The study analyzes flood risk in the Danube Delta, a fragile ecosystem at the intersection of natural processes and anthropogenic interventions, increasingly exposed to the effects of climate change. Using 2D hydraulic modeling with HEC-RAS software, the research assessed the extension of floodplains for 10, 100 and 1000-year return periods, based on hydrological data collected between 1971 and 2005. The results obtained demonstrate a significant increase in the areas affected by floods, especially in scenarios with a low probability of occurrence, thus highlighting the increased vulnerability of the region to extreme phenomena. Analysis of water depth in flooded areas indicates that regions with intermediate depths (0.2–0.5 m) are most affected by 100-year flood events, while rare events (1000 years) extend the impact to larger areas and greater depths (0.5–1 m). This evolution confirms the trend of increasing risks in the context of climate change and the increase in the frequency of extreme hydrological phenomena. The hazard maps generated are an essential tool for flood risk management, supporting informed decision-making to protect local communities and the delta’s biodiversity. Among the solutions proposed to reduce the impact of floods are the ecological reconstruction of wetlands, the implementation of effective climate change adaptation policies and the improvement of water management strategies. The study also highlights the need to integrate additional factors, such as the influence of the Black Sea level, in future simulation models, for better risk anticipation and optimization of protection measures.

2D hydraulic modeling, Danube Delta, flood risk, hazard maps, HEC-RAS, risk management

Deltas are dynamic low-lying areas formed at the mouths of large rivers flowing into the receiving seas (Liu et al. 2024). They account for about 1% of the Earth’s land area globally and are home to more than 7% of the world’s population. These coastal areas are dominated by intensive agricultural activities favored by the presence of water and fertile soils (Chen et al. 2021) and are frequently heavily anthropized areas. The deltaic regions are characterized by frequent floods, some of them even showing a seasonality of the phenomenon, so in the context of increased anthropization they show increased flooding (Leduc et al. 2017). Since 2000, about 85% of the coastal deltas have been affected by severe flooding caused by changes in precipitation patterns due to global climate change and environmental change. These floods have temporarily affected a total area of about 260,000 km² of coastal zones. Research has shown that about 50% of coastal deltaic areas are vulnerable to flooding caused by rising sea levels (Edmonds et al. 2020). This vulnerability has increased in recent years, as projected values for sea level rise in the 21^st century and are above the inundability threshold of many deltaic areas (Youssef et al. 2024).

Flood risk assessment for coastal deltaic areas is based on the analysis of the floodplain area and the water depth in the floodplain (Nandam and Patel 2025). These parameters are used in the European Commission’s recommendations for flood hazard and flood risk maps (Velegrakis et al. 2024). The European Commission’s recommendations also include 3 possible scenarios for flood risk analysis (Arvidsson and Johansson 2024), including: i) low or extreme probability floods associated with a return period of once every 1000 years (0.1%); ii) medium probability floods associated with a return period of 100 years (1%); and iii) high probability floods that are associated with a return period of 10 years (10%). At the same time, in the context of global climate change, which leads to a high frequency of extreme meteorological and hydrological events, it is important for flood risk management in deltaic areas to also conduct flood hazard studies for flows associated with the impacts caused by different climate projections.

In the 20^th and 21^st centuries, the effects of human activity on climatic parameters have led to significant changes in ocean current patterns, marine processes and continental stream flows, accelerating the energy balance in the terrestrial atmosphere (Wang et al. 2024). Thus, the structural and functional parameters of coastal deltaic areas are directly and indirectly affected by the climatic changes of the last century. One of the largest coastal delta ecosystems in Europe is the Danube Delta with an area of 5640 km², the second largest delta in Europe. The Danube Delta is also vulnerable to changing climatic parameters and their effects on marine processes and the Danube River (Bănăduc et al. 2023), such as the rising of the sea level. The Danube Delta evolved in a natural regime until 1860 (Bănăduc et al. 2023), with the main branches (Chilia, Sulina and Sfântu Gheorghe from north to south) maintaining their natural structure until the beginning of the 20^th century, when the direction of the Sulina branch was changed to allow navigation to the sea (Armaş and Avram 2009). In the second half of the twentieth century, the delta’s inland water network underwent significant changes through the opening of navigation channels and the removal of large areas from the floodplain. The change in the flood regime led to changes in the natural dynamics of the entire delta system, increasing its vulnerability to flooding. Major floods occurred in 1942, 1970, 1975, 1981, 1988, 2005, 2006 and 2010 accordingly to the publicly available data at Ceatal Ismail.

In the Danube Delta, the frequency of floods has increased over the last decades favored by the geographical position, the hydrological specificity of the area, but also by extreme meteorological phenomena caused by climate change (Crăciun et al. 2022). The extent of flooding in the Danube Delta is directly proportional to the flow/level of the Danube’s waters as they flow into the Black Sea. The hydrological regime of the Danube Delta is primarily influenced by the water discharge of the Danube River, which varies significantly throughout the year. The average annual discharge at the entrance of the delta (Ceatal Izmail) is approximately 6,500 cubic meters per second (m³/s), with seasonal peaks during spring and early summer (April–June) due to snowmelt and rainfall in the Danube Basin. Minimum discharges typically occur during late summer and early autumn, and can drop below 2,000 m³/s in dry years. Water distribution within the delta is managed through three main branches: Chilia, Sulina, and Sfântu Gheorghe, with the Chilia branch carrying the largest share (more than 60% of the total flow). The variable discharges influence sediment transport, water levels, and the ecological dynamics of the delta’s wetlands. Human interventions, such as upstream dams and channel modifications, also impact natural flow patterns, affecting both flood risk and ecosystem health in the region.

The vulnerability of the Danube Delta to flooding can thus be determined on the basis of hydraulic modeling of the hydrological regime of the Danube, identifying both the extent of the flood band and the water depth within it. One of the well-known hydraulic modeling software is HEC-RAS (Hydraulic Engineering Center River Analysis System), developed by the US Army Corps of Engineers (Peker et al. 2024). This software was used, having a 2D configuration for the study of the flood hazard of the Danube Delta by delineating a number of three flood inundability bands for flows with 0.1%, 1% and 10% probabilities of occurrence calculated based on observations at the Ceatal Ismail station between 1971 and 2005.

A short review of the methods used to determine flood inundability in the Danube Delta shows that the techniques applied so far are based on methods such as hydro grades analysis (Mierla et al. 2015), detection of the extent of flood inundability bands using RADAR or optical imagery (Niculescu et al. 2015), as well as 1D hydraulic modeling (Bănescu et al. 2020). In contrast, the present study proposes, for the first time, the use of 2D hydraulic modeling to analyze an extended area of the Danube Delta.

In the next sections the methods used in the study are described together withdata sources used to configure the 2D model and to calculate the return periods for Ceatal Ismail cross-section, which is upstream of Danube Delta branching. Next, the results of the flood modeling and the analysis of the results are presented followed by conclusions and discussions.

The results of the simulations on the floodplains showed a possible evolution of the flooded area within the Danube Delta, depicted in Figs 5–7 and summarized in Table 1. An effective method of analyzing floodability is to compare the values for scenarios with similar return periods, in order to observe the variations in the intensity of the phenomenon. Thus, in the case of simulations for a return period of 100 years, often used in hazard analyses, a significant increase in the area covered by water is observed. In the reference period (1971–2005), this was approximately 2100 km².

Table 2 shows the absolute and percentage distribution of areas occupied by different water depth classes. For the 0–0.1 m class, the Q10-year flood occupies about 19.3% of the modeled area with an area of 346.5 km². At Q100 years, the percentage decreases to 10.4% and the area to 218,551 km². At Q1000 years, the percentage is 4.3% with an area of 94.1 km2. Here, a steady decrease in the share of this class is observed with increasing return period. This could indicate that shallower areas are more prone to frequent floods, but less affected by rare events. Continuing with the 0.1–0.2 m class, the percentages are 19.5% (Q10 years), 15.5% (Q100 years) and 9.2% (Q1000 years), and the corresponding areas are 349.9 km2, 324,851 km² and 203.2 km². Here, too, a decrease in the share is observed, but less accelerated than in the case of the previous class. The 0.2–0.5 m class is the most dominant for Q10 years with 24.0% and 426.9 km². At Q100 years, the percentage increases significantly to 38.5% and the area to 809,281 km², and at Q1000 years it decreases to 30.0% with 660 km². This suggests that this depth class is the most affected by floods with a 100-year return. The intermediate classes (0.5–0.75 m, 0.75–1 m) show interesting increases. For Q1000 years, the 0.5–0.75 m class increases to 21.1% (465.6 km²) compared to 7.7% (138.3 km2) for Q10 years, and the 0.75–1 m class increases to 8.6% (206.2 km²) compared to 5.0% (89.6 km²). These increases indicate a trend towards intermediate depths for rare events. The larger depth classes (1–1.5 m, 1.5–2 m, 2–3 m, 3–5 m and over 5 m) show smaller percentage variations, but still a slight increase in areas at Q1000 years compared to Q10 years. For example, the 2–3 m class increases from 5.0% (89.3 km²) to 7.9% (174.8 km²). This could signal that severe flooding affects deeper areas. For example, the 2–3 m class increases from 5.0% (89.3 km²) to 7.9% (1748.8 km²). This could indicate that severe flooding affects deeper areas. It is also important to note that the total flooded area increases from Q10 years (1792.2 km²) to Q100 years (2101.9 km²) and then to Q1000 years (~2203.3 km²), confirming that rarer events cover a larger area.

Figure 5. 

Flood hazard map in the Danube Delta for the 100-year return period determined by observations during 1971–2005.

Figure 6. 

Flood hazard map in the Danube Delta for the 10-year return period determined by observations during 1971–2005.

Figure 7. 

Flood hazard map in the Danube Delta for the 1000-year return period determined by observations during 1971–2005.

The study conducted through 2D hydraulic modeling with HEC-RAS software highlighted the dynamics of flood risk in the Danube Delta (Figs 5–7), for different return periods, taking into account the most accurate topography of the area. The main results indicate a significant increase in floodable areas with a decrease in the probability of occurrence of events (from 10 to 1000 years). For the 1000-year return period, the flooded area reached approximately 2203 km², compared to 1792 km² for decadal events. These differences highlight the increased vulnerability of the delta to extreme phenomena.

The analysis of water depth revealed that areas with intermediate depths (0.2–0.5 m) are predominantly affected by 100-year return period floods, while rare events (1000 years) extend the impact to greater depths (0.5–1 m) and larger areas. Also, the decrease in the share of shallow depth classes (0–0.1 m) at lower probabilities suggests that marginal areas are exposed more frequently, but with limited intensities.

The hazard maps generated constitute an essential tool for risk management, facilitating the identification of critical areas and the implementation of adaptation strategies. Ecological restoration of wetlands and policies to reduce the impact of climate change are confirmed as vital measures to protect local communities and the deltaic ecosystem.

The limitations of the study, such as the exclusion of marine influence and the variable resolution of the digital terrain model, indicate the need to improve future approaches by integrating additional factors (e.g., Black Sea level variations). Further research could include climate projections and socio-economic analyses to optimize adaptive management plans. In conclusion, the application of 2D modeling in HEC-RAS has demonstrated its efficiency in hydrological hazard assessment, providing a solid basis for policy decisions and sustainable protection of the Danube Delta.