Challenging terrestrial biosphere models with data from the long-term multifactor Prairie Heating and CO₂ Enrichment...

Area of Science:

• Ecology
• Climate Change Science
• Computational Modeling

Background:

• Terrestrial biosphere models (TBMs) are crucial for predicting ecosystem responses to global change.
• Previous TBM evaluations have primarily used single-factor experiments, limiting their assessment under complex environmental conditions.
• Multifactor experiments are advocated for advancing TBM development, but their utility in constraining models remains underexplored.

Purpose of the Study:

• To evaluate the performance of ten TBMs against data from the multifactorial Prairie Heating and CO2 Enrichment (PHACE) experiment.
• To investigate how multifactor experiments can constrain TBMs and identify pathways for model improvement.
• To assess TBMs' ability to simulate grassland responses to combined CO2 enrichment and warming under varying water availability.

Main Methods:

• Applied ten different TBMs to simulate ecosystem processes at the PHACE experimental site.
• Compared model outputs for above-ground net primary productivity (ANPP) against observational data under ambient and single-factor treatments.
• Analyzed model performance in capturing phenological shifts, carbon allocation, and the influence of water stress on plant responses.

Main Results:

• TBMs exhibited poor performance under ambient conditions, with a wide range in simulated ANPP, indicating significant model uncertainty.
• Models failed to accurately represent carbon allocation, phenology, and the impact of water stress, even under single-factor treatments.
• Models inaccurately simulated phenological responses to warming and CO2, and failed to capture subtle interactive effects of multifactor treatments.

Conclusions:

• Multifactor experiments like PHACE are essential for revealing TBM limitations in simulating complex ecosystem dynamics.
• Significant improvements are needed in TBMs' representation of carbon allocation, phenology, and water stress interactions.
• Future experimental data should be used to refine TBMs for more accurate predictions of grassland responses to global change.