Elsevier

Land Use Policy

Volume 99, December 2020, 104868
Land Use Policy

The effects of changing land use and flood hazard on poverty in coastal Bangladesh

https://doi.org/10.1016/j.landusepol.2020.104868Get rights and content

Highlights

  • We present a spatiotemporal appraisal of poverty in the coastal zone in Bangladesh.

  • Flood risk is positively correlated with land use/land cover (LULC) change.

  • The expected annual damage of 2005 is estimated to be more than double by 2030.

  • Flood risk and patterns of LULC change have a negative effect on wealth index.

  • The rate of increase of wealth index is likely to be low in the future.

Abstract

The construction of polders in the coastal region of Bangladesh has significantly modified the patterns of flooding, as well as leading to significant land use/land cover (hereinafter, LULC) changes. The impact of LULC change and flooding on poverty is complex and poorly understood. This study presents a spatiotemporal appraisal of poverty in relation to LULC change and pluvial flood risk in the south western embanked area of Bangladesh. A combination of logistic regression (LR), cellular automata (CA), and Markov Chain models were utilised to predict future LULC based on historical data. Flood risk assessment was performed at present and for future LULC scenarios. A spatial regression model was developed, incorporating multiple parameters to estimate the wealth index (WI) for present-day and future scenarios. In the study area, agricultural lands reduced from 34 % in 2005 to 8% in 2010, while aquaculture land cover increased from 17 % to 39 % during the same time. The rate of LULC change was relatively low between 2010 and 2019. Based on the recent trend, LULC was predicted for the year 2030. Flood risk was positively correlated with LULC and the expected annual damage (EAD) was estimated at $903 million in 2005, which is likely to increase to $2096 million by 2030, considering changes in LULC scenarios. The analysis further showed that the EAD and LULC change were negatively associated with the WI. Despite consistent national GDP growth in Bangladesh in recent years, the rate of increase of WI is likely to be low in the future because flood risk and patterns of LULC change have a negative effect on WI.

Introduction

It is widely recognised that poor people are disproportionately exposed to environmental hazards (Winsemius et al., 2018). There are several possible reasons for this. For instance, poor people tend to inhabit remote low-lying floodplains, due to the limited development opportunities and relatively cheaper lands (Dasgupta, 2007). Their livelihoods and assets are less protected (Bangalore et al., 2019; Hossain et al., 2012), and thus, they have relatively a low capacity to cope with property losses resulting from flooding (Brouwer et al., 2007).

Bangladesh is located in the floodplain of three major rivers — the Ganges, Brahmaputra, and Meghna. The combined discharge generated of these three rivers is the highest in the world. The peak run-off depth is also the highest, which, combined with storm surges generated from the Bay of Bengal. This makes a major portion of the country is prone to flooding (Dasgupta, 2007). Flood processes in the coastal region of Bangladesh are complex, as it can occur from multiple sources such as intense precipitation during the monsoon, high water levels in the rivers, and cyclone induced storm surges (Adnan et al., 2019). Different environmental stresses create biophysical and socioeconomic challenges in the coastal region. For instance, frequent flooding and increasing soil salinity limit agricultural productivity, which is the main source of livelihoods in coastal Bangladesh (Rahman et al., 2020).

Flood management approaches in the coastal region of Bangladesh include both structural and non-structural measures (Paul and Rashid, 2017; Rahman and Salehin, 2013). Major surge events induced by cyclones in the 1950s forced the then government to invest in the Coastal Embankment Project (CEP) in the 1960s. The CEP aimed at increasing agricultural production to ensure food security, by preventing salinity intrusion in the coastal region particularly during the dry season. As a part of the CEP, 139 polders (enclosed coastal embankments) were created in between the 1960s and 1980s (Islam et al., 2016; Warner et al., 2018). The construction of the polders has brought both beneficial and harmful effects on society and the environment. The protection from flooding afforded by embankments led to an increase in agricultural productivity until the 1980s (Adnan et al., 2020). Embankments have demonstrably protected the polder area against storm surges and fluvio-tidal floods of moderate severity (Adnan et al., 2019). However, the separation of floodplains from adjacent rivers caused geomorphological changes in the polder areas, exacerbating land subsidence inside polders (Auerbach et al., 2015). Accelerated land subsidence and inadequate drainage are accountable for frequent pluvial flooding (locally called ‘waterlogging’) (Adnan et al., 2019).

Generally, the construction of structural flood control measures, such as polders, shapes the pattern of human settlements and land use, which in turn impacts the extent of flood risk. Such flood control measures create the so-called “levee effect” (White, 1945). Whilst people tend to settle in less flood-prone areas, presence of structural flood defence system encourages floodplain development by engendering a sense of safety (Di Baldassarre et al., 2013; Montz and Tobin, 2008). Therefore, the failure of structural systems in the form of overtopping or breaching of embankments may exacerbate flood damages (Hui et al., 2016).

The pattern of land use/land cover (LULC) in the coastal region of Bangladesh has experienced major changes over the past half-century, following the construction of polders (Abdullah et al., 2019; Huq et al., 2015; Khan et al., 2015; Parvin et al., 2017; Rahman et al., 2017). Such changes largely occurred due to frequent and diverse natural hazards (e.g., floods) and increases in inundation, soil salinity, and land erosion (Brouwer et al., 2007; Khan et al., 2015). For instance, about 1% of agricultural land along the south western coast was transformed into non-agricultural use in each year over the past four decades due to the occurrence of frequent flooding (Rahman et al., 2017). The transformation of agricultural land to shrimp culture has been a common practice in the area since the 1980s as it can be more profitable (Khan et al., 2015). However, such land transformation has reportedly been leading to an increase in soil salinity, reducing agricultural production (Khan et al., 2015; Rahman et al., 2017).

Whilst anthropogenic drivers profoundly change the pattern of LULC, such transformation of land may affect local flooding processes (Wheater and Evans, 2009). The pattern of LULC determines the amount of runoff generated during a precipitation event, thus, influencing the water balance in an area. Hence, LULC may affect both the probability of flooding and its consequences (McColl and Aggett, 2007; Szwagrzyk et al., 2018). Flood losses are not only dependent on extreme hydro-meteorological conditions of a region, unplanned land use can multiply property damages (Lee and Brody, 2018). In coastal Bangladesh, unplanned LULC change may lead to environmental degradation such as soil salinization, disappearance of seasonal lagoons, and deterioration of water quality by increasing salinity (Islam et al., 2015).

Generally, flooding and poverty coexist particularly within rural communities, as damages caused by recurring flood events deplete assets, negatively impact agricultural incomes and thus lower quality of life of communities (Dube et al., 2018). It has been hypothesised that increasing flood risk and unplanned LULC change may create a poverty trap in the coastal region of Bangladesh (Ahmed, 2018; Borgomeo et al., 2017), inhibiting long-term development prospects (Parvin et al., 2017). Marginalised farmers could not generate adequate income through agricultural activities, whilst being unable to transform their agricultural land into aquaculture due to high cost associated with such change (Islam et al., 2015). As a result, they are unable to migrate out of such areas due to social and economic constraints and related costs (Dasgupta, 2007).

Regulating LULC change is an intervention to reduce flood risk, which has been adopted in different coastal cities (Adnan and Kreibich, 2016). Therefore, it is essential to understand the association between LULC and flood risk. Risk-based flood management approaches have received attention globally due to recent experience of several catastrophic events in many regions across the world (Hall et al., 2015, 2003b; Poussin et al., 2015), as well as the projected increase in the frequency and severity of flooding due to climate change-induced sea level rise (Koks, 2018). An empirical analysis of flood risk can support decision-makers to appraise and sequence investments for flood management (Dawson et al., 2011; Hall et al., 2003a, 2019; Hino and Hall, 2017; Sayers et al., 2002). The methods used in research and practice for quantifying flood hazard and vulnerability range from simple approaches (with numerous simplifying assumptions) to very complex applications, which are both data and time-intensive and computationally expensive (Apel et al., 2009; Dewan, 2013).

In the existing literature, the association between flood risk and poverty has been comprehended primarily by estimating exposure of poor people to flooding at various geographical scales (Bangalore et al., 2019; Brouwer et al., 2007; Qiang et al., 2017; Winsemius et al., 2018). In the case of coastal Bangladesh, a few studies have applied quantitative approaches (based on household survey data) to show how poverty exacerbates flood vulnerability/risk (Akter and Mallick, 2013; Brouwer et al., 2007). However, little is known about (i) how the pattern of LULC change influences flood risk at present and in the future; (ii) what is the association between LULC change and risk of flooding, and how they impact poverty spatially. We address these questions by estimating: (i) flood risk in relation to current and future LULC scenarios; and (ii) the change in poverty in relation to a change in LULC and flood risk.

Section snippets

Materials and methods

This study was conducted in three stages. First, a model was established to analyse spatiotemporal patterns of LULC change and predict future LULC. Second, pluvial flood hazard was modelled to simulate the depth and extent of inundations for various return periods of monsoonal precipitation. Then flood risk was estimated at each LULC scenario (historical and future), for different flood return periods. Finally, a spatial regression model was developed to estimate poverty, incorporating

Temporal change of LULC

Fig. 4(a) shows temporal changes of observed LULC from 2005 to 2019 and their spatial variations are presented in Figure S2 (see supplementary document). From 2005–2010, a significant decrease in agricultural land was observed, while the proportion of aquaculture category increased substantially. More than 50 % of agricultural lands transformed into aquaculture use, with another 25 % into rural settlements. Contrarily, LULC change from 2010 to 2019 was relatively stable, when the main

Discussion

Monitoring and managing LULC changes have been recognised as an essential geographic phenomenon for guiding socio-economic development (Corner et al., 2014; Shahbazian et al., 2019). This study analysed and simulated LULC changes in the south western embanked area of Bangladesh to understand their association with flood risk and poverty. The study results indicated that the proportion of agricultural lands decreased significantly between 2005 and 2019. This result is similar to a few other

Conclusion

This study quantified the degree of influence of LULC change and flood risk on poverty in the south western embanked area of Bangladesh. Poverty was estimated, in terms of WI, for the present-day and for future LULC and flood risk scenarios. The analysis indicated that the area has been experiencing a rapid LULC change, resulting in a significant decrease in agricultural lands, while the proportion of aquaculture lands increased consequently. Based on the recent pattern of changes, LULC was

CRediT authorship contribution statement

Mohammed Sarfaraz Gani Adnan: Conceptualization, Methodology, Software, Formal analysis, Validation, Writing - original draft, Writing - review & editing. Abu Yousuf Md Abdullah: Investigation, Resources, Software. Ashraf Dewan: Investigation, Resources, Software. Jim W. Hall: Conceptualization, Supervision, Validation.

Acknowledgements

This work is an output from the REACH programme (www.reachwater.org.uk) funded by UK Aid from the UK Department for International Development (DFID) for the benefit of developing countries (Aries Code 201880). However, the views expressed, and information contained in it are not necessarily those of or endorsed by DFID, which can accept no responsibility for such views or information, or reliance placed on them.

References (79)

  • J.K. Poussin et al.

    Effectiveness of flood damage mitigation measures: empirical evidence from French flood disasters

    Glob. Environ. Change

    (2015)
  • R. Rojas et al.

    Climate change and river floods in the European Union: socio-economic consequences and the costs and benefits of adaptation

    Glob. Environ. Change

    (2013)
  • L. Sang et al.

    Simulation of land use spatial pattern of towns and villages based on CA–Markov model

    Math. Comput. Model.

    (2011)
  • J.F. Warner et al.

    Cutting dikes, cutting ties? Reintroducing flood dynamics in coastal polders in Bangladesh and the netherlands

    Int. J. Disaster Risk Reduct.

    (2018)
  • H. Wheater et al.

    Land use, water management and future flood risk

    Land Use Policy

    (2009)
  • A.Y.M. Abdullah et al.

    Modeling soil salinity using direct and indirect measurement techniques: a comparative analysis

    Environ. Dev.

    (2018)
  • A.Y.M. Abdullah et al.

    Spatio-temporal patterns of land Use/Land cover change in the heterogeneous coastal region of Bangladesh between 1990 and 2017

    Remote Sens.

    (2019)
  • S.G. Adnan et al.

    An evaluation of disaster risk reduction (DRR) approaches for coastal delta cities: a comparative analysis

    Nat. Hazards

    (2016)
  • M.S.G. Adnan et al.

    The potential of tidal river management for flood alleviation in South Western Bangladesh

    Sci. Total Environ.

    (2020)
  • S. Ahmed

    Shrimp farming at the interface of land use change and marginalization of local farmers: critical insights from southwest coastal Bangladesh

    J. Land Use Sci.

    (2018)
  • B. Ahmed et al.

    Simulating land cover changes and their impacts on land surface temperature in Dhaka, Bangladesh

    Remote Sens.

    (2013)
  • M.A. Akber et al.

    Changes of shrimp farming in southwest coastal Bangladesh

    Aquac. Int.

    (2017)
  • M.A. Akber et al.

    Impact of land use change on ecosystem services of southwest coastal Bangladesh

    J. Land Use Sci.

    (2018)
  • H. Apel et al.

    Flood risk analyses - how detailed do we need to be?

    Nat. Hazards

    (2009)
  • J.J. Arsanjani et al.

    Integration of logistic regression, Markov chain and cellular automata models to simulate urban expansion

    Int. J. Appl. Earth Obs. Geoinf.

    (2013)
  • L.W. Auerbach et al.

    Flood risk of natural and embanked landscapes on the Ganges-Brahmaputra tidal delta plain

    Nat. Clim. Change

    (2015)
  • M. Bangalore et al.

    Exposure to floods, climate change, and poverty in Vietnam

    Econ. Disasters Clim. Change

    (2019)
  • BBS

    Bangladesh Disaster-related Statistics 2015

    (2015)
  • E. Borgomeo et al.

    Avoiding the water-poverty trap: insights from a conceptual human-water dynamical model for coastal Bangladesh

    Int. J. Water Resour. Dev.

    (2017)
  • R. Brouwer et al.

    Socioeconomic vulnerability and adaptation to environmental risk: a case study of climate change and flooding in Bangladesh

    Risk Anal.

    (2007)
  • R.J. Corner et al.

    Monitoring and Prediction of Land-use and Land-cover (LULC) Change, Dhaka Megacity

    (2014)
  • A. Dasgupta

    Floods and poverty traps: evidence from Bangladesh

    Econ. Polit.

    (2007)
  • S. Dasgupta et al.

    Climate change and soil salinity: the case of coastal Bangladesh

    Ambio

    (2015)
  • A. De Bono et al.

    A Global Exposure Model for GAR 2015, the Global Assessment Report on Disaster Risk Reduction 2015

    (2014)
  • A. Dewan

    Floods in a Megacity: Geospatial Techniques in Assessing Hazards, Risk and Vulnerability

    (2013)
  • G. Di Baldassarre et al.

    Socio-hydrology: conceptualising human-flood interactions

    Hydrol. Earth Syst. Sci.

    (2013)
  • E. Dube et al.

    Flooding and poverty: two interrelated social problems impacting rural development in Tsholotsho district of Matabeleland North province in Zimbabwe

    Jàmbá: J. Disaster Risk Stud.

    (2018)
  • G. Feng et al.

    Trend analysis and forecast of precipitation, reference evapotranspiration, and rainfall deficit in the Blackland Prairie of Eastern Mississippi

    J. Appl. Meteorol. Climatol.

    (2016)
  • J.W. Hall et al.

    A methodology for national-scale flood risk assessment

    Proc. Inst. Civil Eng. Water Maritime Eng.

    (2003)
  • Cited by (131)

    View all citing articles on Scopus
    View full text