Article by Khairun Nahar Diba
Journal of Jahangirnagar University Geography and Environment Review, Jahangirnagar University, Savar, Dhaka, Bangladesh, 2016
Image classification for land use/ land cover is an important tool for many policy planning and m... more Image classification for land use/ land cover is an important tool for many policy planning and management activities related to human-environment relationship. The broad objective of this study is to clarify an image classification system of LULC for the study area that can be explained easily with strong scientific justification. MultiSpec was employed to find out the classification which is a processing system for multispectral image data such as the one produced for the study area from Landsat image. This study followed the Maximum Likelihood classification procedure for image classification that assumes the statistics for each class in each band as normally distributed and calculates the probability of a given pixel belongs to a specific class. Confusion Matrix is a key focus of this research as the table shows accuracy level, Kappa Statistic, actual & predicted classifications. In this study the confusion matrix gives 95.9% overall accuracy, Kappa Statistic of 94% and Kappa Variance of 0.000043 which in combine means a good statistical and scientifically reliable result for image classification. Kappa coefficient for this study is 0.94 which validate a good agreement as in the scale for Kappa a highly reliable range is in between 0.80 to 1.00. The results of this study show a possibility of a simple but well-built image analysis procedure that can be followed in related research areas.
In high relief areas, there is seen many different types of variables such as altitude, aspect an... more In high relief areas, there is seen many different types of variables such as altitude, aspect and slope which strongly influence both human and physical environments. To present the relationship between these two, a 3D data model is therefore essential because it present a more realistic representation of the real world. Here, ArcGIS is used to present an approach for representing complex 3D objects.
Papers by Khairun Nahar Diba
Overseas Development Institute, 2019
• 'Forecast-based early action' (FbA) is emerging among humanitarian and disaster risk management... more • 'Forecast-based early action' (FbA) is emerging among humanitarian and disaster risk management practitioners as an approach that can reduce the impact of shocks on vulnerable people and their livelihoods, improve the effectiveness of emergency preparedness, response and recovery efforts, and reduce the humanitarian burden. • This paper investigates the technical, economic and institutional challenges to scaling up FbA in Bangladesh. Taking a political economy approach it examines the structures and policies around disaster management in the country, options for financing, and the forecasting infrastructure and dissemination systems. • The concept of FbA is not new to Bangladesh but triggers for action are often unclear. The Cyclone Preparedness Programme has long used forecasts to trigger early warning, preparedness and evacuation. However, in most other cases triggers for action are determined more subjectively. • The risk of 'acting in vain' is a major perceived barrier to scaling up FbA. Taking early action when forecasts prove inaccurate has potential implications for accountability and perceived misallocation of limited resources. But if targeted at poor groups, actions could anyway help to enhance resilience. • Institutional incentives and finance are still skewed towards relief. Post-disaster response is seen as more visible and defensible, forming a barrier to early actions. Scaling up of FbA could help to reform prevailing cliental biases in relief by making targeting and delivery of aid more transparent, equitable and needs-based. • Value for money? Stakeholders are demanding for better evidence on the (cost)-effectiveness of FbA approaches. Pilots in Bangladesh suggest improved food security, reduced lending costs and lower anxiety/depression among those taking early action before disasters. • Forecasting is limited but has future potential. Tidal influence makes it difficult to forecast flooding in the southern and coastal zones, while the inaccuracy of cyclone forecasting leaves a limited window for early actions. Riverbank erosion and flash flood forecasts have future potential, along with efforts to improve impact-based forecasting.
Thesis Chapters by Khairun Nahar Diba
The Jahangirnagar Review: Journal of The Faculty of Social Sciences, 2017
From past; the successive event can be known with other circumstance. The event with a successful... more From past; the successive event can be known with other circumstance. The event with a successful ending is written in history. But there are some interesting events also happened that can be ended up with a story of sorrow. The famous discovery of Christopher Columbus is almost the same story which can be ended up with a massive accident but fortunately find a successful joy. The coincident of Columbus’s voyage can be explained in the view of a geographer as he miscalculated the measurement of scale and did the wrong interpretation of contemporary maps. Columbus tried to find a western route as he thought it could be a shorter distance according to his miscalculation. In this research paper, the reason behind Columbus’s miscalculation procedure and his thoughts about the geographic orientation is discussed. The comparative view of the real world and Columbus’s assumed world is presented with the two scales that he got confused. To find the knowledge gap and difference between the two scale; the reproduction of some ancient maps have done and the contemporary geographic knowledge gap is discussed.
Uploads
Article by Khairun Nahar Diba
Papers by Khairun Nahar Diba
Thesis Chapters by Khairun Nahar Diba