We analysed temporal variations of trace element concentrations in groundwater from a 101 m-deep borehole (HA01) in northern Iceland during 2010–2018 and compared them with seismic and volcanic events that occurred in the same period to identify potential hydrogeochemical precursors. An increase of B, Al, V, Li and Mo concentrations started from eight months to one month before the 2014 Bárðarbunga eruption (~115 km from HA01), a major rifting event in central Iceland, while Ga and V concentrations began to increase one day and one month after the onset of the event, respectively. We also found that concentrations of some trace elements (Li, B, Ga, Mo, Sr, Rb and Fe) significantly increased before an Mw 5.0 earthquake that occurred ~80 km from the borehole in 2018. However, other notable hydrogeochemical changes were detected during the monitoring period without apparent correlation with the seismic and volcanic events in the region. This study shows that the systematic long-term hydrogeochemical monitoring in seismic and volcanic areas is critical to advance the science of seismic and eruptive precursors. Furthermore, the use of statistical tools, such as Principal Component Analysis (PCA) and Change Point (CP) detection can help identify the most useful chemical elements and validate the trend variability of those elements in the time series, reducing arbitrary choices of pre-seismic and pre-volcanic hydrogeochemical anomalies as potential precursors.
Na-carbonate (NaCW) waters are not concentrated in well defined areas, but usually widespread in areas where other water types (e.g., Ca-carbonate) are dominant. NaCW are the product of long-term water-rock interaction with dissolution of Na-silicates in presence of phyllosilicates, silica phases, and calcite. NaCW circulating in calcite-bearing sediments very probably have a Ca-carbonate parent with moderate to low PCO2, which changes its composition assuming increasing Na character as the water-rock interaction proceeds. However, the moderate to low PCO2 values of the potential parent Ca-carbonate waters do not account for the high Na content of many NaCW. The higher PCO2 required may be due to oxidation of organic matter of the sediments and, perhaps, to further addition of CO2 coming from deeper crustal levels, from the mantle, or from other sources. Na-Ca exchange involving a Na-exchanger could be an alternative genetic hypothesis. At present, however, at least for some areas (e.g., Northern Apennines, Italy), this hypothesis is not supported by mineralogical evidence.
