Geomorphometric Analysis of Baseflow Recharge in Aquifer Groundwater Assessment

Abstract

The inadequate understanding of geomorphometric impacts on the formation of groundwater baseflow recharge calls for an intuitive tool for managing the invisible dynamic water formation system. This study investigates the factors responsible for baseflow recharge formation and also determines the influence of geomorphometric parameters in the formation of baseflow in groundwater recharge of the Mthatha River Catchment (MRC) in South Africa. Specifically, the study evaluates the relationship between morphometric, geological, land-use, and hydrometeorological variables and determines their influence on baseflow recharge formation. The arc hydro-statistics tools in quantum geographical information system (QGIS) were used to process the aster digitized elevation model (DEM) for land, soil, and geology with meteorological rainfall and temperature data in (QSWAT) to process the correlation factors that influence baseflow recharge formation. The results show bifurcation ratio, drainage density, and relief ratio values of 0.0034, 0.0056, and 1.232, respectively. The strong correlation coefficient of 0.76 obtained for land use/landcover variables and other hydrogeological characteristics of the area depicts a significant contribution in circulation flow movement for the likelihood of sufficient water yield for the study area. Decision-makers would find the study’s outcome useful in visualizing the invisible controlling variables for baseflow recharge and runoff generations.

1. Introduction

Among the major components of streamflow is the baseflow. Baseflow is an essential constituent of stream or river runoff. It is the water from the delayed subsurface flow at a shallow depth above the groundwater table (GWT), joining a nearby stream [1]. Different aquifers in groundwater discharge into streams when the water table rises above the streambed. This process is often referred to as the base flow. Since groundwater contains all the water beneath the surface saturation zone (zone of aeration). In contrast to the saturated zone, which is the area that is bounded by the water table. The zone of aeration occupies the layer from which the water table to the ground surface, and it consists mainly of three sections: the soil-water zone, intermediate zone, and capillary fringe. Most of the baseflow formation water is found in-between the soil-water zone, and capillary fringe. The amount of base flow a stream receives is closely linked to the permeability of rock or soil in the catchment area [2, 3]. Essentially, baseflow is crucial for maintaining the stability of the stream and ensuring its continuous flow [4]. The stream flow consists of the direct runoff—which lasts for hours or days depending upon the catchment size, while the base flow emerges from delayed subsurface runoff and groundwater resources.

Groundwater resources have played a fundamental role in water supply augmentation schemes [5]. It has been used to support most economic activities and human development in rural and urban environments [6]. However, there exists a dearth or few quantitative studies that have evaluated the controlling variables responsible for base flow recharge in aquifer groundwater assessment and runoff generation in catchment water formation. This may be due to various induced complexity, bias, and varying intricate factors controlled by surface water—groundwater interaction with the catchment characteristics such as land cover/land use [7, 8]. Many past studies have shown that hydro-climatic variables, geomorphometric, and catchment topography do influence base flow [8, 9]. For instance, groundwater inflow may be received from a stream in a wet climate, while a stream in a similar physiographic setting in an arid climate might lose water to groundwater [10]. Also, meteorological factors, particularly rainfall, and temperature [11], considering their seasonality and the intensity of individual rainfall events [12, 13, 14] do influence base flow. In addition, some studies have highlighted the effect of topography as defined by the catchment morphometric analysis such as relief of the area, slope steepness, and drainage porosity as contributors to base flow recharge/discharge [15, 16]. Li et al. [17] have highlighted the importance of lithographic features such as size and permeability beneath aquifers as major controlling factors that contribute to base flow. The varying contributing factors/processes that could potentially affect base flow recharge depend on catchment characteristics and regional lithographic features [17]. Some authors have elucidated the significant roles of soil internal thermal mechanism and soil moisture content in determining the baseflow recharged rate for an area [18, 19]. Since, there is currently no consensus on determinant factors that influence or control the groundwater and surface water interaction in determining baseflow formation recharge from works of literature [19, 20, 21], and the least understood, mainly because of the difficulty of measuring the different spatiotemporal variables requirement. Most groundwater base flow recharge simulation requires a wide range of catchment characteristic data such as climatic, geologic, hydrologic, and physiographic [11, 22, 23, 24]. Hence, there is a need to develop methods to better quantify present-day base flow recharge formation in predicting future imbalances between water supply and demand in a catchment. Moreso, since large areas of water development and sustainability rely on a proxy of groundwater aquifers discharge to streams or rivers for its flow stability and sustainability. Thus, this study’s importance cannot be over-emphasized. The study investigates the factors responsible for baseflow recharge formation and their controlling parameters using geomorphometric and hydrologic characteristics.

Globally, groundwater resources only account for 3% of the world’s freshwater and yet it is the largest and most widely distributed source of freshwater [25]. For a country like South Africa which has been classified as a semi-arid country and the 30th driest country in the world [26], the impact of drought has negatively impacted the growth and development of many sectors of the economy, including the aquatic ecosystems [11, 25, 26]. Moreover, the country’s water resources have been under increasing threat from pollution in recent years, due to rapid demographic changes and the establishment of human settlements lacking appropriate sanitary infrastructure [27, 28, 29]. This applies especially to peri-urban areas, which surround the larger metropolitan towns in the country, where many of the settlements have developed with no proper water supply and sanitation services. People living in these areas, as well as downstream users, often utilize the contaminated surface water for drinking, recreation, and irrigation, which creates a situation that poses a serious health risk to the people [30, 31, 32]. Hence, this necessitates the country hydrogeologists to tap into groundwater resources as an alternate source of water supply in addition to the availability of traditional surface water.

Consequently, most of the existing studies on groundwater yield estimation in South Africa have been targeted to achieve regulated development both in rural and urban municipal water supply augmentation [33, 34]. Furthermore, many of the previous studies like [35, 36, 37, 38] have sought contributions to the background information about groundwater potential development and recharge quantification. However, few or limited studies have assessed both the quantity and the quality of groundwater conjunctive use. Among the past methods that have been employed in the estimation of baseflow are the use of the statistical approach, the use of analytical and predictive algorithms, and non-mechanical-oriented models, [39, 40, 41, 42, 43, 44]. Despite these avalanche approaches, the use of statistical time series modeling has provided an intuitive alternative to simulate or forecast many of the complex hydrologic-induced impacts of morphometric variables on the catchments system [29, 33, 45], while the use of a non-mechanical-oriented model proves to be simpler, depending on the catchment size and specific nature of the system’s structured data [45, 46]. Even though cumulative errors still permeate most of the prevailing available methods for estimating baseflow recharge due to uncertainty and general assumptions in their use in groundwater visual modeling [47].

The use of statistical techniques has found great applications in baseflow recharge estimation [48]. Methods like the use of a regression approach to combat the problem of multicollinearity and confounding variables [49, 50, 51, 52, 53]. The stepwise regression analysis has proved useful in isolating a set of hydrogeological parameters that constitute the major yield predictor variables for any study area [52]. The multiple linear regression (MLR) analyses are particularly useful for addressing issues relating to prediction such as identifying a set of predictors that will maximize the amount of variance explained in the criterion [54]. In general, the regression analyses that include relative weight analysis and dominance weight analysis have proved useful in minimizing errors and bias toward determining if there exists any relationship between the determining factor and other independent variables.

Other emerging methods used to estimate groundwater recharge use multiple statistical techniques and compare the results for validation [48]. The complexity reduction in datasets has made the use of MLR, principal component analysis (PCA), and factor analysis (FA) to be purposely used to identify outliers and to reduce the variety of collected data matrices into a few selected derived components in a statistical approach [2, 55, 56]. The newly derived component known as the “principal component” (PC) variables can now form a true representative of the original sets [55]. The PCA has been used for mitigating the problem of multi-dataset reduction into orthogonal components [56]. Hence, this study investigates and determines the influencing geomorphometric parameters necessary for baseflow recharge formation using geologic, hydrologic, climatic, and physiographic data parameters correlation in forming a new set of orthogonal elements that can serve as controlling factors responsible for baseflow recharge in the catchment formation system. The study is of significance in base flow estimation as it contributes to predicting/estimating the amount of base flow rate needed to maintain the stability of streamflow and runoff generation. No doubt, the study research outcome will be useful for making informed decisions in groundwater resources management and conjunctive surface water uses. Stakeholders, engineers, and decision-makers would the study useful in determining peak runoff, flood risk mitigation, and combating drought impact both for local and global water use.

2. Study area

The Mthatha River Catchment (MRC) in the King Sabatha Dalindyebo Local Municipality is situated in the Eastern Cape Province, South Africa (Figure 1). The river is endowed with natural water resources that are of immense value to all its riparian users, both for their domestic supply, irrigation, tourism, and industrial uses for their subsistence livelihoods [57].

Figure 1.

Map of the study Mthatha River catchment.

Figure 2 depicts the catchment streamflow network from the processed raster DEM image to delineate the catchment into 89 sub-catchments and 289 hydrological response units (HRU) after rigorous arc hydro-GIS analysis in QGIS. Among other catchment factors affecting the distribution of groundwater are the vegetation and the catchment drainage network [7]. The drainage network characteristics such as size, area, shape, slope, and stream network distribution orders are useful in estimating surface flow discharge. The stream flow network (Figure 2) is of radial drainage forms where water flows downward from a hill or mountain, thereby providing linked to a wheel circular network of parallel channels flowing away from a central high point.

Figure 2.

The Mthatha River and streamflow network.

Figure 3 illustrates the digitized elevation model with its lithographic geologic features classification for the area.

Figure 3.

The digital elevation model of the study area in meters.

The study area elevations have a minimum elevation of 4 m; maximum elevation of 1741 m; and mean elevation of 768.03 m with a standard deviation of 242.06 m for the whole MRC. The MRC geology is hilly with deep valleys and incised gorges. The dominant geological formations in the region are the Dwyka foundation and the Beaufort group. The Dwyka foundation is fine-grained and compact, and thus forms a poor aquifer, with boreholes yielding less than 0.7 l/s. The average yield of boreholes in this rock type is 0.5 l/s [42]. However, the Beaufort Group is mostly sandstone. Figure 4 illustrates the land cover classification summary.

Figure 4.

The Mthatha River catchment landcover and land classification.

The land use reclassification based on QSWAT -GIS was aggregated to eight major land use/land cover (Figure 5). The map codes classified AGRL, PAST, FESC, BARR, URBN, WATR, RNGE, and UCOM to represent agriculture, pastures, forest, barren, urban settlements/residential, and water occupied area with varying percentages while Figure 5 depicts the catchment’s soil-type characteristics codes of Bh11-1b-27, Fr8-2-3b-51, and Ge22-a-55 for the sandy loam, loam, and sandy clayey loam for the area.

Figure 5.

The Mthatha River catchment soil type and characteristics.

The major soil types include open-texture sandy-loam soils, which will tend to be associated with higher infiltration volumes than fine-grained closely compacted clay soil. The soil type is light silt loam and dark to dark red clay soils, which seem to have different distributions.

3. Materials and methods

3.1 Research design overview

This study uses a descriptive research design which allows a proper description using a quantitative research strategy to identify the controlling factors for base flow recharge formation. After preliminary site visitation, the arc hydro-statistic tools in the GIS environment were used to process the resample raster images for DEM, soil, land use/land cover, and geology features. The topography morphometric analysis was carried out for relief, aerial, and linear parameters, while the quantum soil water assessment tool (QSWAT) embedded in the QGIS environment was used for the catchment groundwater hydrological routing for water balance configuration. The QSWAT models the base flow recharge as a function of the hydraulic conductivity of the aquifer with the distance from the ridge or sub-catchment divide for the groundwater system to the main channel, and the water table height [58]. In all these, seven factors were identified from literature works (Table 1). These factors and their variables compilation were based on similar studies [3, 7, 8, 9]. In all, the combined identified 34 parameters that could potentially contribute to base flow recharge and runoff formation in the study area were estimated after rigorous arc hydro-statistics analysis. Table 1 depicts the identified catchment controlling factor that could affect the baseflow recharged in aquifer groundwater formation.

Controlling factors	Value
Morphometric parameters	Degree
Topographic	%
Soil relief/type	%
Land use cover	%
Geology	%
Vegetation cover	%
Hydrometeorological	mm/year

Table 1.

Catchment controlling factors for baseflow recharge formation.

These controlling catchment factors and their variables are important in characterizing the drainage network and base flow recharge for the area. As a first step in identifying the influencing controlling factors for baseflow recharge, the PCA was used to screen and reduce the dimensionality of the potential parameters as indicated by the factor loading analysis. These controlling parameters were derived from variables with high factor loading (>0.75) and were assumed to be the parameters that best represent the catchment baseflow recharge contributors. The parameters were derived from detailed hydrological routing assessment for hydrometeorology, topography, soil texture, land-use type, and geological combined datasets. Only components with an eigenvalue greater than one that was retained [55]. Thereafter, the multiple linear regression analysis (MLR) in XL-STAT statistical software was used to model the strength of possible relationships between the high-loading parameters that may be responsible for baseflow recharge. Next, Pearson’s correlation coefficient was used to determine the strength of each variable (predictor) in explaining their significant contribution to the baseflow in recharge formation. Correlation was used to assess the direction and strength of relationships across different units.

3.2 Data and QGIS hydro-statistical analysis

The various stations’ source data were processed and subjected to a rigorous scientific method to test their accuracy, reliability, homogeneity, consistency, and localization gaps as part of preliminary data cleaning and integration before use. The years 2012–2020 represent the common base and the corrected available data used for this study. Remote sensing (RS) source data for DEM, land, and soil were processed in a geographic information system (GIS), and digitized by their global positioning system (GPS) was used for the catchment characterization mapping and were used for the catchment characterization and delineation of drainage patterns for a better understanding of the hydrologic system. Several studies have used it and have proved to be an efficient tool in water resources management and its planning [59, 60, 61]. The use of the digital elevation model (DEM) and its associated derived dataset (slope, aspect, etc.) has been the starting point for defining the catchment boundary, delineation, flow direction, and stream network, which are usually prepared as input for hydrologic models [60, 61, 62]. The catchment boundary and delineation start by completing the filled sink, flow direction, flow accumulation, creation of catchment pour point, snaping the pourpoint, stream link, and order creation for stream definition and segment in the catchment grid delineation following the polygon process in the creation of drainage line, adjacent catchment and drainage profile. Details of stream order extrapolation as proposed by Strahler [61] and the method for morphometric analysis in arc hydro-statistic as a module in ArcGIS surface analysis are well documented in Refs. [60, 61, 62]. Various morphometric parameters such as linear aspects of the drainage network: stream order (Nu), bifurcation ratio (Rb), stream length (Lu), and areal aspects of drainage basin: drainage density (D), stream frequency (Fs), texture ratio (T), elongation ratio (Re), circularity ratio (Rc) form factor ratio(Rf) of the catchment were computed.

4. Results and discussion

4.1 Identification of controlling factors

In all, seven controlling factors were identified as potential contributing variables to these factors that influence base flow recharge in works of literature review [1, 3, 4, 5, 6, 7, 8]. This includes topographic morphometric parameters such as catchment slope, relief, land-use and land cover type, geologic, soil type, vegetation cover, and hydrometeorological variables. Details of these factors give pertinent parameters necessary for hydrogeology correlation that give insights into baseflow recharge for water formation. The catchment morphometric factor consists of relief, aerial, and linear parameters; while the topography class factor (slope steepness and drainage porosity); the soil relief/type denotes the sandy loam, loam, and sandy clayed loam texture for the area while the land use and vegetation cover as factor type include percentage area for agriculture, pastures, forest, barren, urban settlements/residential, and water occupied area. The geologic factor includes variables for the percentage area occupied by the beneath rock formation (undeferential basement complex, fractured sedimentary rock - Karoo Supergroup; unconsolidated aquifers and percentage area of fracture granite- basement complex occupied for the area). The different geologic formation determines their size, permeability, and hydraulic conductivity, while the vegetation cover as a factor was interpreted to include other land cover vegetation. The hydrometeorological factor class denotes the dry and wet seasons’ rainfall and temperature parameters with the total yearly rainfall amount.

4.2 Catchment characteristic parameters

Table 2 depicts catchment morphometric parameter analysis with their employed method. The various catchment characteristic parameters were generated after rigorous arc hydro-statistical analysis for the different morphometric variables. Many of these parameters give insight into the topography and the geologic history of the rocks underlying the study area.

Variables	Morphometric Parameters	Method	References
Linear	Stream order (u)	Hierarchical order	Strahler, 1964
	Stream length (Lu)	Length of the stream	Horton, 1945
	Stream length $(R_{\mathit{SL}}$)	$R_{\mathit{SL}}=L_{\mathit{st}}/L_u-1;$where, $L_{\mathit{st}}=$ Total stream length of order (u), $L_u-1$=The total stream length of its nest lower order	Horton, 1945
	Bifurcation Ratio $\left(R_b\right)$	$R_b=N_d/N_u+1$; where, $N_d$ =Number of stream segments of the next higher order	Schumn, 1956
Relief	Basin relief	Vertical distance between the lowest and highest points of the basin	Schumn, 1956
	Relief Ratio $\left(R_h\right)$	$R_h=B_t/L_b$ where, $B_h$ =Basin relief, $L_b$ =Basin Length	Schumn, 1956
	Ruggedness Number $\left(R_n\right)$	$R_n$ = $R_b$x D where, $D_d$ = Drainage density, $L_b$ = Basin length	Schumn, 1956
	Drainage density $D_d$	$D_d$=L/A, where, L = Total length of stream, A =Area of the basin	Horton, 1945
	Stream frequency $F_s$	$F_s$ =N/A where, N=Total number of stream, A =Area of the basin	Horton, 1945
Aerial	Texture Ratio (T)	T=N/P where $N_i$ =Total number of stream, A =Area of the basin	Horton, 1945
	Form Factor $F_f$	$F_f$ = A/($L_h)$2 where, A=Area of basin, Lb=Basin Length	Horton, 1945
	Circulatory Ratio ($R_c)$	$R_c=4\mathit{\pi A}/P^2$ where, A=Area of the basin, $\pi$=3.14, P=Perimeter of basin	Schumn 1956
	Elongation Ratio $\left(R_e\right)$	$R_e=\raisebox{1ex}{$\sqrt{\frac{A}{\pi }}$}\!\left/ \!\raisebox{-1ex}{$L_b$}\right.$Where A=Area of watershed, $\pi$=3.14, $L_b$=Basin	Schumn, 1956
	Length of overland flow $L_o$	Length $L_o$= $\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2D_d$}\right.$	Horton, 1945
	Constant Channel Maintenance	$L_{\mathit{cd}}$ = $\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$D_d$}\right.$, where $D_d$=Drainage density	Horton, 1945

Table 2.

Morphometric parameter analysis and their method [61, 62].

4.3 Hydrologic characteristic controlling parameters

Table 3 presents the identified factors and parameters for the catchment morphometric, detailed topographic features, land use/soil texture cover, geologic, and hydrometeorology variables description these identified potential parameters were collated from similar works of literature review [1, 3, 4, 5, 6, 7, 8]. In all, these 34 parameters could serve as controlling prerequisites for baseflow recharge in water formation for the area.

Morphometric parameters	Variable name	Value
Linear Properties	Basin area (km^2)	169,564.390
	Basin perimeter (km)	27,160.450
	Length of mainstream (km)	506.862
	Bifurcation ratio	2.250
Properties of the catchment
	Drainage density (km/km^2)	1.103
	Stream frequency (km^2)	0.002
	Texture ratio (km)	5.201
	Drainage intensity (km)	0.053
	Form factor	0.660
	Circulatory ratio	0.002
	Elongation ratio	0.459
	Length of overland flow (km)	0.002
	Constant channel maintenance	334.568
Relief properties
	Basin relief (km)	2.830
	Relief ratio	0.011
	Ruggedness number	0.006
Geological variables
	Percentage area of Dwyka foundation and undeferential basement complex	40.230
	Percentage area of Beaufort group with fractured sedimentary rock.	30.430
	Percentage of area under unconsolidated aquifers along the coasts	14.090
	Percentage area of fracture granite- basement complex	15.250
Land use and land cover
	Percentage area of agriculture	6.790
	Percentage area of pastures	28.960
	Percentage area of forest	45.200
	Percentage area of barren	4.330
	Percentage area of urban commercial	0.970
	Percentage area of settlements	9.020
	Percentage of water area	0.220
Hydrometeorological
	Dry season rainfall (mm)	147.000
	Wet season rainfall (mm)	453.200
	Total rainfall (mm)	700.000
	Dry season temperature (°C)	14.400
	Wet season temperature (°C)	33.170
	Mean temperature (°C)	24.408

Table 3.

Hydrological parameters compilation at MRC for 2000–2020.

The accuracy degree of these morphometric parameters analysis is required for baseflow recharge estimation. The bifurcation ratios express the ratio of the number of streams of any given order to the number of streams in the next order. It gives insight into the geologic structure does not distort the drainage pattern [62]. The ratio is usually between 3.0 and 5.0 for a basin. The mean bifurcation ratio value is 2.250 for the study area (Table 3), which implies the geological structures are less disturbing the drainage pattern. The length of the mainstream is a significant hydrological parameter of a catchment, as it reveals how flatter the catchment gradients are. A shorter stream length is an indication of an area with larger slopes and finer textures characterized the surface runoff. The stream network properties help study the landform-making process while the drainage pattern is useful in determining storm hydrograph and peak runoff in a catchment. Also, water concentration capacity in a rich drainage system is greater as it enables water to run a shorter distance to the streams [33, 45, 46]. The study drainage density shows the closeness of spacing of channels, hence providing a quantitative measure of the average length of stream channel for the whole catchment. It depicts both the geologic and climatic types that an area is prone to. The drainage density of the study area is 1.103 km/sq.km depicting low drainage density. A low drainage density is more likely to occur in areas with high resistance of high permeable subsoil material under dense vegetative cover and vice versa. Another important drainage morphometric analysis is the texture ratio (T). This parameter depends on the relief aspect, infiltration capacity, and the underlying lithology of an area. The study area texture ratio is 5.201, which is classified as moderate. Both the stream frequency and the drainage density of an area are indications of stream population correlation.

Based on the pre-screening of the data using PCA, the collected datasets were classified into three main components, namely PC1, PC2, and PC3. The PCA reveals the hidden structure of the dataset. The outcome of the PCA helped to identify which factors have a dominant influence and provide insights into variables influencing base flow recharge. Factor analysis (FA) as a choice from PCA screens these 34 catchment parameters into three major PCs (Table 4). The principal component (PC) with the highest factor loading value as shown in Table 4, explains the order of their importance. It can be observed from Table 4, that PC1 explains most of the variability associated with the variables. The bold value in PC1 denotes topography, land use type, geology, and hydrometeorology variables that are of pertinent importance, while other values are of lesser variance factors.

	Base flow recharge formation
Variables name	Symbol	PC1	PC2	PC3
Basin area	BA	0.102	0.004	0.235
Basin perimeter (km)	BP	0.086	0.006	0.000
Length of mainstream (km)	LMS	0.160	0.186	0.009
Bifurcation ratio	BR	0.629	0.072	−0.369
Drainage density	DD	0.741	0.441	0.013
Stream frequency	SF	0.155	0.169	0.150
Texture ratio	TR	0.130	0.098	−0.028
Drainage intensity	TI	0.137	0.014	0.236
Form factor	F	0.140	0.027	0.000
Circulatory ratio	CR	0.038	0.008	0.046
Elongation ratio	ER	0.036	0.016	0.011
Length of overland flow	LOF	0.003	0.037	0.033
Constant channel maintenance	CCM	0.142	0.061	0.019
Basin relief	BRE	0.136	0.018	0.003
Relief ratio	RER	0.645	−0.279	−0.035
Ruggedness number	RN	0.041	0.049	0.147
Percentage area of Dwyka foundation and undeferential basement complex	UBC	0.704	−0.122	0.011
Percentage area of Beaufort group and fractured sedimentary rock	FSR-KoG	0.547	−0.857	0.020
Percentage of area under unconsolidated aquifers - along the coasts	UnCo
Percentage area of fracture granite- basement complex	FGBC	0.022	0.006	0.013
Percentage area of agriculture	AGRIC	0.026	0.016	0.020
Percentage area of pastures	PAST	0.177	0.014	0.236
Percentage area of forest	FORT	0.000	0.010	0.000
Percentage area of barren	BA	0.011	0.057	0.002
Percentage area of urban commercial	UC	0.714	0.026	−0.410
Percentage area of settlements	SET	0.625	0.502	0.000
Percentage of water area	WAT	0.162	0.004	0.235
Dry season rainfall	DSR	0.869	0.006	0.000
Wet season rainfall	WSR	0.606	0.486	0.009
Total rainfall (mm)	To PPT	0.027	0.372	0.369
Dry season temperature	DS Temp	0.119	0.041	0.013
Wet season temperature	WS Temp	0.152	0.169	0.150
Mean temperature	Mean T	0.030	0.198	0.028

Table 4.

Principal component analysis with base flow.

*

Values in bold correspond to each variable with high factor loading, the factor for which the squared cosine is the largest.

According to the PCA, the degree of boldness corresponds to the variables in which factor loading is the highest (greater than 0.75) and with spatial eigenvalue degree variation greater than 1. The PCA derivative was based on the orthogonal matrix correlation of standardized data. The negative signs imply loss of water and should be interpreted as variables responsible for sub-catchment where lateral deep percolation exceeds inflow rainfall (precipitation–runoff). Table 5 depicts the eigenvalue factor analysis for the major parameters that contributed positively to baseflow recharge. The result shows bifurcation ratio (0.629), drainage density (0.741), and relief ratio (0.645) denote variables for topographic morphometric properties and their variation degree in influencing base flow recharge, while the percentage of Dwyka foundation for undeferential basement complex (0.794) and fractured sedimentary rock for Beaufort group (0.542) as the factor for beneath geological variables. The urban commercial percentage area (0.704) and the settlements percentage area with (0.605) denotes-variation for land use/soil type and vegetation area cover factors while the rest variation accounts for the dry season (0.869) -variable and wet season rainfall variable (0.606)-represent major contributing variables for hydrometeorological factor. Table 6 depicts the newly selected parameters with high factor loadings only, whether it is negative or positive loading.

Variables (%)	PC1	PC2	PC3	PC4	PC5	PC6	PC7	PC8	PC9
Eigenvalue	4.15	1.72	1.04	0.79	0.58	0.33	0.23	0.13	0.05
Variability	46.10	19.07	11.56	8.81	6.42	3.62	2.52	1.39	0.51
Cumulative	46.10	65.18	76.73	85.55	91.99	95.58	98.10	99.49	100.00

Table 5.

Factor analysis of the eigenvalue.

Parameter	F1	F2	F3
Bifurcation ratio	0.984	0.001	0.010
Drainage density	0.924	0.032	0.003
Relief ratio	0.936	0.007	0.046
% Area of urban/commercial	0.978	−0.005	0.016
% Area of settlements/residential	0.989	0.004	0.006
Dry season rainfall	0.885	0.014	0.099
Wet season rainfall (mm)	0.105	0.890	0.003
% area of undeferential basement complex	0.377	0.014	−0.236
% area of fractured sedimentary rock	0.440	0.027	0.000

Table 6.

Contributing hydrological parameters.

The bold figures depict respective variables to the factor for which the squared cosine is the largest, this signifies the most significant variables that affect discharge flow.

The major contribution variables of bifurcation ratio (0.629), drainage density (0.741), and relief ratio (0.645) account for 46.10% variability in PC1. The bifurcation ratio, drainage density, and relief ratio denote variables for topographic morphometric properties in their order of importance, while the percentage of area for fractured sedimentary rock and unconsolidated aquifers account for 65.18% variation factor for beneath geological variables, while the percentage area of urban commercial and settlements account 76.73% variation for land use/soil type vegetation area cover factors. The eigenvalue with the highest loading factor consists of the morphometric variables for bifurcation, and drainage density followed by land use/vegetation cover factor for % area of urban and rural residential settlement in PC1. Furthermore, agriculture and forest land-use types are the best representative variables in F2, while the percentage of urban land-use type is the only variable contributing to the variance in F3. The bold squared cosine values depict the most significant variables that affect baseflow recharge and account for the surplus water that is readily available both for infiltration and runoff.

Besides, the data samples were later grouped by the factor analysis into cluster classification as depicted in Table 6. In all, meteorological, morphometric, and land use/land cover with geologic factors are the most contributing variables based on potential factor loading F1 while other clusters depict the minor axis with associated degree of variables component.

It can be observed from Table 6, that factor loading explains most of the variability associated with the variables. The bold squared cosine values depict the most significant variables that affect discharge flow. Figure 6 shows three major cluster factors “F1, F2, and F3” with cumulative variability of “81.90, 3.14, and 0.86%” for the hydrogeological variables’ classification, respectively.

Figure 6.

Latent factors variability of the cluster factors.

The factor analysis loadings (F1) depict the most significant variables that affect base flow recharge flow. The negative data samples sign denotes the discharge result for deeper aquifer drainage properties that denote less significant variables.

4.4 Dominant factor analysis results

Furthermore, correlation analysis was applied to examine the types of associations that exist among the selected contributing hydrological factors (bifurcation ratio, drainage density, relief ratio, percentage of urban/commercial area, percentage area of settlements/residential, percentage area of undeferential basement complex percentage area of fractured sedimentary rock - Karoo Supergroup and dry and wet season rainfall, which constitute dominant factors. Table 7 shows the correlation matrix’s significant level for the different contributing hydrogeological factors. The correlation matrix for the selected hydrogeological variables (topography, land use/vegetation cover, with other meteorological, and geological data). The lithographic factors presented by the bifurcation ratio had a strong, positive correlation with wet season rainfall (0.7480), percentage of the urban area (0.6456), percentage of undeferential basement complex (0.5439), indicating an increase in baseflow storage. The wet season runoff and relief ratio (-7471) negatively correlated with wet season runoff meaning water lost. The correlation matrix of these hydrological and geological variables depicts statistical similarity for producing valid findings of the influence of watershed characteristics on base flow.

	BF	Dd	Rr	%Ur	%S	DSR	WSR	% UBC	% FSR
BF	1
Dd	0.5456	1
Rr	0.6547	0.8357	1
%Ur	0.6456	0.7418	0.9438	1
%S	0.6541	0.6043	0.8744	0.6735	1
DSR	0.1234	0.7486	0.7571	0.5133	−0.925	1
WSR	0.7480	0.9485	−0.747	0.5439	0.9785	0.7480	1
% UBC	0.5439	0.9485	−0.747	0.5439	0.9785	0.9785	0.5431	1
% FSR	0.9785	0.9485	−0.747	0.5439	0.9785	0.654	0.9785	0.9785	1

Table 7.

Correlation matrix of hydrological and geological variables.

Confidence = 95% (Significance level = 0.05).

Generally, the geomorphometric forces are driven by the fluvial process, which is governed principally by relief and underlying geology of the area. The strong positive correlations among morphometric and meteorological parameters depict the degree of dependency on rainfall and vice versa. Also, the correlation matrices for these hydrogeological variables depict a positive strong relationship which implies that as these variables increase, there is a likelihood of the groundwater yield to increase. Further evidence from the fractured sedimentary rock, Karoo Supergroup; and unconsolidated aquifers that exist along the coasts reach of the study area may imply deep depth for borehole digging and water yield, which may only hold for a limited level of deepness after which the position stops functioning [45]. No doubt, the contribution of topographic morphometric variables, land use, hydro meteorologic, and geologic parameters had been responsible for the base flow recharge in this study, their strong positive correlation results show their likelihood of increased base flow recharge for the area.

5. Key findings of the study

The results of our geomorphometric parameters analysis in the GIS environment clearly show lateral hydraulic transport mechanisms from topography factor as the main properties, followed by land use, and land cover properties as major contributing variables in the orthogonal cluster classification in determining the controlling factor necessary for baseflow recharge formation while the minor axis is lithological/geological with vegetation cover as moderate factors for low runoff and low filtrations in the relief properties. Also, the morphometric analysis of the MRC area results in drainage density, texture ratio, circulatory ratio, and elongation ratio shows that the texture of the basin is low, and the shape of the catchment is almost flat. Besides, the given areas are of moderate flat terrain, the catchment characteristic that is associated with a late topographic stage for erosion development. This stream network property helps in studying the landform-making process, while the drainage pattern is useful in determining the water concentration capacity. The study area has a rich drainage system which enables greater water to run a shorter distance to the streams.

In addition, the correlation matrices for these hydrogeological variables depict a positive strong relationship, which implies that these variables increase their likelihood as major factors as controlling parameters to baseflow recharge. This also corresponds to the core principles of hydrogeology [32, 60]. Furthermore, it is evidence from the fractured sedimentary rock, Karoo Supergroup; and unconsolidated aquifers that exist along the coasts reach the study area may imply deeper depth for borehole digging and water yield, which may only hold for a limited level of deepness after which the position stops functioning [47]. In conclusion, baseflow recharge and runoff generation in South Africa may not so be equally spread due to geological formations that do not contain major aquifers and disparities exist from one location to another, this might hinder groundwater development on a national scale. This study has contributed to investigating the processes responsible for estimating groundwater yield and factors contributing to base flow recharge for the study area. The study outcome is of utmost benefit for analyses of peak runoff determination, flood area estimation, and mitigation against deep erosion abstraction for sustainable water resources management in the catchment.

6. Direction for further study

The current study has focused on factors that influence geomorphometric parameters in baseflow recharge formation at MRC in King Sabatha Dalindyebo local municipality, future research studies may examine the impacts of a transboundary aquifer system and water transfer into the new aquifer. Also, the use of an automatic, online instrument in providing continuous data in characterizing baseflow recharge is necessary for real-time assessment of borehole yield. Also, future baseflow recharge quantification and quality assessment can be investigated using other simulation and optimization methods. This would improve the understanding and modeling of groundwater variability and emerging catchment pollutants trace.

7. Limitations of the study

The accuracy of secondary data and thematic satellite data analyses from the GIS environment limited the values in the Tables and Figures. Also, the study did not consider the different sites’ uncertainty for the source data and GIS process data. The identification of recharge zones, quantification of recharge rate, and baseflow formation estimate were simulated from the QSWAT to which the results presentation is beyond the present study scope but are presented in another sister manuscript titled base flow recharge estimation and validation processes. Also, most of this study’s calculations had largely been based on the average value, accomplished theories, and principles by other hydrogeology similar scholars’ work. To this effect, the interpretation of the results had largely been based on intuitive reasoning, and complementary documents from the Department of Water and Sanitation at Mthatha, South Africa.

8. Conclusions

This study investigates the influence of geomorphometric parameters in baseflow recharge formation at MRC in the King Sabatha Dalindyebo Local Municipality, South Africa. The study has elucidated more on the likelihood impacts of different controlling parameters that may explicitly influence the baseflow recharge formation. Although limited ephemeral consideration for temporal and spatial variation in climatic conditions was inculcated, the major controlling parameters for the study are catchment characteristic topographic-morphometry feature: bifurcation ratio, drainage density, and relief ratio for correlation matrices for geologic and hydro-metrological parameters is a major variable in establishing the nexus between hydrogeological indicators and baseflow recharge for the study area. Also, considering the extent of GIS usage, the various delineation results and study outcomes are useful for determining local peak runoff, estimation of surcharge flood depth, and maximum flow surcharge coverage area can be determined. The study’s contribution to the body of knowledge lies in the characterization of land use and land cover variables, which can play a significant role in hydraulic structure design and seasonal variability factor as it affects groundwater yield. In conclusion, there is a need for up-to-date morphometric analysis and mapping investigation to know the contributing variables in baseflow recharge quantification. Also, there is a need for proper planning to review the policy and regulatory framework for managing groundwater aquifer supply and to assess the different hydro-geological parameters that may influence baseflow recharge formation. Stakeholders, engineers, and decision-makers would find the employed tools useful in determining the likelihood of baseflow recharge and aquifer formation yield in the area and Walter Sisulu University through the risk and vulnerability science center for logistic support.

Acknowledgments

The authors would like to acknowledge the support from the Department of Water and Sanitation (DWS), Mthatha, and South Africa Weather Bureau Services (SAWS), for providing the data that was used in this study and Walter Sisulu University through the risk and vulnerability science center for logistic support.

Conflict of interest

The authors declare that the research was conducted without any commercial or financial relationships that could be construed as a potential conflict of interest.