3.1 Flood regime estimation
To estimate the flood regime of the Jangdan and Binae Wetlands, water level data from Paju-si (Tongil Bridge), located 5.5 km upstream of the Jangdan Wetland and Chungju-si (Mokgye Bridge) located 6.8 km upstream of the Binae Wetland were obtained. Although both stations are located upstream of the wetlands, there is a difference of ± 3 cm between each station and wetland according to a field survey conducted with Trimble R10 RTK GPS. The year-round water level and LDC time series for both wetlands are shown in Fig. 3.
The flood regime of Jangdan and Binae Wetlands showed different behaviors based on their water level, as depicted in Fig. 3(a-d). The Jangdan Wetland (Fig. 3(a-b)) had an average water level between 1.5 and 12.3 EL.m throughout the years, which seems to be influenced by the tide due to its proximity to the Imjin River estuary. During normal conditions, the water level was maintained at a constant level between 1.5 and 2.5 EL.m, with an average annual water level of approximately 1.9 EL.m. The water level could rise to approximately 2.0 EL.m during normal conditions due to the influence of the tide but can reach up to 10.0 El.m or more for a short period during a flooding event.
The Binae Wetland, located in the middle of the Namhan River (Fig. 1), exhibited a different behavior from the Jangdan Wetland, as shown in Fig. 3(c–d). The Binae Wetland had a constant water level between 52.5 and 52.8 EL.m throughout the year, with an average annual water level of about 52.6 EL.m and did not exceed 53.0 EL.m. During the flood season (June to September), the average water level rose to 53.0 EL.m or more for about three months. Therefore, the Jangdan Wetland shows a constant water level throughout the year and only experiences a high-water level during only flood events. At the same time, the Binae Wetland maintains a constant water level throughout the year, with a rise in water level during overall flood season.
3.2 Quantification of flood regime on wetland
The flood regime diagram method discussed in Section 2.2 was used to quantify the Jangdan and Binae Wetlands’ flood regimes. Firstly, the LDCs were estimated using at least 13 years or more of historical records. Secondly, the inundated area of wetlands corresponding to each water level in the LDCs was calculated. Lastly, the inundated area ratio curve was plotted directly, corresponding to the LDCs. To calculate the inundated area ratio, the digital elevation model (DEM) for each wetland and the nearby area was constructed using a 1:1,000 numerical topographic map provided by the National Geographic Information Institute (NGII, 2021) that was calibrated with a field survey using Trimble R10 RTK GPS. The resulting flood regime diagrams for each wetland are shown in Fig. 4.
The flood regime diagrams in Fig. 4. allowed for the quantification of the flood regime of each wetland. For the Jangdan Wetland (Fig. 4(a–b)), it was observed that inundation occurs in approximately 20% of the area, with the water level rising over 4.0 EL.m or more for approximately 18 days a year. Less than 5% of the area was constantly inundated most of the year, and the remaining 95% was flooded for only a few days annually. This suggests that the Jangdan Wetland is mostly territorialized, with flooding occurring only for a few days each year. On the other hand, the Binae Wetland (Fig. 4(c–d)) had approximately 45% of its area constantly inundated throughout the year, and 50–65% of the area was inundated annually for more than two months (60 days). However, approximately 30% of the wetland area is not inundated in ordinary years, and 10% of the area is inundated by large floods that occur once every few years. This means that about 50% of the Binae Wetland area is constantly inundated, up to 70% is regularly flooded for about two months annually, and the remaining 30% is territorialized and only rarely inundated.
The quantification of the flood regime of wetlands, such as those analyzed in this study, can provide crucial data for restoration and maintenance efforts. For instance, the Binae Wetland is home to the endangered Aster altaicus, a Korean endemic plant species vulnerable to inundation. The habitat of this species is susceptible to flooding, which can cause significant fluctuations and damage. Thus, restoring a 30% area of the wetland to avoid flooding would be an effective measure to protect this valuable species.
3.3 Wetland vegetation zone considering flood regime
Flood regimes are crucial for wetlands as changes in the water level or inundation impacts vegetation on or near the wetlands (Hu et al., 2015; Farrell et al., 2010). Considering that, inundated plants and humus plants are generally less affected by flooding (Kim et al., 2010) while flood regimes (water level and inundation) are important constraints on the growth of terrestrial plants. Zoning wetland areas according to the flood regime is vital for reliable management, and in this study, spatial quantification of the flood regime was performed to facilitate this. Using the flood regime diagram discussed in Section 3.2, nine zones were classified based on the corresponding inundation depth and plant species that died during the inundation period (Table 1). To determine the threshold period for vegetation species growing in the riparian zone, the critical period that could withstand inundation was used as a criterion. However, it is difficult to consider all growth and living conditions owing to the lack of relevant studies.
Table 1
Classification of vegetation communities according to inundation conditions.
Area No. | Exceedance Probability for Water Level | Inundation days per Yr | Statistics (Normal Distribution) | Vegetable communities |
Ⅰ | ~ 0.02 | 7↓ | -2 σ | Beckmannia syzigachne, Poa pratensis, Lolium perenne, Festuca arundinacea |
Ⅱ | ~ 0.04 | 15 | | Stellaria alsine, Cymbopogon torti, Acorus calamus L., Typha orientalis |
Ⅲ | ~ 0.06 | 21 | | Lespedeza bicolor, Lolium multiflorum, Miscanthus sinens, Kummerowia striata, Equisetum hyemale, Iris pseudoacorus L., Iris sanguinea |
Ⅳ | ~ 0.08 | 30 | | Amorpha fruticosa, Wisteria floribunda, Pennisetum alopecuroides, Hemerocallis lilioasphodelus, Zoysia japonica Steud., Hemerocallis lilioasphodelus, Iris setosa Pallas ex Link, Salix koreensis |
Ⅴ | ~ 0.16 | 60 | -1 σ | Miscanthus sacchariflorus, Iris ensata, Iris pseudoacorus L., Phragmites japonica, Phragmites communis, Iris pseudoacorus L. |
Ⅵ | ~ 0.20 | 72 | | Bromus ciliatus, Helianthus grosseserratus, Phalaris arundinacea, Typha latifolia, Eutrochium maculatum, Aneilema japonicum, |
Ⅶ | ~ 0.30 | 110 | | Persicaria thunbergii, Persicaria hydropiper, Zizania caduciflora, Scirpus tabernaemontani, Typha orientali & Typha angustifoli |
Ⅷ | ~ 0.50 | 183 | µ | |
Ⅸ | ~ 0.75 | 275↑ | | |
※ σ and µ are indicate that standard deviation and median, respectively. ※ Reference from Park, M (2017); Kercher and Zedler (2004); Han et al. (2009); Park and Choi (2001); Nam et al., (2015); Park, C. M. (2002); Lee et al., (2014); Park et al., (2008); Shim, S. (2015); Kim et al., (2014) and Park et al., 2017 are suggested annual total inundation period, not a continuously period |
The investigation on the effect of flooding on riparian vegetation was expanded in this study by conducting a literature survey on 43 riparian vegetation species. In this study, tree species that were less likely to be wholly inundated during flooding were excluded, and mainly herbs and shrubs were investigated. To determine the impact of flooding on the vegetation species, it was observed whether they died or not when they were inundated by 50% or more of the height of the vegetation.
Based on the survey results, it was observed that terrestrial vegetation species, such as American sloughgrass (Beckmannia syzigachne) and lawns (Poa pratensis, Lolium perenne, Festuca arundinacea), died even after inundation within a week, indicating their high vulnerability to flooding. On the other hand, sweet flag (Acorus calamus L.), cattail (Typha orientalis), and flea sprouts (Stellaria alsine) were observed to die after the inundation of more than two weeks. Some plant species, such as lespedeza (Lespedeza bicolor), day lily (Hemerocallis lilioasphodelus), and montane aster (Aster koraiensis Nakai), were found to survive even after inundated for 3–4 weeks. Common reeds (Miscanthus acchariflorus), runner reeds (Phragmites japonica), reeds (Phragmites communis), and reed canary grass (Phalaris arundinacea) have been observed to survive immersion for nearly two months, and some vegetation survives inundation for more than three months. Based on the inundation tolerance of the investigated vegetation species, the flood regimes of the wetland were classified into nine zones (Table 1), and the corresponding water levels are shown in Fig. 5.
The DEM was compared with the exceedance probability in Table 1 and the corresponding water level to classify each zone. The inundation criterion was based on literature surveys, where shrubs and herbs were considered when the water level reached 50% of their height. As the average height of these vegetation types was 0.8 m, areas where the inundation depth exceeded 0.8m from the ground level were considered inundated. The resulting inundation map with a resolution of 10 m × 10 m is presented in Fig. 6.
From Fig. 6, the spatial flooding regime of each wetland can be determined. For instance, the Jangdan Wetland in Fig. 6(a) shows that most wetlands are Zone 1, which is only flooded for seven days of the year. Thus, it can be assumed that the wetland is only inundated for a few days during the year, and stagnant water is the primary source of water in the wetland. Zones 4–9 are situated northwest of the wetland, indicating that they serve as a primary route for intermittent flooding and inflow during floods. Meanwhile, Zones 2 to 5 are located in the southern part of the wetland and may function as a puddle and flood inflow/outflow passage. Considering the diagram in Fig. 4(a–b), it appears that the Jangdan Wetland experiences total inundation every year.
For the Binae Wetland, zone 9 (Fig. 6(b)), located on the entire right bank, is inundated for approximately eight months of the year due to the main flow of the Namhan River. Zone 1 is located between the left bank and the center, while Zones 3–6 are located between the center and right bank. Thus, in the case of the Binae Wetlands, the right bank was frequently inundated because of the main flow of the Namhan River, while the left bank and center were considered areas that are rarely inundated. It is estimated that Zone 1, which accounts for approximately 20% of the total area, is not flooded yearly but is only inundated during major floods that occur once every few years, as shown in Fig. 4(c–d).
The quantification of the flood regime is crucial in providing preliminary data for the restoration and management of wetlands. This is because year-round changes in the water level of wetlands and the classification of flooded areas are important constraints for vegetation and ecosystems. However, wetland surveys of flood regimes have not been conducted because of insufficient methodology for quantifying hydrological conditions. Even in the fourth detailed survey of inland wetlands of Korea (National Institute of Ecology, 2020), the hydrological condition investigation was limited to established master plans for future management, which focused on water resource amount, facilities used, flood volume, average flow, depth, and velocity.
Applying the suggested flood regime method allows field engineers and designers to easily analyze the detailed flood regime of the target wetland spatially. In the case of the Jangdan Wetland, which is home to several endangered species, such as the spoonbill (Platalea minor), Eurasian spoonbill (Platalea leucorodia), white-tailed sea eagle (Haliaeetus albicilla), and crane (Grus japonensis), the detailed flood regime can help identify feeding and nesting areas that require protection. Similarly, for the Binae Wetland, restoration efforts for vulnerable species such as Aster altaicus can be better targeted by identifying areas that are rarely inundated, such as the left bank. Overall, quantifying the flood regime provides essential data for the restoration and management of wetlands, which is critical for maintaining healthy ecosystems and preserving biodiversity.
The study had limitations greatly influenced by the reliability and accuracy of the data used, such as the DEM and water level. The accuracy of the topography of wetlands using the 1:1,000 topographic map provided by the National Geographic Institute may be limited, and this uncertainty was considered in the study. The DEM was calibrated and verified using RTK GPS, but further improvements can be made using drone surveys (Shelestov et al., 2021). Regarding water level data, the locations of the stations were constrained. The Jangdan and Binae wetlands have water level stations in close proximity and could be used. However, in other wetlands such as Wicheon Wetland, there is no water level station nearby; therefore, obtaining historical record data is impossible. Although long-term hydrological analysis and hydraulic modeling, such as HEC-RAS, can be an alternative for calculating the water level of the river, they may not be simple and practical for conservation biologists and engineers. Nevertheless, with the acquisition of fundamental datasets, the suggested method can be used to quantify flood regimes easily.