Most of the available freshwaters on Earth are stored in the underground, consequently groundwater represents the main resource in term of water supply. The exploitation of groundwater bodies will increase to face the significant increasing of the global water demand, which has been predicted as a consequence of the future economic expansion, population growth, and urbanization (Rosegrant et al., 2002). Furthermore, the reliance on this resource is continuously growing given the key role that groundwater plays for mitigating the climate change/variability. The estimation of the entity of these effects is mandatory for a reliable management of this crucial resource, which must be protected by suitable actions in order to guarantee safe water supplying for the next generations (Doveri et al., 2016). This work is focused on the water resources destined to the drinkable water distribution, by studying possible empirical relationship between meteorological parameter and groundwater quantity indices. This activity is in the wider context of a research for the development of support tools for the management of the resources under specific climate scenarios. Furthermore, for what regards carbonate aquifer, the impact of climate change can be very significant, given the high sensitivity caused by their karst features. In this work, flowrate of the Cartaro spring (draining a karst aquifer of the Apuan Alps, northwestern Tuscany) and meteorological timeseries (both historical and synthetic scenarios) in the relevant hydrogeological basin were used. Flowrate measurement were provided by the Tuscan Water Authority (AIT) and GAIA ApA (Integrated Water Service), while synthetic meteorological scenarios were provided by Consorzio LaMMA. This work describes the data-driven approach experimented with the collected time series, essentially based on multi-variate analysis techniques and on a simpliﬁed machine learning scheme based on neural networks. In fact, a preliminary test of a data-driven approach based on Multi Layer Perceptron Neural Networks (MLP-NN) is described here. Dedicated techniques for data pre-processing, training and validation have been experimented. In particular, a strong hypothesis of linearity and time-invariance of the system under observation was done, and MLP-NNs were essentially used as non-linear approximators. A further activity regarded the assessment of a performance metric for the evaluation of multiple MLP-NNs with respect to independent test sets, based on either historical or synthetic data. Results are shown in terms of predicted ﬂowrates in a given time window (up to 90 days in our case study), and are organized according to different scenarios of total rainfall quantity.
This research has been supported in part by the Project of National Interest NextData of the MIUR (Italian Ministry for Education, University and Research) and in part by the AIT (Tuscan Water Authority).