IAP-26-130

Can we grow more resilient riverbanks? How plant traits, invasion and drought shape riverbank erosion

Riverbank erosion is a natural process, but excessive erosion can threaten infrastructure, degrade habitats and increase sediment delivery to rivers. Riparian vegetation can help stabilise riverbanks through below-ground root reinforcement, while above-ground growth biomass affects channel roughness and soil moisture, influencing bank stability (Gurnell, 2014). However, vegetation does not always increase bank stability as its effects are dependent on complex interactions between plant characteristics, soil conditions and river hydrology (Simon & Collison, 2002).

Native and invasive non-native plants may influence bank stability in different ways because they differ in traits such as rooting structure, morphology, growth rate and seasonal ground cover (Hardwick et al. 2025). Some plants may reinforce banks through their roots, whereas others may indirectly increase erosion if they displace vegetation that provides more persistent cover (Hardwick et al. 2026). Invasion can therefore reshape the trait composition of riparian plant communities, yet few studies link these shifts to root or ground-cover traits and bank erosion. The influence of vegetation on riverbank erosion may further change under increasingly extreme hydrological conditions, including drought followed by flooding.

Recent reviews of vegetation–river interactions have highlighted the need for a standardised, trait-based framework to understand both how vegetation responds to environmental conditions and how it affects river processes (Corenblit et al., 2024a, 2024b). Such a framework would use consistent methods to measure plant characteristics and assess their relationships with river processes, enabling comparisons across species and environmental conditions. Yet, connecting plant traits, community change under invasion, drought responses and bank erosion remains a challenge.

This project will combine field observations, potential greenhouse and flume experiments, and modelling to test how the traits of selected native and invasive non-native plants, and of invaded communities, influence erosion at individual and patch scales under contrasting hydrological conditions. The findings will inform invasive plant management and riparian restoration by identifying plant traits associated with bank resistance and assessing how these effects may change under drought and flooding. This evidence could help guide the selection and establishment of native riparian vegetation, including trees, under future climate conditions, while recognising that restoration outcomes depend on plant characteristics, establishment stage and local river conditions.

Aim and research questions

Aim: To determine how plant traits, and trait shifts caused by invasion, influence riverbank erosion under contrasting drought and flood conditions, and whether a trait-based approach predicts erosion better than species identity or origin alone.

The project will address three questions:
1. How well do plant functional traits, including community-level trait shifts under invasion, explain differences in erosion resistance compared with species origin alone? We will test whether traits associated with root reinforcement and ground cover explain erosion better than origin, including across a gradient of invasion density.
2. How do drought conditions alter the effects of vegetation on riverbank erosion during subsequent floods? This will examine which plant traits and their response to drought affect bank erosion during subsequent high flow events.
3. Do relationships identified in controlled experiments predict erosion on natural riverbanks better than a species-identity model? We will evaluate whether relationships identified in controlled experiments can explain erosion patterns observed in the field.

A variety of methods will be used in this research project. The student has the opportunity to focus on specific experimental set-ups based on their interests and scientific requirements needed to meet the project aim.

1. Field observations: vegetation and riverbank change
Sites in Scotland and northern England will span native-dominated and invaded (e.g. Impatiens glandulifera, Reynoutria japonica) riparian vegetation, including a gradient of invasion density where feasible, while documenting bank material, geometry and hydraulic exposure. Traits will be measured at plant and patch scales using consistent protocols, potentially including root biomass or density, rooting depth where feasible, above-ground biomass, cover and bare ground, and community-weighted mean traits will be calculated. Buffer width will be recorded as a secondary variable. Repeated bank profile surveys and river flow data, ideally before and after high flow events, will quantify bank change.

2. Greenhouse experiments: plant responses to drought
Controlled greenhouse experiments will investigate how selected plants respond to contrasting soil moisture conditions. Plants will be grown under standardised conditions and assigned to treatments, including a well-watered control, controlled drying and rewatering to assess recovery.
A focused set of traits will be measured at consistent stages. These may include plant growth and survival, above-ground cover, root biomass, root distribution, root tensile strength and soil moisture. The protocol will distinguish response traits, which change with environmental conditions, from effect traits, which influence bank resistance or erosion. Some traits may fulfil both roles. Species selection will prioritise contrasting functional characteristics and experimental feasibility rather than attempting to represent all native or invasive non-native plants. This will help establish which traits, rather than species identity alone, are most informative.

3. Flume experiments: testing effects of traits on erosion
Standardised model banks will be planted with single species and with assemblages along an invader-cover gradient to test whether invasion displaces or adds effect traits. Bank material and geometry will be standardised as far as possible, allowing erosion responses to be compared between selected plant types and moisture treatments.
Measurements will quantify bank retreat or sediment loss, alongside relevant soil and vegetation characteristics. The experiments will test whether drought increases erosion. Outcomes may depend on plant traits, soil properties and the duration of drying. Experimental replication and treatment numbers will be balanced against the time and resources available, with the aim of isolating the most important mechanisms.

4. Analysis and modelling
Results from the greenhouse and flume experiments will inform a mechanistic model linking a focused set of plant traits and environmental conditions to bank erosion. Model predictions will be evaluated against independent experimental observations, with field measurements providing a further test of whether the identified relationships are relevant and reflect riverbank dynamics. The emphasis will be on explaining and predicting erosion under the conditions studied, rather than developing a universal model for all rivers or plant species. Uncertainty and the limits to transferring findings between sites will be considered explicitly.

Year 1

Literature review and/or meta-analysis, with collection of quantitative data if possible (months 1-4); Preparing for fieldwork, including initial consultation with site managers and mapping study sites (month 5); Field Studies Council ID course (month 6); vegetation and river surveys (months 6-8); assigning trait data and planning for flume/greenhouse experiments; winter field surveys (months 11-13).

Year 2

R training course (14); Set up for experimental greenhouse or flume work (14-15), greenhouse or flume experiments (16-19); Field surveys (20-22); Data collection for greenhouse or flume work (23).

Year 3

Data analysis and writing (24-26); Conference preparation and attendance (27); Data analysis and writing (28-32); Field surveys or additional flume/greenhouse work(33-34), Data analysis and writing (35-36).

Year 3.5

Writing publications and thesis submission (months 37-42).

The student will receive training from an multidisciplinary supervisory team, including key skills noted by NERC as being most wanted:

• Fieldwork: this PhD includes a large fieldwork element, and the student will be become experienced in a wide variety of sampling techniques, including remote sensing-based fieldwork if requested
• Taxonomic Identification: to understand wider biodiversity impacts the student will learn identification skills for plants, further supported by attendance on an FSC field course.
• Data management, modelling and numeracy: the student will attend an R training course and will learn a variety of statistical and modelling techniques supported by the supervisory team and wider expertise at the University of Stirling and Newcastle University.
• Translating research into practice: The output of this PhD will inform riparian restoration schemes and the student will receive training in science communication to multiple audiences (e.g. policy makers, non-governmental organisations (NGOs) and the wider public).
• Interdisciplinarity: the project includes and integrates both ecology and geomorphology to inform riparian restoration and function and uses tools from multiple disciplines across different spatial and temporal scales.