Future changes in the intensity of tropical cyclones (TCs) under global warming are uncertain, although several studies have projected an upward trend in TC intensity. In this study, we examined the changes in the strength of TCs in the twenty-first century based on the Hurricane Maximum Potential Intensity (HuMPI) model forced with the sea surface temperature (SST) from the bias-corrected CMIP6 dataset. We first investigated the relationship between the mean lifetime maximum intensity (LMI) of major hurricanes (MHs) and the maximum potential intensity (MPI) using the SST from the Daily Optimum Interpolation SST database. The LMI of MHs and the MPI in the last two decades was, on average, 2–3% higher than mean values in the sub-period 1982–2000, suggesting a relationship between changes in MPI and LMI. From our findings, the projected changes in TC intensity in the near-future period (2016–2040) will be almost similar for SSP2-4.5 and SSP5-8.5 climate scenarios. However, TCs will be 9.5% and 17% more intense by the end (2071–2100) of the twenty-first century under both climate scenarios, respectively, compared with the mean intensity over the historical period (1985–2014). In addition, the MPI response to a warmed sea surface temperature per degree of warming is a 5–7% increase in maximum potential wind speed. These results should be interpreted as a projection of changes in TC intensity under global warming since the HuMPI formulation does not include environmental factors (i.e., vertical wind shear, mid-level moisture content and environmental stratification) that influence TC long-term intensity variations. Highlights: The maximum potential intensity (MPI) of tropical cyclones is a predictor of their climatological intensities. Tropical cyclones will be 17% more intense than today by the end of the 21st Century. The maximum potential wind speed will increase by 5–7%/ºC under global warming.
The short-term prediction of precipitation is a difficult spatio-temporal task due to the non-uniform characterization of meteorological structures over time. Currently, neural networks such as convolutional LSTM have shown ability for the spatio-temporal prediction of complex problems. In this research, we propose an LSTM convolutional neural network (CNN-LSTM) architecture for immediate prediction of various short-term precipitation events using satellite data. The CNN-LSTM is trained with NASA Global Precipitation Measurement (GPM) precipitation data sets, each at 30-min intervals. The trained neural network model is used to predict the sixteenth precipitation data of the corresponding fifteen precipitation sequence and up to a time interval of 180 min. The results show that the increase in the number of layers, as well as in the amount of data in the training data set, improves the quality of the forecast.