Phosphorus (P) is one of the most limiting nutrients for plant growth, especially in tropical ecosystems. It plays a critical role in shaping productivity and functioning of terrestrial ecosystems by regulating photosynthesis, respiration, carbon (C) allocation and some other physiological processes. However, influences of P cycle on C and nitrogen (N) cycling have not been well addressed so far. To better understand how and to what extent P cycling interacts with C and N cycling, we incorporated P processes into a process-based land ecosystem model, Dynamic Land Ecosystem Model (DLEM). Since previous studies have demonstrated tropical and subtropical areas are mainly P-limited, we selected two low latitude ecological sites with highly weathered soils (Amazon in Brazil, Hawaii in United States ) to test the model performance. The DLEM model with both coupled and decoupled P-C-N processes was applied to the above sites to explore how model simulation results would be changed with and without considering phosphorus effects.
Results/Conclusions
Our preliminary results show that the P-C-N coupled model simulations were more consistent with field observations than the decoupled model did, e.g. annual GPP would be greatly overestimated if no P cycle was coupled in model simulations. These results suggest that it is essential to couple P processes in ecological models particularly for simulating P-deficient ecosystems. To better understand the role of P dynamics in terrestrial ecosystems and improve the coupled P-C-N model, the DLEM model will be tested with observation data from a variety of ecosystems. We will apply the improved model and newly developed regional database to investigate the dynamics of P cycle and its control over terrestrial productivity at regional level.