1 Introduction

In the framework of flood risk management, one of the important strategies to reduce flood hazards, especially in rural landscapes, is the management of small watercourses. One of the rational bases for the management of small watercourses is knowledge of their hydromorphological conditions, and an understanding of the hydrological behaviour of a watercourse in the context of the entire river basin (Smith and Ward 1998; Solín and Rusnák 2020) is the starting point. Furthermore, knowledge from fluvial geomorphology is important in the context of sediment management in the watercourse channel (Sear et al. 1995, Fasseta et al. 2009, Nuehold et al. 2009, Benito and Hudson 2010). The assessment of the watercourse morphology itself is based on geomorphic watercourse types (Montgomery and Buffington 1997; Montgomery and MacDonald 2002), and knowledge of the vegetation cover and its functions is necessary for the management of watercourse vegetation. The vegetation along the watercourses has several functions, such as anti-erosion, anti-collision and anti-deflation activity, protection against overgrowing and siltation of the riverbed, ensuring water quality, influencing the self-cleaning capacity of the watercourse, providing fauna refuge, aesthetic and production function, natural bio-corridor function, recreational function or hygienic function (Šlezinger and Úradníček 2009; Šlezinger 2010). The appropriate species composition of the vegetation, the spatial arrangement of the vegetation cover of the watercourse as well as the passability of the riparian vegetation itself can enhance the aesthetic enjoyment of the riparian landscape. In general, a watercourse is an important landscape element, and its management, in addition to reducing flood hazards, must also take into account the preservation of the biodiversity of the watercourse and its ecosystem services, including aesthetic and socio-recreational functions (Buisson et al. 2008; Moore and Rutherfurd 2016). However, the provision of such a high number of functions may not always be in line with flood hazard management practices.

The evolutionary development of conceptual frameworks and scientific principles for the assessment of the hydromorphological conditions of watercourses was analysed by Fryirs (2015). According to the author, an evaluation was carried out in the late 1980s and early 1990s on the basis of an inventory (audit) of hydromorphological conditions, and the identified status was compared with predetermined parameters or objectives. Although such an approach allowed the current state of the watercourse to be determined, the evaluation framework did not enable an interpretation of the current state in the context of its causes. This limitation was removed by the conceptual framework of design and development (early 2000s), which has recently been transformed into an application and use phase (Fryirs 2015). In the context of the application and use phase, the author suggested several challenges associated with this evaluation framework. In particular, the challenge of setting a benchmark against which to assess the extent of changes in the hydromorphological conditions of watercourses is crucial. However, determining the pressure indicators (which examine the causes or triggers of change in hydromorphological conditions) and response indicators (which record the symptoms of changes in the conditions from a reference) is also important.

As reported by Belletti et al. (2015), the hydromorphological assessment of watercourses started to develop rapidly with the adoption of the EU Water Framework Directive (Directive 2000). A wide range of hydromorphological assessment methods have been used in the context of attribute processing on watercourses in various scientific disciplines, such as ecology, biology, hydrology and geomorphology (e.g. Rinaldi et al. 2013, 2017; Gurnell et al. 2020). A detailed analysis of the hydromorphological evaluation methods of watercourses, in terms of concept, purpose, spatial scale, nature of data, their collection and processing, was previously performed by Belleti et al. (2015). The authors classified these methods into four basic groups: (i) physical habitat assessment, (ii) riparian habitat assessment, (iii) morphological assessment and (iv) assessment of hydrological regime alternation.

The hydrological and geomorphological attributes of the watercourses were assessed primarily in the context of achieving a good ecological status of the watercourses and restoring rivers for effective catchment management. However, there is still a lack of studies assessing the characteristics of watercourses in the context of flood hazards. As an example, the effectiveness of geomorphological analysis for flood studies was analysed by Lastra et al (2008). For another example, see Fasseta et al. (2009), who discussed the hazards associated with the transport of wood. In this study, the flood hazard of a watercourse is understood more comprehensively than that caused by floating trees and branches alone, and this work encompasses a broad set of watercourse attributes that influence flood hazards. The aim of this study is to present a possible way of mapping the flood hazard potential of small watercourses based on their hydromorphological attributes in a geographic information system (GIS) environment. Flood hazard mapping is an important precondition for the maintenance of watercourses (Moore and Rutherfurd 2016) because it creates a rationale for the sediment, vegetation and debris management of river reaches (EA Report 2015). It is considered to be particularly valuable and beneficial for local authorities, especially in connection with the paradigm of increasing the resilience of municipalities to flood risks (Tobin 1999; Cutter et al. 2014).

The paper is further structured as follows: In Sect. 2, the studied area is defined, and its characteristics are given. In Sect. 3, the methodological section clarifies the principles of mapping the flood hazard potential of small watercourses. In Sect. 4, the results are presented, and in Sect. 5, the results are discussed and concluding remarks are presented.

2 Study area

Mapping the flood hazard potential of small watercourses is presented in the form of a case study in a mountain region in the north-western part of Slovakia (Fig. 1), in the upper course of the Kysuca River and its tributary (Predmieranka). The upper stream of Kysuca is 22.16 km long and flows through the cadastral territory of three municipalities: Makov (8.99 km), Vysoká nad Kysucou (6.88 km) and Turzovka (6.29 km). The Predmieranka is 15.39 km long and flows through the cadastral territory of the Klokočov (11.19 km) and Turzovka (4.2 km) municipalities. The total area of the cadastral areas is 176 km2 (Fig. 1). Geologically, the area consists primarily of flysch rocks, on which soils with low permeability are created (Solín 2004). Thus, stream hydrographs are characterised by sudden and abrupt flood waves and indicate that the dominant drainage process in the river basin is surface or direct runoff (Solín 2019).

Fig. 1
figure 1

Schematic map of the studied area (source data ZBGIS Slovakia, GKU 2018)

Full size image

The area was chosen primarily because it is characterised by frequent floods, several of which were caused by local factors. From 1996 to 2020, floods were recorded 24 times, with 7 times in Klokočov, 8 times in Turzovka, 4 times in Makov and 5 times in Vysoká nad Kysucou. With the exception of the municipality of Klokočov, the other three municipalities were included in the area of potentially significant flood risk undertaken within the preliminary flood risk assessment (MoE SR 2018) in accordance with EU Directive on the Assessment and Management of Flood Risks (Directive 2007). From a geomorphological perspective, the study area is situated in highlands, with an elevation ranging from 450 to 1060 m above sea level. Based on land cover data from the GIS core database of Slovakia (GKU 2018), the area is dominated by forests, which make up 68% of the total area, followed by meadows (13%), arable land (0.73%) and built-up areas (0.60%).

3 Research design

In the context of flood hazard mapping, two approaches can be distinguished. In the first, the hydrological approach, the flood hazard is understood on the basis of probability. Flood hazard maps in this case cover the areas that could be flooded with a low, medium and high probability (Directive 2007). In the second, the environmental approach, the flood hazard is understood as the disposition or potential to flood (Weingartner et al. 2003; Minár et al. 2005; Solín 2019), which is conditioned by landscape attributes. In the case of watercourses, the flood hazard potential is driven by the attributes of the watercourse, influencing the flow capacity of the channel and the potential negative consequences of a flood. The mapping process in this case consists of determining the status of attributes and expressing their impact on the flood hazard. Such an approach is rational above all from the point of view of the management of watercourses where intervention management into sediment, vegetation and debris is required.

Our study focuses on mapping the flood hazard potential based on the hydromorphological attributes of watercourses. The mapping process consists of the following steps: (i) selection of watercourse variables influencing flood hazard potential (pressure indicators), (ii) division of the watercourse into reaches in terms of the status of their variables, (iii) evaluation of the attributes of the variables in terms of their impacts on flood hazards and (iv) determination of the flood hazard potential index (FHI) for river reaches based on the aggregation of the impacts of variables.

3.1 Selection of watercourse variables

The following set of variables was used to map the status of the watercourse: the morphology of the watercourse channel, the land cover of the watercourse floodplain, the transport infrastructures along the watercourse, the type of bank and channel vegetation, the negative features of the channel vegetation and the debris in the channel. Each of the seven variables is represented by a set of attributes (Table 1) and 34 attributes in total. The selection of the variables was based on the authors' expert judgement of their impact on the flow capacity of the channel or on the negative consequences of flooding.

Table 1 Variables and their attributes for flood hazard mapping in the study area
Full size table

The attributes of variable floodplain land cover and type of bank vegetation were determined based on high-resolution orthophoto maps with a ground sampling distance (GSD) of 20 cm/pixel (GKÚ and NLC 2017). To determine the attributes of transport infrastructure, we used orthophoto maps and the ZBGIS Slovakia database (GKU 2018). The attributes of channel morphology were also determined based on orthophoto maps, and in the second step, by field research, their photographic documentation was carried out. The attributes of channel vegetation, negative features of vegetation and debris were determined based on field research, including the extensive photographic documentation of river reaches.

3.2 Division of the watercourse into reaches

In the GIS, each watercourse was divided into river reaches based on the attributes of the variables, using a hierarchical structure. At the first hierarchical level, the morphology of the watercourse channel was considered. The river reaches of the 1st order were then divided at the second hierarchical level, according to the land cover of the floodplain. At the third hierarchical level, the differentiation of river reaches was conducted in terms of the attributes of the transport infrastructures (roads, railway) located near the watercourse. At the last hierarchical level, the attributes of the watercourse vegetation were applied.

3.3 Evaluation of the variable attributes in terms of the impact on flood hazards

3.3.1 Logical rules

In the context of assessing the impact of each variable, the following rules were used: Channel morphology. Based on the morphological types of river channels, we can infer the nature of the processes taking place in the river basin and channel. Of seven natural channel morphologies (Montgomery and Buffington 1997; Montgomery and MacDonald 2002), we identified three morphological types on the Kysuca and Predmieranka watercourse, and extensively modified channels were also determined (Fig. 2). The plane bed channels of the watercourses consist of boulders and cobbles. Pool riffle channels have a heterogeneous bed that contains bars and pools. Transport in this morphology type is limited, and sand and gravel sediments comprise the bank and central bars. Sedimentation in the watercourse reduces the channel flow capacity, leading to an increase in the water level and, consequently, a higher risk of flooding. The morphology of forced pool riffle channels is forced by obstructions such as large woody debris. Extensively modified channels may have the character of an artificial channel, where both banks and the bottom are modified, or they may have a modified left or right bank only. Increasing the cross-sectional profile of the channel and making concrete or stone adjustments to its banks and bottom will increase the channel flow capacity, resulting in a smooth passage of the flood wave across the modified river reach. However, a flood wave that passes through a modified reach without any transformation increases the flood hazard of the reach below because the channel has difficulty in transporting the flood wave.

Fig. 2
figure 2

Watercourse morphologies identified in the study area (following Montgomery and Buffington 1997; Montgomery and MacDonald 2002)

Full size image

Land cover of the floodplain. The severity of flood hazard consequences varies depending on the type of land cover. For example, it is higher in the case of river reaches in residential, industrial or commercial zones than in meadows and forests. The land cover of the floodplains on the left and right sides of the watercourse is assessed separately. The land cover classes were determined according to Feranec et al. (1996).

Transport infrastructure. The severity of a flood hazard is also increased by the damage to the line transport infrastructure (roads, railway) located near the watercourse (at an approximate distance of 20 m from the watercourse). The severity of flood hazards in river reaches with infrastructures of greater importance for traffic is higher (e.g. railway, first-class order) than that in reaches with infrastructures of lower importance (e.g. third-class road or local and purpose-built communication).

Bank and channel vegetation. The anti-erosion and protective functions of the bank vegetation are considered (Šlezinger and Úradníček 2009; Šlezinger 2010). The anti-erosion effect of the root systems of vegetation against the flowing water strengthens the banks of the channel and thus helps to stabilise the flow channel capacity. Vegetation, especially at smaller watercourses, prevents excessive light access, partially shades the channel and effectively prevents the creation of conditions suitable for the overgrowth of weedy aquatic plants and the sedimentation of soil particles. The impact of the channel vegetation slows down the flow rate and increases the sedimentation of soil particles, resulting in a decreased flow channel capacity.

Negative aspects of vegetation. There are specific aspects of each type of vegetation cover that negatively affects the flow channel capacity. For example, trees on the bank of a channel with a washed and exposed root system are susceptible to being uprooted during floods, creating obstructions in the channel. The flow channel capacity is also reduced by obstructions caused by overgrown vegetation or overhanging branches of trees and shrubs above the water level.

Debris. Debris on the banks or in the river channel is transported away during floods and can cause the clogging of bridges over watercourses, thus decreasing flow channel capacity and increasing local flooding.

For the latter two variables mentioned, only the occurrence (yes/no) of organic and inorganic waste materials on the banks and uprooted trees or branches in the channel of the river reach was determined, not the frequency of their occurrence. However, the occurrence of the variables themselves is not the subject of the impact evaluation but rather their attributes (RHG, WERS, OHBW, respectively, DOIMB, UTBCH), in case they occur on the river section. When a given variable does not occur on a river reach, it does not have an effect on the flood hazard and has a value of zero in the evaluation process.

3.3.2 Evaluation scheme

The impact of each attribute on the flood hazard potential of the river reaches was assessed according to logical rules, as described in the previous paragraph. Using logical rules, the ordinal flood hazard impact scale of individual attributes was first established (1—high impact, 7—very low impact). This ordinal scale was then transformed into a quantitative weight of impact (score of impact) by the rank-sum method (Malczewski 1999), using the following equation:

$$w_{j} = \frac{{n - r_{j} + 1}}{{\sum {(n - r_{k} + 1)} }},$$
(1)

where wj is the normalised weight for the jth attribute, n is the number of attributes (k = 1, 2, …, n) and rj is the rank position of the attribute.

The range ordinal scale of impact is not fixed but varies depending on the numbers of attributes of the variable. The range of the ordinal scale affects the size of the values of wj; they will be lower for variables with more attributes than for those with fewer attributes, which distorts their impact. To eliminate this distortion, a second-stage transformation was performed, and norm wj values were determined using the following equation:

$$norm{\text{ w}}_{j} = \frac{{w_{j} }}{{w_{\max } }}$$
(2)

Table 2 shows the impact of individual attributes of selected watercourse variables on the flood hazard, using the ordinal flood hazard impact scale and a quantitative weight of impact.

Table 2 Assessing the impacts of variable attributes on flood hazards
Full size table

To assess the influence of attributes of individual watercourse variables on the flood hazard, it is worth mentioning that among the methods for expressing the weights of attributes of individual variables (Malczewski 1999), we used the rank-sum method, which belongs to the group of ranking methods. The reasons for its use are simplicity and logicality as it does not require any additional information to express the normalised weight wj. In addition to expressing the weight of the influence of the individual attributes of the variables, it is also possible to differentiate the weight of the variables themselves. However, in the case of the variables themselves, the formulation of rules to express the weighting of their influence is not as straightforward as in the case of their attributes. Therefore, we assume that all variables have the same weight.

3.3.3 Determination of FHI

The FHI is a multicriteria function that aggregates the influence of flood hazard attribute variables into a single flood hazard mapping characteristic. It thus enables a mutual comparison of river reaches in terms of their flood hazard potential. The value of the FHI for each river reach was determined by the following relationship:

$$FHI={\sum }_{i=1}^{n}norm {w}_{ji},$$
(3)

where norm wji is the normalised value of the attribute j of the ith variable of the river reach.

4 Results

4.1 Differentiation of river reaches in terms of the impacts of their attributes on flood hazards

Based on a combination of the variables influencing the flood hazard (channel morphology, land cover of floodplains, transport infrastructures, technical adjustments of the channel and vegetation), the watercourse Kysuca was divided into 104 unequal reaches and Predmieranka into 57 river reaches. The numbers of reaches are listed in Table 3, together with an indication of the minimum and maximum lengths of each individual reach according to the cadastral areas.

Table 3 Number of river reaches in the cadastral areas
Full size table

The mapping of the impact of river reach attributes on the flood hazard (value of norm wj) is presented in more detail in Figs. 3, 4, 5 and 6. In the context of flood risk management at the local level, we find it useful to interpret the results obtained in relation to the municipalities through which the watercourses flow. The graphs in the figures show the proportion of the impact of individual classes of the given attribute on the flood hazard from the total length of the watercourse in the municipality's cadastral area. In relation to the status of river reaches in terms of selected variables and their impacts on the flood hazard potential, several interesting facts can be derived in the context of cadastral areas. The following can be stated about the river sections in the individual municipalities of the area of interest:

Fig. 3
figure 3

Impacts of channel morphology a and channel modifications b on the flood hazard

Full size image
Fig. 4
figure 4

Impacts of floodplain land cover a and infrastructures b on the flood hazard

Full size image
Fig. 5
figure 5

Impacts of the bank a and channel vegetation b on the flood hazard

Full size image
Fig. 6
figure 6

Impacts of negative vegetation a and debris attributes b on the flood hazard

Full size image

The river reaches with a channel morphology of bed plane and pool riffles of the Kysuca watercourse have average slopes ranging from 2.11–4.38% to 2.01–3.31%, respectively, and those of the Predmieranka watercourses range from 3.02–4.35% to 2.31–3.24%, respectively. The flood hazard potential resulting from the watercourse morphology is generally low in the study area (Fig. 3a). In the case of the Kysuca watercourse, the share of bed plane reaches with a low impact on the flood hazard (norm wi = 0.25 and 0.5) ranges from 81 to 93% of the total length of the watercourse in the municipalities. In the case of the Predmieranka watercourse, this share ranges over 73–78%. The reaches of the Kysuca watercourse with bank and central bars that increase the flood hazard (norm wi = 0.75 and 1) make up 19% in Turzovka, whereas they account for 7% of the total length of the watercourse in Makov and Vysoká nad Kysucou. In the case of the Predmieranka watercourse, the share of these river reaches is 27% in Klokočov and 22% in Turzovka.

Figure 3b shows the flood hazard potential of the river reaches in the context of technical adjustment. In Makov, Vysoká nad Kysucou and Turzovka, the shares of the river reaches with technically adjusted channels reducing the flood hazard (norm wj = 0.25 and 0.50) are 12%, 13% and 14%, respectively, of the length of the Kysuca watercourse in the cadastral areas. In the case of the Predmieranka watercourse, the shares of technically modified reaches in the Klokočov and Turzovka municipalities are only 1% and 2%, respectively.

Figure 4a shows the differentiation of river reaches in terms of the severity of the flood hazard due to the floodplain land cover. Given that the land cover types on the left and right sides of the river reach were identified, the value of the flood hazard severity of the river reach was determined as the arithmetic average of the values of the norm wj on the left and right sides of the river reach. Therefore, the average values of the norm wj of the river reaches can be divided into interval classes. In the case of the Kysuca watercourse, river reaches with a high severity of flood hazard were identified in Vysoká nad Kysucou and Turzovka and accounts for 92% and 88% of the total length of the watercourse in the cadastral area, respectively. In Makov, this value is 54%. A slightly different situation was found in the cadastral areas through which the Predmieranka watercourse flows. In Klokočov, the river reaches with a high severity of flood hazard make up only 11% of the total length of the watercourse in the cadastral area, whereas this value reaches 68% in Turzovka.

Figure 4b shows the severity of the flood hazard due to infrastructure damage along the river sections. River reaches with a high severity of flood hazard (norm wj = 1 or 0.75) due to the presence of railways and first- and second-class roads along the river reaches account for 10% (Makov) to 36% (Vysoká nad Kysucou) of the total length of the Kysuca watercourse in the municipalities. In the case of the Predmieranka watercourse, river reaches with a high level of flood hazard severity constitute 28% and 15% of the total length of the watercourse in Turzovka and Klokočov, respectively.

Figure 5a presents the differentiation of the river reaches in terms of the impact of bank vegetation on the flood hazard. The bank vegetation was identified on the left and right sides of the river reach, and the flood hazard potential of the reach was determined as the arithmetic mean of the norm wj values on the left and right sides of the river reach. The average values of norm wj were grouped into five interval classes. The scale with a low to very low impact on flood hazard (norm wj in the range of 0–0.285 or 0.286–0.50) belongs to river reaches with shrub–tree vegetation on both banks and to a combination of tree, shrubby and herbaceous vegetation on one of the banks. In the case of the Kysuca watercourse, the largest share of the total length of these classes was observed for Makov, Vysoká nad Kysucou and Turzovka, (45%, 65% and 55%, respectively). The river reaches with herbaceous or herbaceous–shrub vegetation that have a very high impact on the flood hazard (norm wj in the range of 1–0.73) account for 23%–28% of the total length of the Kysuca watercourse in the cadastral areas. In the case of the Predmieranka watercourse, the occurrence of river reaches with a low to very low impact on flood hazard dominates. In Klokočov and Turzovka, they account for 88% and 70%, respectively, of the total length of the Predmieranka watercourse in the cadastral areas.

River reaches with some form of channel vegetation are rare in the case of the Kysuca watercourse (Fig. 5b). The reaches without channel vegetation dominate in Makov and Vysoká nad Kysucou, accounting for 94% and 93%, respectively, of the total length of the watercourse in the cadastral area. The river reaches with channel vegetation occur mostly in Turzovka, accounting for 31% of the total length of the Kysuca watercourse in the cadastral area. The prevailing vegetation form is herbaceous–shrub vegetation, with a very high impact on the flood hazard (norm wj = 1), accounting for 14% of the total length of the watercourse in the cadastral area. In the case of the Predmieranka watercourse, river reaches with some form of channel vegetation make up 25% and 18% of the total length of the watercourse in the cadastral area in Klokočov and Turzovka, respectively.

The length of river reaches with negative attributes of bank vegetation account for a large percentage of the total length of watercourses, especially in the Vysoká nad Kysucou (64%) and Turzovka (74%) municipalities in the case of the Kysuca watercourse. These river reaches accounted for an even higher percentage (up to 87%) in the cadastral area of Turzovka in the Predmieranka watercourse. In these municipalities, the river reaches with OHBW combined with RHG or ISL, having a low impact on flood hazard potential (norm wj = 0.25), are predominant (Fig. 6a). However, the occurrence of river sections in which OHBW and RHG is combined with the occurrence of WERS, with a very low impact (norm wj = 1.00), is also significant (Vysoká nad Kysucou). A significantly lower proportion of river reaches with negative vegetation attributes of the total length of the watercourse in the cadastral area was observed for the municipalities of Makov (39%) and Klokočov (26%).

In the municipalities of the Kysuca watercourse, the river reaches with debris account for 21–25% of the total length of the watercourse. Overall, the reaches without the presence of waste in watercourses dominate. Organic and inorganic material waste, which impact the flood hazard to a medium level (norm wj = 0.50), are the predominant type of debris on the banks (Fig. 6b). In the case of the Predmieranka watercourse, the share of river reaches with debris is smaller, accounting for 4% and 12% of the total length of the watercourse in Klokočov and Turzovka, respectively.

4.2 Differentiation of river reaches in terms of FHI

The FHI values of the river reaches determined by Eq. (3), aggregating the norm wji values of the impact of attributes on flood hazard, are the key aspects of flood hazard potential mapping. They were divided into five interval classes (Fig. 7), where the interval class with the lowest FHI values reflects a very low flood hazard potential and that with the highest FHI values reflects a very high flood hazard potential. In particular, river reaches with FHI values in the interval classes 3.29–4.00 and 4.01–5.23 (high and very high flood hazard potential) are considered as reaches where management intervention is primarily needed. In the case of the Kysuca watercourse, such reaches make up 35% of the watercourse’s total length in the Makov municipality, 53% in Vysoká nad Kysucou and 54% in Turzovka. Regarding the cadastral areas of the Predmieranka watercourse, river reaches with high and very high flood hazards make up 14% of the total length of the watercourse in the cadastral area in Klokočov and 44% in Turzovka.

Fig. 7
figure 7

Flood hazard potential index (FHI) of the river reaches of the Kysuca and Predmieranka watercourses

Full size image

River reaches with high to very high flood hazard potential (FHI > 3.28) are slightly dominated by bed-stone morphology, whereas those with pool riffle morphology with bank and central bars are also common. The landscape cover of the floodplains of these river reaches is dominated by classes with discontinuous urban structure and a complex of cultivation patterns and traditional forestland–scrub. The riparian vegetation of river reaches is dominated by herbaceous–scrub vegetation, and river sections with negative attributes of riparian vegetation are common. The number of river reaches with overgrown riverbeds and debris in the channel is also not negligible.

River reaches with a very low to low flood hazard potential (FHI ≤ 3.28) are those with plane bed channel morphology, without technical channel modification. A linear infrastructure along river sections occurs only sporadically. The floodplain of the river reaches is mainly covered by forests, a complex of cultivation patterns, with the rare occurrence of transitional woodland–scrub and a discontinuous urban fabric. The bank vegetation of the river reaches is dominated by trees and tree–scrub vegetation, whereas the river bottoms are generally free of vegetation and debris. Based on the negative attributes of the bank vegetation, OHBW, RHG or ISL is occasionally found in the river reaches.

5 Discussion

One of the important methodological aspects of the hydromorphological assessment of watercourses is the length of the surveyed river reach. For example, Raven et al. (1998) divided their studied watercourse into 500-m-long reaches. The same reach length has also been applied in the widely used Modular River Physical or MoRPh survey (Gurnell et al. 2020). The length of the river reach in the MoRPh survey is a function of the width of the watercourse, namely twice the river width (width < 5 m, length = 10 m; width ≥ 5 m and < 10 m, length = 20 m; width ≥ 10 m and < 20 m, length = 30 m; width ≥ 20 m and < 30 m, length = 40 m). The primary river modules defined in this way can then be aggregated into subsections, e.g. MoRPh5 includes five contiguous MoRPh modules. However, in our study, we did not consider a fixed a priori length of the reach, but similar to Rinaldi et al. (2013), we applied a hierarchical division of the watercourse in the reaches of unequal length in relation to its attributes. In contrast to the approach with the same length of the river reaches, the applied approach of unequally long river reaches enabled the identification of relatively homogeneous reaches in terms of a given attribute and the creation of certain types of river reaches in combination with other attributes.

Another important aspect is whether the hydromorphological assessment of the watercourse concerns only the channel (bottom and banks) or also covers a wider area along the watercourse. For instance, the MoRPh survey extends 10 m away from the river on both banks. However, in the context of the negative consequences caused by river flooding, a buffer zone of 10 m is not sufficiently wide. The assessment of the severity of the flood hazard within the wider area, i.e. the floodplain, therefore appears to be justified.

The relativity of the rating scales for the influence of watercourse attributes on flood hazard can also lead to some debate. The rating scales were created within the sets of attributes related to the watercourses that were the subject of the case study. In the context of the systematic mapping of the flood hazard potential of watercourses throughout the whole territory of the state, it would certainly be appropriate to base the evaluation scheme for the influence of their attributes on flood hazard on an exhaustive set of attributes representing the watercourses of the whole territory of the state. However, such an evaluation scheme does not yet exist and requires a separate study. To present the actual process of mapping the flood hazard potential itself in the form of a case study, which is the intention of this study, we consider the application of relative rating scales to be justified.

Although the proposed mapping process has the features of an inventory (audit), the identified status is not in any way compared to predetermined parameters, objectives or benchmarks but to the phenomenon—the flood hazard. We see the contribution of the study mainly in the context of flood hazard assessment. We have already stated above that flood hazards can be understood on the basis of probability or as a disposition or potential for flood. Modelling of the water level for floods of different probabilities of occurrence and the corresponding extent of inundation is primarily intended to contribute to the determination of the amount of damage caused (direct, indirect) and the regulation of economic activities in areas exposed to flooding. The assessment of the flood hazard potential based on catchment attributes or other (administrative) units to identify the variability of flood hazards at national, regional or local level also contributes to the spatial rationalisation of flood risk management decisions.

The results for our study area indicate that the river reaches with a high and very high flood hazard potential account for a high proportion of the total length of the watercourses in three municipalities, namely Makov, Vysoká nad Kysucou and Turzovka. This is an important piece of information, especially in terms of determining the expected expenditure on the maintenance of watercourses. It equally puts the argument in the hands of local authorities to ask the watercourse manager to urgently take action to manage critical reaches of the river, particularly with regard to the maintenance of vegetation, sediment and debris. However, if the riparian vegetation and channel sediments are not systematically maintained, the watercourse is isolated from the remaining area. This is the case for the reaches of the Kysuca watercourse in the urban area of Turzovka (Fig. 8a), which do not act as an integral part but rather as a separate element in the urban landscape. An example of appropriate watercourse management is the Dunajec tributary in the municipality of Szczawnica in Poland (Fig. 8b). In this case, the watercourse is an integral part of the urban landscape and an important ecological and aesthetic element of the riverine landscape. In the present study, the occurrence of waste in the channel, which was mainly found for the cadastral areas of the Kysuca watercourse, indicates that the population has a low awareness of flood hazards due to debris. This is valuable information for the local government regarding measures that focus on increasing the population's awareness of flood hazards.

Fig. 8
figure 8

Riverine landscapes

Full size image

6 Conclusions

The flood hazard mapping of small watercourses presented in this study introduces a further complementary aspect to the flood hazard assessment, namely the assessment of the flood hazard potential of the watercourses themselves. The presented mapping based on the current trends of hydromorphological assessment of watercourses enables the division of the watercourse into reaches that differ from each other by their flood hazard potential as well as the identification of river reaches that are critical in terms of flood hazards. This assists in eliminating a group of local flood hazard factors associated with watercourses and is mainly considered important for flood hazard management at the local level, especially in connection with the paradigm of increasing the resilience of society to flood risks. Thus, local authorities should acquire a comprehensive body of knowledge on the flood hazards and risks of the cadastral areas they manage. Mapping the impacts of river reach attributes on the flood hazard potential creates a knowledge base that is important for the intervention management of watercourses. Such an approach is a rational basis, especially for the management of sediment, vegetation and debris of river reaches.