Please login first
EVALUATING TRENDS IN GROUNDWATER DISCHARGES FUNCTIONING THROUGH MACHINE LEARNING TOOLS APPLIED TO SPRINGS IN KARSTIC AQUIFERS: RESULTS OBTAINED IN LAS LORAS UNESCO GLOBAL GEOPARK (SPAIN)
* 1 , 2 , 1 , 2 , 3 , 3 , 4 , 5 , 6 , 7 , 8, 9
1  Centro Nacional Instituto Geológico y Minero de España, Consejo Superior de Investigaciones Científicas, (IGME-CSIC). Ríos Rosas 23, 28003 Madrid, Spain.
2  Department of Geodynamics, Stratigraphy and Paleontology. Faculty of Geological Sciences. Jose Antonio Novais s/n, 28040 Madrid, Spain.
3  Geoparque Mundial UNESCO Las Loras. 34800 Aguilar de Campoo, Palencia.
4  Birzeit University – Ramallah / Palestine.
5  Water, Energy and Environment center, The University of Jordan. Jordan.
6  Cadi Ayyad University, Faculty of Sciences Semlalia of Marrakech, Earth Sciences Dept. Marocco.
7  Springs Stewardship Institute 414 N Humphreys St. Flagstaff, AZ 86001
8  Queensland Herbarium (DES), Mt Coot-tha Road, Toowong, QLD 4066, Australia;
9  Department of Biological Sciences, University of Queensland, St Lucia, QLD 4072, Australia;
Academic Editor: Junye Wang

Abstract:

The UNESCO Global Geopark Las Loras (Palencia-Burgos, 960 km², Spain) constitutes a significantly sensitive area to changes caused by the impact of climate change, as it is located at the transition between the Atlantic and Mediterranean sides of the biogeographical regions in Spain and is a notably depopulated area.

The analysis of the impact of climate change on groundwater resources has been carried out by applying global climate models to the precipitation and temperature data series available at the Aguilar de Campoo weather station. The obtained projections for future potential climate scenarios (time intervals 2021-2040 and 2051-2070) in Las Loras UGGp at a monthly scale have been based on AEMET Free and open data, Representative Concentration Pathway (RCP), and Global Circulation Models (GCM).

The prevalence of karstic aquifers identified in the Las Loras UGGp renders this area particularly vulnerable to potential declines in water resources. The conservation of springs is an essential indicator of the efficiency in groundwater management. Using the springs as an indicator of water management, ML tools have been applied to estimate the number of active springs versus inactive springs as a first theoretical approach, which is expected to be checked in the near future using field campaigns and other complementary methodologies. The number of known springs amounts to more than 200 according to information obtained from IGN maps. However, the number of active springs compared to those that have disappeared is not well known. This work presents the results obtained as a first approach. The correlation level is higher than 90%, and this methodology could be extrapolated to other areas.

Keywords: Climate change, Representative Concentration Pathway (RCP), Global Circulation Models (GCM), Machine Learning (ML), springs, hydrogeology.
Top