Introduction

Land use has generally been considered as a local environmental issue, but it is becoming a force of global importance (Foley et al. 2005). Rates of deforestation, agriculturalization, urbanisation, wetlands drainage, and several other types of land use changes have accelerated as a function of the growth of human populations and economic development (Eshleman 2004).

Land use and land cover (LULC) changes strongly affect catchment hydrology. Additionally, studies have also demonstrated that the environmental effects of LULC changes have potentially large impacts on hydrologic processes such as sediment load and nutrient concentration, evapotranspiration, groundwater recharge, base flow, and runoff (Aghsaei et al. 2020; De Oliveira Serrão et al. 2022).

The factors that influence surface runoff in a basin are many and varied and can be classified into two general groups: natural characteristics and anthropic interventions. The first group involves shape and extension of the basin, geology, morphology, slope, depth of the water table, vegetation cover, and previous humidity of soil (Davis and De Wiest 1971). The second group of factors involves streets, hydraulic works, urbanisation, and landscape alteration for agriculture, among others. Urbanisation and other anthropic interventions tend to increase surface runoff by creating more impermeable surfaces. The opening of streets as well as the changes in land use to less dense vegetation cover change the dynamics of runoff and increase their volume (Shuster et al. 2005). According to Foley et al. (2005), the surface runoff and river discharge generally increase when natural vegetation (especially forest) is cleared. The lack of planning regarding the use and occupation of basins increases the problems of water availability and demand, as well as its quality (Barbosa et al. 2019). Therefore, an assessment of the impacts of anthropogenic LULC alteration on hydrology is crucial for sustainable river basin management (Woldesenbet et al. 2017; Aghsaei et al. 2020).

The study of the morphometric parameters of a basin, namely shape, relief and drainage, allows for a correct and broader interpretation of the general functioning of the hydrological system. In addition, knowledge of runoff processes is essential to respond to environmental problems such as water erosion and floods. Both issues are key elements when considering the planning of land use and an integrated management of water resources. In this respect, one of the most widely used methodologies to study changes in hydrological processes caused by human activities is the Curve Number method (CN) (Guo et al. 2019).

The study area is hilly and consists of the upper zone of a drainage basin. In order to prevent flooding in nearby areas, resulting from the rapid and abundant concentration of rainfall, a dam was built at the mouth of Lago del Fuerte basin (Ruiz de Galarreta et al. 2010). Over time, the area has become a site of commercial, urban, and tourist interest, thus turning into one of the main expansion areas of Tandil city.

Lago del Fuerte is one of the main tourist attractions of the region. In recent years, the area near the lake has experienced a significant population increase, mainly due to the growth of residential land use and the settlement of tourist-recreational enterprises (Rodríguez 2014). However, this growth has not been accompanied by water and sewer services, except in one small sector of the basin, nor by a comprehensive territorial planning, generating impacts on the dynamics of the basin and associated environmental problems (Rodríguez et al. 2013; Guerrero et al. 2015; Cifuentes et al. 2022a).

As mentioned above, urban growth and its associated interventions, such as changes in land use and cover, tend to increase surface runoff by creating more impermeable surfaces. This situation can alter the dynamics of the water resources of the lake’s contribution basin and of the reservoir itself. Therefore, without proper planning and management of LULC, the reservoir’s responsiveness could be compromised, further exacerbated by the accumulation of sediments (Rodríguez et al. 2008; Cifuentes 2020; Cifuentes et al. 2022b).

Several researches about the study area have been carried out, especially on the lake basin, which described and characterised the hydrogeology of the area (CINEA 2009) as well as the changes in land use and urban expansion and their associated environmental impacts (Ulberich 2007; Miranda del Fresno and Ulberich 2010, 2011; Rodríguez 2014). Creparula (2006) focused on the relationships between land uses and diatom algae communities in one of the lake sub-basins while Bogetti (2015) carried out an environmental diagnosis of the dynamics and quality of water in the two contributing streams of the reservoir. La Macchia (2014) conducted a modelling of urban drainage in the city of Tandil which includes the area of study as the head basin. Other studies on Lago del Fuerte reservoir focused on water quality and eutrophic level (IHLLA 1995; 2005a, 2005b; Albornoz et al. 2009; Cifuentes 2012; 2020), bathymetry and sediment volume (Rodríguez et al. 2008), nutrient balance, especially phosphorus, and phytoplankton (Cifuentes 2020) and ichthyological characters (Grosman 1999; Berasain and Padín 2014; Bertora et al. 2016).

Although there are studies that have applied GIS tools for the analysis of changes in land use and land cover and their implications for surface runoff in the region (Gaspari and Bruno 2003; Vagaría et al. 2012; Delgado et al. 2013, 2020; Senisterra et al. 2015), this has not been developed in Lago del Fuerte basin. Furthermore, there are some studies by the working group in the area (Rodriguez 2014; Cifuentes 2020; Cifuentes et al. 2022a, b) which constitute partial advances because they have not jointly analysed the impact of territorial changes on surface runoff and the reservoir.

The objective of this study was to perform a spatial and temporal analysis of the changes in land use and land cover and their incidence on surface runoff in Lago del Fuerte basin (Tandil, Argentina). This work represents a contribution to environmental territory planning, and is useful for decision makers towards the integrated water resources management in a basin.

Study area

Lago del Fuerte basin (Fig. 1) is located on the North face of the Tandilia hilly system, at the South of Tandil city, in the South-East of Buenos Aires province (Argentina). It covers an area of 2,000 hectares and has a defined and integrated surface drainage. The direction of runoff is towards the North-East, according to the region slope (Ruiz de Galarreta et al. 2010).

The Tandilia system includes two geological units: the crystalline basement consisting of plutonic igneous rocks, and the sedimentary cover which includes Pampean sediments of the loessoid silt type (Teruggi and Kilmurray 1980).

Fig. 1
figure 1

Location of Lago del Fuerte basin and the sub-basins San Gabriel and La Cascada

Full size image

Regarding the regional geomorphology, three morphological units are distinguished within the Tandilia system: mountains, piedmont, and plains (Ruiz de Galarreta 2006). The mountainous sector is characterised by pronounced relief, isolated hills, and valleys with the occurrence of faulted blocks. Watersheds are recognizable, and the valleys of the streams are well-defined. The study area is partially located in this geomorphological region, where the crystalline basement is shallow, and significant outcrops are observed in certain areas. The hills are followed by the piedmont with a gentler slope and the presence of large blocks and alluvial cones. The flat zone has low slopes and extends towards the North-East. The elevation of the study ranged from 197 to 486 m above sea level (Fig. 2).

Fig. 2
figure 2

Digital Elevation Model (DEM) of the study area

Full size image

According to the geology and geomorphology of the region, the soils can be classified in three subregions. The first, corresponding to mountainous morphology, is characterised by rocky outcrops, surface stoniness, and areas with a thin layer of loess, where Lithic Hapludolls and Petrocalcic Hapludolls predominate. The piedmont subregion, which occupies the largest area of the watershed, includes hilly areas and small plain sectors, primarily represented by Typical Argiudolls. The third zone corresponds to the plain with the presence of Tapto Natric Hapludolls and Typical Natracuols (INTA 1989).

The region presents a typical subhumid-humid, mesothermal climate, with a low water deficit in summer (Ruiz de Galarreta et al. 2010). Tandil city had an annual average rainfall of 849 mm for the period 1900–2016. The precipitation is distributed as follows: 29.1% in summer, 27.8% in autumn, 26.7% in spring, and 16.4% in winter for the period under consideration (Fig. 3). The month with the lowest average monthly precipitation is July (< 43 mm), while March experiences the highest precipitation levels (98 mm) (Cifuentes 2020).

Fig. 3
figure 3

Average monthly precipitation and temperatures of Tandil city (1900–2016) (Cifuentes 2020)

Full size image

Lago del Fuerte basin is composed of two sub-basins of San Gabriel (SG) and La Cascada (LC) streams (Fig. 1). Both streams flow into the Del Fuerte lake, which constitutes an artificial reservoir of 19 hectares originated by the construction of a dam aimed to regulate the floods that affected the city. The dam, called Dique del Fuerte, was built between 1957 and 1962 (Ruiz de Galarreta et al. 2010). From the exit of the dam, the water flows through Del Fuerte stream which crosses the city piped (Fig. 1).

The reservoir has two direct runoff zones on both East and West banks. Thus, surface water and groundwater drain directly into the lake instead of contributing to the streams (Fig. 1).

Materials and methods

Morphometric parameters

Calculations of the morphological features of Lago del Fuerte basin (shape, relief and drainage parameters) were developed, focusing on their influence on the surface runoff (Table 1). The Arc Hydro Tools extension for ArcGis (v 10.5) was used together with cartographic support for those calculations. This software has specialised tools for territorial, topological, and hydrological analysis.

Digital Elevation Models (DEM) provided by the Geological Service of the United States (USGS 2014), type SRTM 1 Arc-seg (spatial resolution of 30 m), were used, which were compared with the Topographic Chart Nº 3760-29-2 Sierras del Tandil, scale 1: 50.000 (IGN 1957). Direct observation was carried out in the field in order to identify and solve inconsistencies between the DEM and the chart detected during the analysis of the sub-basins. The layers were georeferenced by a reference system based on the IGN Posiciones Geodésicas Argentinas network (POSGAR ‘07 zone 5), which uses a metric system.

Table 1 Morphometric parameters for Lago del Fuerte basin.
Full size table

Curve number (CN) calculation

The Soil Conservation Service Curve Number method (CN; USDA, 1972) was used to estimate the runoff. CN is an empirical method useful for computing the transformation from rain into runoff according to the following properties: hydrologic soil type, use and soil treatment, surface condition of the soil, and moisture condition antecedent (Ponce 1989). CN ranges from 1 to 100, where higher values of CN refers to greater levels of runoff.

The CN method has been developed for more than 50 years and has been applied as well as reviewed in numerous research (Ponce and Hawkins 1996). According to Hawkins et al. (2009), the CN method has undergone a series of unavoidable adjustments, improvements, redefinitions and clarifications over the years. High positive correlations (p ≤ 0.05) between the runoff estimated by the CN methodology and the observed values have validated the use of this methodology (Nonglait and Tiwari 2016).

CN method was applied for the years 2003 and 2020 on Lago del Fuerte basin to analyse and compare the surface runoff generated by a reference precipitation, according to the modifications of the land cover and land uses in the basin under study. The period was chosen for the availability of satellite images needed to identify the land use and land cover and also for the LULC changes evidenced in previous studies.

In order to calculate the CN, a HEC-GeoHMS extension for AcrGis was applied (Fleming and Doan 2003). Georeferenced vector shapes were carried out with the surveyed territorial information. The information layers developed were Hydrological Group (HG), land use and land cover (LULC), and digital elevation model (DEM) (Fig. 2):

  1. a)

    Hydrological Group (HG). The CN method requires classifying soil types according to their permeability or infiltration capacity, in order to indicate their capacity for surface runoff. Thus, four hydrological soil groups were established (USDA 1972):

    Group A: Soils with low potential for surface runoff. Water is easily transmitted in the edaphic profile. They are sandy or sandy-loamy soils.

    Group B: Soils with a moderately low potential for runoff. Infiltration has no impediments. They are medium-depth sandy loam soils or deep loamy soils.

    Group C: Soils with moderately high runoff. The permeability of the profile presents limitations or impediments. The texture has clay loam characteristics or with some impervious soil horizon.

    Group D: Impervious soils with high surface runoff. The typical texture of this group is clay type, or with lithological characteristics that totally or partially restrict water infiltration in the event of events.

    The edaphological characterization was carried out considering the Instituto Nacional de Tecnología Agropecuaria (INTA) Soil Chart Nº 3760-29-2 (Sierras del Tandil) and the elaboration of the textural triangle from the edaphic profile (Villaroya et al. 2013). The rocky outcrop in mountains was considered as areas with great restrictions to the infiltration of the water resource. Each type of soil was assigned to a HG according to its characteristics.

  2. b)

    Land use and land cover (LULC). This research considered CN values as used by different authors who adapted them to land use and vegetation cover (Ponce 1989; Chow et al. 1994; Gaspari et al. 2013). Thus, a land use and land cover mapping was carried out using satellite images provided by Google Earth for the periods of May 2003 and June 2020. This process was complemented by field visits, photographic documentation, and georeferenced surveys, providing greater specificity and precision to the LULC maps. The minimum mapping area was 0.1 ha. Based on the data collected, 13 types of land use and land cover were identified. A CN value was assigned to each LULC according to their corresponding HG. Table 2 presents the description of each category.

  3. c)

    Digital elevation model (DEM). The DEM was used to calculate morphometric parameters (Table 1). The layer provides elevation and slope information with a spatial resolution of 30 m.

The CN calculation was carried out by applying the ‘Generate CN Grid’ tool of the HEC-GeoHMS extension for ArcGis (Fleming and Doan 2003). A vector layer of polygons was created that ‘intersected’ (such as the GIS tool name) the surveyed territorial information with the HG shape. Each polygon of the new layer had a unique combination of land use and land cover and hydrological group. Slope percentages were calculated with the DEM, and related to the previously created vector layer.

Fig. 4
figure 4

Methodological steps for CN calculation

Full size image

The ‘Generate CN Grid’ tool assigned a previously established CN value for each polygon (Table 2), according to Ponce (1989), Chow et al. (1994) and Gaspari et al. (2013). As a result, a new column with CN data is added to the attribute table of the polygon layer and a raster layer was generated with CN information in each cell.

The differences between the CN values for the study years were estimated with the ArcGIS raster calculator. The methodological steps for CN calculation are shown in Fig. 4.

Table 2 CN associated with each land use land cover and hydrological group.
Full size table

The average value of CN for the basin was obtained from the following equation (Gaspari et al. 2013):

$$CN=\frac{1}{A}{\sum }_{i=1}^{n}{A}_{i}{CN}_{i}$$
(1)

where CN is the basin curve number, A is the basin area (m2), Ai is the area of each land use and land cover polygon (m2), CNi is the curve number for each land use and land cover polygon, and n is the number of polygons calculated.

A descriptive statistical analysis of the rasters generated by the geoprocessing analysis was carried out in order to characterise the mean value of CN calculated by sub-basin and the standard deviation. The ‘Band Collection Statistics’ tool, from the ArcGis spatial analysis tools, was used.

In order to represent the significance of the CN values obtained, the expected runoff calculation was performed in typical rainfall events of the region. The variation in the period 2003–2020 and the significance of the changes in land use land cover was analysed.

Typical events were selected from the regional study of Collazos and Cazenave (2015) who elaborated intensity, duration, and frequency curves (IDF) with correction coefficients for the center of Buenos Aires province, where the study area is located. Based on that work, it was considered rainfall lasting 60 min with return periods (T) of 5 and 25 years, which resulted in events of 49 mm and 63.6 mm respectively.

Equations 2 and 3 were applied to estimate surface runoff for both events (USDA 1986).

$$S=254\left(\frac{100}{CN}-1\right)$$
(2)
$$Q\left( {mm} \right) = \frac{{{{\left( {P - 0.2S} \right)}^2}}}{{P + 0.8S}}$$
(3)

where CN is the basin curve number, S is the maximum possible retention, Q is the runoff flow (mm) and P is the precipitation (mm).

Results and discussion

Basin morphometric parameters

The Lago del Fuerte basin, with an area of 19.94 km2, corresponds to the classification of a micro-basin (Campos Aranda 1992). In relation to the basin shape, the compactness coefficient (Kc = 1.51) indicates it is oblong (Gaspari et al. 2013). However, according to the shape index (If = 0.77) and the elongation ratio (Re = 0.99) the basin resulted in a rounded shape (Table 3). The elevation of the basin ranged between 197 and 486 m above sea level, with an average slope of 5.41%. The basin presented a smooth relief according to the classification by Ortiz Vera (2004).

In terms of drainage parameters, the hypsometric curve indicated that the basin presents an advanced stage of hydrologic evolution. The basin reached a maximum stream order of 4 following the ranking method by Strahler (1964). The density of the drainage (Dd = 1.24 km.km− 2) was between low and moderate (Fuentes Junco 2004), with a concentration time (Ct) of 45 min and a torrential coefficient (Tc) of 0.85 (Table 3).

In the field, modifications of the drainage networks as a result of anthropic actions were identified. The increasing expansion of residential land use and the subsequent opening of streets has caused alterations in the drainage of some transitory tributaries. The streams have been modified with respect to their original channel, thus increasing the values of stream order number and drainage density.

Table 3 Lago del Fuerte basin morphometric parameters
Full size table

According to shape parameters, the basin presented an intermediate dangerous situation in case of high intensity rainfall events. In addition, the high number of first-order streams in the basin (17) generated a high torrential coefficient (Tc = 0.85), which implies a high susceptibility to erosion, a shorter time to reach the peak of flood and a greater generation of torrential flows (Busnelli and Horta 2014). These conditions motivated the construction of the dam in 1962.

Hydrological groups (HG) of the basin

Three hydrological groups (HG) of soils were identified in the basin (Fig. 5): Group B (71 ha.), Group C (1454 ha.), and Group D (469 ha.). HG C had the largest area covering 73% of the basin. Four main soil series were identified in this group: Sierra de los Padres (SP6), Tandil (Ta19), Tandil (Ta7), and Tres Esquinas (TEs3). These series present, respectively, Lithic Hapludol, typical Argiudol, typical Argiudol, and vertic Argiudol as main soil taxonomy (INTA 1989).

The second largest HG was D covering 24% of the area. It is entirely composed of the Rock (R) series. HG B was the smallest group with 4%. It is formed by La Alianza (La13) soil series which present a superficial petrocalcic Paleudol as its main taxonomy (INTA 1989).

In SG predominated by far the HG C (94%) followed by the HG D (6%) and there was no presence of HG B. In the LC basin the three HG were found with 7% for B, 58% for C and 35% for D.

Fig. 5
figure 5

Hydrological Groups in Lago del Fuerte basin

Full size image

Changes in land uses and land covers

Table 4; Fig. 6 show the thirteen identified land uses and land cover (LULC) areas and the variations for the study period.

Table 4 Land uses and land cover areas for 2003 and 2020
Full size table
Fig. 6
figure 6

Land uses and land covers in Lago del Fuerte basin for 2003 and 2020

Full size image

Figure 7 shows the covered area (%) for each use and land cover identified with respect to the total area of the basin for the years 2003 and 2020 and their temporal variation. Approximately 30% of the study area had rocky outcrops which remained unchanged during the study period. Brush-forbs-grass was the most predominant coverage in 2003 (30%), followed by grassland (12%), and open spaces (8%). However, grassland and open spaces were the most affected LULC; their areas were reduced by 2020 because of the partial replacement by growing high and low density residential uses, with 7% and 2%, respectively. The crop land use was in general low, covering around 2% for 2003 with a reduction of the surface by 2020.

Fig. 7
figure 7

Occupation and temporal variations of the land use and cover in the study area for the years 2003 and 2020

Full size image

These results are consistent with other works carried out on the basin. Ulberich (2007) analysed the changes in land use in the period 1973–2006 in a region which includes part of the study area and evidenced a growth of the residential land use, an increase in the concentration of buildings, and a replacement of the crops by buildings for residential and recreational use.

Rodríguez (2014) analysed the changes in land use within the LC sub-basin and found that, from 2003 to 2013, the variations were characterised especially by the increase in residential, tourist-commercial, livestock, and not-defined uses, while farming use presented the highest decrease.

Curve number (CN) identification and analysis

Curve Number (CN) maps showed values from 48 to 98. These results were grouped into categories and showed, according to their colour, areas with similar runoff conditions (Fig. 8).

The average CN estimated for the Lago del Fuerte basin was 79 for the year 2003 and 79.2 for 2020. This estimation showed that, although there was no perceptible variation in the mean value of the CN, the general conditions of the basin tends to favour runoff processes over infiltration.

Fig. 8
figure 8

CN maps for Lago del Fuerte basin in 2003 and 2020

Full size image

Although the temporal variation between the average CN values in each sub-basin was not considerable, the analysis of each direct runoff zone (Fig. 1) showed great diversity of results. Thus, while for the direct runoff zone of the LC sub-basin the CN ranged from 79.6 to 80.6, in the direct runoff zone of the SG sub-basin it varied from 76.5 for 2003 to 78.9 for 2020 (Table 5).

Figure 9 shows the difference of CN values in the study period. Some areas of the basin had lower CN values in the year 2020, while other areas showed higher values (Table 2). The decrease in CN values was generally associated with stopping a use or an activity such as crops and with the renaturalization of an area. Thus, there is a successional growth of the natural vegetation species which would allow greater infiltration. However, there was also a decrease in the CN when moving from fallow to low-density residential since the former is associated with bare soil, while in the second, an open area is always allocated with some kind of cover. Furthermore, the increase of the CN values occurred in areas where residential use was intensified, but especially when an area was transformed from scrub vegetation or grassland to urban use.

While, in general, changes in land use and land cover tend to increase CN values, there are studies in which decreases are recorded, as occurred in the present research. Havrylenko et al. (2006) identified a seasonal decrease in CN values depending on land use and land cover in the Arrecifes River basin (Buenos Aires, Argentina). The highest CN values occurred in spring where 49% of the surface had bare soils while in summer with 92% of the surface covered by crops the CN values were lower. In addition, De Antueno et al. (2020) and Bottan et al. (2023) showed a decrease in CN values due to changes in land use that applied conservation improvements in agricultural activities, thus favouring infiltration in the upper basin of the Sauce Chico River (Buenos Aires, Argentina).

The analysis of CN values shows averages with high standard deviation values, which indicates a significant amplitude in the CN values throughout the sub-basins. These results are consistent with Gallegos Reina and Perles Roselló (2019) who worked in Malaga province, Spain. They affirmed that, although the values of the runoff threshold remained practically the same during a study period of 50 years, these moderate differences were the result of a regional compensation and masked the reality of intense changes that took place in certain areas where the differences are higher.

Fig. 9
figure 9

CN variation in Lago del Fuerte basin between 2003 and 2020

Full size image

Main areas with notable variations in CN values were identified. In SG sub-basin these areas are coincident with a residential area that changed from low to high density. While in LC sub-basin the changes primarily occured along the route of a main street, where both residential and tourist-recreational buildings have been constructed.

However, the differences are compensated due to the areas in which agricultural activity was stopped and their subsequent renaturalization in both sub-basins.

Table 5 shows the transformation of the CN to surface runoff, according to the typical precipitation considered for the region. As a result, large variations are homogenized and offset along the sub-basins.

However, a significant variation in surface runoff can be observed in the areas of direct contribution with a smaller surface. In both events of T = 5 and T = 25 years, an increase in the range of 4.6% and 2.2% was observed in the percentage of surface runoff from San Gabriel and La Cascada basins, respectively.

The increase in the direct runoff zone could accentuate soil erosion processes and, therefore, a greater accumulation of sediments in Lago del Fuerte, thus reducing its efficiency in water regulation and increasing nutrient inputs.

Table 5 Estimated runoff according to CN in each sub-basin and direct runoff zones
Full size table

These results regarding changes in land use and land cover and their connection to surface runoff constitute the first precedent in the Lago del Fuerte basin and serve as a starting point for future research.

Conclusions

This work consisted of a comprehensive hydrological characterization of Lago del Fuerte basin and its tributaries, considering the morphometry of the basin, the LULC changes and the surface runoff analysis. An intermediate dangerous situation in case of a flood event was identified.

Spatial and temporal changes in land uses and covers were identified. CN values increased mainly due to both replacement of green areas by urbanised ones and an increase in urbanisation density. Inversely, CN values decreased when a use or activity stopped and consequently the area is renaturalized.

While CN values varied widely in each sub-basin, the average CN value, and therefore the surface runoff value, did not show great variations at the total basin level, which is explained by a compensation between areas where the CN increased and those where it decreased.

It is highlighted the need to assess the progress of land use and modification as well as possible impacts. If the trend of population growth and urban-touristic development in the basin continues, together with its morphometric characteristics, the surface runoff could increase and affect the functionality of the reservoir.

This work is expected to contribute to the analysis of the integrated water resource management of the basin in order to protect its environmental quality and the functionality of the reservoir. This information is essential for the monitoring, control, and management of surface and groundwater resources in the basin and particularly for the reservoir maintenance and usefulness.