Online monitoring of shallow groundwater for use in climate adaption solutions

Project Details


This project is based on the results from a Short Scientific Mission under the auspices of the C18 project which was a part of Coast to Coast Climate Challenge. Here it was proved that fluctuations in the sealevel had a significant influence on the shallow groundwater level in the coastal areas.

The purpose of this project is to measure the changes in the groundwater level and evaluate these in relation to precipitation events and other relevant parameters, aiming to help making smarter decisions when choosing climate adaption solutions. This is done through manual analyses of relevant parameters and an examination of the groundwater response. Additionally, a model is developed with the purpose of predicting the local changes in the groundwater level, based on a large amount of data from Juelsminde. This model builds on a machine Learning algorithm, which is chosen based on performance. The modelled values will be visualised as a groundwater table, which is autogenerated based on the predicted measurements.

Layman's description

Previous studies show that the groundwater level in very coastal areas in Juelsminde are influenced by changes in the ocean level. Based on these observations, a prediction of the groundwater level calculated from forecast data (precipitation and ocean level) is wanted.

Key findings

Investigation of the relevant parameters and an subsequent development of a machine learning model resulted in a prediction of the groundwater table respectively 1, 3 and 5 days into the future. The predicted as well as the observed groundwater table is moreover possible to plot in QGIS. This is both possible as a simple model of the groundwater table but also as a map showing the areas in danger of experiencing flooding.
Effective start/end date01/01/2131/12/22


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.