Land use (LU) and atmospheric factors, such as precipitation and temperature, exert a huge impact on the amount of sediment yield on practically every basin. Studies conducted in many regions of the world [1
] have proven that surface runoff is a serious threat to soil resources in the world [8
]. This phenomenon is particularly dangerous in arable lands because the maintenance of long-term crop production depends on the soil’s production capacity, which is negatively affected by leaching of the topsoil and organic matter, and increased water outflow [10
]. Until recently, in temperate zones, frost and snow cover were playing a similar role during winter as vegetation cover for the rest of the year. Nowadays, even in mountainous areas, frost and snow are becoming rarer and the erosive effect of rain is increasing during this period [14
]. The problem is even more acute in sub-mountain and mountainous areas, where large slopes accelerate the leaching of soil particles. The correct identification of such areas is not simply easy because even relatively small river basins can display huge variability in terms of land and climate features that greatly affect sediment yield calculations [17
]. Forecasts indicate that both climate and LU change will be very dynamic this century [19
]. In turn, changes in LU will depend on local economics, population migration, arable land quality, and their location relative to urban and protected areas, as well as national and international policies [23
]. Both of these stressors will interact by summation or balancing, depending on the region, basin, or even sub-basin area. However, knowledge about their combined impact is still limited due to the small amount of comprehensive research [13
]. The mountainous area, where the Raba River basin is located, is very specific in terms of land use forecasts, meteorological phenomena, intensity of water erosion [31
], as well as activities to mitigate this process [32
]. Until now, most similar studies have been conducted in catchments where the main problem was the development of cities/agriculture. Meanwhile, forecasts for the Raba basin are quite opposite. Forests are slowly beginning to dominate the landscape of this area [33
], taking the place of agriculture, and the falling number of inhabitants contributes to a reduction of urban area impact on the environment, despite the proximity of a large urban agglomeration—the Krakow Metropolitan Area.
The goal of this study is to answer the question of whether LU changes will be able to compensate for sediment yield increase predicted by the future climate change scenarios on the individual sub-basin scale. The modeling tool used in this study, the Macromodel DNS (Discharge–Nutrient–Sea) combined with the SWAT module (Soil and Water Assessment Tool), enabled also the discussion on seasonal trends of this process.
The obtained Macromodel DNS/SWAT results enabled discussion of sediment yield temporal and spatial distribution in the Raba River basin. The detailed model response for each sub-basin for each of the scenario variables (temperature, precipitation, urban, and forest area predictions) and their combination have been included in the Mendeley Data [88
]. As for the baseline scenario, considered as the reference point for subsequent examination of the climate and land use changes, important differences between seasons and sub-basins have been revealed (Figure 8
, Table 4
). Highest average sediment yields have been observed for the spring period, both in the upper and lower parts of the basin (0.92 +/− 0.27 t/ha and 0.57 +/− 0.34 t/ha, respectively). They were followed by the summer sediment yield elevated values (0.58 +/− 0.30 t/ha and 0.41 +/− 0.23 t/ha, respectively) to reach minima during the autumn-winter period (at the range of 0.28–0.29 and 0.21–0.25, respectively). Although statistically significant differences (Kruskal-Wallis and Bonferroni tests, 95% confidence level) between the upper and lower parts of the Raba River basin have been detected only for the spring season, notably higher sediment yields in the upper part are also visible during the summer period. However, it should be noted that extremely high sediment yields have also been detected in one of the lower sub-catchments (No. 17).
The climate forecasts (Section 2.3.1
) imposed on the baseline scenario further emphasized differences between seasons and sediment yield origins. The implications of the RCP 4.5 and RCP 8.5 predictions for the two-time perspectives (2021–2050 and 2071–2100) have been assessed for individual sub-basins as a sediment yield change in relation to the baseline calculations (Figure 9
). The largest increases of the sediment yields have been observed in the upper part of the basin during the spring and winter periods. Particularly, in sub-basins No. 25, 30, and 32, where the predicted sediment yield is expected to grow by 0.15–0.30 t/ha (scenarios C1.1 and C2.2, respectively). The increase in sediment values has been also predicted for the selected lower sub-basins, especially during the winter period for No. 4 and 17 (0.21–0.26 t/ha). Sub-basin No. 17 stands out also during the summer period (scenario C1.1) displaying the biggest changes across the scenarios, from 0.21 t/ha in C1.1, to −0.18 t/ha in C2.2. Generally, the summer period is distinguished by a decrease of sediment yields, particularly noticeable in the C2.2 scenario, reaching values of −0.36 t/ha in sub-basin No. 25.
As for the land use scenarios (LU, Section 2.3.2
), a noticeable sediment yield reduction in the majority of the Raba River sub-basins has been observed when compared to the baseline scenario (Figure 10
). Substantial average decreases have been observed in the upper part of the basin, especially in sub-basin No. 34, during the spring and summer periods (0.22 +/− 0.09 t/ha and 0.13 +/− 0.06 t/ha, respectively). As for the autumn and winter periods, the impact of LU scenarios on sediment yields was much smaller and remained below 0.1 t/ha. However, the presence of areas without any changes in sediment yield under LU scenarios should also be noted. Such a situation concerned sub-basins No. 35 (Upper Raba), and No. 4, 9, 10, and 15 (Lower Raba) characterized by a very small share of forest area, or its complete lack of.
Since the assumed climate and land use changes are predicted to occur simultaneously, the final modeling approach focused on superimposition of previously described scenarios. As comparing seasonal average values for the upper and lower portions of the Raba River basin (Table 5
) the general decrease of the sediment yields for the upper part has been noted in the range of 0.03–0.25 t/ha, except for the winter period under all scenarios. The scenario combination for the lower part resulted in a slight increase of sediment yields (0.01–0.06 t/ha) for the majority of the modeled cases. However, an average decrease by 0.01 has been also observed under RCP 4.5 scenarios (C1.1+LU and C1.2+LU) during the summer and autumn periods. Particularly, a large decrease of the seasonal average value in this part of the basin has been detected for the summer period under long term RCP 8.5 predictions (C2.2+LU).
To answer the main question of this study, the results obtained for the combined climate change and land use scenarios were compared with the baseline scenario for each sub-basin, taking into account the increase or decrease in sediment yield. The negative results (sediment yield change below zero; green color Figure 11
) signify the ability of the LU scenario to compensate for the sediment yield changes induced by climate change. While, positive results (sediment yield change above zero; red color Figure 11
) indicate a lack of such a response in individual sub-basins. During the spring season, significant differences (Kruskal-Wallis and Bonferroni tests, 95% confidence level) have been detected between the upper and lower Raba River parts under all four scenarios. Moreover, in the majority of the upper sub-basins, the LU changes counterbalanced climate change effects, except for sub-basins No. 31, 33 and 35). The same pattern has been detected during the summer period with all the upper sub-basins showing positive effects of the LU changes on total sediment yield. Moreover, such an effect has been also observed for all lower sub-basins under the RCP 8.5 long-term scenario (C2.2 + LU), except for sub-basin No. 3. For the autumn months, the Macromodel predictions display again the ability of the LU changes to compensate for climate RCP 4.5 scenarios (C1.1 + LU and C1.2 + LU) for almost the entire Raba River basin. While for the winter period, the majority of sub-basins, in both parts, displayed positive sediment yield changes, signifying that future LU modifications will not counterbalance climate change effects.
The Raba River basin has been selected as a modeling venue to assess the combined effects of climate and land use change scenarios on sediment yield at the sub-basin level. This basin covers an area elevated from 145 a.s.l. to 1272 a.s.l., and therefore, provides a unique opportunity to study a wide spectrum of land and climate features. The baseline scenario for all the 36 sub-basins revealed high sediment yields for the upper part of the area, especially for sub-basins No. 29–32 (Figure 8
), where slopes exceed 25% (Figure 2
a), and agricultural use is about 40% on average in these sub-basins (Figure 3
). Although precipitation averaged 1400 mm in this part of the basin, clay soils dominated these sub-basins (Figure 2
c), and therefore, are less prone to water erosion and prevented over-excessive leaching of particles. The average sediment yields for these sub-basins reached 1.38 t/ha, while the highest value (1.55 t/ha) was detected at sub-basin No. 17, located in the lower part of the Raba River basin. Although only 10% of its area is covered by high slopes (>25%) and average rainfall reaches 1000 mm, 80% of this area is used for agricultural activities (with 20% covered by potato crops), its soil cover (light dusty clay, light loamy sand, and dusty soil; Figure 2
c) appears to be more vulnerable to water erosion. As a combination of land features can account for the record sediment yields, the impact of climate attributes is more visible when seasonal differences are discussed (Figure 8
). High levels of precipitation and torrential rainfall during the spring-summer period amplify sediment yield differences between the upper and lower parts of the basin. When a decrease in heavy rain frequency and intensity is observed during autumn, the sediment yield values are considerably reduced, to reach a minimum during winter due to soil freezing, snow cover presence, and low precipitation. Comparing the obtained results to previous analyses related to sediment load and its seasonal variability in the closing profile of the upper Raba catchment [44
], it can be concluded that sediment loads and sediment yields do not coincide in particular seasons. As observed previously, the winter sediment loads can reach elevated values, while the winter yields are low (Figure 8
). Such differences between phenomena occurring on the catchment surface (yields), and in the river itself (loads), could be associated with the specificity of the Carpathian catchment where deposition processes intensively occur; additionally supported by numerous, but mostly overfilled and improperly managed, anti-erosion structures (Figure 1
]. They temporarily interrupt the journey of large amounts of sediments washed away to surface waters during the spring and summer precipitation increase. Therefore, during autumn, deposition rapidly increases sediment amounts in such structures. While in winter, along with the frequent occurrence of so-called anomalously wet seasons (AWS) and the local flow increase [72
], previously stored sediments begin to move down river. In addition, the erosion of the riverbed is rapidly increasing behind the sediment barriers. It causes an increase in sediment load especially in sub-basins No. 20, 26, 27, 30, 32, and 34.
The spring sediment yield maxima are supposed to be further exacerbated under the adopted climate change scenarios in both time perspectives (Figure 9
). Since the predicted temperature changes show a similar increasing pattern for both parts of the basin (Figure 5
), this change will be provoked mainly by precipitation increase, especially in the long-term perspective (2071–2100) (by over 25% for RCP 8.5). This is particularly evident in sub-basins No. 25, 30, 32, and 33 where spring sediment yield increased by 0.33 t/ha, compared to the baseline scenario. A similar or even higher precipitation increase has been predicted for the winter season. Thus, the winter sediment yields will require particular attention in this basin. While the forecasts indicate a large increase in rainfall in winter and spring, this increase will be much smaller during summer, and in the case of the long-term forecasts for the Upper Raba, rainfall will decrease even by 3% (Figure 5
). Such a summer decrease is characteristic of all mountain areas in this region of Europe, and the higher the analyzed sub-basin is located above sea level, the higher decrease of precipitation should be expected [78
]. Therefore, the summer sediment yield decrease is noticeable in many Raba River sub-basins (Figure 9
), especially for the long-term RCP 8.5 prediction, with the highest decrease of 62 t/ha in sub-basin No. 31.
In land use scenarios (LU), in which the increase of forest and urban areas was predicted to occur at the expense of arable lands, the maximum sediment yield reduction is expected in spring and summer (Figure 10
). The analyses showed that even a 15% increase of the residential medium- and low-density areas has a negligible impact on the size of the sediment yield. Therefore, its reduction will result almost exclusively by the growth of forest areas [91
]. According to the modeled scenario such changes will be noticeable in the entire Raba River basin, but the most intense changes will concern its upper part (Figure 6
). Particularly, sub-basins No. 24–28, 30, 32, 34, and 36 will display a reduction in spring and summer sediment yields, reaching even 0.37 t/ha (sub-basin No. 34). The predicted 30% increase of forest areas may therefore reduce sediment yields by more than 35% in selected sub-basins, which proves a very good effectiveness of the model’s forest soil loss mitigation function.
The concurrent discussion of how the combined climate and land use changes impact sediment yields on the basin scale is still not very common in scientific publications. Moreover, due to variability of precipitation, temperature, and land use projections, related to geographic and economic factors, it is difficult to indicate the uniform relationship between these projections and basin response. Nevertheless, the results from very contrasting areas in China [17
], the USA [25
] or the UK [18
] indicate that most likely changes in land use are the key driving force of changing sediment yield. Such a pattern is also clearly visible in the Raba River basin. However, its extent varies greatly in both parts. In its upper, the applied liberal land use forecast, i.e., the increase in forest area by as much as 30%, generally compensated for the sediment yield changes induced by the climate predictions (Figure 11
), except for winter. Even in case of the spring months, when sediment production is abundant in sub-basins with high slopes, intense agriculture, and exposed to high precipitation, the yield reduction can reach 10.5–23.2% on average, depending on the scenario. Since the predicted temperature increases are almost uniform among the seasons, the precipitation variations should also be taken into consideration when the effectiveness of land use changes is further discussed. Especially, when the RCP 8.5 long-term summer precipitation decrease, causing over 40% reduction of the sediment yield, is followed by the autumn precipitation increase (Figure 5
). This situation results in sediment yield growth which cannot be further attenuated by the forest area increase (sub-basins No. 24, 26, 28, 31–33, and 35). As for the lower part of the Raba River basin the opposite trend could be observed. In the majority of the sub-basins, the land use changes did not compensate for the sediment yields induced by climate change, except for the selected scenarios when only minimal rainfall increases were predicted (summer under C1.2 scenario, and autumn under C2.1, and C2.2; Figure 5
). Since, this part of the basin has mainly an agricultural character, with fertile soils and highly prone to erosion (loess), the predicted growth of the forest area share will not be effective to balance out washing of sediments.
The most unique situation in the studied basin is, however, created by the winter predictions. The obtained results clearly show that regardless of the chosen scenario, the imposed land use changes will not compensate for the sediment yields induced by the climate predictions in the majority of the sub-basins. Although the lack of plant cover during this period, protecting soil particles against washing out during the vegetation period, is usually accounted for by this phenomenon [16
], meteorological conditions should also be taken into consideration. The predicted temperature increase during the winter months will reduce the time-span of snow accumulation in the higher altitudes and eventually will replace snowfall with rainfall in the entire basin. This situation, combined with the lower infiltration of soils and energy-limited evapotranspiration in low temperatures will increase runoff induced soil leaching [96
]. The final response of the upper part of the Raba River basin shows a distinct increase of sediment yield, especially in the long-term forecasts, reaching 23–30% on average. While, in the lower part of the basin, such an increase is less pronounced (16–18%). As for the sub-basins displaying that land use changes could still be effective in attenuating sediment yield increase induced by climate change (Figure 10
), it should be noticed that both sub-basin groups differ notably in terms of their response to climate change. As visible in Figure 9
, the selected upper part sub-basins (No. 21, 23, 24, 34, and 36) display a relatively small sediment yield increase when compared to the rest. Since the sediment yield changes exerted by the land use are uniform for the entire basin, therefore, the impact of climate-induced change will be more decisive. Likewise, in the lower part, the distinctive sub-basins (No. 2, 6–8, 11–12, and 15) are marked by relatively low sediment yield change, and subsequently is even lowered when the land use scenario is applied. Therefore, it can be concluded that while from spring to autumn land use changes have a decisive impact on sediment yields, winter climate changes exert greater importance.
The example of the Raba River (Carpathian Mts., Poland) demonstrates that even relatively small river basins can display huge variability in terms of land and climate features, which greatly affect sediment yield calculations. Therefore, it seems to be very purposeful to divide the modeled area into smaller units (sub-basins), which could bring even more specific answers to factors controlling the overall basin sediment leaching response. This may especially occur when the river basin can be divided into parts with contrasting features, which here encompass the Carpathian upper part and the sub-montane lower portion of the Raba River basin. The general land use forecasts for this region predict gradual afforestation of the studied basin at the expense of agricultural areas, and a reduced impact of urban areas due to decreasing number of inhabitants. However, the extent of these phenomena is much more visible in the upper part of the basin. Moreover, the temperature and precipitation predictions display noticeable differences, with higher variability of changes for the mountainous part of this area, which is specific to the Carpathian Mts. The response of 36 individual sub-basins to climate scenarios created a mosaic of negative and positive sediment yield changes in comparison to the baseline scenario [88
]. The overlapped land use predictions allowed us to indicate those sub-basins where land use changes could balance out sediment yields under climate change predictions, and those where it will not be possible. The general response for the combined scenarios revealed that sediment yields could be altered even by 49% in the selected upper sub-basins during the spring-summer months, while for the lower sub-basins the predicted changes would be less effective, up to 3% on average.
Since these effects are based on the assumed replacement of agricultural areas by forests, it should be noted that it will be a continuously progressive process and not necessarily feasible in all parts of the basin. Moreover, future research should focus on simulations relating the sediment yield to the forest growth and structure. In addition, it must be remembered that replacing agriculture with forest areas is possible in practice, in a limited area. While it is possible on weak and infertile soils on high slopes, this solution will not be possible on fertile lowland catchments due to economic issues. The seasonal changes taken into consideration here also prove that special attention must be paid to the winter months. Under the adopted climate changes the frost and snow cover protecting soils against erosion will become exceptional even in the mountainous part of the basin, while completely disappearing in the sub-mountain part in the long-term perspective. These phenomena will also significantly affect the soil particle transport within the studied basin. It seems to be therefore extremely important to underline that sediment transport modeling, based on the averaged values of temperature, precipitation, and land use changes, can lead to significant errors in the scale of the entire basin. Therefore, it is highly recommended to individualize forecasts at the sub-basin level, as performed here. Moreover, the extent and magnitude of land-use and land management practices vary depending on the needs of local communities, which also prompts a downscaled assessment based on modeling tools. The approach proposed in this current article is the first of its kind in the Carpathian Mts., and must be continued with further analyses for the mountain/sub-mountain basins.