PhD: Unravelling the drivers of groundwater recharge variability in the Indo-Gangetic Plain

Summary

This project will develop and then interrogate a large dataset to unravel the drivers of groundwater recharge variability in the Indo-Gangetic Plain, one of the most productive aquifers in the world that underpins regional food and water security.

Project background

The Indo-Gangetic Plain (IGP) is one of the most productive groundwater systems in the world. However, the expansion of irrigation over the last few decades to support the region’s growing population has led to declining groundwater levels in many locations, threatening regional food and water security and farmer livelihoods (Bhattarai et al., 2023). This project will use advanced data science techniques and machine learning to advance our understanding of the IGP groundwater system, to evaluate the impact of future environmental change on regional water resources, and to identify potential measures to ensure sustainable groundwater extraction in one of the most densely populated areas on Earth.

The IGP receives around 80% of its rainfall during the summer monsoon. Groundwater is therefore a vital source of water during the dry winter season, enabling farmers to grow two or three crops annually. There is observational evidence that the frequency of precipitation extremes during the monsoon is increasing (Singh et al., 2014), placing additional pressure on crop management and potentially changing the partition between surface and subsurface runoff. In the Himalayas, a shift from snow to rain and accelerating glacier retreat is changing the timing and magnitude of streamflow entering the IGP. At the same time, population growth, economic development and technological innovation are putting regional water resources under increasing pressure.

Unravelling the various drivers of groundwater recharge is vital to understanding how the aquifer will behave in the future. However, the relative influence of the various pressures on regional groundwater availability is poorly understood. This PhD will exploit large observational datasets collected over many years by the supervisory team to increase our understanding of the IGP. You will join a vibrant community of hydrologists and hydrogeologists at the British Geological Survey and the University of Edinburgh. The PhD project aligns with the TerraFIRMA project (https://ukesm.ac.uk/terrafirma/), and you will therefore have the opportunity to interact with scientists across the UK working on challenges at the intersection of climate change, resilience and sustainable development.

Research questions

1. What is the relative influence of large-scale climate variability, evolving mountain hydrology, irrigation water withdrawal, and hydrogeology on groundwater storage across the IGP?
2. What are the linkages between the Himalayas and groundwater recharge in the IGP, and how are these evolving in response to climate change?
3. What is the role of hydrometeorological extremes on groundwater recharge and groundwater resources?
4. How might the interactions between climate change and socioeconomic change impact future groundwater availability?

Methodology

1. Data collection (M1-M12): The first task will be to compile a regional spatiotemporal dataset containing historical measurements of groundwater level and multiple potential drivers of recharge variability. This will include historical climate as well as spatiotemporal estimates of irrigation source and depth and other socioeconomic variables. You will benefit from existing data held by the project supervisors, but additional collection of legacy data in India is likely to be an important part of the project. key research challenge will be to develop methods for handling incomplete and noisy data. The complete dataset will provide significant opportunities for additional analysis depending on the interests of the student.
2. Identify dominant drivers (M13-M24): You will develop a regional probabilistic ML model of groundwater recharge using the identified drivers. You will apply feature engineering to ensure the ML model is exposed to the most relevant formation. Using interpretable machine learning methods, you will unravel the drivers of groundwater recharge variability in the region and assess their variability in time and space.
3. Future prediction (M25-M36): In the last step, you will use the ML model and other methods to understand and predict the future behaviour of the system using projections of socioeconomic and climate change. A key challenge will be to translate the regional projections of change to the local scale. The modelling framework will enable the identification of hotspots of future change, and evaluate potential measures to achieve sustainable groundwater use.

Training

A comprehensive training programme will be provided comprising both specialist scientific training and generic transferable and professional skills. Additional training will be tailored to the profile of the student, but may cover topics such as hydrogeology, scientific programming, geostatistics and machine learning. During the PhD you will develop expertise in these areas, preparing you for a career in research or industry.

Requirements

You should have a background in the physical sciences, engineering, or mathematics. Some experience with a programming language such as R, Python or MATLAB is desirable, but most important is a willingness to learn. You should be willing to travel to India to collect legacy data and collaborate with local partners.

References

Bhattarai et al., Warming temperatures exacerbate groundwater depletion rates in India.Sci. Adv.9, eadi1401(2023). https://doi.org/10.1126/sciadv.adi1401

MacAllister, D.J., Krishan, G., Basharat, M. et al. A century of groundwater accumulation in Pakistan and northwest India. Nat. Geosci. 15, 390–396 (2022). https://doi.org/10.1038/s41561-022-00926-1

MacDonald, A., Bonsor, H., Ahmed, K. et al. Groundwater quality and depletion in the Indo-Gangetic Basin mapped from in situ observations. Nature Geosci 9, 762–766 (2016). https://doi.org/10.1038/ngeo2791

Singh, D., Tsiang, M., Rajaratnam, B. et al. Observed changes in extreme wet and dry spells during the South Asian summer monsoon season. Nature Clim Change 4, 456–461 (2014). https://doi.org/10.1038/nclimate2208

Turner, A., Annamalai, H. Climate change and the South Asian summer monsoon. Nature Clim Change 2, 587–595 (2012). https://doi.org/10.1038/nclimate1495

Supervisors

E4 supervisors are happy to hear from candidates who would wish to adapt the project to their own ideas and research background.