BBS: Developing methods to simulate biome boundary shifts

Tree species migration can considerably slow down biome boundary shifts. We develop methods to simulate migration reliably and efficiently with coarse resolution.

Frame

The project is part of the GLP (Global land project) fast track action (http://bbs2008.wikidot.com) “Decreasing uncertainty in predicting biome boundary shifts” which aims at improving the simulation of biome boundary shifts at large spatial scales, working group “Migration”.
The long-term goal is to improve existing vegetation models or to develop new models that are reliable and robust and can be included in Earth System models for studying biosphere-atmosphere feedbacks.  

Project members and partners

• Heike Lischke
  

Collaboration with Hisashi Sato (Yokohama, Jp), Takashi Kohyama (Sapporo, Jp), Paul Moorcroft (Harvard, USA), George Hurtt (New Hampshire, USA), Sean MacMahon (Durham, USA)

Rationale

Because of the nature of terrestrial plant population and community dynamics and dispersal, and the pace of climate change, predicting the future distribution of plant species is challenging. Many coupled GCM's assume simply that the boundaries between major terrestrial biomes are either static, or adjusted non-mechanistically to follow the change of climate without time lags.  In some DGVM's, a non-mechanistic treatment of biome boundaries is employed with assumed delays.
Recent model simulations with both explicit seed dispersal and population and community dynamics suggest that range shifts of forest biomes will be both complex and extremely delayed (several millennia delay for centennial warming).

Research topics

1. the effect of plant population processes and dispersal on migration
2. the effect of spatial heterogeneity (e.g. fragmentation or barriers) on dispersal and migration
3. methods to incorporate these effects into large scale models like such as DGVM's
4. the lags due to species migration and their effects on feedbacks to the earth system

Methods

Starting from the forest landscape model TreeMig which describes tree species migration by explicitly simulating seed dispersal on a grid of 1km wide cells, we develop numerical approaches to describe migration across heterogenous grid cells. These approaches are either aggregated models of within-cell migration speed, e.g. derived from meta-modelling, or simulating spread in a subset of smaller cells within each grid cell and then extrapolating to the larger cell.
We test our methods with simulations on south-north transects in Siberia and assess the effect of species migration on the feedbacks to the earth system.

Publications

Epstein, H.E.; Kaplan, J.O.; Lischke, H.; Yu, Q., 2007: Simulating Future Changes in Arctic and Subarctic Vegetation. Comp. Sci. Eng. July/August 2007: 13-23.
Lischke, H.; Zimmermann, N.E.; Bolliger, J.; Rickebusch, S.; Löffler, T.J., 2006: TreeMig: A forest-landscape model for simulating spatio-temporal patterns from stand to landscape scale. Ecol. Model. 199: 409-420.
Neilson, R.P.; Pitelka, L.F.; Solomon, A.M.; Nathan, R.; Midgley, G.F.; Fragoso, J.M.V.; Lischke, H.; Thompson, K., 2005: Forecasting Regional to Global Plant Migration in Response to Climate Change. BioScience 55, 9: 749-759.