Article
Environmental Sciences
Solmaz Fathololoumi, Mohammad Karimi Firozjaei, Huijie Li, Asim Biswas
Summary: This study aims to improve the accuracy of land cover/land use classification by fusing different surface biophysical features and combining land cover maps from different scenarios. The results demonstrate that the fusion-based method significantly increases the overall accuracy of land cover/land use classification.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Ecology
Jianbo Tan, Jiaqi Zuo, Xinyao Xie, Meiqing Ding, Zhuokui Xu, Fangbin Zhou
Summary: The study compared the performance of random forest, decision tree, support vector machine, and artificial neural network algorithms in mapping three typical landscapes in Hunan Province, China, where random forest showed the best accuracy across various terrains with minimal human intervention.
ECOLOGICAL INFORMATICS
(2021)
Article
Environmental Sciences
Bo Zhong, Aixia Yang, Kunsheng Jue, Junjun Wu
Summary: Long time series of high-quality and consistent land cover datasets are in great demand for analyzing long-term climate, environmental, and ecological changes. This study proposed a new time series land cover mapping method based on machine learning, utilizing Landsat satellite imagery and the LCMM method, which was successfully applied to Heihe River Basin to produce a long time series land cover dataset with an average precision of about 90%. This dataset, with 30m resolution, is the longest time series land cover map at HRB and demonstrates good time continuity and stability.
Article
Environmental Sciences
Vasco Mantas, Claudia Caro
Summary: Land cover in mountainous regions is influenced by a complex range of stressors. This study presents the development of a land cover database for the Junin National Reserve in Peru using satellite data and in situ observations. The database was used to identify ecosystem services provided by different land cover classes.
Article
Environmental Sciences
Adam Wasniewski, Agata Hoscilo, Milena Chmielewska
Summary: Monitoring land cover is crucial for environmental management, resource assessment, protection, planning, and sustainable development. This study explores the impact of a hierarchical approach on the accuracy of land cover mapping using machine learning algorithms and Sentinel-2 imagery in central Poland. The results indicate that the hierarchical approach leads to higher overall accuracy compared to the flat approach.
Article
Environmental Sciences
Jianfeng Luo, Chunyu Dong, Kairong Lin, Xiaohong Chen, Liqiang Zhao, Lucas Menzel
Summary: This study presents a new algorithm based on machine learning technology to improve the accuracy of binary snow cover mapping in forests. The proposed algorithm shows high performance in forest BSC mapping, retrieving 67% of all real forest snow pixels compared to only 8-14% by the NDSI-based approach. The algorithm's performance is sensitive to changes in solar illumination conditions and forest coverage, suggesting that machine learning with the fusion of optical remote sensing and ground-based observations is an effective approach for improving the accuracy of forest snow cover mapping at regional scales.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Green & Sustainable Science & Technology
Muhammad Junaid, Jianguo Sun, Amir Iqbal, Mohammad Sohail, Shahzad Zafar, Azhar Khan
Summary: Pakistan has a high deforestation rate of 4.6%, with an increase from 1.8-2.2% in the past two decades. KPK province holds 31% of the country's forest resources, mainly natural forests. The Malam Jabba region has experienced significant changes in land cover, with an increase in agricultural and built-up land and loss in forest and agricultural land due to flood disasters. This study proposes a framework using remote sensing data to assess vegetation cover changes in Malam Jabba region, and the results show a significant decrease in forest and woodland cover.
Article
Environmental Sciences
Sara Dahhani, Mohamed Raji, Mustapha Hakdaoui, Rachid Lhissou
Summary: This study demonstrates the efficiency of machine learning in improving land use/cover classification from SAR satellite imagery, which is particularly useful in sub-Saharan countries with frequent cloud cover. The researchers focused on mapping land use and land cover in the agricultural areas of the Kaffrine region in Senegal using S-1 time-series data. They evaluated the performance and processing time of three machine-learning classifiers and found that RF classification using the S-1 time-series data achieved the highest accuracy, although it was slower compared to KDtKNN.
Article
Environmental Sciences
Zhenyu Zhang, Georg Hoermann, Jinliang Huang, Nicola Fohrer
Summary: Understanding land use/cover change (LUCC) in watersheds is crucial for sustainable development. The machine learning-based CA-Markov model comprehensively evaluates the factors influencing LUCC, identifies patterns under different scenarios, and can serve as a helpful tool for watershed management.
Article
Geochemistry & Geophysics
Yilang Shen, Jingzhong Li, Rong Zhao, Fengfeng Han
Summary: In this study, a superpixel-based land cover mapping method for remote sensing images is proposed, which can more effectively achieve multiresolution mapping by considering the geometric, topologic, and semantic characteristics of land parcels.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Mathematics, Applied
Dawei Li, Ruifang Zhang, Peng Liu, Tianye Liu, Wei Fang
Summary: A simple and effective algorithm RFDCNN based on random forests is proposed to address issues in deep learning methods for remote sensing, achieving perfect classification images with a high overall accuracy of 0.90 through decision-level fusion and compatible weights in experiments with different feature sets.
JOURNAL OF NONLINEAR AND CONVEX ANALYSIS
(2021)
Article
Environmental Sciences
Devanantham Abijith, Subbarayan Saravanan
Summary: The study utilized GEE, TerrSet, and GIS tools to analyze LULC changes on the Northern TN coast between 2009-2019 and 2019-2030, revealing trends of decreased water bodies, increased built-up areas, and conversion of barren land and vegetation into built-up areas. The overall accuracy was above 89%, providing valuable insights for urban development planning and coastal flooding prevention.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2022)
Article
Computer Science, Information Systems
Francisco Javier Lopez-Andreu, Juan Antonio Lopez-Morales, Manuel Erena, Antonio F. Skarmeta, Juan A. Martinez
Summary: This paper presents a methodology for utilizing Sentinel-2 data for agricultural monitoring, transforming the data into information through machine learning and remote sensing techniques to manage agricultural policy aid and achieve efficient resource allocation.
Article
Environmental Sciences
Eleni Papadopoulou, Giorgos Mallinis, Sofia Siachalou, Nikos Koutsias, Athanasios C. Thanopoulos, Georgios Tsaklidis
Summary: This study aims to design, develop, and evaluate two deep learning architectures for agricultural land cover and crop type mapping. The results show that these architectures outperformed the traditional random forest algorithm in terms of accuracy. The study also highlights the importance of sampling strategy for handling dataset imbalance and spectral variability.
Article
Environmental Sciences
Wiwin Ambarwulan, Fajar Yulianto, Widiatmaka Widiatmaka, Ati Rahadiati, Suria Darma Tarigan, Irman Firmansyah, Muhrina Anggun Sari Hasibuan
Summary: This study aims to identify the land use/land cover (LULC) changes in Cisadane Watershed, Indonesia, and simulate future LULC for 2030 and 2050. The results reveal the trade-off between maintaining food security and conserving natural resources. Efficient land use planning in the future is important to meet increasing resource demand due to population growth.
EGYPTIAN JOURNAL OF REMOTE SENSING AND SPACE SCIENCES
(2023)
Article
Environmental Sciences
Karimon Nesha, Martin Herold, Veronique De Sy, Sytze de Bruin, Arnan Araza, Natalia Malaga, Javier G. P. Gamarra, Kristell Hergoualc'h, Anssi Pekkarinen, Carla Ramirez, David Morales-Hidalgo, Rebecca Tavani
Summary: This study assessed the availability, temporal distribution, and extent of national forest inventories (NFIs) in 236 countries and analyzed the latest NFI design characteristics in 46 tropical countries. The study found significant NFI availability globally, with most multiple NFIs found in temperate and boreal countries. However, NFIs in the tropics were relatively recent and less frequent. The existing NFI designs pose challenges for statistical integration with remotely sensed data in the tropics.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Biodiversity Conservation
Nicolas Labriere, Stuart J. Davies, Mathias Disney, Laura Duncanson, Martin Herold, Simon L. Lewis, Oliver L. Phillips, Shaun Quegan, Sassan S. Saatchi, Dmitry G. Schepaschenko, Klaus Scipal, Plinio Sist, Jerome Chave
Summary: This study aims to establish a global forest biomass reference measurement system. To successfully implement this system, uniform data collection and processing standards, inclusive and equitable system establishment and management, as well as mandatory training and involvement of site partners in downstream activities are emphasized.
GLOBAL CHANGE BIOLOGY
(2023)
Article
Environmental Sciences
Anne-Juul Welsink, Johannes Reiche, Veronique de Sy, Sarah Carter, Bart Slagter, Daniela Requena Suarez, Ben Batros, Marielos Pena-Claros, Martin Herold
Summary: Illegal logging is a major cause of tropical forest loss. Satellite-based alert systems can accurately estimate tree cover loss in logging concessions using 10 m scale satellite data, but reliability is lower in areas with few disturbances. There is a trade-off between aggregation level and accuracy in estimating logging volumes, which presents a challenge for remote verification of logging activities.
ENVIRONMENTAL RESEARCH LETTERS
(2023)
Article
Biodiversity Conservation
Daniela Requena Suarez, Danae M. A. Rozendaal, Veronique De Sy, Mathieu Decuyper, Natalia Malaga, Patricia Duran Montesinos, Alexs Arana Olivos, Ricardo De la Cruz Paiva, Christopher Martius, Martin Herold
Summary: Amazonian forests play a vital role as reservoirs of biomass and biodiversity, contributing to climate change mitigation. This study examines the impact of disturbances on forest biomass and biodiversity in the Peruvian Amazon, using tree-level data and remotely sensed monitoring. The results show that disturbance intensity negatively affects tree species richness and biomass recovery. Surprisingly, time since disturbance has a small negative effect on species richness. Approximately 15% of Peruvian Amazonian forests have experienced disturbance since 1984, with an increase in biomass of 4.7 Mg ha(-1) year(-1) during the first 20 years.
GLOBAL CHANGE BIOLOGY
(2023)
Article
Environmental Sciences
Stefano Tebaldini, Mauro Mariotti d'Alessandro, Lars M. H. Ulander, Patrik Bennet, Anders Gustavsson, Alex Coccia, Karlus Macedo, Mathias Disney, Phil Wilkes, Hans -Joachim Spors, Nico Schumacher, Jan Hanus, Jan Novotny, Benjamin Brede, Harm Bartholomeus, Alvaro Lau, Jens van der Zee, Martin Herold, Dirk Schuettemeyer, Klaus Scipal
Summary: The TomoSense experiment, funded by the ESA, studied remote sensing of forested areas using SAR data, with a focus on using TomoSAR to study the vertical structure of vegetation. The experiment used a temperate forest in the Eifel National Park in Germany, with dominant species being beech and spruce trees. The dataset includes SAR data as well as lidar data for comparison and analysis.
REMOTE SENSING OF ENVIRONMENT
(2023)
Article
Ecology
Na Chen, Nandin-Erdene Tsendbazar, Daniela Requena Suarez, Jan Verbesselt, Martin Herold
Summary: Characterization of regrowing forests is essential for understanding forest dynamics and supporting sustainable forest management. This study analyzed the effects of environmental and human factors on regrowing forests in Brazil. The results showed that the time since disturbance interpreted from satellite time series is the most important predictor for characterizing aboveground biomass and tree cover of regrowing forests.
REMOTE SENSING IN ECOLOGY AND CONSERVATION
(2023)
Article
Ecology
Kim Calders, Benjamin Brede, Glenn Newnham, Darius Culvenor, John Armston, Harm Bartholomeus, Anne Griebel, Jodie Hayward, Samuli Junttila, Alvaro Lau, Shaun Levick, Rosalinda Morrone, Niall Origo, Marion Pfeifer, Jan Verbesselt, Martin Herold
Summary: Climate change and human activities are affecting ecosystems and biodiversity. Quantitative measurements of essential biodiversity variables and climate variables are used to monitor and evaluate interventions. Spaceborne measurements lack detailed information on three-dimensional vegetation structure at local scales, but ground-based laser scanning shows potential for systematic monitoring.
REMOTE SENSING IN ECOLOGY AND CONSERVATION
(2023)
Review
Multidisciplinary Sciences
Osamu Ochiai, Benjamin Poulter, Frank Martin Seifert, Stephen Ward, Ian Jarvis, Alyssa Whitcraft, Ritvik Sahajpal, Sven Gilliams, Martin Herold, Sarah Carter, Laura Innice Duncanson, Heather Kay, Richard Lucas, Sylvia N. Wilson, Joana Melo, Joanna Post, Stephen Briggs, Shaun Quegan, Mark Dowell, Alessandro Cescatti, David Crisp, Sassan Saatchi, Takeo Tadono, Matt Steventon, Ake Rosenqvist
Summary: Space-based remote sensing can play a crucial role in monitoring greenhouse gas emissions and removals from AFOLU sector and addressing climate change through the UNFCCC Paris Agreement. International cooperation, led by CEOS, is essential for the development and realization of a long-term roadmap for observations. This paper identifies useful data and information for supporting the global stocktake of the Paris Agreement and provides a workflow for harmonization and contribution to greenhouse gas inventories and assessments.
Article
Multidisciplinary Sciences
Akane O. Abbasi, Xiaolu Tang, Nancy L. Harris, Elizabeth D. Goldman, Javier G. P. Gamarra, Martin Herold, Hyun Seok Kim, Weixue Luo, Carlos Alberto Silva, Nadezhda M. Tchebakova, Ankita Mitra, Yelena Finegold, Mohammad Reza Jahanshahi, Cesar Ivan Alvarez, Tae Kyung Kim, Daun Ryu, Jingjing Liang
Summary: Planted forests in East Asia, which account for approximately 36% of global planted forest area, play a critical role in climate change mitigation and timber/non-timber production. However, there is limited information available on the geographic distribution and tree species composition of these planted forests. This study presents the first spatial database of planted forests in East Asia, based on extensive data collection and modeling. The maps generated in this study provide valuable information for understanding the role of planted forests in climate change mitigation and guiding forest conservation and management decisions.
Article
Remote Sensing
Arnan Araza, Martin Herold, Sytze de Bruin, Philippe Ciais, David A. Gibbs, Nancy Harris, Maurizio Santoro, Jean-Pierre Wigneron, Hui Yang, Natalia Malaga, Karimon Nesha, Pedro Rodriguez-Veiga, Olga Brovkina, Hugh C. A. Brown, Milen Chanev, Zlatomir Dimitrov, Lachezar Filchev, Jonas Fridman, Mariano Garcia, Alexander Gikov, Leen Govaere, Petar Dimitrov, Fardin Moradi, Adriane Esquivel Muelbert, Jan Novotny, Thomas A. M. Pugh, Mart-Jan Schelhaas, Dmitry Schepaschenko, Krzysztof Sterenczak, Lars Hein
Summary: This study assessed the net Delta AGB derived from four global multi-date AGB maps over the past decade. The comparison between LiDAR data and maps showed reasonable agreement, while the comparisons using NFI only had some agreements at smaller aggregation levels. Disagreement between maps is still large in key forest regions. The results suggest that the AGB assessed from current maps can be biased and any use of the estimates should consider this.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Environmental Sciences
Adugna Mullissa, Johannes Reiche, Martin Herold
Summary: The paper proposes a rapid forest disturbance detection approach for tropical dry forests using Sentinel-1 SAR and deep learning. The method shows effectiveness in detecting disturbances in near real-time, outperforming traditional methods based on Landsat data in terms of timeliness. The proposed approach has potential for operational use in generating large area disturbance alerts in the dry tropics.
REMOTE SENSING OF ENVIRONMENT
(2023)
Correction
Multidisciplinary Sciences
Camille S. Delavaux, Thomas W. Crowther, Constantin M. Zohner, Niamh M. Robmann, Thomas Lauber, Johan van den Hoogen, Sara Kuebbing, Jingjing Liang, Sergio de-Miguel, Gert-Jan Nabuurs, Peter B. Reich, Meinrad Abegg, Yves C. Adou Yao, Giorgio Alberti, Angelica M. Almeyda Zambrano, Braulio Vilchez Alvarado, Esteban Alvarez-Davila, Patricia Alvarez-Loayza, Luciana F. Alves, Christian Ammer, Clara Anton-Fernandez, Alejandro Araujo-Murakami, Luzmila Arroyo, Valerio Avitabile, Gerardo A. Aymard, Timothy R. Baker, Radomir Balazy, Olaf Banki, Jorcely G. Barroso, Meredith L. Bastian, Jean-Francois Bastin, Luca Birigazzi, Philippe Birnbaum, Robert Bitariho, Pascal Boeckx, Frans Bongers, Olivier Bouriaud, Pedro H. S. Brancalion, Susanne Brandl, Roel Brienen, Eben N. Broadbent, Helge Bruelheide, Filippo Bussotti, Roberto Cazzolla Gatti, Ricardo G. Cesar, Goran Cesljar, Robin Chazdon, Han Y. H. Chen, Chelsea Chisholm, Hyunkook Cho, Emil Cienciala, Connie Clark, David Clark, Gabriel D. Colletta, David A. Coomes, Fernando Cornejo Valverde, Jose J. Corral-Rivas, Philip M. Crim, Jonathan R. Cumming, Selvadurai Dayanandan, Andre L. de Gasper, Mathieu Decuyper, Geraldine Derroire, Ben DeVries, Ilija Djordjevic, Jiri Dolezal, Aurelie Dourdain, Nestor Laurier Engone Obiang, Brian J. Enquist, Teresa J. Eyre, Adande Belarmain Fandohan, Tom M. Fayle, Ted R. Feldpausch, Leandro V. Ferreira, Markus Fischer, Christine Fletcher, Lorenzo Frizzera, Javier G. P. Gamarra, Damiano Gianelle, Henry B. Glick, David J. Harris, Andrew Hector, Andreas Hemp, Geerten Hengeveld, Bruno Herault, John L. Herbohn, Martin Herold, Annika Hillers, Euridice N. Honorio Coronado, Cang Hui, Thomas T. Ibanez, Ieda Amaral, Nobuo Imai, Andrzej M. Jagodzinski, Bogdan Jaroszewicz, Vivian Kvist Johannsen, Carlos A. Joly, Tommaso Jucker, Ilbin Jung, Viktor Karminov, Kuswata Kartawinata, Elizabeth Kearsley, David Kenfack, Deborah K. Kennard, Sebastian Kepfer-Rojas, Gunnar Keppel, Mohammed Latif Khan, Timothy J. Killeen, Hyun Seok Kim, Kanehiro Kitayama, Michael Kohl, Henn Korjus, Florian Kraxner, Diana Laarmann, Mait Lang, Simon L. Lewis, Huicui Lu, Natalia V. Lukina, Brian S. Maitner, Yadvinder Malhi, Eric Marcon, Beatriz Schwantes Marimon, Ben Hur Marimon-Junior, Andrew R. Marshall, Emanuel H. Martin, Olga Martynenko, Jorge A. Meave, Omar Melo-Cruz, Casimiro Mendoza, Cory Merow, Abel Monteagudo Mendoza, Vanessa S. Moreno, Sharif A. Mukul, Philip Mundhenk, Maria Guadalupe Nava-Miranda, David Neill, Victor J. Neldner, Radovan V. Nevenic, Michael R. Ngugi, Pascal A. Niklaus, Jacek Oleksyn, Petr Ontikov, Edgar Ortiz-Malavasi, Yude Pan, Alain Paquette, Alexander Parada-Gutierrez, Elena I. Parfenova, Minjee Park, Marc Parren, Narayanaswamy Parthasarathy, Pablo L. Peri, Sebastian Pfautsch, Oliver L. Phillips, Nicolas Picard, Maria Teresa T. F. Piedade, Daniel Piotto, Nigel C. A. Pitman, Irina Polo, Lourens Poorter, Axel D. Poulsen, Hans Pretzsch, Freddy Ramirez Arevalo, Zorayda Restrepo-Correa, Mirco Rodeghiero, Samir G. Rolim, Anand Roopsind, Francesco Rovero, Ervan Rutishauser, Purabi Saikia, Christian Salas-Eljatib, Philippe Saner, Peter Schall, Dmitry Schepaschenko, Michael Scherer-Lorenzen, Bernhard Schmid, Jochen Schongart, Eric B. Searle, Vladimir Seben, Josep M. Serra-Diaz, Douglas Sheil, Anatoly Z. Shvidenko, Javier E. Silva-Espejo, Marcos Silveira, James Singh, Plinio Sist, Ferry Slik, Bonaventure Sonke, Alexandre F. Souza, Stanislaw Miscicki, Krzysztof J. Sterenczak, Jens-Christian Svenning, Miroslav Svoboda, Ben Swanepoel, Natalia Targhetta, Nadja Tchebakova, Hans ter Steege, Raquel Thomas, Elena Tikhonova, Peter M. Umunay, Vladimir A. Usoltsev, Renato Valencia, Fernando Valladares, Fons van der Plas, Tran Van Do, Michael E. van Nuland, Rodolfo M. Vasquez, Hans Verbeeck, Helder Viana, Alexander C. Vibrans, Simone Vieira, Klaus von Gadow, Hua-Feng Wang, James V. Watson, Gijsbert D. A. Werner, Susan K. Wiser, Florian Wittmann, Hannsjoerg Woell, Verginia Wortel, Roderik Zagt, Tomasz Zawila-Niedzwiecki, Chunyu Zhang, Xiuhai Zhao, Mo Zhou, Zhi-Xin Zhu, Irie C. Zo-Bi, Daniel S. Maynard
Article
Remote Sensing
Johannes Balling, Martin Herold, Johannes Reiche
Summary: Accurate information about tropical forest disturbances is crucial for forest management and law enforcement. Monitoring forest disturbances using cloud-penetrating SAR imagery has shown effective results. However, current methods based on backscatter values may cause omission errors. This research proposes a method to quantify the heterogeneity of neighboring pixel values using textural features, in order to overcome omission errors caused by post-disturbance tree remnants or debris. The combination of SAR-based textural features and backscatter improves the consistency and timeliness of forest disturbance monitoring.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Remote Sensing
Natalia Malaga, Sytze de Bruin, Ronald E. McRoberts, Alexs Arana Olivos, Ricardo de la Cruz Paiva, Patricia Duran Montesinos, Daniela Requena Suarez, Martin Herold
Summary: This study evaluates the use of a global aboveground biomass (AGB) map as auxiliary information for subnational AGB estimates in the Peruvian Amazonia and analyzes the sources of uncertainty. The results show that the calibrated map can improve the precision of AGB estimates, but the contribution of within block variability is significant.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Environmental Sciences
Milutin Milenkovic, Johannes Reiche, John Armston, Amy Neuenschwander, Wanda De Keersmaecker, Martin Herold, Jan Verbesselt
Summary: Two satellite LiDAR missions, GEDI and ICESat-2, provide global measurements of forest height and structure. This study utilized both missions' data to assess regrowth rates in regrowing forests of different ages in Rondonia, Brazil. A calibration model was derived to improve the accuracy of satellite height measurements, and the results showed reliable estimates of forest regrowth rates over large areas.
SCIENCE OF REMOTE SENSING
(2022)