Elsevier

Environmental Science & Policy

Volume 89, November 2018, Pages 163-175
Environmental Science & Policy

Impacts of climate and land use changes on flood risk management for the Schijn River, Belgium

https://doi.org/10.1016/j.envsci.2018.07.002Get rights and content

Highlights

  • A sound understanding of climate change is required in support of current flood risk management.

  • The hydrological behaviour of the Schijn River, Belgium, was studied from urbanization and climate scenarios perspectives.

  • Peak flows for rural and urban runoff were significantly higher than existing rainfall-runoff model (PDM) flows because of faster and more peaked urban runoff response.

  • The impact of climate change on current and future conditions was assessed by estimating peak flows with respect to return periods from the flood frequency curve.

  • Climate change impacts were found to contribute the most in producing peak flows in coming years, while increased urbanization takes the second place for both hourly and daily values.

  • Results on urbanization effect and climate change impact assessment are useful to the water managers for spatial planning, emergency planning and insurance industry.

Abstract

Flooding is the most common natural disaster in Europe. Modern flood risk management relies not only infrastructure development but also on governmental and non-governmental actors applying legal, economic and communicative water management instruments. Within the European Union (EU), flood management closely relies on policy set at the EU and national levels. It is now recognized that a sound understanding of climate change is required in addition to current management by taking into account land use change and socio-political context, as climate and land use changes have major impacts on hydrological responses.

This paper investigates the hydrological behavior due to urbanization under current and future climate scenarios of high summer and high winter rainfall for 20 sub-catchments of the Schijn River, located in the Flanders region near Antwerp, Belgium. As urbanization increases and existing rainfall-runoff models neglecting the specific behavior of urban runoff, a hydrological model was developed based on a basic reservoir concept and applied to the existing rainfall-runoff model (PDM) flow to examine the specific urban contribution. Results revealed that peak flow for urban runoff and the total peak flow (i.e. rural and urban runoff) were significantly higher (i.e. ranges from 200% to 500%) than the existing rainfall-runoff model (PDM) flows, because of faster and more peaked urban runoff response. The impact of climate change on current and future conditions was also assessed by estimating peak flows with respect to return periods from the flood frequency curve. The predicted peak flow of high summer future climate scenario was significantly higher (i.e. ranges from 200% to 250%) than that of the current climatic condition for this region. Furthermore, hourly peak flow and daily volume ratios of 100-year return period for the highest, lowest and average impervious area were projected for the time horizon of the year 2100. It is concluded that climate change impacts contribute the most in producing peak flow in coming years, while increased urbanization takes the second place for both hourly and daily values. Results on urbanization effect and climate change impact assessment are useful to the water managers for spatial planning, emergency planning and insurance industry.

Introduction

Flooding is the greatest economic natural disaster in Europe (Guha-Sapir et al., 2013) via damage and property and infrastructure, as well as physical injury and loss of life. As discussed below, EU flood policies took their roots in more than 100 major floods which occurred in the years 2000–2005 in Europe, among those, 9 floods were classified as extreme (Barredo, 2007). Major flood events resulted in 155 casualties and economic losses of more than €35 billion (Barredo, 2007). Material damage by floods in Europe in 2002 is estimated to be higher than in any previous year (Barredo, 2007). Damages caused by extreme floods have increased more than double in the last 50 years (Munich Re, 2005). (Feyen et al., 2009) estimated that economic losses caused by flooding in the EU are €6.5 billion per year, while the estimated annual damage is projected to rise to at least twice this amount by the end of this century. In May and June 2013, an extreme flood hits Central Europe in the Elbe and Danube River catchments and caused the highest water levels ever recorded (ICPDR, 2014). Subsequently, these floods highlighted the challenges related to Flood Risk Management (FRM) and fuelled the necessity for effective action programmes driven by policy in Europe. FRM is defined as a process of ‘holistic and continuous societal analysis, assessment and mitigation of flood risk’ (Schanze, 2006; Merz et al., 2010). It aims at managing the whole flooding system to reduce flood risks and providing environmental, social and economic benefits both for present and future (Sayers et al., 2014). In this case, accurate and updated data is necessary for decision-making and that’s why the implementation of FRM strategies is quite challenging for practitioners, policy makers and researchers.

The DPSIR model (EEA, 1999) is widely used to conceptualize environmental changes which set risk management rules. In this context, socio-economic developments are the driving forces (D), leading to environmental pressures (P) such as, increasing temperature and precipitation, which themselves lead to changes in environmental state (S) such as, inundation and flood, impact (I) refers to the effects on the environment of the pressures that are exercised on the system such as damage of property, ecosystem and loss of life, and response (R) consists of the actions taken to improve the status of the system by the society or policy makers such as strict rules for construction, maintaining natural floodplain etc. Some potential drivers of change are identified by (Merz et al., 2014) in Fig. 1.

Land use changes such as shifts from forestry to agriculture, from pasture to arable land, from rain fed to irrigated agriculture or from agricultural use to urbanized areas act as drivers for changes (EEA, 2016). Climate change impacts are increasingly considered in flood management along with other drivers such as, land cover changes and increasing water demand (Quevauviller, 2011).

In this respect, Global Climate Models (GCM) and Regional Climate Models (RCM) have shown that the magnitude and frequency of high precipitation extremes are likely to increase for Northern Europe and for Central and Southern Europe in winter (Dankers and Feyen, 2008; IPCC, 2013). For 2071–2100, projected precipitation extremes highlight an increase in Northern Europe, especially during winter (Kundzewicz et al., 2013) leading to increased flooding across most of North, Central and Eastern Europe (Lehner et al., 2006). Decreased flooding is projected for some parts of Central and Southern Europe (Dankers and Feyen, 2008). Alfieri et al., (2015) report that floods with return periods of 100 years are projected to increase double in frequency within 3 decades.

At the European Union (EU) level, the water policy is governed by the Water Framework Directive (WFD), which aims to achieve good status for all waters in Europe (European Commission, 2000). The 2015 objectives have only been partly achieved in the 1st River Basin Management Plan (RBMP, 2009–2015) and are being now pursued in the 2nd RBMP (2015–2021). Flooding was not explicitly addressed in the WFD, nor climate change or its impacts. RBMPs represent the water management instrument and implementation of the WFD in all EU Member States. While climate change was not considered in the first cycle of RBMP (2009–2015), it has gradually been introduced in the policy discussions. In particular, climate impact has been discussed from 2009 onward through Common Implementation Strategy (CIS), composed of policy makers, experts from the Member States and of the European Commission (CIS, 2009) and recommended to integrate this dimension into the second (2015–2021) and third (2021–2027) cycles of RBMP (European Commission, 2013) to meet WFD goals under future projected climatic conditions. This approach is a kind of climate-proofing of the water policy (Quevauviller, 2014).

Recognizing the continued risks of flooding, specifically after the most devastating flood event in Central Europe in August 2002 and at the request of the EU Member States, the EC proposed the Flood Directive (FD) to set rules for the risk assessment and management of flooding (European Commission, 2007) in Europe. Complementing the WFD, the FD aims at reducing the adverse consequences of floods to human health, the environment and economic activity, taking into account the future changes in the risk of flooding as a result of climate change. Three steps are described in the EU Flood directive - preliminary flood risk assessment, flood hazard and risk map and FRM planning (Fig. 2).

As flood risk is not constant over time, flood risk maps and plans need to be revised every 6 years (De Moel et al., 2009) corresponding to the RBMP cycle. The principal information on flood and FRM at EU level is based on the reporting under the FD, which contains the Flood Hazard and Risk Maps and the draft of FRMPs i.e. flood-related action programs have to be embedded into the second RBMP (European Commission, 2015a). So the preliminary flood risk assessment would ideally consider climate change impacts and urbanization, that would have an impact on flooding consequences (European Commission, 2015b).

In this regard, many studies have been focused on the assessment of the impacts of future trends of land use and climate change at different scales on flood risk management (Klijn et al., 2012; Feyen et al., 2012; Alfieri et al., 2016). Most studies focused on the future flood risk assessment and estimation of flood related damages (De Kok and Grossmann, 2010; Feyen et al., 2009). Recent studies have investigated climate change impacts on hydrological responses e.g. (Ashraf Vaghefi et al., 2014; Teferi et al., 2015) indicating that climate change has a larger impact on the hydrological cycle than land-use change for the catchments like Schijn River (Kim et al., 2013; Khoi and Suetsugi, 2014). Therefore, it is indispensable to investigate the hydrological responses to land use changes under future climatic conditions. A variety of hydrological models, e.g., VHM (Willems, 2014), NAM (DHI, 2007), PDM (Moore, 2007) were used to simulate rainfall-runoff processes on a basin scale and used for climate change impact assessment.

Science policy interfacing plays an important role at a number of levels within the WFD context, including national, regional and local policy implementation. Several steps were undertaken by the EU to improve the interaction between science and policy through EU-funded research projects, which are considered to be an essential support to policies, particularly in the water sector (Quevauviller, 2010a). The FLOOD Site Integrated Project was carried out in 2004–2009 at the EU level and provided a solid knowledge basis for the development of the Flood Directive. This was complemented by improved knowledge on climate change impacts gained within the FP6 WATCH project and regional assessments with the CIRCE project. Further research actions focused on specific issues such as flash floods (IMPRINTS project), resilience (CORFU project), and governance (STARFLOOD project) etc. All these projects are described and referenced in Quevauviller et al., (2012b). The STARFLOOD project has informed the policy community about progress on FRM at EU level by regular contacts with the CIS Working Groups on floods, including scientists, flood risk managers and stakeholders. Thus, science-policy interactions seem to be an essential component (Quevauviller, 2010b) in the EU regulatory framework and need to be implemented in the second RBMP (2015–2021) and onwards.

Floods occur in Belgium mainly due to tides and fluvial and pluvial runoff (Zagonari, 2013). In Flanders, the percentage of built-up land is expected to increase to 30–50% of land use (Poelmans et al., 2010) by 2050. Between 1976 and 2000, the urbanization ‘sealing’ process increased surface runoff by ∼ 20% and with future developments, it will only increase (Poelmans et al., 2010). Annual precipitation has increased by 0.55 mm/year with a total increase of 94 mm/year in Flanders from 1833 to 2014, while the number of days with heavy precipitation has doubled since the 1950 s (VMM, 2015a). On average, about 800 mm rainfall annually falls in Low and Mid-Belgium and up to 1400 mm in High-Belgium (the Royal Meteorological Institute of Belgium (KMI)), 2013. There is already risk of urban pluvial flooding (e.g. extreme precipitation with heavy thunderstorms) in Antwerp (Uytven et al., 2015). Over 220,000 peoples are directly affected by flooding (VMM, 2015b). More than 67,000 citizens live in the area with medium probability of flooding (i.e. once in 100 years) and about 10,000 reside in the area with major probability of flooding (i.e. once in 10 years. The average annual flood damages in Flanders region of Belgium currently have exceeded 50 million Euros (VMM, 2015b). The Schijn River Valley and the Antwerp suburbs will always be more sensitive to pluvial than fluvial flooding due to the increased rate of urbanization and river systems with specifically dimensioned pumping stations and relatively open valleys. Moreover, climate change will lead to more likely frequent extreme summer storms so this effect is expected to increase in the future.

Hydrodynamic models (i.e. hydrology and hydraulics) are required to prepare flood risk maps. To fulfill the requirements of the EU Flood Directive by generating the flood risk maps, the PDM model (CEH Wallingford, 2000) is used as the hydrological component in the Info Works River System software (Innovyze) that simulates stream flow in the River. This PDM model works well for rural environments, but in urban environments, the urban drainage networks lead to faster runoff, creating a bi-modal response i.e. fast runoff from impervious areas and slow runoff from pervious areas (Dahl et al., 1996). In order to estimate river flows, both rural and urban runoff need to be calculated separately owing to large differences of the peak flows observed between rural and urban runoff (Dahl et al., 1996). This can be solved by using separate models for the urban and rural runoff in the same catchment by adopting an urban boundary (Vaes et al., 2009), an implementation of the Remuli model (Vaes,1999). Application of this urban boundary (Infoworks RS, Innovyze) can make a significant difference as e.g. observed on the Mandel River in Roeselare (Vaes et al., 2009).

In the past, fully integrated models, including urban drainage and river systems, as well as overland flow were built (e.g. Woluwe basin (HydroScan, 2016)). However, most currently available models include only river hydrology and hydraulics and do not include urban drainage network because the responsible river authorities are up till now only interested in the winter flooding from the rivers and not in the interactions with the urban drainage systems which are the responsibility of other authorities. Moreover, these detailed integrated models used for rivers as well as urban drainage require much more efforts and calculation time. So, it is chosen here to use very simple models with continuous long term simulations to maintain the first order accuracy (Vaes et al., 2001). This study demonstrated the influence of urban areas on the hydrological model flow for the Schijn River.

The Flemish legislature established the ‘Co-ordination Committee on Integrated Water Policy (CIW)’ to implement the WFD by adopting a Decree (18 July 2003). Prior to the EU Flood Directive, the CIW already introduced the FRMPs in Flanders through water assessments. The implementation of the FD just strengthened its instruments (e.g. water test1 by the water manager for every building permit within flood prone areas, duty to inform, signal areas and re-parceling with land swap) as well as measures (e.g. emergency intervention plan and general fire insurance). So, Flanders opted to skip the first phase of FD i.e. the preliminary flood risk assessment. They were able to make use of existing flood risk assessments and flood risk and hazard maps subject to certain conditions described in the Flood Directive (European Commission, 2012). Later on, Flanders decided to form an expert group within the CIW, which is responsible for flooding and cooperate with the working group of EC. Thus, the CIW plays the central role in planning and execution of water policy at the River basin level in Flanders. Taking account of reports required under the EU flood directive, flood hazard and risk maps for the Schijn River were published by the Flemish Environmental Agency (VMM) for low, medium or high probability in the case of the current climate, and a medium and high climate change scenario (VMM, 2016). Hence, the VMM can be considered as the active motor of the CIW in the integration of water management and spatial planning.

This paper presents an analysis of the existing flood risk management practices in the Schijn River based on the EU policy framework, assesses the peak flows under different scenarios taking into account both rural and urban runoff, evaluates the impacts of climate change on flood risk by analyzing peak flows and proposes new policy features and adaptation measures for flood risk management in the studied area.

Section snippets

Description of the study area

The international Scheldt River originates in France, crossing Belgium and the Netherlands and discharging in the North Sea (Fig. 3). The River has a tidal influence, covering a gradient from salt to brackish to freshwater areas (Cox et al., 2006). The study sites are located on the Schijn River, a tributary of the Scheldt River. This small River with a mean discharge of 0.67 m3/s, flows through the highly urbanized northern part of Belgium and enters the Scheldt River in the city of Antwerp (

Analysis of flood risk management practices in the Schijn River

The prominent paradigm for Flanders is ‘creating space for water’ and the basic principle of ‘retaining, storing and discharging’ moved towards ‘Multi-Layer Water Safety’ (MLWS). The MLWS focuses on the 3 P approach - prevention, protection and preparedness in FRM that brings shared responsibilities among actors. The Schijn River is classified as non-navigable category managed by VMM.

To manage flood risk for the Schijn River, there are four concepts such as, avoiding downstream flooding,

Conclusions and recommendations

For an effective FRM policy, the impact of land use and climatic change on flood risk for the Schijn River in Belgium was investigated. The following conclusions can be drawn from this study.

Urbanization has increased significantly since the last decades along the Schijn River, reaching one-third to half of the sub-catchment area. A high degree of land sealing decreases rainwater infiltration, thereby making it more vulnerable to flooding. The integrated hydrological model clearly shows that

Acknowledgement

The authors would like to thank the VMM for the catchment and rainfall data and the PDM models that was used to perform this study.

References (75)

  • S. Ashraf Vaghefi et al.

    Analyses of the impact of climate change on water resources components, drought and wheat yield in semiarid regions: Karkheh River Basin in Iran

    Hydrol. Process.

    (2014)
  • J.I. Barredo

    Major flood disasters in Europe: 1950-2005

    Nat. Hazards

    (2007)
  • V. Bruggeman

    De schadeloosstelling van slachtoffers van natuurrampen: België als wenkend voorbeeld?

    Milieu-en Energierecht

    (2010)
  • V.T. Chow et al.

    Applied Hydrology

    (1988)
  • CIS

    Common Implementation Strategy for the Water Framework Directive, River Basin Management in a Changing Climate

    (2009)
  • A. Dahl et al.

    Joint probability of flooding

  • R. Dankers et al.

    Climate change impact on flood hazard in Europe: an assessment based on high-resolution climate simulations

    J. Geophys. Res. Atmos.

    (2008)
  • J.L. De Kok et al.

    Large-scale assessment of flood risk and the effects of mitigation measures along the Elbe River

    Nat. Hazards

    (2010)
  • H. De Moel et al.

    Flood maps in Europe - methods, availability and use

    Nat. Hazards Earth Syst. Sci.

    (2009)
  • H. De Moel et al.

    Flood risk assessments at different spatial scales

    Mitig. Adapt. Strat. Glob. Change

    (2015)
  • N. Devroede et al.

    Flood risk management planning in Flanders

    Compr. Flood Risk Manag.

    (2013)
  • DHI

    MIKE11, A Modeling System for Rivers and Channels. Reference Manual

    (2007)
  • EEA

    Environmental indicators : typology and overview

    Eur. Environ. Agency

    (1999)
  • EEA

    Flood Risks and Environmental Vulnerability - Exploring the Synergies Between Floodplain Restoration, Water Policies and Thematic Policies. EEA Rep. No 1/2016

    (2016)
  • T. Endreny

    Rainfall and runoff

    Water Encyclopedia

    (2005)
  • European Commission

    Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy

    Off. J. Eur. Parliam.

    (2000)
  • European Commission

    Directive 2007/60/EC of the European parliament and of the council of 23 October 2007 on the assessment and management of flood risks

    Off. J. Eur. Union

    (2007)
  • European Commission

    Assessment of Data and Information Reported by Member States on Their Preliminary Flood Risk Assessments and Identification of Areas of Potentially Significant Flood Risk Under the Floods Directive, Member State Report: Belgium

    (2012)
  • European Commission

    Communication From the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions - an EU Strategy on Adaptation to Climate Change

    (2013)
  • European Commission

    Commission Staff Working Document-report on the Progress in Implementation of the Floods Directive - Accompanying the Document- Communication From the Commission to the European Parliament and the Council

    (2015)
  • European Commission

    Commission Staff Working Document- Report on the Progress in Implementation of the Floods Directive- Accompanying the Document- Communication From the Commission to the European Parliament and the Council, the Water Framework Directive and the Flood Direc

    (2015)
  • L. Feyen et al.

    Implications of global warming and urban land use change on flooding in Europe

    Water & Urban Development Paradigms - Towards an Integration of Engineering, Design and Management Approaches

    (2009)
  • L. Feyen et al.

    Fluvial flood risk in Europe in present and future climates

    Clim. Change

    (2012)
  • FGIA, 2016....
  • D. Fotakis et al.

    Integration of a hydrological model within a geographical information system: application to a forest watershed

    Water (Switzerland)

    (2014)
  • D. Guha-Sapir et al.

    Annual Disaster Statistical Review 2013: the Numbers and Trends

    (2013)
  • HydroScan

    Impact of the Changing Rainfall Patterns on Local Rivers (in Dutch). Study for the City of Antwerp

    (2016)
  • Cited by (76)

    • Quantifying the flood risk index of the Malaysian “rice bowl”

      2023, Journal of Hydrology: Regional Studies
    View all citing articles on Scopus
    View full text