**2. Materials and Methods**

Field studies were conducted in the years 2014 and 2019 on the Wełna River, located in the belt of central European lowlands (Figure 1a). The Wełna River is a right bank tributary of the Warta River (ranking third in Poland in terms of river size). The Wełna River is 118 km in length, and its catchment area is 2621 km2. It flows from a lake located 10 km north-east of the town of Gniezno. Analyses were conducted on a selected river reach of 250 m in length, situated near the town of Coto ´n. The study object is between a tree-covered area and an area with no tree or shrub cover, no water bodies, and no hydraulic structures, and there is no inflow from another watercourse. The investigated cross-sections no. 1 and 2 were shown on Figure 1a. In this way, any potential effect on disturbance of flow between the investigated cross-sections was eliminated (Figure 1b, Figure 1c).

**Figure 1.** Location of the studied site in: (**a**) the digital elevation model (DEM), (**b**) the digital surface model (DSM), and (**c**) an aerial photograph.

Field measurements consisted of the assessment of watercourse hydrometric parameters. Within this study, the leaf area index (LAI) was also determined for vegetation in a tree-lined river reach, while the number and species composition of macrophytes found in the investigated river cross-sections were recorded [44]. Based on field measurements, the volume of flow and velocity distributions in cross-sections were determined along with the substrate grain-size characteristics and river bottom morphology. Moreover, a model was constructed to illustrate variability in the shade range within a river reach lined with trees. The shading analysis was based on DEM and a DSM developed on airborne LIDAR data. The results of this study enable us to establish the relationship between river shading by vegetation covering the bank zone and changes in the hydromorphological parameters of the river channel.

Based on LIDAR data, a DEM of the investigated river reach was developed (Figure 1a), while the location of trees was given in the DSM (Figure 1b). The current vegetation status in the analyzed Wełna River reach is presented in an aerial photograph (Figure 1c). Based on the forest stand description and a forest map, on the right bank of the river reach there is a forest subcompartment of 0.17 ha. Black alder (*Alnus glutinosa*) is the dominant species. The stand is approximately 70 years old. Common sallow (*Salix cinerea*) is the shrub species found in that area. In turn, a forest subcompartment of 7.7 ha is located on the left bank of the river. That subcompartment is a belt of swamps adjacent to the river at a length of approximately 50 m.

Another subcompartment of 12.35 ha is located farther from the riverbank. It is a mixed forest with the predominance of oak (*Quercus* L.) and pine (*Pinus sylvestris* L.). Two measurement cross-sections were established in the selected river reach. Cross-section 1 was situated in a location with no trees or shrubs, while cross-section 2 was located in a site with banks lined with trees and shrubs. The distance between cross-section 1 and cross-section 2 is 93 m. The locations of the cross-sections are marked in the presented maps (Figure 1). Figure 2 gives available aerial photographs of the investigated Wełna River in the years 2009–2019. They show an unchanged status of tree cover in this river reach.

**Figure 2.** Changes of land cover and shading of the watercourse in different periods and years (images obtained from GoogleEarthPro).

Flow measurements were taken in the individual cross-sections. For this purpose, a miniature ADCP StreamPro stream flow meter was used. The ADCP StreamPro instrument measures distributions of velocity, discharge intensity, and river bottom geometry. It operates based on the emission of acoustic signals.

Waves reach suspensions floating in water and are reflected by them, as a result returning to the emitter. The instrument processes thus obtained data and next determined the hydraulic parameters of discharge in the river bed. Analyses of river channel geometry consisted in measurements of the location and ordinates of points on the riverbanks and the river bottom using a GPS real-time kinematic (RTK) device accurate to ± 3 mm for vertical measurements.

In order to determine the sediment composition analyses were conducted on the material collected from the river bottom. The grain-size distribution was analyzed using the sieve method according to the PN-B-00481:1988 (Polish standard) guidelines. To determine the grain-size composition, dried and ground samples were passed through a set of sieves (0.063, 0.125, 0.25, 0.5, 1.0, 2.0, 3.15, 4.0, 6.0, 8.0, 10.0, 16.0 mm). Samples were dried in 105–110 ◦C. In order to determine organic matter content, the dried samples were precisely weighed accurate to 0.01 g. Next, they were again subjected to heat—they were incinerated (600–800 ◦C) in a furnace. In this way the weight of the prepared samples made it possible to determine the content of organic matter. Samples were sieved again. Thus, the contents of minerals could be determined more precisely, excluding organic matter. Based on the recorded results, curves for fraction contents were plotted, and grain-size index (*Cu*) and grain-shape index (*Cc*) were calculated.

Thus, collected data were implemented in a graphic computer program, in which the shape of the cross-sections was plotted. Flow velocity was determined using an ADCP StreamPro device [45], applying the phenomenon of sound wave propagation in the water environment. Measurement results were used to determine flow values and to plot river channel cross-sections; next, wetted perimeter *O*, hydraulic radius *Rh*, and water table gradient *I* were calculated. At cross-section 2, with dense tree and shrub cover, regular measurements were taken for LAI on both riverbanks. These measurements were recorded during the vegetation season using a LAI 2000 device. The results provided the basis for an assessment of area cover by the tree canopy (including leaves). In this manner information was obtained on the availability of light-reaching plants, including macrophytes, at the ground level.

In the next stage, model calculations for solar radiation were made in the ArcGIS environment. The input data for the analysis were the Digital Surface Model (DSM) (Figure 1a). The DSM model was developed on the basis of the LIDAR data. The DSM is characterized with spatial resolution of 1 m and was provided by the Head Office of Geodesy and Cartography, Poland. The DSM contains information concerning objects on the surface, such as vegetation and buildings, which are crucial for shading analysis. The area for analysis was extended to make sure all of the shadows cast on the riverbed were included in the calculations. Solar radiation was calculated using the Area Solar Radiation tool. The output of the tool was a global radiation raster calculated for the whole year, vegetation period, and separate months from April to September with a one-hour interval in WH·m<sup>−</sup>2. The results were assigned to the points of the cross-sections in the river.

#### **3. Results**

Results of flow measurements recorded using a miniature ADCP StreamPro flow meter were analyzed using the WinRiverII program. On this basis, values of flow, river bottom structure in the cross-sections, and the distribution of velocity were determined. Using the ADCP results and geodesy measures, graphs were prepared for the river cross-sections (Figure 3, Figure 4).

For the purpose of further calculations based on the above figures for each cross-section, their wetted perimeter, *O*, and cross-section, area *A*, were calculated. In the course of the measurements of water table ordinates, it was found that the water table gradient is so small that it is impossible to determine the slope at the river reach between the cross-sections. In order to determine slope, precise levelling was performed and the distance between the recorded measurements of water table height was increased to 380 m. At the river reach of that length, the mean hydraulic gradient was *I* = 0.32%.

**Figure 3.** Bed changes in cross-section no. 1 (in an area with no tree cover—with submerged and emergent vegetation).

**Figure 4.** Bed changes in cross-section no. 2 (in a forested area).

In Figures 3 and 4 the cross-section was plotted for the status in 2014 and 2019. In the course of the study the cross-sections were also measured using a range pole. Cross-section 2 is wider, but slightly shallower than cross-section 1, in which, due to the bank overgrowth, flow is concentrated in the stream centerline of the watercourse. As a result of the large deposition of organic sediments the cross-sections measured using a range pole are slightly deeper (as they were probed to reach the so-called hard bottom). In this manner, the thickness of organic sediments deposited in the river channel could be assessed. Sediment thickness in cross-section 2 was much greater than in cross-section 1, with a maximum of 0.85 m. In contrast, in cross-section 1, the sediment thickness did not exceed 0.4 m. The analysis of the cross-sections also shows that within the previous five years channel shallowing was observed over the entire river reach. This trend is particularly evident in cross-section 2, where the bottom in the riverbank zone was elevated by as much as 0.6 m. Also, in cross-section 1 channel shallowing by 0.4 m was recorded. River bottom elevation and shallowing of the cross-sections may be connected with small hydraulic gradients in the investigated river reach, which is reflected in small flow velocities, at mean values for both cross-sections ranging from 0.06 m/s (cross-section 1) to 0.07 m/s (cross-section 2).

This is related, among other things, to vegetation overgrowth in the river reach in the vicinity of cross-section 2, being the downstream measuring site for the investigated reach. Thus, in the reach the predominant processes are the accumulation of sediments, to a considerable extent composed of organic matter.

A large proportion of the riverbank vegetation (trees and shrubs), i.e. those that are casting shadow over the entire river channel, causes shading, eliminating any aquatic vegetation within the river channel and banks. On the other hand, we may observe considerable accumulation of organic matter in the river reach. This also results from the small flow velocities. In cross-section 1, with the greatest share of riverbank and bottom vegetation, respective changes were observed in flow conditions. The flow velocity concentrates in the stream centerline zone, which also results in intensified erosion processes. This cross-section is the most compact.

Values of LAI were measured on the riverbanks at the tree-lined reach immediately above the ground. LAI is a measure for the total area of leaves per unit ground area, and it is directly related to the amount of light that can be intercepted by plants [44]. It is an important variable used to predict photosynthetic primary production and evapotranspiration, while it is also as a reference tool for crop growth. As such, LAI plays an essential role in theoretical production ecology. An inverse exponential relation has been established between LAI and light interception, which is linearly proportional to the primary production rate. Following the adopted methodology, the first measurement was taken in an open area with no vegetation (the reference measurement), while the next 12 were taken at the riverbanks, where trees and shrubs were found. For the right bank, the averaged LAI value was 3.03 [m2/m2], whereas for the left bank it was 3.06 [m2/m2]. The values indicate dense overgrowth reducing access to light in the investigated cross-section.

In each of the cross-sections the river bottom was raked using garden rakes in order to identify species of the bottom vegetation. The belt of the raked river bottom was approximately 1 m wide. Plant samples were weighed using scales. For cross-section 1 the obtained plant sample was approximately 1.5 kg. In cross-section 2 no bottom vegetation was found due to considerable shading. Samples of floating vegetation and plants growing on the banks were also collected. Identified plant species in the individual cross-sections are listed in Table 1.


Debris samples were collected from the river bottom. Substrate samples were subjected to sieve analysis in order to determine the grain-size composition. The sieve analysis was performed for samples from cross-sections 1 and 2. On this basis the uniformity index *Cu* was determined:

$$\mathcal{C}\_{\mu} = \frac{d\_{60}}{d\_{10}}\tag{1}$$

along with the curvature index *Cc*:

$$\mathcal{C}\_{\mathfrak{c}} = \frac{{d\_{30}}^2}{{d\_{10}}^2 d\_{60}} \tag{2}$$

where: *d*10, *d*<sup>30</sup> and *d*<sup>60</sup> are diameters of particles, which together with smaller ones, account for 10, 30, and 60% of soil mass, respectively (Table 2).


**Table 2.** Debris parameters.

For cross-section P1 the uniformity index *Cu* value classified the substrate as a grained substrate (1≤ *Cu* ≤ 5). The curvature index *Cc* indicates a poorly grained substrate. For cross-section P2 the recorded *Cu* value corresponds to a uniformly grained substrate. The *Cc* value corresponds to the values from the range for well-grained soils (*Cc* = 1–3). Results obtained after five years are similar and indicate a certain balance of river channel forming processes. Individual samples were incinerated in a furnace to determine organic matter content (MO). The following results were recorded: cross-section P1–2.4% MO (2014) and 4.1% (2019); cross-section P2–26.5% MO (2014) and 31.9% (2019).

The percentage share of organic matter in the sample from cross-section P2 is considerably greater compared to the sample from cross-section P1. This results from the fact that sample 2 was collected from the cross-section, in which trees and shrubs were growing on both banks, while dying plant parts accumulated on the watercourse channel bottom. The increase of organic matter content after five years indicates intensive accumulation of organic substance in the investigated cross-sections of that river.

The mean roughness coefficient was obtained using the results of hydrometric and geodesy analyses, applying the Manning formula for individual cross-sections. The results for the calculations of measurements both from 2014 and 2019 are presented in Table 3. When field measurements were taken, the water table was found in the main river channel. In cross-section 1, aquatic vegetation was growing on the channel banks and the river bottom. Paradoxically, the values of the roughness coefficient for that cross-section are lower than those obtained for cross-section 2, in which no aquatic vegetation was found either on the bottom or in the bank zone. This results from the parameters of that cross-section (greater width) and the damming effect due to the cross-section with a high proportion of vegetation, found in the immediate vicinity. This results in a reduced flow velocity, which under the assumption of the constant averaged value of the hydraulic gradient provides a higher value of the roughness coefficient. The resulting roughness coefficients are typical of overgrown river channels, in which the presence of vegetation drastically reduces the flow capacity of the river channel (Table 3).


**Table 3.** Results of calculations of roughness coefficient.

Solar radiation was calculated using the Area Solar Radiation tool in ArcGIS 10.5. The results were assigned to the points of the cross-sections of the river in the monthly configuration (Figure 5) as well as the vegetation period (Figure 6a) and annual configurations (Figure 6b).

Calculations showed that solar radiation in the annual period at cross-sections no. 1 and no. 2 amounted to 847 <sup>×</sup> 10<sup>3</sup> and 135 <sup>×</sup> 103 WH·m−2, respectively (Table 4). In the analyzed river reach, total solar radiation varied, and ranged from 50 <sup>×</sup> 103 to 857 <sup>×</sup> 103 WH·m−2. From point no. 11, solar radiation was greater and ranged from 514 to 857 <sup>×</sup> 10<sup>3</sup> WH·m−<sup>2</sup> with the mean value of <sup>743</sup> <sup>×</sup> <sup>10</sup><sup>3</sup> WH·m<sup>−</sup>2, which was connected with the exposure of the river channel. In the tree-lined river reach, total radiation ranged from 50 to 512 <sup>×</sup> <sup>10</sup><sup>3</sup> WH·m−<sup>2</sup> with the mean of 245 <sup>×</sup> 103 WH·m<sup>−</sup>2. Results of calculations for individual months and the vegetation period are presented in Table 4. The value of radiation in the period from April to September ranged from 38 <sup>×</sup> 103 WH·m−<sup>2</sup> at point

no. 6 to <sup>687</sup> <sup>×</sup> 103 WH·m−<sup>2</sup> at point no. 13. In cross-section no. 2, monthly values of solar radiation in the period from April to September ranged from 14 to 21 <sup>×</sup> 103 WH·m−<sup>2</sup> at the total value of <sup>110</sup> <sup>×</sup> <sup>10</sup><sup>3</sup> WH·m<sup>−</sup>2. In turn, in cross-section no. 1 solar radiation in individual months was almost fiveto over eightfold higher, with a mean of over sixfold higher. Greater insolation in cross-section no. 1 contributed to more intensive growth of river channel vegetation.

**Figure 5.** Distribution of total solar radiation (WH·m<sup>−</sup>2) reaching the substrate in the analyzed Wełna River reach in individual months.

**Figure 6.** Distribution of total solar radiation (WH·m<sup>−</sup>2) reaching the substrate in the investigated reach of the Wełna River for: (**a**) the period of April–September, (**b**) the annual period.


**Table 4.** Solar radiation (103·WH·m<sup>−</sup>2) at points located along the investigated reach of the Wełna River.

#### **4. Discussion**

The obtained results indicate the importance of shading for macrophyte growth. Similarly, as in a study by Jusik and Staniszewski [17], it was shown that increased shading perhaps limited macrophyte biodiversity and total cover in the river channels. The development of aquatic plants in the river channels is limited by shading [23]. Ali et al. [23] suggested that the management of tree vegetation might control incoming solar radiation, affecting submerged macrophytes. Similarly to the present case study, other studies shown that the impact of trees on the growth of aquatic plants depends on the group of macrophytes and the range of shadow impact [18,46]. Also, Jusik and Szoszkiewicz [47] observed significant correlations between the level of morphological modifications and shadowing of lowland river channels. Together with the increase in modifications of watercourses, the rate of shadowing decreased due to the changes of land use and simplification of the riparian species structure, commonly related to river regulation works. Kurtz et al. [48] conducted a shading experiment over a vegetation period to measure the effects of light reduction on *Vallisneria americana* in Perdido Bay (Florida–Alabama). The results showed that a 92% light reduction led to a decrease in chlorophyll *a* concentration, biomass, and leaf dimensions of macrophytes. Schneider et al. [49] showed that submerged macroscopic algae (*Chara intermedia* and *C. contraria*), which grow in an upright position, are taller at higher light intensities. In turn, Tan et al. [50] showed that reduced light reduced the growth of *M*. *aquaticum*; all of the analyzed growth indicators were significantly higher in sunlight treatment compared to the shading treatments. Kankanamge et al. [51] observed differences in shade tolerance between native and non-native species. In high shade (≥ 90%), the reduction in lateral spread and an increase in main stem length for non-native macrophytes was observed, while native species showed no response for these traits.

The river's shading is spatially and temporally heterogeneous. Spatial variations are related to landscape (channel width, orientation, and hillshade) and canopy characteristics (canopy extent, structure, and height) [41,52]. Temporal variations are related to temporal variability of canopy structure and sun position in terms of days, months, and years [42]. As it was shown in this study, only the use of a suitable model allows reliable assessment of the amount of energy reaching the ground.

During the planning of river conservation or regulation works, it is crucial to maintain proper light conditions for macrophytes in view of the potential problems with both absence and limitation of aquatic plant development [4]. Excess light, which can even cause full overgrowth of the river channel, respectively changes water flow conditions [34] and causes a decrease in channel permeability [14]. The phenomenon of river channel overgrowth is very intensive in lowland rivers flowing through agricultural areas [16], with the simultaneous presence of strong sunlight conditions (lack of trees, removed during regulation works) and significant input of nutrients due to the surface flow [14].

Flow resistance for a specific type of aquatic vegetation is a function of many variables, including flow velocity, plant shape (habit), and roughness of the river bottom and channel walls. In recent decades hydraulic flow conditions in open river channels overgrown with vegetation have been investigated by many research centers [7,53,54]. Similarly, as in the present study, most research on the subject has been experimental [55,56]. For example, the authors of numerous papers analyzed flow conditions in overgrown river channels at an increase in the roughness coefficient [57] or in terms of changes in debris transport processes [55]. Many papers also indicate the need to combine the stream transport capacity with the degree of shading of the cross-section by aquatic vegetation [58]. In literature on the subject [58], an association was noted between the degree of cross-section shading by vegetation and values of the roughness coefficient, and biomass of aquatic plants [57]. Similarly, as in the case of these results, Łoboda et al. [59] showed that the presence of plants within the river channel cross-section slightly reduces the cross-section surface area, but markedly changes the distribution of flow velocity. This observation was also confirmed by studies on the effect of river channel dredging on values of the roughness coefficient and distributions of velocity [57,60].

In a similar way as by Przyborowski et al. [61], results obtained in this study may be referred to hydromorphological changes, in the analyzed case connected with transport and sedimentation of fine organic particles. The presence of plants has a considerable effect on the above-mentioned processes.

Calculations conducted according to the method proposed by Rickert [36] for the location at ϕ = 52.5◦ northern latitude (comparable to the location of the reach in Coto ´n with ordinates 52.678◦N 17.631◦E) for the cross-section shaded by trees showed that the actual solar radiation reaching the water surface decreases within 24-hour periods to values comparable to those obtained in the model for the shaded river reach. The values of total photosynthetically active radiation for the shaded river reach confirmed a lack of potential conditions for hydrophyte growth in that part of the river reach.

#### **5. Conclusions**

This paper presents an example of studies on the effect of factors causing watercourse channel shading on water flow conditions and the dynamics of hydromorphological processes. The presented analysis indicates the importance of local conditions determining availability of light energy for plants on the hydromorphological conditions found in a watercourse. The selected reach of the Wełna River is characterized by high variability in river channel shading, resulting from the presence of a dense forest complex growing in the upstream river stretch (LAI = 3). Immediately outside the forest edge, the river flows among meadows and arable fields, where there are no trees. These conditions affect river channel morphology and flow conditions. In the forested river reach, bottom and riverbank vegetation is completely absent. The river channel is wide; however, because of the vicinity of trees and shrubs, considerable accumulation of organic matter is observed in the river channel (26.5% MO compared to 2.4% MO in the treeless cross-section in 2014, and 31.9% MO to 4.1% in 2019). This is also promoted by the very low flow velocity (approximately 0.1 m/s). This velocity results from the damming effect as a result of overgrowth in the river reach located immediately outside the forest. The downstream reach is characterized by a large amount of vegetation overgrowing the banks and the river bottom. This cross-section is more compact, and flow velocities in the streamline are slightly higher (although with reference to the entire cross-section together with the vegetation (dead) zones the mean velocity is also approximately 0.1 m/s). The developed model of light availability makes it possible to determine river channel overgrowth conditions, and thus to predict potential dynamics of hydrodynamic processes. Analysis of total radiation reaching the water surface through trees, ranging from 50 to <sup>512</sup> <sup>×</sup> <sup>10</sup><sup>3</sup> WH·m−<sup>2</sup> at the mean value of 245 <sup>×</sup> <sup>10</sup><sup>3</sup> WH·m<sup>−</sup>2, confirmed a significant role of shading, as limiting macrophyte growth, which in turn causes changes in the character of hydromorphological processes.

**Author Contributions:** Conceptualization, T.K.; methodology, T.K. and R.W.; validation, T.K., S.Z. and M.H.; investigation, T.K., S.Z. and M.H.; writing—original draft preparation, T.K., M.S., R.W., J.J., S.Z. and M.H.; writing—review and editing, T.K., M.S. and J.J.; visualization, S.Z. and R.W.; supervision, T.K. and M.S.; project administration, T.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** The publication was co-financed within the framework of the Ministry of Science and Higher Education program "Regional Initiative Excellence" in the years 2019–2022, Project No. 005/RID/2018/19.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 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 (http://creativecommons.org/licenses/by/4.0/).
