Figure 1. Number of peer reviewed journal articles per year using the combined terms ”riparian vegetation” + ”geomorphology” + ”river*”, as archived on Clarivate Web of Knowledge (all databases) (https://www-webofscience-com.inee.bib.cnrs.fr/). A.M. Gurnell’s first article using these terms was published in 1997 (see vertical black arrow) and her name comes out first with more than 12% of all articles authored or co-authored. Accessed on 11th November 2023.
The objective of this article is to present a set of studies and results from the past 20 years obtained by the authors and many collaborators, including Angela M. Gurnell, on a panel of French rivers: Tech, Garonne, Isère and Allier Rivers. In particular, feedbackmechanisms between fluvial morphodynamics and riparian vegetation dynamics were investigated directly in the field and used high resolution remote sensing at the scale of individual plants, populations, communities and landscapes as well as during semi-controlled ex situ experiments at the scale of individual plants. Collectively, the authors’ research contributed to elucidate some key aspects of feedback dynamics between the lowest and highest levels of the riparian ecosystem organisation. This article presents and discusses those key aspects.
Feedbacks between vegetation and fluvial geomorphology at nested spatiotemporal scales
Riparian ecosystems are among the most dynamic and variable environments on Earth’s surface in terms of structure and function, making them a target of many ecological restoration projects (Amoros and Petts, 1993; Naiman and Décamps, 1997; Muñoz-Mas et al., 2017; Gonzalez et al., 2018; González, et al. 2022; O’Briain et al., 2023). They represent land-water interface zones where matter and energy are concentrated, transit, are stored, and transformed. The flows of water, sediment, organic matter – from dissolved organic matter to large woods –, and seeds that pass through, both cyclically (hydrological regime) and stochastically (floods), contribute to maintain a high degree of heterogeneity in the mosaic of habitats (Townsend, 1989; Gurnell et al., 2005). This spatiotemporal heterogeneity generates strong ecological gradients and sometimes exceptional levels of biodiversity, including a variety of survival and reproduction strategies in a shifting mosaic (Ward, 1989; Schnitzler et al., 1992; Ward et al., 2002; Stanford et al., 2005; Bornette et al., 2008).
Hydroecological theoretical models are based on the postulate that the structure and functioning of aquatic and riparian communities can be explained by considering the links between life history traits of species and the geomorphological structures and processes that govern flows of matter and energy (Vannote et al., 1980; Newbold et al., 1982; Pickett and White, 1985; Amoros et al., 1987; Junk et al., 1989; Townsend, 1989; Ward, 1989; Tockner et al., 2000). Models such as the River continuum concept, Flood pulse and Flow pulse, Nutrient spiralling, Patch dynamics, were nevertheless developed according to a type of Physical habitat template approach (sensu Southwood, 1977), that is, focusing strictly on the a priori unidirectional effect of the physicochemical heterogeneity on biodiversity and ecosystem functioning. Previous research on riparian ecosystems has not sufficiently considered how and to what extent the three spatial dimensions of riverine ecosystems are controlled by riparian vegetation.
Certain plant species and assemblages have a profound and lasting impact on the structure and function of riparian ecosystems by modulating matter and energy fluxes and modifying substrate cohesion (Gurnell et al., 2012). Riparian plant species (e.g., the black poplar Populus nigra L. in the Northern hemisphere) that significantly affect hydrogeomorphological processes and fluvial landforms have been identified as ecosystem engineers (sensu Jones et al., 1994), keystone species (sensu Paine, 1966), and, more recently, as potential niche constructors sensu Odling-Smee et al. (2003) (Edwards et al., 1999; Gurnell and Petts, 2006; Corenblit et al., 2009b; Francis et al., 2009).
How, and to what extent, the structure and function of riparian ecosystems are controlled by abiotic-biotic feedbacks across various spatiotemporal scales has become an important issue in fluvial biogeomorphology. This question must be addressed from the fine scale of the individual plant, the microhabitat and single hydrological event to the coarse scale of plant community mosaic, the fluvial corridor and the hydrosedimentary regime. In particular, the mechanisms by which engineer plants respond to, and shape, the mosaic of riparian habitats are not fully understood. Some evolutionary adaptations of plants in response to the hydrogeomorphological disturbance regime such as floating seeds, synchronization of reproduction with the flood regime (Junk et al., 1986; Guilloy-Froget et al., 2002), vegetative reproduction in addition to sexual reproduction strategies (Barsoum, 2002; Francis and Gurnell, 2006; Moggridge and Gurnell, 2009), and biomechanical stem flexibility in order to avoid uprooting during flood events (Karrenberg et al., 2002; Lytle and Poff, 2004; Bornette et al., 2008), help riparian plants to survive despite high physical constraints. These adaptations, combined with phenotypic plasticity, allow engineer plants to mechanically resist to drag force in the flow and to influence their riverine environment. By trapping sediments, organic matter, and seeds, engineer plants can create, build, and maintain riparian habitats that are beneficial to other plant species, including other engineer plants.
In addition to the role of vegetation assemblages (i.e., community scale) in shaping the hydrogeomorphic features of riverine ecosystems, specific life history, morphological and biomechanical traits related to distinct engineer species also play an important role. In the absence of large, destructive floods, the short-range actions of plants – within their own stands and immediately downstream – typically result under adequate conditions in the accumulation of fine sediment, nutrients, organic matter, which improve habitat quality. In addition, with the deposition of seeds (e.g., Goodson et al., 2001, 2002; Gurnell et al., 2004, 2006; Corenblit et al., 2009a, 2016a), the improved habitat quality, in turn, promotes the establishment and growth of new plant species, including other engineer species. This process can increase the likelihood of pioneer woody engineer species such as P. nigrareaching sexual maturity (Corenblit et al., 2014a, 2018).
The synchronized process of fluvial landform construction and plant establishment/growth/succession can lead to the emergence of characteristic biogeomorphological landforms such as fluvial islands that are the result of engineering effects of riparian plants (Gurnell et al., 2001). These positive abiotic-biotic feedback dynamics which potentially affect the fluvial style (e.g., island braided riverssensu Gurnell et al. 2019) are consistent with the action of ecosystem engineers (sensu Jones et al., 1994; Gurnell and Petts, 2006) and, potentially, niche constructors (sensu Odling-Smee et al., 2003; Corenblit et al., 2015). In turn, the effects of riparian engineer plants on the hydrogeomorphological components of riverine ecosystems involve a set of responses in these plants, from the individual to the community mosaic level and vice versa. This abiotic-biotic feedback is organized in three stages (Fig. 2):
(1) Initial response of plants to hydrogeomorphological constraints: at the beginning of growth and succession, plants must respond to a variety of hydrogeomorphological constraints, including mechanical forces (e.g., stream flow, erosion and sediment burial) and physiological stress factors (e.g., prolonged immersion and drought). These constraints act as a powerful filter for the colonisation, establishment and growth of engineer plants in exposed areas on alluvial bars. However, they can also stimulate morphological and biomechanical responses that increase plant resistance to these constraints.
(2) Effects of engineer plants on hydrogeomorphological components: the mechanical resistance allows plants to act passively on their hydrosedimentary and geomorphological environment. Once established, engineer plants can modify the hydrogeomorphological components of their environment in a variety of ways. For example, their roots can increase substrate cohesion, which can help to stabilize bars and banks and prevent erosion. Their trunks, branches and leaves can promote trapping of sediment and organic matter, which can help to build up islands, floodplains and create new habitats.
(3) Plant response to changes in the hydrogeomorphological environment: as engineer plants modify their environment, they can create positive feedback loops that further enhance their impact and growth (Corenblit et al., 2009a, 2014, 2018) as well as negative feedback loops translated into the transition toward stabilized and mature stages of the vegetation succession on raised alluvial surfaces (Bendix and Hupp, 2000; Corenblit et al., 2007; Corenblit et al., 2009a). For example, the increased stability of river bars and banks and their vertical accretion provided by engineer plants can reduce erosion and sediment burial, which can create more favourable conditions for the establishment of other competitive plants. This can lead to the development of complex and diverse riverine ecosystems with increased beta diversity at the landscape scale (Gurnell et al., 2005).