University of Twente Student Theses


Assessment of drought impacts on groundwater table in the Netherlands using gridded datasets in Google Earth Engine (GEE)

Naeimi, Golnar (2021) Assessment of drought impacts on groundwater table in the Netherlands using gridded datasets in Google Earth Engine (GEE).

Link to full-text:
(only accessible for UT students and staff)
Abstract:Drought impacts in different regions in the world have been affected natural ecosystems and people's properties negatively. As it progresses in every region specifically, there is no universal definition for this hazard. In the Netherlands, in recent years, the drought caused damages to many stakeholders' properties, so governmental budgets were allocated to alleviate the impacts as the current drought indicator has limitations. Twente region is located in the east of the Netherlands in higher elevation areas with sandy soil material. Drainage has made the groundwater resources of the Twente more vulnerable to drought impacts. Also, this region is far from the main surface water storage, so the high-quality primary water is the groundwater. To decrease the wickedness, this study focuses on the permanent largest grasslands in the Twente region and assesses drought impact on the groundwater using global models in Google Earth Engine (GEE). Actual climate variables of precipitation and evaporation were derived from global models' of ERA5 and GLDAS 2.1, available in Google Earth Engine (GEE). The timeseries of daily mean precipitation deficit (PD) over 20 years (2001-2021), cumulative PD, and driest year with the highest PD and lowest surplus of 2018-2019 were analyzed for the two models and KNMI data. Daily mean values over 20 years of the models were evaluated separately with the reference data of KNMI using performance metrics of R coefficient, RMSError, and Mean Absolute Error(MAE). The precipitation ERA5 shows a stronger correlation with KNMI than the GLDAS2.1, respectively correlation coefficient R 0.683, 0.532. However, for the evaporation, especially GLDAS 2.1 indicates a very strong 0.988 and ERA5 a bit lower 0.959. The actual evaporation from the models is strongly in accordance with the reference evaporation in KNMI. Finally, the time series of the groundwater wells from different places in the Twente region and cumulative PD (2001-2021) were analyzed using in-situ measurements data in DINOloket. The groundwater measurements and cumulative PD from ERA5 data were correlated. All three groundwater wells data show that groundwater fluctuations persistently decline (Sep2005-Dec2019). The standard anomalies derived from groundwater show that drought propagation in the region is different. For instance, the B34G0251 in the southern part where the groundwater is shallowest than two others; the groundwater table is more vulnerable to drought impacts. In the first years of the study, this well has the highest inverse correlation that responds with a lag compared to other wells to PD. It probably happened due to the natural flow direction of groundwater towards that area. Again, in the well B34G0251, anomalies near the end of the study period and some years before 2018 became more intensive, and the lowest groundwater table occurred earlier than the highest PD. The earlier response reflects non-climatic factors that caused higher anomaly intensity in this area. The results indicate that drought progressed to the shallow groundwater area in the southern part of the Twente. For future research, the possibility of artificial groundwater recharge using an annual surplus is recommended to prevent more progress of drought to the deeper groundwater table.
Item Type:Essay (Master)
Faculty:ITC: Faculty of Geo-information Science and Earth Observation
Programme:Spatial Engineering MSc (60962)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page