Global net biome CO₂exchange predicted comparably well using parameter–environment relationships and plant functional types

Abstract

Despite their importance for understanding the role of terrestrial ecosystems in a changing climate, forecasts of net biome CO2 exchange are hindered by uncertainty in model parameters. Here, we compare the traditional plant functional type (PFT)-based parameterization approach to a novel top-down, machine learning-based "environmental filtering" (EF) approach.We find that the EF-based approach matches or outperforms the PFT-based approach at a narrow majority of vegetated pixels across the globe. KEYWORDSThis article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as

Are Plant Functional Types Fit for Purpose?

“…(2022) found PFTs to exhibit strong control on spatial variation of peatland fluxes, while a PFT approach for modeling global fluxes produced similar results to a more complicated ”environmental filtering” framework in a study by Famiglietti et al. (2023).…”

“…Conversely, recent studies have not necessarily produced results consistent with PFTs being fully obsolete (Thomas et al, 2019). Laine et al (2022) found PFTs to exhibit strong control on spatial variation of peatland fluxes, while a PFT approach for modeling global fluxes produced similar results to a more complicated "environmental filtering" framework in a study by Famiglietti et al (2023).…”

Ecosystem Water‐Saving Timescale Varies Spatially With Typical Drydown Length

“…The strong relationship between τ and D max (Figure 4) also motivates the parametrization of stomatal closure in models as a function of local climate, not just using plant functional types, as is commonly done. This approach is known as environmental filtering (e.g., Famiglietti et al., 2023; Walker et al., 2017). Our findings are consistent with previous findings of instantaneous stomatal parameters being related to aridity (Bassiouni et al., 2023; Lin et al., 2015; Y. Liu et al., 2022) and rainfall intensity and frequency (Bassiouni et al., 2023).…”

Impacts of climate timescale on the stability of trait–environment relationships

“…Madani et al, 2018) and inform trait-based modeling efforts across scales (e.g. Famiglietti et al, 2023). Fig.…”

“…While such relationships have myriad applications in trait-based ecology, including flexibly parameterizing large-scale ecological models (e.g. Verheijen et al, 2013Verheijen et al, , 2015Famiglietti et al, 2023), as well as extrapolating between sparse in situ data (Borgy et al, 2017), their potential utility is hampered by their relatively low predictability (Anderegg, 2023).…”