Can artificial intelligence improve flood risk management?

Climate change is having an influence on extreme weather events (both severity and frequency) which is leading to increased impacts from environmental disasters such as flooding [1-7]. Severe flooding events are on the increase. Depending on frequency and magnitude, flooding can be either beneficial or detrimental e.g., it can maintain or enhance soil fertility by depositing fresh layers of alluvium and flushing salts out. However, flooding can also be destructive, leading to widescale damage and an increase in the mobility of potentially harmful elements (PHEs), whether these originate from natural or anthropogenic sources. Flood-waters can also redistribute contaminants such as sewage pathogens, metals and radionuclides from different industrial practices and medical uses for example. Such remobilisation was seen during flooding associated with hurricane Katrina [8]. Transport of contaminants in floodwaters and the redistribution of contaminated sediments is a potentially serious human and environmental health issue [9]. Exposure occurs via direct ingestion, dermal contact, and inhalation [10]. Transfer to foodstuffs in urban and peri-urban agriculture systems is increasingly a problem [11].

Our previous research has shown flooding events (or cycles of wetting and drying) can increase mobility of PHEs [e.g., 12]. A review of the literature on flooding-induced contaminant fate showed an increase in contamination levels in soils and pore-water after flooding [13]. These studies have improved our understanding of the biogeochemical processes occurring during and after flood events. Characterising changes in, for example, the solid phase distribution and the bioaccessibility of chemical and biological contaminants before, during and after drying and wetting has provided new knowledge and most recently [12], we have used Artificial Intelligence (AI) tools to successfully predict contaminant mobility (McGlade, pers comm).

In this proposed PhD, our objectives are to:

1) Address the challenges of predicting contaminant mobility and bioaccessibility changes within flood risk areas at a national/landscape scale using artificial intelligence modelling.
2) Make AI based predictions (as maps) of hazard and risk of remobilisation of contaminants nationally.
3) Undertake fieldwork to test and validate the AI based predictive maps.

The student should have a background in environmental sciences with a keen interest/understanding of the effects and consequences of flooding. Ideally, the student will have some experience in the application of AI based tools, but they will benefit from supervisor expertise in flooding, environmental contaminant mobility, hazard and risk assessment and the development and application of AI tools.

Click on an image to expand

Image Captions

Aerial view of flooding in Cumbria British Geological Survey P711398,Flooding in New Orleans caused by Hurricane Katrina Image by 12019 from Pixabay


Initially, spatial databases containing geochemical and flood risk information belonging to BGS, the Environment Agency and the Scottish Environment Protection Agency (SEPA) will provide national scale inputs into the AI tool(s). These will be supported then by targeted fieldwork with subsequent laboratory experiments and sample analysis to determine the drivers influencing contaminant mobility and bioaccessibility, which will be used to train artificial intelligence modelling tools. Key parameters are likely to be soil and water physiochemical properties such as pH and organic carbon. These factors highly influence aspects for example such as i) contaminant ion exchange with soil and sediments, ii) the dissolution of oxides (e.g., iron), which binds with contaminants, and iii) the formation of metal sulfide complexes. These factors all impact a contaminant’s mobility and bioaccessibility. These soil physiochemical properties differ spatially, influenced by soil texture, geology and mineralogy, anthropogenic inputs, and flood frequency. Such heterogeneities in spatial features make modelling changes in contaminant mobility and bioaccessibility complex, which is where artificial intelligence comes in to address these highly unstructured data.

Field study
To help validate the AI modelling output, at one location, a section of each field site will be partitioned off and inundated using pumped river water for a known period (3-5 days), with specific flow and depth parameters. Following a period of inundation, the residual floodwater will be pumped offsite. After a period of rest/drying (approximately 1 month) the site will be flooded for second time. Soil and flood water samples will be collected at up to ten sampling points across each test site pre- and post-flooding on both occasions. All samples will be analysed as below.

Laboratory study
Building on previous work, we expect that soil samples from field sites across the UK will be subjected to simulated flooding under different extreme weather scenarios in the laboratory to test the AI model predictions for chemical remobilisation. Plastic cells filled with the air-dried soil samples will be inundated with simulated river water under different biotic and abiotic conditions (e.g., different soil temperatures or varying soil to solution ratio to simulate faster/slower flooding), followed by periods of drying and subsequent re-inundation. Saturation and drying times will reflect available values from the literature. Physico-chemical parameters will be monitored in real time and soils and inundation fluids will be sampled at each time-step and subjected to sample analysis as described below to determine contaminant mobility and availability for uptake.

Sample analysis
Each sample from the laboratory or field will be subject to a range of analytical tests to determine how different PHEs (e.g., heavy metals, radionuclides, faecal indicator organisms) may be affected by the wet/drying regimes applied. Solid phase distribution of contaminants will be determined using sequential extraction. It is expected that analytical tests such as the following will be used to address the objectives.

• Real-time physico-chemical analysis using a Hana Flow Cell (DO, TDS, pH, ORP, Temperature), Infra-red spectrometry, particle size, organic matter, C, N analysis – to determine the influence of these parameters on contaminant mobility and availability
• Total elemental concentrations using ICP-MS or XRF – to determine PHEs present
Additionally, using the ISO standard [14] the potential human availability of PHE will be determined. The bioaccessibility protocol and sequential extraction data will be used to also determine the availability to certain wildlife species (e.g., mammals). The results of these analyses will be utilised to test the AI modelling.

Data processing and modelling
Using the data collected, the student will predict the environmental and human health impacts of contaminants mobilised and redistributed during flood events. Risk will be assessed by comparison with numeric criteria (e.g., soil guideline values, bathing water criteria etc.) and input into risk assessment modelling software such as PC Cream 2008 [15] and CLEA [16] for humans, ERICA [17] for wildlife to determine the potential impact of PHEs. This risk predictor also needs to be combined with information on the flood probability along with the understanding the drivers for remobilising the PHEs and their associated bioaccessibility (obtained from artificial intelligence), along with information on population dynamics and people’s habits to produce risk maps.

We will further develop and test data driven AI models using traditional regression analysis or machine learning techniques to relate the data for the different PHEs and the co-factors such as temperature, water volume etc.

The resulting outputs will increase our knowledge of the behaviour and fate of contaminants affected by flooding and enable us to provide advice to people growing food for example in areas prone to flooding.

Project Timeline

Year 1

Induction, training needs assessment, literature review, preliminary laboratory work (initial exploratory soil wetting/drying to test different parameters e.g., flooding rate etc.) and training in AI tools and their application for predicting flooding related chemical remobilisation, planning for fieldwork in year 2 to test the AI modelling predictions.

Year 2

Fieldwork nationally to explore the AI model predictions at a range of biogeochemically different catchments, training in and application of bioaccessibility testing, development of AI modelling tools building on the field survey results. Conference attendance to present interim results.

Year 3

Validation of the AI model predictions, production of national hazard and risk maps using the AI modelling approaches. Presentation at an international conference, preparation of publications.

Year 3.5

Thesis write-up and submission. Preparation of publications.

& Skills

The student will benefit from a mix of field, laboratory, and AI computing techniques, all of which are professional transferable skills. Further skills development will be supported through IAPETUS specific provision and external courses. Example courses include statistical analysis with R, conference skills (e.g., networking, poster, and oral presentation skills), Geographic Information Systems (GIS), and grant writing. The supervisory team are highly experienced in environmental assessment, ecological and human health impacts, and AI techniques with access to a breadth of facilities at the Universities of Stirling and Heriot-Watt and the British Geological Survey (BGS). The student will also be able to engage with large research projects at Stirling, Heriot-Watt and BGS (e.g., https://www.camelliawater.org/) involving their supervisors.

References & further reading

[1] https://www.niehs.nih.gov/research/programs/climatechange/health_impacts/weather_related_morbidity/index.cfm

[2] https://www.ipcc.ch/site/assets/uploads/2018/03/SREX-Chap3_FINAL-1.pdf

[3] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9013542/

[4] https://esd.copernicus.org/articles/9/757/2018/

[5] https://www.sciencedirect.com/science/article/pii/S0304389421016563

[6] https://www.nature.com/articles/s41598-022-06929-7

[7] http://dx.doi.org/10.1016/j.scitotenv.2020.142040

[8] https://link.springer.com/article/10.1007/s10653-009-9282-1

[9] https://doi.org/10.1002/esp.3442

[10] https://assets.publishing.service.gov.uk/media/5a7b96a6ed915d131106039e/scho0508bnqy-e-e.pdf

[11] https://doi.org/10.1080/13604813.2013.827843

[12] https://storre.stir.ac.uk/handle/1893/30384

[13] https://link.springer.com/article/10.1007/s11368-012-0552-7

[14] https://www.iso.org/standard/64938.html

[15] https://assets.publishing.service.gov.uk/media/5a8017c040f0b6230269177c/HPA-RPD-058_June_2015.pdf

[16] https://assets.publishing.service.gov.uk/media/5a7ce9eae5274a724f0be48b/scho0508bnqw-e-e.pdf

[17] https://doi.org/10.1016/j.jenvrad.2015.12.011

Apply Now