Geospatial Datasets for Assessing Vulnerability of Bangladesh to Climate Change and Extremes
Description
The present dataset provides necessary indicators of climate change vulnerability of Bangladesh in raster form. Geospatial databases have been created in Geographic Information System (GIS) environment mainly from two types of raw data; socioeconomic data from Bangladesh Bureau of Statistics (BBS) and biophysical maps from various government and non-government agencies. Socioeconomic data have been transformed to raster database through the Inverse Distance Weighted (IDW) interpolation method in GIS. On the other hand, biophysical maps have been directly recreated as GIS shapefiles and eventually the biophysical raster database have been produced. These geospatial datasets have been analyzed to assess the spatial vulnerability of Bangladesh to climate change and extremes. 30 socioeconomic indicators have been considered, which has been obtained from the Bangladesh Bureau of Statistics (BBS, 2012; BBS, 2013; BBS, 2016). All socioeconomic data were incorporated in GIS database in order to generate maps. 12 biophysical system indicators have also been classified based on the collected information from different sources and literature. The coefficient of temperature and precipitation variability have been extracted from the work of Institute of Water and Flood Management (IWFM, 2014) and mapped according to the climatic sub-regions produced by Rashid (1991). A five-class drought class map of the whole country has been recreated from Comprehensive Disaster Management Program (CDMP, 2006). The cyclone risk map used in this study, a four-class relative risk map, has been adopted from Center for Environment and Geographic Information Services (CEGIS, 2006). The Sea Level Rise (SLR) risk map have been produced from the elevation map collected from United States Geological Survey (USGS). Different types of flood risk maps have been reproduced from the maps of Bangladesh Agricultural Research Council (BARC, 2001) and Bangladesh Water Development Board (BWDB, 2010). Erosion prone areas with relative risks (BWDB, 2010) and salinity intrusion map of 1 to 5 ppt salinity line (SRDI, 2010) also have been recreated in this study. Finally, a general hazard class map covering all over the country, with a 1 to 5 relative hazard proneness, has been adopted from Bangladesh Center for Advanced Studies (BCAS, 2008).
Files
Steps to reproduce
Digitization, Database Creation and Raster Conversion: Since spatial assessment of vulnerability is adopted in the present study, the base map of Bangladesh is firstly digitized along with its all district centers. The socioeconomic data, collected from different BBS publications, are transformed into desired units of measurement and then incorporated in GIS database. As the dataset is presented based on districts, the GIS database is also created basing district centers. On the other hand, all the biophysical data were collected from different published maps which were not corresponding to the district map of Bangladesh. Therefore, a new GIS database was created for biophysical indicators followed by the incorporation of quantified scale derived from those reference maps. For the suitability of spatial analysis, all vector maps from created databases, both from socioeconomic and biophysical, were converted to raster datasets using ArcMap’s conversion toolbox. However, for GIS analysis, ArcGIS 10.5 desktop version is used in all over the present study. Normalization of the Indicators: Normalization is important for multivariate statistical analysis as some variables have large range of variance and some of them have a small range of variance. To avoid the influence of one variable to other variables, the dataset has been normalized. For the normalization of raster datasets in ArcMap 10.5, raster calculator was used which is a widely used tool under Map Algebra of the Arc toolbox. For each raster dataset, the following expression was used; (“x” - “x”.minimum)/(“x”.maximum – “x”.minimum) Where, x = Raster name. All of the created normalized raster data are then stored in a new database for further analysis.