Heat waves over India occur during the months of March-June. This study aims at the real-time monitoring and prediction of heat waves using a multi-model dynamical ensemble prediction system developed at Indian Institute of Tropical Meteorology, India. For this, a criterion has been proposed based on the observed daily gridded maximum temperature (Tmax) datasets, which can be used for real-time prediction as well. A heat wave day is identified when either (1) Tmax (a)≥ its climatological 95^th percentile (calculated from daily values during March-June and for 1981–2010), (b) >36 °C, and (c) its departure from normal is >3.5 °C, Or, (2) when the Tmax >44 °C. Three heat wave prone regions, namely, northwest, southeast and northwest-southeast regions are recognized and heat wave spells of minimum consecutive six days are identified objectively for each region during 1981–2018. It is noticed that the prediction system has reasonable skill in predicting the heat waves over heat wave prone regions of India. Forecast verification of heat wave spells during 2003–2018 reveals that the prediction system has great potential in providing overall indication about the onset, duration and demise of the forthcoming heat wave spell with sufficient lead time albeit with some spatio-temporal error.

To better represent organized convection in the Climate Forecast System version 2 (CFSv2), a stochastic multicloud model (SMCM) parameterization is adopted and a 15 year climate run is made. The last 10 years of simulations are analyzed here. While retaining an equally good mean state (if not better) as the parent model, the CFS‐SMCM simulation shows significant improvement in the synoptic and intraseasonal variability. The CFS‐SMCM provides a better account of convectively coupled equatorial waves and the Madden‐Julian oscillation. The CFS‐SMCM exhibits improvements in northward and eastward propagation of intraseasonal oscillation of convection including the MJO propagation beyond the maritime continent barrier, which is the Achilles Heel for coarse‐resolution global climate models (GCMs). The distribution of precipitation events is better simulated in CFSsmcm and spreads naturally toward high‐precipitation events. Deterministic GCMs tend to simulate a narrow distribution with too much drizzling precipitation and too little high‐precipitation events.

This study investigates the seasonal climatology of the three-dimensional distribution and transport pathways of dust aerosols over West Asia (WA). Dust column loading over WA exhibits strong seasonality, with markedly high (weak) loading during summer (winter). The summer dust features over WA include the (i) dust reaching up to the 500 hPa level between the Red Sea (RS) and the west coast of the Indian subcontinent (IS); (ii) a slantwise advection of dust aerosols between 850 and 700 hPa levels over the Arabian Peninsula (AP) and Arabian Sea (AS); and (iii) a prominent mid-tropospheric zonal transport of AP dust toward the IS. Maximum column integrated horizontal dust mass flux (DMF) over WA is observed in summer. The intraday changes in the intensity and spatial spread of the DMF over the AP are mediated by the out-of-phase evolution of the surface winds and low-level Shamal jets. Furthermore, the diurnal changes in the strength of the inversion layers located above the monsoon boundary layer and associated wind shear regulate the spatial patterns and intensity of the DMF over the AS. The findings will support future studies aiming at quantifying the radiative effects of dust on the regional climate.

The National Centre for Environmental Prediction (NCEP) Climate Forecast System (CFS) is being used for operational monsoon prediction over the Indian region. Recent studies indicate that the moist convective process in CFS is one of the major sources of uncertainty in monsoon predictions. In this study, the existing simple cloud microphysics of CFS is replaced by the six‐class Weather Research Forecasting (WRF) single moment (WSM6) microphysical scheme. Additionally, a revised convective parameterization is employed to improve the performance of the model in simulating the boreal summer mean climate and intraseasonal variability over the Indian summer monsoon (ISM) region. The revised version of the model (CFSCR) exhibits a potential to improve shortcomings in the seasonal mean precipitation distribution relative to the standard CFS (CTRL), especially over the ISM region. Consistently, notable improvements are also evident in other observed ISM characteristics. These improvements are found to be associated with a better simulation of spatial and vertical distributions of cloud hydrometeors in CFSCR. A reasonable representation of the subgrid‐scale convective parameterization along with cloud hydrometeors helps to improve the convective and large‐scale precipitation distribution in the model. As a consequence, the simulated low‐frequency boreal summer intraseasonal oscillation (BSISO) exhibits realistic propagation and the observed northwest‐southeast rainband is well reproduced in CFSCR. Additionally, both the high and low‐frequency BSISOs are better captured in CFSCR. The improvement of low and high‐frequency BSISOs in CFSCR is shown to be related to a realistic phase relationship of clouds.