Title: Upscaling Soil Moisture from Field to Catchment Scale: Integrating Ground Sensors, Satellite Data, and Drone Surveys for Hydrological Applications
Lead Supervisor: Dr Ponnambalam Rameshwaran, UK Centre for Ecology and Hydrology
Email: ponr@ceh.ac.uk
Co-supervisors: Professor Andrew Wade, Department of Geography and Environmental Science, University of Reading; Dr Neeraj Sah, UK Centre for Ecology and Hydrology; Dr Thomas R Nisbet, Forest Research; Dr Alejandro Dussaillant, UK Centre for Ecology and Hydrology
UKRI funding only covers Home fees which increase annually. International students may still apply to this project, but will be required to meet the difference between the International and Home student fees themselves.
The COSMOS-UK network operates 50+ fixed cosmic ray neutron sensors (CRNS) measuring field-scale soil moisture (~12 hectares per station) continuously since 2013. However, hydrological models used for flood and drought research require catchment-scale distributed soil moisture information—a critical gap that limits model calibration, validation, and prediction accuracy. This project aims to develop systematic upscaling methods combining fixed CRNS measurements with roving CRNS surveys, satellite soil moisture products, and drone-based soil property mapping to create catchment-scale soil moisture maps with quantified uncertainties.
With over a decade of continuous measurements from fixed CRNS stations across diverse UK landscapes, we have unprecedented data on how field-scale soil moisture responds to weather, seasons, and land management. Recent technological advances enable connections between measurement scales: roving CRNS map spatial patterns across larger areas (Schrön et al. 2018), satellite products provide catchment-scale context, and drone hyperspectral sensors characterise soil properties controlling moisture dynamics. Building on foundational work such as Dimitrova-Petrova et al. (2021), which combined fixed and roving CRNS to assess catchment-scale heterogeneity, this PhD introduces key novelties: leveraging the full decade of COSMOS-UK data for long-term insights; integrating multiple data sources (ground sensors, satellites, drones) across scales; testing transferability across UK land uses and soil types; and delivering soil moisture maps with uncertainty estimates suitable for ensemble hydrological modelling.
Field campaigns using roving CRNS will be conducted at selected COSMOS-UK sites representing contrasting land uses (grassland, arable, woodland) and soil types. Sheepdrove (permanent grassland) serves as the primary intensive site for method development, building on existing data and methods. Transferability testing at selected arable and/or woodland COSMOS-UK sites (e.g., Rothamsted, Stoughton, Fincham, Alice Holt) ensures upscaling methods apply across different land uses and management practices. The Chess catchment, an FDRI (Floods and Droughts Research Infrastructure) working catchment, provides validation at larger catchment scales. Each campaign will map soil moisture spatial patterns (integrated over top 10-50 cm depth) under varying wetness conditions, combined with soil sampling and infiltration tests.
Satellite soil moisture products (SMAP, SMOS, Sentinel-1) will be integrated as both validation targets and spatial constraints for upscaling—comparing coarse-scale satellite retrievals (~10-40 km) with aggregated roving CRNS observations tests whether upscaled patterns are consistent with independent remote sensing. Satellite-derived spatial patterns will inform interpolation between sparse fixed stations, while temporal dynamics validate whether upscaled fields capture seasonal variations. Drone hyperspectral surveys (enabled by FDRI infrastructure) will map soil texture, organic matter, and surface characteristics—conducted pre-crop emergence for optimal soil visibility on arable sites. These soil property maps become input layers for physically-based upscaling algorithms, transforming empirical interpolation into process-informed spatial prediction.
The project will evaluate geostatistical and machine learning upscaling techniques to create distributed catchment-scale soil moisture maps. Process-based hydrological models (Hydrus 2D/3D, MIKE SHE, or SWAT) will validate upscaled moisture fields and quantify uncertainty propagation. The modelling tests how upscaled soil moisture distributions improve representation of hydrological processes including spatial patterns in runoff generation, catchment wetting and drying dynamics during notable dry/wet periods, and thresholds controlling rapid response.
Results will address fundamental scaling questions—how do soil properties, land management, and vegetation affect moisture distributions from fields to catchments?—while delivering practical outputs: validated multi-sensor upscaling methods for translating sparse fixed CRNS observations into distributed catchment information; soil moisture maps with quantified uncertainty bounds for model calibration and validation; and transferable protocols applicable beyond instrumented sites. These maps will enable improved representation of antecedent conditions in hydrological models used for flood and drought research, while also supporting applications in agricultural water management (including trafficability and GHG emissions modelling) and understanding controls on natural flood management effectiveness.
Research will be conducted primarily at UK Centre for Ecology & Hydrology (Wallingford) with University of Reading and FDRI support, utilising COSMOS-UK network infrastructure and roving CRNS technology.
The student will develop highly sought skills in field instrumentation, spatial data analysis, satellite remote sensing, hydrological modelling, and scientific programming (Python/R), opening pathways to careers with Environment Agency, Met Office, and environmental consultancies—sectors experiencing growing demand as climate risks intensify.
Training opportunities:
The student will gain expertise in cutting-edge field instrumentation (cosmic ray neutron sensors, drones, soil sampling), satellite remote sensing data processing, spatial geostatistics, and process-based hydrological modelling. Field campaigns across diverse UK landscapes develop practical environmental monitoring skills. The student will access specialised training in digital infrastructure and innovation methods, placements across FDRI catchments, and networking with UK hydrological research communities. Collaboration opportunities include UKCEH and university diverse expertise required for this PhD. Conference presentations at British Hydrological Society, European Geosciences Union, and American Geophysical Union build international profile.
Student profile:
This project would be suitable for students with a degree in Environmental Sciences, Hydrology, Soil Science, Physical Geography, Earth System Science, Agricultural Sciences, or closely related environmental or physical sciences. Strong quantitative and analytical skills are essential, along with enthusiasm for fieldwork in diverse landscapes. Prior experience with data analysis (Python, R, or MATLAB) would be advantageous but not essential, as comprehensive training will be provided. Interest in remote sensing, hydrological modelling, environmental monitoring, or climate change adaptation would be valuable. The interdisciplinary nature means students excited by combining field science, satellite observations, and computational modelling would thrive. UKRI funding only covers Home fees which increase annually. International students may still apply to this project, but will be required to meet the difference between the International and Home student fees themselves.
Sponsorship details:
This project receives support from the Floods and Droughts Research Infrastructure (FDRI), including access to infrastructure, training, placements, expertise across catchments (Upper Tweed, Upper Severn, Chess), and ÂŁ5,000 towards training expenses.
References:
Dimitrova-Petrova, K., Geris, J., Soulsby, C., Wilkinson, M.E., and Lilly, A. (2021). Combining static and portable Cosmic ray neutron sensor data to assess catchment scale heterogeneity in soil water storage and their integrated role in catchment runoff response. Journal of Hydrology, 601, 126659. https://doi.org/10.1016/j.jhydrol.2021.126659
Schrön, M., Köhli, M., Scheiffele, L., Iwema, J., Bogena, H.R., Lv, L., Martini, E., Baroni, G., Rosolem, R., Weimar, J., Mai, J., Cuntz, M., Rebmann, C., Oswald, S.E., Dietrich, P., Schmidt, U., and Zacharias, S. (2018). Cosmic-ray neutron rover surveys of field soil moisture and the influence of roads. Water Resources Research, 54(9), 6441-6459. https://doi.org/10.1029/2017WR021719
