Get access

Riparian vegetation dynamics: insight provided by a process-based model, a statistical model and field data

Authors

  • F. Ye,

    1. State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-environmental Sciences (RCEES), Chinese Academy of Sciences (CAS), Beijing, China
    Search for more papers by this author
  • Q. Chen,

    Corresponding author
    1. China Three Gorges University, Yichang, China
    • State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-environmental Sciences (RCEES), Chinese Academy of Sciences (CAS), Beijing, China
    Search for more papers by this author
  • K. Blanckaert,

    1. State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-environmental Sciences (RCEES), Chinese Academy of Sciences (CAS), Beijing, China
    2. Department of Limnology of Shallow Lakes and Lowland Rivers, Leibniz-Institute of Freshwater Ecology and Inland Fisheries (IGB), Berlin, Germany
    Search for more papers by this author
  • J. Ma

    1. State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-environmental Sciences (RCEES), Chinese Academy of Sciences (CAS), Beijing, China
    Search for more papers by this author

Correspondence to: Q. Chen, Research Center for Eco-environmental Sciences (RCEES), Chinese Academy of Sciences (CAS), Beijing, 10085, China.

E-mail: qchen@rcees.ac.cn

ABSTRACT

The dynamics of riparian vegetation in a reach of the Lijiang River, China, are investigated. A new process-based model is developed based on cellular automata, which simulates the key processes in the life cycle of the ten most occurring plant species: germination, normal growth, response to floods and droughts, destruction by high velocities, consumption of resources, colonization and competition. The parameterization of these processes is based on controlled experiments on plants sampled in the study area. A traditional statistical model is also developed, which relates the vegetation state to four flow-related variables. Both models are assessed based on data from 12 field surveys in the period from 2009 to 2011, during the dry season, the wet season and at the end of the growing season. Both models predict satisfactorily the spatial distribution of the vegetation cover at the end of the growing season. Although the statistical model is by definition limited to the steady state conditions at the end of the growing season, the process-based model also satisfactorily simulates the temporal dynamics of the vegetation during the dry season and the wet season. Contrary to the statistical model, the process-based model also satisfactorily simulates the vegetation cover outside the area used for the model parameterization. Thus, process-based models are more robust under flow regimes with spatial heterogeneity and important temporal variations. Field observations and process-based model predictions indicate that the regime of dry season and wet season floods is the main regulator of the vegetation cover in the study area. Copyright © 2012 John Wiley & Sons, Ltd.

Ancillary