Next Article in Journal
A Multi-Method Approach to Geophysical Imaging of a Composite Pluton in North Portugal
Previous Article in Journal
Potentially Toxic Elements (PTEs) Dispersion in Alluvial Deposits from Abandoned Mining Sites
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Minerals
Volume 14
Issue 4
10.3390/min14040338
minerals-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Fluorite and Gibbsite Solubility Controls the Vertical Transport of Fluoride and Aluminum during Rainwater Percolation through Ashfall Deposits in La Palma (Canary Islands, Spain)

by
Javier Sánchez-España
1,*,
Ana M. Nieto Castillo
2,
M. Pilar Mata
3,†,
Javier Martínez-Martínez
4 and
Jose F. Mediato
5
1
Department of Planetary Geology and Habitability, Centro de Astrobiología (CAB, CSIC-INTA), Torrejón de Ardoz, 28850 Madrid, Spain
2
Geochemistry Lab, IGME-CSIC, Calera 1, Tres Cantos, 28760 Madrid, Spain
3
Geological Repository, IGME-CSIC, Calera 1, Tres Cantos, 28760 Madrid, Spain
4
Department of Geological Resources, IGME-CSIC, Calera 1, Tres Cantos, 28760 Madrid, Spain
5
Department of Geological Resources, IGME-CSIC, Rios Rosas 23, 28003 Madrid, Spain
*
Author to whom correspondence should be addressed.
†
Deceased 19 February 2024.
Minerals 2024, 14(4), 338; https://doi.org/10.3390/min14040338
Submission received: 16 January 2024 / Revised: 5 March 2024 / Accepted: 21 March 2024 / Published: 26 March 2024
(This article belongs to the Section Environmental Mineralogy and Biogeochemistry)
Browse Figures
Versions Notes

Abstract

:
This study addresses the in situ mobility of fluoride and aluminum in two different ashfall deposits accumulated during the 2021 eruption of the Tajogaite volcano (La Palma, Canary Islands, Spain), which were exposed to contrasting conditions of ambient humidity and precipitation. We selected one site to the east of the volcanic emission center, located near the top of Cumbre Vieja Ridge and exposed to continuous humidity and rain, and another site to the west of the volcano situated in a lowland and characterized by much drier conditions. The mobility of fluoride and aluminum is markedly different at both sites, with the first sequence suggesting a downwards migration of Al and F, and the second sequence showing no sign of mobility. The migration of aluminum and fluorine results from the dissolution of different fluoride salts (mostly AlF3 and CaF2, as suggested by scanning electron microscopy) followed by vertical transport as ionic complexes (AlF3, AlF2+, AlF4−) during the percolation of rainwater through the ashfall deposits. Geochemical calculations suggest that the mobility of fluorine at neutral to alkaline conditions (pH 7.0–9.0) is likely limited by the solubility of fluorite (CaF2), whereas at slightly acidic conditions (pH < 6.5), the aqueous concentration of aluminum seems to be controlled by the solubility of gibbsite (Al(OH)3). This study demonstrates that aluminum and fluoride can be transported from volcanic ash to the underlying soil or groundwater, which is an environmental concern that should be followed in the future.

Graphical Abstract

1. Introduction

The eruption that occurred in the Tajogaite Volcano (Cumbre Vieja Ridge, La Palma, Canary Islands, Spain) between September and December 2021 released massive amounts of tephra and lava flows, which posed a serious threat for the local population [1,2,3] (Figure 1). An impressive eruptive column of tephra rose heights of 2400 to 4600 m a.s.l. (8500 m on some specific days), and lava flows traveled downslope to the west of the new vent (Figure 1). The most important accumulations of ashfall occured towards the west and southwest of La Palma island due to the direction of the prevailing winds. The eruptive column included, in addition to ashfall particles, volcanic gases such as sulfur dioxide (SO2), which was released at rates of 6000–11,500 tons SO2 in the initial stages of the volcanic eruption, reached peak values of 50,000 tons/day, and fluctuated between 900 and 40,000 tons per day in October and November 2021 [2]. Other gases like HF and HCl were also abundant in the eruptive column, though no information exists on emission rates for these gases.
Civil authorities (i.e., the La Palma Cabildo) were soon concerned about the possible consequences of the accumulation of thick sequences of volcanic ashfall (estimated to be around 8–9 million of m3 only during the first month [2]) in a vast portion of the island. The priority targets to protect from the potentially harmful effects of the ashfall were (1) human health, (2) infrastructure, (3) agriculture (e.g., banana plantations), and (4) natural ecosystems. The first experimental studies conducted by an IGME-CSIC research team [4] used a series of ashfall leachate tests to investigate the mobility and potential toxicity of different elements as a natural consequence of ashfall leaching by rainwater. This type of study provides a fast and reliable proxy to the leaching of freshly deposited ashfall by rainwater, and therefore, they have become a key tool to evaluate the potential release of volcanogenic toxic elements to the environment during rain episodes after eruptive processes [5,6,7,8,9].
Basic questions addressed in that first study by our group included (1) whether the ashfall accumulated in La Palma during the 2021 eruption could be a suitable fertilizer (e.g., for banana plantations), or (2) if, on the other hand, the elements released by the ashfall during rainy periods could represent a risk for the population (e.g., effects on water supply) or the environment (e.g., effects on aquatic ecosystems, vegetation or soils). The cited study concluded that the most evident and potentially dangerous effect of the ashfall leaching is the release of aluminum and fluoride [4]. These two elements (which travel together in the form of aluminum fluoride complexes) are known for their separate and combined toxicity and may inflict deleterious effects to natural ecosystems, agricultural soils, streams, and groundwaters, as well as to human safety, although the toxicity of these pollutants depends on many different factors, such as element (Al, F) concentration, route, dose, and time of exposure, etc. [4,10,11,12,13,14,15]. The mobility of aluminum and fluoride is therefore of concern for the island authorities and may influence decision making as regards the management of ashfall deposits accumulated in the island during the volcanic eruption, as well as the design of appropriate and effective measures of environmental protection.
However, the existing knowledge on Al and F mobility in our previous work was based on leaching tests conducted on 34 surface ashfall samples taken in different parts of the island, including roads, trails, or soils. These samples were seen to vary significantly in composition and grain size depending on the sampling site (e.g., fine-grained ashfall is usually more abundant in more distal areas, whereas coarse-grained ashfall predominates in proximal zones) and on the moment of the volcanic eruption (e.g., higher F and Al concentrations in the ashfall erupted in the first stages when the volcanic plume had higher contents of acid volatile phases like HF, HCl and SO2, and lower contents of these elements were present in the final stages of the eruption [4]). Moreover, the sampling of that study was conducted during the first weeks of the violent eruptive episode while the emergency situation and travel restrictions in the island were at their maximum; therefore, it only included surface samples taken from different sites that could be visited. Thus, it is still unknown how aluminum and fluoride may have behaved in thick sequences of ashfall accumulated in different parts of the island during the rainy periods that followed the volcanic episode, and if this mobility is homogeneous throughout the island or, on the other hand, if it shows differences depending on factors like (i) the local humidity degree or (ii) amount of precipitation received per unit of surface area. In fact, despite the abundance of studies assessing the mobility of potentially toxic elements from volcanic ashfall through leaching tests, the monitoring of the actual in situ mobility of volcanically derived toxic elements like fluoride or aluminum some time after volcanic eruptions is commonly lacking; therefore, the vertical migration of these elements through ashfall deposits and underlying soils is not usually known. Furthermore, the geochemical controls existing on this mobility of pollutants in the environment, like, for example, pH or mineral solubility, are usually not addressed at all.
In order to solve these questions, which are of clear practical interest, the present manuscript is focused on the mobility of aluminum and fluoride in two separate ashfall sequences studied in two different parts of the island with contrasting humidity and precipitation regimes (Figure 2). The two sampling sites, situated to the east and west of the Tajogaite volcano (Figure 2), were selected to compare the local-scale mobility of fluoride and aluminum under (i) humid conditions in a zone with relatively abundant precipitation (exemplified by the Enrique sequence to the east of Tajogaite) with (ii) the mobility of these two elements occurring under drier conditions in an area with more scarce precipitation (illustrated by the Cementerio sequence). The Enrique sequence is located at the Cumbre Vieja Ridge, a north–south elongated polygenetic volcanic range, and it is representative of natural conditions with autochthonous flora and fauna (e.g., pine trees), as well as an aquifer reception area. The Cementerio sequence is a lowland in the agricultural zone and more anthropized (e.g., abundant wells to extract groundwater for irrigation).
In addition, the studied ashfall sections represent the two main volcano-sedimentary deposit types formed in the island from a granulometric point of view. These physical differences in the volcanoclastic deposits are due to the existence of prevailing northeast winds, which cause preferential ash and lapilli deposition in the southwest side of the island. Only occasionally do southwest winds move the pyroclastic plume to the east. The Enrique and Cementerio sequences are located at a similar distance to the emission cone, but one of them is located southwest (Cementerio), and the other one is located northeast (Enrique).
The working hypothesis of this study was that these two climatically different zones would provide a representative window of environmental conditions in which to compare the behavior of these two potentially toxic contaminants. This more locally focused study will allow for a more rigorous assessment of the polluting potential of recent volcanic ashfall in La Palma. We believe that our sequence-specific approach could be valid and applied in other volcanic areas affected by ashfall deposition, particularly those containing environmentally hazardous pollutants and where sharp climatic contrasts between different sites could make any homogenous or general management action or remediation measure without previous on-site validation invalid.

2. Climatic, Geographic and Geological Framework

La Palma island features a warm–temperate subtropical climate (Cfb, according to the Köppen classification) characterized by mild winters, moderate summers, and precipitation distributed across the seasons, with a drier period in summer. This climate is influenced by trade winds and the local topography, bringing about distinct microclimates around the island. Western areas experience milder temperatures and reduced precipitation, whereas mountainous regions and the eastern side are cooler and wetter due to the Atlantic fronts that enter from the north of La Palma island [16]. The Foehn effect is produced on the Cumbre Nueva and Cumbre Vieja; the cold and humid northeasterly trade winds find the mountain range in their way and are forced to rise; and as the temperature decreases with height, the moist air becomes saturated and condenses to form clouds that deposit precipitation on the mountain [17].
The basic geological features of the Canary Islands in general, and those of La Palma in particular, have been summarized in [4] and can be found elsewhere [18,19,20]. In short, the Canary Islands are large independent volcanic edifices whose emerged part constitutes the top of a pile of submarine materials that, as a whole, can reach more than 5–6 km in height from the sea bottom, which is formed by 150–180 Ma old (i.e., Jurassic) oceanic lithosphere. La Palma island is the north-westernmost and (together with El Hierro) youngest island of the Canary Archipelago (Figure 2). The highest peak, with 2426 m a.s.l., is located at the top of Taburiente Caldera in the Roque de los Muchachos Peak. The Upper Pleistocene–Holocene volcanic rift in the southern part of the island (the Cumbre Vieja Ridge), with a north–south direction, is where all the historical eruptions have occurred and where the most recent volcanic activity is concentrated [18,19,20].
The ashfall deposits studied in the present paper were ejected from the Tajogaite Volcano in the eruption that happened between 19 September 2021 (14:10 UTC) and 13 December 2021 (22:21 UTC), which is the longest one in the historical record of eruptions in La Palma Island [21]. The eruption at Tajogaite was of a fissure-like and Strombolian type, with some phreatomagmatic phases that expelled lava flows in the more effusive phases that reached the western coast of the island, forming two lava deltas and destroying hundreds of houses, crops, and industries in an area of around 1200 Ha. Alternating with these effusive phases, there were more explosive stages that generated eruptive columns reaching heights up to 8500 m a.s.l. [1]. These columns travelled to the west, transported by the prevailing winds, and formed thick deposits of tephra, which covered the whole island, also reaching the islands of El Hierro, La Gomera, Tenerife, and Gran Canaria due to the prevailing winds [4]. The thickness of these tephra deposits was ~1 m in many places throughout the southwestern side of the island, although it reached maximum thicknesses of 2.7 m around the volcanic cone (1 km around). The most affected municipalities were those of El Paso and Los Llanos de Aridane, where tephra was the first volcanic material removed from streets and roads by the island authorities.
Tephra deposits alternate between millimetric and centimetric layers of lapilli, with ashes that have decreasing thickness and grain size towards more distal parts of the volcano. The chemical composition, mineralogy, and grain size distribution of these ashes are beyond the scope of this work and have been already reported in previous studies [4,22]. Briefly, the tephra and ashfall of the Tajogaite eruption included abundant particles of olivine (<80% forsteritic), clinopyroxene (augite, Ti-augite), magnetite, titanomagnetite, bytowinitic plagioclase and kaersutite-type amphibole, in addition to minor F-apatite crystals. In addition, the ashfall samples have been observed to contain significant amounts of micron- to submicron-size soluble salts, including Na- and K-rich chlorides (e.g., halite, carnallite), Ca-sulfates (mostly gypsum) and fluorides (NaF, CaF2, and Al-rich fluorides), which are interpreted to result from post-eruption processes like adsorption and/or the interaction of acid volatile phases (SO2, HF, HCl) with the volcanic glass contained in the ashfall particles of the eruptive plume [4]. Among the fluorides, crystalline aggregates of AlF3 and CaF2, with sizes of a few microns and varied morphology, have been detected in many samples.

3. Materials and Methods

3.1. Sampling of Ashfall Deposits

The most representative volcano-stratigraphic levels from the Enrique (ENR) and Cementerio (CEM) deposits were sampled. First, a detailed volcano-stratigraphic description was provided for both columns, outlining the sequence of tephra layers and their corresponding thickness (Figure 2). The different layers were defined according to (1) particle size, discerning between fine, medium, and coarse ash and fine and coarse lapilli; (2) grain-sorting sequences observed within each layer; (3) color; and (4) predominant particle shape, distinguishing between rounded, needle-like, and irregular particles.
The total thickness studied was 23 and 37 cm in the Enrique and Cementerio sites, respectively. Twelve samples were obtained from the Enrique sequence and eighteen samples were taken from the Cementerio sequence. In both cases, samples were obtained from the most significant layers (according to textural and petrological criteria), and they were labeled with increasing sequential number from the bottom to the top (Figure 2). Sampling was carried out with a spatula and a brush, avoiding sample contamination from adjacent levels. Samples were bagged, labeled, and stored for later chemical and mineralogical analysis in the lab (see next section).
The meteorological variability occurred in La Palma during the months of the eruptive period (September–December 2021) is schematically represented in Figure 3. Different wet and dry periods alternated, showing an irregular distribution of humidity on the island with the preferential concentration of rain and wet microenvironments on the center and eastern side of the island. This irregular distribution is characteristic of La Palma island and is associated with the existence of preferred wind directions (from northeast–east) and the north–south alignment of the mountain ridge that determines the orography of the island.
Figure 3 also includes information about the eruptive evolution of the Tajogaite volcano [1]. The evolution of the explosiveness and in the type of emission determines the type and characteristics of the layers sampled in the CEM and ENR volcano-stratigraphic sequences.

3.2. Ashfall Leaching Tests

The leaching tests followed the recommendations originally summarized by [5] and later revised by [9]. These recommendations are also available from the International Volcanic Health Hazard Network (IVHHN) (downloadable from https://www.ivhhn.org/, accessed on 24 December 2021). In short, the ashfall leaching procedures followed in this study included (i) extractions conducted at a ratio of 1:25 (3 g ash in 75 mL ultrapure, deionized MilliQ water with resistivity > 18 MΩ.cm), (ii) shaking of the flasks gently by hand for 10 s, (iii) agitation of the samples for 90 min using an automatic shaker, (iv) settling of the obtained suspensions for 10 min, (v) determination of pH and conductivity in an unfiltered aliquot of the leachates immediately after the tests, (vi) filtering of samples through 0.45 µm Millipore membrane filters, and (vii) collection of separate aliquots in HDPE bottles for cation and anion analysis, as described below. The leaching tests were always conducted in closed, trace-metal-free flasks at room temperature and under atmospheric conditions.
We conducted a total of 30 leaching tests in samples taken from the two selected ashfall deposits (18 samples from Cementerio and 12 samples from Enrique taken at different depths to cover the whole columns; Figure 2). The objective of this sampling strategy was to compare the amount of adsorbed, water-soluble, and Al- and F-containing salts in different parts of the ashfall deposits, which could then be used to infer possible trends of fluoride and aluminum mobility within these deposits as a result of repeated washing by rainwater. The leaching tests were conducted at a pH of 5.3, which is the original value of the MilliQ water used in the experiments. Two blanks consisting of pure MilliQ water without any ashfall were also subjected to the same protocol as the rest of samples and analyzed at the same time in order to correct any analytical artifact in the calculations (e.g., introduced by filtration, shaking of the flasks, etc.).

3.3. Chemical Analyses of Leachates

The final pH of the leachates was measured immediately after the leaching tests with a CRISON micropH 2001 instrument (Crison Instruments, Barcelona, Spain) calibrated with pH 7.0 and pH 4.0 buffers. Conductivity was also measured with the same equipment in a few selected leachates to evaluate the increase in dissolved solid content after the leaching tests.
Ion concentrations in leachates were measured by ion chromatography using an 881 Compact IC Pro-Metrohm chromatograph for major anions and by continuous flow analysis (CFA) using an Alliance-Futura analyzer for major cations. Trace elements were analyzed by inductively coupled mass spectrometry (ICP-MS) using an Agilent 7900 (Agilent Technologies, Santa Clara, CA, USA). Aluminum and strontium were measured by inductively coupled plasma atomic emission spectrometry (ICP-AES) in Agilent 5800 (Agilent Technologies, Santa Clara, CA, USA) Quality control was assessed by participation in the international proficiency test Aquacheck program (Laboratory of the Government Chemist, LGC, Teddington, Middlesex, UK). Certified reference materials were analyzed during routine laboratory work on a regular basis. The accuracy obtained during chemical analyses was in the range of 90%–110% for most of the studied elements.

3.4. Geochemical Modelling

Calculation of saturation indices (SI) of selected mineral phases was conducted with PHREEQCI (Version 3.0.5-7748, US Geological Survey, Reston, VA, USA [23]). The goal of these calculations was to infer possible mineral phases acting as solubility controls on fluoride and metal mobility during water/ashfall interaction. The calculations were conducted using the MINTEQA2 4.0 database (USEPA, Washington, DC, USA [24]). The concentrations of the different elements (e.g., Al, F, Ca, SO4) in the leaching solutions were input to the program. In the particular case of calcium, in the samples where this element was below detection (<1 mg/L), we selected an arbitrary concentration of 0.5 mg/L Ca to estimate maximum values of SI for Ca-containing minerals (fluorite and gypsum). The saturation indices were calculated using the solubility products (log Ksp values) included in the MINTQA2.V4 database.

4. Results and Discussion

4.1. Vertical Profiles of pH, Sulfate, and Bicarbonate Concentration

The pH, conductivities, and major ion concentrations obtained in the leachates from the different ashfall samples are given in Table S1 (Supplemental Material), and the most relevant elements (pH, SO42−, HCO3−, Al, F) have been graphically represented as vertical profiles through the ashfall sequences in Figure 4, Figure 5 and Figure 6 to evidence their geochemical behavior during the leaching tests.
The pH profiles obtained for the ENR and CEM ashfall sequences (Figure 4A,D) show a general trend of decreasing pH with increasing depth within the sequence. The pH obtained near the surface of the ENR deposit was 8.8 and decreased to values of ~6.2 at 20 cm below the surface (Figure 4A). In CEM, an upper layer in the first 3 cm yielded a pH of 6.1–6.2, below which a sharp increase (to pH 7.7 at 5 cm) occurred, and then a smooth but continuous decrease to values of 6.1 at 36 cm was again evident (Figure 4D).
It is worth noting that these pH profiles do not represent actual pH values of pore waters within the ashfall sequences (which were both extremely dry and did not contain any interstitial water at the time of sampling), but the pH resulting from leaching tests of the ashfall samples taken at different depths (Figure 2). Thus, considering that the starting pH in all cases was 5.3, the profiles shown in Figure 4 show a maximum alkalinizing capacity of the uppermost (and therefore younger and more recently deposited) ashes and a general trend of decreasing alkalinizing capacity with depth as the ashes were progressively aged and compacted upon burial below newly deposited ashfall.
Previous leaching tests by our group have already demonstrated that the ashfall from the Tajogaite volcano usually shows a certain alkalinizing capacity, which may induce significant pH increases in the leaching solutions (e.g., from 5.3 to 5.5–6.5 [4]). In the apparent absence of alkaline minerals such as calcite or dolomite (which were not to be found by XRD analyses nor by SEM investigations in the cited paper), this alkalizing capacity was attributed to the partial dissolution of volcanic glass (Equation (1)), which takes place at elevated aqueous fluoride concentrations due to the formation of aqueous Al-fluoride complexes that increase aluminosilicate solubility [4,25,26]. This partial dissolution of volcanic glass is a proton-consuming reaction, which may lead to pH increases in the leaching solutions, according to the reaction:
SiAl0.3O2(OH) + H+ + H2O → H4SiO4 + 0.3Al3+
In this study, however, the correlation between SiO2 and Al concentrations for the whole data set of solutions including both sequences is very poor (R2 = 0.31; Figure S1), which is clearly influenced by the high amount of samples with SiO2 concentrations close to or below the limit of detection (0.1 mg/L).
On the other hand, the geochemical data obtained in this study suggest that the ashfall samples taken in the ENR and CEM profiles may include a significant amount of carbonates like calcite or dolomite, which could have exerted some influence on the pH profiles. This is evidenced in the profiles of the bicarbonate (HCO3−) concentration shown in Figure 4B,E (also given in Table S1), which apparently suggest the release of significant HCO3− during leaching in a considerable number of samples. In particular, the concentrations of bicarbonate in some of the samples reached relatively high values (e.g., 8–15 mg/L HCO3− in ENR, 12–37 mg/L HCO3− in CEM), which are far above the background values obtained in the blank samples (4 and 1 mg/L HCO3−, respectively; Figure 4B,E) and suggest that a significant amount of carbonate minerals like calcite or dolomite may have been partly dissolved during the leaching tests. The detection of significant quantities of calcium and magnesium in the resulting leachates (<1 to 6 mg/L Ca, 1–5 mg/L Mg) would also be coherent with this idea, although these major cations show a poor correlation with bicarbonate (R2 = 0.17–0.23, not shown). These cations could also be partly derived from partial dissolution of some other minerals such as sulfates like gypsum (Ca) or aluminosilicates like plagioclase, amphibole, or pyroxene (Ca, Mg).
The apparent presence of carbonates in the studied ashfall sequences is significantly different with respect to our previous study, which did not detect any trace of carbonate minerals by XRD or SEM investigations [4]. The explanation for this difference is likely related to the age of the studied ashfall, which consisted of freshly deposited ashfall (hours or days after having been expelled from Tajogaite volcano) deposited on soil surfaces in our first study but comprise older ashfall samples (weeks to months) in thicker sequences in the present paper. This is because carbonate minerals are not natural components of volcanic ashes but are usually transported by aeolian dust from the Sahara Dessert to soils in the Canary Islands [27]. Thus, the ashfall sequences sampled in ENR and CEM sites would have been affected by the atmospheric deposition of allochthonous particles (including carbonates) that would have been admixed with the ash particles. Thus, the resulting pH of the obtained leaching solutions after the interaction of deionized water with the ashfall particles seems to have been a combination of at least two major geochemical processes, including (1) aluminosilicate-rich glass dissolution (e.g., Equation (1) [25]), and (2) carbonate dissolution (exemplified by Equation (2)), which would also tend to produce an increase in pH in the leaching solutions [28].
CaCO3 + H+ →Ca2+ + HCO3−
The profiles of sulfate (SO42−) concentration obtained in the leachates of both ashfall sequences do not show any clear trend (Figure 4C,F) but displayed generally significant concentrations, which were notably higher in the CEM sequence (3–51 mg/L SO42− in CEM vs. 2–22 mg/L SO42− in ENR; Figure 4C,F and Table S1). These concentrations suggest a significant dissolution of sulfates, which, based on the obtained leachate chemistry (Table S1), would include Na-, Ca-, and Mg-rich sulfates. These minerals have been already described as common efflorescent salts in the vicinity of the Tajogaite volcano [29], with gypsum (CaSO4·2H2O) being especially common [4] (see also Figure S2). Sulfate was usually the dominant ion in the majority of the leaching solutions (Table S1), and therefore the one contributing the most to the conductivity of these solutions, as evidenced in binary plots that show a fairly good correlation between SO42- and conductivity (R2 = 0.77; Figure S3A), which is much better than the one displayed by bicarbonate (R2 = 0.39; Figure S3B). In any case, despite their different relative contribution, it is clear from Figure 5 (which considers the sum of [SO42− + HCO3−] and shows a very strong correlation of R2 = 0.91) that conductivity in the analyzed leachates is chiefly controlled by the concentration of these two ions, which were by far the most abundant in the solutions (Table S1).

4.2. Aluminum and Fluoride Profiles

The profiles obtained for water-extractable aluminum and fluoride in the two ashfall sequences are displayed in Figure 6. In the ENR ashfall sequence, and despite some internal variability between adjacent layers, a general trend of increasing concentration with increasing depth is evident for both aluminum and fluoride (Figure 6A,B). Aluminum concentration in the resulting leachates increased from low values of 2–31 µg/L Al in the top layer of the sequence to values of a few hundred (126–650 µg/L Al) in the lower part and, finally, a peak value of 4733 µg/L Al in the bottommost sample (Figure 6A and Table S2). Similarly, fluoride concentration increased from values <1 mg/L F in the top layer of the sequence to a peak value of a 16 mg/L in the bottommost sample (Figure 6B and Table S1).
On the other hand, the results for the CEM ashfall sequence did not show a clear vertical trend as regards the water-extractable content of these two elements (Figure 6C,D; Tables S1 and S2). In this case, a topmost layer (upper 6 cm of the sequence) displayed lower concentrations of aluminum (85–875 µg/L Al) and fluoride (1–3 mg/L F) in the resulting leachates, compared to the rest of the sequence (284–3953 µg/L Al and 2–14 mg/L F in the ashfall samples from the remaining 30 cm of the sequence; Figure 6C,D).
Overall, the aluminum and fluoride concentrations measured in the leachates of both ashfall sequences are notably lower than those measured in ashfall samples taken directly from soil surfaces in the days and weeks immediately following the onset of the volcanic eruption, between 25 September 2021 and 17 October 2021 (maximum concentrations of 28 mg/L Al and 67 mg/L F−, average values of 8 mg/L Al and 25 mg/L F− [4]). In fact, as shown in Figure 7, with the exception of sulfate, the concentrations obtained in the leaching tests of this study for most major ions (e.g., F−, Cl−, Al, Ca, Na) were always much lower than those previously found in leaching tests conducted in our first study on recently deposited ashfall samples [4]. These results suggest that fluorides and chlorides would have been more soluble than sulfates under the conditions typical of rainwater (usually pH 5.5–6.0). Consequently, their respective ionic constituents would have travelled further downwards through the ashfall sequences with respect to other, less mobile elements like, e.g., Si, Mg, C, or N (Table S1).
This marked difference in the water-extractable concentration of most major ions between the studied ashfall sequences in the ENR and CEM sites and the ashfall samples collected in September–October 2021 is likely the result of different factors—(1) firstly, the marked decline in the water-extractable concentration of major ions is a natural consequence of the decreasing content of adsorbed salts (chlorides, sulfates, fluorides) in the expelled ashfall as the eruption progresses, which, in turn, is the result of a decreasing content of acid volatile phases (HCl, SO2, HF) in the eruptive column due to chemical fractionation in the underlying magmatic chamber [6,13]; (2) secondly, the major ion concentration in the leachates from the 2021 samples could partly result from an overall lower pH of the fresh ash leachates due to abundant sulfuric acid droplets on the ash, which would have totally disappeared in the washed ashes of ENR and CEM sites; and (3) thirdly, the low content of adsorbed salts in the ENR and CEM ashfall profiles in comparison to the firstly erupted ashfall likely reflects the continuous washout of salts in ashfall exposed to atmospheric agents after successive rainy periods. The rainfall infiltrated in the ashfall sequences could have dissolved (totally or partially) the soluble salts existing in the upper layers, transporting these solutes through the ashfall sequence to the bottommost layers or even the underlying soils. The decline in water-extractable concentration of major ions (e.g., Al, F), as a consequence of the lesser content of adsorbed salts (chlorides, sulfates, fluorides), was already evidenced during the first month of the eruption [4]. With respect to ashfall leaching by infiltrated rainwater, this seems to be also suggested by the vertical profiles of the Enrique sequence, which show a general trend of increasing fluoride and aluminum concentration with depth, which may have been caused by a downwards vertical flow path of these two solutes as a result of the dissolution/precipitation processes (Figure 6A,B). This trend, however, is not observed in the CEN sequence, which is ascribed to the lack of rainfall discharge received in the westernmost and dry side of the island (Figure 3). In this case, the differences in aluminum and fluoride concentration between successive layers within the Cementerio ashfall sequence could just reflect the natural variability of salt content in the ashfall deposited in different moments during the eruptive episode. Unlike the observations of our first study [4], a “grain size effect” (e.g., higher Al and F contents in fine-grained and more reactive ashes, and lower contents in coarse-grained ashes) was not observed in the vertical trends. The poor sorting observed in the ENR sequence, where fine ashes alternate with coarse ashes and coarse lapilli (Figure 2), does not match with the vertical trends of increasing Al and F concentration with depth in this sequence (Figure 6A,B). This observation strongly suggests a clear post-depositional origin (i.e., downwards migration of fluoride and metals via infiltration and percolation of rainwater and the dissolution/precipitation processes) of the observed vertical trends.
In any case, a strong correlation exists between the fluoride and aluminum molar concentrations in the ash leachates of both sequences (R2 = 0.95 in ENR, R2 = 0.98 in CEM; Figure 8), which indicates a close chemical association between these two elements in the whole concentration range. This close correlation reflects (1) that both elements are derived from the dissolution of the same mineral carriers (aluminum fluoride salts present in the ash samples), and (2) that once incorporated into the aqueous phase, these two elements show a strong tendency to form stable aluminum fluoride complexes (e.g., AlF3, AlF2+, AlF4−) and thus remain chemically linked during transport [30]. With respect to the possible precursor mineral phases, the slope of the regression lines in both plots of Figure 8 is around 0.2, which represents a molar proportion of [F/Al] = 5. This molar ratio could be derived from the combined dissolution of mineral phases like AlF3 and CaF2 (which have both been found in the studied samples, as discussed in a later section). The correlation between Ca and F, however, is very poor (R2 = 0.015 in the CEM sequence; see Figure S4 in supplemental material). A plausible explanation for the lack of correlation between Ca and F is that, unlike F and Al, which have a purely volcanic origin and have not experienced post-depositional inputs (e.g., from aeolian dust), calcium has been probably introduced by external, non-volcanic sources during the time span covered in this study; therefore, its stoichiometric relation with F does not reflect the original volcanic source.
The results obtained in this study can be used to estimate the amount of contaminants potentially released to the underlying soils by rainwater/ashfall interaction during the one-year period elapsed between our first study [4], where we found very high concentrations of adsorbed elements in freshly deposited ashfall during the initial stages of the volcanic eruption and this work. A very simplistic calculation could consider the total amount of ash accumulated in La Palma during the Tajogaite eruption (estimated to exceed 10 million of m3 [1,2,3]), a density range of 400–700 kg/m3 for the dry ashes, and the difference between the average value of adsorbed Al and F in the fresh ashes of 2021 and the average of adsorbed Al and F in the aged (washed) ashes analyzed in this study (see Figure 7 for comparison). This calculation suggests that around 2500 to 4500 Mt (metric tons) of fluoride and between 720 and 1260 Mt of aluminum (depending on the ash density considered) could have been released to the environment during one year of the washing of ashfall in La Palma. The calculations for other major elements would include, for example, 2200–3800 Mt of chloride or 1500–2600 Mt of calcium. This, of course, is an oversimplification that ignores many aspects like (1) the spatial variability of ashfall composition, grain size, density, and chemical reactivity [4]; (2) geographic variation in precipitation volume across the island (Figure 3); (3) the accumulation of aeolian dust or clay minerals that could have acted as impermeable barriers avoiding or limiting infiltration, or (4) the partial removal of ashfall from roads or urban areas by civil authorities. However, this gross estimation may help visualize the magnitude of the potential accumulation of these pollutants in the soils, plants, or aquifers of the island.

4.3. pH Control of Aluminum and Fluoride Concentrations

Another clear difference observed between the ENR and CEM ashfall sequences refers to the pH control of aluminum and fluoride concentrations. The correlation between these two elements and pH is reasonably good in the case of the ENR leachates (R2 = 0.75–0.79; Figure 9A,B), whereas it is non-existent in the case of the leachates from CEM (R2 = 0.006–0.011; Figure 9C,D). This fact is considered to be another part of the indirect evidence of the marked difference between the two ashfall sequences as regards the extent of reactive transport of solutes by infiltrated rainwater. In the ENR ashfall sequence, abundant rainfall and humidity conditions appear to have favored a reactive flow path, whereby soluble aluminum fluorides (and other salts) would have been dissolved in the upper (more recent) ashfall layers, and the resulting F− and Al3+ ionic complexes incorporated to the interstitial solutions would have percolated downwards and precipitated in the lower layers as a direct consequence of pH-controlled solubility equilibrium (this is discussed in more detail in Section 4.5).
On the other hand, the extremely dry conditions prevailing in the CEM sampling site would have precluded a continuous percolation of rainwater, and this would have, in turn, prevented a noticeable vertical transport of aqueous aluminum and fluoride through the sequence. Thus, the fluoride and aluminum content of ashfall samples in the CEM sequence would not be controlled by pH nor enhanced by fluid circulation. Instead, and as stated previously, differences of fluoride and aluminum concentration between successive layers in the CEM sequence are likely introduced by the natural variability of the ashfall composition.

4.4. Al- and Ca-Fluorides Observed by SEM-EDX

As in the previous study on recently deposited ashfall on streets, roads, and soil surfaces [4], our SEM investigations have revealed the widespread presence of fluorides in the form of aggregates of fine-grained (submicron) crystals around glass and silicate particles (Figure 10 and Figure 11).
The aggregates are not monomineralic but more often include finely intergrown crystals of fluorite (CaF2) and aluminum fluoride (AlF3). Due to this close association of different fluorides, it is usually difficult to obtain “clean” semiquantitative chemical analyses of these aggregates by EDS (Figure 10C,D). The EDX spectra often reflect an average composition dominated by F, Al, and Ca (in accordance with the fluoride-rich mineralog) but may also include minor amounts of other elements like Na, Cl, Si, Fe, or Mg due to the presence of other halides (e.g., NaCl, MgCl2) or by contamination from adjacent glass or silicate mineral particles. The most efficient method to differentiate between different fluorides in these ash samples is with element mappings (as deduced by EDX) that evidence the dominance of one specific element in certain aggregates and which can then be used as diagnostic of specific minerals (e.g., Ca in the case of fluorite, Al in the case of AlF3; Figure 11B–D).
However, we could not confirm the presence of other aluminum-rich secondary minerals by SEM-EDX, such as aluminum hydroxide (e.g., Al(OH)3 (am), gibbsite), which are often found in Al-rich solutions at pH > 5.0 and are known to control aluminum solubility under the most natural conditions [31,32,33,34,35,36,37,38]. We did not detect any gypsum either, which was observed in the form of tabular (platy) crystals with diameters around 10–15 µm in the ashfall erupted from the Tajogaite volcano during the first days [4].
Overall, the SEM-EDX observations are coherent with the geochemical trends shown in Figure 6A,B and suggest that the mobility of aluminum and fluoride through the Enrique ashfall sequence must have taken place via dissolution of AlF3 and CaF2 crystals in the upper levels followed by downwards transport through the ashfall deposits during rainwater percolation.

4.5. Geochemical Modeling of Solubility Equilibrium through the Ashfall Profiles

The results of the geochemical modeling calculations are shown in Figure 12, which shows the variation in the saturation index for selected minerals (fluorite, gibbsite, gypsum) with respect to pH (Figure 12A,B) and also with respect to the concentration of major cations (Al, Ca; Figure 12C) and major anions (F−, SO42−; Figure 12D). The saturation index is defined as SI = log [IAP/Ksp], where IAP is the ion activity product, and Ksp is the known solubility product constant of a given mineral phase. The SI values provide useful information about the saturation state of the aqueous solutions (the leachates) with respect to given mineral phases. In this case, we selected mineral phases that have been widely observed in the ashfall samples of La Palma (e.g., fluorite, gypsum [4]) and others that are well-known solubility controls of Al transport in acidic waters (e.g., gibbsite [31,32,33,34,35,36,37,38]). The SI values plotted on Figure 12A,B were calculated on the basis of the element concentrations obtained from the leachate compositions (Tables S1 and S2). An exception had to be made for calcium, since this element was below detection (<1 mg/L) in most samples. In this case, we assigned an arbitrary concentration of 0.5 mg/L, which could at least allow for a rough estimation of the saturation state of Ca-containing minerals (i.e., fluorite and gypsum). Even if these SI values for fluorite and gypsum could have been slightly overestimated, they were still valid for comparison with the Al-containing phase gibbsite. Thus, the plots of Figure 12A,B show that the leachates were always strongly undersaturated with respect to gypsum (SIgypsum = −3 to −5) at the low Ca concentrations considered. Fluorite, however, showed near-equilibrium conditions (SIfluorite = −0.4 to −0.6, Figure 12A) for most samples in the ENR sequence and in some samples of the CEM sequence (Figure 12B). These results suggest that fluorite could have acted as an upper solubility limit for both F and Ca in the pH range of 7.0–9.0. This is in good agreement with our previous study [4] and with the known control of fluorine concentrations in natural waters, which is usually controlled by fluorite solubility [14].
On the other hand, gibbsite appears to be strongly saturated in this pH range in most samples (SIgibbsite = 2–3; Figure 12A,B). However, at more acidic conditions (i.e., pH < 6.5), some leachates plotted on the line of SI = 0, which indicates equilibrium, suggesting that this mineral likely acts as an upper solubility limit for Al3+ at lower pH, which is in line with many previous studies on the mobility of aluminum in natural waters [31,32,33,34,35,36,37,38]. This apparent gibbsite saturation could also imply a certain underestimation of the mobility of Al in the studied ashes since the concentration of this metal in some leachates could have been biased by the chosen ashfall/water ratio in the leaching experiments (1:25) due to limited salt dissolution under Al-saturated conditions. In any case, and despite the above stated limitations, the plots of Figure 12A,B strongly support the idea of different mineral phases exerting a solubility control at different pH conditions, with fluorite controlling the maximum concentrations of F− and Ca2+ at near-neutral conditions and gibbsite limiting Al3+ concentration in the percolating fluids at a lower pH. This is also illustrated in the plots of Figure 12C,D, which point to an apparent equilibrium with respect to fluorite solubility for calcium concentrations lower than around 5–10 mg/L Ca, as well as near-equilibrium with respect to gibbsite solubility for aluminum concentrations lower than around 1–2 mg/L Al (in both cases, the chosen conditions of the theoretical percolating fluids were pH = 7 and F− = SO42− = 10 mg/L; Figure 12C). Regarding the restrictions imposed by major anion concentrations, the plot of Figure 12D suggests that the precipitation of fluorite requires fluoride concentrations at least above 20 mg/L F− (for conditions of pH 7 and Ca = 1 mg/L). Interestingly, at these geochemical conditions, gibbsite precipitation would not take place at fluoride concentrations greater than around 30 mg/L F− (Figure 12D). The most plausible explanation for this apparent limitation is the competing effect of different mineral phases for the incorporation of Al3+ ions from the solution. Thus, the precipitation of Al-containing fluorides (e.g., AlF3) at higher F- concentrations would obviously decrease the availability of free Al3+ ions for the formation of gibbsite (Figure 12D). The undersaturation of gypsum in all leachates implies a small influence of this mineral on the mobility of calcium and sulfate.
The random position of points in the Cementerio sequence suggests the absence of clear mineral solubility controls. This is in line with the above-commented scarcity of aqueous fluids in this drier site, which would not have allowed for any significant water/mineral interaction nor the subsequent vertical transport and mobility of fluoride and aluminum in this sequence.
A final observation about Figure 12A refers to the relation of the SI trends in the ENR sequence with pH, which are parallel to the X-axis (hence pH-independent) for fluorite and gypsum but show a very good correlation (R2 = 0.99) for gibbsite. This difference is just a consequence of the precipitation reactions of these minerals, which do not imply the release or consumption of protons in the case of the former minerals (Equations (3) and (4)) but involves the release of protons in the case of the latter (as indicated in Equation (5)):
Ca2+ + 2F− → CaF2 (s)
Ca2+ + SO42− + 2H2O → CaSO4·2H2O
Al3+ + 3H2O → Al(OH)3 (s) + 3H+
The stoichiometry of Equations (3)–(5) also implies that the dissolution or precipitation of fluorite or gypsum does not have any effect on solution pH, whereas gibbsite will act as a buffer system in rainwater percolated through the ashfall profiles, acidifying the solutions upon precipitation and increasing the pH during dissolution.

5. Conclusions

Previous studies of ashfall leachate have already highlighted the importance of sampling across different tephra deposits and characterizing the stratigraphy and internal arrangement, mineralogy, and granulometry of the ashfall to ensure the quality and comparability of collected leachate data sets after eruptive episodes in volcanic regions [39]. The studied ashfall deposits at the Enrique and Cementerio sites in La Palma island offer an excellent opportunity to compare the in situ mobility of fluoride and aluminum, two potentially polluting and environmentally relevant elements, under different climatic conditions of contrasting humidity and precipitation. These two ashfall sequences were deposited during three months of explosive volcanic eruption in the newly formed Tajogaite volcano, which expelled millions of cubic meters of volcanic ashes (in addition to lava flows) that were transported by prevailing winds to different parts of the island and beyond. The mobility of both fluoride and aluminum has been shown to be markedly different at both sites, with the Enrique ashfall sequence showing evident signs of mobility and downwards migration of Al and F (travelling together as different ionic complexes like AlF3, AlF2+, or AlF4−), and the Cementerio sequence displaying no evidence of mobility. In the later sequence, the observed differences in Al and F concentrations at different depths more likely reflect a natural internal variability of the sequence as regards the content of Al- and F-containing salts in the successive layers of ashfall deposited in this site. Therefore, a major conclusion of this study is that fluoride and aluminum mobility from ashfall deposits to downstream sites or to underlying soils or groundwaters is restricted to sites where the presence of interstitial water (either derived indirectly from the condensation of local ambient humidity or directly as precipitation) has occurred more or less regularly during a given period of time.
The examination of vertical geochemical trends and saturation index calculations in the Enrique ashfall sequence suggests that the mobility of fluoride and aluminum is firstly controlled by the dissolution of different mineral carriers (fluorite, AlF3) in the uppermost levels and secondly by solubility equilibrium reactions that limit the maximum concentrations of dissolved fluorine (fluorite, CaF2) and aluminum (gibbsite, Al(OH)3) in the interstitial waters during downwards percolation. The aqueous content of fluorine and aluminum is also strongly pH-controlled, in agreement with previous studies [36,37,38,40]. Thus, the dissolution or precipitation of fluorite during ashfall/water interaction seems to dictate the aqueous content of fluoride at neutral to alkaline pH (7.5–9.0), whereas at slightly acidic conditions (pH < 6.5), the aqueous concentration of Al3+ in interstitial waters within the sequence seems to be determined by the solubility of gibbsite (Al(OH)3), which also acts as a chemical buffer due to the release of hydrogen ions during precipitation or consumption during dissolution.
An obvious concern that can be extracted from our study is that aluminum and fluoride cannot only migrate downwards through the ashfall deposits accumulated in different parts of the island but may also be transported from these recent volcanic deposits to the underlying soils or neighboring vegetal cover, groundwaters, or natural ecosystems. Aluminum and fluorine have been both reported to be highly toxic for plants, livestock, and humans [41,42,43,44,45], although the effect that these two volcanically derived pollutants may have on living organisms and natural resources will largely depend on several factors like element concentration, time and route of exposure, or tolerance to fluoride and/or aluminum toxicity. In either case, the environmental monitoring of naturally or economically valuable areas (e.g., natural parks, banana plantations, aquifers) that are still covered by ashfall is strongly recommended. Monitoring programs should focus on the possible accumulation of fluoride, aluminum, or both in the water resources and agricultural soils of the island.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/min14040338/s1, Table S1: Field parameters and major ion concentrations measured in leachates obtained in ashfall samples of the Enrique and Cementerio profiles; Table S2: Trace element concentrations measured in leachates obtained from ashfall samples of the Enrique and Cementerio profiles; Figure S1: Binary plot of SIO2 vs. Al content in leachates of the Enrique and Cementerio field sites. Figure S2: SEM images and EDX spectrum of gypsum crystals found in ashfall particles from the Enrique sampling site. Figure S3: Binary plots showing the correlation of sulfate and bicarbonate with conductivity for the whole data set of 30 samples. Figure S4: Binary plots showing the correlation of calcium and calcium plus aluminum with fluoride concentration in the Cementerio sequence.

Author Contributions

Conceptualization, J.S.-E.; methodology, J.S.-E., A.M.N.C., M.P.M., J.M.-M. and J.F.M.; validation, A.M.N.C.; formal analysis, A.M.N.C. and M.P.M.; investigation, J.S.-E., M.P.M., J.M.-M. and J.F.M.; data curation, J.S.-E. and A.M.N.C.; writing—original draft preparation, J.S.-E.; writing—review and editing, J.S.-E., A.M.N.C., J.M.-M. and J.F.M.; supervision, J.S.-E. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material; further inquiries can be directed to the corresponding author/s.

Acknowledgments

The field work and laboratory analyses for this work were internally funded by IGME-CSIC. The authors wish to thank the technicians at the IGME-CSIC laboratories for the sample processing and lab analyses. We sincerely thank the comprehensive and highly constructive comments of one of the anonymous reviewers, which have considerably helped increase the quality of the final version of this manuscript. Finally, we wish to dedicate this paper to our colleague, Pilar Mata, who passed away in February 2024.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. PEVOLCA. Actualización de la Actividad Volcánica en Cumbre Vieja (La Palma). Informe del Comité Científico del PEVOLCA del 25 de diciembre de 2021. Available online: https://www.gobiernodecanarias.org/infovolcanlapalma/pevolca/ (accessed on 1 July 2022).
  2. Bennis, K.L.; Venzke, E. (Eds.) Report on La Palma (Spain). In Bulletin of the Global Volcanism Network; No. 2; Smithsonian Institution, Global Volcanism Program: Washington, WA, USA, 2022; Volume 47. [Google Scholar] [CrossRef]
  3. Taddeucci, J.; Scarlato, P.; Andronico, D.; Ricci, T.; Civico, R.; Del Bello, E.; Spina, L.; D’Auria, L.; Asensio-Ramos, M.; Calvo, D.; et al. The explosive activity of the 2021 Tajogaite eruption (La Palma, Canary Islands, Spain). Geochem. Geophys. Geosyst. 2023, 24, e2023GC010946. [Google Scholar] [CrossRef]
  4. Sánchez-España, J.; Mata, M.P.; Vegas, J.; Lozano, G.; Mediato, J.; Martínez, J.; Galindo, I.; Sánchez, N.; del Moral, B.; Ordóñez, B.; et al. Leaching tests reveal fast aluminum fluoride release from ashfall accumulated in La Palma (Canary Islands, Spain) after the 2021 Tajogaite eruption. J. Volcanol. Geotherm. Res. 2023, 440, 107959. [Google Scholar] [CrossRef]
  5. Witham, C.S.; Oppenheimer, C.; Horwell, C.J. Volcanic ash-leachates: A review and recommendations for sampling methods. J. Volcanol. Geotherm. Res. 2005, 141, 299–326. [Google Scholar] [CrossRef]
  6. Bagnato, E.; Aiuppa, A.; Andronico, D.; Cristaldi, A.; Liotta, M.; Brusca, L.; Miraglia, L. Leachate analyses of volcanic ashes from Stromboli volcano: A proxy for the volcanic gas plume composition? J. Geophys. Res. 2011, 116, D17204. [Google Scholar] [CrossRef]
  7. Cabré, J.; Aulinas, M.; Rejas, M.; Fernandez-Turiel, J.L. Volcanic ash leaching as a means of tracing the environmental impact of the 2011 Grímsvötn eruption, Iceland. Environ. Sci. Pollut. Res. 2016, 23, 14338–14353. [Google Scholar] [CrossRef]
  8. Ruggieri, F.; Fernández-Turiel, J.-L.; Saavedra, J.; Gimeno, D.; Polanco, E.; Naranjo, J.A. Environmental geochemistry of recent volcanic ashes from Southern Andes. Environ. Chem. 2011, 8, 236–247. [Google Scholar] [CrossRef]
  9. Stewart, C.; Damby, D.E.; Tomašek, I.; Horwell, C.J.; Plumlee, G.S.; Armienta, M.A.; Hinojosa, M.G.; Appleby, M.; Delmelle, P.; Cronin, S.; et al. Assessment of leachable elements in volcanic ashfall: A review and evaluation of a standardized protocol for ash hazard characterization. J. Volcanol. Geotherm. Res. 2020, 392, 106756. [Google Scholar] [CrossRef]
  10. Cronin, S.J.; Manoharan, V.; Hedley, M.J.; Loganathan, P. Fluoride: A review of its fate, bioavailability, and risks of fluorosis in grazed-pasture systems in New Zealand. N. Zeal. J. Agric. Res. 2000, 43, 295–321. [Google Scholar] [CrossRef]
  11. Cronin, S.J.; Nell, V.E.; Lecontre, J.A.; Hedley, M.J.; Loganathan, P. Environmental hazards of fluoride in volcanic ash: A case study from Ruapehu volcano, New Zealand. J. Volcanol. Geotherm. Res. 2003, 121, 271–291. [Google Scholar] [CrossRef]
  12. D’Alessandro, W. Human fluorosis related to volcanic activity: A review. WIT Trans. Biomed. Health 2006, 10, 21–30. [Google Scholar] [CrossRef]
  13. Oskarsson, N. The interaction between volcanic gases and tephra: Fluorine adhering to tephra of the 1970 Hekla eruption. J. Volcanol. Geotherm. Res. 1980, 8, 251–266. [Google Scholar] [CrossRef]
  14. Edmunds, W.M.; Smedley, P.L. Fluoride in natural waters. In Essentials of Medical Geology; Selinus, O., Ed.; Springer: Dordrecht, The Netherlands, 2013. [Google Scholar] [CrossRef]
  15. Wenzel, W.W.; Blum, W.E.H. Fluorine speciation and mobility in F-contaminated soils. Soil Sci. 1992, 153, 357–364. [Google Scholar] [CrossRef]
  16. García-Herrera, R.; Gallego, D.; Henández, E.; Gimeno, L.; Ribera, P. Influence of the North Atlantic Oscillation on the Canary Islands Precipitation. J. Clim. 2001, 14, 3889–3903. [Google Scholar] [CrossRef]
  17. Troll, V.R.; Carracedo, J.C. (Eds.) The Geology of La Palma. In The Geology of the Canary Islands; Elsevier: Amsterdam, The Netherlands, 2016; pp. 101–180. [Google Scholar] [CrossRef]
  18. Carracedo, J.C.; Badiola, E.R.; Guillou, H.; de La Nuez, J.; Pérez-Torrado, F.J. Geology and volcanology of la Palma and El Hierro, Western Canaries. Estud. Geológicos 2001, 57, 175–273. [Google Scholar]
  19. Carracedo, J.C.; Rodríguez Badiola, E.; Guillou, H. Memoria de la Hoja nº 1085-III/IV (Isla de La Palma. El Pueblo), 1st ed.; Mapa Geológico de España E. 1:25.000, Segunda Serie; Insituto Geológico y Minero de España: Madrid, Spain, 2015; 170p. [Google Scholar]
  20. Carracedo, J.C.; Troll, V.R.; Day, J.M.D.; Geiger, H.; Aulinas, M.; Soler, V.; Deegan, F.M.; Perez-Torrado, F.J.; Gisbert, G.; Gazel, E.; et al. The 2021 Eruption of the Cumbre Vieja Volcanic Ridge on La Palma, Canary Islands. Geol. Today 2022, 38, 94–107. [Google Scholar] [CrossRef]
  21. Hernández-Pacheco, A.; Valls, M.C. The historic eruptions of La Palma Island (Canaries). Arquipélago. Série Ciências da Natureza 1982, 3, 83–94. [Google Scholar]
  22. Mata, M.P.; Ordóñez-Casado, B.; del Moral, B.; Mediato Arribas, J.F.; Bellido Martín, E.; Castillo, M.; Vegas, J.; Martínez Martínez, J.; Sánchez-España, J.; Pérez, R.; et al. Composición del material piroclástico de la erupción de 2021 en Cumbre Vieja (Isla de La Palma). Macla 2022, 26, 116–117. [Google Scholar]
  23. Parkhurst, D.L.; Appelo, C.A.J. Description of input and examples for PHREEQC version 3-A computer program for speciation, batch-reactions, one-dimensional transport, and inverse geochemical calculations. In Groundwater Book 6, Modeling Techniques; U.S. Geological Survey: Denver, CO, USA, 2013; p. 497. [Google Scholar]
  24. Allison, J.D.; Brown, D.S.; Novo-Gradac, J. MINTEQA2/PRODEAFA2, A Geochemical Assessment Model for Environmental Systems: User Manual Supplement for Version 4.0.; U.S. EPA, NERL: Athens, GA, USA, 1999; 42p. [Google Scholar]
  25. Wolff-Boenish, D.; Gislason, S.R.; Elkers, E.H. The effect of fluoride on the dissolution rates of natural glasses at pH 4 and 25 °C. Geochim. Cosmochim. Acta 2004, 68, 4571–4582. [Google Scholar] [CrossRef]
  26. Wolff-Boenish, D.; Gislason, S.R.; Elkers, E.H.; Putnis, C.V. The dissolution rates of natural glasses as a function of their composition at pH 4 and 10.6, and temperatures from 25 to 74 °C. Geochim. Cosmochim. Acta 2004, 68, 4843–4858. [Google Scholar] [CrossRef]
  27. Huerta, P.; Rodríguez-Berriguete, A.; Martín-García, R.; Martín-Pérez, A.; La Iglesia, A.; Alonso-Zarza, A.M. The role of climate and aeolian dust input in calcrete formation in volcanic islands (Lanzarote and Fuerteventura, Spain). Palaeogeogr. Palaeoclimatol. Palaeoecol. 2015, 417, 66–79. [Google Scholar] [CrossRef]
  28. Pokrovsky, L.S.; Golubev, S.V.; Schott, J.; Castillo, A. Calcite, dolomite and magnesite dissolution kinetics in aqueous solutions at acid to circumneutral pH, 25 to 150 °C and 1 to 55 atm pCO2: New constraints on CO2 sequestration in sedimentary basins. Chem. Geol. 2009, 265, 20–32. [Google Scholar] [CrossRef]
  29. Martínez-Martínez, J.; Mediato, J.F.; Mata, M.P.; Ordóñez, B.; del Moral, B.; Bellido, E.; Pérez-López, R.; Rodríguez-Pascua, M.A.; Vegas, J.; Lozano, G.; et al. Early fumarolic minerals from the Tajogaite volcanic eruption (La Palma, 2021). J. Volcanol. Geotherm. Res. 2023, 435, 107771. [Google Scholar] [CrossRef]
  30. Kockum, P.C.F.; Herbert, R.B.; Gislason, S.R. A diverse ecosystem response to volcanic aerosols. Chem. Geol. 2006, 231, 57–66. [Google Scholar] [CrossRef]
  31. Nordstrom, D.K.; Ball, J.W. The geochemical behavior of aluminum in acidified surface waters. Science 1986, 232, 54–56. [Google Scholar] [CrossRef] [PubMed]
  32. Rotteveel, L. Trends, Patterns, and Drivers of Freshwater Aluminium Concentrations: Revisiting the Conceptual Model of Freshwater Acidification. Master’s Thesis, Dalhousie University, Halifax, NS, Canada, 2020. [Google Scholar]
  33. Botté, A.; Zaidi, M.; Guery, J.; Fichet, D.; Leignel, V. Aluminium in aquatic environments: Abundance and ecotoxicological impacts. Aquat. Ecol. 2022, 56, 751–773. [Google Scholar] [CrossRef]
  34. Driscoll, C.T.; Schecher, W.D. The chemistry of aluminum in the environment. Environ. Geochem. Health 1990, 12, 28–49. [Google Scholar] [CrossRef] [PubMed]
  35. Driscoll, C.T.; Postek, K.M. The chemistry of aluminum in surface waters. In The Environmental Chemistry of Aluminum, 2nd ed.; Sposito, G., Ed.; CRC Press: London, UK, 1996; p. 56. [Google Scholar]
  36. Sánchez-España, J. The behavior of iron and aluminum in acid mine drainage: Speciation, Mineralogy, and Environmental Significance. In Thermodynamics, Solubility and Environmental Issues; Letcher, T.M., Ed.; Elsevier B.V.: Amsterdam, The Netherlands, 2007; pp. 137–149. [Google Scholar] [CrossRef]
  37. Sánchez-España, J.; Yusta, I.; Gray, J.; Burgos, W.D. Geochemistry of dissolved aluminum at low pH: Extent and significance of Al-Fe(III) coprecipitation below pH 4.0. Geochim. Cosmochim. Acta 2016, 175, 128–149. [Google Scholar] [CrossRef]
  38. Sánchez-España, J.; Yusta, I.; Burgos, W.D. Geochemistry of dissolved aluminum at low pH: Hydrobasaluminite formation and interaction with trace metals, silica and microbial cells under anoxic conditions. Chem. Geol. 2016, 441, 124–137. [Google Scholar] [CrossRef]
  39. Ayris, P.M.; Delmelle, P.; Pereira, B.; Maters, E.C.; Damby, D.E.; Durant, A.J.; Dingwell, D.B. Spatial analysis of Mount St. Helens tephra leachate compositions: Implications for future sampling strategies. Bull. Volcanol. 2015, 77, 60. [Google Scholar] [CrossRef] [PubMed]
  40. D’Alessandro, W.; Bellomo, S.; Parello, F. Fluorine adsorption by volcanic soils at Mt. Etna, Italy. Appl. Geochem. 2012, 27, 1179–1188. [Google Scholar] [CrossRef]
  41. Loganathan, P.; Hedley, M.J.; Wallace, G.C.; Roberts, A.H.C. Fluoride accumulation in pasture forages and soils following longterm applications of phosphorous fertilisers. Environ. Pollut. 2001, 115, 275–282. [Google Scholar] [CrossRef] [PubMed]
  42. Mackowiak, C.L.; Grossi, P.R.; Bugbee, B.G. Biogeochemistry of fluoride in a plant-solution system. J. Environ. Qual. 2003, 32, 2230–2237. [Google Scholar] [CrossRef] [PubMed]
  43. VROM. Aluminum Fluoride: Summary Risk Assessment Report; CAS No 7784-18-1, EEC 793/93, Ministry of Housing, Spatial Planning and the Environment (VROM): Bilthoven, The Netherlands, 2008; 12p. [Google Scholar]
  44. Exley, C.; Clarkson, E. Aluminum in human brain tissue from donors without neurodegenerative disease: A comparison with Alzheimer’s disease, multiple sclerosis and autism. Sci. Rep. 2020, 10, 7770. [Google Scholar] [CrossRef] [PubMed]
  45. Gensemer, R.W.; Playle, R.C. The bioavailability and toxicity of aluminium in aquatic environments. Crit. Rev. Environ. Sci. Technol. 1999, 29, 315–450. [Google Scholar] [CrossRef]
Minerals 14 00338 g001
Figure 1. General views of the eruptive column and tephra deposits. (a) Vertical eruptive column (view from the southwest); (b) hole dug in the tephra deposit for a volcano-stratigraphic study; (c) eruptive column leaning towards west on a windy day; (d) eruptive column leaning towards northeast on a windy day.
Figure 1. General views of the eruptive column and tephra deposits. (a) Vertical eruptive column (view from the southwest); (b) hole dug in the tephra deposit for a volcano-stratigraphic study; (c) eruptive column leaning towards west on a windy day; (d) eruptive column leaning towards northeast on a windy day.
Minerals 14 00338 g001
Minerals 14 00338 g002
Figure 2. Geographic location and volcano-stratigraphic columns of the Cementerio and Enrique tephra deposits studied in this study, with indication of the sampled levels. Diamonds, circles and squares in level numbers indicate the mean particle size: fine ash (diamonds), coarse ash-fine lapilli (circles) or coarse lapilli (squares).
Figure 2. Geographic location and volcano-stratigraphic columns of the Cementerio and Enrique tephra deposits studied in this study, with indication of the sampled levels. Diamonds, circles and squares in level numbers indicate the mean particle size: fine ash (diamonds), coarse ash-fine lapilli (circles) or coarse lapilli (squares).
Minerals 14 00338 g002
Minerals 14 00338 g003
Figure 3. Schedule of the singular volcanic events, environmental conditions, and maximum registered rainfall (“Volc. Events”, “Env. Cond.” and “Max. Rainfall” in the table) during the three months of volcanic eruption in the Tajogaite volcano. The (A–P) maps show the spatial distribution of precipitation during the wet periods (maps obtained from www.HDMeteo.com; accessed on 24 December 2021). (A–P) The corresponding date of each map is included in the table.
Figure 3. Schedule of the singular volcanic events, environmental conditions, and maximum registered rainfall (“Volc. Events”, “Env. Cond.” and “Max. Rainfall” in the table) during the three months of volcanic eruption in the Tajogaite volcano. The (A–P) maps show the spatial distribution of precipitation during the wet periods (maps obtained from www.HDMeteo.com; accessed on 24 December 2021). (A–P) The corresponding date of each map is included in the table.
Minerals 14 00338 g003
Minerals 14 00338 g004
Figure 4. Vertical profiles of pH, bicarbonate, and sulfate concentration obtained for the Enrique (A–C) and Cementerio (D–F) ashfall deposits, as deduced from leaching tests conducted on samples taken at different depths. The background value (b.v.) of bicarbonate and sulfate concentrations (as deduced from blanks conducted on pure MilliQ solutions; Table S1) are given for comparison.
Figure 4. Vertical profiles of pH, bicarbonate, and sulfate concentration obtained for the Enrique (A–C) and Cementerio (D–F) ashfall deposits, as deduced from leaching tests conducted on samples taken at different depths. The background value (b.v.) of bicarbonate and sulfate concentrations (as deduced from blanks conducted on pure MilliQ solutions; Table S1) are given for comparison.
Minerals 14 00338 g004
Minerals 14 00338 g005
Figure 5. Binary plot showing the strong correlation between the sum of sulfate and bicarbonate ions [SO42− + HCO3−] in mM and conductivity for the whole data set of 30 samples (12 samples from the Enrique sequence and 18 samples from the Cementerio sequence).
Figure 5. Binary plot showing the strong correlation between the sum of sulfate and bicarbonate ions [SO42− + HCO3−] in mM and conductivity for the whole data set of 30 samples (12 samples from the Enrique sequence and 18 samples from the Cementerio sequence).
Minerals 14 00338 g005
Minerals 14 00338 g006
Figure 6. Vertical profiles of water-extractable aluminum and fluoride concentration across the Enrique (ENR, (A,B)) and Cementerio (CEM, (C,D)) ashfall sequences studied in La Palma.
Figure 6. Vertical profiles of water-extractable aluminum and fluoride concentration across the Enrique (ENR, (A,B)) and Cementerio (CEM, (C,D)) ashfall sequences studied in La Palma.
Minerals 14 00338 g006
Minerals 14 00338 g007
Figure 7. Boxplot diagram comparing the water-extractable content of major ions (F−, Cl−, SO42−, Al, Ca, Na) measured in samples of the first ashfall expelled by the Tajogaite volcano (September–October 2021, taken from [4]; boxes on the left and labelled as “fa”) with the results obtained in this study on ENR and CEM ashfall sequences (boxes on the right and labelled as “ec”). The data points outside the whiskers of the box plots represent outliers.
Figure 7. Boxplot diagram comparing the water-extractable content of major ions (F−, Cl−, SO42−, Al, Ca, Na) measured in samples of the first ashfall expelled by the Tajogaite volcano (September–October 2021, taken from [4]; boxes on the left and labelled as “fa”) with the results obtained in this study on ENR and CEM ashfall sequences (boxes on the right and labelled as “ec”). The data points outside the whiskers of the box plots represent outliers.
Minerals 14 00338 g007
Minerals 14 00338 g008
Figure 8. Scatterplots of fluoride vs. aluminum molar concentrations in the obtained leachates from the Enrique (ENR, (A)) and Cementerio (CEM, (B)) ashfall sequences. The regression lines and correlation coefficients (R2) are indicated in both cases.
Figure 8. Scatterplots of fluoride vs. aluminum molar concentrations in the obtained leachates from the Enrique (ENR, (A)) and Cementerio (CEM, (B)) ashfall sequences. The regression lines and correlation coefficients (R2) are indicated in both cases.
Minerals 14 00338 g008
Minerals 14 00338 g009
Figure 9. Scatterplots showing the correlation of pH with aluminum and fluoride concentration in the water-extractable fraction of ashfall samples in the Enrique (A,B) and Cementerio (C,D) sequences. The correlation coefficient (R2) is indicated in all cases.
Figure 9. Scatterplots showing the correlation of pH with aluminum and fluoride concentration in the water-extractable fraction of ashfall samples in the Enrique (A,B) and Cementerio (C,D) sequences. The correlation coefficient (R2) is indicated in all cases.
Minerals 14 00338 g009
Minerals 14 00338 g010
Figure 10. SEM images (A,B) and EDS spectra (C,D) of aggregates formed by finely intergrown crystals of aluminum fluoride (AlF3, yellow arrows) and fluorite (CaF2, white arrows) found in the studied ashfall particles from La Palma.
Figure 10. SEM images (A,B) and EDS spectra (C,D) of aggregates formed by finely intergrown crystals of aluminum fluoride (AlF3, yellow arrows) and fluorite (CaF2, white arrows) found in the studied ashfall particles from La Palma.
Minerals 14 00338 g010
Minerals 14 00338 g011
Figure 11. SEM image (A), element mapping for F, Ca, and Al (B–D) and the EDS spectrum (E) of an aggregate of fluorite (CaF2, white dashed line) microcrystals coexisting with adjacent particles of aluminum fluoride (AlF3, yellow dashed line).
Figure 11. SEM image (A), element mapping for F, Ca, and Al (B–D) and the EDS spectrum (E) of an aggregate of fluorite (CaF2, white dashed line) microcrystals coexisting with adjacent particles of aluminum fluoride (AlF3, yellow dashed line).
Minerals 14 00338 g011
Minerals 14 00338 g012
Figure 12. Variation in the saturation index (SI) for selected minerals (fluorite, gibbsite, gypsum) with pH (A,B) and ion concentration (C,D) in the ashfall profiles ((A,C,D); Enrique; (B), Cementerio). In (A,B), all SI values have been calculated using the respective chemical compositions of the samples (given in Tables S1 and S2) except when calcium was below detection (<1 mg/L), where we used an arbitrary value of 0.5 mg/L Ca to provide at least a “maximum SI value”. In (C), the SI values are plotted against Al or Ca concentration (in mg/L) and have been calculated for constant values of pH (7.0) and anion concentration (F− = SO42− = 10 mg/L). In (D), the SI values are plotted against F- or SO42− concentration (in mg/L) and have been calculated for constant values of pH (7.0) and cation concentration (Ca = 1 mg/L, Al = 5 mg/L). In the four plots (A–D), the shaded area around the horizontal line of SI = 0 depicts the near-equilibrium zone (defined as SI = −0.5 to +0.5).
Figure 12. Variation in the saturation index (SI) for selected minerals (fluorite, gibbsite, gypsum) with pH (A,B) and ion concentration (C,D) in the ashfall profiles ((A,C,D); Enrique; (B), Cementerio). In (A,B), all SI values have been calculated using the respective chemical compositions of the samples (given in Tables S1 and S2) except when calcium was below detection (<1 mg/L), where we used an arbitrary value of 0.5 mg/L Ca to provide at least a “maximum SI value”. In (C), the SI values are plotted against Al or Ca concentration (in mg/L) and have been calculated for constant values of pH (7.0) and anion concentration (F− = SO42− = 10 mg/L). In (D), the SI values are plotted against F- or SO42− concentration (in mg/L) and have been calculated for constant values of pH (7.0) and cation concentration (Ca = 1 mg/L, Al = 5 mg/L). In the four plots (A–D), the shaded area around the horizontal line of SI = 0 depicts the near-equilibrium zone (defined as SI = −0.5 to +0.5).
Minerals 14 00338 g012
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Sánchez-España, J.; Castillo, A.M.N.; Mata, M.P.; Martínez-Martínez, J.; Mediato, J.F. Fluorite and Gibbsite Solubility Controls the Vertical Transport of Fluoride and Aluminum during Rainwater Percolation through Ashfall Deposits in La Palma (Canary Islands, Spain). Minerals 2024, 14, 338. https://doi.org/10.3390/min14040338

AMA Style

Sánchez-España J, Castillo AMN, Mata MP, Martínez-Martínez J, Mediato JF. Fluorite and Gibbsite Solubility Controls the Vertical Transport of Fluoride and Aluminum during Rainwater Percolation through Ashfall Deposits in La Palma (Canary Islands, Spain). Minerals. 2024; 14(4):338. https://doi.org/10.3390/min14040338

Chicago/Turabian Style

Sánchez-España, Javier, Ana M. Nieto Castillo, M. Pilar Mata, Javier Martínez-Martínez, and Jose F. Mediato. 2024. "Fluorite and Gibbsite Solubility Controls the Vertical Transport of Fluoride and Aluminum during Rainwater Percolation through Ashfall Deposits in La Palma (Canary Islands, Spain)" Minerals 14, no. 4: 338. https://doi.org/10.3390/min14040338

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop