Anil KumarMalik, Anurag2019-04-222019-04-222019-01http://krishikosh.egranth.ac.in/handle/1/5810100769Drought is a natural disaster which disturbs the entire ecosystem and adversely affects various sectors, such as agriculture, hydropower generation, water supply and industry. Occurrence of drought and its forecasting are critical components of hydrology which play a major role in risk management, drought preparedness and mitigation. This study was conducted using monthly rainfall and streamflow data of Almora, Bageshwar, Chamoli, Champawat, Dehradun, Haridwar, Nainital, Pauri Garhwal, Pithoragarh, Rudraprayag, Tehri Garhwal, U.S. Nagar/Pantnagar, Uttarkashi, Naula and Kedar stations located in Uttarakhand State, India, with the specific objectives to determine the spatiotemporal trends in hydro-meteorological data, find the best fit probability distribution, characterize meteorological and hydrological drought and wet conditions using Standardized Precipitation Index (SPI), Effective Drought Index (EDI), and Streamflow Drought Index (SDI), demarcate the homogeneous areas using Agglomerative Hierarchical Clustering (AHC), and predict hydro-meteorological drought and wet conditions using soft computing and statistical techniques. The results of trend analysis revealed significant positive (rising) and negative (falling) trends with different magnitudes in monthly, seasonal and annual rainfall time series data at 1%, 5% and 10% significance levels for 13 stations, while negative trend in monthly, seasonal and annual streamflow time series data at 1%, 5% and 10% significance levels at Naula and Kedar stations. The Kolmogorov-Smirnov test (K-S) statistic showed gamma distribution fitted well to 1-, 3-, 6-, 9-, 12-1 and 24-month rainfall and streamflow data series at 1% and 5% significance levels. The gamma distribution was used for analysis of hydrometeorological drought and wet conditions based on SPI, EDI, and SDI at 1-, 3-, 6-, 9-, 12-, and 24-month time scales for study stations. The occurrence of severe and extreme hydrometeorological drought and wet conditions were minimum, while normal, moderate drought and wet conditions occurred most frequently at 1-, 3-, 6-, 9-, 12-, and 24-month time scales for all the stations. The AHC analysis showed minimum three clusters (1, 2 and 3) and maximum four clusters (1, 2, 3 and 4) of similar characteristics in the study region. The performance of CANFIS model, followed by MLPNN, was found to be the best for prediction of hydro-meteorological drought or wet conditions based on the multi-scalar SDI, SPI and EDI values for most of the stations. The results of trend analysis and prediction of hydro-meteorological drought and wet conditions would help the local stakeholders, hydrologists, water managers and policy maker to understand the risks and vulnerabilities related to climate change and anthropogenic activities in the study region.ennullThesis