The role of secondary data in estimating groundwater levels in the Iberian Peninsula

Mapping groundwater levels at the regional scale is often hindered by the lack of high-density direct head measurements. In this study, we explore the value of incorporating different secondary data with cokriging to assess the spatial variability of groundwater levels for the entire Iberian Peninsula, a region characterized by pronounced hydroclimatic variability and rising water demand. The analysis uses more than 60 years of groundwater level measurements from 3,822 observation wells across the region, combined with readily available secondary data, including a digital elevation model, precipitation data, and hydrogeological information. Because of the time series length, the analysis was divided into four periods, each containing a different number of measurements. The performance of the geostatistical models was rigorously evaluated through cross-validation and statistical metrics. Results showed that for earlier periods, characterized by a limited number of observation wells, incorporating groundwater data from subsequent periods and hydrogeological information led to the most significant improvement in groundwater level mapping. Average annual precipitation data did not correlate with groundwater level data and had minimal impact on the interpolation, whereas digital elevation models had a mixed effect. Overall, the study underlines the value of incorporating secondary data through cokriging to map groundwater levels at a regional scale when in-situ data are scarce.