Recognising the various sources of nitrate pollution and understanding system dynamics are fundamental to tackle groundwater quality problems. A comprehensive GIS database of twenty parameters regarding hydrogeological and hydrological features and driving forces were used as inputs for predictive models of nitrate pollution. Additionally, key variables extracted from remotely sensed Normalised Difference Vegetation Index time-series (NDVI) were included in database to provide indications of agroecosystem dynamics. Many approaches can be used to evaluate feature importance related to groundwater pollution caused by nitrates. Filters, wrappers and embedded methods are used to rank feature importance according to the probability of occurrence of nitrates above a threshold value in groundwater. Machine learning algorithms (MLA) such as Classification and Regression Trees (CART), Random Forest (RF) and Support Vector Machines (SVM) are used as wrappers considering four different sequential search approaches: the sequential backward selection (SBS), the sequential forward selection (SFS), the sequential forward floating selection (SFFS) and sequential backward floating selection (SBFS). Feature importance obtained from RF and CART was used as an embedded approach. RF with SFFS had the best performance (mmce=0.12 and AUC=0.92) and good interpretability, where three features related to groundwater polluted areas were selected: i) industries and facilities rating according to their production capacity and total nitrogen emissions to water within a 3km buffer, ii) livestock farms rating by manure production within a 5km buffer and, iii) cumulated NDVI for the post-maximum month, being used as a proxy of vegetation productivity and crop yield.

No abstract

No abstract

In the present work, spectral analysis has been applied to determine the presence and statistical significance of climate cycles in long-term data series from different rainfall and gauging stations located in the Tramuntana Range, in the north-western sector of the island of Majorca. Climate signals recorded previously in the Mediterranean region have been identified: the ENSO, NAO, HALE, QBO and Sun Spot cycles as well as others related to solar activity; the most powerful signals correspond to the annual cycle, followed by the 6-month and NAO cycles. The incorporation of data derived from gauging stations contributes to better climate signal detection as local and exceptional influences are eliminated. Simulations have been performed for each rainfall/gauging station, using the most significant climate cycles obtained by means of the power spectrum. A good correlation between rainfall/flow values and simulated cycles has been obtained. The NAO and ENSO cycles are the most influential in the rainy periods, and specifically the NAO cycle, where a good correlation between episodes of high rainfall/flow and high values of ANAOI can be observed. At a second stage, landslides dated and recorded in the Tramuntana Range since 1954 (174 events) have been correlated with the simulated cycles obtaining good results, as the landslide events match rainfall peaks well. The correlation for the past decade (since 2005), when a detailed landslide inventory is available, also

This work integrates detailed geological and hydrogeological information with PSI data to obtain a better understanding of subsidence processes detected in the detrital aquifer of the Vega de Granada (SE Spain) during the past 13 years. Ground motion was monitored by exploiting SAR images from the ENVISAT (2003ENVISAT ( -2009, Cosmo-SkyMed (2011-2014 and Sentinel-1A (2015Sentinel-1A ( -2016 satellites. PSInSAR results show an inelastic deformation in the aquifer and small land surface displacements (up to À55 mm). The most widespread land subsidence is detected during the ENVISAT period (2003)(2004)(2005)(2006)(2007)(2008)(2009), which coincided with a long, dry period in the region. The highest displacement rates recorded during this period (up to 10 mm/ yr) were detected in the central part of the aquifer, where many villages are located. For this period, there is a good correlation between groundwater level depletion and the augmentation of the average subsidence velocity and slight hydraulic head changes (<2 m) have a rapid ground motion response. The Cosmo-SkyMed period (2011-2014) coincided with a rainy period, and the land subsidence is only concentrated in some points. Rates of average subsidence up to 11.5 mm/yr are obtained for this period and are anthropogenic in origin, being related to earthmoving works. During the Sentinel-1A monitoring period (2015-2016) most of the region showed no deformation, except for some points of unknown origin in the NE sector. A general conclusion is that there is a clear lithological control in the spatial distribution of ground subsidence; all the subsiding areas detected are located where a higher clay content was identified. Although the SE sector of the aquifer had more intense groundwater exploitation, no land subsidence processes were detected, as coarse-grained sediments predominate in the substratum. This research will contribute to the drawing-up of a management plan for the sustainable use of this strategic aquifer, taking into account critical levels of groundwater depletion to avoid land subsidence in the areas identified as vulnerable. The European Space Agency satellite Sentinel-1A could be an effective decisionmaking tool in the near future.

Geostatistical estimation (kriging) and geostatistical simulation are routinely used in ground water hydrology for optimal spatial interpolation and Monte Carlo risk assessment, respectively. Both techniques are based on a model of spatial variability (semivariogram or covariance) that generally is not known but must be inferred from the experimental data. Where the number of experimental data is small (say, several tens), as is not unusual in ground water hydrology, the model fitted to the empirical semivariogram entails considerable uncertainty. If all the practical results are based on this unique fitted model, the final results will be biased. We propose that, instead of using a unique semivariogram model, the full range of models that are inside a given confidence region should be used, and the weight that each semivariogram model has on the final result should depend on its plausibility. The first task, then, is to evaluate the uncertainty of the model, which can be efficiently done by using maximum likelihood inference. The second task is to use the range of plausible models in applications and to show the effect observed on the final results. This procedure is put forth here with kriging and simulation applications, where the uncertainty in semivariogram parameters is propagated into the final results (e.g., the prediction of ground water head). A case study using log‐transmissivity data from the Vega de Granada aquifer, in southern Spain, is given to illustrate the methodology.

No abstract

No abstract