Introduction

Freshwater is an essential resource for the functioning of ecosystems, for sustaining human life, and for fostering economic growth1,2. The effect of man-made impacts and climate change on fresh water resources in arid and semi-arid regions have been reported to be significant, posing a growing challenge to meet increasing water demands3,4,5,6,7. Desertification and high evaporation losses, coupled with the impacts of climate change, also exacerbate water scarcity in such environments. Besides the increase in population and the growth of industries, Southern African countries face significant challenges in managing the effect of the continuous expansion of urban areas, which results in an increase in water demand for industrial and potable uses. The population of Botswana, the focus of this study, has increased from 2.024 million in 2011 to 2.35 million in 20238, indicating a consumerate steady growth in the need for water. The need for water has also increased due to the growth in mining industries, the expansion of irrigation schemes in various parts of the country, and the development of the tourism industry. These increases in water demand aggravate water stress on the existing water supply, which sourced from surface water (35%) and groundwater (65%). Currently, these sources do not meet the demand arising from various water sector needs, underscoring the need for additional water sources. The actual demand for water in Botswana is presently estimated at 272 Mm3 (million cubic meters) and is expected to increase to 340 MCM by 20359.

To offset this water scarcity and enhance sustainable water supplies, alternative solutions are necessary. Among the alternative solutions is the construction of sand dams, which allow for the harvesting and storage excess rainwater10,11,12,13,14,15,16. Such water harvesting techniques, on which this research is focused, have the potential to boost the availability of water in arid and semi-arid climatic areas. It is a method that provides additional water sources to areas with limited existing water resources, thereby meeting the diverse demands such as domestic, livestock’s and agricultural needs.

A sand dam is an inexpensive concrete or stone-masonry wall dam constructed across intermittent rivers. According to Maddrell and Neal10 it is a cost-effective water harvesting structure. Once the structure is built during the season of floods, the dam fills with water that carries sediments of different sizes. Coarse grain-size sediments and sand settle behind the dam, while fine-grained sediments are carried downstream. This process continues until the dam is filled with sand as far as the spillway. Rains and overland flows percolate through the accumulated sand and water is stored in the voids among sand grains behind the structure and in the riverbanks forming an artificial sandy aquifers. During times of water scarcity/or when additional water supply is required, local communities can obtain water using simple and inexpensive groundwater extraction techniques and use this stored subsurface water for multiple purposes. Besides creating subsurface reservoirs, the deposited sands behind the dam structure can also function as a recharge medium in areas where the underlying geological formations are porous and permeable.

All the works done before, as cited in the above literatures, focus on the social, environmental, and economic benefits and planning of sand dams, mainly from a surface water use perspective. The current study examines the potential additional uses of sand dams in groundwater recharge and the creation of artificial sand aquifer media, in addition to the common practice of sand dam construction in alluvial river beds. The groundwater recharge and the creation of artificial sand aquifer media are achieved by integrated analysis of geomorphological and geological controls of surface and groundwater movement.

This study was conducted in the Motloutse River basin in eastern Botswana with the objective of assessing the suitability of sub-watersheds in the basin for the construction of sand dams and prioritizing them for the formation of artificial sandy aquifers. Sand dams constructed at appropriate sites on tributaries of the Motloutse River will scale up the potential of stored water by either creating artificial sandy aquifers or serving as a recharge medium to the underlying formations, thereby increasing the available water resources in the basin. Besides creating artificial sandy unconfined aquifers, the sand dams also minimize the effect of evaporation, improve the quality of water stored in the pores of the accumulated sand, enhance the overall ecosystem status, and improve the welfare of the community in the area17.

Study area characteristics

Location, topography, drainage, climate and existing water supply

The Motloutse River basin is a sub-basin of the Limpopo River Basin located in the eastern part of Botswana on the northwestern side of the Limpopo River Basin and is irregularly bounded by the latitude 21º to 23º S and longitude 25º to 29º E (Fig. 1). It has a catchment area of 19, 586.6 sq km18 constituting roughly 4.7% of the total Limpopo River Basin.

Fig. 1
figure 1

Location map of the study area within the Limpopo River Basin.

Full size image

The Motloutse River basin is found in the country’s hardveld physiographic unit, which is formed by Archaean and Paleoproterozoic rocks. Geomorphologically, it has a mainly flat to undulating surface with patchy ridges and hills that developed on the Archaean rocks. The Motloutse River and its tributaries, which are ephemeral, drain the basin easterly to the Limpopo River. The Motloutse River has a total stream length of approximately 288 km. The rivers found within the basin include Motloutse (288 km long), Moenyenane (46 km long), Thune (138 km long), Makutlagage (25 km long) and Mhwana (35 km long)19. Generally, the basin is characterized by a dendritic drainage pattern.

According to Edwin20 the mean annual rainfall and potential evapotranspiration of the basin is 430 mm and 2000 mm, respectively. On the basis of the United Nations Environment Programme (UNEP)21 formula and classification scheme, the Aridity Index (AIU) of the basin is 0.215, which indicates that the climate of the basin is semi-arid. However, rainfall within the basin exhibits spatial and temporal variations. In the eastern parts, the annual average rainfall typically falls within the range of 337 to 373 mm, whereas northwards, it increases and ranges from 393 to 457 mm. The months from October to March are rainy months whereas April to September are dry months marked by the occurrence of an insignificant amount of and/or no rainfall.

Both surface water and groundwater serve as sources of water supply for the communities residing in various areas of the basin: with groundwater from four wellfields and surface water from two dams. Four wellfields are developed on the western and eastern parts of the basin, supplying water for domestic and agricultural uses to Talana Farms and Bobonong villages in the east and to Paje and Mmashoro viallges in the west. The A-Cap Energy Limited wellfield, which comprises 17 exploration boreholes, is also located in the western part of the basin22. The A-Cap Energy Limited wellfield; however, it has been developed as a potential water supply for the prospective uranium mine project in the area. The 15 boreholes in the Talana Farms wellfield were drilled on the banks of the Limpopo and Motloutse rivers supply groundwater for irrigation throughout the year to the Talana Farms, a commercial farming enterprise in Botswana located in the Tuli Block area. The Bobonong wellfield supplies groundwater to Bobonong and surrounding villages (Damochojena, Gobojango, Lentswelemoriti, Molalatau, Moletemane, Motlhabaneng, Mabolwe, Mathathane, Robelela, Sefhophe, Semolale, Tobane, Tsetsejwe, and Tshokwe). Groundwater is pumped from 31 boreholes. Among these production boreholes only 7 are found within the Bobonong village whereas the remaining boreholes are found scattered north-east and south east of Bobonong village. The Morupule Power Station, which supplies power to the country, is supplied with groundwater from the Mmashoro and Paje wellfields, where the groundwater is pumped from 35 boreholes23,24.

In the sub basin, among the different geological formations that constitute the basin, fractured and weathered crystalline rocks, the Ntane sandstone of the Karoo strata and alluvials along the Motloutse River and its major tributaries serve as aquifers20,22,23,24,25,26. The aquifer system in fractured and weathered crystalline rocks and the Ntane sandstone is both confined and unconfined, whereas in the recent sediments, which is sand river aquifer, it is entirely unconfined. According to the above authors, the borehole depths range from 23 m to 240 m with a mean depth of 178 m. The depth to the static water level ranges from 14 m to 58 m with a mean depth of 18 m. The measured borehole yields range from 1.5 m3/hr to 40.5 m3/hr with a mean of 18 m3/hr.

Two dams were constructed across the Motloutse River and its tributary, Thune River, and supply water within and outside the basin: Letsibogo and Thune dams, respectively. Letsibogo dam, constructed on the Motloutse River, has a storage capacity of 100 MCM. It was built initially to supply water to Selebi-Phikwe town and surrounding villages. Currently as part of the North-South Carrier Water Project, it is providing the capital of the country, Gaborone, linked by a 400 km pipeline26. The Thune dam was constructed near the village of Molalatau on the Thune River. It supplies several villages in the Bobirwa area.

Geology

The Motloutse River basin region is underlain by Archean rocks that are overlain by the sedimentary and volcanic rocks of the Karoo Supergroup (Table 1). The region exhibits complex geological structures associated with the deformed and metamorphosed Archean terranes of the Limpopo Belt27. Van Reenen et al.28 describe the Limpopo belt as an ENE trending granulite gneiss terrane located between the Zimbabwe and Kaapvaal Cratons. The midsections of the Motloutse River basin are underlain by the Archean rocks of the North Marginal Zone of the Limpopo Belt and the Zimbabwe Craton (Fig. 4). According to Jelsma and Dirks29 and Blenkinsop et al.30 the Zimbabwe Craton is a granite-greenstone terrane containing 26 distinct greenstone belts and associated granitoids. The Limpopo Belt includes volcano-sedimentary sequences and granitoids, which are complexly deformed and metamorphosed at granulite and amphibolite facies27.

The western and eastern parts of the Motloutse River basin are underlain by rocks of the Karoo Super Group32,33 (Fig. 4).

The Karoo Supergroup sequence in the Motloutse River basin is located within the northern Kalahari Karoo Basin, and is characterized by the Late Carboniferous to Early Permian Dukwi Formation at the bottom. The Dukwi Formation is a series of glacial and periglacial deposits that are related to the late Palaeozoic Gondwana glaciation34. This formation does not outcrop in the sub-basin. Nxumalo34 describes the Ecca Group, which is early to late Permian in age. This unit comprises substantial coal deposits associated with deltaic, continental slope, and lacustrine deposits. The upper Tlhapana Formation, the middle Mea Formation, and the lower Tswane Formation are the three formations that constitute the Ecca Group in the northern Kalahari Karoo Basin. Overlying the Ecca Group is the Beaufort Group consisting of the Tlhabala Formation, which is Middle Permian to Middle Triassic in age and is dominated by dark grey basal mudstone with minor carbonaceous partings, coaly streaks and siltstone lenses31. The Lebung Group, which is Late Triassic and fluvial and aeolian origin, overlies the Tlhabala Formation and is represented by two sandstone formations: the bottom, which is marked by the Mosolotsane Formation dominated by purplish muddy siltstones that are fine- to medium-grained, well sorted, pinkish red to purplish grey sandstones, marls, and conglomerates. The top is marked by the Ntane Sandstone Formation; consist of pinkish to red-brown, fine-grained, well-sorted, porous sandstone representing a uniform, widespread aeolian unit27,31,34. In the northeast Kalahari Karoo Basin, the Stormberg Lava Group overlies the Karoo Supergroup27,31,35. The lavas form the uppermost unit of the Karoo Supergroup, consisting of a very extensive and often very thick sequence of tholeiitic flood basalts which mark the end of the Karoo deposition. An extensive system of post-Karoo dolerite dykes and sills cut through the study area predominantly in a WNW direction (Fig. 4).

Table 1 Geology and stratigraphy of Northern Kalahari Karoo basin After27,31.
Full size table

Methodology

Geomorphological and geological methods were utilized for the assessment and prioritization of the sub-watersheds for sand dam suitability (Fig. 2). A base map of the Motloutse River basin and its tributaries was acquired from the Shuttle Radar Topographic Mission (SRTM) Digital Elevation Model (DEM), which has a resolution of 30 m (Fig. 3). Secondary data such as geology and stratigraphy of northern Kalahari Karoo Basin and geology of the Motloutse basin20,22,23,24,25,26,27,28,29,30,31,32,33,34,35 general geomorphological information’s, the different rivers found within the in the Motloutse River basin, and the Motloutse River basin as part of the Limpopo basin18,19 hydrological information of the Motloutse basin20 existing sources of water supply in the basin: available dams and wellfields, the boreholes in the wellfields, main aquifers and associated groundwater information’s20,22,23,24,25,26 were collected. Geological field work was conducted to collect ground truth data on the geology of the sub-watersheds by paying particular attention on the outcrops at the river banks and in the riverbed geology at the outlet of the sub–watersheds.

Fig. 2
figure 2

Methodological flow chart.

Full size image

Geomorphological analyses

The geomorphological method employed in this study was morphometric analysis. SRTM-DEM (Fig. 3) was used to generate drainage networks and delineate sub-watersheds using the Spatial Analyst tool of ArcGIS Pro 3.4 software (https://ubbw.maps.arcgis.com). Further SRTM-DEM data (Fig. 3) was processed and morphometric analyses of each sub-watershed were conducted using the same tool.

In the Motloutse River network, streams are assigned sequential numbers based on their position in the hierarchy using the Strahler36 stream order model. Although different methods exist for assigning stream numbers, due to its improved clarity and reduced ambiguity, and consistent ordering, this model was adopted. In this model, each fingertip channel is classified as a first-order stream, and when two streams of the same order converge, they form the next higher order. If streams of different orders meet, the resulting stream is assigned the higher order of the two (Supplementary Fig. 1, SW 6). Following this, the Motloutse River basin was then sub divided into ten sub-watersheds (SW 1 – SW 10) based on elevation, drainage flow direction and slope, allowing for the delineation of each sub-watershed. Subsequently, morphometric analyses of each sub-watershed were performed utilizing several empirical formulas (Table 2) to extract aerial, relief and linear aspects of the sub-watersheds.

Fig. 3
figure 3

Digital elevation model of the Motloutse River basin.

Full size image

Once all the morphometric values were extracted, the subsequent tasks involved determining the ranking of individual parameters within each sub-watershed. Because of indirect or direct relationship that morphometric parameters have with soil erosion, peak flow, and runoff, the extracted morphometric aspects were employed to rank the sub-watersheds in terms of their suitability for sand dam construction.

Since the relief and linear morphometric aspects are directly related to soil erosion, the sub-watershed with the highest value in linear and relief characteristics was assigned the first rank (1). The second-highest value was ranked second (2), the third-highest value was ranked third (3), and so forth. The shape or areal characteristics have an indirect relationship with soil erosion, therefore the sub-watershed that exhibited the lowest value was designated as the first rank (1), while the one with the second-lowest value was assigned the second rank (2), and the one with the third-lowest value was assigned the third rank (3), and so forth.

Table 2 Morphometric parameters used in this study, and their methods determination.
Full size table

The next step from thereon was to determine the compound parameter value for each sub-watershed after obtaining all ranks for individual parameters in each sub-watershed. To achieve this, all ranks in each sub-watershed were added independently and divided by the number of characteristics. The sub-watershed with the lowest compound value was ranked 1, followed by the sub-watershed with the next lowest value ranked 2, and so forth.

Geological analyses

The results of geological investigations are provided as geological maps produced using Arc GIS (Fig. 4). The geological maps cover the entire Motloutse River basin. A sub-watershed geological cover analysis was conducted to assess the suitability of each sub-watershed for sand dam construction. This was done by giving more weight to the outlet geology since the outlet is crucial as it is where peak flow, runoff, and eroded material from the entire catchment are captured.

Results

Morphometry analyses

Linear parameters

Linear parameters listed in Table 2 were computed for the 10 sub-watersheds, and are summarized and given in Table 3.

Stream order, stream length and stream length ratio

According to Strahler stream order model, the Motloutse River Basin contains a fifth-order stream system with a total of 31, 57, 133, 63, 101, 41, 45, 75, 55, and 73 streams across sub-watersheds 1 through 10, respectively. For each sub-watershed the stream lengths of each stream order and the total stream length of the sub-watershed were also computed (Table 3). Accordingly, the total stream length of the sub-watersheds ranges from 195.04 (SW 1) to 652.25 km (SW 5). The mean stream length ratio values of the 10 sub-watersheds in the Motloutse River basin range from 0.38 to 47.32.

Table 3 Linear parameters of the ten sub-watersheds (SW 1 to SW 10).
Full size table

Bifurcation ratio and Rho coefficient

In the Motloutse River basin, bifurcation ratios range from 2.91 to 22.50, suggesting minimal structural disturbances in some sub-watersheds, such as SW 6, 4, and 1, while others, such as SW 10, 9, and 2, show greater geological and structural disturbances as the bifurcation ratios in these watersheds is > 5, characteristic of elongated basins where geological structures significantly control the shape of the basins. The mean bifurcation ratio varies from 1.45 to 5.62 across the basin. According to Horton37 the Rho Coefficient is designed as a means of understanding the relationship between the drainage density and the physiographic development of watersheds. This is because the physiographic development affects the storage capacity of the drainage network. Therefore, the Rho coefficient can be used to understand the storage capacity of a drainage network; a value greater or equal to 0.50 indicates higher hydrologic storage during floods. In the Motloutse River basin, the Rho value of the sub-watersheds ranges from 0.13 to 10.51. In sub-watersheds 9, 8, 7, 6, 5, 4 and 1 the Rho values are less than 0.50, signifying lower hydrologic storage during floods, whereas sub-watersheds 10, 3 and 2 have Rho values > 0.50, suggesting higher hydrologic storage during floods. Rho values less than 0.50 indicate that the sub-watersheds are characterized by a higher potential for surface runoff and erosion: revealing the availability of surface runoff and sediment yield. The Rho values in the sub-watersheds 10, 3 and 2, however, indicate low potential surface runoff and sediment yield morphological characteristics of the sub-watersheds. Nevertheless, this value of Rho in these three sub-watersheds also signifies the high recharge potential of these sub-watersheds, if their geological coverages prevails percolation, which enhances the availability of groundwater in them.

Areal aspects

The areal parameters computed for each sub-watershed are summarized and given in Tables 4 and 5.

Area

The area of the 10 sub-watersheds ranges from 946.40 to 3042.72 km2. SW 1 has the smallest area whereas SW 5 has the largest area.

Stream frequency and drainage density

The computed stream frequency values for all the 10 sub-watersheds vary from 0.027 to 0.050, and denote the number of streams per area. It is influenced by the lithological units that underlie the catchment. The Motloutse River basin exhibits low stream frequency, suggesting fewer causeways and dispersed stream distribution, which indicates a lower level of hydrogeological activity. The drainage density quantifies the nearness of the spacing of channels and is an average measure of stream channel length for the entire basin43,44. It reveals the land infiltration capacity, as well as the sediment and water output from the basin area and indicates the basin’s susceptibility to erosion.

The computed drainage density of the sub-watersheds ranges from 0.17 to 0.26 (Table 4). According to Asfaw and Workineh45 a higher drainage density is associated with to fine texture with impervious underlying geological materials, scanty and sparse vegetation coverage, and high relief. These condition that lead to higher erosion potential and runoff. Conversely, low drainage density is related to coarser texture with significantly porous/permeable underlying geological materials, dense vegetation coverage, and lower relief. In this context the computed drainage density values of the sub-watersheds, which are relatively low, suggests a coarser basin texture with permeable sub-soils, better vegetative cover, and low relief, as supported by the findings of Kaur et al.46 Nag and Chakraborty47 Prasad et al.48 and Vaidya et al.49 based on studies conducted at different river basins in different states in India.

Table 4 Areal parameters of the Motloutse river sub-basin (SW 1 to SW 10).
Full size table
Table 5 Results of additional areal morphometric parameters analysis of sub-watersheds of the Motloutse river sub-basin (SW 1 to SW 10).
Full size table

Drainage texture and length of overland flow

The computed drainage texture values of the sub-watersheds range from 0.200 to 0.487, where the maximum is at SW 3 and the minimum at SW 6. This index enables the assessment of the drainage characteristics of a drainage system in terms of carrying capacity. The sub-watersheds of the Motloutse River basin can be characterised as very coarse to coarse.

According to Sukristiyanti et al.50 three classes of length of overland flow are recognized: low value (< 0.2), moderate value (0.2–0.3), and high value (> 0.3). Those authors further state that topographical feature such as high relief and hydrological features such as less infiltration, short flow paths and high runoff that make prone to the flash flooding resulted low values. In contrast topographical feature such as gentle slope topography and opposite hydrological features of the lower values resulted high values. The computed length of overland flow values for all the sub-watersheds ranges from 1.947 to 2.897. This high values revealed that the sub-watersheds in the Motloutse River basin are characterized by long flow paths and area of low relief with gentle slopes.

Drainage intensity, infiltration number and circularity ratio

The calculated drainage intensity value for the ten sub-watersheds varies from 0.439 to 2.628. As shown in Table 4, SW 6 has the lowest infiltration number, and SW 8 has the highest infiltration number. Circularity ratio helps to indicate the shape of the catchment, and its values range from 0 to 1.039, with higher values representing a more circular basin shape, while lower values indicate a more elongated basin shape. The circulatory ratio analysis of the 10 sub-watersheds in the Motloutse River basin shows that they have low circularity ratios (Table 5) that range from 0.331 to 0.575. As highlighted by Ali et al.51 Aparna et al.52 and Soni53 the circularity ratio equals 1.0 when the basin is a perfect circle, and it falls between 0.4 and 0.5 for strongly elongated basins that are highly permeable and consisting of homogeneous geological materials. In line with this assertion, the computed circularity ratio values for the 10 sub-watersheds in the Motloutse River basin suggest they are significantly elongated.

Texture ratio, basin length and form factor

The computed texture ratio for all 10 sub-watersheds varies 0.097 to 0.194 and the computed basin length of the ten sub-watersheds ranges from 64.319 km to 124.861 km (Table 5). These computed texture ratio values revealed that the sub-watersheds in the Motloutse River basin are characterized low rainfall and area of low relief.

The form factor also provides insight into basin shapes and their related hydrological characteristics. It indicates the flow intensity within a basin. For a finely circular basin, the value of the form factor must be above 0.7844,54. High peak flows over a shorter duration are the hydrological characteristics of such basins. According to Chopra et al.55 a lower form factor value (< 0.45) indicates a more elongated basin. According to Farhan and Anaba56 hydrologically such a basin is characterized by lower peak flows for a longer duration. The computed form factor for the whole sub-watersheds is less than 0.45, with values ranging between 0.20 and 0.21. These low values suggest that the sub-watersheds are indeed elongated. This finding is consistent with the elongation ratio results described below, supporting the conclusion that the sub-watersheds have an elongated shape, leading to lower peak flows but with longer durations.

Elongation ratio

The elongation ratio provides insight into the shape of the river basin and is influenced by climatic and geological conditions45. On a wide range of geological structures and climatic conditions, the elongation ratio values mostly vary from 0.6 to 1.044. High values of elongation ratio indicate regions that have a circular shape and of very low relief, and are hydrologically characterized by minimum runoff and high infiltration capacity44,57,58. Mountainous and rugged areas with steep slopes usually have lower elongation ratio values, typifying more elongated shapes that are hydrologically characterized by high runoff and low infiltration capacity56. According to Asfaw and Workineh45 three categories are recognized on the basis of elongation ratio values: elongated (< 0.7), oval (0.9–0.8), and circular (> 0.9). In the Motloutse River basin the elongation ratio of the sub-watersheds ranges from 0.499 to 0.540, making the entire sub-watersheds elongated, with SW5 being the most elongated and SW1 the least elongated.

Shape index, lemniscate ratio, compactness coefficient and constant channel maintenance

The values of shape index for the ten sub-watersheds range from 4.371 to 5.124. According to Shekar and Mathew41 the lemniscate ratio can be used to compute the slope of the catchments. The calculated value of this ratio in the ten sub-watersheds varies from 1.093 to 1.281. The compactness coefficient, according to Horton37 is the amount of the perimeter of the catchment to the circumference of an equivalent circular area. For the ten sub-watersheds, the highest calculated compactness coefficient is 1.739 in SW 5, and the lowest one is 1.319 in SW 9. Constant channel maintenance is an inverse of drainage density38,53. The computed constant channel maintenance values for the ten sub-watersheds (Table 5) revealed that the maximum is 5.793 in SW 2, and the minimum is 3.893 in SW 7.

Relief parameters

Table 6 presents four relief parameters calculated for each sub-watershed in the Motloutse River basin: relief, relative relief, ruggedness number, and relief ratio. The maximum and minimum elevations were obtained through ArcGIS analysis using the Digital Elevation Model (DEM).

The Basin relief serves as a landform denudation indicator, reflecting the existence of active processes such as erosion, weathering, and mass wasting59. As shown in Table 6, relief ranges from 201 m in SW 1 to 408 m in SW 4. The relief ratio varies from 2.36 in SW 2 to 4.39 in SW 7. The central part of the basin shows a higher relief ratio, indicating steeper slopes, while the northwest and northeast sides of the basin exhibit lower relief ratios, representing areas with gentler slopes. The relative relief value ranges from 361.30 (SW 10) to 761.32 (SW 1). These extremely high relief values in all the sub-watersheds indicate the highest erosion potential in the Motloutse River basin.

The Motloutse River basin sub-watersheds relief was characterized using the ruggedness number, which is a measure of surface unevenness and its susceptibility to soil erosion60. According to Gutema et al.61 the ruggedness number reflects the structural complexity of the terrain. The computed ruggedness number values for the ten sub-watersheds are notably high, ranging from 20.57 in SW 2 to 42.37 in SW 4, indicating a rugged topography that is prone to soil erosion and structural instability. Adhikari62 reported a high value of ruggedness number, which is 20.5, in his study conducted in the Ghatganga River in Nepalhe river basin.

Table 6 Relief parameters of the ten sub-watersheds in Motloutse river sub-basin (SW 1 to SW 10).
Full size table

Slope, like the other relief parameters, is an important parameter that determines the potential for soil erosion and defines the steepness of the topography of an area. The general slope distribution for each sub-watershed of the Motloutse River basin and the values of ground slope for each sub-watershed are shown in supplementary Fig. 2. The degree of slope in all the sub-watersheds varies from 0–8.2º to 0–32.2º. Most of the area in all the sub-watersheds is characterized by very gentle slopes (< 5º), but few step mountains are associated with the sub-watersheds and increase the range. Table 7 presents the percentage of the area in each sub-watershed that falls within the slope range of 1–2. According to Practical Action63 because of the maximum water storage yields, drainage basins characterized by slopes between 1 and 2% are ideal sites for the location of sand dams. The percentage of these areas in the sub-watersheds of the Motloutse River basin varies from 9 to 53%, with SW 3 having the largest area covered by slopes of 1–2, and SW 5 having the least percentage.

Table 7 Slope distributions of the sub-watersheds from 0–1° and 1–2°.
Full size table

Discussions

Ranking of sub-watersheds

Morphometric – compound parameter

The morphometric ranking was calculated using the 20 parameters listed in Table 8, with the ranking results summarized in Table 9. The sub-watersheds were ranked from 1 to 9 based on their compound parameter values: lowest ranks were given to sub-watersheds with the highest compound value and vice versa. Accordingly, therefore sub-watershed 7, with a compound value of 4.20, is ranked one and sub-watershed 3, with a 6.50 compound value, which is the highest, was given the lowest rank, 9. This analysis revealed the ranking of all the sub-watersheds from one to nine (SW7 ranking 1 whereas SW 3 ranking 9) in terms of their suitability for sand dams with respect to their hydrologic characteristics such as runoff, peak flow and sediment yield/soil erosion. The sub-watershed with the highest ranking is the most vulnerable to higher peak flow, runoff and soil erosion, which are important for the sediment yield of the watershed for the formation of artificial sandy aquifers after the dam structures. However, this morphometry analysis does not rule out any of these sub-watersheds for sand dam construction.

Table 8 Twenty parameters and ranking for different sub-watersheds.
Full size table
Table 9 Calculation of compound parameters, prioritization, and ranking of sub-watersheds.
Full size table

Sub-watershed geology cover analysis

Since sand dams are usually built across the riverbeds, affixed to the bedrock, analyses of the sub-watershed geology cover are indispensable, in addition to the morphometric analyses of the sub-watersheds. Intact massive bedrock is required to ensure the storage of water within the sand dam to totally avoid or minimize losses through leakage from the bottom. Additionally, the geology of the watershed also determines, along with slope and discharge characteristics, the grain sizes of the sediments that will be deposited in the sand dam. According to Practical Action63 riverbeds with volcanic rocks have lower yields compared to riverbeds with crystalline (basement) rocks. Furthermore, to reduce the cost of construction of this structure, bedrock that outcrops to the surface is preferred. On this basis, a sub-watershed geological cover analysis was conducted by giving more weight to the outlet geology since the outlet is crucial as it is where peak flow, runoff, and eroded material from the entire catchment can be best captured, allowing for the assessment of the suitability of each sub-watershed for the sand dam construction.

The ranking was done based on the following:

  • Entire sub-watersheds geological coverage analyses on the basis of water holding and transmission capacity;

  • Sediment yields (coarse sand and gravel) offering nature of the entire sub-watersheds geology; and,

  • Sub-watersheds outlet geology and its degree of exposure to the surface (exposed bedrock that outcrops to the surface is preferred to reduce the cost of construction).

The geology of each sub-watershed was analyzed by overlaying the geological map with the same sub-watersheds used for morphometric analysis (labelled SW 1–10) (Fig. 4), and their areal coverage was also quantified (Table 10). Following this, a field survey was conducted to collect ground truth data on the sub-watershed geology (supplementary Fig. 3 and Fig. 4).

Fig. 4
figure 4

Geology map with ten sub-watersheds of the Motloutse sub-basin.

Full size image
Table 10 Ranking of sub-watersheds - based on sub-watershed geological cover analyses, their water holding and transmission capacity, and sediment yield (coarse sand and gravel) offering nature.
Full size table

Applying the results of the geology cover analyses and field data, the sub-watersheds were ranked based on their suitability for the siting of sand dams.

Accordingly, the first rank (1) is given to the sub-watersheds that have entire watershed coverage of basement rocks (SW 7 and SW 9), and outlet geology of basement rocks and the remaining sub-watershed areas are covered by both basement and different formations of Karoo Supergroup rocks in different proportions (SW 2 and SW 4).

The second rank (2) is given to the sub-watersheds that have entire watershed coverage of different formations of Karoo Supergroup rocks with the outlet geological coverage of the volcanic rocks (SW 8). This rank is also given to those sub-watersheds that have outlet geology of volcanic rocks of the Karoo Supergroup and the remaining sub-watershed areas are covered by both basement and different formations of Karoo Supergroup rocks in different proportion (SW 3 and SW 6).

The third rank is assigned to the sub-watersheds that have entire watershed coverage of different formations of Karoo Supergroup with the outlet geological coverage of the clastic sedimentary rocks of the Karoo Supergroup (SW 5). This rank is also assigned to those sub-watersheds that have outlet geological coverage of the clastic sedimentary rocks of the Karoo Supergroup and the remaining sub-watershed areas are covered by both basement and different formations of Karoo Supergroup rocks in different proportion (SW 1 and SW 10). Accordingly, on this basis, SW 2, SW 4, SW 7 and SW 9 fall in the first rank, SW 3, SW 6 and SW 8 in the second rank and SW 1, SW 5 and SW 10 in the third rank (Table 10).

Sub-watershed geological cover analysis revealed that sand dams are feasible in most of the sub-watersheds for creating artificial sandy unconfined aquifers in the respective reservoir pool areas, except rank 3 sub-watersheds. In rank 3 sub-watersheds, the geological cover of the outlets comprised of clastic sedimentary rocks that are not intact enough but exhibit significant percolation that enhances the groundwater availability in the underlying aquifers.

Furthermore, among SW 1, SW 5, and SW 10, the availability of water for recharge is more in SW 1 and SW 5 than in SW 10. Besides, due to the availability of more sediment yield in these two sub-watersheds, the artificial sandy aquifers created in these sub-watersheds as a result of sand dam construction can serve as an additional thickness to the existing thickness of the underlying aquifers. The artificial sandy aquifers created in these two sub-watersheds play a role as storage medium besides their role as a recharge medium. All these enhance the availability of groundwater in these two sub-watersheds: groundwater resource availability will be more in the SW 1 and SW 5 than in the SW 10 if sand dam construction is going to be done for the purpose of groundwater recharge.

The integration of sub-watershed geological cover and morphometric analysis revealed that SW 7 falls within the same ranking and the remaining nine sub-watersheds, however, show no correlation and differ in their ranking under sub-watershed geological cover and morphometric analyses (Table 11). Albeit these differences in ranking, the combined morphometric and geological findings highlighted SW 7 as the most suitable sub-watershed for sand dam construction and subsequent formation of artificial sandy aquifer, followed by SW 2, SW 3, SW 4, SW 6, SW 8 and SW 9. Specific site suitability within these sub-watersheds varied based on geological coverage. For example sub-watersheds SW 3, SW 6, SW 7, and SW 9, underlain predominantly by crystalline and volcanic formations, offer multiple viable sites for sand dam deployment and consequent formation of artificial sandy aquifers. In SW 2 and SW 4, sand dams are best suited for the middle and lower sections, respectively, whereas in SW 8, the northeastern part is deemed unsuitable, while other sections are also appropriate.

Table 11 Summary of suitable sub-watersheds in the Motloutse river sub-basin using morphometric and sub-watershed geological cover analyses.
Full size table

Generally, the findings highlight the potential application of an integrated analysis of geomorphological and geological controls of surface and groundwater movement for understanding the dynamics and use of sand dams in groundwater recharge and creation of artificial sandy aquifer media. It is a further contribution beyond the current available knowledge and practice of sand dam construction in alluvial river beds that focus on the social, environmental, and economic benefits and planning of sand dams, mainly from a surface water use perspective as highlighted in various literature10,11,12,13,14,15,16.

Conclusions

The morphometric analysis of the Motloutse River basin using linear, areal and relief aspects and sub-watershed geological coverage analyses revealed that among the analyzed ten sub-watersheds, seven of them are suitable for sand dam construction and formation of artificial sandy aquifers. Among these, SW7 is the most suitable one followed by SW 2, SW 3, SW 4, SW 6, SW 8 and SW 9. All these seven sub-watersheds offer multiple feasible sites for sand dams and subsequent artificial sandy aquifers that are deposited overlaying the bed rocks. These formations of artificial sandy aquifers at multiple sites will boost the groundwater availability in the basin and also enhance the overall ecosystem status.

Sub-watersheds SW 1, SW 5, and SW 10, which are found to be underlain by highly porous and fractured clastic sedimentary rocks, are best suited for groundwater recharge in the reservoir pool area, and they are found to be recommendable for groundwater recharge that promotes percolation into the local groundwater system. However, because of the availability of more water for recharge and sediment yield in SW 1 and SW 5 than in SW 10, the availability of groundwater resources will be greater in these two sub-watersheds if sand dam construction is going to be done for the purpose of groundwater recharge.

During the implementation of the sand dam structure in those sub-watersheds, detailed specifications of the structure to the specific site should be made separately by considering other important parameters (peak flow analyses, stream dimension measurements, contouring, etc.) that are vital to the design of structure where the structure is constructed.