Project highlights

  • Headwater and intermittent small streams are two key riverine types, partly overlapping, representing the vast majority of river networks worldwide, yet less monitored and understood than larger rivers; this project would be of interest internationally 
  • The project will require developing state-of-the-art technical skills and approaches (e.g. UAV data collection and analysis, which could involve AI-based image classification) alongside a thorough understanding of the hydrology and ecology of headwater/intermittent streams 
  • The project will benefit from access to UKCEH and University of Birmingham facilities, equipment and expertise (e.g. high performance cluster computing, drones) 

Overview

Intermittent streams (c. 50% of world rivers) and headwaters (c. 90% of rivers) are key riverine systems for hydrology, ecology and water quality, which are poorly monitored and lack data (Dugdale et al. 2022; Mainstone et al. 2018; Datry et al. 2016). Streams are often flagged as intermittent based on downstream flows, but where or how they dry is not known. Capturing the evolution of river habitat and connectivity as streams dry out is particularly important (Dugdale et al. 2022) for process hydrology, river ecology (e.g. refugia during drought, loss of physical in-channel habitat, loss of longitudinal connectivity, loss of lateral connectivity with riparian area, floodplain and wetlands), and water quality (pools are considered to favour algae blooms). Unpiloted Aerial Vehicles (UAVs), more commonly called drones, are a potential way to address these issues (Kelleher et al. 2016). River habitat monitoring can also benefit from using UAV data, which can provide fast higher resolution updates than traditional techniques. Figure 1 shows an example of mapping an intermittent chalk river in the UK. This research is relevant to other water-dependent systems faced with similar issues (e.g. wetlands) and could complement research on UAV-based water quality surveys in larger rivers. Beyond the riverine environment, the project would also explore potential to link with terrestrial UAV applications (e.g. biodiversity). 

The figure gives two examples of mapping an intermittent stream using a drone with optical and thermal infrared cameras. The top map shows the river when it was wet, the bottom map shows the same river three months later when it was drier. One can see how river habitat is lost when intermittent rivers start to dry out.

Figure 1: Drone-based mapping of a stretch on the intermittent river Beane, near Luffenhall, UK showing the loss of habitat between wetter conditions (14/12/24; top) and drier conditions (26/03/24; bottom); map was derived by combining unsupervised classifications of the RGB, TIR and DSM orthomosaics; blue, open flowing water class, green, emerging macrophytes/vegetated deposits class; map superimposed on the RGB orthomosaic. 

CENTA Flagship

This is a CENTA Flagship Project

Host

UK Centre for Ecology & Hydrology

Theme

  • Climate and Environmental Sustainability
  • Organisms and Ecosystems

Supervisors

Project investigator

Co-investigators

How to apply

Methodology

A selection of river stretches located on small intermittent rivers will be surveyed as they dry and replenish (re-wetting) during a typical drought cycle. This ideally requires repeated surveys at high, medium, and low flows plus at least a couple of surveys during the intermittent phase. The total number of sites and surveys will be commensurate with the available time and resources. Sites will be surveyed on the ground too. The objective is to build a time series of surveys ideally capturing several cycles. Different types of drone-mounted sensors will tested, with likely types being optical, TIR, and multi-spectral cameras. They will be assessed not only from a scientific perspective but also against the context of future data collection (e.g. potential for citizen science). A consistent image classification approach will be developed. Possible route could be combining thermal imagery and very high–resolution visible imagery followed by AI-based classification to identify and segment the different habitat units. Developing an AI classifier would require using Python packages like PyTorch (e.g. Carbonneau et al., 2020a, 2020b). The supervision team is currently involved in a variety of active projects on these topics (e.g. looking at intermittent chalk rivers for a water company) and the student will be able to capitalise on this experience. We also have our own pool of drones, which will be available for the project. 

Training and skills

DRs will be awarded CENTA Training Credits (CTCs) for participation in CENTA-provided and ‘free choice’ external training. One CTC can be earned per 3 hours training, and DRs must accrue 100 CTCs across the three and a half years of their PhD.  

The student will benefit from access to the full portfolio of trainings available to UKCEH staff, from technical skills to personal professional development. Relevant trainings to this project will likely include advanced coding and data analysis (e.g. R, Python), statistics, machine learning, GIS, remote sensing software, field work practices, first aid. The student would be supported to learn how to pilot UAVs as far as practical and appropriate (e.g. working towards a GVC) but would get support from qualified pilots in the interim. 

Further details

For any enquiries related to this project please contact Dr Cedric Laize ([email protected]).  

The successful applicant would be registered at the University of Birmingham. 

To apply to this project:  

  • You must include a CV with the names of at least two referees (preferably three) who can comment on your academic abilities.  
  • Please submit your application and complete the host institution application process via: https://sits.bham.ac.uk/lpages/LES068.htm.   Please select the PhD Geography and Environmental Science (CENTA) 2025/26 Apply Now button. The CENTA Studentship Application Form 2025 and CV can be uploaded to the Application Information section of the online form.  Please quote CENTA 2025-UKCEH2 when completing the application form.  

Applications must be submitted by 23:59 GMT on Wednesday 8th January 2025.  

Possible timeline

Year 1

Site selection; flight campaign planning; test surveys; initial image analysis and survey protocol; literature review.

Year 2

Repeated surveys; development and consolidation of image processing approach; initial analysis of mapped geodata.

Year 3

Final surveys; comparison UAV and alternative EO data (e.g Lidar); final analysis of mapped geodata.

Further reading

Carbonneau, P., Belleti, B., Micotti, M., Lastoria, B., Casaioli, M., Mariani, S., Marcheti, G., and Bizzi, S. (2020) ‘UAV-based training for fully fuzzy classification of Sentinel-2 fluvial scenes’. Earth Surf. Process. Landforms, 45, 3120–3140. 

Carbonneau, P., Dugdale, S., Breckon, T, Dietrich, J., Fonstad, M., Miyamoto, H., Woodget, A. (2020) ‘Adopting deep learning methods for airborne RGB fluvial scene classification’. Remote Sensing of Environment, 251. 

Datry, T., Corti, R., Foulquier, A., von Schiller, D., and Tockner, K. (2016) ‘One for all, all for one: A global river research network’, Eos, 97, https://doi.org/10.1029/2016EO053587. 

Dugdale, S., Klaus, J., and Hannah, D. (2022) ‘Looking to the skies: Realizing the combined potential of drones and thermal infrared imagery to advance hydrological process understanding in headwaters’. Water Res Research, 58. 

Kelleher, C., Scholz, C. A., Condon, L., and Reardon, M. (2018) ’Drones in geoscience research: The sky is the only limit’, Eos, 99, https://doi.org/10.1029/2018EO092269. 

Mainstone, C. et al. (2018) Developing a coherent framework for assessing priority freshwater habitats in England. Natural England JP016.