Application of SWAT+ model to assess the hydrology of irrigated agricultural catchments in Western Cape, South Africa

Mabohlo, Sakikhaya

Title: Application of SWAT+ model to assess the hydrology of irrigated agricultural catchments in Western Cape, South Africa
Creator: Mabohlo, Sakikhaya
Subject: Uncatalogued
Date Issued: 2024-10-11
Date: 2024-10-11
Type: Academic theses
Type: Master's theses
Type: text
Identifier: http://hdl.handle.net/10962/464986
Identifier: vital:76563
Description: Agricultural activities can modify hydrological processes in catchments. An in-depth investigation of the impact of agriculture on hydrological processes is thus crucial for sustainable water management and utilisation planning in agricultural catchments. Hydrological models are widely used as practical tools to estimate and understand hydrological processes at various scales in catchments. However, it is often challenging to represent hydrological processes accurately in agriculture-dominated areas due to their complexity and the structural inadequacy of models. Complexities driven by agricultural developments such as reservoirs, cultivation irrigation, and water transfers are often intricate and difficult to represent in detail. The current study applied the SWAT+ hydrological model to the agriculture-dominated Twee and Leeu catchments in the Western Cape Province of South Africa. The SWAT+ model was preferred because it is versatile in representing multiple reservoirs and irrigation from multiple sources. The model application aimed to enhance the understanding of hydrological processes and the impact of agricultural water use on catchment hydrology, explore the capabilities and performance of this relatively new version of the SWAT in a complex catchment and provide baseline hydrology input for systems models that were used to formulate a water management plan for the catchment. The SWAT+ model was able to represent the significant number of reservoirs in the catchment. In some cases, small farm dams had to be lumped together to deal with the model's structural inadequacy. The model performed well in terms of observed and simulated streamflow comparison. NSE and R2 values above the standard of 0.5 were obtained for both catchments. Moreover, an NSE of 0.7 at a daily time-step simulation was obtained for the Leeu Catchment validation period, indicating a generally impressive SWAT+ performance. The model indicated that hydrological processes in the area were largely dominated by evapotranspiration, which is expected of semi-arid regions. 55% of the total water input into the Twee quaternary catchment is lost through evapotranspiration, while 22% is converted into surface runoff. Interflow and groundwater flow account for 9% and 9% of the catchment water, respectively. The remaining 5% is distributed to soil moisture and groundwater storage. For the Leeu, the model simulated a loss of 67% of the precipitation through evapotranspiration. Interflow and surface runoff were estimated to be 25% and 24%, respectively. Groundwater contribution to the stream accounts for 7%. Therefore, evapotranspiration and surface runoff are the dominant processes from a water balance perspective, whereas groundwater flow is also significant. Agricultural activities impacted the hydrological system significantly, and this is mainly attributed to the construction of numerous reservoirs and the subsequent intensive irrigation in the dry summer periods. The model simulation revealed that the reservoir and irrigation reduced the annual average streamflow by 71% in the Twee Catchment. In the Leeu Catchment, reservoirs and irrigation resulted in an annual average streamflow reduction of 77%. Therefore, agriculture has significantly altered flow patterns in the catchment, particularly downstream areas, during drier years. In conclusion, understanding of hydrological processes, reservoir and irrigation impact on catchment flow dynamics are the key outcomes that support decision-making regarding sustainable water management and utilisation planning. The broader outcomes, including spatially distributed flows and irrigation demand for different crop types, were used as baseline inputs for systems models that explored various water management options. SWAT+ thus played a crucial role in developing a water management plan for the area. Although the model exhibited some structural deficiencies in representing some catchment complexities, it fairly represented and enhanced our understanding of the hydrological dynamics of the Twee and Leeu Catchments. Additionally, the model has proven to be a versatile tool that can be applied for practical catchment water management in the agriculture-dominated catchments of South Africa.
Description: Thesis (MSc) -- Faculty of Science, Institute for Water Research, 2024
Format: computer
Format: online resource
Format: application/pdf
Format: 1 online resource (142 pages)
Format: pdf
Publisher: Rhodes University
Publisher: Faculty of Science, Institute for Water Research
Language: English
Rights: Mabohlo, Sakikhaya
Rights: Use of this resource is governed by the terms and conditions of the Creative Commons "Attribution-NonCommercial-ShareAlike" License (http://creativecommons.org/licenses/by-nc-sa/2.0/)

Hits: 34
Visitors: 33
Downloads: 1

Collections

RU Institute for Water Research, including the Unilever Centre for Environmental Water Quality (IWR)

		Thumbnail	File	Description	Size	Format
View Details Download			SOURCE1	MABOHLO-MSC-TR24-199.pdf	4 MB	Adobe Acrobat PDF	View Details Download