Article
Environmental Sciences
Pan Zhou, Zhibin Sun, Xiongqing Zhang, Yixiang Wang
Summary: This study provides a novel framework for planning thinning operations in a pure managed forest using Unmanned Aerial Vehicle (UAV) remote sensing techniques. The framework allows obtaining forest attributes and optimizing thinning areas, intensities, and cut-trees. Results from a case study in a subtropical Chinese fir plantation show the potential of low-cost UAV-acquired RGB images in depicting forest structure.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Environmental Sciences
Hu Li, Chengxin Qin, Weiqi He, Fu Sun, Pengfei Du
Summary: This paper investigates the sub-daily dynamics of cyanobacterial harmful algal blooms (CyanoHABs) in Taihu Lake using a combination of high-frequency time-series remote sensing and hydro-ecological modelling. The results show that the distribution of CyanoHABs is patchy and dynamic, largely influenced by the migratory behavior of cyanobacteria. The hydro-ecological model successfully reproduces the observed pattern and trend, with photosynthesis rate and respiration rate identified as the most influential model parameters. The study highlights the need for microscale modelling to better understand the physiological processes of CyanoHABs.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2022)
Article
Plant Sciences
Ruiyang Yu, Xicun Zhu, Xueyuan Bai, Zhongyu Tian, Yuanmao Jiang, Guijun Yang
Summary: By comparing the accuracy of two methods for inverting reflectance at multiple wavelengths, it was found that the Ground-UAV-Linear Spectral Mixture Model (G-UAV-LSMM) is more suitable for obtaining canopy inversion reflectance.
JOURNAL OF PLANT RESEARCH
(2021)
Article
Environmental Sciences
Shijie Yan, Linhai Jing, Huan Wang
Summary: This study investigates the potential of using high-resolution satellite remote sensing data in combination with a convolutional neural network to recognize individual tree species. The modified GoogLeNet model achieved an overall accuracy of 82.7% in recognizing six individual tree species, and this method avoids manual tree crown delineation.
Article
Environmental Sciences
Tengfei Yang, Jibo Xie, Peilin Song, Guoqing Li, Naixia Mou, Xinyue Gao, Jing Zhao
Summary: The ecological environment is crucial for human survival and development, and effective monitoring methods are essential for human settlements. Traditional data and methods have limitations, while social media data can supplement them. This paper proposes a framework that integrates social media, remote sensing, and other data to monitor the ecological environment. The study extracts relevant information from social media data and constructs a social semantic network for comprehensive analysis.
Article
Environmental Sciences
Cristian Rossi, Luke Bateson, Maral Bayaraa, Andrew Butcher, Jonathan Ford, Andrew Hughes
Summary: The demand for green metals such as lithium is increasing as the world works to reduce reliance on fossil fuels. This research aims to track lithium mass from its source to its greatest concentration in the nucleus, with a focus on the Salar de Uyuni in Bolivia. The study highlights the importance of groundwater flow in the formation of lithium-brine deposits in the Lithium Triangle.
Article
Geography, Physical
Jianhua Guo, Qingsong Xu, Yue Zeng, Zhiheng Liu, Xiao Xiang Zhu
Summary: Urban tree canopy maps are crucial for providing urban ecosystem services. This study developed a semi-supervised deep learning method to robustly segment urban trees from high-resolution remote sensing images in order to better serve Brazil's urban ecosystem. The results showed that the urban tree canopy coverage in Brazil ranges from 5% to 35%, with an average coverage of approximately 18.68%. These canopy maps quantified the nationwide urban tree canopy inequality problem in Brazil. It is expected that these maps will encourage research on Brazilian urban ecosystem services, support urban development, and improve inhabitants' quality of life to achieve the Agenda for Sustainable Development goals.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Forestry
Yongchao Liu, Ruyun Zhang, Chen-Feng Lin, Zhaochen Zhang, Ran Zhang, Kankan Shang, Mingshui Zhao, Jingyue Huang, Xiaoning Wang, You Li, Yulin Zeng, Yun-Peng Zhao, Jian Zhang, Dingliang Xing
Summary: Tree species diversity is crucial for ecosystem functions, but traditional field-based approaches limit our ability to map it. However, recent advances in spaceborne remote sensing provide new opportunities. In this study, using Sentinel-2 satellite data and forest plot inventory data, we developed regression models to predict tree diversity in an eastern Chinese natural reserve.
Article
Environmental Sciences
Lili Liu, Meng Chen, Pingping Luo, Weili Duan, Maochuan Hu
Summary: This study constructs an ecological security network based on multi-factor ecological sensitivity using machine learning, remote sensing, geographic information systems, analytic hierarchy process and principal component analysis. The results demonstrate a coupling relationship among comprehensive ecological sensitivity, ecological security patterns, and administrative districts. The study provides a quantitative foundation for urban and rural ecological spatial planning in Yangxian and contributes to the sustainable development of ecological planning in the Qinling region.
Article
Ecology
Kushan Aravinda Bellanthudawa, Ni-Bin Chang
Summary: Understanding the long-term impact of intermittent weather events on diverse forest and plant communities is crucial in sustainability science. Remote sensing technology can be used to assess these impacts and compare the response of different land uses. Research findings show that land surface temperature can affect the biophysical and biochemical properties of canopies, as well as the gross primary productivity.
ECOLOGICAL INFORMATICS
(2022)
Article
Construction & Building Technology
Jianhua Guo, Zhiheng Liu, Xiao Xiang Zhu
Summary: This study creates a high-resolution urban tree canopy cover dataset using deep learning based on remote sensing images, revealing the heterogeneity of tree canopy cover in Brazilian cities. The results show regional variations in tree canopy cover and a difference between old and new urban areas, with most urban populations exposed to low tree canopy coverage. Climate factors play a major role in determining the tree canopy cover patterns. Therefore, the study suggests the Brazilian government should focus on greening renovation in old urban areas and formulate effective tree irrigation policies for cities with limited rainfall.
SUSTAINABLE CITIES AND SOCIETY
(2024)
Article
Economics
Jesus Crespo Cuaresma, Sebastian Uljas Lutz
Summary: The study identifies economic and demographic covariates as robust predictors of digital variables, and predicts that convergence trends related to digital technology access will continue, but the digital divide in Europe is not expected to disappear in the coming years without specific policy interventions.
Article
Geochemistry & Geophysics
Qi Wang, Wei Huang, Xueting Zhang, Xuelong Li
Summary: This article proposes a new explainable word-sentence framework for remote sensing image captioning, decomposing RSIC into a word classification task and a word sorting task, which is more in line with human intuitive understanding.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Chen Xu, Xiaoping Du, Xiangtao Fan, Zhenzhen Yan, Xujie Kang, Junjie Zhu, Zhongyang Hu
Summary: This research analyzed the processing flow of remote sensing big data from the perspective of computer science and remote sensing science, proposing a modular framework. By introducing computation ready data as a dynamic data type to connect key modules of the framework, it significantly reduces experimental costs for remote sensing researchers.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Donatella Occorsio, Giuliana Ramella, Woula Themistoclakis
Summary: In this study, a framework was proposed to assess recent image resizing methods for remote sensing applications. Extensive experiments were conducted on multiple public remote sensing image datasets and two new datasets included in the framework to evaluate the performance of each method in terms of image quality and statistical measures.
Article
Geosciences, Multidisciplinary
Mari Myllymaki, Mikko Kuronen, Tomas Mrkvicka
Summary: A Monte Carlo version of testing for a covariate effect in a parametric point process model is proposed, with two different test statistics suggested for local investigation of the covariate effect in the globally fitted model. Simulation scheme resembling the permutation inference for GLMs is used to obtain the null distribution of the statistics, followed by a Monte Carlo test with graphical interpretation to determine the global significance of the covariate and the spatially significant areas. The proposed statistics are shown to be valuable for model construction through application to simulated and real point pattern data.
SPATIAL STATISTICS
(2021)
Article
Forestry
Roope Ruotsalainen, Timo Pukkala, Annika Kangas, Mari Myllymaki, Petteri Packalen
Summary: The study found that increasing carbon prices and reducing error levels led to decreased losses in NPV. Inclusion of carbon payments in maximizing NPV reduced the impact of errors on losses, indicating that the value of collecting more accurate forest inventory data may decrease as carbon prices rise.
CANADIAN JOURNAL OF FOREST RESEARCH
(2021)
Article
Biodiversity Conservation
Soyeon Bae, Lea Heidrich, Shaun R. Levick, Martin M. Gossner, Sebastian Seibold, Wolfgang W. Weisser, Paul Magdon, Alla Serebryanyk, Claus Baessler, Deborah Schafer, Ernst-Detlef Schulze, Inken Doerfler, Jorg Mueller, Kirsten Jung, Marco Heurich, Markus Fischer, Nicolas Roth, Peter Schall, Steffen Boch, Stephan Woellauer, Swen C. Renner, Joerg Mueller
Summary: Despite increasing interest in beta-diversity, the mechanisms underlying species turnover at different spatial scales are not fully understood. Our study revealed that environmental factors and landscape spatial structure have different effects on community composition, and these effects vary among different functional groups.
DIVERSITY AND DISTRIBUTIONS
(2021)
Article
Mathematical & Computational Biology
Mikko Kuronen, Mari Myllymaki, Adam Loavenbruck, Aila Sarkka
Summary: This article aims to construct spatial models for sweat gland activation in healthy subjects and those with peripheral neuropathy using videos of sweating. Two point process models and inference methods are proposed, with the acknowledgement of potential errors in identifying sweat gland locations in the data. Image analysis steps and robust estimation procedures are necessary to address these errors.
STATISTICS IN MEDICINE
(2021)
Article
Environmental Sciences
Mikko Kuronen, Aila Sarkka, Matti Vihola, Mari Myllymaki
Summary: A hierarchical log Gaussian Cox process (LGCP) is proposed to study the influence of one set of points on another set in point patterns, where each point in the influencing set has a parameterized signal forming an influence field. Parameters of the model are estimated in a Bayesian framework using Markov chain Monte Carlo, with Laplace approximation for the Gaussian field. The model is applied to analyze the impact of large trees on regeneration success in uneven-aged forest stands in Finland.
ENVIRONMENTAL AND ECOLOGICAL STATISTICS
(2022)
Article
Mathematical & Computational Biology
Tomas Mrkvicka, Mari Myllymaki, Mikko Kuronen, Naveen Naidu Narisetty
Summary: Permutation methods are commonly used to test the significance of regressors in GLMs for functional data sets, and new multiple testing methods have been proposed to improve power and robustness. The methods rely on sorting the permuted functional test statistics based on pointwise rank measures.
STATISTICS IN MEDICINE
(2022)
Article
Agriculture, Multidisciplinary
Sandra Mueller, Martin M. Gossner, Caterina Penone, Kirsten Jung, Swen C. Renner, Almo Farina, Lisa Anhaeuser, Manfred Ayasse, Steffen Boch, Falk Haensel, Janine Heitzmann, Christoph Kleinn, Paul Magdon, David J. Perovic, Nadia Pieretti, Taylor Shaw, Juliane Steckel, Marco Tschapka, Juliane Vogt, Catrin Westphal, Michael Scherer-Lorenzen
Summary: Understanding drivers and monitoring changes of biodiversity is crucial for evidence-based management and policy recommendations. Ecoacoustic monitoring offers the potential for resource-efficient ecosystem monitoring. Acoustic diversity has been shown to correlate with species richness and vegetation and landscape structure. This study found that land-use intensity and landscape structure affect species richness and composition of birds and orthopterans, indirectly impacting acoustic diversity and composition.
AGRICULTURE ECOSYSTEMS & ENVIRONMENT
(2022)
Letter
Mathematical & Computational Biology
Tomas Mrkvicka, Mari Myllymaki
STATISTICS IN MEDICINE
(2022)
Article
Computer Science, Artificial Intelligence
Maximilian Freudenberg, Paul Magdon, Nils Noelke
Summary: We propose a deep learning-based framework for individual tree crown delineation in aerial and satellite images. The method creates irregular polygons for tree crown delineation instead of bounding boxes and provides a tree cover mask for areas that are not separable. It is trainable with low amounts of training data and does not require 3D height information. The results show that the approach can efficiently delineate individual tree crowns in high-resolution optical images.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Forestry
Petteri Packalen, Jacob Strunk, Matti Maltamo, Mari Myllymaki
Summary: In airborne laser scanning (ALS)-based forest inventories, there is often a discrepancy between the circular plot shape used for model fitting and the square shape of population elements used for predictions. This study found that for equal area square and circular plots, there was no evidence of systematic prediction error when a model fitted to one shape was used to predict for the other shape. However, using a model fitted to circular plots to predict for square plots slightly underestimated the root mean square error (RMSE) value.
Article
Ecology
Lukas Drag, Ryan C. Burner, Jorg G. Stephan, Tone Birkemoe, Inken Doerfler, Martin M. Gossner, Paul Magdon, Otso Ovaskainen, Maria Potterf, Peter Schall, Tord Snall, Anne Sverdrup-Thygeson, Wolfgang Weisser, Joerg Mueller
Summary: Climate, topography, and forest structure play important roles in shaping local species communities, but little is known about how environmental characteristics affect the functional traits of wood-living beetles involved in wood recycling. In this study, we used ecological and morphological traits of saproxylic beetles and airborne laser scanning data to investigate the factors driving the distributions of over 230 species in European temperate forests. We found that elevation and the proportion of conifers were important factors determining species occurrences, while habitat heterogeneity and forest complexity had less influence. Ecological traits, such as canopy niche, wood decay niche, and host preference, were found to be more important in shaping species responses to forest structure and environmental variation, while morphological traits showed little association with environmental characteristics. These findings highlight the potential impacts of climate and tree species composition changes on saproxylic beetle communities, and the importance of ecological traits in predicting species responses to future environmental change.
FUNCTIONAL ECOLOGY
(2023)
Article
Biodiversity Conservation
Niklas Hagemann, Paul Magdon, Sebastian Schnell, Arne Pommerening
Summary: This study analyzed canopy gap dynamics in the Krycklan catchment area in northern Sweden using the Boolean model and landscape metrics. The results showed no significant trend of harmful development in forest resources between 2006 and 2015, with evidence of gaps decreasing, becoming more random, and canopy cover increasing.
ECOLOGICAL INDICATORS
(2022)
Article
Environmental Sciences
Javier Muro, Anja Linstaedter, Paul Magdon, Stephan Woellauer, Florian A. Maenner, Lisa-Maricia Schwarz, Gohar Ghazaryan, Johannes Schultz, Zbynek Malenovsky, Olena Dubovyk
Summary: This study successfully predicted above-ground biomass and species richness in grassland using remote sensing sensors and machine learning methods. The feed-forward deep neural network (DNN) outperformed the random forest (RF) algorithm in terms of prediction accuracy and generalizability.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Environmental Sciences
J. Kostensalo, L. Mehtatalo, S. Tuominen, P. Packalen, M. Myllymaki
Summary: This study developed a framework for building structurally representative tree maps using airborne laser scanning data and ground measurements, which can accurately map the attributes of forests. Compared to other methods, this approach improves the accuracy of mapping forest attributes.
REMOTE SENSING OF ENVIRONMENT
(2023)
Article
Remote Sensing
Lauri Mehtatalo, Adil Yazigi, Kasper Kansanen, Petteri Packalen, Timo Lahivaara, Matti Maltamo, Mari Myllymaki, Antti Penttinen
Summary: This study discusses methods for estimating tree-level characteristics using a Horvitz-Thompson-like estimator and a new method based on a sequential spatial point process model. The new method showed comparable performance to the HT-like estimator, with lower computational demands, making it attractive for practical use.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)