Introduction
Tropical ecosystems, including forests and savannas, play a pivotal role in the global carbon cycle. However, their carbon storage capacity exhibits significant inter-annual variability. This study investigates the "memory" of these ecosystems—specifically, how lagged effects from past environmental conditions influence current carbon balances. To quantify these effects, the researchers developed and utilized the DALEC2a model within the CARDAMOM framework.
The DALEC2a Model
The Data Assimilation Linked Ecosystem Carbon model version 2a (DALEC2a) is a sophisticated "box model" designed to track the movement of carbon and water through six interconnected pools: Labile (temporary storage), Foliar (leaves), Fine Roots, Wood, Litter, and Soil Organic Matter (SOM).
Key Advancements
- Water Module: Unlike previous versions, DALEC2a includes an explicit plant-available water module. It calculates a water stress threshold based on precipitation, evapotranspiration, and runoff, which directly limits Gross Primary Production (GPP).
- Fire Module: The model incorporates fire dynamics, estimating immediate carbon emissions and "fire-induced mortality." This mortality shifts carbon from living biomass to litter and soil pools, leading to delayed CO2 release through decay.
Datasets and Optimization
The model parameters are optimized using the CARDAMOM framework, which "ingests" diverse observational datasets to ensure ecological realism. Key constraints include MODIS Leaf Area Index (LAI), biomass benchmarks from Saatchi et al. (2011), and soil carbon data from the Harmonized World Soil Database (HWSD).
Additionally, Solar-Induced Fluorescence (SIF) from GOSAT serves as a critical proxy for GPP, allowing the model to validate its simulation of photosynthetic activity across the tropics.
Key Insights: Structural Carbon Memory
The study reveals that ecosystem "memory" is physically stored within carbon pools. For instance, environmental stressors like drought may not cause immediate mortality but can deplete labile carbon stores, increasing vulnerability to future stressors. Fire acts as a major redistributor, moving carbon from stable living pools to decomposing dead pools, thereby creating a multi-year "hangover" effect on carbon emissions.
Model Validation and Results
Validation against independent datasets confirms the model's high fidelity. The framework accurately reproduces observed patterns of biomass and soil carbon stocks across tropical regions.
The model also successfully attributes the variability of Net Biospheric Exchange (NBE) to specific pool contributions, confirming that lagged effects are the dominant driver of inter-annual variability in the 2010–2015 tropical carbon balance.
Conclusion
By explicitly modeling carbon pools and water-fire dynamics, DALEC2a provides a robust tool for quantifying the legacy of climate events. This framework offers a pathway to improve global climate models, which currently struggle to predict the future of the tropical carbon sink under increasing environmental extremes.
Summary updated