Members’ Publication

This study examines how mobility-challenged persons (MCPs) navigate the limitations of transportation systems, using an integrated choice and latent variable (ICLV) model. Building on the experience of MCPs in a South Asian megacity context, that of Dhaka, Bangladesh, the empirical strategy accounts for perceptions of mode-specific challenges, which arguably vary by mobility impairment. These perceptions are posited as latent variables. Drawing on survey data collected from 400 MCPs living in Dhaka, the latent variables were constructed via a factor analysis of 18 statements about the experienced severity (ranked on a scale from 1 to 5) of mode-specific challenges. Holding socio-demographic and travel-related factors constant, we find that perceptions of mode-specific challenges significantly influence mode choices – while the degree of impairment alone, and related mobility aid needed, do not. Perceived limitations of the walking infrastructure shift MCPs' travel demand towards the bus, whereas bus fare-related issues encourage the use of non-motorized and powered three-wheelers. We recommend that mode choice models include latent variables related to MCPs’ perceptions of various modes to more accurately inform universal access policies.

Identifying urban buildings in high-resolution RGB images presents challenges, mainly due to the absence of near-infrared bands in UAVs and Google Earth imagery and the diversity in building attributes. Deep learning (DL) methods, especially Convolutional Neural Networks (CNNs), are widely used for building extraction but are primarily pixel-based. Geographic Object-Based Image Analysis (GEOBIA) has emerged as an essential approach for high-resolution imagery. However, integrating GEOBIA with DL models presents challenges, including adapting DL models for irregular-shaped segments and effectively merging DL outputs with object-based features. Recent developments include tabular DL models that align well with GEOBIA. GEOBIA stores various features for image segments in a tabular format, yet the effectiveness of these tabular DL models for building extraction still needs to be explored. It also needs to clarify which features are crucial for distinguishing buildings from other land-cover types. Typically, GEOBIA employs shallow learning (SL) classifiers. Thus, this study evaluates SL and tabular DL classifiers for their ability to differentiate buildings from non-building features. Furthermore, these classifiers are assessed for their capacity to handle roof heterogeneity caused by sun exposure and roof materials. This study concludes that some SL classifiers perform similarly to their DL counterparts, and it identifies critical features for building extraction.

Sustainable management of resources (SRM) is most crucial in achieving sustainable development goals in countries with a high population density like Bangladesh. This study aimed to summarize the natural and manmade influences and interferences in SRM in Bangladesh through a systematic literature review. The study also proposed a framework for linking natural and manmade interferences. This study reviewed the natural and man-made hazards in Bangladesh to highlight their impact on SRM. Result shows that environmental degradation is primarily caused by activities for economic reasons, including shrimp cultivation, deforestation, hill-cutting, brick-kilns, and the discharge of waste into rivers. Rapid population growth, industrialization, and urbanization are man-made factors that are putting pressure on land and water resources, leading to deforestation, soil erosion, and pollution. Natural factors, such as climate change, geography, and natural disasters, pose challenges to SRM by affecting water, agriculture, and fisheries. Both these natural and man-made events generate some drastic consequences, like soil erosion, food insecurity, salinity, coastal hazards, degradation of underground water levels, increased pollution, fire hazards, road accidents, building collapse, reduction of soil fertility, damage to biodiversity, and so on. Political interventions like unskilled manpower, corruption, ill-leadership, inadequate accountability and transparency, resource distribution, etc. also affect resource management in Bangladesh. Overall study demonstrated the need for a multi-faceted approach to address these challenges, including investments in sustainable development, disaster risk reduction, and environmental conservation. By defining the necessary initiatives, like financial support, community-based autonomy, empowerment of local government bodies, and sector-based policies, this study will help minimize these losses.

Hard rock systems are typically characterized by undulated landscapes, short flow path, dense drainage network, and low aquifer storage. If located in water-limited environments (WLE), then typically have variable rainfall characteristics, frequent floods and droughts, and savannah type of vegetation capable to uptake groundwater. In such environments, surface–groundwater interactions are of primary importance. In this study, surface–groundwater interactions are analyzed focusing on spatiotemporal net recharge dependence on hydrotopes, where a hydrotope is conceptualized as spatial unit (or object), which at a given scale of assessment can be assumed as internally homogeneous (e.g. tree canopy or rock outcrop) but hydro(geo)logically different than the adjacent area. The study is carried out in the small (7.6 ha) Trabadillo study area (TSA) in Spain, representative for hard rocks of WLE using 3D-GSFLOW integrated hydrological model (IHM) at very fine grid (5x5m), extracted from coarser grid (100x100m) of Sardon 3D-GSFLOW model (Hassan et al., 2014, Journal of Hydrology 517, 390–410). The IHM was calibrated using: (i) 20-year head observation; (ii) 8-year soil moisture observations at 4 profiles, each at different hydrotope and each at 4 different depths; and (iii) MODIS satellite earth observation of evapotranspiration. Despite applying spatially uniform rainfall as forcing, surface–groundwater interactions were highly spatially variable, primarily, because of hydrotope-differences in rainfall interception loss and secondarily, because of differences in transpiration, in subsurface evaporation and in hydrological properties of subsurface. The high temporal variability of rainfall, ranging yearly from 310 to 925 mm, resulted in high temporal variability of all water fluxes, including groundwater fluxes; e.g. gross recharge (Rg) varied annually from 2 to 445 mm, net recharge (Rn), from −50 mm to + 89 mm, while the 20-year mean Rn was close to zero, because Rg was counterbalanced by groundwater exfiltration and less by groundwater evapotranspiration. The high temporal variability of groundwater fluxes and many dry years with largely negative Rn, in combination with low aquifer storage, emphasized TSA dependence on rainfall and its vulnerability to droughts. The hydrotope modeling concept applied at the very fine scale, emphasized also large Rn dependence on land cover type, especially on density and species composition of trees, invisible at coarser scale modeling. As such, the proposed hydrotope modelling approach can be applied as a tool to test whether in a given environment planting (or cutting) trees will result in increase or decrease of water resources.

There were several attempts in identifying the urban poor communities in the cities and towns of Bangladesh in order to fix a correct target for poverty alleviation programmes. However, due to methodological problems, it was always problematic to set a correct target. It was felt by the authority and donor agencies that there was hardly any correct figure of the number of poor and realistic statistics on their characteristics. This paper highlights the methods of mapping and collection of some basic data on poor settlements. It has been possible to identify the poor easily as the poor themselves were involved in the process of identification of their own settlements. The community people, most of whom were little educated successfully conducted the task entrusted to them. This helped them understand the nature and issues of poverty which they can utilize when the actual poverty alleviation programme is launched. Although the study was conducted on 27 cities, this paper explains the mapping and poverty situation of only one city Chattagram. The exercise identified 301,527 households as poor in about 5,778 poor settlements in the Chattagram City Corporation area and estimates about 53.6% of the city’s total population were poor of which 44.6% of the poor households were extremely poor.