**1. Introduction**

Soil water content is an important component of the hydrological cycle. The formation of water resources in the catchment area is greatly influenced by the amount of precipitation, evapotranspiration, temperature as well as soil properties (water storage capacity, texture, structure), management practices and the existing vegetation [1,2]. The main source of soil water content is precipitation through infiltration and surface runoff [3]. Temperature, on the other hand, influences the evapotranspiration process [4].

**Citation:** Badora, D.; Wawer, R.; Nieróbca, A.; Król-Badziak, A.; Kozyra, J.; Jurga, B.; Nowocie ´n, E. Simulating the Effects of Agricultural Adaptation Practices onto the Soil Water Content in Future Climate Using SWAT Model on Upland Bystra River Catchment. *Water* **2022**, *14*, 2288. https://doi.org/10.3390/ w14152288

Academic Editors: Alban Kuriqi, Luis Garrote and Ian Prosser

Received: 7 March 2022 Accepted: 12 July 2022 Published: 22 July 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

There are numerous studies focusing on the calculation of soil water content using the Soil and Water Assessment Tool (SWAT) model [5,6]. These authors used SWAT to simulate soil water content at levels of large catchments (Vistula, Odra). They demonstrated the ability to generate long-term series of soil water content even in the absence of comparative data. On the other hand, for a small catchment located in Poland, one of the few similar studies to the present one in terms of climatic scenarios as well as parameters studied (soil water content, actual evapotranspiration) is a publication concerning the Barycz and Upper Narwia catchments [7].

In the publication on soil water retention and drought risk assessment based on water balance for the area of the Lower Silesian province [1], soil retention parameters were determined: Available Water Capacity (AWC), Wilting Point (WP), Field Capacity (FC) for soil species found in Poland. The retention parameters were determined by expert methods [1].

The aim of the article is to analyze five adaptation scenarios (AS-1, AS-2, AS-3, AS-4, AS-5) in relation to the 2041–2050 climate projections GCMs/RCMs for the RCP 4.5 and RCP 8.5 climate change scenarios described as scenario 0 (S-0) [8], as well as their assessment against the current state of knowledge related to research involving similar adaptation studies. Adaptation scenarios 1–5 are modifications of Scenario 0.

The need for such studies of small catchments (up to a few hundred km2) is due to the small number of studies that would be based on adequate preparation of soil parameters (e.g., retention). Moreover, for the Polish area, there are no studies on adaptation scenarios that would attempt to increase the water content of soil and minimize the adverse effects of climate change (RCP 4.5, RCP 8.5) in future decades.

Among the many hydrological models in use today, the SWAT model, widely used by scientists and developed by the USDA [9,10], was selected for this study because of its ability to predict the impact of practices of land management onto the hydrology and water quality in the catchment area.

Much research is currently being conducted on climate change and the associated unpredictability of extreme weather events. This raises legitimate concerns about the possible emergence of environmental, social and economic threats in the decades to come. These changes may also have an impact on agriculture in Poland [11]. The increase in air temperature, which was observed in recent decades, contributed to the increase in potential evapotranspiration, especially in the last decade 2011–2020. A large increase in potential evapotranspiration and an increase in the variability of this indicator were found [8,12,13]. Recent decades also brought observations of climate change in Poland resulting from the world global warming, changes in precipitation and a number of weather extremes [14–16].

These changes also concern the extension of the growing season in Poland. For the years 1971–2000, the length of the growing season was 218 days (from March 31 to November 4) [17]. According to studies on the change in the growing length in Poland [17], the length of the growing season will extend by 18–27 days in the perspective of 2050 compared to the years 1971–2000.

The increase in evapotranspiration, temperature and precipitation in the coming decades will, to a greater or lesser extent, also apply to all European countries [18,19].

According to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [20], the average temperature of the Earth's surface will reach 1.5 degrees Celsius in the coming decades above pre-industrial levels. Moreover, in Poland, a projected 10-fold increase in the occurrence of droughts by 2020 [21] is observed in the data; hence, the predictions using climate change models seem to reflect changing climate quite well for Poland [22]. Until recently, climate change adaptation received less attention in Poland than climate change mitigation. The vast majority of national communications have been devoted to climate projections, vulnerability and impacts. However, recently there has been increasing attention to adaptation measures in agriculture, among others [16].

There is a need to look for solutions that will reduce the negative impact of climate change [23], inter alia, the occurrence of weather extremes, including drought [12,24,25] in the coming decades. Climate change adaptation in agriculture is associated with a number of preventive measures (adapting crops to changing thermal and water conditions). These include changes in adaptation practices and the introduction of new varieties. Protecting the soil and its water resources is also extremely important. Soil moisture can be maintained through mulching and water conservation through efficient irrigation and water storage (small retention, filter strips). Soil fertility and its potential for water storage can also be increased by increasing soil organic matter [16].

For the sake of this paper, we chose the Bystra river catchment (South-Eastern Poland) as the study area. In order to check the effectiveness of the designed adaptation solutions, it was necessary to develop boundary conditions that would indicate the reference level [8]. These conditions show the behavior of the hydrosystem of the Bystra catchment in the Business as Usual scenario. It takes into account changes in the hydrological cycle caused solely by climate change while maintaining unchanged conditions of human activity. The described boundary conditions for the 2050 horizon must be based on simulation modeling, which is calibrated on archival data. The appropriate tool for this is the SWAT model.

The article presents a comparison of the results of soil water content (profile 1.5 m) for five adaptation scenarios obtained via a simulation of a calibrated and validated SWAT model [8] for three regional climate models derived from the global EC-EARTH climate model for the years 2041–2050 (S-0). Then, the results of scenario 0 were compared with the results of adaptation scenarios 1–5, which included land use changes and protective measures.

The publication is presented as follows: Section 1 presents the Introduction; Section 2 introduces the methodology and describes the study area. Section 3 describes the results, and Section 4 presents the discussion in terms of results regarding soil water content, total runoff and sediment yield, followed by conclusions in Section 5.

#### **2. Material and Methods**

This section is divided in 5 sub-sections. The first one describes the study area; the second and the third describes the SWAT model and SUFI-2 model; the fourth presents climate change scenarios, and finally the fifth presents climate change adaptation scenarios 1–4.

#### *2.1. Characterization of the Study Area*

The Bystra catchment area is situated in the north-western part of the Lubelskie Province (Figure 1). The length of the Bystra River is 33 km, and it is the right tributary of the Vistula river. According to the generated SWAT model, the lowest point of the catchment area is 126 m above sea level, and the highest point is 246 m above sea level. The catchment area delineated froma5mresolution DEM is 296.6 km2 [8].

The Bystra catchment area is part of the Lublin Upland [26–28]. The valley of the Bystra river and its tributaries are strongly carved in a thick loess layer overlaying calcareous bedrock. It consists of numerous valley forms with a constant or episodic tributary. The largest valley with a constant tributary, the Bystra valley, is 35 km long. In the part where the Bystra valley flows into the Vistula, it cuts up to 35 m in rocks and marls [29–31].

The upland nature of the Bystra catchment area, consisting mostly of loess soils, with a high slope of the slopes at the mouth of the Vistula, poses a high risk in terms of medium and very strong water and surface erosion [32].

Most of the Bystra catchment area is made of loess up to 20 m. In the deeper layers, there are Quaternary Pleistocene sediments: water-glacial sand and gravel and, at a little deeper level, tilts. On the other hand, there are geodes under the clays. Under the geysers, on the other hand, there are deposits of the Upper Cretaceous: rocks with lime inserts [33].

The study area consists mainly of podzolic and lessivage (49%) soils, which extend mainly in the south-eastern part of the catchment as well as cambisols (47%) in the northwestern part. The predominant soil texture in the catchment area is loess (73%) [34–36] and silt (18%) [8].

In the Bystra catchment area, arable land (78%) and forests (16%) dominate [8]. The largest part of agricultural land is arable land beyond the reach of irrigation facilities (52%);

large areas are also orchards and plantations (11%), complex systems of arable plots (9%) and meadows and pastures (6%) [8].

**Figure 1.** Location of the study area, Bystra catchment, with marked main tributaries and their catchments (own study).

#### *2.2. Description of SWAT Model and SUFI-2 Model*

SWAT was used to model and examine the water balance of the Bystra river catchment area. SWAT is a model [9,10] developed by the USDA Agricultural Research Service [37]. The model operates on assigning one resource to another (physical, chemical, biological) using mathematical formulas that were developed to predict the impact of management practices on water efficiency and agricultural chemistry at the catchment scale [38,39]. We used the QSWAT3 v1.1 model with an interface in Quantum GIS 3.10.13 Coruna [40]. However, the calculations of the SWAT model were performed in the SWAT Editor on 10 December 2012 [41].

The water balance is the fundamental driving force behind all the processes that take place in the catchment area regardless of the choice of the SWAT model analysis. SWAT modeling for the catchment area is carried out in the land phase [42] and in the routing phase [43].

One of the formulas that is used in the SWAT model is the water balance equation:

$$SW\_t = SW\_0 + \sum\_{i=1}^{t} \left( P\_d - S \underline{I} \underline{R} \underline{Q} - E - w\_{sep} - GW \underline{Q} \right)^2$$

where: *SWt* is the final water content of the soil (mm); *SW*<sup>0</sup> is the initial water content of the soil (mm); t is the time in days; *Pd* is precipitation (mm); *SURQ* is surface runoff (mm); *E* is evapotranspiration (mm); *wseep* is the amount of water entering the wad zone from the soil profile (mm); *GWQ* is the groundwater flow (mm) [10].

Calibration and validation in the SWAT-CUP program are used to adjust the SWAT model to real conditions in the catchment area. The commonly used example of calibration is stream flow, which includes water balance processes. The calibration process is used to adjust the relevant parameters so that the simulated results are consistent with the observational data. Validation involves running the model using the parameters that were used during the calibration process. The purpose is to compare simulated results with observed data that were not used in calibration [44–46]. The SWAT-CUP program is used to analyze the uncertainty and sensitivity of the model [44,45] using the SUFI-2 algorithm, also used in small catchments [44,47,48].

#### *2.3. Application of SWAT and SUFI-2*

To simulate the water balance in the SWAT model, data were obtained from many sources (Table 1), which were used to build the SWAT model.


**Table 1.** Input data used in SWAT model (own study).

The SWAT model generated for this study consists of 31 generated partial catchments (Figure 1) [8]. The soil map was developed on the basis of digital soil and agricultural maps (scale 1:25,000 and 1:100,000) and geological data describing lithology. Descriptive soil data were collected within the statutory research projects of IUNG-PIB. Available water capacity and wilting point values were obtained from the study "Assessment of water retention in

soil and the risk of drought based on the water balance for the Lower Silesian Voivodeship", which was developed in 2013 by the employees of the Department of Soil Science, Erosion and Land Protection of IUNG-PIB in Pulawy [1].

The land use map was developed on the basis of Corine Land Cover maps with additional vectorization of land cover and land use using an orthophoto-map and Open Street Map data.

Based on the generated maps of soils, lands and slopes, 484 HRU (Hydrological Response Units) areas were created. When creating HRU areas, the land cover class of agricultural areas beyond the reach of irrigation CRDY was additionally separated with WWHT winter crops (43%), BARL spring crops (31%), CANP rape (14%) and other CRDY (12%) [58]. APPL apple orchards were separated from the land use class of ORCD [58]. On the other hand, forests were divided into coniferous FRSE forests (49%), deciduous FRSD forests (13%) and mixed FRST forests (38%) [59].

After generating HRU areas, the following meteorological data were used in the SWAT model: daily precipitation totals (mm); daily minimum and maximum air temperature ( ◦C); average daily wind speed (m/s); average daily relative humidity; daily sums of total solar radiation (MJ/m2) (Table 2) [8].

**Table 2.** Meteorological data for the Bystra catchment [8].


In the SWAT model, the parameters related to the point discharge of sewage, as well as for water bodies located outside the river network, for water bodies, rivers, and parameters for planned non-irrigated arable land management operations (WWHT, BARL, CANP, CRDY) were supplemented and corrected. The current value of CO2 concentration was also entered.

In the next stage, the SWAT model simulation was run for the period of 2010–2017 in a monthly step, with a five-year model start-up period.

Then, calibration and validation of the obtained SWAT model for the Bystra catchment area [8] was performed using the SWAT-CUP program. To obtain a more accurate coverage of the model with reality, the average monthly flow velocities (m3/s) obtained under the statutory projects of IUNG-PIB, obtained near the mouth of the Bystra River to Vistula for 2010–2014 (calibration) and 2015–2017 (validation), were used. A five-year warm-up period was used. Calibration and validation were performed in a monthly increment. This resulted in parameter ranges that fell within the ranges of calibration and validation accuracy [44,60,61]. The NSE coefficients (calibration: 0.58; validation: 0.70) and R2 (calibration: 0.60; validation: 0.71) for calibration and validation [8] were within the satisfactory ranges [60].

The results concerning the value of potential evapotranspiration were also analyzed with the results of the statutory service of IUNG-PIB implemented under the project Agricultural Drought Monitoring System [62]. It was found that the SWAT model for the Bystra catchment area accurately reflects the potential evapotranspiration in the study area.

Additionally, the results concerning the soil water content were compared with the available values of water capacity and the wilting point, which were obtained from the study prepared in 2013 by the employees of the Department of Soil Science, Erosion and Land Protection, IUNG-PIB in Pulawy [1].

#### *2.4. Climate Change Scenarios*

The daily grid climate data used in the SWAT model were prepared and tested in the recent paper on SWAT model calibration in the Bystra catchment [8]. Three RCM (Regional Climate Models)—RACMO22E, HIRHAM5 and RCA4—were selected for further study. They were selected to cover the range of the available two climate scenarios RCP (Representative Concentration Pathways) in terms of temperature increase and precipitation— RCP 4.5, RCP 8.5 (Table 3)—reflecting extreme and average variants of climate change, hence covering the widest range of uncertainty about possible scenarios (three RCM × two RCP). Most of the data were obtained at a spatial resolution of 0.11 degrees from the EURO-CORDEX database for the years 1951–2050 (widely available via the ESGF—Earth System Grid Federation, https://esgf-data.dkrz.de/search/cordex-dkrz for Europe) (accessed on 3 March 2021) [18,63].

**Table 3.** Description of GCM/RCM simulation with its division depending on radiative forcing. Comparison of temperature and precipitation changes in 2021–2050 in GCM/RCM simulation for RCP 4.5 and RCP 8.5 to the base period 1971–2000 (own study).


Climate scenario daily meteorological derivatives (minimum and maximum daily air temperature, daily precipitation, solar radiation, daily average wind speed, relative humidity) are based on the RCM for two RCPs (three RCM x two RCP). The RCMs are powered by one GCM (General Circulation Model): EC-EARTH. The RCP corresponds to the radiative forcing values in 2100 compared to pre-industrial values of +4.5 W m−<sup>2</sup> (RCP4.5) while RCP8.5 to + 8.5 W m−<sup>2</sup> (RCP8.5) [18,64,65] (Table 3). Table 3 also presents the boundary values of changes in the characteristics of selected models for the period 2021–2050 in relation to the period up to the base period 1971–2000.

Climate projections that were used in the SWAT model were extracted from grid cells that correspond to weather stations' location. Air temperature and precipitation data were additionally corrected by the SMHI (Swedish Meteorological and Hydrological Institute) using the DBS (Distribution-Based Scaling) method [48] and regional MESAN reanalysis (MESoscale Analysis) for the 1989–2010 dataset [66]. The data used were taken in a rotated polar grid. Therefore, we used bilinear interpolation to remap the dataset to a common latitude/longitude grid. CDO (Climate Data Operators) software [67] was used for this purpose.

For the analysis of the climate projections (RCP 4.5.1, RCP 8.5.1, RCP 4.5.2, RCP 8.5.2, RCP 4.5.3 and RCP 8.5.3), one iteration in SWAT-CUP was used for the set of the best calibration parameters for the years 2021–2050 in the prepared scenarios (Table 3) [8]. In the RCP 4.5 and RCP 8.5 scenarios, CO2 concentrations were changed for the periods 2021–2030, 2031–2040 and 2041–2050, developed by the Potsdam Institute for Climate Impact Research [68,69].

#### *2.5. Climate Change Adaptation Scenarios 1–5*

For the main purpose of this article, 5 scenarios for the adaptation of agriculture to climate change were prepared, which assume changes in land use (adaptation scenario 1 and 2) and protective measures (adaptation scenario 3, 4, 5) in the area of the Bystra catchment. The first adaptation scenario (AS-1) assumes an increase in afforestation on soils from the agricultural usefulness complex of soils 6 (temporarily too dry), 7 (permanently too dry) and 8 (temporarily too wet). The second adaptation scenario (AS-2) assumes the creation of a forested buffer for the Bystra River and its tributaries. The third adaptation scenario (AS-3) shows one of the erosion prevention practices at the riverbed, the so-called filter strips. The fourth adaptation scenario (AS-4) assumes the reduction in plowing on agricultural land. The fifth adaptation scenario (AS-5) involves increasing soil organic carbon to 2%. Adaptation scenarios are aimed at checking the possibility of increasing the soil water content in the 2041–2050 perspective. In doing so, the effects of adaptation scenarios on total runoff, sediment yield and actual evapotranspiration were also checked.

In the zero scenario (S-0), the Bystra catchment area is dominated by agricultural land (78%) and forests (16%). The largest part of agricultural land is arable land beyond the range of irrigation facilities (52%); a large area is also orchards and plantations (11%), complex systems of cultivating plots (9%) and meadows and pastures (6%) (Table 4). For adaptation scenarios 1 (AS-1) and 2 (AS-2), there will be changes in land use compared to scenario 0 (S-0), which are described later. In contrast, adaptation scenarios 3 (AS-3), 4 (AS-4) and 5 (AS-5) remain unchanged in terms of changes in land use.

**Table 4.** Division of the land cover and land use as well as the percentage of land use in the Bystra catchment generated in the QSWAT interface. CLC code 112–142 means artificial surfaces; code 211–243 means agricultural areas; code 313–324 means forest and semi natural areas; code 411 means wetlands, and code 511 is water bodies (own study).


In the first adaptation scenario (AS-1), the land use on all soils of complexes (representing soil habitats in Polish soil-agricultural mapping)—6 (semi-dry), 7 (permanently dry) and 8 (semi-wet) (6Bw-pgl.ps, 7Bw-ps, 8A-l)—was changed to mixed forest. The soils where the land use was changed are described in more detail in Table 1 of the publication on the water balance of the Bystra catchment [8]. Replacement of the above-mentioned soils is made through delineating the ranges of these soils on the land use maps and changing the attributes to mixed forests. After this change, afforestation in the Bystra catchment area increased by 3.31% (Table 4).

In the second adaptation scenario (AS-2), a forested buffer strip 80 m wide along the bank of the Bystra River was created and a smaller buffer strip 50 m wide for its tributaries [70–72]. The creation of buffer zones by rivers consisted of deleting the ranges of buffer zones on the land use maps and changing the attributes to mixed forests. The afforestation area compared to the zero scenario increased by 1.03% (Table 4).

In the third adaptation scenario (AS-3), filter strips were used, which are one of the protective measures used to drain water slowly from the field, thanks to which larger particles, including soil and organic material, may be deposited [73].

Filter strips [9,74] are areas covered with vegetation that are located between surface water bodies (rivers, ponds, lakes) and arable land, pastures and forests. They are generally found in areas where runoff leaves the field to filter sediment, organic material, nutrients and chemicals from the runoff. Filter strips are also known as vegetative filters or buffer strips. Due to the retention of sediment and the establishment of vegetation, nutrients can be absorbed into the sediment that settles and remain in the field landscape, making it possible for plants to take it up [73].

A protective treatment is also tillage without plowing [73], which is the fourth adaptation scenario (AS-4).

Plowing is defined as the mechanical disturbance of soil for crop production that has a significant impact on soil properties such as soil water behavior, soil temperature, infiltration and evapotranspiration [75]. In the long term, tillage can lead to soil degradation [76]. An alternative to traditional plowing is protective treatments (tillage without plowing, minimal mechanical disturbance of the soil) which consist of maintaining the surface soil cover by retaining crop residues. Retention of harvest residues protects the soil from direct exposure to raindrops and sunlight, while minimal soil disturbance improves soil biological activity and air and water movement in the soil [75].

No plowing cultivation was implemented in WWHT, BARL, CANP and CRDY arable land and simulated in SWAT.

In the fifth adaptation scenario (AS-5), the soil organic carbon content was increased from 1% to 2%. The original soil organic carbon values were studied as part of IUNG-PIB statutory research [8]. Soils in Poland are characterized by low soil organic carbon content. According to the European Soil Bureau (ESB), an organic carbon content of about 1% (Bystra catchment area) is a very low or low value [77]. The decrease in organic matter in soils and the associated decrease in organic carbon content result in increased CO2 emissions (exacerbating the greenhouse effect). The opposite situation, i.e., sequestration of CO2 in the soil, causes carbon to bind to soil organic matter for a longer period of time. Particularly large amounts of carbon are stored in peats, organic soils and organic-mineral soils [77].

#### **3. Results**

Section 3.1 describes the analysis of soil water content in S-0 for the period 2041–2050.

For the 10-year period (2041–2050), Table 5 presents a comparison of the seasonal soil water content in the Bystra catchment for each climate projection GCMs/RCMs under the RCP 4.5 and RCP 8.5 climate scenarios.

For the 10-year period (2041–2050), Figure 2 shows the average soil water content (1.5 m) for each season of DJF, MAM, JJA, SON for the GCMs/RCMs climate projections under the RCP 4.5 and RCP 8.5 climate scenarios, while Figure 3 shows the spatial comparison of average soil water content in 31 sub-catchments for the SWAT simulation period 2010–2017 and 2041–2050 for the GCMs/RCMs climate projections under the RCP 4.5 and RCP 8.5 climate scenarios.

Section 3.2 describes the climate change AS-1, AS-2, AS-3, AS-4, AS-5 analysis for the period 2041–2050.

For the period 2041–2050, Table 6 presents a comparison of AS-1, AS-2, AS-3, AS-4, AS-5 with respect to S-0 for seasonal soil water content in the Bystra catchment for the RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, and RCP 8.5.3 projections.

**Figure 2.** Seasonal average soil water content (1.5 m) for 2041–2050 and for the SWAT 2010–2017 model for individual climate projections RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3 (own study).

**Figure 3.** Comparison of average soil water content in 31 sub-catchments during the SWAT simulation period 2010–2017 and 2041–2050 for individual climate projections RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3 (own study).

**Table 5.** Comparison of average soil water content by season for the SWAT 2010–2017 simulation period with climate projections (RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3) for the years 2041–2050 in the Bystra catchment. Bold numbers indicate soil water content, while shaded numbers indicate percentage change (red is % decrease in content; blue is % increase in content). Dark red and dark blue shading means large changes, while light red and light blue shading means small changes (own study).


**Table 6.** Comparison of average soil water content by season between scenario 0 (S-0) and adaptation scenarios 1–5 (AS-1, AS-2, AS-3, AS-4, AS-5) for 2041–2050 in the Bystra catchment for climate projection RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3. Bold numbers indicate soil water content, and shaded numbers indicate percentage change (red indicates % decrease in content and blue indicates % increase in content). Dark red and dark blue shading indicates large changes, while light red and light blue shading indicates small changes (own study).


Table 7 presents a comparison of total runoff by season for S-0 and AS-1, AS-2, AS-3, AS-4, AS-5 for the period 2041–2050 in the Bystra catchment. Next, Table 8 compares sediment yields by season for S-0 and AS-1, AS-2, AS-3, AS-4, AS-5 for 2041–2050 in the

Bystra catchment. In turn, Table 9 compares actual evapotranspiration by season for S-0 and AS-1, AS-2, AS-3, AS-4, AS-5 for the years 2041–2050 in the Bystra catchment.

**Table 7.** Comparison of seasonal total runoff between scenario 0 (S-0) and adaptation scenarios 1–5 (AS-1, AS-2, AS-3, AS-4, AS-5) for 2041–2050 in the Bystra catchment for climate projections RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3. Bold numbers indicate soil water content, and shaded numbers indicate percentage changes (red indicates % decrease in content, and blue indicates % increase in content). Dark red and dark blue shading indicates large changes, while light red and light blue shading indicates small changes (own study).


**Table 8.** Comparison of seasonal sediment yield between scenario 0 (S-0) and adaptation scenarios 1–5 (AS-1, AS-2, AS-3, AS-4, AS-5) for 2041–2050 in the Bystra catchment for climate projections RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3. Bold numbers indicate soil water content, and shaded numbers indicate percentage changes (red indicates % decrease in content, and blue indicates % increase in content). Dark red and dark blue shading indicates large changes, while light red and light blue shading indicates small changes (own study).



**Table 8.** *Cont.*

**Table 9.** Comparison of seasonal actual evapotranspiration between scenario 0 (S-0) and adaptation scenarios 1–5 (AS-1, AS-2, AS-3, AS-4, AS-5) for 2041–2050 in the Bystra catchment for climate projections RCP 4.5.1, RCP 4.5.2, RCP 4.5.3, RCP 8.5.1, RCP 8.5.2, RCP 8.5.3. Bold numbers indicate soil water content and shaded numbers indicate percentage changes (red indicates % decrease in content, and blue indicates % increase in content). Dark red and dark blue shading indicates large changes, while light red and light blue shading indicates small changes (own study).

