Publication Details

Category Software Publication
DOI 10.5281/zenodo.6373776
Title (Primary) Parsimonious Canopy Model (PCM) v1.0
Version v1.0
Author Bahrami, B.; Hildebrandt, A.; Thober, S.; Rebmann, C.; Fischer, R. ORCID logo ; Samaniego, L. ORCID logo ; Rakovec, O. ORCID logo ; Kumar, R. ORCID logo
Source Titel Zenodo
Year 2022
Department OESA; CHS
Topic T5 Future Landscapes
Abstract Temperate forest ecosystems play a crucial role in governing global carbon and water cycles. However, unprecedented global warming presents fundamental alterations to the ecological functions (e.g., carbon uptake) and biophysical variables (e.g., leaf area index) of forests. The quantification of forest carbon uptake, gross primary productivity (GPP), as the largest carbon flux has a direct consequence on carbon budget estimations. Part of this assimilated carbon stored in leaf biomass is related to the leaf area index (LAI), which is closely linked to and is of critical significance in the water cycle. There already exist a number of models to simulate dynamics of LAI and GPP; however, the level of complexity, demanding data, and poorly known parameters often prohibit the model applicability over data-sparse and large domains. In addition, the complex mechanisms associated with coupling the terrestrial carbon and water cycles poses a major challenge for integrated assessments of interlinked processes (e.g., accounting for the temporal dynamics of LAI for improving water balance estimations and soil moisture availability for enhancing carbon balance estimations). In this study, we propose a parsimonious forest canopy model (PCM) to predict the daily dynamics of LAI and GPP with few required inputs, which would also be suitable for integration into state-of-the-art hydrologic models. The light use efficiency (LUE) concept, coupled with a phenology submodel, is central to PCM (v1.0). PCM estimates total assimilated carbon based on the efficiency of the conversion of absorbed photosynthetically active radiation into biomass. Equipped with the coupled phenology submodel, the total assimilated carbon partly converts to leaf biomass, from which prognostic and temperature-driven LAI is simulated. The model combines modules for the estimation of soil hydraulic parameters based on pedotransfer functions and vertically weighted soil moisture, considering the underground root distribution, when soil moisture data are available. We test the model on deciduous broad-leaved forest sites in Europe and North America, as selected from the FLUXNET network. We analyze the model's parameter sensitivity on the resulting GPP and LAI and identified, on average, 10 common sensitive parameters at each study site (e.g., LUE and SLA). The model's performance is evaluated in a validation period, using in situ measurements of GPP and LAI (when available) at eddy covariance flux towers. The model adequately captures the daily dynamics of observed GPP and LAI at each study site (Kling–Gupta efficiency, KGE, varies between 0.79 and 0.92). Finally, we investigate the cross-location transferability of model parameters and derive a compromise parameter set to be used across different sites. The model also showed robustness with the compromise single set of parameters, applicable to different sites, with an acceptable loss in model skill (on average ±8 %). Overall, in addition to the satisfactory performance of the PCM as a stand-alone canopy model, the parsimonious and modular structure of the developed PCM allows for a smooth incorporation of carbon modules to existing hydrologic models, thereby facilitating the seamless representation of coupled water and carbon cycle components, i.e., prognostic simulated vegetation leaf area index (LAI) would improve the representation of the water cycle components (i.e., evapotranspiration), while GPP predictions would benefit from the simulated soil water storage from a hydrologic model.
linked UFZ text publications
Persistent UFZ Identifier
Bahrami, B., Hildebrandt, A., Thober, S., Rebmann, C., Fischer, R., Samaniego, L., Rakovec, O., Kumar, R. (2022):
Parsimonious Canopy Model (PCM) v1.0
Version: v1.0 Zenodo 10.5281/zenodo.6373776