Modeling the interactions of riparian vegetation and fluvial processes.

Financiador

Organização

Publicações

Modeling the distribution of riverine vegetation in regulated rivers - from dynamic to static equilibrium

Publication . Ochs, Konstantin; Egger, Gregory; Ferreira, Maria Teresa

While methodological advances in ecosystem modeling reflect the growing recognition in the importance of accounting for dynamic change in river ecosystems, it is also recognized that various forms of regulation measures have completely disrupted its natural dynamics. In this context the underlying research question of this PhD is how river regulation affects the spatial distribution of riverine vegetation (aquatic and riparian) and whether rather simple static models that assume equilibrium between vegetation and environmental factors are adequate tools for its prediction. In a first step, we presented a systematic, quantitative literature review on models to predict the distribution of riverine vegetation on reach scale and identified research gaps to guide the further development of the thesis. Then, we developed and tested a habitat suitability model for aquatic vegetation based on hydrological variables. We concluded that during artificially stabilized (static) low flows the vegetation is in equilibrium with the physical instream condition and showed how the model can be used to define a flow threshold that reduces the risk of species invasion and proliferation. Further, we reconstructed the historic succession dynamics of a large river floodplain using a dynamic vegetation model and showed that typical regulation measures led to a steady progression of the vegetation communities toward mature phases without regression to younger stages. Finally, we applied different static and dynamic modeling approaches for the distribution of floodplain vegetation to the same study area and concluded from the comparison of their results that due to regulation measures the relevance of succession dynamics and disturbance stochasticity for the prediction of vegetation patterns is much reduced. Consequently, from a river manager ́s perspective, static models seem to be an adequate option for the modeling of the distribution of riverine vegetation in artificially stabilized environments since they show high accuracy, need relatively few resources (data, time, expert knowledge) when compared to dynamic models and are reproducible

2021Tese de doutoramento

Acesso aberto

Ver mais

Entidade financiadora

Fundação para a Ciência e a Tecnologia

Número da atribuição

PD/BD/114354/2016