| Abstrakt: |
Terrestrial gross primary production (GPP) represents the magnitude of CO2 uptake through vegetation photosynthesis, and is a key variable for carbon cycles between the biosphere and atmosphere. Light use efficiency (LUE) models have been widely used to estimate GPP for its physiological mechanisms and availability of data acquisition and implementation, yet each individual GPP model has exhibited large uncertainties due to input errors and model structure, and further studies of systematic validation, comparison, and fusion of those models with eddy covariance (EC) site data across diverse ecosystem types are still needed in order to further improve GPP estimation. We here compared and fused five GPP models (VPM, EC‐LUE, GOL‐PEM, CHJ, and C‐Fix) across eight ecosystems based on FLUXNET2015 data set using the ensemble methods of Bayesian Model Averaging (BMA), Support Vector Machine (SVM), and Random Forest (RF) separately. Our results showed that for individual models, EC‐LUE gave a better performance to capture interannual variability of GPP than other models, followed by VPM and GLO‐PEM, while CHJ and C‐Fix were more limited in their estimation performance. We found RF and SVM were superior to BMA on merging individual models at various plant functional types (PFTs) and at the scale of individual sites. On the basis of individual models, the fusion methods of BMA, SVM, and RF were examined by a five‐fold cross validation for each ecosystem type, and each method successfully improved the average accuracy of estimation by 8%, 18%, and 19%, respectively. Plain Language Summary: Plants uptake carbon dioxide through photosynthetic process from atmosphere, and the amount of carbon fixed is called terrestrial gross primary production (GPP) which is a key factor in global carbon cycle. People have developed many process‐based models to simulate GPP at regional or global scale, and Light use efficiency (LUE) GPP models have been widely used for they are easy to implement and the input data could be conveniently accessed. How to improve the capacity of GPP model to get more precise estimation remains a change for a long time. In this study, for the purpose of improving single model capacity, we ran five popular individual GPP models (VPM, EC‐LUE, GOL‐PEM, CHJ, and C‐Fix) across eight plant types based on site‐measured and remote sensed data, and then fuse them together using three machine learning methods: Bayesian Model Averaging (BMA), Support Vector Machine (SVM), and Random Forest (RF). We found that EC‐LUE performed best to simulate GPP as an individual model. RF was the best fusion method to improve estimation of GPP, followed by SVM and BMA, with the increased average accuracy of estimation by 8%, 18%, and 19%. Key Points: The performance of individual gross primary production (GPP) models varied to capture interannual variability of GPP across different plant functional typesMachine learning algorithm could significantly improve the capacity of estimating GPP by combining the advantages of all individual models [ABSTRACT FROM AUTHOR] |
Plný text