Arif Masrur

Bio: Arif Masrur is an academic researcher from Pennsylvania State University. The author has contributed to research in topics: Immersive technology & Tundra. The author has an hindex of 6, co-authored 12 publications receiving 199 citations. Previous affiliations of Arif Masrur include University of Northern Iowa.

Spatio-Temporal Patterns of Land Use/Land Cover Change in the Heterogeneous Coastal Region of Bangladesh between 1990 and 2017

Abstract: Although a detailed analysis of land use and land cover (LULC) change is essential in providing a greater understanding of increased human-environment interactions across the coastal region of Bangladesh, substantial challenges still exist for accurately classifying coastal LULC. This is due to the existence of high-level landscape heterogeneity and unavailability of good quality remotely sensed data. This study, the first of a kind, implemented a unique methodological approach to this challenge. Using freely available Landsat imagery, eXtreme Gradient Boosting (XGBoost)-based informative feature selection and Random Forest classification is used to elucidate spatio-temporal patterns of LULC across coastal areas over a 28-year period (1990–2017). We show that the XGBoost feature selection approach effectively addresses the issue of high landscape heterogeneity and spectral complexities in the image data, successfully augmenting the RF model performance (providing a mean user’s accuracy > 0.82). Multi-temporal LULC maps reveal that Bangladesh’s coastal areas experienced a net increase in agricultural land (5.44%), built-up (4.91%) and river (4.52%) areas over the past 28 years. While vegetation cover experienced a net decrease (8.26%), an increasing vegetation trend was observed in the years since 2000, primarily due to the Bangladesh government’s afforestation initiatives across the southern coastal belts. These findings provide a comprehensive picture of coastal LULC patterns, which will be useful for policy makers and resource managers to incorporate into coastal land use and environmental management practices. This work also provides useful methodological insights for future research to effectively address the spatial and spectral complexities of remotely sensed data used in classifying the LULC of a heterogeneous landscape.

Remote Sensing & GIS Based Spatio-Temporal Change Analysis of Wetland in Dhaka City, Bangladesh

Abstract: Landscape of Dhaka city—one of the fastest growing mega cities in the world, is continuously changing due to un-planned urbanization. For example, the wetlands of the city have been shrinking. This study evaluates wetland changes in Dhaka Metropolitan Area (DMA), Bangladesh, between 1978 and 2009. Spatial and temporal dynamics of wetland changes were quantified using four Landsat images, a supervised classi?cation algorithm and the post-classi?cation change detection technique in GIS environment. Accuracy of the Landsat-derived wetland maps ranged from 87% to 92.5%. The analysis revealed that area of wetland and Rivers & Khals in Dhaka city decreased significantly over the last 30 years by 76.67% and 18.72% respectively. This changing trend of wetlands makes the drainage system of Dhaka City vulnerable, creating water logging problems and their consequences. Land filling and encroachment were recognized to be the main reasons for shrinking of the wetlands in the city. Development and alteration of the existing water bodies should consider the natural hydrological conditions.

Space-Time Patterns, Change, and Propagation of COVID-19 Risk Relative to the Intervention Scenarios in Bangladesh.

Abstract: The novel coronavirus (COVID-19) pandemic continues to be a significant public health threat worldwide, particularly in densely populated countries such as Bangladesh with inadequate health care facilities. While early detection and isolation were identified as important non-pharmaceutical intervention (NPI) measures for containing the disease spread, this may not have been pragmatically implementable in developing countries due to social and economic reasons (i.e., poor education, less public awareness, massive unemployment). Hence, to elucidate COVID-19 transmission dynamics with respect to the NPI status-e.g., social distancing-this study conducted spatio-temporal analysis using the prospective scanning statistic at district and sub-district levels in Bangladesh and its capital, Dhaka city, respectively. Dhaka megacity has remained the highest-risk "active" cluster since early April. Lately, the central and south eastern regions in Bangladesh have been exhibiting a high risk of COVID-19 transmission. The detected space-time progression of COVID-19 infection suggests that Bangladesh has experienced a community-level transmission at the early phase (i.e., March, 2020), primarily introduced by Bangladeshi citizens returning from coronavirus epicenters in Europe and the Middle East. Potential linkages exist between the violation of NPIs and the emergence of new higher-risk clusters over the post-incubation periods around Bangladesh. Novel insights into the COVID-19 transmission dynamics derived in this study on Bangladesh provide important policy guidelines for early preparations and pragmatic NPI measures to effectively deal with infectious diseases in resource-scarce countries worldwide.

Circumpolar spatio-temporal patterns and contributing climatic factors of wildfire activity in the Arctic tundra from 2001–2015

Abstract: Recent years have seen an increased frequency of wildfire events in different parts of Arctic tundra ecosystems. Contemporary studies have largely attributed these wildfire events to the Arctic's rapidly changing climate and increased atmospheric disturbances (i.e. thunderstorms). However, existing research has primarily examined the wildfire–climate dynamics of individual large wildfire events. No studies have investigated wildfire activity, including climatic drivers, for the entire tundra biome across multiple years, i.e. at the planetary scale. To address this limitation, this paper provides a planetary/circumpolar scale analyses of space-time patterns of tundra wildfire occurrence and climatic association in the Arctic over a 15 year period (2001–2015). In doing so, we have leveraged and analyzed NASA Terra's MODIS active fire and MERRA climate reanalysis products at multiple temporal scales (decadal, seasonal and monthly). Our exploratory spatial data analysis found that tundra wildfire occurrence was spatially clustered and fire intensity was spatially autocorrelated across the Arctic regions. Most of the wildfire events occurred in the peak summer months (June–August). Our multi-temporal (decadal, seasonal and monthly) scale analyses provide further support to the link between climate variability and wildfire activity. Specifically, we found that warm and dry conditions in the late spring to mid-summer influenced tundra wildfire occurrence, spatio-temporal distribution, and fire intensity. Additionally, reduced average surface precipitation and soil moisture levels in the winter–spring period were associated with increased fire intensity in the following summer. These findings enrich contemporary knowledge on tundra wildfire's spatial and seasonal patterns, and shed new light on tundra wildfire–climate relationships in the circumpolar context. Furthermore, this first pan-Arctic analysis provides a strong incentive and direction for future studies which integrate multiple datasets (i.e. climate, fuels, topography, and ignition sources) to accurately estimate carbon emission from tundra burning and its global climate feedbacks in coming decades.

Space-time patterns, change, and propagation of COVID-19 risk relative to the intervention scenarios in Bangladesh

Abstract: The novel coronavirus (COVID-19) pandemic continues to be a significant public health threat worldwide. As of mid-June 2020, COVID-19 has spread worldwide with more than 7.7 million confirmed cases and more than 400,000 deaths. The impacts are substantial particularly in developing and densely populated countries like Bangladesh with inadequate health care facilities, where COVID-19 cases are currently surging. While early detection and isolation were identified as important non-pharmaceutical intervention (NPI) measures for containing the disease spread, this may not be pragmatically implementable in developing countries primarily due to social and economic reasons (i.e. poor education, less public awareness, massive unemployment). To shed light on COVID-19 transmission dynamics and impacts of NPI scenarios, e.g. social distancing, this study conducted emerging pattern analysis using the space-time scan statistic at district and thana (i.e. a sub-district or 9upazila9 with at least one police station) levels in Bangladesh and its capital Dhaka city, respectively. We found that the central and south eastern regions in Bangladesh are currently exhibiting a high risk of COVID-19 transmission. Dhaka megacity remains as the highest risk "active" cluster since early April. The space-time progression of COVID-19 infection, when validated against the chronicle of government press releases and newspaper reports, suggests that Bangladesh have experienced a community level transmission at the early phase (i.e., March, 2020) primarily introduced by Bangladeshi citizens returning from coronavirus-affected countries in the Europe and the Middle East. A linkage is evident between the violation of NPIs and post-incubation period emergence of new clusters with elevated exposure risk around Bangladesh. This study provides novel insights into the space-time patterns of COVID-19 transmission dynamics and recommends pragmatic NPI implementation for reducing disease transmission and minimizing impacts in a resource-scarce country with Bangladesh as a case-study example.

Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations—A Review

Abstract: Rapid and uncontrolled population growth along with economic and industrial development, especially in developing countries during the late twentieth and early twenty-first centuries, have increased the rate of land-use/land-cover (LULC) change many times. Since quantitative assessment of changes in LULC is one of the most efficient means to understand and manage the land transformation, there is a need to examine the accuracy of different algorithms for LULC mapping in order to identify the best classifier for further applications of earth observations. In this article, six machine-learning algorithms, namely random forest (RF), support vector machine (SVM), artificial neural network (ANN), fuzzy adaptive resonance theory-supervised predictive mapping (Fuzzy ARTMAP), spectral angle mapper (SAM) and Mahalanobis distance (MD) were examined. Accuracy assessment was performed by using Kappa coefficient, receiver operational curve (RoC), index-based validation and root mean square error (RMSE). Results of Kappa coefficient show that all the classifiers have a similar accuracy level with minor variation, but the RF algorithm has the highest accuracy of 0.89 and the MD algorithm (parametric classifier) has the least accuracy of 0.82. In addition, the index-based LULC and visual cross-validation show that the RF algorithm (correlations between RF and normalised differentiation water index, normalised differentiation vegetation index and normalised differentiation built-up index are 0.96, 0.99 and 1, respectively, at 0.05 level of significance) has the highest accuracy level in comparison to the other classifiers adopted. Findings from the literature also proved that ANN and RF algorithms are the best LULC classifiers, although a non-parametric classifier like SAM (Kappa coefficient 0.84; area under curve (AUC) 0.85) has a better and consistent accuracy level than the other machine-learning algorithms. Finally, this review concludes that the RF algorithm is the best machine-learning LULC classifier, among the six examined algorithms although it is necessary to further test the RF algorithm in different morphoclimatic conditions in the future.

A review of machine learning applications in wildfire science and management

Abstract: Artificial intelligence has been applied in wildfire science and management since the 1990s, with early applications including neural networks and expert systems. Since then, the field has rapidly ...

A review of machine learning applications in wildfire science and management

Abstract: Artificial intelligence has been applied in wildfire science and management since the 1990s, with early applications including neural networks and expert systems. Since then the field has rapidly progressed congruently with the wide adoption of machine learning (ML) in the environmental sciences. Here, we present a scoping review of ML in wildfire science and management. Our objective is to improve awareness of ML among wildfire scientists and managers, as well as illustrate the challenging range of problems in wildfire science available to data scientists. We first present an overview of popular ML approaches used in wildfire science to date, and then review their use in wildfire science within six problem domains: 1) fuels characterization, fire detection, and mapping; 2) fire weather and climate change; 3) fire occurrence, susceptibility, and risk; 4) fire behavior prediction; 5) fire effects; and 6) fire management. We also discuss the advantages and limitations of various ML approaches and identify opportunities for future advances in wildfire science and management within a data science context. We identified 298 relevant publications, where the most frequently used ML methods included random forests, MaxEnt, artificial neural networks, decision trees, support vector machines, and genetic algorithms. There exists opportunities to apply more current ML methods (e.g., deep learning and agent based learning) in wildfire science. However, despite the ability of ML models to learn on their own, expertise in wildfire science is necessary to ensure realistic modelling of fire processes across multiple scales, while the complexity of some ML methods requires sophisticated knowledge for their application. Finally, we stress that the wildfire research and management community plays an active role in providing relevant, high quality data for use by practitioners of ML methods.

Can Simulated Nature Support Mental Health? Comparing Short, Single-Doses of 360-Degree Nature Videos in Virtual Reality With the Outdoors.

Abstract: Nature exposure in virtual reality (VR) can provide emotional well-being benefits for people who cannot access the outdoors. Little is known about how these simulated experiences compare with real outdoor experiences. We conduct an experiment with healthy undergraduate students that tests the effects of 6 min of outdoor nature exposure with 6 min of exposure to a 360-degree VR nature video, which is recorded at the outdoor nature exposure location. Skin conductivity, restorativeness, and mood before and after exposure are measured. We find that both types of nature exposure increase physiological arousal, benefit positive mood levels, and are restorative compared to an indoor setting without nature; however, for outdoor exposure, positive mood levels increase and for virtual nature, they stay the same. The nature-based experience shows benefits above and beyond the variance explained by participants' preferences, nature and VR experiences, and demographic characteristics. Settings where people have limited access to nature might consider using VR nature experiences to promote mental health.

Real time surveillance of COVID-19 space and time clusters during the summer 2020 in Spain.

Abstract: On June 21st de-escalation measures and state-of-alarm ended in Spain after the COVID-19 first wave. New surveillance and control strategy was set up to detect emerging outbreaks. To detect and describe the evolution of COVID-19 clusters and cases during the 2020 summer in Spain. A near-real time surveillance system to detect active clusters of COVID-19 was developed based on Kulldorf’s prospective space-time scan statistic (STSS) to detect daily emerging active clusters. Analyses were performed daily during the summer 2020 (June 21st – August 31st) in Spain, showing an increase of active clusters and municipalities affected. Spread happened in the study period from a few, low-cases, regional-located clusters in June to a nationwide distribution of bigger clusters encompassing a higher average number of municipalities and total cases by end-August. STSS-based surveillance of COVID-19 can be of utility in a low-incidence scenario to help tackle emerging outbreaks that could potentially drive a widespread transmission. If that happens, spatial trends and disease distribution can be followed with this method. Finally, cluster aggregation in space and time, as observed in our results, could suggest the occurrence of community transmission.