Multi-criteria assessment of factors affecting the reduction of retention capacity of dam reservoirs

Drought is defined as a lack of water compared to normal conditions that can occur in different parts of the hydrological cycle (Tallaksen and van Lanen, 2004; Van Loon et al., 2016). This phenomenon occurs in a given area under the condition of many natural factors: climatic (shortage or lack of precipitation combined with high air temperatures and high evaporation), physicogeographical (density of the river network, longitudinal slopes of catchments and watercourses), geological and hydrogeological (catchment lithology, substrate permeability), vegetation cover (which determines the rate of infiltration, evaporation, and surface runoff), the presence of lakes and wetlands (influencing the distribution of precipitation and increasing runoff during low water levels, and in mountain catchments deciding on lower low flows) (Tokarczyk et al., 2022a). The development of droughts is also strongly influenced by the related impacts resulting from human activities (use of water resources and changes in land use: deforestation, urbanization, opencast, and underground mining), which have a significant impact on water retention capacity of the catchment (Tokarczyk et al., 2022a, 2022b). This is why water deficits (or droughts) are the result of a complex interaction between meteorological anomalies, land surface processes, and human inflows, outflows, and storage changes (Van Loon et al., 2016).

Past experience shows that drought has become a global problem (Mondal et al., 2023) and is therefore being studied by scientists all over the world. The dominance of the aspect of monitoring and evaluation of this phenomenon and its effects is noticeable (Hao et al., 2022; Jiyoung et al., 2022; Tareke and Awoke, 2022; Abolafia-Rosenzweig et al., 2023; Hailesilassie et al., 2023; Reyniers et al., 2023). For this purpose, a number of drought management tools have been developed, included in the “Handbook of Drought Indicators and Indices” by the World Meteorological Organization and the Global Water Partnership (Svoboda and Fuchs, 2016). Some methods are systematically verified and improved in response to the current hydrological and meteorological conditions (Zhang et al., 2022a). The issue of the origin of drought is less frequently addressed, but also in this aspect, the identification of drought-causing factors is undertaken (Orimoloye et al., 2022)—including the proposed combined drought indicator (Sepulcre-Canto et al., 2012), or the assessment of the impact of human activity on its propagation, from meteorological drought to hydrological drought (Zhang et al., 2022b). At the same time, drought monitoring and early warning systems are based on several indicators, but there is still a lack of assessment of how they are related to the impact (Torelló-Sentelles and Franzke, 2022).

The provisions of Directive 2000/60/EC of the European Parliament and of the Council of October 23, 2000, establishing a framework for community action in the field of water policy imply the possibility of developing detailed programs and plans concerning individual aspects of water management (Directive 2000/60/EC, 2000). On this basis, the need to take into account drought protection in water management planning has also been recognized in Poland, and for this purpose, a methodology was prepared to propose the introduction of organizational, methodological, operational, and implementational aspects in the field of drought prevention (Tyszewski et al., 2008). A study included uniform nationwide rules for conducting an analysis of the possibility of increasing available water resources, the method of identifying and prioritizing areas at risk of drought and those exposed to its effects, and also presented assumptions regarding the construction or reconstruction of hydraulic structures allowing for increasing the level of water retention (Stolarska et al., 2017). On the basis of the methodology in question, in 2021, the “Drought Effects Counteracting Plan” was developed and implemented in Poland, which defines the directions of actions to ensure the appropriate amount and at least good quality of water necessary for the society, the environment and all sectors of the national economy (Regulation of the Minister of Infrastructure, 2021). The document also indicates areas at risk of various types of drought (agricultural, hydrological, and hydrogeological), and among the proposed tasks—which may serve to limit or reduce possible losses in the event of drought occurrence—the need to include the processes shaping water resources in catchments has also been indicated. This approach is consistent with the conceptual model of retention reservoir capacity management, which recommends taking into account technical and nontechnical elements in the area of water retention and sediment management (Pieron et al., 2021). The technical elements taken into account in counteracting or limiting the phenomenon of drought are the construction of retention reservoirs (the multifunctional infrastructure that also often provides flood protection) and taking measures in their catchment area to determine the loss of their capacity (Kondolf et al., 2014). On the other hand, nontechnical measures include, for example, changes in the management of reservoirs to optimize their operation and save water necessary for use during droughts (Woś et al., 2022). The query of available literature shows that relevant studies have been prepared on the subject of water use in the context of counteracting the effects of drought; however, they focus mainly on the issues of retention development and do not necessarily include an element of investigating the causes of fluvial processes (e.g., sediment transport) in individual catchments, leading, among others, to loss of water storage capacity. Nevertheless, their essential provisions and designated directions were the starting point for the carried out work, which resulted in the methodology for conducting a multi-criteria assessment of factors affecting the reduction of retention capacity of dam reservoirs proposed in this article. It was implemented in the catchment area of reservoirs located in the southern part of Poland: Goczałkowice on the Vistula, Rożnów on the Dunajec, and Tresna on the Soła, where anthropogenic pressures and natural factors potentially affecting the reduction of the capacity of individual objects were subjected to detailed spatial analyses: land use, river network, hydraulic development, and land slopes. The subject is particularly important due to the fact that maintaining the appropriate level of retention in the catchment area is important in counteracting the effects of drought.

The methodology of conducting a multi-criteria assessment of factors affecting the reduction of retention capacity of dam reservoirs was based on analyses carried out within selected facilities in Central Europe. For this purpose, catchment basins of retention reservoirs created on 3 different rivers in the Vistula basin, located in the southern part of Poland and the northern part of Slovakia were used: Goczałkowice on the Vistula, Rożnów on the Dunajec, and Tresna on the Soła (Figure 1).

In terms of physicogeography, the catchment area of the Tresna reservoir is located in the area of the Western Beskids macroregion (part of the Carpathians), similarly to fragments of the catchment area of the Goczałkowice and Rożnów reservoirs (Richling et al., 2021). The Western Beskids are built of Outer Carpathian flysch rocks with frequent landslide potentials. This region is characterized by the mountain climate with its visible layering (there are 5 climate zones). The remaining part of the catchment area of the Goczałkowice reservoir is located in the area of the West Beskid Upland and Oświęcim Basin macroregions. The first one is geologically built of less resistant rocks than in the Beskids, but landslides are also possible. The climate is relatively mild with only 1 vertical climate zone. The Oświęcim Basin is more diverse because it belongs to the Upper Silesian sinkhole, filled with sediments of the Miocene Sea. The climate is transitional, conditioned by the influx of air masses from various directions. On the other hand, the Rożnów reservoir and the adjacent catchment area are located in the Central Beskid Upland macroregion, whose geological features are similar to the West Beskid Upland macroregion, and in terms of climate, there are 2 vertical zones: the valley and the moderately warm (foothills). The remaining part of the catchment area of the Rożnów reservoir is part of the following macroregions: Orawa-Podhale Depression (a tectonic sinkhole with a mountain climate) and the Tatra Chain (in geological terms, it is made of crystalline rocks, limestones and dolomites, and climatically the area is characterized as high mountain).

The analyzed reservoirs are located in an area threatened by hydrological drought (Regulation of the Minister of Infrastructure, 2021; Absalon et al., 2023). Each of the reservoirs performs many functions related to water management, and the basis for their construction was flood protection of the areas located downstream, and currently, in the context of climate change, the use of water collected in them to counteract the effects of drought is playing an increasingly important role.

The choice of reservoirs located in mountainous areas or in their foreground results from the high rate of sediment transport in their catchments, in which the basic component of the clastic material is suspended load (even 90%–95%; Łajczak, 1999). The key aspect was also the selection of reservoirs located in a similar area and characterized by a varying degree of capacity reduction compared to the initial value (at the time of commissioning). In each of the considered reservoirs, a loss of capacity was recorded: large—at the level of as much as 31.9% (72.9 million m³) in the Rożnów reservoir and moderate—amounting to 9.1% (9.3 million m³) in the Tresna reservoir and 1.1% (1.1 million m³) in the Goczałkowice reservoir (Pieron et al., 2021). This means that, on average, each year they lose, respectively, 0.4% (0.923 million m³), 0.17% (0.172 million m³), and 0.02% (0.027 million m³) of their water storage capacity (Pieron et al., 2021).

The area and capacity of the analyzed reservoirs at the maximum damming level are as follows:

• Goczałkowice: 32 km² and 161.30 million m³ (RWMA Gliwice, 2021).

• Rożnów: 16 km² and 155.70 million m³.

• Tresna: 10 km² and 92.70 million m³ (RWMA Kraków, 2021).

The total area of the studied reservoirs is 58 km² and they can store almost 410 million m³ of water. In both aspects (area and capacity), the largest object is the Goczałkowice reservoir. However, the situation is different in terms of capacity at the normal damming level and catchment area. The Rożnów reservoir has the largest capacity—155.77 million m³ (it does not have a flood reserve; RWMA Kraków, 2021), while the Goczałkowice stores 118.10 million m³ of water in such conditions (RWMA Gliwice, 2021), and the Tresna 53.90 million m³ (RWMA Kraków, 2021). In turn, in terms of catchment area, the Rożnów reservoir has the largest catchment area—4900 km² and it significantly exceeds the catchment area of the Tresna reservoir—1100 km² and the Goczałkowice reservoir—430 km² (Map of Hydrographic, 2023).

Geospatial analyses (GIS) tools in the grid of basic fields were used to analyze the factors influencing the assessment of the decrease in retention capacity of the Goczałkowice, Rożnów, and Tresna reservoirs. They made it possible to perform a multi-criteria assessment of the factors causing the loss of reservoir capacity. The following grid model was adopted for each studied basin of the reservoir: field shape—hexagon, area of one field equal to 2.593 km² and width equal to 2 km.

The steps of analysis using the primary field grid are as follows:

1. For each of the parameters (percentage of individual land use areas, density of the river network, number of transverse structures, length of longitudinal development, and land slopes), the results are assigned to the grid of basic fields.

2. Then, each field is assigned an appropriate point value (rank) depending on the value of the analyzed parameter. When defining the rank ranges in the analyzed catchments, all the studied factors were assigned the following parameters: the best, average, and the weakest. In the case of land use, the percentage share in the hexagon of the area of a given form of land development was significant (the dominant share decides the qualifying class). In the analysis of the river network, density was used as the key parameter. For the hydraulic development, the presence of a given type of structure on watercourses (transverse or longitudinal) or their absence was taken into account (in the case of 2 different types of infrastructure, results are assigned to the one ranked higher). Land slopes were divided in terms of size using statistical analysis, for which the XLSTAT software was applied (descriptive statistics were used, fed with land slope data). Thus, each field receives a score for all subsequent criteria. Unclassified areas have a basic field value of zero (e.g., no hydraulic structures in the basic field).

3. The area hierarchization procedure is also carried out in the basic fields. The maximum value of the parameter determines the qualification of a given field to the parameter class (this has been clarified above for forms of land development and hydrotechnical development, and the reasons for choosing such a scheme of operation are also included in the further part of the study). The results of the analysis are maps of the range of hierarchized catchment areas of reservoirs for each parameter and a map of summary factors influencing the assessment of the reduction of the retention capacity of reservoirs. The final classification of the primary field is the maximum value of the points/criteria obtained.

To sum up, the analysis of the resulting GIS maps (land use, river network, hydraulic development, land slopes) together with the described factors affecting the susceptibility to reducing retention of reservoirs allowed for a score for the described parameters (criteria). The obtained results are intersected and entered into the grid of basic fields (hexagon grid).

It is also worth noting that the analyzed attributes were of different natures: point (transverse structures), linear (river network and longitudinal structures), and surface (land use and slopes), so for this reason the maximum value of the class obtained in the basic field was adopted for the final classification. This is due to the differentiation of the attributes of the analyzed databases. In the point and linear databases, it was assumed that the absence of such objects in the hexagon are unclassified areas, which have a basic field value of zero. On the other hand, surface databases covered all basic fields. It was therefore assumed that unclassified areas with a value of zero will underestimate the average classification value but are significant in terms of the impact on the loss of reservoir capacity in accordance with the established criteria. In turn, the average value of the points obtained in each hexagon was adopted for the final classification.

Multifaceted analyses required obtaining basic materials from various institutions. The database developed by the European Environment Agency (CORINE, 2022) is the source of data on the use of catchment basins of the studied reservoirs. Data from the Map of the Hydrographic Division of Poland in on the Scale of 1:10000 (Map of Hydrographic, 2023) and the Slovak basic database for the geographic information system (ZBGIS, 2022) were used to analyze the river network. When assessing the existing hydraulic structures (transverse and longitudinal on the main river and its tributaries, and longitudinal on the floodplain) in Poland, the digital Database of Topographic Objects was used, containing the location and types of transverse and longitudinal structures (BDOT10k, 2023), and for the section of the Dunajec data from the National Geoportal of Slovakia was also used (National Geoportal, 2022). The European Digital Elevation Model from the Copernicus Global Land Service (Copernicus, 2023) was used to determine the longitudinal slopes of the catchment. A digital terrain model was created, which was then used in statistical analyses to calculate the basic height parameters of the study areas and the most important element, that is, the inclination of the catchment in per mille. Whereas information on the basic parameters of retention reservoirs is presented mainly on the basis of data obtained from the State Water Holding Polish Waters: Regional Water Management Authority in Gliwice (RWMA Gliwice, 2021) and Regional Water Management Authority in Kraków (RWMA Kraków, 2021).

The first stage in defining and preparing the methodology for conducting a multi-criteria assessment of factors affecting the reduction of the retention capacity of dam reservoirs was the selection of elements that have the greatest impact on sediment transport within the catchment area of the reservoirs: Goczałkowice, Rożnów, and Tresna. In this type of analysis, anthropogenic pressures (land use and hydraulic development) and natural impacts (river network, slopes, and geological structure) are examined (Figure 2). The previous experience of the authors allowed for the selection of the leading factors for further research, such as land use, river network, hydraulic structures, and land slopes.

The next important step was to categorize the above factors. Three-level ranks were established to classify the individual examined parameters in terms of their impact on reducing sediment transport to reservoirs, indicating the best, average, and weakest values (Table 1).

In terms of land use, it was found that the supply of sediment to a given watercourse and later to the reservoir is the smallest among forest areas and wetlands. Agricultural areas are responsible for supplying sediment to watercourses, while anthropogenic areas potentially concentrate and accelerate surface runoff, favoring greater movement of material. In the case of the river network, it was assumed that its greater density favors the transport of sediments to a given reservoir, and the ranks of individual basic fields were determined according to the available classification (Chirala et al., 2012). On the other hand, the applied hydraulic development (in the form of various types of weirs, thresholds, steps, rapids, and debris barriers) is designed to stabilize the bottom and at the same time limit the movement of sediment. Its impact also breaks the continuity of the river and dams up water upstream of some of these objects (DHI Poland, 2019). Among the longitudinal structures shore bands made of various types of material predominate, which counteract the undermining of the shore and the resulting movement of sediments. Their impact is also related to direct interference in the riverbed or bank and breaking the river continuity (DHI Poland, 2019). On the other hand, the rate and intensification of the supply of material to a given watercourse and, consequently, to the studied reservoirs depends on the diversified longitudinal slopes in the catchment. Therefore, on the basis of the prepared digital terrain model (Figure 1), statistical analyses were carried out, which allowed for the diversification of the study area in terms of the inclination of the catchment.

The last element was to compile a list of all the factors taken into account, that is, the so-called “intersection” of the obtained partial results (Figure 2). Therefore, on the basis of a multi-criteria analysis, the components affecting the reduction of the retention capacity of dam reservoirs were assessed, while determining parts of the catchment with the best and worst properties. For this purpose, the average values of all tested components were adopted, in accordance with the current classification of parameters (Table 2).

The largest percentage of the catchment area of the Goczałkowice reservoir (38.1%) is occupied by forest areas and it slightly exceeds the area of agricultural land (37.1%). The share of urbanized areas, which is 13%, is also much higher than in the catchment area of the Rożnów and Tresna reservoirs. In addition, areas covered with water are observed (mainly the Wisła Czarne reservoir and fish ponds) and, to a lesser extent, other forms of land use (CORINE, 2022; Absalon et al., 2023). The catchment area of the Rożnów reservoir is covered mainly by forests, which constitute almost half of its area (45.6%). As much as 41.4% is occupied by areas related to agriculture and it is the largest share among all the analyzed catchments. On the other hand, the situation is reversed in the case of urbanized areas, the percentage of which is only 4.9%. Water areas, despite the presence of the large Czorsztyn Niedzica and Sromowce Wyżne reservoirs, cover 0.8% of the catchment area, and the share of other types of land use does not exceed 0.4% (CORINE, 2022; Absalon et al., 2023). On the other hand, more than half (50.5%) of the catchment area of the Tresna reservoir is occupied by forests and this indicator is higher than the catchment area of the Goczałkowice and Rożnów reservoirs. Compared to them, the smallest share of agricultural land (28.3%) is also significant. On the other hand, urbanized areas cover 8.5% of the analyzed area, and inland waters 0.9% (CORINE, 2022; Absalon et al., 2023).

The tests performed using hexagons show that the largest percentage of areas with the best parameters (dominating forests, waters, and wetlands) occurs in the catchment area of the Tresna reservoir—64.1% (Figure 3). They cover more than half of the area of the remaining basins of the reservoirs under consideration: Rożnów—54.3% and Goczałkowice—53.4% (Table 2). In the case of areas used for agriculture (average conditions), they constitute the following percentage of the catchments of the reservoirs: Rożnów—43.9%, Goczałkowice—39.6%, and Tresna—31.7% (Table 3). Another important factor is the relatively small share of areas with dominant anthropogenic elements (the weakest parameters), whose share in the catchment area of the Goczałkowice reservoir is 7.0%, and in the catchment areas of the Tresna and Rożnów reservoirs is at a lower level (4.2% and 1.8%, respectively; Table 3). The average value of the land use criterion for the entire study area (all 3 catchments) for individual parameters was also calculated: the best—55.8%, average—41.6%, and the weakest—2.6% (Figure 3).

The longest river network, which is 3,220 km, is characteristic of the catchment area of the Rożnów reservoir (Map of Hydrographic, 2023; Woś et al., 2022), which definitely distinguishes it from other areas. Nevertheless, its density is the lowest and amounts to 0.66 km/km² (Map of Hydrographic, 2023; Woś et al., 2022). This is due to the large catchment area of the Rożnów reservoir. Clearly, the most developed river network is located in the catchment area of the Goczałkowice reservoir, with a density of 0.99 km/km², which consists of a total of 427 km of watercourses (Woś et al., 2022). In turn, the river network in the catchment area of the Tresna reservoir is almost 4 times shorter than that recorded in the catchment area of the Rożnów reservoir (868 km), and its density is 0.79 km/km² (Woś et al., 2022).

Verification of the component concerning the density of the river network showed that in total 61.5% of the study area had the best parameters (below 0.5 km of watercourses per 1 km²), 20.4% (from 0.5 to 1.5) average, and 18.1% the weakest (above 1.5; Table 3). The poor development of the river network in the catchment area of the Rożnów reservoir was also reflected in the results obtained in the form of basic fields, where as much as 2/3 of its area obtained the best parameters (Figure 4). For the same reason, the share of the weakest areas is the smallest among the considered areas and amounts to 17.4%, with a similar level for the catchment area of the Goczałkowice (20.2%) and Tresna (20.3%) reservoirs (Table 3). It is also worth noting that the density of the river network in the case of the Goczałkowice catchment is higher than in the catchment area of the Tresna reservoir, but their spatial distribution does not reflect this in the case of classes below 0.5 km of watercourses per 1 km² of the catchment area (51.9% and 45.0%, respectively; Table 3).

In the catchment area of the Goczałkowice reservoir, a total of 351 transverse structures were built in the riverbeds (Absalon et al., 2023; BDOT10k, 2023). This is definitely less than in the case of other catchments, but it results from the smallest area of the analyzed area. Thus, the combination of both of these factors results in a fairly high density of transverse hydraulic development, which is 0.82 per km² of the catchment area and 0.82 per km of watercourses within it (Absalon et al., 2023). In the initial course of the Vistula, there is the Wisła Czarne reservoir, and in the remaining section and other rivers, there are, among others, weirs, thresholds, steps, and debris dams limiting the transport and delivery of sediment. In the case of the catchment area of the Rożnów reservoir, there are numerous hydraulic structures located across the watercourses, and their total number is 1,758 (Absalon et al., 2023; BDOT10k, 2023). This is definitely more than the values recorded in the catchments of the Goczałkowice and Rożnów reservoirs. Despite this, due to the large area of the catchment, their density should be considered relatively low: 0.36 objects per km² of the area occupied by it and 0.55 per km of watercourses occurring within it (BDOT10k, 2023). The existing types of hydraulic structures include, among others: weirs, thresholds, steps, debris dams, and retaining walls, and apart from the Rożnów reservoir, there is also a complex of reservoirs Czorsztyn Niedzica—Sromowce Wyżne. In terms of hydraulic development, there are 0.47 objects of this type per km² of the catchment area of the Tresna reservoir, and 0.59 per km of watercourses located in its area (Absalon et al., 2023). This is due to the presence of 512 transverse structures in the riverbeds (Absalon et al., 2023; BDOT10k, 2023) in the form of for example, weirs, thresholds, steps, and debris dams. On the other hand, the longitudinal infrastructure is only fragmentary.

The analysis of the hydraulic development showed that the areas with the best parameters account for a total of 17.4% of the area of the studied catchments, with average ones—10.8% and the weakest parameters, which definitely prevail, account for 71.8% (Figure 5). However, there were large differences in this respect broken down into individual areas. The results in the catchment area of the Goczałkowice reservoir are the most favorable, where the share of areas of the above classes is as follows: 31.8% (with transverse hydraulic structures), 28.6% (existing longitudinal structures), and 39.6% (without hydraulic structures; Table 4). In the case of the catchment area of the Tresna reservoir, the calculations for the indicated classes are, respectively, 20.6%, 16.4%, and 63.0% (Figure 5). There are visible areas where there are no hydraulic structures, but the catchment area of the Rożnów reservoir is by far the worst among the considered areas. The areas with the best conditions cover 15.1% of the area, the average ones cover 7.8%, and more than three quarter (77.1%) have the weakest parameters (Figure 5).

On the basis of the created Digital Terrain Model (Copernicus, 2023) and statistical analyses carried out based on it, the basic parameters of the catchment areas of the Goczałkowice, Rożnów, and Tresna reservoirs were determined. Their relief is mountainous, and the sources of the rivers on which the dams were created are in the Carpathians (Dunajec in the Western Tatras, Soła in the Żywiec Beskids, and Wisła in the Silesian Beskids). In the case of the Goczałkowice reservoir catchment, the maximum height is 1,207 m above sea level, and the minimum is 243 m above sea level (Digital Terrain Model data). Therefore, the recorded denivelation is 964 m. In turn, the average height of the terrain is 450 m above sea level. In the catchment area of the Rożnów reservoir, the denivelation is the highest among the studied areas and amounts to 2,367 m. This is due to the much higher maximum elevation (2,608 m above sea level) and the lowest point—located at a similar level as in the case of the catchment area of the Goczałkowice reservoir—with an ordinate of 241 m above sea level. This also contributes to the by far highest average height of the terrain at 765 m above sea level. On the other hand, the highest points in the catchment area of the Tresna reservoir occur at an altitude of 335 m above sea level and 1,521 m above sea level, and the difference between them is 1,186 m. The average altitude of the area is 670 m above sea level.

The performed calculations showed that in terms of the slope value, the studied catchments have similar properties. The distribution of individual classes in the total area of the entire analyzed area is as follows: the best parameters—23%, average—51%, and the worst—26% (Figure 6). Areas with the lowest slopes are most numerous in the catchment area of the Goczałkowice reservoir (27.9%), slightly less in the case of the Tresna reservoir (27.5%), and the worst in the catchment area of the Rożnów reservoir (25.7%; Table 5). Medium value parameters prevail in the case of 2 areas associated with the reservoirs: Tresna (52.8%) and Rożnów (50.7%) and clearly dominate in the catchment area of the Goczałkowice reservoir (49.2%; Table 5). The remaining part consists of areas characterized by the greatest slopes and their share is for the basins of the following reservoirs: Rożnów (23.6%), Goczałkowice (22.9%), and Tresna (19.7%; Table 5).

The multi-criteria assessment of the factors responsible for the supply of sediment and reducing the retention capacity of the dam reservoirs: Goczałkowice, Rożnów, and Tresna (land use, river network, hydraulic development, and land slopes) allowed us to conclude that land slopes are the key component in this respect. The distribution of areas with different inclinations largely coincides with the results of the “intersection” of all the tested elements (Figure 7). The analysis showed that this factor is to the greatest extent convergent with the final results obtained, and the verification was carried out in 2 ways—an advantage over other tested elements is visible both in terms of the number of hexagons and their area (Table 6).

It should be noted, however, that the obtained results are significantly affected by the extensive river network observed throughout the area and the hydraulic structures shown on the result map, mainly in the form of linear sequences related to the course of individual watercourses (Table 6, Figure 7). In turn, land use plays a very important role in increasing the value (in terms of reducing the ability to initiate erosion processes) of other areas and qualifying them to the medium category. This is mainly due to the fact that forested areas (the most favorable ones) are located mainly on steep slopes, devoid of hydraulic development (Figure 7), and rated low in terms of their impact on limiting the supply of sediments to retention reservoirs. In total, the areas with the best parameters in terms of limiting sediment supply to the Goczałkowice, Rożnów, and Tresna reservoirs cover 13.6%, average conditions are present in most of the area (85.6%), and areas with the weakest criterion 0.8 (Figure 7).

Taking into account partial results from the present study, the most favorable result was recorded within the catchment area of the Goczałkowice reservoir (Figure 7), which is consistent with the rate of its capacity loss compared to the other considered objects. Almost 1/4 of its area (23.4%) is characterized by the best parameters, that is, 9.8% points more than in the case of the Tresna reservoir and 10.9% points more than the catchment area of the Rożnów reservoir (Table 7). On the other hand, the share of areas with average parameters is dominant in all areas and amounts to 75.7%, 86.6%, and 85.9%, respectively (Table 7). The occurrence of areas with the weakest parameters is negligible in all catchments and they occupy less than 1% of their area (Figure 7). At the same time, in the case of the Rożnów and Tresna reservoirs, the results are so similar (Figure 7) that it is difficult to clearly indicate which of them has better conditions.

Analyses using hexagons turned out to be an effective method for conducting a multi-criteria assessment of factors affecting the reduction of the retention capacity of dam reservoirs: Goczałkowice on the Vistula, Rożnów on the Dunajec, and Tresna on the Soła. An undoubted advantage of the proposed solutions is a comprehensive approach to considering all factors taken into account and the objectivity of the results obtained. In this aspect, the level of detail (hexagon size) can be adapted to the individual requirements of the subject of the study, related to, for example, the area or type of a given catchment. At the same time, it should be remembered that in this type of research, the availability of data may be a limitation. In the case of this article, publicly available databases developed by European, Polish, and Slovak public institutions were used. Due to this, complete and coherent materials were available, enabling the planned activities to be carried out. On the other hand, for the area for which complete materials are not available, it is necessary to perform detailed studies of all the analyzed factors within the studied catchment area (land use, river network, hydraulic development, and land slopes) along with detailed characteristics. In some situations, this can be a time-consuming and costly task, especially if it involves fieldwork.

The use of a methodology for conducting a multi-criteria assessment of factors affecting the reduction of the retention capacity of dam reservoirs, taking into account land use, river network, hydraulic development, and land slopes, allowed for the determination of the catchment parameters selected for analysis. It also made it possible to indicate that the key component in maintaining the proper water storage capacity in the Goczałkowice, Rożnów, and Tresna reservoirs are terrain slopes. This result may be surprising, because in the literature opinions in this regard are divided. On the one hand, it is indicated that the relief imposes a specific type of water circulation (and in mountainous areas, its essential feature is a relatively quick outflow), but the slope inclination does not play a direct role in varying the size of surface runoff or is a minor factor (Słupik, 1973). On the other hand, contradictory information can be found that areas with an increased slope show a much higher rate of erosion to the stream channels than plain areas (Kolli et al., 2021), and the intensity and variability of denudation generally increases with the average inclination of the catchment (Delunel et al., 2020). However, only the application of the authors’ methodology made it possible to verify and define the role of the discussed factor in terms of its impact on sediment management in the catchment.

It is also worth paying attention to land use, which, in the light of the obtained results, contributes to a large extent to increasing the value of areas with the poorest parameters (including those with large slopes and without hydraulic development). Earlier analyses emphasized the importance of a broader approach to this issue but did not specify its impact on fluvial processes in the study area (Absalon et al., 2023). Comparing the obtained results with the experiences of other scientists from the analyzed area, one can come across the same and even more radical conclusions. The method of land use (particularly agricultural) influences the amount of suspended solids outflow in the Carpathian part of the Vistula basin, causing the leveling of the impact of other elements of the geographical environment on sediment transport in the catchment (Łajczak, 1989). Thus, the introduction of changes in the structure and use of the study area could counteract the loss of storage capacity of the Goczałkowice, Rożnów, and Tresna reservoirs. At the same time, it should be remembered that various attempts have been made in the world to reduce the flow of sediment to reservoirs through changes in land use (mainly through afforestation and changes in agricultural practices), but the benefits of reducing sediment transport have not been clearly demonstrated everywhere (Kondolf et al., 2014).

The research takes into account the above assumptions and presents a new approach to the method of assessing the factors affecting the reduction of the retention capacity of dam reservoirs. The analysis of the “intersection” of the results in the field of land use, river network, hydraulic development, and land slopes allowed to obtain comprehensive information on the analyzed catchments. It can be said that the previous research covered only selected elements in the field of, for example, sediment transport or retention capacity. A very often used method is USLE (Universal Soil Loss Equation), which refers to soil erosion caused by water (except gully erosion and landslides), specifying their degradation on a small and large scale (Alewell et al., 2019). This process can lead to a loss of capacity of retention reservoirs, and in the course of work, extended versions of RUSLE (Revised Universal Soil Loss Equation) and MUSLE (Modified Universal Soil Loss Equation) have also been used for its analysis (Benavidez et al., 2018). The SWAT (Soil and Water Assessment Tool) model is also useful in landscape research, determining the routes of water streams and determining soil dynamics (Douglas-Mankin et al., 2010). In turn, the FLDAS system (Famine Early Warning Systems Network Land Data Assimilation System) can be used to calculate various indicators related to the drought phenomenon and covering the catchment areas and the reservoirs themselves (Albarakat et al., 2022). In addition, we can come across many different morphometric indices for the analysis of sediment dynamics in catchments, including erosion volume and dynamics (Marchi and Dalla Fontana, 2005), for which, as in this article, for example, GIS tools are used. However, this multi-criteria assessment considers many factors affecting the reduction of the retention capacity of dam reservoirs. It has been described in a way that allows its use also on other reservoirs, and in the first place, it is recommended to implement it on catchments of upland and lowland reservoirs.

In addition, GIS spatial analyses using a grid of basic fields (hexagons) are used in various fields. The conducted query of scientific studies shows that they are used, among others, in meteorology and climatology (Martinez et al., 2023), urban planning (Feick and Robertson, 2015), environmental research (Annis et al., 2017), or even medical sciences (Altonen, 2017). The grid of basic fields in the shape of a hexagon was used throughout Poland also in the field of hydrological analyses. This method made it possible to identify areas at risk of drought (agricultural, hydrological, and hydrogeological) as part of the Drought Effects Counteracting Plan (Regulation of the Minister of Infrastructure, 2021). In turn, the authors used this research method in the evaluation of pressure types impacted on sediment supply to dam reservoirs (in the field of hydraulic development; Absalon et al., 2023) and spatial multi-criteria analysis of water-covered areas (Janczewska et al., 2023).

The use of the proposed methodology at the design stage of dam reservoirs should bring measurable effects in terms of assessing their life span. By adopting the presented assumptions, it is possible to increase the level of retained water and to store it for a longer time as a result of partial elimination or limitation of factors that may affect the loss of initial parameters for certain dams. In this way, it is possible to maintain the functions of reservoirs in the future (Kondolf et al., 2014), which also includes counteracting the effects of drought. In this aspect, it should also be remembered that along with the growing demand for water retention and the decreasing number of places available for new reservoirs, the loss of capacity within existing facilities poses a threat to the durability of the water supply (Annandale, 2013). On the other hand, one of the goals of the United Nations is to ensure sustainable development in the world, that is, meeting current needs without compromising the ability to meet them also for future generations (United Nations, 1987). At the same time, in many regions water retention in reservoirs will have to play an important role in mitigating climate change, also aimed at providing water, food, and energy and reducing the risk of flooding (Randle et al., 2021). Therefore, there is an undeniable need to maintain the current capacity of reservoirs also for the purpose of counteracting the potential effects of drought.

In order to manage the capacity of reservoirs, it is also possible to apply various technical measures, taking into account environmental requirements: directing sediments through or around the reservoir (e.g., bypass channel or tunnel, sluicing, and flushing), removal of accumulated sediments in the reservoir to regain capacity (dredging), and minimizing the amount of sediment flowing into the tank. The use of all these methods makes it possible to maintain the designed parameters of the facility—however, in the case of the first two, unlike the last, the supply of material to the lower course of the river will be maintained (Kondolf et al., 2014). It is also worth noting that reducing the amount of sediment transported to the reservoir is possible by using hydraulic structures on the tributaries to the reservoir (including, above all, on the main river) like the ones used in the analyzed catchments, for example, smaller control weirs or debris dams. Such structures do not necessarily have to be concrete thresholds and walls—in the construction of such smaller and more densely located objects, local rock material and wood can be successfully used. In the field of hydraulic development, examples from Poland and around the world show that its effective use can extend the life of retention reservoirs, that is, ensuring full capacity to perform strictly defined functions. The use of appropriate regulatory structures made it possible to retain 70% of the debris in the upper course of the Yangtze River, and thus 164 million tons of load reached the Three Gorges Dam annually. At the same time, as a result of the introduction of the sediment management strategy (including its greater discharge during floods), the accumulation in the reservoir decreased by 33% (Shi et al., 2021). A positive effect in the form of reduced transport of sediment to the Czchów reservoir—located on the Dunajec River, in the vicinity of the Rożnów reservoir—was brought about by engineering works on the Łososina River (Radecki-Pawlik et al., 2019).

The study utilized GIS to spatially analyze the available data about land use, river network density, hydraulic development, and land slopes in 3 catchments in South Poland with water reservoirs (Goczałkowice on the Vistula, Rożnów on the Dunajec, and Tresna on the Soła) in order to determine that the proposed multi-criteria methodology can be an effective solution for conducting an assessment of factors affecting the reduction of the retention capacity of dam reservoirs. The selection of the catchments is justified by their location in an area threatened by hydrological drought, and management of the catchments is crucial to prevent a reduction in water storage capacity in the context of climate change. In addition, the catchments are known for high sediment transport regimes due to their mountainous terrain, and data are available for them. The results show that the slope of the catchment is a critical parameter for the loss of capacity of the studied retention reservoirs. In order to counteract the reduction of retention capacity, it is worth taking into account the issues of changes in land use and the use of appropriate hydraulic development. While the slopes of the land cannot be changed, some technical solutions can be implemented to slow down the speed of the water flow using locally available resources, such as small control weirs or debris dams.

The use of the newly developed methodology allows the determination of the best, average, and worst parameters in terms of sediment delivery to retention reservoirs for each unit (hexagon) of the studied catchments. In total, these areas occupy, respectively, 13.6%, 85.6%, and 0.8% (same results as for the Tresna reservoir). The best parameters were determined in the Goczałkowice reservoir, which is consistent with the rate of its capacity loss compared to the other considered reservoirs. Conducting this type of research is important in the aspect of sediment management within the catchment, which has a significant impact on maintaining or developing retention capacity in the context of counteracting the effects of drought. The study provides a “recipe” for data analysis that can be implemented in other areas to determine the influencing factors of sediment transport and reduction of water holding capacity of reservoirs and implement proper management techniques. However, this can be limited by data availability for specific catchments.

The hexagons data of this study are in the supplemental file of the article.

The supplemental files (Supplemental—hexagons tables) for this article include data on the hexagons used in the article: their geometry, numbering and areas, as well as the classes of individual factors (land use, river network, hydraulic structures, and land slopes), and the final evaluation.

This research was funded by the funds granted under the Research Excellence Initiative of the University of Silesia in Katowice.

The authors have declared that no competing interests exist.

Contributed to conception and design: ŁP, DA, MM.

Contributed to acquisition of data: ŁP, DA, MM.

Contributed to analysis and interpretation of data: ŁP, DA, MM.

Drafted the article: ŁP.

Revised the article: DA, MM.

Approved the submitted version for publication: ŁP, DA, MM.

How to cite this article: Pieron, Ł, Absalon, D, Matysik, M. 2024. Multi-criteria assessment of factors affecting the reduction of retention capacity of dam reservoirs. Elementa: Science of the Anthropocene 12(1). DOI: https://doi.org/10.1525/elementa.2023.00069

Domain Editor-in-Chief: Steven Allison, University of California Irvine, Irvine, CA, USA

Associate Editor: Chuan-Chou Shen, Department of Geosciences, National Taiwan University, Taipei, Taiwan

Knowledge Domain: Ecology and Earth Systems