A new modeling approach to evaluate the effects of climate change on plant functional diversity in the Amazon rainforest

Climate change is impacting all regions of the world, particularly tropical ecosystems like the Amazon rainforest. The Amazon rainforest, being the largest tropical forest globally, plays a crucial role in acting as a carbon sink and mitigating the effects of climate change. However, studies indicate that the increasing levels of CO₂ in the atmosphere can disrupt the ability of ecosystems to act as effective carbon sinks. The use of vegetation models, known as Dynamic Global Vegetation models (DGVMs), has become increasingly frequent in understanding the impact of climate change on vegetation through computer simulation of ecological and physiological processes. Some DGVM models, referred to as trait-based vegetation models, also allow the representation of different plant functional strategies within an ecological unit. Given the immense influence of mega-diverse ecosystems like the Amazon rainforest on the global carbon cycle and atmosphere CO₂ concentrations, it is crucial to study the connections between the forest's ecosystem functioning and climate change. For this, we present a novel approach called “inverse modeling”. Conventional vegetation modeling approaches treat traits as parameters (i.e. independent variables) and carbon storage and productivity as outputs (i.e. dependent variables), however, in this approach, we will invert this arrangement. So, through the development of an inverse modeling framework we aim to identify the combination of functional traits that best maintain the Amazon forest's capacity as a carbon sink and ensure essential processes such as evapotranspiration, which impacts rainfall and the water cycle at the local level under climate change scenarios. The inverse algorithm will be developed using two trait-based DGVMs to ensure reproducibility and enhance the algorithm's robustness: CAETÊ (Carbon and Ecosystem functional Trait Evaluation model) and the LPJ-GUESS-NTD (Nutrient-Trait Dynamics). The main input data of this algorithm will be the values of the processes of (i) net primary productivity (NPP), (ii) biomass, and (iii) evapotranspiration rates. Three functional traits, related to productivity, carbon stock and evapotranspiration processes, will be considered to evaluate the functional composition: Specific Leaf Area (SLA, m²/g), Wood Density (WD, g/cm³), Specific Root Length (SRL, cm/g), and the parameter that describes the plant water use strategy related to CO₂ assimilation rates (g1). And, simulations will be made under the climate change predicted according to the IPCC Sixth Assessment Report (2021; increase [CO₂] and average temperature and reduces precipitation to Amazon region). Therefore, we will present to the scientific community an innovative approach to applying ecosystem modeling that allows testing and elucidating new hypotheses about climate change and its impacts on terrestrial ecosystems of global relevance, such as the Amazon. Highlighting the significance of plant functional traits in sustaining ecosystem functioning and resilience, and contributing to discussions on effective management and restoration techniques and methods of modeling.