@phdthesis{,
    author = {Balcha, Yonas Abebe},
    title = {Hydro-Climatic Analysis and Future Water Resource Projections : Integrating CMIP6 Models and Hydrological Models in the Upper Awash Sub-Basin, Ethiopia},
    editor = {},
    booktitle = {},
    series = {},
    journal = {},
    address = {},
    publisher = {},
    edition = {},
    year = {2024},
    isbn = {},
    volume = {},
    number =  {},
    pages = {},
    url = {},
    doi = {},
    keywords = {CMIP6, SWAT, TAR, ANFIS, NARX, Climate Change},
    abstract = {In this study, an investigation into the prospective water resource conditions of the study area was conducted by integrating climate model outputs from the new CMIP6 archive with hydrological models. Comparative assessments were made between the outcomes of hydrological models utilizing data-driven approaches and those derived from the Soil and Water Assessment Tool (SWAT). The motivation for this research stems from the discrepancies in the outputs of numerous past studies in the sub-basin that primarily aimed to elucidate past hydro-climatic conditions. To gain insight into future water availability within the sub-basin, twelve climate models were initially selected and evaluated based on their capacity to simulate observed historical climates. A meticulous downscaling process was implemented under the CDO platform, followed by bias correction using the QM technique on all climate model outputs. Following a rigorous evaluation, five climate models (ECEARTH3, GFDL-ESM4, MPI-ESM1-2-HR, MRI-ESM2, and INM-CM5-0) were identified for their relatively robust potential in capturing various characteristics of the observed climate series. Time series models, categorized under regime switching and data-driven approaches, were scrutinized for potential. Self-Exciting TAR (SETAR) and Logistic-Smooth TAR (LSTAR) models, variations of TAR models, were applied to the flow data of selected stations in the sub-basin with a commendable degree of accuracy. Among the data-driven models (DDMs), the Adaptive Neuro Fuzzy Inference System (ANFIS) model exhibited the greatest potential in simulating the flow series of selected gauging stations in the sub-basin. Ultimately, the Nonlinear Auto-Regressive with eXogenous inputs (NARX) model from DDMs was chosen for its minimal data requirements and ease of model setup. Additionally, the SWAT model underwent calibration and validation. Subsequently, the NARX model, in conjunction with the SWAT model, was employed to analyze future water resource conditions in the sub-basin. Using the trained NARX model and the previously validated SWAT model, the examination of future water resource conditions took place at the sub-basin outlet with the assistance of new CMIP6 climate scenarios. These scenarios were derived by ensembling the outputs of the five selected climate models. The analysis projected an increase in both spring and summer flow for all scenarios (SSP1.26, SSP2.45, and SSP5.85) across near (2022-2039), mid (2040-2069), and far (2070-2099) time periods. Additionally, the outputs indicated a decrease in dry period flow. Results from the Standardized Discharge Index (SDI) analysis on simulated future water availability suggested a higher likelihood of wet events compared to dry ones. Considering existing land use and water resource management conditions in the sub-basin, these findings imply an increased probability of future flooding in the low-lying areas of the sub-basin and successive downstream parts of the Awash basin.},
    note = {},
    
    school = {Universität der Bundeswehr München},
    
}