Solar Radiation Triggers the Bimodal Leaf Phenology of Central African Evergreen Broadleaved Forests

Liyang Liu, Philippe Ciais, Fabienne Maignan, Yuan Zhang, Nicolas Viovy, Marc Peaucelle, Elizabeth Kearsley, Koen Hufkens, Marijn Bauters, Colin A. Chapman, Zheng Fu, Shangrong Lin, Haibo Lü, Jiashun Ren, Xueqin Yang, Xianjin He, Xiuzhi ChenJournal of Advances in Modeling Earth Systems 2024 · DOI

TL;DR

Central African rainforests have a unique "double annual cycle" of leaf growth and carbon uptake, which existing models struggle to replicate. This paper introduces a new phenology module for the ORCHIDEE land surface model, driven by solar radiation, that successfully captures this bimodal behavior, significantly improving simulations of leaf phenology and photosynthesis.

What Problem Does This Paper Address?

Tropical rainforests play a crucial role in global carbon and energy cycles. While many rainforests have relatively stable canopies, those in Central Africa near the equator exhibit a distinct bimodal phenology – meaning they have two distinct periods of leaf shedding and regrowth each year. This phenomenon, along with its impact on carbon uptake, is not well understood, and importantly, current land surface models (LSMs) fail to capture this double seasonality. This gap limits our ability to accurately model and predict how these vital ecosystems will respond to climate change.

Methods

The researchers modified the ORCHIDEE land surface model by developing a new leaf phenology module, dubbed ORCHIDEE-AFP. The core innovation is using short-wave incoming radiation (SWd) as the primary driver for leaf shedding and canopy rejuvenation, a departure from previous models that often relied on factors like water availability or vapor pressure deficit. They also introduced a new, four-stage leaf efficiency model that accounts for leaf age and its impact on photosynthetic capacity. This new module was then integrated into the ORCHIDEE framework.

Datasets

The study utilized a combination of:

  • In-situ field data: Long-term leaf phenology observations from two sites in Central Africa (Yangambi and Kibale).
  • Satellite-derived data:
    • Enhanced Vegetation Index (EVI) from MODIS, used as a proxy for young leaf area index (LAI<sub>Young</sub>).
  • Six gridded products related to Gross Primary Production (GPP) and its proxies, including FLUXCOM GPP, FLUXSAT GPP, BESS GPP, and satellite-derived Solar-Induced Chlorophyll Fluorescence (SIF) and Near-Infrared Reflectance of Terrestrial Vegetation (NIRv).
  • Climate forcing data: Two datasets were used: CRUJRA v2.0 and WFDE5, to assess the impact of climate input uncertainty.

Key Insights and Ideas

  • Solar radiation is the key driver: The study strongly suggests that solar radiation (SWd) is the dominant factor triggering the bimodal leaf phenology in Central African rainforests, rather than water availability, which is often a primary driver in other tropical regions.
  • A new phenology paradigm for Central Africa: The standard ORCHIDEE model, designed for more general tropical forests, failed to capture the bimodal pattern. The new SWd-driven approach is specifically tailored to the unique climatic and phenological characteristics of this region.
  • Leaf age matters for photosynthesis: The refined leaf efficiency model, which considers how photosynthetic capacity changes with leaf age, is crucial for accurately simulating the bimodal carbon uptake.
  • Regional differences are critical: The findings highlight the need to differentiate phenological parameterizations for different tropical regions (e.g., Central Africa vs. Amazonia) within LSMs, as their climate seasonality and drivers differ significantly.

Key Results

  • Improved Leaf Phenology: The ORCHIDEE-AFP model successfully reproduced observed leaf turnover (R = 0.45) and young leaf abundance (R = 0.74) at the site scale. Regionally, the proportion of grid cells showing a significant positive correlation between modeled LAI<sub>Young</sub> and observed EVI increased dramatically from 0.2% in the standard model to 27% with the new model.
  • Enhanced Photosynthesis Simulation: The model significantly improved the representation of the bimodal seasonality of GPP. The proportion of grid cells with significant positive correlations between modeled and observed GPP seasonality ranged from 26% to 65% across the six evaluation products, a substantial improvement over the standard model.
  • Climate Forcing Impact: While both CRUJRA and WFDE5 forcings showed improvements with the new model, the CRUJRA forcing generally led to better correlations, suggesting that parameter calibration is sensitive to the specific climate input.

Limitations and Future Work

  • Data Scarcity: The authors acknowledge the limited availability of in-situ measurements for leaf phenology and photosynthesis in Central Africa, which restricts robust model evaluation.
  • Uncertainty in Remote Sensing Data: Uncertainties in satellite-derived data and a lack of detailed ecological knowledge about the specific processes driving this bimodal phenology also pose limitations.
  • Asynchronous Climate Drivers: The model's performance was weaker in regions where rainfall and solar radiation are asynchronous, suggesting that future models might need to incorporate more complex interactions between water stress and radiation.
  • Future Directions: The authors recommend:
  • Collecting more in-situ data to further refine and validate the model.
  • Developing more sophisticated phenology schemes that account for compound impacts of water and radiation.
  • Creating a more detailed vegetation map for Africa that incorporates floristic information and climate synchronicity regimes to allow for region-specific parameterizations.
  • Extending this approach to other tropical regions, like Southeast Asia, to improve pan-tropical modeling.

Summary updated

Abstract

Abstract Central African evergreen broadleaved forests around the equator exhibit a double annual cycle for canopy phenology and carbon uptake seasonality. The underlying drivers of this behavior are poorly understood and the double seasonality is not captured by land surface models (LSM). In this study, we developed a new leaf phenology module into the ORCHIDEE LSM (hereafter ORCHIDEE‐AFP), which utilizes short‐wave incoming radiation (SWd) as the main driver of leaf shedding and partial rejuvenation of the canopy, to simulate the double seasonality of central African forests. The ORCHIDEE‐AFP model has been evaluated by using field data from two forest sites and satellite observations of the enhanced vegetation index (EVI), which is a proxy of young leaf area index (LAI Young ) with leafage less than 6 months, as well as six products of GPP or GPP proxies. Results demonstrate that ORCHIDEE‐AFP successfully reproduces observed leaf turnover ( R = 0.45) and young leaf abundance ( R = 0.74), and greatly improve the representation of the bimodal leaf phenology. The proportion of grid cells with a significant positive correlation between the seasonality of modeled LAI Young and observed EVI increased from 0.2% in the standard model to 27% in the new model. For photosynthesis, the proportions of grid cells with significant positive correlations between modeled and observed seasonality range from 26% to 65% across the six GPP evaluation products. The improved performance of the ORCHIDEE‐AFP model in simulating leaf phenology and photosynthesis of central African forests will allow a more accurate assessment of the impacts of climate change in tropical forests.
BibTeX
@article{liu2024solar,
  author = {Liyang Liu and Philippe Ciais and Fabienne Maignan and Yuan Zhang and Nicolas Viovy and Marc Peaucelle and Elizabeth Kearsley and Koen Hufkens and Marijn Bauters and Colin A. Chapman and Zheng Fu and Shangrong Lin and Haibo Lü and Jiashun Ren and Xueqin Yang and Xianjin He and Xiuzhi Chen},
  title = {{Solar Radiation Triggers the Bimodal Leaf Phenology of Central African Evergreen Broadleaved Forests}},
  journal = {Journal of Advances in Modeling Earth Systems},
  year = {2024},
  doi = {10.1029/2023ms004014},
}
Figure