High-latitude Vegetation Change
![[Translate to English:] Drone-based acquisitions for forest structure](/fileadmin/user_upload/AWI/Forschung/Geowissenschaft/Polare_Terrestrische_Umweltsysteme/Land_Cover/Vegetation_and_Forest_inventory_-_IMG_9485_Drohnenaufnahmen_zur_Waldstrukturuntersuchung_von_offenen_L%C3%A4rchenw%C3%A4ldern_Chukotka2018_Stefan_Kruse.JPG "[Translate to English:] Drone-based acquisitions for forest structure")
The climate in the high latitudes is changing rapidly, profoundly affecting vegetation and its geographic distribution. Our research focuses on the taiga–tundra ecotone in North America and Eurasia, the sensitive transition zone at the Arctic tree line. We study the Arctic and mountain treelines, tracing the gradient from treeless tundra through scattered trees and open woodlands to the dense forests of the southern taiga.
Our goal is to understand the ecosystem functions and services of the taiga–tundra ecotone, in the past, present, and future. We investigate processes ranging from the individual tree to the community level, from the protection of permafrost to the role in the global carbon cycle.
Our methods
![[Translate to English:] Probennahme von Lärchennadeln](/fileadmin/_processed_/1/7/csm_labor_genetics_3b6b26258a.png "[Translate to English:] Probennahme von Lärchennadeln")
![[Translate to English:] Dendrometer Logger Auslesung](/fileadmin/_processed_/2/f/csm_dendrometer_f3cf0c855b.png "[Translate to English:] Dendrometer Logger Auslesung")
Head
Kunyan Hao (scientific assistant)
Antonia Gfrörer (BSc., student assistant)
Elisabeth Riegel (MSc., student assistant)
Amely Wernitz (BSc., student assistant)
Nelly Zens (MSc., student assistant)
Prof. Dr. Elisabeth Dietze
Dr. Alison Beamish
Dr. Xianyong Cao
Dr. Léa Enguehard
Dr. Rongwei Geng
Dr. Ramesh Glückler
Dr. Sarah Haupt
Dr. Simone Stuenzi
Dr. Ximena Tabares
Dr. Fang Tian
Dr. Iuliia Shevtsova
Jakob Broers (scientifc assistant)
Timon Miesner (scientific assistant)
Femke van Geffen
Josias Gloy
Dr. Sarah Haupt
Prof. Dr. Luidmila Pestryakova
Prof. Dr. Xingqi Liu
Prof. Dr. Jian Ni
Dr. Natalya Rudaya
Dr. Kai Li
Dr. Yury Dvornikov
Dr. Evgenii Zakharov
Projects, Cooperations
ForestUNLOCK integrates terrestrial, airborne, and satellite‑borne sensor data to create a multi‑modal, multi‑scale benchmark dataset for AI‑driven monitoring of boreal forests and carbon accounting. The project aims to develop algorithms that automatically derive forest density, wood volume, and carbon stocks from remote sensing data, improving the representation of greenhouse gas balances in boreal ecosystems.
Contact: Dr. Stefan Kruse
Partner:
- Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI)
- German Aerospace Center (DLR)
- Helmholtz Centre for Geosciences (GFZ)
Duration: 2026
Funding: Funded within the UNLOCK benchmarking projects of the Helmholtz Association, from the Helmholtz Initiative and Networking Fund (INF) as well as with in‑house funds of the participating Helmholtz centres.
Further information: https://helmholtz-imaging.de/project/forestunlock/
POINTR develops AI‑driven methods for the automated, tree‑individual recording of forest structures from high‑resolution remote sensing data (e.g., drone‑borne LiDAR and aerial imagery). The project aims to create precise 3D models of trees and stands to analyse growth, stress conditions, and carbon storage at the individual‑tree level, thus improving the understanding of forest responses to climate change and disturbances.
Contact: Dr. Stefan Kruse
Partner:
- Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI)
- German Aerospace Center (DLR)
- Helmholtz Centre for Geosciences (GFZ)
Duration: 2025-2028
Funding: Funded as a Helmholtz Imaging project from the Helmholtz Initiative and Networking Fund (INF) as well as with in‑house funds of the participating Helmholtz centres.
Further information: https://helmholtz-imaging.de/project/pointr/
FAIR AIMS improves the accessibility and quality of sample data by developing an automated system for the registration and management of International Geo‑Sample Numbers (IGSNs). The project creates an IGSN online template generator, domain‑specific metadata profiles for Earth and Environment samples, and an integrated connection to the HZB sample database to generate standardized, machine‑readable sample metadata and support Open Science practices.
Contact: Dr. Birgit Heim
Duration: 2024–2026 (HMC‑project phase)
Funding: Funded within the HMC project cohort (Helmholtz Metadata Collaboration) by the Helmholtz Association.
Further information: https://helmholtz-metadaten.de/de/inf-projects/fairaims
FAIR WISH develops standardized, domain‑specific IGSN metadata schemas and automated workflows to simplify the registration and management of International Generic Sample Numbers (IGSNs) for physical samples in the Helmholtz Association. The project covers a wide range of sample types in Earth and Environment research (e.g., rock, soil, vegetation, water, and sediment samples) and supports all digitization levels – from individual researchers’ field‑book notes to fully digital sample‑management systems in databases.
Contact: Dr. BIrgit Heim
Partners: GFZ, AWI, Hereon
Duration: 2020–2024 (HMC‑project cohort)
Funding: Funded within the Helmholtz Metadata Collaboration (HMC) by the Helmholtz Association.
Further information: https://helmholtz-metadaten.de/de/inf-projects/fair-wish
This project investigates how topography controls aboveground carbon allocation and stocks in northern boreal forests along the Tundra–Taiga Ecotone (TTE) and whether these relationships remain stable over several decades. Using generalized additive models, it analyses links between aboveground biomass (AGB) and terrain features and combines long‑term field data with remote‑sensing‑based reconstructions to test the stability of topographic effects. The individual‑based vegetation model LAVESI is extended to predict future AGB trajectories in the TTE under climate change.
Contact: Dr. Stefan Kruse
Duration: 2026-2028
Funding: Funded by the German Research Foundation (DFG) within the individual research grants programme.
Further information: https://gepris.dfg.de/gepris/projekt/563037533?language=en
This project investigates how the elevational treeline in the Ural Mountains shifts in response to climate change and which factors – such as growing conditions, micro‑topography, snow dynamics, and disturbance regimes – limit the advance of the spruce treeline. Using field surveys, tree‑ring data, genomic analyses, and modelling, it examines the contributions of genetic adaptation and ecological constraints to treeline dynamics, in order to better assess the responsiveness of boreal forests in the Ural under future climate scenarios.
Contact: Dr. Stefan Kruse
Duration: 2021–2025
Funding: Funded by the German Research Foundation (DFG) within the individual research grants programme.
Further information: https://gepris.dfg.de/gepris/projekt/448651799?language=en
This project investigates how plant diversity and vegetation composition on the southeastern Tibetan Plateau changed during the late Pleistocene and Holocene warming and which climatic and ecological factors controlled the structure of plant communities. Using sedimentary ancient DNA (sedaDNA), pollen, and other proxy data, it analyses the responses of high‑mountain and alpine plant assemblages to cold and warm phases, snowmelt, and cryosphere loss, in order to better assess the vulnerability of this unique biodiversity to future warming.
Contact: Dr. Stefan Kruse
Duration: 2019–2024 (or as listed in GEPRIS)
Funding: Funded by the German Research Foundation (DFG) within an individual research grant (project number 410561986).
Further information: https://gepris.dfg.de/gepris/projekt/410561986
AI-vergreens developped a wide range of FAIR DATA publications for specific northern boreal forests of Siberia, Alaska, and Canada, supporting downstream tasks of satellite data based classifications for summergreen needleleaf and evergreen needleleaf forests.. A highly valuable component of AI-vergreen is the provision of labeled satellite training and validation datasets, as well as extensively tested benchmark datasets. These are based on UAV forest structure data and derived fromentinel-2, Sentinel-1, and Harmonized Landsat–Sentinel-2 (HLS) satellite data. Machine learning enables the downstream task of deriving boreal forest structure and boreal forest types for the specific “Northern Forest Edge” boreal forest type.
Kontakt: Dr. Birgit Heim
Partner:
- Alfred‑Wegener‑Institut, Helmholtz‑Zentrum für Polar‑ und Meeresforschung (AWI)
- Deutsches Zentrum für Luft‑ und Raumfahrt (DLR)
- Technische Universität Berlin (TUB)
Duration: 2022-20285
Funding: Funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWE)