Mapping Long Term Changes in Mangrove Cover and Predictions of Future Change Under Different Climate Change Scenarios in the Sundarbans, Bangladesh

Abstract

The Sundarbans mangrove forest is an important resource for the people of the Ganges Delta. It plays an important role in the local as well as global ecosystem by providing ecological services and economic goods. However, this mangrove ecosystem is under threat, mainly due to climate change and anthropogenic factors. The aims of this Thesis are: (1) to apply remote sensing techniques and open source mid-resolution data, as cheap and reliable data, to identify and map mangrove composition at species level, (2) to monitor the impact of climate variability in this ecosystem , (3) assess spatial and temporal dynamics of tidal channels in the Bangladesh Sundarbans and finally (4) to predict the magnitude of mangrove area loss and future impacts on mangrove species composition and distribution due to a rise in Mean Sea Level (MSL<br/>Chapter one evaluates the efficacy of mid-resolution Landsat satellite image combined with traditional classification algorithms to produce an acceptable accuracy at species level mapping of mangroves. A maximum likelihood algorithm was employed to identify and map mangrove species composition using open-source mid-resolution Landsat data, taking Bangladesh Sundarbans as a case study. The classified image achieved an overall accuracy of 89.10% and kappa coefficient of 0.87 for the five-identified species, viz. <i>Heritiera fomes</i>, <i>Ceriops decandra</i>, <i>Excoecaria agallocha</i>, <i>Sonneratia apelatala</i>, and <i>Xylocarpus mekongensis</i>, which is higher than the required minimum overall accuracy of 85% deemed suitable to use in most of the natural resource mapping applications. Based on our result, it can be concluded that mid-resolution images, such as Landsat, and the traditional classification algorithm can be applied with confidence for the identification and classification of mangrove forest resources at species level as an alternative to the high resolution satellite images.<br/>The second research chapter is about mapping the long term changes in mangrove species composition in the Sundarbans. Maximum likelihood classifier technique was employed to classify images recorded by the Landsat satellite series and used post classification comparison techniques to detect changes at the species level. The image classification resulted in overall accuracies of 72%, 83%, 79% and 89% for the images of 1977, 1989, 2000 and 2015, respectively. We identified five major mangrove species and detected changes over the 38-year (1977–2015) study period. During this period, both <i>Heritiera fomes</i> and <i>Excoecaria agallocha</i> decreased by 9.9%, while <i>Ceriops decandra</i>, <i>Sonneratia apelatala</i>, and <i>Xylocarpus mekongensis</i> increased by 12.9%, 380.4% and 57.3%, respectively.<br/>The third research chapter presents the relationship between temperature, rainfall pattern and dynamics of mangrove species in the Sundarbans, Bangladesh, assessed over a 38 year time period from 1977–2015. A three stage analytical process was employed to monitor the impact of climate variability in this ecosystem. Primarily, the trend of temperature and rainfall over the study period were identified using a linear trend model; then, the supervised maximum likelihood classifier technique was employed to classify images recorded by Landsat series and postclassification comparison techniques were used to detect changes at species level. The rate of change of different mangrove species was also estimated in the second stage. Finally, the relationship between temperature, rainfall and the dynamics of mangroves at species level was determined using a simple linear regression model. A significant statistical relationship between temperature, rainfall and the dynamics of mangrove species was obtained. The trends of change for <i>Heritiera fomes</i> and <i>Sonneratia apelatala</i> show a strong relationship with temperature and rainfall, while <i>Ceriops decandra</i> shows a weak relationship. In contrast, <i>Excoecaria agallocha</i> and <i>Xylocarpus mekongensis</i> do not show any significant relationship with temperature and rainfall. This chapter concluded that temperature and rainfall are important climatic factors influencing the dynamics of three major mangrove species viz. <i>Heritiera fomes</i>, <i>Sonneratia apelatala</i> and <i>Ceriops decandra</i> in the Sundarbans.<br/>The fourth research chapter focuses on the spatial and temporal dynamics of tidal channels in the Bangladesh Sundarbans. Parts of the Passur River system were considered for this investigation. Tidal channel bank layers were extracted from aerial photographs from 1974 and 2011, and a Sentinel-2 image from 2017. Remote Sensing and Geographic Information System (GIS) platforms were used to analyse, interpret, and visualize data on accretion and erosion, as well as the locations of the tidal channel bank over different years. The results revealed that erosion was severe in the larger channels, whereas accretion was dominant in the smaller channels. Displacement of the tidal channel bank has had a profound impact on the Sundarbans mangrove ecosystem, and continued erosion and accretion processes are of concern for the future sustainability of biodiversity in the Sundarbans. While in the short term these changes may not have much impact, over decades the dynamics of tidal channels may significantly contribute to the imbalance of fauna and flora in the Sundarbans.<br/>A synthesis of the forcing mechanisms of tidal channel dynamics in the context of natural and anthropogenic forces and their implications on the Sundarbans delta floodplain mangrove forest comes in the fifth research chapter. Natural tidal channel dynamics driving forces viz: tectonic and subsidence, sea level rise, tides, storms, cyclones and other climatic factors are discussed in this synthesis. Human induced morphodynamic factors that affect erosion and accretion processes in the Sundarbans tidal channel system are also discussed. Based on our discussion it can be concluded that natural and anthropogenic forces such as tides, storms and cyclones, fluctuations in seasonal rainfall, tectonic and subsidence forces, sea level rise, infrastructure development and changing pattern of land use plays a vital role in the erosion accretion processes in tidal channel dynamics in the study area, and subsequently have important implications on the sustainability of the Sundarbans mangrove ecosystem. Precise effects of these natural and anthropogenic forces are recommended for future research.<br/>The sixth research chapter predicts the magnitude of mangrove area loss and future impacts on mangrove species composition and distribution due to a rise in Mean Sea Level (MSL). In this study, a geospatial model of potentially inundated areas was developed using Digital Elevation Model (DEM) data to assess the potential impacts of sea level rise (SLR) on the future spatial distribution of mangrove species and estimates the potential inundation and subsequent mangrove area loss. The mangrove areas of 2646 ha, 9599 ha and 74720 ha are projected to be inundated and subsequently lost by the end of the 21st century for the low, medium and high SLR scenarios respectively under the net subsidence rate -2.4 mm/year relative to the baseline year 2000. All the major five mangrove species of the Bangladesh Sundarbans will be affected and that can potentially contribute to a change in the present species composition and biodiversity of the forest. Results suggest that, under the extreme scenario, inundation and subsequent loss of different mangrove species will be substantial and this can bring a massive change in the species composition and their spatial distribution in the Bangladesh Sundarbans.<br/>In conclusion, long term changes in mangrove cover at species level, and prediction of future spatial distributions under different climate change scenarios in the Bangladesh Sundarbans are mapped and analysed in this thesis. In addition, a newly developed geospatial model shows the future impact of sea level rise on species composition and their spatial distribution. This model can be used in future for the impact assessment of sea level rise on mangrove species in the Sundarbans and other parts of the world. Thus, this research provides some invaluable insights and techniques for the development of a proper monitoring strategies in the future for sustainable management of the forest in response to climate variability and change. Different relevant agencies of Bangladesh government, such as Bangladesh Forest Department and Ministry of Environment, can follow the approach and incorporate the outcomes of this research to develop a continuous and proper monitoring system in a time saving, efficient and cost effective way. Hope outcome of this study will be a stepping stone in the studies of the Sundarbans mangroves and its sustainable management using remote sensing techniques.