Forest regeneration within Earth system models: current process representations and ways forward

Abstract

Earth system models must predict forest responses to global change in order to simulate future global climate, hydrology, and ecosystem dynamics. These models are increasingly adopting vegetation demographic approaches that explicitly represent tree growth, mortality, and recruitment, enabling advances in the projection of forest vulnerability and resilience, as well as evaluation with field data. To date, simulation of regeneration processes has received far less attention than simulation of processes that affect growth and mortality, in spite of their critical role maintaining forest structure, facilitating turnover in forest composition over space and time, enabling recovery from disturbance, and regulating climate-driven range shifts. Our critical review of regeneration process representations within current Earth system vegetation demographic models reveals the need to improve parameter values and algorithms for reproductive allocation, dispersal, seed survival and germination, environmental filtering in the seedling layer, and tree regeneration strategies adapted to wind, fire, and anthropogenic disturbance regimes. These improvements require synthesis of existing data, specific field datacollection protocols, and novel model algorithms compatible with global-scale simulations. Vegetation demographic models offer the opportunity to more fully integrate ecological understanding into Earth system prediction; regeneration processes need to be a critical part of the effort.

Simulating environmentally‐sensitive tree recruitment in vegetation demographic models

“…1). F repro already exists in VDMs and it is possible to use existing litter fall data to empirically constrain it in tropical forests (Hanbury‐Brown et al ., 2022). F seed has not been quantified at the ecosystem or PFT level, but is measurable from field observations (Wenk et al ., 2017), highlighting the need and opportunity to quantify it at VDM‐relevant scales.…”

“…Vegetation demographic models are ‘a special class of DGVM, which include representation/tracking of multiple size‐classes or individuals of the same PFT, which can encounter multiple light environments within a single climatic grid cell’ (Fisher et al ., 2018) (PFT, plant functional type). In contrast to the sophisticated algorithms VDMs use to predict growth and mortality, their representations of recruitment lack a sufficiently process‐based approach (McDowell et al ., 2020; Hanbury‐Brown et al ., 2022). Gap models, forest landscape models and mechanistic species distribution models have successfully represented key regeneration processes influencing recruitment such as seed production, dispersal, germination and seed decay in stand‐ and landscape‐scale simulations (Mladenoff, 2004; Larocque et al ., 2006; Lischke et al ., 2006; Lischke & Löffler, 2006; Scheller et al ., 2007; Holm et al ., 2012; Mok et al ., 2012), but their modeling approaches generally are less suitable for large‐scale ESM‐coupled simulations because they are computationally expensive and do not conserve carbon.…”

Nutrient Dynamics in a Coupled Terrestrial Biosphere and Land Model (ELM-FATES)

“…Unlike the other organs, the nutrient to carbon ratio of the reproductive tissues and storage (o = re, so) are not defined directly by parameter constants. FATES, like many vegetation demography models, does not mechanistically resolve germination or other processes of plants below a minimum recruitment size (Hanbury-Brown et al, 2022); instead it assumes that a fraction of carbon flux allocated to reproduction emerges as new recruits at some time later. We extend this approach to nutrients as well.…”

A machine learning approach targeting parameter estimation for plant functional type coexistence modeling using ELM-FATES (v2.0)

“…Reproductive features that enhance competitive exclusion tendencies have been illustrated to affect coexistence (Maréchaux and Chave, 2017;Fisher et al, 2018). Hanbury-Brown et al (2022) discussed the importance of the representation of forest regeneration, including improving parameters and algorithms for reproductive allocation, dispersal, seed survival and germination, environmental filtering in the seedling layer, and tree regeneration strategies adapted to wind, fire, and anthropogenic disturbance regimes. Besides, both growth-survival and stature-recruitment tradeoffs are critical to accurately predict successional patterns in tropical forest structure and competition (see details in Rüger et al, 2020), which should also be better considered in future model development.…”