Developing remote sensing to map forest stress using data from multiple sensors and platforms 2024-2026

Ongoing climate change, with warmer, drier summers, favors spruce bark beetles, reducing spruces’ defense against attacks. Early detection of stressed forests is crucial for strategic control. In this project, we will combine drone and satellite data for improved detection. This innovative approach aims to quantify forest stress and vulnerability from individual trees to landscapes. By protecting forests against biotic and abiotic damage, we support climate-resilient and sustainable forestry in Sweden and Europe.

Role: Co-applicant.
Applicants: Langning Huo, Henrik Persson, Eva Lindberg.

New volume functions based on harvester measurements 2024-2026

Today, national volume functions in Sweden, developed by Göran Brandel in 1990, allow calculating the tree volume based on diameter, height, and species. These functions, derived from an empirical sample, show systematic errors when applied to smaller areas than intended, depending on location. We aim to create new volume functions for Sweden, factoring in location (coordinates and altitude). Using new volume estimations from harvesters across Sweden, we can now develop such functions. Accurate volume calculations are crucial for timber flow planning and economic outcomes. While large companies can use laser scanning and extensive harvesting data for forecasts, smaller forest owners face unequal conditions. These new functions could also enhance national estimates of forest carbon storage, impacting climate reporting.

Role: Main applicant.
Applicants: Henrik Persson, Björn Hannrup, Liviu Ene, Maria Nordström.

Use of satellite data to characterize spatial and temporal variation in grasslands managed by grazing or mowing 2024-2024

In this project, we aim to investigate the possibilities of improving the knowledge regarding the significance of different soil conditions and dynamics over time in species-rich meadows and pastures, both for production and biodiversity. By utilizing readily available remote sensing data, we can create the opportunity to scale up information from environmental monitoring samples to comprehensive landscape data that can be used for planning and policy purposes.

Role: Main applicant.
Applicants: Henrik Persson, Per Toräng, Anders Glimskär.

DikesPlan: a decision support for management of forest ditches 2023-2025

The project led by Skogforsk will develop decision support of forest ditches for the operational forest planning. It will cover a resource analysis, digital mapping, a decision support system prototype, and validation of the performed decisions by using field inventoried reference data.

Role: Main applicant.
Applicants: Henrik Persson, Maria Nordström, Carolina Offenbacher.

Skogforsklabbet – a forest data lab for the industry 2023-2025

The data lab will provide a safe data storage and data sharing platform for forest related data, with the ambitions to meet the security needs of the industry. The lab is managed by Skogforsk.

Role: Main applicant.
Applicants: Henrik Persson, Maria Nordström, Carolina Offenbacher.

Using radar measurements to analyze water exchange in forests – a PhD project 2023-2026

The project will investigate how radar measurements from the ground, from a tower (www.borealscat.se), and from satellites can be used to understand the water exchange in the forest. The exchange happens day and night throughout the vegetation season, while measurements during the frozen winter conditions are more affected by external weather driven factors, e.g., precipitation as rain and snow, wind and low temperatures. The project is co-located with two large international infrastructures (ICOS and ACTRIS) that are used for studying the carbon cycle and changes in the atmosphere. The co-location enables excellent conditions for studies of evapotranspiration in the context of global warming and changing forest conditions. The knowledge derived through this project will support improved estimates of the forest at large scale through wall-to-wall maps.

Role: Main applicant.
Applicants: Henrik Persson, Langning Huo, Ritwika Mukhopadhay, Hjalmar Laudon, Lars Ulander, Albert Monteith.

Developing hyperspectral drone imagery for improved monitoring of forest insect damages 2023-2025

Granbarkborrar är den mest förödande skadegöraren i Europas skogar och orsakar årligen skador för 7 miljarder kronor. Genom att avverka och transportera ut angripna träd tidigt (medan borrarna fortfarande befinner sig i träden) kan man minska skadorna. Det är därför avgörande att tidigt identifiera angripna träd. Avdelningen för skoglig fjärranalys vid SLU forskar nu på hur barkborreattacker kan identifieras med drönare och vi fann att med multispektrala kameror kunde dubbelt så många attackerade träd identifieras, jämfört med fältbesök. Vi vill i detta projekt förbättra metoden ytterligare genom att nyttja bättre sensorer med fler våglängder. En hyperspektral kamera mäter ett kontinuerligt spektrum som inte begränsas till förbestämda band. Det gör att vi dels kan bedriva grundforskning på trädens spektrala signatur i olika förhållanden och specifikt utveckla metoden för tidig detektion av granbarkborreattacker.

Role: Co-applicant.
Applicants: Langning Huo, Henrik Persson, Eva Lindberg, Jonas Bohlin.

RETURN: Restoring EnvironmenTs in UkRaiNe 2023-2024

Capacity building and knowledge exchange with respect to environmental assessment and restoration due to the war in 2022. We will work together to identify areas in need of restoration, discuss and then conduct a preliminary environmental assessment there, and then together develop tailored methodologies based on our collective experience for how these sites can be restored. If possible, the Swedish and Ukrainian specialists will also travel to make field visits and work on the methodologies.

Role: Co-applicant.
Applicants: Brian Kuns, Nataliia Kulatska, Jens Olsson, Anders Glimskär, Richard johnson, Lenka Kuglerova, Henrik Persson, Eddie von Wachenfeldt, Marine Elbakidze, Per Toräng, Thomas Keller, SLU.
Victor Karamushka, Olena Maslyukivska, National University of Kyiv Mohyla Academy.

The ForestWard Observatory to Secure Resilience of European Forests (FORWARDS) 2023-2027

FORWARDS will prototype The ForestWard Observatory to provide (a) timely and detailed information on European forests’ vulnerability to climate change impacts, (b) science-based knowledge to guide management using the principles of climate-smart forestry, ecosystem restoration, and biodiversity preservation (CSF & Restoration), and (c) stakeholder engagement and public participation in decision-making processes. We capitalize on data from existing networks (e.g. ICP Forests) and expand this with a Network of Pilot sites through 5 FORWARDS Demo cases plus ~50 trials established via grants to third parties. We will reconcile the current divide between forest information obtained from the ground and remote sensing by incorporating the concept of Monitoring Supersites and novel approaches to more comprehensively characterize cause-effect relationships of forest disturbances. Tools for European-wide forward-looking and spatially explicit projections on forests as well as regionalized CSF & Restoration trajectories will be developed jointly with stakeholders to evaluate synergies and trade-offs of conversion and restoration activities. These will be used to provide good practice guidance on effective CSF & Restoration management practices. The ForestWard Observatory will be constructed under the principle of co-design to address the information needs by users and stakeholders. FORWARDS interacts with several established networks on CSF & Restoration and effectively utilizes five dedicated grant calls to implement forest observations and test CSF & restoration measures. The ambition is for The ForestWard Observatory to become a long-lasting legacy of FORWARDS, which supports decision making across scales to boost the uptake of good CSF & Restoration management practice throughout Europe (local scale for management practice), while efficiently informing about climate change and disturbance impacts and resilience of European forests (regional to EU scale for policy making).

Role: Co-applicant, (PI Ruben Valbuena, SLU).
Applicants:
Swedish University of Agricultural Sciences – SLU, (Sweden), European Forest Institute (Finland), Universita Degli Studi Di Firenye (Italy), Universita Degli Studi Del Molise (Italy), WSL (Switzerland), Johann Heinrich von Thünen Institute (Germany), GFZ German Research Centre For Geosciences (Germany), Wageningen University (Netherlands), Research Institute of Wageningen (Netherlands), Forestry Commision Research Agency (United Kingdom), Latvian State Forest Research Institute – Silava, (Latvia), Fondazione ICONS (Italy), Bangor University (United Kingdom), Natural Resources Institute -LUKE, (Finland), Universität für Bodenkultur Wien – BOKU (Austria), Environmental Social Science Research Group – ESSRG, (Hungary), Lunds University (Sweden), Prospex Institute (Belgium), Research Institute for Nature and Forest (Belgium).

RESDINET (Network for novel remote sensing technologies in forest disturbance ecology) 2023-2025

RESDINET_logo

The project enhances networking activities between the Institute of Forest Ecology, Slovak Academy of Sciences (IFE SAS) and the top-class counterparts Finnish Geospatial Research Institute, The University of Eastern Finland and Swedish University of Agricultural Sciences. The project will enhance IFE SAS staff management capacities, administrative skills and scientific capabilities in the use of novel remote sensing technologies (RST) in forest disturbance ecology (FDE). Rigorous analyses of severe insect-induced disturbances using novel RST will be carried out in test areas representing different forest and climate types: mountain forests in Slovakia and boreal forests in Finland and Sweden. We will integrate in situ UAV and drone acquired remotely sensed data, existing multitemporal geospatial information and field data, particularly data on bark beetle population density, visible infestation symptoms linked to outbreak phases, and tree physiology parameters measured using
electronic dendrometers or sapflow meters. The combined dataset will be used to develop new tools for landscale-level early bark beetle attack identification and for designing bark beetle infestation risk assessment model. We will draw on the latest advances in drone technologies and image analytical tools, including deep Convolutional Neural Networks based machine learning techniques and Artificial Intelligence algorithms. We expect to obtain important scientific results and contribute new knowledge to this scientific field.

Role: Co-applicant
Applicants: Rastislav Jakus, Miroslav Blazenec et al. (IFE SAS), Langning Huo, Henrik Persson, Eva Lindberg et al. (SLU), Eija Honkavaara et al. (FGI), Päivi Lyytikäinen-Saarenmaa et al. (UEF).

Mapping of biodiversity indicators in forests using remote sensing 2022

This project will derive maps of biodiversity indicators in forests from remotely sensed data and data from the National Forest Inventory. The long-term goal is to derive biodiversity indicators for the whole of Sweden. This will contribute to the environmental quality objectives Sustainable Forests and A Rich Diversity of Plant and Animal Life.

Role: Co-applicant
Applicants: Eva Lindberg, Henrik Persson, Langning Huo, Jörgen Wallerman, Mats Nilsson

Planning of a pilot project for remote sensing based forest assortment estimates – a simulation study 2022-2023

The project will investigate the impacting conditions for carrying out regular airborne laser scanning forest inventories to generate volume estimates for a forest company’s timber assortment, using remote sensing. Simulations are used to address cost parameters, accuracy and geographical distribution in relation to clustering of the forest.

Role: Main applicant
Applicants: Henrik Persson, Johan Holmgren, Patrik Ulvdal, Göran Ståhl

A prolonged observation series of bark beetle infestations from the field and remote sensing 2022-2023

The project will continue the spruce bark beetle research at the test site Remningstorp. The specific objectives include spreading patterns and early detection using drone and satellite images.

Role: Co-applicant
Applicants: Langning Huo, Eva Lindberg, Jonas Bohlin, Henrik Persson

Multi-phase inventory of forest using high-resolution laser scanning 2022-2023

This project investigates how the use of high-resolution laser scanning of forest acquired from airplanes and helicopters can be used to estimate tree properties for large areas. The methods are applied to the single-tree level and include the variables diameter breast height, stem volume, tree species and Lorey’s height.

Role: Main applicant
Applicants: Henrik Persson, Johan Holmgren, Kenneth Olofsson

Establishing a forest damage database for advancing the use of remote sensing in monitoring damages 2021-2022

This project investigates how a reference database of forest damages can be established to improve remote sensing based methods for monitoring forest damages. The project is in collaboration with the Swedish Forest Agency.

Role: Main applicant
Applicants: Henrik Persson, Eva Lindberg

Improved detection and prediction of spruce bark beetle infestations 2022-2023

The objective of this project is to develop methods for bark beetle attack detection and prediction, using drones and satellite images. The temporal, spectral, and spatial characteristics of the infestations will be detailed studied from the individual tree scale to the canopy scale (groups of trees). Refined methods for the detection and prediction using high-resolution images from different sensors will be proposed. Damage maps and forecasting maps will be generated for southern Sweden and results will be validated using harvester data.

Role: co-applicant
Applicants: Langning Huo, Eva Lindberg, Henrik Persson

Mapping of state and changes with time series of satellite radar data for future forestry planning 2021-2022

The project aims to investigate time series of X-, C- and L-band synthetic aperture radar (SAR) data, separately and in combination, from the satellites TanDEM-X, Sentinel-1 and ALOS/-2, respectively, for tree height, diameter, basal area, stem volume and above-ground biomass estimation, and their change over time with special focus on diameter distribution. The time series will be used to map forest changes and they will be the base for proposing appropriate silviculture treatments. The change studies will also include detection and mapping of clear-cuts, thinning and growth.

Role: co-applicant
Applicants: Johan Fransson, Lars Ulander, Henrik Persson, Ivan Huuva.

Satellites and drones for future remote sensing research 2021-2022

The project will operate at the test site Remningstorp and is divided in two parts. 1. An experiment with controlled bark beetle infestations and collect related reference data for remote sensing research. The aim is to delineate infested single trees. 2. Inventory of forest stands that were recently managed, in order to enable time series analyses of forest changes using ALOS-2 PALSAR-2. The hypothesis that the ratio of the polarization channels HV/VV can be used to correct disturbing weather affects during the acquisition. This has been verified for an airborne setting and this project will verify this also for space borne images.

Role: main applicant and project leader
Applicants: Henrik Persson, Johan Fransson, Langning Huo, Eva Lindberg.

Development of methods for satellite based research and monitoring of carbon and water cycle in boreal forest ecosystems 2021-2024

The project will use a multi-band and multi-polarization radar mounted on a 50 m high tower combined with extensive measurements of the surrounding forest, air, water and carbon flux. It is colocated with ICOS and SITES measurement at the test site Krycklan, located in northern Sweden..

Role: co-applicant
Applicants: Johan Fransson, Mats Nilsson, Hjalmar Laudon, Tomas Lundmark, Matthias Peichl, Henrik Persson, Lars Ulander.

Measuring forest vitality with radar remote sensing 2021-2023

In the past few years, Sweden’s forests have experienced unprecedented levels of degradation due to ecological stress, fires and insect damage. These effects are a consequence of unusually hot summers due to our changing climate. New techniques for monitoring forest health are necessary to proactively manage forests and to sustain the environmental and economic value of Sweden’s forests. Recent advances in radar remote sensing of forests have shown that information about the biological activity of forests can be observed using radars. With world leading radar and forest research infrastructures in Sweden, methods for monitoring forest vitality using radar data can be developed. In the proposed project, high-quality, co-located radar, ecological, atmospheric and hydrological data will be collected. The data will be used to gain a quantitative, systems-based understanding of how ecological processes, forest disturbances and radar observations are coupled at timescales of minutes to years. The work in
this project will lead to more efficient and cost-saving forest management practices based on radar remote sensing.

Role: project participant
Applicants: Lars Ulander, Johan Fransson, Mats Nilsson, Henrik Persson.

Forest inventory using airborne laser scanning from low elevation 2021-2022

The project will develop and evaluate a new method for forest inventory using very dense laser scanning data acquired from low elevation. Current forest inventories typically use a design-based field sample of their inventory, which represents their entire forest. The field inventory normally consists of manual measurements on random or systematically distributed circular plots. Most commonly, diameter, height and tree species are registered. For many of the Swedish companies, this implies manual visits of more than 16 000 field plots. This project will evaluate a strip sampling approach, where a helicopter carrying a laser scanner will fly across the stands at low elevation. The hypothesis is that this may provide more or similar information to a lower price, compared to manual field visits. The project will develop algorithms for detection of single trees, classify the tree species, and estimate forest height, diameter and stem volume.

Role: project leader

Improved understanding of the influence of reference data accuracy when used in remote sensing 2018-2020

The use and quality of remote sensing for forest monitoring improves all the time and currently, remote sensing may sometimes provide as accurate predictions as those obtained from field inventories. Yet, field data are still used as reference when new remote sensing methods are evaluated. This project will develop a framework for quantifying various error sources related to the field inventory, and correct for these when field data are used in remote sensing based predictions of forest. The dominant error sources will be identified and we will quantify their influence. Then, methods for correcting for these errors will be proposed, in order to objectively evaluate various remote sensing estimates, regardless of sensor or study area.

Role: project leader

New forest tree species maps from time series of satellite images

The aim of this project is to develop and evaluate a methodology to derive tree species maps from freely available Sentinel-1 (S1) and S2 satellite data using dense multispectral airborne laser scanning (ALS) as training data in a middle step and tree species information from field sample plots as the final training data. The power of this approach is the individual tree analysis of ALS data. Only a small field dataset is needed, but since tree species is identified for individual trees detected in ALS data, a large number of tree species combinations can be automatically estimated for satellite pixels within transects and then used for training of wall-to-wall satellite data

Role: co-applicant

Tree species classification from remote sensing – next generation forest maps for ecology and management

Sustainable forest management requires accurate geospatial information about forest state. Estimates of size related forest variables such as stem volume and tree height can be derived with high accuracy from Airborne
Laser Scanning (ALS) data and surface models from matching of aerial images. However, tree species has not yet been estimated with sufficient accuracy for user needs. Thus, there is a need for a system that can map tree species at nationwide scale and at regular time intervals.
New technologies offer the potential for this. The new Sentinel-2 satellites provide multi-spectral data with short intervals, enabling utilization of phenological differences between tree species. Very High Resolution (VHR) images from satellite and aerial images make analysis of tree crowns structure possible. Multispectral ALS data provide both geometric and spectral information, and single-photon-counting (SPC ) laser data enable dense 3D points clouds over large areas with the possibility to analyze the 3D structure of tree crowns. SAR-data are available for large areas and have the potential to contribute to tree species classification.
This project will develop methods to derive tree species information from remotely sensed (RS) data. The intention is that the methods should be feasible for nationwide use. The final methodology will combine RS data that are identified as most suitable for tree species maps. The design of the maps will be determined with the help of a user group.

Role: co-applicant

Detection of spruce bark beetle attacks using remote sensing

This project will evaluate various methods for using remote sensing sensors to detect spruce bark beetle attacks at the test site Remningstorp. The focus is development of methods based on satellite images, especially Sentinel-1 and Sentinel-2, to more efficiently monitor and react to related damages.

Role: co-applicant

New Space Digital Economy Innovation Center (KvarkenSpaceEco)

The “Kvarken Space Eco” project will implement a long-lasting regional economic development structure for space-based business and innovation, i.e. a Kvarken Space Center. The planned center’s primary objective is to support regional businesses to develop opportunities within the “new space economy” and commercialise existing space-based data. Factors have developed in the space industry that allow much less expensive processes to be implemented in space relative to the past. To aid the region in taking a stake in these new economic activities as well as available data and services, the center will assure that an understanding of the latest technologies is available, along with the capacity to implement them. The center will share knowledge and implement demonstration projects to bring the regional businesses to the level needed to independently manage their own space business activities. The center will work closely with regional education systems and development companies and will promote the necessary themes to build and sustain a workforce capable of advancing regional participation and value-creation in new space.

Role: co-applicant

Can spruce bark beetle attacks be detected in radar tower time series?

This project will investigate if spruce bark beetle attacks can be detected with radar. This project is possible due to time series of C- L- and P-band radar measurements acquired from a tower within the project BorealScat http://www.borealscat.se/ Within the time frame the project has been active (since 2016), the observed forest has also been infested by bark beetles, which has created an interesting setup to monitor changes in the radar signal during the entire process before, during and past attacks.

Role: co-applicant

Mapping above-ground biomass in natural forest and shrub lands in central Chile

The project goal is mapping and monitoring carbon stocks in terms of above-ground biomass in natural forest and shrub lands in central Chile. Partners: Pontificia Universidad Católica de Chile.

Role: collaborator

BorealScat

This project is enables continuous monitoring of boreal forest series using C- L- and P-band radar measurements acquired from a 50 m high tower. http://www.borealscat.se/ This project enables a separation of reasons causing changes in the observed radar signal, especially due to precipitation, snow, wind and temperature. The project is lead by Chalmers Technical University, with Prof. Lars Ulander and Albert Monteith. SLU is a collaborator responsible for ground surveys and forest inventories.

Role: collaborator

Some finished projects

(Since the publication procedure in the field of forest remote sensing may be months or years, some finished projects may still generate new publications, due to the write and review process continuing beyond the project end. The following list of projects is not complete.)

  • Technical Assistance in the Implementation of a C-band Convoy Mission Demonstration Campaign. Financed by ESA. Partners: NIBIO (Norway), SLU, Chalmers (Sweden).
  • GlobBiomass, http://globbiomass.org/ Mapping global biomass and biomass change. 2015-2017.
    The main purpose is to better characterize and to reduce uncertainties of AGB estimates by developing an innovative synergistic mapping approach in five regional sites for the epochs 2005, 2010 and 2015 and for one global map for the year 2010.
  • Forest biomass and biomass change with spaceborne SAR. 2015-2018. Partners: SLU, Chalmers, Gamma RS.
  • Retrieval of forest biomass and biomass change with spaceborne SAR. 2013-2016. Partners: SLU, Chalmers, Gamma RS.
  • AdvancedSAR – Advanced Techniques for Forest Biomass and Biomass Change Mapping Using Novel Combination of Active Remote Sensing Sensors. 2013-2017.