Oil palm modelling in the global land-surface model ORCHIDEE-MICT

Yidi Xu, Philippe Ciais, Le Yu, Wei Li, Xiuzhi Chen, Haicheng Zhang, Chao Yue, Kasturi Devi Kanniah, Arthur P. Cracknell, Peng Gong

TL;DR

This paper introduces a new, detailed model for oil palm within the ORCHIDEE-MICT global land-surface model. By incorporating oil palm's unique "phytomer" structure, age-specific physiology, and management practices, the model significantly improves the simulation of its growth, yield, and ecological impact, offering a crucial tool for understanding and managing this globally important crop.

What Problem Does This Paper Address?

Oil palm is a dominant global oil crop, but its rapid expansion has significant environmental consequences, including deforestation and carbon emissions. Current global land-surface models (LSMs) represent oil palm as a generic tropical tree, failing to capture its specific biological and management characteristics. This leads to inaccuracies in simulating its growth, yield, and its impact on carbon, water, and energy cycles. The authors address this by developing a specialized oil palm Plant Functional Type (PFT) for the ORCHIDEE-MICT model, enabling more realistic simulations.

Methods

The core of this work is the development of a new oil palm PFT within the ORCHIDEE-MICT land-surface model. Key methodological advancements include:

  • Phytomer Structure: Introducing a "phytomer" structure, which represents the palm's basic growth unit (leaf, branch, fruit bunch), allowing for more accurate modeling of its sequential development and harvest.
  • Age-Specific Parameterization: Developing age-specific parameterizations for photosynthesis, respiration, and carbon allocation, reflecting how oil palm physiology changes as it matures.
  • Management Integration: Incorporating key management practices like pruning, fruit harvesting, and the typical 25-30 year rotation cycle.
  • Leaf Phenology: Adapting a leaf age cohort-based phenology scheme to better represent seasonal variations in leaf activity.
  • Carbon Allocation: Modifying the carbon allocation scheme to specifically support the growth of branch, leaf, and fruit components within each phytomer.

Datasets

The study utilized data from 14 observation sites across Indonesia and Malaysia, with one site in Benin. These sites provided measurements for:

  • Leaf Area Index (LAI)
  • Gross Primary Productivity (GPP)
  • Net Primary Productivity (NPP)
  • Fruit bunches (yield)
  • Biomass (total, above-ground, and below-ground) at different ages.

These datasets were used for model calibration and validation.

Key Insights and Ideas

  • The "Phytomer" is Key: Recognizing the phytomer as the fundamental unit of oil palm growth and development is crucial for accurately modeling its productivity and lifecycle.
  • Age Matters Dramatically: Oil palm's physiological processes (photosynthesis, carbon allocation) change significantly with age, and capturing this age-dependency is vital for realistic simulations.
  • Management is Not Trivial: The managed nature of oil palm plantations, including regular harvesting and pruning, fundamentally differs from natural forests and needs explicit representation in models.
  • Generic Models Fall Short: Representing oil palm as a generic tropical tree leads to significant biases, highlighting the need for specialized PFTs for agriculturally important species.

Key Results

The developed ORCHIDEE-MICT-OP model demonstrated significant improvements over the default ORCHIDEE-MICT version:

  • Improved NPP Simulation: The model reproduced NPP with a Normalized Mean Bias Error (NMBE) of 12.87% and an R² of 0.9 across sites, a substantial improvement over the default.
  • Good Yield Reproduction: Simulated fruit yields showed good agreement with observations, with an NMBE of 6.1%, ranging from 12.2-21.4 t DM ha⁻¹ yr⁻¹ compared to observed yields of 13.0-22.1 t DM ha⁻¹ yr⁻¹.
  • Accurate Biomass Estimates: Simulated biomass was more consistent with observations, with an NMBE of 10.4% (after excluding one outlier site).
  • Better GPP Partitioning: The model showed improved partitioning of GPP into respiration and NPP components, aligning better with observations.
  • Realistic Biomass Allocation: The model better represented the allocation of NPP to different biomass components (stem, leaf, root, fruit).

Limitations and Future Work

The authors acknowledge several limitations and suggest future research directions:

  • Simplified Growth Structure: The model assumes a constant maximum phytomer number (40) throughout the life cycle, which may not reflect reality where this number can decrease with age.
  • Inflorescence Gender and Abortion: Factors like inflorescence gender and abortion rate, which influence yield, were not explicitly modeled due to limited understanding.
  • Fixed Management Parameters: Certain management practices (e.g., harvest interval, pruning frequency) are set as fixed parameters, representing a compromise between plant growth and management. Future work could explore making these dynamic.
  • Lack of Nutrient/Water Cycles: The current model does not explicitly include nitrogen and phosphorus cycles, limiting the representation of fertilization effects. Similarly, detailed water and nutrient dynamics, especially in peat soils, could be further refined.
  • Genotype and Plantation Scale: Differences in genotypes, plantation scales, and planting densities, which influence biomass and yield, are not fully captured.
  • Future Work: The authors suggest implementing more management practices (fertilization, irrigation), further parameterizing biophysical variables, and exploring the impact of oil palm conversion on water and energy cycles. They also highlight the potential for modifications to canopy structure to better understand biophysical changes.

Summary updated

Abstract

Oil palm is the most productive oil crop that provides ~40 % of the global vegetable oil supply, with 7 % of the cultivated land devoted to oil plants. The rapid expansion of oil palm cultivation is seen as one of the major cause for deforestation emissions and threatens the conservation of rain forest and swamp areas and their associated ecosystem services in tropical areas. Given the importance of oil palm in oil production and its adverse environmental consequences, it is important to understand the physiological and phenological processes of oil palm and its impacts on the carbon, water and energy cycles. In most global vegetation models, oil palm is represented by generic plant functional types (PFT) without specific representation of its morphological, physical and physiological traits. This would cause biases in the subsequent simulations. In this study, we introduced a new specific PFT for oil palm in the global land surface model ORCHIDEE-MICT (v8.4.2). The specific morphology, phenology and harvest process of oil palm were implemented, and the plant carbon allocation scheme was modified to support the growth of branch, leaf and fruit component of each phytomer. A new age-specific parameterization scheme for photosynthesis, autotrophic respiration, and carbon allocation was also developed for the oil palm PFT, based on observed physiology, and was calibrated by observations. The improved model generally reproduces the leaf area index, biomass density and fruit yield during the life cycle at 14 observation sites. Photosynthesis, carbon allocation and biomass components for oil palm also agree well with observations. This explicit representation of oil palm in global land surface model offers a useful tool for understanding the ecological processes of oil palm growth and assessing the environmental impacts of oil palm plantations.
BibTeX
@inproceedings{xu2021oil,
  author = {Yidi Xu and Philippe Ciais and Le Yu and Wei Li and Xiuzhi Chen and Haicheng Zhang and Chao Yue and Kasturi Devi Kanniah and Arthur P. Cracknell and Peng Gong},
  title = {{Oil palm modelling in the global land-surface model ORCHIDEE-MICT}},
  booktitle = {Geoscientific},
  year = {2021},
  doi = {10.5194/gmd-14-4573-2021},
}

Loading related content...

Figure