Tree mortality submodels drive simulated long-term forest dynamics: assessing 15 models from the stand to global...

Area of Science:

• Ecology
• Forestry
• Climate Change Science

Background:

• Dynamic Vegetation Models (DVMs) are essential for predicting forest futures.
• Quantitative research on the impact of tree mortality submodels is limited.
• Assessing forest dynamics requires accurate representations of tree growth, mortality, and regeneration.

Purpose of the Study:

• To evaluate 15 dynamic vegetation models (DVMs) for their sensitivity to different tree mortality submodels.
• To analyze model performance under various climate change scenarios.
• To understand the uncertainty introduced by mortality submodels in forest projections.

Main Methods:

• Tested 15 DVMs (stand, landscape, continental scales) with at least two mortality submodels each.
• Validated model behavior against empirical time series data.
• Subjected models to different climate change scenarios to assess future forest dynamics.

Main Results:

• Most DVMs accurately simulated past data, regardless of the mortality submodel used.
• Different mortality submodels led to significantly divergent future forest trajectories under climate change.
• DVMs showed higher sensitivity to mortality submodels (10-170% deviation) than to climate change scenarios (2-3x lower).

Conclusions:

• Tree mortality is a primary source of uncertainty in predicting forest responses to climate change.
• Enhanced data and process understanding of tree mortality are vital for robust future forest simulations.
• Comparing alternative mortality formulations in DVMs reveals critical process uncertainties impacting simulated forest dynamics.