Elsevier

Science of The Total Environment

Volume 562, 15 August 2016, Pages 666-677
Science of The Total Environment

Flood projections within the Niger River Basin under future land use and climate change

https://doi.org/10.1016/j.scitotenv.2016.04.021Get rights and content

Highlights

  • Modeling suggests increased floods of the Niger for future climate and land-use.

  • Dry areas of the Sahelian and Sudanian region show a particularly high sensitivity.

  • Projections for climate and land-use change result in diametrical impacts.

  • Modeled land-use change effects were smaller than climatic effects.

  • Ecohydrological modeling is a valuable tool to disentangle the impacts and drivers.

Abstract

This study assesses future flood risk in the Niger River Basin (NRB), for the first time considering the simultaneous effects of both projected climate change and land use changes. For this purpose, an ecohydrological process-based model (SWIM) was set up and validated for past climate and land use dynamics of the entire NRB. Model runs for future flood risks were conducted with an ensemble of 18 climate models, 13 of them dynamically downscaled from the CORDEX Africa project and five statistically downscaled Earth System Models. Two climate and two land use change scenarios were used to cover a broad range of potential developments in the region. Two flood indicators (annual 90th percentile and the 20-year return flood) were used to assess the future flood risk for the Upper, Middle and Lower Niger as well as the Benue. The modeling results generally show increases of flood magnitudes when comparing a scenario period in the near future (2021–2050) with a base period (1976–2005). Land use effects are more uncertain, but trends and relative changes for the different catchments of the NRB seem robust. The dry areas of the Sahelian and Sudanian regions of the basin show a particularly high sensitivity to climatic and land use changes, with an alarming increase of flood magnitudes in parts. A scenario with continuing transformation of natural vegetation into agricultural land and urbanization intensifies the flood risk in all parts of the NRB, while a “regreening” scenario can reduce flood magnitudes to some extent. Yet, land use change effects were smaller when compared to the effects of climate change. In the face of an already existing adaptation deficit to catastrophic flooding in the region, the authors argue for a mix of adaptation and mitigation efforts in order to reduce the flood risk in the NRB.

Graphical abstract

Fig. 1 Projected changes in flood magnitudes in the Niger River Basin between a near future period (2021–2050) and the base period (1976–2005 driven with 18 Regional Climate projections. The results of a limited emission (RCP 4.5) (top left) and a high emission (RCP 8.5) (top right) scenario are depicted as medians for each sub-region as arrow, with red for increasing and green for decreasing trends. The angle of the arrows are relative to the highest increase (Middle Niger limited emission: + 30.4%  90°). For land use and land cover changes a regreening scenario (bottom left) and a business as usual scenario (bottom right) with strong agricultural expansion are depicted as changes from the mean of both climate scenarios. The angles of the arrows are relative to the highest decrease (Middle Niger regreening: − 21.2%   90°).

Image 1
  1. Download : Download high-res image (217KB)
  2. Download : Download full-size image

Introduction

Catastrophic flooding in the Niger River Basin (NRB) is increasingly perceived as a major threat, affecting people at a scale of millions (Aich et al., 2014a, Tarhule, 2005, Tschakert et al., 2010). Simultaneously increasing vulnerability, population growth and discharges in absolute terms were identified as reasons for this augmented flood risk (e.g. Aich et al., 2014a, Descroix et al., 2012, Di Baldassarre et al., 2010). A study of Jury (2013) for the entire Sub-Saharan African region found a general return to wetter conditions when looking at streamflows during the last decades, including the Niger River. More specifically for the Niger River, Tarhule et al. (2015) found increasing flows in all parts since the extremely dry period of the 1970s and 1980s. However, for most regions of the NRB, the water levels do not reach the levels observed before this dry period (e.g. Paturel et al., 2003). Hydrological projections for the West African region have recently been reviewed by Roudier et al. (2014). For the NRB, they found a tendency towards increasing river flows with a mean increase among all studies above 5%, yet with strong variability and high spatial heterogeneity. Roudier et al. (2014) concluded that a general agreement exists on the region's high vulnerability to climate change. However, this vulnerability analysis lacks projections of future floods. In addition, they state that there is an urgent need to take into account other factors influencing runoff, especially water consumption and land use changes, in order to acquire a more comprehensive flood risk assessment. Especially for the drier parts of the basin, in the Sahelian Zone, Amogu et al. (2010) and Descroix et al. (2013) and Séguis et al. (2004) identified Land Use Land Cover (LULC) change as the main driver of the increasing flood magnitudes. To assess these past changes, Aich et al. (2015) used a modeling attribution approach to quantify the share of LULC change and climatic variability responsible for the recent flood increases and concluded that LULC and climatic changes have approximately equal shares affecting the flood increase. Future flood projections for the NRB should therefore include both drivers, climate and LULC, in order to support the elaboration of sound flood mitigation and adaptation strategies.

The present study intends to assess this research gap by addressing the question of how flood magnitudes in the Niger basin will evolve under the influence of expected climatic and LULC changes. In order to answer this question, projected changes in LULC and climate dynamics are analyzed first and, based on these scenarios, a process-based, ecohydrological model is employed to assess future high flows. The number of modeling studies focusing on floods has increased in the recent past (Bellu et al., 2016, Chen et al., 2012, Kundzewicz et al., 2013, Vansteenkiste et al., 2014), as has the corresponding methodology for calibration and validation with its specific focus on high flows (Crochemore et al., 2015, Giuntoli et al., 2015, Hattermann et al., 2014, Kay et al., 2015, Vetter et al., 2015, Westerberg et al., 2011). The selected Soil and Water Integrated Model (SWIM) has proved to be generally able to simulate high flows (95th and 99th percentile discharges), flood events (10- and 30-year return periods) and even extreme floods (50-year return period) well ((Hattermann et al., 2011, Krysanova et al., 2015). However, regional characteristics and data scarcity can impede or hinder such exercises. In order to judge the model's efficiency for high flows and floods, a thorough validation for this specific feature of the flow regime is necessary. Crochemore et al. (2015) and Kay et al. (2015) recently provided measures for a sound validation of model ability which are also applied in this study.

The experimental setup includes two scenarios for the climate and LULC, respectively, which have been combined in the study. In addition to these four scenario combinations, stable LULC as of 2005 has also been combined with both climate scenarios, allowing a quantification of the impact of climate on its own. In total, this results in six different scenario combinations. Due to the huge extent and diversity of the study area, four different example regions of the NRB are analyzed (Fig. 1). The projections concentrate on the near-future period of 2021–2050, because of its specific relevance for adaptation measures. In particular, the near-future period considered for the analysis corresponds approximately to the occurrence of + 2 °C global warming with respect to the pre-industrial period. The + 2 °C global temperature increase is still considered a critical guidance for mitigation strategies, in order to avoid the most severe socio-economic impacts of climate change. The results are being discussed and interpreted, taking into account the uncertainties. Finally, implications for the management of future flood risks in the NRB are discussed.

Section snippets

Study area

The NRB in West Africa covers an area of approximately 2,156,000 km2, of which only approximately 1,270,000 km2 contribute to the river discharge (Fig. 1). Guinea, Ivory Coast, Mali, Burkina Faso, Benin, Niger, Chad, Cameroon and Nigeria fall within this active basin. These countries are members of the Niger Basin Authority, an intergovernmental body which coordinates the management and development of the NRB's water resources, including the development and implementation of a joint flood policy.

Climate

The historical simulations of the CORDEX Africa models were evaluated against station data by Kim et al. (2013) and Nikulin et al. (2012). They stated that, in general, all RCMs provide a reasonable description of the observed climatology and that the mean of the ensemble outperformed the individual models. They recommended, especially for impact studies, an evaluation of the models' performance individually for each region and all relevant parameters in order to assess the uncertainty. This

Flood trends

The results imply mostly an increase in flood magnitudes in the various regions of the Niger basin. These findings complete and confirm the overall picture of a return to wet conditions in West Africa, as found by Tarhule et al. (2015), Aich et al. (2014b), Jury (2013) and Roudier et al. (2014). Diverging from our approach, in a modeling study by Hirabayashi et al. (2013), global climate models of RCP 8.5 were used to drive a global river routing model without considering LULC. They found a

Conclusions

The simulation study projects that the risk of catastrophic floods in the modeled sub-catchments of the NRB is likely to increase in most climate and LULC scenarios. It also stresses the need for and importance of detailed regional studies in order to understand and quantify the at times opposing or counterbalancing impacts of climate and LULC changes in different vegetational and climatic settings of the NRB.

The modeling results come with substantial uncertainty. Still, the great majority of

Acknowledgments

We thank the IMPACT2C (Grant Agreement 282746) project for financing this study and the Niger Basin Authority for providing data.

References (79)

  • J.E. Nash et al.

    River flow forecasting through conceptual models part I — a discussion of principles

    J. Hydrol.

    (1970)
  • D. Ramier et al.

    Towards an understanding of coupled physical and biological processes in the cultivated Sahel — 1. Energy and water

    J. Hydrol.

    (2009)
  • J. Schuol et al.

    Estimation of freshwater availability in the West African sub-continent using the SWAT hydrologic model

    J. Hydrol.

    (2008)
  • T. Vansteenkiste et al.

    Intercomparison of five lumped and distributed models for catchment runoff and extreme flow simulation

    J. Hydrol.

    (2014)
  • V. Aich et al.

    Floods in the Niger basin — analysis and attribution

    Nat. Hazards Earth Syst. Sci. Discuss.

    (2014)
  • V. Aich et al.

    Comparing impacts of climate change on streamflow in four large African river basins

    Hydrol. Earth Syst. Sci.

    (2014)
  • V. Aich et al.

    Climate or land use?—attribution of changes in River flooding in the Sahel zone

    Water

    (2015)
  • O. Amogu et al.

    Increasing River flows in the Sahel?

    Water

    (2010)
  • I. Andersen et al.

    The Niger Basin: A vision for Sustainable Management, the World Bank; Directions in Development

    (2005)
  • J.C.M. Andersson et al.

    Improving crop yield and water productivity by ecological sanitation and water harvesting in South Africa

    Environ. Sci. Technol.

    (2013)
  • ArcSWAT

    ArcSWAT | ArcGIS-ArcView extension and graphical user input interface for SWAT

    (2012)
  • A. Awotwi et al.

    Assessing the impact of land cover changes on water balance components of White Volta Basin in West Africa

    Water Environ. J.

    (2014)
  • E. Bartholomé et al.

    GLC2000: a new approach to global land cover mapping from Earth observation data

    Int. J. Remote Sens.

    (2005)
  • D.F. Bauer

    Constructing confidence rank sets using statistics

    J. Am. Stat. Assoc.

    (1972)
  • K. Beven et al.

    The uncertainty cascade in flood forecasting

  • X. Chen et al.

    Uncertainty intercomparison of different hydrological models in simulating extreme flows

    Water Resour. Manag.

    (2012)
  • S. Coles et al.

    An Introduction to Statistical Modeling of Extreme Values

    (2001)
  • L. Crochemore et al.

    Comparing expert judgement and numerical criteria for hydrograph evaluation

    Hydrol. Sci. J.

    (2015)
  • L. Descroix et al.

    Impact of drought and land – use changes on surface – water quality and quantity: the Sahelian paradox

  • G. Di Baldassarre et al.

    Flood fatalities in Africa: from diagnosis to mitigation

    Geophys. Res. Lett.

    (2010)
  • J. Doherty

    PEST—Model-Independent Parameter Estimation, User Manual

    (2005)
  • EEA (European Environment Agency)

    Precipitation extremes [WWW document]

  • FAO et al.

    Harmonized world soil database v 1.2 [WWW document]

  • G. Favreau et al.

    Land clearing, climate variability, and water resources increase in semiarid southwest Niger: a review

    Water Resour. Res.

    (2009)
  • E. Gilleland

    extRemes: Extreme Value Analysis

    (2012)
  • I. Giuntoli et al.

    Future hydrological extremes: the uncertainty from multiple global climate and global hydrological models

    Earth Syst. Dyn.

    (2015)
  • F.F. Hattermann et al.

    Model-supported impact assessment for the water sector in Central Germany under climate change—a case study

    Water Resour. Manag.

    (2011)
  • F.F. Hattermann et al.

    Modelling flood damages under climate change conditions — a case study for Germany

    Nat. Hazards Earth Syst. Sci.

    (2014)
  • B. Hewitson et al.

    CORDEX-Africa: A Unique Opportunity for Science and Capacity Building

    (2012)
  • Cited by (92)

    View all citing articles on Scopus
    View full text