Precipitation obtained from rain gauges is an essential input for hydrological modelling. It is often sparse in highly topographically varying terrain, exhibiting a certain amount of uncertainty in hydrological modelling. Hence, satellite rainfall estimates have been used as an alternative or as a supplement to station observations. In this study, an attempt was made to evaluate the Tropical Rainfall Measuring Mission (TRMM) and Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), employing a semi-distributed hydrological model, i.e., Soil and Water Assessment Tool (SWAT), for simulating streamflow and validating them against the flows generated by the India Meteorological Department (IMD) rainfall dataset in the Gurupura river catchment of India. Distinct testing scenarios for simulating streamflow were made to check the suitability of these satellite precipitation data. The TRMM was able to better estimate rainfall than CHIRPS after performing categorical and continuous statistical results with respect to IMD rainfall data. While comparing the performance of model simulations, the IMD rainfall-driven streamflow emerged as the best followed by the TRMM, CHIRPS-0.05, and CHIRPS-0.25. The coefficient of determination (R²), Nash–Sutcliffe efficiency (NSE), and percent bias (PBIAS) were in the range 0.63 to 0.86, 0.62 to 0.86, and −14.98 to 0.87, respectively. Further, an attempt was made to examine the spatial distribution of key hydrological signature, i.e., flow duration curve (FDC) in the 30–95 percentile range of non-exceedance probability. It was observed that TRMM underestimated the flow for agricultural water availability corresponding to 30 percent, even though it showed a good performance compared to the other satellite rainfall-driven model outputs.

This study aims to construct the multiple time-scale joint distributions of drought duration and severity using two-dimensional copulas and compare the drought characteristics in India’s two contrasting climate regions: the arid Rajasthan and humid, tropical Kerala. The drought occurrences were defined by the standardized precipitation index (SPI) with a threshold below $-0.8$ at time scales of 3, 6, 12, and 24 months for 1900–2016. Significant correlations were noted between the drought severity and drought duration in both regions. The Clayton copula gave a better fit than other copulas for modeling the dependence among the observed drought duration and severity. The results indicate that the probability of short-term droughts (SPI-3 and SPI-6) is more significant than those of long-term droughts (SPI-12 and SPI-24) for an identical drought event in both regions. Also, the probability of severe drought events with greater duration and severity for long-term droughts (SPI-12 and SPI-24) is higher in Kerala than that in western Rajasthan. For all the time-scale SPIs, the conditional probability of drought severity for a given duration exceeding a threshold showed an increasing trend in both regions. Furthermore, the conditional probability of the drought duration given the severity for short-term droughts is greater than that of the long-term droughts for the same drought event. For short-term droughts, the conditional return period of an identical drought event is lower in Kerala than in western Rajasthan. In contrast, the conditional return period of long-term droughts is lower in western Rajasthan. Additionally, copula-based nonexceedance conditional distributions for the major crops were established based on rainfall.

Drought is a natural hazard that is characterized by a low amount of precipitation in a region. In order to evaluate the drought-related issues that cause chaos for human well-being, drought indices have become increasingly important. In this study, the monthly precipitation data from 1964 to 2013 (about 50 years) of the Jodhpur district in the drought-prone Rajasthan state of India was used to derive the effective drought index (EDI). The machine learning models hybridized with evolutionary optimizers such as the genetic algorithm adaptive neuro-fuzzy inference system (GA-ANFIS) and particle swarm optimization ANFIS (PSO-ANFIS) were used in addition to the generalized regression neural network (GRNN) to predict the EDI index. Using the partial autocorrelation function (PACF), models for forecasting the monthly EDI were constructed with 2-, 3- and 5-input combinations to evaluate their outcomes based on various performance indices. The results of the different combination models were compared. With reference to 2-input and 3-input combination models, both GA-ANFIS and PSO-ANFIS show better performance results with R2 = 0.75, while among the models with 5-input combination, GA-ANFIS depicts better performance results compared to other models with R2 = 0.78. The results are presented suitably with the aid of scatter plots, Taylor's diagram and violin plots. Overall, the GA-ANFIS and PSO-ANFIS models outperformed the GRNN model.

The present study focuses on the dependence modeling of hydroclimatic variables such as the El Niño–Southern Oscillation (ENSO) index, precipitation, tidal height, and groundwater level (GWL) in humid tropical coastal region of India. The rank-based correlation coefficient was used to determine the dependence between the pairs of cumulative monsoon precipitation of June–July–August–September (P_JJAS) and the postmonsoon groundwater level (PMGWL), ENSO–P_JJAS, ENSO–PMGWL, and GWL–tidal height. The results indicated that P_JJAS–PMGWL, ENSO–PMGWL, and GWL–tidal height had significant dependence, whereas P_JJAS–ENSO had no significant dependence. The best fit distributions for P_JJAS, PMGWL, and tidal height were found to be lognormal, extreme value, and generalized extreme value distributions, respectively, whereas for the ENSO index, it was the normal kernel-density function. The Archimedean families of copulas were used for dependence modeling, and it was observed that the ENSO–PMGWL was best modeled by the Frank copula, the P_JJAS–PMGWL by the Gumbel-Hougaard copula, and the GWL–tidal height by the Frank copula. The copula-based conditional probability for the Gumbel-Hougaard and Frank copulas for GWL were obtained to understand the risk associated with other hydroclimatic variables. Thus, copula-based dependence modeling could be useful for understanding the risk among hydroclimatic variables including groundwater.