Spatiotemporal patterns of the gross primary production in the salt marshes with rapid community change: A coupled modeling approach

Coastal salt marshes are among the most productive ecosystems in the world. However, rapid changes in the vegetation communities of salt marshes caused by exotic species invasion and plant propagation requires a better understanding of how these shifts affect the landscape-scale variations of gross primary production (GPP). In the Yangtze Estuary of eastern China, we firstly compared 2-years GPP values obtained through eddy covariance measurements, satellite-based estimations and model simulations in two vegetation mixtures consisting of exotic Spartina alterniflora (C-4 plant) and dominant native Phragmites australis and Scirpus spp. (C-3 plants) The results indicated that the low-resolution remote sensing data with light-use efficiency method did not represent well the seasonal course of GPP relative to flux measurements and underestimated the annual amount GPP by 25-32%. In contrast, a detailed process based vegetation model with species-specific parameterizations could identify the proportions of GPP from exotic C-4 and native C-3 vegetation and accurately reproduce the seasonal course and annual amount of GPP in the mixtures. The slopes of the linear regressions between the measured and modeled GPP were close (1.09 and 0.89 for the two mixtures, respectively) to the 1:1 line. To further evaluate the variations in GPP throughout the salt marshes, we coupled high-resolution remote sensing data to the vegetation model by transforming the vegetation index into the leaf area index (LAI) for different species. The coupled model reproduced the spatiotemporal dynamics of GPP in the salt marshes with rapid community change during the period of 2000-2008 and identified the variations of GPP from different species. The simulations indicated that the contribution rate of exotic S. alterniflora to GPP has been greater than that of native species since 2003-2004. Therefore, we suggest that this method is useful for high-resolution estimations of regional GPP in other coastal marshes with invasive S. alterniflora in China. (C) 2015 Elsevier B.V. All rights reserved.