Article
Environmental Sciences
Putri Setiani, Luhur Akbar Devianto, Fatwa Ramdani
Summary: This study utilized Google Earth Engine platform to monitor forest fire events in Mount Arjuno, Indonesia during 2016-2019, estimating a total emission of 2.5 x 10(3) tCO(2)/km(2). Higher carbon dioxide emissions were observed in areas affected by the fires, consistent with higher local surface temperatures during the period.
ENVIRONMENTAL MONITORING AND ASSESSMENT
(2021)
Article
Remote Sensing
Yulin Shangguan, Xiyu Li, Yi Lin, Jinsong Deng, Le Yu
Summary: This study successfully extracted and analyzed the nationwide soybean planting areas in Argentina from 2016 to 2019 using the Google Earth Engine and pixel-based machine learning method random forest. The results showed the importance of NDVI and NIR features in the classification. This research provides an effective method for accurately and rapidly retrieving the soybean planting area.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Review
Environmental Sciences
Qiang Zhao, Le Yu, Xuecao Li, Dailiang Peng, Yongguang Zhang, Peng Gong
Summary: Earth system science has evolved rapidly due to global environmental changes and the emergence of Earth observation technology. Google Earth (GE) and Google Earth Engine (GEE) have become essential tools for monitoring, analyzing, and modeling Earth observation data, with GEE experiencing faster growth in applications compared to GE. These tools are widely used in multidisciplinary research areas, with GEE focusing more on big data and time-series analysis, while GE is primarily used for visualization purposes.
Article
Remote Sensing
Nelson Diaz, Omar Gallo, Jhon Caceres, Hernan Porras
Summary: This study introduces two ground filtering algorithms based on normal vectors and voxel structure, respectively, for 3D modeling with point clouds. Comparisons were made in terms of execution time, effectiveness, and efficiency, showing that the voxel structure-based algorithm outperformed the normal vector ground filtering algorithm in all aspects.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Environmental Sciences
Andres Kuusk, Allan Sims
Summary: This study validates the hot-spot theoretical model in the Jarvselja RAMI pine stand using extensive terrestrial laser scanning (TLS) measurements. A point cloud of laser hits with a resolution of 1 cm was created to describe the spatial structure of the crown layer. The study found that the determined value of the hotspot parameter agrees well with the value estimated indirectly from measurements of reflectance profile.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Plant Sciences
Miro Demol, Kim Calders, Hans Verbeeck, Bert Gielen
Summary: This study evaluated the use of terrestrial laser scanning (TLS) for estimating tree volumes, finding that the method was slightly more reliable compared to reference volumes and allometric scaling models (ASMs), but with some errors in finer details.
Article
Chemistry, Analytical
Dimitrios Panagiotidis, Azadeh Abdollahnejad, Martin Slavik
Summary: This study compared two different methods for accurate estimation of tree heights and diameters, and evaluated their accuracy in assessing total tree stem volume. Results showed a strong relationship between the measured and estimated stem volumes, indicating the feasibility of using fine-resolution remote sensing data for precise forestry.
Article
Environmental Sciences
Aline Pontes-Lopes, Ricardo Dalagnol, Andeise Cerqueira Dutra, Camila Valeria de Jesus Silva, Paulo Mauricio Lima de Alencastro Graca, Luiz Eduardo de Oliveira E Cruz de Aragao
Summary: Fire is a major factor contributing to forest degradation in the Amazon, and understanding how post-fire canopy changes affect spectral signals is important. This study found that spectral indices can improve the prediction of post-fire tree loss, and enhance the accuracy of carbon emission estimates. The integration of structural and spectral-based spatial data has potential in studying post-fire ecological processes in the Amazon.
Article
Environmental Sciences
Yuan Qi, Shiwei Li, Youhua Ran, Hongwei Wang, Jichun Wu, Xihong Lian, Dongliang Luo
Summary: The study simulated the dynamics of active layer thickness, top temperature, and maximum frozen soil depth in the Qilian Mountains from 2004 to 2019, showing a decrease in permafrost area and thickness over this period. TTOP and ALT increase with decreasing elevation, while the rate of permafrost loss accelerated during 2004-2019.
Article
Environmental Sciences
Jie Cheng, Nan Jia, Ruishan Chen, Xiaona Guo, Jianzhong Ge, Fucang Zhou
Summary: Seaweed aquaculture plays a significant role in achieving sustainable development goals, but its large-scale development and unreasonable use may have negative impacts. Therefore, accurate monitoring of the seaweed aquaculture industry is crucial. This study used remote sensing and random forest algorithm to map the distribution of seaweed aquaculture along the Jiangsu coast of China, revealing a significant reduction in aquaculture scale under policy restrictions.
Article
Environmental Sciences
Farzane Mohseni, Meisam Amani, Pegah Mohammadpour, Mohammad Kakooei, Shuanggen Jin, Armin Moghimi
Summary: In this study, a wetland map of the GL region was created using Sentinel-1/2 datasets and the Google Earth Engine. A supervised machine learning classification workflow was used, with two main steps to accurately classify wetland and non-wetland areas. The overall accuracy and kappa coefficient of the classification results demonstrated the effectiveness of the proposed method.
Article
Environmental Sciences
Hui Liu, Mi Chen, Huixuan Chen, Yu Li, Chou Xie, Bangsen Tian, Chu Wang, Pengfei Ge
Summary: This paper utilizes multiple machine learning algorithms, supported by the Google Earth Engine platform and based on Landsat 8 time series image data, to obtain the spatial variation information of agricultural land in Shandong Province from 2016 to 2020. The results show that the multi-spatial index time series method and the ensemble learning method have higher accuracy in obtaining phenological characteristics of agricultural land and classification.
Article
Remote Sensing
Chao Zhang, Jinwei Dong, Yanhua Xie, Xuezhen Zhang, Quansheng Ge
Summary: This study developed a remote sensing-based method to map irrigated croplands in China using a synergetic training sample generating method and machine learning classifier. The resulting map showed high accuracy and can provide valuable information for understanding the distribution of irrigated croplands in China and assisting water resource management and agricultural decision-making.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Agriculture, Multidisciplinary
Chengkang Zhang, Hongyan Zhang, Sijing Tian
Summary: Accurate and consistent mapping of paddy rice is crucial for food security and market stability. This study proposed a flexible Phenology-assisted Supervised Paddy Rice (PSPR) mapping framework on Google Earth Engine (GEE) to overcome the challenges of sparse observations, weather contamination, and shortage of training samples. High-resolution 30-m paddy rice maps of Heilongjiang Province in China from 1990 to 2020 were successfully generated and validated. The results showed improved performance compared to previous studies and revealed a significant northward shift in rice planting earlier than expected.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Environmental Sciences
Indishe P. Senanayake, In-Young Yeo, George A. Kuczera
Summary: Australian inland riparian wetlands located east of the Great Dividing Range face rapid decline due to water competition, climate change, and lack of reliable data. This study presents a method using a random forest-based multi-index classification algorithm to construct a long-term time record of inundation maps, providing crucial information for wetland ecosystem management.
Article
Ecology
Jessica M. Stitt, Andrew T. Hudak, Carlos A. Silva, Lee A. Vierling, Kerri T. Vierling
Summary: The study tested a method for quantifying canopy gaps around snags and live trees, finding that snags had more gaps surrounding them than live trees. It suggests that incorporating lidar-derived canopy gap analyses can improve snag modelling and enhance understanding of gap dynamics in closed-canopy forests. Highest differences in canopy gaps were observed at mid-canopy heights and smallest footprint size.
METHODS IN ECOLOGY AND EVOLUTION
(2022)
Article
Environmental Sciences
Jessica M. Stitt, Andrew T. Hudak, Carlos A. Silva, Lee A. Vierling, Kerri T. Vierling
Summary: The study evaluated the feasibility of inferring snag characteristics using ALS data, and found that ALS data alone were not sufficient to identify top intactness for large snags. Future research is encouraged to combine ALS data with other remotely sensed data to improve the classification of snag characteristics important for wildlife.
Article
Environmental Sciences
Arjan J. H. Meddens, Michelle M. Steen-Adams, Andrew T. Hudak, Francisco Mauro, Paige M. Byassee, Jacob Strunk
Summary: This study investigates the preferred geospatial data characteristics for monitoring and managing forests and fuels across large landscapes by natural resource professionals. The survey and workshop results show that attributes related to species composition, total biomass/timber volume, and vegetation height are the most preferred, with preference varying slightly based on employment type.
ENVIRONMENTAL RESEARCH LETTERS
(2022)
Article
Environmental Sciences
Jonathan L. Batchelor, Todd M. Wilson, Michael J. Olsen, William J. Ripple
Summary: We have developed new measures of structural complexity using single point terrestrial laser scanning (TLS) point clouds. These metrics, which include depth, openness, and isovist, can accurately capture the structural complexity of forests without observer bias. They have the potential to quantify structural change in forest ecosystems, measure the effects of forest management activities, and describe habitat for organisms.
Article
Chemistry, Analytical
Hastings Shamaoma, Paxie W. Chirwa, Jules C. Zekeng, Abel Ramoelo, Andrew T. Hudak, Ferdinand Handavu, Stephen Syampungani
Summary: This study utilized multi-date multispectral Unmanned Aerial Systems (UAS) imagery and object-based image analysis (OBIA) to classify dominant canopy tree species in the wet Miombo woodlands of Zambia. The results demonstrated that using a combination of different time periods and spectral indices improved the accuracy of species classification.
Review
Environmental Sciences
Ewane Basil Ewane, Midhun Mohan, Shaurya Bajaj, G. A. Pabodha Galgamuwa, Michael S. Watt, Pavithra Pitumpe Arachchige, Andrew T. Hudak, Gabriella Richardson, Nivedhitha Ajithkumar, Shruthi Srinivasan, Ana Paula Dalla Corte, Daniel J. Johnson, Eben North Broadbent, Sergio de-Miguel, Margherita Bruscolini, Derek J. N. Young, Shahid Shafai, Meshal M. Abdullah, Wan Shafrina Wan Mohd Jaafar, Willie Doaemo, Carlos Alberto Silva, Adrian Cardil
Summary: Protecting and enhancing forest carbon sinks is important for mitigating climate change, but droughts caused by climate change can threaten their stability and growth. The use of unmanned aerial vehicles (UAVs) has the potential to bridge the gap between field inventory and satellite remote sensing for assessing forest characteristics and their responses to drought conditions. UAVs can also help optimize forest carbon management with climate change adaptation and mitigation practices.
Article
Environmental Sciences
Yifan Qiao, Guang Zheng, Zihan Du, Xiao Ma, Jiarui Li, L. Monika Moskal
Summary: Accurate classification of tree species is crucial for monitoring, managing, and conserving forest resources. This study utilized ALS data and hyperspectral data to extract four categories of indicators and applied them to the random forest algorithm for tree species classification, achieving an overall accuracy of 84.4%. By introducing individual-tree structure parameters into the constant allometric ratio (CAR) biomass model, biomass models for three tree species were established, and the model-fitting effects were improved after incorporating crown parameters.
Article
Environmental Sciences
Kleydson Diego Rocha, Carlos Alberto Silva, Diogo N. Cosenza, Midhun Mohan, Carine Klauberg, Monique Bohora Schlickmann, Jinyi Xia, Rodrigo Leite, Danilo Roberti Alves de Almeida, Jeff W. Atkins, Adrian Cardil, Eric Rowell, Russ Parsons, Nuria Sanchez-Lopez, Susan J. Prichard, Andrew T. Hudak
Summary: This study compared crown metrics derived from terrestrial and airborne laser scanners, as well as a combination of both, for describing the crown structure and fuel attributes of longleaf pine forest in Florida, USA. The results showed that both terrestrial and airborne laser scanner data accurately predicted tree attributes with good correlation and low errors.
Article
Environmental Sciences
Jonathan L. Batchelor, Eric Rowell, Susan Prichard, Deborah Nemens, James Cronan, Maureen C. Kennedy, L. Monika Moskal
Summary: Lidar can be used to estimate the moisture content of dead forest litter by scanning with lasers. The study found a strong correlation between lidar intensity and standard deviation of intensity per sample tray, and the moisture content of the dead leaf litter. Lidar has the potential to detect and quantify fuel moisture levels in real-time and create spatial maps of wildland fuel moisture content.
Article
Ecology
Eva Louise Loudermilk, Scott Pokswinski, Christie M. Hawley, Aaron Maxwell, Michael R. Gallagher, Nicholas S. Skowronski, Andrew T. Hudak, Chad Hoffman, John Kevin Hiers
Summary: This study demonstrates the use of single-scan terrestrial laser scanning (TLS) combined with physical measurements to predict understory vegetation and fuel biomass, providing an efficient estimation method for managing prescribed fire and studying its effects.
Article
Ecology
Chad M. Hoffman, Justin P. Ziegler, Wade T. Tinkham, John Kevin Hiers, Andrew T. Hudak
Summary: This study compared the performance of four spatial interpolation methods for mapping fine-scale fuel loads and found that regression kriging outperformed the other approaches.
Article
Forestry
Aaron M. Sparks, Alistair M. S. Smith, Andrew T. Hudak, Mark V. Corrao, Robert L. Kremens, Robert F. Keefe
Summary: To improve our understanding of ecological effects following fire and inform natural resource management decisions, it is necessary to integrate pre-, active-, and post-fire measurements to quantify fire effects across multiple spatial scales. Most studies use low-resolution optical multispectral data, which is limited in its ability to quantify tree-level effects and changes in forest structure. Furthermore, most studies do not integrate active fire behavior observations, hindering their ability to identify mechanisms of tree injury and predict fire effects. Combining active fire observations and structural measurements derived from multitemporal airborne laser scanning (ALS) data can potentially address these limitations and provide a scalable method for assessing fire effects on tree structure and growth.
FOREST ECOLOGY AND MANAGEMENT
(2023)
Article
Environmental Sciences
Jonathan L. Batchelor, Andrew T. Hudak, Peter Gould, L. Monika Moskal
Summary: This study demonstrates that single-scan TLS plots can effectively quantify fine-scale forest structure elements relevant to threatened species habitats, and can be used to inform larger area models using airborne lidar.
