Article
Biochemical Research Methods
Ben G. Weinstein, Sarah J. Graves, Sergio Marconi, Aditya Singh, Alina Zare, Dylan Stewart, Stephanie A. Bohlman, Ethan P. White
Summary: The study established a benchmark dataset to evaluate crown detection and delineation methods for canopy trees in dominant forest types in the United States, consisting of thousands of image-annotated crowns and field-annotated crowns, as well as training crowns. Standardizing evaluation metrics helped streamline comparisons between different methods.
PLOS COMPUTATIONAL BIOLOGY
(2021)
Article
Ecology
Alwin A. Hardenbol, Lauri Korhonen, Mikko Kukkonen, Matti Maltamo
Summary: In a landscape dominated by intensive forestry with some protected areas, multifunctional forestry with retention trees plays a crucial role in nature conservation. This study successfully detected and classified retention trees using nationwide Finnish airborne laser scanning data in combination with unrectified color-infrared aerial imagery. Different detection rates were observed for dead and living trees, and overall accuracy improved when spectral information was added to the data analysis.
METHODS IN ECOLOGY AND EVOLUTION
(2023)
Article
Forestry
Hans Ole Orka, Endre Hansen, Michele Dalponte, Terje Gobakken, Erik Naesset
Summary: This study utilized airborne laser scanning and hyperspectral data along with field sample plots to provide species composition in forest management inventories. The results showed acceptable accuracy in estimating species-specific volume proportions, with improved results compared to traditional approaches. Dominant species were classified with an overall accuracy of 91%, indicating the potential of using remotely sensed data for estimating tree species composition and volumes.
Article
Environmental Sciences
Kai O. Bergmueller, Mark C. Vanderwel
Summary: This study used spectral information from UAV imagery to predict tree mortality in 38 forest stands in western Canada. The inclusion of multispectral indices improved the prediction accuracy, and different tree species had varying levels of prediction performance. However, all models tended to overpredict tree mortality.
Article
Environmental Sciences
Alika Polyakova, Svetlana Mukharamova, Oleg Yermolaev, Galiya Shaykhutdinova
Summary: This study examines the potential for discriminating tree species in coniferous-deciduous forests using Sentinel-2 data and two automated recognition methods. The results indicate that the random forest method has higher recognition accuracy.
Article
Environmental Sciences
Marcin Kluczek, Bogdan Zagajewski, Tomasz Zwijacz-Kozica
Summary: Europe's mountain forests, valuable for their biodiversity and natural characteristics, are undergoing significant changes. Monitoring these forests requires up-to-date information on species composition, extent, and location, as well as the selection of appropriate remote sensing data.
Article
Environmental Sciences
Aaron M. Sparks, Mark Corrao, Alistair M. S. Smith
Summary: This study evaluated the accuracy of seven individual tree detection methods in coniferous forest stands, showing that higher ALS pulse density data resulted in higher ITD accuracy. Omission errors were mainly related to stand density, and the use of simple canopy height model methods could reduce omission errors.
Article
Environmental Sciences
Y. Wang, L. Suarez, T. Poblete, V Gonzalez-Dugo, D. Ryu, P. J. Zarco-Tejada
Summary: This study investigated the combination of different hyperspectrally derived proxies for leaf nitrogen (N) to assess N status in a 1200-ha almond orchard across two growing seasons. The results showed that the RTM-based chlorophyll a + b content and solar-induced fluorescence (SIF) were the most important and consistent predictors for leaf N compared to other leaf biochemical and biophysical traits. The combination of non-collinear SIF and chlorophyll a + b content significantly improved the predictions of leaf N variability.
REMOTE SENSING OF ENVIRONMENT
(2022)
Article
Environmental Sciences
Aishwarya Chandrasekaran, Guofan Shao, Songlin Fei, Zachary Miller, Joseph Hupy
Summary: This paper presents a methodology for automated measurement of tree height and crown area in broadleaf tree plantations using unmanned aerial systems imagery. The analysis shows higher accuracy in tree measurements with the datasets derived from multi-rotor platform than with the fixed wing platform. The automated method can efficiently calculate tree-level biometric estimations for a large number of trees based on UAS-SfM derived images.
Article
Geography, Physical
Guillaume Lassalle, Matheus Pinheiro Ferreira, Laura Elena Cue La Rosa, Rebecca Del'Papa Moreira Scafutto, Carlos Roberto de Souza Filho
Summary: This study presents the advances in mapping mangrove species using multispectral and hyperspectral imagery. A new framework based on convolutional neural network is proposed for accurate classification at pixel and object level. The study highlights the importance of spatial and spectral resolutions, especially the short-wave infrared bands.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Jay R. Malcolm, Braiden Brousseau, Trevor Jones, Sean C. Thomas
Summary: This study successfully modeled the tree species composition in over 51,000 forest stands in south-central Ontario, Canada by combining multispectral satellite information with aerial-photo interpreted forest resource inventories (FRI), showing significantly higher accuracy compared to traditional FRI models. By incorporating field data, the accuracy of the models was further improved, demonstrating the potential for modeling multivariate tree species compositions.
Article
Environmental Sciences
S. L. Barnsley, A. A. Lovett, L. V. Dicks
Summary: Wild pollinator numbers are positively associated with flower-rich habitat at landscape level, and having a baseline understanding of the temporal and spatial availability of resources can allow targeted habitat management. Very high-resolution remote sensing can accurately map fine-scale foraging resources, but challenges remain in classifying co-flowering species and quantifying floral unit density to calculate nectar sugar supply. The research provides a prototype approach for mapping pollinator foraging resources in an agricultural context and lays the foundation for developing a remote sensing pipeline for valuable data on nectar-rich flowering plant species availability throughout the year.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2022)
Article
Agriculture, Multidisciplinary
J. L. Pancorbo, M. Alonso-Ayuso, C. Camino, M. D. Raya-Sereno, P. J. Zarco-Tejada, I. Molina, J. L. Gabriel, M. Quemada
Summary: Early prediction of crop production by remote sensing can assist in planning the harvest and ensuring food security. This study aims to improve the quantification of yield, grain protein concentration, and nitrogen output in winter wheat using RS imagery. The results showed that the visible and short-wave infrared region had similar accuracy to hyperspectral and Sentinel-2 imagery in yield estimation. The SWIR bands were important for estimating grain protein concentration, and red-edge-based NDSIs improved the estimation of nitrogen output.
PRECISION AGRICULTURE
(2023)
Article
Forestry
Xavier Gallagher-Duval, Olivier R. van Lier, Richard A. Fournier
Summary: This study aimed to determine the optimal approach for estimating stem diameter distributions (SDD) from airborne laser scanning (ALS) data using different metrics. The results showed that the performance of SDD modality classification models was consistent and the best SDD function parameter models were generally fitted with a combination of metrics. It was found that CHM texture metrics can improve the estimate of SDD parameters and differentiating for modality prior to estimating SSD is especially beneficial in stands with bimodal SDD.
Article
Geochemistry & Geophysics
Fei Tong, Yun Zhang
Summary: This article proposes a spectral-spatial and cascaded multilayer random forests (SSCMRF) method for classifying tree species in high-spatial-resolution hyperspectral images. The method achieves superior classification results by fully utilizing spatial information from shape-adaptive superpixels and shape-fixed patches, integrating two different types of spatial information.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Multidisciplinary Sciences
Hans Ole Orka, Janis Gailis, Mathias Vege, Terje Gobakken, Kenneth Hauglund
Summary: Today's availability of large amounts of high-resolution satellite imagery necessitates effective preprocessing methods. One such method is mosaicking, which is required in many applications utilizing optical satellite imagery from the Landsat and Sentinel-2 archives. Maintaining the original data structure and preserving metadata is advantageous for further analysis, and creating mosaics that match the phenological state of natural phenomena can be beneficial for certain applications. This work presents Geomosaic, an open-source tool coded in Python that produces analysis-ready satellite data mosaics from Landsat and Sentinel-2 imagery.center dot The produced mosaics preserve pixel metadata for further analysis and can be used on multiple platforms.
Review
Biodiversity Conservation
Nelson Grima, Marie -Claude Jutras-Perreault, Terje Gobakken, Hans Ole orka, Harald Vacik
Summary: Ecosystem services make significant contributions to human well-being and economies, but measuring and quantifying them accurately is challenging. The use of indicators as proxies to estimate ecosystem services has been proposed as a solution. This study conducted a literature review and generated a list of 85 indicators for measuring ecosystem services, categorized according to the CICES (v5.1) classification system. The study also identified which of these indicators can be derived from remotely sensed data, with some being directly related, some indirectly related, and others currently not derivable.
ECOLOGICAL INDICATORS
(2023)
Article
Forestry
Ana de Lera Garrido, Terje Gobakken, Marius Hauglin, Erik Naesset, Ole Martin Bollandsas
Summary: The aim of this study was to analyze the accuracy of forest attribute predictions from the nationwide forest attribute map (SR16). Field observations from 33 forest inventory projects in Norway were used for validation. The overall results showed satisfactory accuracy, but there were large differences in accuracy among different inventory projects, with forest structure being the most influential factor.
SCANDINAVIAN JOURNAL OF FOREST RESEARCH
(2023)
Article
Forestry
Marie-Claude Jutras-Perreault, Erik Naesset, Terje Gobakken, Hans Ole Orka
Summary: This study utilized ALS data and vegetation indices from optical images to predict the presence of standing dead trees in a managed forest in Southern Norway. Area-based regression models were initially tested but proved to be statistically insignificant due to limited ground reference information. A tree-based approach, however, successfully identified standing dead trees based on ALS point cloud data and vegetation indices.
SCANDINAVIAN JOURNAL OF FOREST RESEARCH
(2023)
Article
Environmental Sciences
Marie-Claude Jutras-Perreault, Terje Gobakken, Erik Naesset, Hans Ole orka
Summary: This study proposes a tree-based approach that combines remote sensing data to predict the presence of standing dead trees (SDT) in forests. By comparing different remotely sensed data sources, it was found that NDVI calculated from aerial images accurately predicts the presence of SDT, while NDVI calculated from satellite images is less accurate.
Article
Environmental Sciences
Victor F. Strimbu, Erik Naesset, Hans Ole Orka, Jari Liski, Hans Petersson, Terje Gobakken
Summary: This study proposes an integrated methodology to estimate changes in forest carbon pools at the level of forest stands by combining field measurements and ALS data. The results demonstrate that ALS data can be used indirectly through a chain of models to estimate soil carbon changes at the primary level of forest management.
CARBON BALANCE AND MANAGEMENT
(2023)
Article
Remote Sensing
Endre Hansen, Julius Wold, Michele Dalponte, Terje Gobakken, Lennart Noordermeer, Hans Ole Orka
Summary: The study applied area-based approaches to predict rot occurrence, rot severity, and rot volume. Random Forest models were built and validated using remotely sensed data and ground reference data. The results showed that rot volume models performed better due to the correlation between timber volume and rot volume.
EUROPEAN JOURNAL OF REMOTE SENSING
(2023)
Article
Environmental Sciences
Ritwika Mukhopadhyay, Erik Naesset, Terje Gobakken, Ida Marielle Mienna, Jaime Candelas Bielza, Gunnar Austrheim, Henrik Jan Persson, Hans Ole orka, Bjorn-Eirik Roald, Ole Martin Bollandsas
Summary: Due to climate change, treelines are shifting to higher altitudes and latitudes. Accurately estimating the biomass of trees and shrubs in alpine areas is crucial for carbon reporting. This study utilized remotely sensed data, such as airborne laser scanning (ALS) and digital aerial photogrammetry (DAP), to estimate aboveground biomass (AGB) in a treeline ecotone in Southern Norway. Despite weak fit of the prediction models, the estimates showed adequate precision with relatively narrow confidence intervals (CIs). The results suggest that ALS and DAP data can be effectively used for AGB estimation in treeline ecotones.
Article
Forestry
Marie-Claude Jutras-Perreault, Terje Gobakken, Erik Naesset, Hans Ole Orka
Summary: Centuries of forest exploitation have caused significant loss of natural forests in Europe, leading to a decline in populations for many species. The Norwegian government has set a target of protecting 10% of forested area, yet less than 2% of Norway's forests are natural. A study using ALS data and NFI measurements aimed to predict the presence of natural forests in southeastern Norway, finding that semi-natural forests had the highest accuracy.
Article
Forestry
Victor F. Strimbu, Tron Eid, Terje Gobakken
Summary: This paper introduces a software tool called GAYA 2.0 for simulating forest development and performing scenario analysis to evaluate carbon sequestration potential, net present value, and climate change impacts under different management regimes. A case study in Norway is presented to demonstrate the tool's potential, showing changes in optimal management strategies and future predictions of forest carbon balance.
Article
Forestry
A. Q. Nyrud, K. M. A. Heltorp, Anders Roos, Francisco X. Aguilar, Katja Lahtinen, Noora Viholainen, Sami Berghall, Anne Toppinen, B. J. Thorsen, Matleena Kniivila, Antti Haapala, Elias Hurmekoski, T. Hujala, H. F. Hoen
Summary: This study investigated public attitudes towards multi-storey wood buildings in seven European countries and found that the level of knowledge about wood buildings was lower in countries where brick, stone, and concrete were commonly used in construction, such as the United Kingdom, Germany, and Denmark. Finland and Sweden had the most positive attitudes towards wood buildings. The study also identified factors such as fire vulnerability, material solidity, indoor environment, and moisture vulnerability that influenced people's perception of wood buildings as a nice place to live.
SCANDINAVIAN JOURNAL OF FOREST RESEARCH
(2024)
