Modelling the influence of environmental conditions on gross primary production and land-atmosphere carbon exchange
File(s)
Author(s)
Mengoli, Giulia
Type
Thesis
Abstract
Vegetation regulates the land-atmosphere exchanges of water, energy and carbon. Land surface models (LSMs), used to represent these exchanges in climate models, face continuing challenges. Most represent only the (experimentally measured) fast responses of photosynthesis, respiration and transpiration, and neglect longer-term (acclimation and adaptation) processes by which plants are physiologically adjusted to their growth environment, including its seasonal variations and geographic patterns. The proposition of this thesis is that model simulations can be improved when plant processes are represented via eco-evolutionary optimality (EEO) principles. An existing EEO-based light-use efficiency model for gross primary production (GPP), the P model, is further developed here to simulate GPP at the sub-daily timescale required for implementation in LSMs. The resulting model has several advantages. By allowing photosynthetic capacities to vary systematically with environmental conditions, it avoids the need to prescribe different (fixed) parameter values for different vegetation types. It is shown to reproduce diurnal cycle of GPP recorded by eddy-covariance flux towers, across five different well-watered biomes, well – and better than a current, operational LSM. However, soil moisture limitations on GPP degrade the model’s performance in dry climates. A semi-empirical soil-moisture stress function is derived to correct this. The function depends on aridity, with plants in arid climates using water more conservatively while maintaining the ability to extract water under drier conditions than plants from wet climates. The resulting model simulates observed seasonal cycles of GPP well across the full range of climatic aridity. Run at global scale, the model is shown to be able to reproduce major features of the spatial and temporal variations of net ecosystem exchange, NEE (inferred from satellite-derived column concentrations of CO2) and of the differences in NEE patterns between El Niño and La Niña years.
Version
Open Access
Date Issued
2024-02
Date Awarded
2024-11
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Prentice, Iain
Harrison, Sandy
Publisher Department
Department of Life Sciences
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)