSHFmax has an appreciable correlation with ITCs’ intensity during the growing stage whereas it has negligible correlation with ITCs’ intensity during the decaying stage pointing towards the influence of sensible heat flux in the growing stage of ITCs. It is also noticed that during rapid decay (RD) the SLHFmax is highly correlated with ITCs’ intensity possibly due to cold wakes which rapidly diminishes the SLHF but during rapid intensification the SLHF does not increase so rapidly due to the sluggish WISHE feedback, hence the SLHFmax during rapid intensification (RI) is not appreciably correlated with ITCs’ intensity.

We have attempted to investigate the spatiotemporal patterns and Cuctuations in summer thermal heat stress/discomfort over the Indian region-based universal thermal climate index (UTCI) in this study. We have calculated UTCI using hourly ERA5 data of 10 m wind speed, 2 m air temperature, 2 m dew point temperature, and solar radiation for the period 1990-2020. To determine the eAect of radiation Cuxes and soil moisture on temperature and UTCI, we have used ERA5 data on cloud fraction (CF), surface heat Cuxes (SLHF and SSHF), and soil moisture (SM) for the study period. Maximum heating and discomfort have been reported in May for most of the regions. Except for the west region, the progress of the monsoon provides some relief in June. Maximum discomfort is observed around 08-09Z. We have observed over 50% of India experiencing a decreasing trend in UTCI in different summer months despite over 50% of India experiencing an increasing trend in temperature. This is due to the inCuence of factors such as solar radiation, cloudiness, wind speed, soil moisture, etc., on UTCI. The UTCI in summer months demonstrates spatial heterogeneity. UTCI increases significantly in the west region in April and the east region in June. In April and May, some portions of the South-Central region, particularly Maharashtra, exhibit an increasing trend in UTCI. The majority of the North-Central region has a noticeable decreasing tendency in UTCI in all the summer months. We have not found any significant trend in the frequency of days with 'very high heat stress' or higher discomfort. Except in the eastern region, there is no noticeable trend in the frequency of discomfort hours with UTCI in the ranges 38 \ UTCI \ 46°C and UTCI [ 46°C. The Eastern region exhibits an increasing trend in the frequency of discomfort hours with UTCI in the range of 38 \ UTCI \ 46°C in April. The Eastern region has a rising trend in the frequency of discomfort hours, with UTCI in the range of 38 \ UTCI \ 46°C in April.

This study has been undertaken to find out the variation of central pressure (intensity) of intense Tropical Cyclones (TCs) with Sea Surface Temperature (SST), Mid-tropospheric Relative Humidity (MRH), Mid-tropospheric Instability (MI), Vertical Wind Shear (VWS), 200-hPa divergence, and Surface Latent Heat Flux (SLHF) during the lifetime intense TCs. This study also aims to determine the most crucial parameter which shows the highest correlation with central pressure (intensity) of intense TCs during their lifetime. Out of all these parameters, SLHF is highly correlated (R = 0.74) with the central pressure (intensity) of intense TCs. Increase and decrease of SLHF correspond to decrease and increase of TCs central pressure (increase and decrease in TCs intensity). The highest SLHF corresponds to the lowest central pressure (highest intensity).

This study has been undertaken to find out the spatiotemporal variations of rainfall and droughts over India in two future periods (2020–2059 and 2060–2099) under two SSP (SSP2-4.5 and SSP5-8.5) scenarios. Here we have selected the four most suitable Coupled Model Inter-comparison Project Phase 6 (CMIP6) Global Climate Models (GCMs) to project rainfall and drought over India. We have found that rainfall over India is on the rise under considered periods and emission scenarios with few spatial departures. The rainfall is higher under higher warming. Central India, particularly Maharashtra and Madhya Pradesh, shows an increase in rainfall in all emission scenarios. This region also shows a higher increase in rainfall variability. The rainfall variability is found to be the maximum in the northeast Himalayan region. In the future, drought frequency does not show any appreciable change but drought duration changes and shows spatial variations. An increase in drought duration is observed over northwestern parts of India, particularly over Gujarat, Rajasthan, parts of MP, UP and Haryana.

To predict the most suitable mode decomposition technique for machine learning in meteorological time series prediction, this study has been carried out. The best mode decomposition technique along with a suitable machine learning algorithm will help predict a time series accurately. For this empirical mode decomposition (EMD), ensemble empirical mode decomposition (EEMD), coupled ensemble EMD with adaptive noise (CEEMDAN), variational mode decomposition (VMD), singular spectrum analysis (SSA) and independent component analysis (ICA) have been taken for comparison. The long short-term memory (LSTM) network has been used for machine learning. The max–min temperature time series have been constructed from Kolkata's max–min temperature data. Those time series have been predicted individually by the LSTM network along with each of these techniques. It has been found that the predicted time series by LSTM decomposed by ICA mostly matches the original time series compared to other techniques.