Article
Environmental Sciences
Yi Yu, Luigi J. Renzullo, Tim R. Mcvicar, Brendan P. Malone, Siyuan Tian
Summary: This study proposed an improved version of the ESTARFM method (ubESTARFM) to generate high spatiotemporal resolution land surface temperature (LST) estimates. Through a comparison evaluation with in-situ observations and ECOSTRESS satellite data, it was found that ubESTARFM can avoid systematic bias accumulation, substantially reduce uncertainty deviation, and maintain a good level of correlation with validation datasets when compared to ESTARFM. This method has the potential to be applied using LST data acquired from geostationary platforms, enabling better farm and regional-scale water management strategies to be implemented.
REMOTE SENSING OF ENVIRONMENT
(2023)
Article
Remote Sensing
Weisheng Li, Xiayan Zhang, Yidong Peng, Meilin Dong
Summary: This paper introduces a spatiotemporal fusion method based on a convolutional neural network, using residual images to train the network and combining multiscale and attention mechanisms to improve fusion accuracy, achieving richer spatial details and more accurate prediction of temporal changes.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Environmental Sciences
Hongwei Zhang, Fang Huang, Xiuchao Hong, Ping Wang
Summary: This article proposes a method to correct input data of spatiotemporal fusion model by learning biases between different sensors through machine learning techniques. By correcting the model biases, detailed spatial information in the data can be enhanced, significantly improving the accuracy and robustness of fusion technology.
Article
Environmental Sciences
Jingbo Wei, Lei Chen, Zhou Chen, Yukun Huang
Summary: This study addresses two issues: although over one hundred spatiotemporal fusion algorithms have been proposed, convolutional neural networks trained with large amounts of data have not shown significant advantages. Additionally, the feasibility of using fused images for change detection has not received enough attention. The study designs a new dataset consisting of nine image pairs to evaluate the accuracy of neural networks trained with one-pair spatiotemporal fusion using neural-network-based models. Notably, the size of each image is significantly larger compared to other datasets used for neural network training.
Article
Engineering, Chemical
Shenglin Li, Jinglei Wang, Dacheng Li, Zhongxin Ran, Bo Yang
Summary: This study evaluated the accuracy of Landsat 8-like LST products generated by different spatiotemporal fusion algorithms, with ESTARFM algorithm showing the highest fusion accuracy. Ground measurement verification showed high consistency between fusion images and actual Landsat 8 LST images.
Article
Environmental Sciences
Qunming Wang, Kaidi Peng, Yijie Tang, Xiaohua Tong, Peter M. Atkinson
Summary: MODIS data is widely used for global monitoring of the Earth's surface due to its daily fine temporal resolution, but the spatial resolution is too coarse. This paper proposes a SU-BR method that removes blocky artifacts to increase prediction accuracy. SU-BR provides a crucial solution to overcome challenges in spatio-temporal fusion.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Geography, Physical
Yang Chen, Ruyin Cao, Jin Chen, Licong Liu, Bunkei Matsushita
Summary: The study introduced a simple and effective method for reconstructing high-quality Landsat NDVI time-series data. This new method showed significant improvements in filling long-term missing values, removing residual noise, and being applicable at a large spatial scale.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Yanghui Kang, Mutlu Ozdogan, Feng Gao, Martha C. Anderson, William A. White, Yun Yang, Yang Yang, Tyler A. Erickson
Summary: This paper presents a high-resolution LAI mapping method based on Landsat images for the Contiguous US, using machine learning and sample balance design to estimate LAI. The approach is validated on 19 NEON sites and eight independent sites, showing promising results with RMSE between 0.52 and 0.91.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Construction & Building Technology
Zakari Aretouyap, Janvier Domra Kana, Franck Eitel Kemgang Ghomsi
Summary: This study applied remote sensing techniques to address sustainability, environment quality, and life comfort issues in the capital district of Cameroon, Yaounde. It highlighted the effects of urban heat island and suggested immediate redesigning of Yaounde and other cities by prioritizing urban forest and vegetation cover to improve environment quality.
SUSTAINABLE CITIES AND SOCIETY
(2021)
Article
Remote Sensing
Shuping Xiong, Shihong Du, Xiuyuan Zhang, Song Ouyang, Weihong Cui
Summary: This study introduces a novel spatial-temporal-spectral integration model that can fuse Landsat-7, 8, and Sentinel-2 sensor data simultaneously. By utilizing an enhanced residual dense network and a time-series-based reflectance adjustment method, the fusion framework is able to fill data gaps, restore more spatial details, and predict tiny reflectance changes.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Environmental Sciences
Marcela Rosas-Chavoya, Pablito Marcelo Lopez-Serrano, Jose Ciro Hernandez-Diaz, Christian Wehenkel, Daniel Jose Vega-Nieva
Summary: This study estimated the near-surface temperature lapse rate (NSTLR) in a mountain ecosystem in Northern Mexico for two seasons and found that it is influenced by aspect, local solar zenith angle (LSZA), and evaporative stress index (ESI). The highest NSTLR was observed in the Northwest and West aspects, while angles close to 90 degrees of LSZA were associated with lower NSTLR. Additionally, ESI values indicating less evaporative stress were related to lower NSTLR.
Article
Engineering, Electrical & Electronic
Chenlie Shi, Ninglian Wang, Quan Zhang, Zhuang Liu, Xinming Zhu
Summary: In this study, a Comprehensive Flexible Spatiotemporal DAta Fusion (CFSDAF) method was proposed to generate a high spatiotemporal resolution urban LST image. The CFSDAF method is able to preserve more spatial details and restore the spatial continuity of urban LST better than other methods. It also has high computational efficiency and can monitor land cover changes. Therefore, CFSDAF has great potential for studying spatiotemporal changes of LST and UHI in an urban area.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Remote Sensing
Yaxu Wang, Xiaobo Luo, Quan Wang
Summary: The spatiotemporal image fusion technique efficiently blends geometrically registered remote sensing images from different sensors into a single image with high temporal and spatial resolutions. The newly proposed BESFM method outperforms existing fusion methods in generating Enhanced Vegetation Index (EVI) images in regions with rapid vegetation changes, showing potential for increasing the availability of fine spatial resolution data.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Remote Sensing
Yaobin Ma, Jingbo Wei, Wenchao Tang, Rongxin Tang
Summary: This study proposed a step-by-step modeling framework and designed three models to separately capture the spatial difference, sensor difference, and temporal difference. Experimental results showed that these difference models all contributed to performance improvement.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Environmental Sciences
David Hidalgo Garcia, Julian Arco Diaz
Summary: This study uses satellite imaging to determine the land surface temperature of Granada, Spain, with the single channel algorithm proving to be more effective and reliable in this aspect.
Article
Meteorology & Atmospheric Sciences
Abu Yousuf Md Abdullah, Md Hanif Bhuian, Grigory Kiselev, Ashraf Dewan, Quazi K. Hasan, M. Rafiuddin
Summary: This study aimed to understand the trends in extreme climatic events in coastal and inland areas of Bangladesh. Results showed significant warming in both areas, with coastal areas experiencing a higher rate of warming. While most extreme rainfall indices did not show significant changes, there was evidence of localized dryness and increased rainfall at individual stations. The decrease in rainfall in the drought-prone northwestern region was contrary to previous studies.
INTERNATIONAL JOURNAL OF CLIMATOLOGY
(2022)
Article
Multidisciplinary Sciences
Dhananjay Deshmukh, M. Razu Ahmed, John Albino Dominic, Mohamed S. Zaghloul, Anil Gupta, Gopal Achari, Quazi K. Hassan
Summary: The objective of this study was to quantify the similarity in meteorological measurements of 17 stations under three weather networks in the Alberta oil sands region. Various methods were used to find correlations and determine the optimal number of stations, which could be critical to rationalize/optimize weather networks in the study area.
Article
Chemistry, Analytical
Ifeanyi R. Ejiagha, M. Razu Ahmed, Ashraf Dewan, Anil Gupta, Elena Rangelova, Quazi K. Hassan
Summary: This study quantified the surface urban heat island (SUHI) in the cities of Calgary and Edmonton, Canada, and analyzed its trends and influencing factors. The results showed that both cities experienced continuous increases in the annual daytime and nighttime SUHI values. Population and built-up expansion were identified as the main factors influencing the SUHI.
Article
Green & Sustainable Science & Technology
Ghafar Salavati, Ebrahim Saniei, Ebrahim Ghaderpour, Quazi K. Hassan
Summary: This study in Sanandaj, Iran, assesses the potential risk of forest and rangeland fires using weights of evidence and statistical index models. The results show that a significant portion of the study area is at moderate to very high risk. The ROC curve analysis validates the models and indicates that the WoE model has slightly higher predictive capability.
Article
Computer Science, Software Engineering
Masoud Abdollahi, Babak Farjad, Anil Gupta, Quazi K. Hassan
Summary: CMIP6-D&A is an open-source software based on R language, with a user-friendly graphical user interface. It can download and process climate data, generate various output files, and is suitable for different regions worldwide.
Article
Remote Sensing
Ali Almagbile, Khaled Hazaymeh
Summary: The COVID-19 lockdown measures in 2020 have significantly reduced the levels of pollutant gases and land surface temperature in Amman. Statistical analysis showed a decrease in NO2 and CO emissions, as well as a reduction in land surface temperature.
GEO-SPATIAL INFORMATION SCIENCE
(2023)
Review
Engineering, Electrical & Electronic
M. Razu Ahmed, Ebrahim Ghaderpour, Anil Gupta, Ashraf Dewan, Quazi K. Hassan
Summary: Understanding land surface temperature (LST) trends is essential for developing strategies to cope with climate change. This article reviews studies on spaceborne sensor-based LST trends using thermal infrared (TIR) and passive microwave (PMW) observations. Most studies use TIR, particularly MODIS observations. Challenges and research gaps in utilizing TIR and PMW observations are identified, along with recommendations for future investigations and directions to overcome limitations.
IEEE SENSORS JOURNAL
(2023)
Article
Chemistry, Analytical
Md. Mahbub Alam, A. S. M. Mahtab, M. Razu Ahmed, Quazi K. Hassan
Summary: This research examined the characteristics of cold days and spells in Bangladesh and quantified their rate of change during the winter months of 2000-2021. It found that cold days were more prevalent in the west-northwestern regions and gradually decreased towards the south and southeast. The highest number of cold spells occurred in the northwest Rajshahi division, while the lowest occurred in the northeast Sylhet division.
Article
Environmental Sciences
Mohamed Shawky, Quazi K. K. Hassan
Summary: This study aims to map and predict flash flood prone areas using the analytic hierarchy process (AHP) that integrates GIS capabilities, remote sensing datasets, the NASA Giovanni web tool application, and principal component analysis (PCA). Nineteen flash flood triggering parameters were considered, and the PCA algorithm was used to reduce subjectivity. The results showed that the AHP model had excellent predictive accuracy of 91.6%.
Article
Environmental Sciences
Hatef Dastour, Anil Gupta, Gopal Achari, Quazi K. K. Hassan
Summary: Stream and river monitoring play a crucial role in various industries such as agriculture, fishing, land surveillance, and oil and gas. This study introduces a new algorithm, Regime Shift Change Detection (RSCD), which can identify periods and regime changes without assumptions about their length. Two specializations of this algorithm, RSCD with Relative Difference (RSCD-RD) and RSCD with Growth Rate (RSCD-GR), were compared and their advantages were discussed. RSCD-RD outperformed RSCD-GR in detecting regime changes with general thresholds for cold and warm months. A regime change was detected in the monthly streamflow data of the Athabasca River at Athabasca, but not below Fort McMurray, suggesting possible factors such as water clarity and industrial water usage.
Article
Chemistry, Multidisciplinary
Md. Mahbub Alam, A. S. M. Mahtab, M. Razu Ahmed, Quazi K. Hassan
Summary: This study used the 10th percentiles of daily minimum and maximum temperatures during 1971-2000 to estimate a threshold for cold days. By applying this threshold to the winter months of 2000-2021, the number and trends of cold days and spells were calculated. The results showed that there were more cold days in the western and northwestern districts of Bangladesh compared to the southern, southeastern, and northeastern districts. Dinajpur and Rajshahi districts had the highest number of extreme and severe cold days, while Rajshahi division had the highest number of cold spells on average. The 10P method proposed in this study could be useful for policymakers in formulating strategies to minimize the impact of cold weather in Bangladesh.
APPLIED SCIENCES-BASEL
(2023)
Article
Environmental Sciences
Khaled Hazaymeh, Ali Almagbile, Ala'a Alsayed
Summary: A cascaded data fusion approach was employed to extract rooftops of buildings in a heterogeneous urban fabric using remote sensing data in Irbid city, Jordan. The method combined support vector machine (SVM) classification, normalized digital surface model (nDSM) calculation, and normalized difference vegetation index (NDVI) filtering to identify the rooftops. The results were evaluated and compared to reference data, showing a high accuracy and completeness. The study proved the effectiveness of the method in mapping rooftops of buildings in diverse urban areas.
EGYPTIAN JOURNAL OF REMOTE SENSING AND SPACE SCIENCES
(2023)
Article
Water Resources
Hatef Dastour, Quazi K. Hassan
Summary: Having a complete hydrological time series is crucial but challenging in data-scarce environments. This study introduces an ensemble machine-learning regression framework to accurately model and predict monthly streamflow using historical data from multiple datasets.
Article
Ecology
M. Razu Ahmed, Quazi K. Hassan
Summary: This study analyzed the trends in forest fire occurrences, burned areas, and seasonality in the forested subregions of Alberta from 1959 to 2021. The results showed that all subregions, except for the Alpine subregion, experienced significantly increasing trends in fire occurrences. For burned areas, nine ecoregions showed decreasing monthly trends for small fires caused by humans, except for one subregion with an increasing trend in May. The study also revealed changes in the start and end of fire seasons, with longer fire seasons observed in five ecoregions. These findings provide valuable insights for fire management agencies and strategic planning.
Article
Green & Sustainable Science & Technology
Hatef Dastour, Quazi K. K. Hassan
Summary: The pace of LULC change has accelerated due to population growth, industrialization, and economic development. Recent advances in deep learning, transfer learning, and remote sensing technology have simplified the LULC classification problem. Deep transfer learning is particularly useful for addressing the issue of insufficient training data.