Project

Back to overview

next-generation Modelling of the biosphere - Including New Data streams and optimality approaches

Applicant Stocker Benjamin
Number 181115
Funding scheme Eccellenza
Research institution
Departement Umweltsystemwissenschaften ETH Zürich ITES / D-USYS
Abt. Landschaftsdynamik und Raumentwicklung Eidg. Forschungsanstalt WSL
Institution of higher education ETH Zurich - ETHZ
Main discipline Other disciplines of Environmental Sciences
Start/End 01.09.2019 - 31.08.2024
Approved amount 1'901'793.00
Show all

All Disciplines (2)

Discipline
Other disciplines of Environmental Sciences
Agricultural and Forestry Sciences

Keywords (6)

Climate; Vegetation modelling; Carbon cycle; Allocation; Tree-rings; Nutrient limitation

Lay Summary (German)

Lead
Der Klimawandel ändert die Umweltbedingungen für Landökosysteme grundlegend. Steigende Temperaturen und zunehmende Hitze- und Trockenextremereignisse belasten die Vegetation bereits heute. Diese Änderungen spielen sich ab vor dem Hintergrund von steigendem atmosphärischen Kohlenstoffdioxid (CO2), was an sich die Photosyntheseraten stimuliert und die Wassernutzungseffizienz von Pflanzen erhöht. Das Zusammenspiel dieser verschiedenen Faktoren ist komplex und die Vorhersagefähigkeit von Modellen limitiert. Dies beeinträchtigt die Abschätzung von Klimarisiken für Forst- und Landwirtschaft und stellt eine der grössten Unsicherheiten in Erdsystemmodellen und Klimaprojektionen dar.
Lay summary
Inhalt und Ziel des Forschungsprojekts
Das übergeordnete Ziel des Projekts MIND ist es, neue Methoden zu entwickeln um bessere Vorhersagen zu treffen, wie terrestrische Ökosysteme auf CO2, Trockenstress, und Nährstofflimitierung reagieren und dadurch globale biogeochemische Kreisläufe beeinflussen. Dafür werden neue Modellierungsansätze entwickelt, die eine systematische Integration von diversen Beobachtungsdaten für Modellvorhersagen ermöglichen, und fundamentale ökologische Prinzipien für die Modellierung verwenden. Diese methodischen Innovationen werden mit einem Fokus auf eine Reihe von Schlüsselfragen angegangen: Wo liegen die Grenzen der Resistenz von Ökosystemen gegen Trockenstress? Führen das steigende CO2 und die beschleunigten Photosyntheseraten zu schnellerem Pflanzenwachstum und zu einem dichteren Baumbestand in unseren Wäldern? Inwiefern limitieren Nährstoffe im Boden die CO2-Fertilisierung des Vegetationswachstums und inwiefern können Pflanzen-Boden-Interaktionen diese Limitierung beeinflussen?
 
Wissenschaftlicher und gesellschaftlicher Kontext des Forschungsprojekts
MIND erarbeitet robuste und durch Beobachtungsdaten breit abgestützte Modelle. Diese werden für globale Simulationen der terrestrischen Biosphäre, sowie für regionale bis lokale Studien zur Anwendung kommen. Die Integration von Modellen und Messungen aus laufenden Datenherhebungsnetzwerken legt den Grundstein für ökologische Prognosen, welche politische Entscheidungsträger in der Planung für Klimaresilienz unterstützen.
Direct link to Lay Summary Last update: 05.08.2019

Responsible applicant and co-applicants

Employees

Abstract

Climate and CO2 impacts on vegetation and the terrestrial carbon (C) cycle are one of the largest uncertainties in climate change projections and pose challenges for planning in forestry and agriculture. The land biosphere is currently absorbing about a quarter of anthropogenic CO2 emissions, but the future of this ecosystem service remains unknown. At the core of this uncertainty are two major scientific questions: (1) To what degree do nutrients limit any positive effect of CO2 on growth and the land C sink? (2) Is the Earth continuing its current greening trend as temperatures increase and droughts become more prevalent? Dynamic Global Vegetation Models (DGVMs) embody our understanding of the behaviour of plants and ecosystems in response to climate and other environmental factors. The limits of this understanding are reflected by the widely diverging model predictions of the land C balance under future climate and CO2 change scenarios and the failure of current models to capture key ecosystem responses observed in manipulation experiments. The increasing realism of models has come at the cost of reduced robustness and reliability of their predictions. General organising principles of terrestrial ecology have been ignored and models lack the capacity to simulate acclimating vegetation responses to shifts in resource availabilities. A re-thinking of the theoretical fundamentals is needed and a systematic use of all available observational data streams is required. MIND will lay the basis for a next-generation global vegetation model by (1) developing a new approach for simulating acclimation and adaptive flexibility in growth based on the principle of optimal functioning under physical constraints and trade-offs, and (2) using previously unexploited large data streams in combination with model-data fusion techniques to unravel how tree growth responds to past and present environmental change. MIND will pave the way to data-driven ecological forecasting and develop a robust theoretical foundation for projections. This will be achieved by work along two major “streams”. The first further develops a new optimality-based vegetation model (SOFUN), which I have started to develop over the last two years. A first objective is to accurately predict how nutrients and other environmental conditions affect allocation of plant-assimilated C into different organs and how this affects C and nitrogen (N) cycling at the global scale and in the long-term. A second objective is to understand and model interactions between increasing CO2 and plant water stress, how this underlies the Earth’s current greening trend and plant control on water availability during dry periods. The second stream of research leverages a new open access biogeochemical vegetation model (LM3-PPA) that explicitly simulates the growth of individual trees and forest stand dynamics. This is a key innovation beyond current DGVMs that do not keep track of the growth of individuals and now enables the direct use of vast additional data resources from tree-rings and forest inventories for Bayesian model-data fusion. These data have been collected by several initiatives led by MIND’s host institutes and will be exploited here to scale from trees to ecosystems and to investigate how allocation and growth has been and is responding to environmental change. MIND will lead the way to a new theoretical understanding of resource limitation and its implementation in models - from a “rigid” concept of resource limitation on plant growth inspired by the century-old Liebig Law of the Minimum, now implemented in DGVMs, to a cost and trade-off based view where plants adjust and acclimate to optimally acquire and allocate resources.
-