The Domuyo volcanic system: An enormous geothermal resource in Argentine Patagonia

Full text

The Domuyo volcanic system: An enormous geothermal resource in
Argentine Patagonia
Giovanni Chiodini
a
, Caterina Liccioli
b,
⁎, Orlando Vaselli
c,d
,SergioCalabrese
e
, Franco Tassi
c,d
, Stefano Caliro
a
,
Alberto Caselli
b,f
,MarianoAgusto
f
, Walter D'Alessandro
g
a
Istituto Nazionale di Geofisica e Vulcanologia sezione di Napoli “Osservatorio Vesuviano,”Via Diocleziano, Napoli 328-80124, Italy
b
Instituto de Paleobiología y Geología de la Universidad Nacional Rio Negro, Rio Negro, Argentina
c
Dipartimento di Scienze della Terra, Università di Firenze, Firenze, Italy
d
CNR-Consiglio Nazionale delle Ricerche, Istituto di Geoscienze e Georisorse, Firenze, Italy
e
Dipartimento di Scienze della Terra e del Mare, Universitàdi Palermo, Palermo, Italy
f
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina
g
Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Palerm o, Palermo , Italy
abstractarticle info
Article history:
Received 24 November 2013
Accepted 5 February 2014
Available online 15 February 2014
Keywords:
Domuyo volcano
Argentine Patagonia
Geothermal potential
Water geochemistry
A geochemical survey of the main thermal waters discharging in the southwestern part of the Domuyo volcanic
complex (Argentina), where the latest volcanic activity dates to 0.11 Ma, has highlighted theextraordinarily high
heat loss from this remote site in Patagonia. The thermal water discharges are mostly Na-Cl in composition and
have TDS values up to 3.78 g L
−1
(El Humazo). A simple hydrogeochemical approach shows that 1,100 to
1,300 kg s
−1
of boiling waters, which have been affected by shallow steam separation, flow into the main drain-
age of the area(Rio Varvarco). A dramatic increase of the mostconservative species such as Na,Cl and Li from the
Rio Varvarco from upstream to downstream was observed andrelated solely to thecontribution of hydrothermal
fluids. The equilibrium temperatures of the discharging thermal fluids, calculated on the basis of the Na-K-Mg
geothermometer, are between 190 °C and 230 °C. If we refer to a liquid originally at 220 °C (enthalpy =
944 J g
−1
), the thermal energy release can be estimated as high as 1.1 ± 0.2 GW, a value that is much higher
than the natural release of heat in other important geothermal fields worldwide, e.g., Mutnovsky (Russia),
Wairakei (New Zealand) and Lassen Peak (USA). This value is the second highest measured advective heat
flux from any hydrothermal system on Earth after Yellowstone.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
In tectonically young and active areas of the Earth, where meteoric
waters infiltrate and circulate deeply, advective heat flux may be the
dominantform of heat transfer andshould be considered inthe evalua-
tion of the geothermal heat flux (e.g., Bodmer and Rybach, 1985;
Čermak and Jetel, 1985; Gosnold, 1990; Bodri and Rybach, 1998;
Brumm et al., 2009; Ingebritsen and Mariner, 2010; Chiodini et al.,
2013). The magnitude and spatial distribution of heat flux are relevant
for evaluating potential geothermal resources and also for inferring
thermal the deep structure underlying volcanic areas, with reference
to the nature and size of heat sources (such as magma or hot intrusive
rocks) at depth (Manga, 1998). The fluid discharge rates of major ther-
mal areas worldwide range from a few liters to several cubic meters of
hot water per second (Taran and Peiffer, 2009). At Yellowstone Caldera
(USA), the total discharge of the thermal waters was estimated to be
3,200 L s
−1
(Fournier, 1989). However, thermal discharges in othergeo-
thermal systems are rarely higher than 100 L s
−1
, this being for instance
the reference value computed for pre-exploitation discharge at
Wairakei (New Zealand; Ellis and Wilson, 1955). Thermal water flow
rates of about 80 L s
−1
were estimated at the Mutnovsky geothermal
field in Kamchatka, Russia (Vakin and Pilipenko, 1986), while about
20 L s
−1
of thermal water flow was measured at the Lassen Peak hydro-
thermal system (USA; Sorey, 1986). Others authors (Manga, 1998)
stressed that cold or “slightly thermal”springs can also strongly
affect the heat budget of a thermal zone, showing that in the central
Oregon Cascades roughly half of the advective geothermal heat flux is
discharged from springs only a few degrees above local ambient
temperature.
In this work, we present the results of a geochemical survey carried
out on boiling water discharges located at the remote volcano of
Domuyo, a scarcely investigated volcanic complex from Argentine Pata-
gonia (Fig. 1). Palacio and Llambias (1978) provided the first geochem-
ical characterization of the Domuyo waters while other chemical data
were reported in two internal reports produced by JICA (1983; 1984)
in the framework of a joint collaboration between the Government of
Journal of Volcanology and Geothermal Research 274 (2014) 71–77
⁎Corresponding author. Tel.: +54 1592984826048.
E-mail address: caterina.liccioli@yahoo.it (C. Liccioli).
http://dx.doi.org/10.1016/j.jvolgeores.2014.02.006
0377-0273/© 2014 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Journal of Volcanology and Geothermal Research
journal homepage: www.elsevier.com/locate/jvolgeores

Argentina and the Government of Japan to evaluate the geothermal re-
sources in the area.
On the western slope of the Domuyo volcano, several boiling ther-
mal springs discharge into roughly ENE-WSW-oriented creeks, which
in turn flow into the main drainage system of the southwestern part
of the volcanic edifice: the N-S-oriented Rio Varvarco (Fig. 2). A geo-
chemical approach similar to those successfully applied to Yellowstone
(Fournier, 1989) and El Chichon (Mexico, Taran and Peiffer, 2009)was
performed in this sector of the Domuyo volcano in order to estimate
its thermal release and geothermal potential. The thermal energy was
computed by (i) evaluating the flux of chloride from the thermal source
into the rivers and (ii) determining the original enthalpy and chloride
content of the geothermal brines. Our results demonstrate that the
thermal discharges account for a surprisingly high-energy release of
1.1 ± 0.2 GW.
2. Regional setting
2.1. Geological setting and volcanic activity
The Domuyo volcanic complex (36°340 S, 70°250 W, 4,709 m high),
nicknamed the “Roof of Patagonia,”is Middle Miocene to Early Pliocene
in age and is located at the northern edge of the Cordillera del Viento
(Fig. 1), a chain uplifted during Cretaceous and Neogene times (Kay
et al., 2006) that, together with the late Miocene to Quaternary Guañacos
fold and thrust-belt mountain system (Suárez and Emparan, 1995,
1997) to its west, constitutes the main geological feature of the study
area. At largerscale, the Domuyo volcanobelongs to the southern volca-
nic zone (SVZ: 33°S to 46°S), which has evolved for the last 20 Ma years
by slightly dextral-oblique convergence between the Nazca and the
South America plates at a rate of ca. 7–9 cm/year (Cembrano and Lara,
2009 and references therein).
The Domuyo volcano, with an estimated volume of ~ 200 km
3
and a
base area of 138 km
2
(Völker et al., 2011), is a broad structural dome
and forms, together with three other tectonic domes, a major anticline,
the “La Cruzada”high (Folguera et al., 2007). The geology of the area,
initially studied by Groeber (1947), is made up of Permian–Triassic sed-
imentary and pyroclastic rocks (Choyoi Group) and acidic intrusive
rocks, the latter intruding older formations. Jurassic and Cretaceous sed-
imentary and pyroclastic rocks unconformably overlie the basement
(Spagnuolo et al., 2012). Tertiary andesitic volcanism covered the Me-
sozoic rocks and intrusive bodies emplaced close to the Domuyo volca-
no. Rhyolitic to dacitic tuff, lapilli tuff and lava flows cover its western
slope. According to Mas et al. (2009), intrusive stocks are centered on
the volcano and outcrop in an area of about 24 km
2
at Domuyo, intrud-
ing the oldersedimentary cover.Volcanism resumed in the late Pliocene
to Quaternary during which time sub-volcanic bodies (K/Ar 2.5 ±
0.5 Ma; Miranda, 1996, 2006) were also emplaced. The volcanic rocks
surrounding the Domuyo volcano belong to two distinct magmatic
series, originating from two different feeding magmas (Brousse and
Pesce, 1982). One of these (Late Pliocene–Early Pleistocene) is calc-
alkaline and yielded andesitic rocks; the other (Middle Pleistocene) is
represented by both shoshonitic and alkaline series and has formed
dacitic and/or rhyolitic rocks of which the main expression is the
Cerro Domo (absolute age 0.72 ± 10 Ma calculated on a rhyolitic
lava flow; Brousse and Pesce, 1982), located on the southern slope of
Domuyo. Younger volcanic products were dated between 0.55 ±
0.10 and 0.11 ± 0.02 Ma (JICA, 1983)bythefission-track method,
suggesting that Domuyo volcanism has been active through the
Late Pleistocene.
2.2. Hydrothermal activity
This remote site in Patagonia has impressive signs and obvious
evidence of high release of thermal energy. High flow rates (hundreds
Fig. 1. (Left panel) The Domuyo volcanic complex, north of the Cordillera del Viento chain, and (right panel) geological map showing the distribution of Permian–Triassic to Pleistocene
rocks; the arrow points to a K-Ar dating location (modified from Miranda et al., 2006).
72 G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77

Ls
−1
) of boiling fluids (Table 1,Fig. 3) enter the hydrographic network,
which are able to maintain the temperature of the creeks up to 30 °C for
5to10kmdownstreamfromthefluid source.
In February 2003, two hydrothermal explosions occurred at El
Humazo (“The Great Smoke”)(Fig. 3a,b), during which blocks up to 1
ton were displaced by theburst. A vapor plume with a maximum height
of about 300 m was visible up to a distance of 20 km (Mas et al., 2009).
This explosive event took place in a fumarolic area that, along with
those at Rincón de la Papas, La Bramadora, Los Tachos (Fig. 3c), Aguas
Calientes and Las Olletas (Fig. 3d), characterizes the SW part of the
Domuyo volcanic complex. The 2003 explosions were attributed to
local sealing of the near-surface channels from where the hydrother-
mal fluids are discharged (Mas et al., 2009). Self-sealing effects in
geothermal fields (Facca and Tonani, 1971) likely play an important
role in triggering hydrothermal eruptions. Nevertheless, it is reason-
able to infer the presence of significant amounts of fluids and energy
at depth, and that the progressive flashing has followed a top–down
model (Lawless and Browne, 2001). Injection of a small batch of
magma (dyke), able to destabilize the hydrothermal system, cannot
be ruled out.
Fig. 2. Location of samplingsites in the SW slopeof Domuyo volcano.(For interpretation of
the references to colour in this figure, the reader is referred to the web version of this
article.)
Table 1
Geographical coordinates, altitudes, flow rates, physico-chemical features and geochemical facies of the studied waters.
Sample Locality Type Latitude Longitude Altitude TpH EC Flow rate Na K Mg Ca Li Cl HCO
3
SO
4
TDS Composition
UTM UTM m °C mS/cm kg s
−1
mg kg
−1
gkg
−1
dor1 Manchana Covunco Creek 361029 5942745 1,995 21.2 7.96 1,641 nd 169 17.8 7.74 113.0 na 245 115 129 0.80 Na-Cl
dor2 Manchana Covunco Creek 360826 5942649 1,952 48.5 8.01 3,860 800 526 53.8 3.04 92.9 5.08 799 100 257 1.84 Na-Cl
dor4 Covunco Creek 361863 5938249 2,038 37.2 na 3,100 840 462 44.4 3.30 168.3 4.23 722 84 220 1.71 Na-Cl
dor5 Covunco Creek 354135 5936607 1,541 23.1 8.15 2,506 nd 404 36.6 2.38 64.8 3.60 629 74 169 1.38 Na-Cl
dor6 Rio Varvarco Rio 352305 5950486 1,581 10.5 8.08 416 6,630 5.98 1.4 2.54 79.1 0.003 1.8 56 145 0.29 Ca-SO
4
dor7 Ahilinco Creek 356191 5945057 1,658 11.5 8.39 147 nd 14.9 2.5 4.89 10.9 na 17.2 55 1.02 0.11 Ca-HCO
3
dor8 Manchana Covunco Creek 355626 5942334 1,585 29.7 8.46 4,160 1,270 416 41.3 3.31 61.7 na 617 98 172 1.41 Na-Cl
dor9 Agua Caliente Creek 356362 5939441 1,739 47.8 7.23 4,800 87 454 28.4 1.57 24.4 na 704 112 62.2 1.39 Na-Cl
dor10 Atreuco Creek 355327 5935187 1,668 16.4 8.08 32 nd 3.1 0.7 0.56 3.1 na 0.5 15 0.94 0.02 Ca(Na)-HCO
3
dor11 Matancilla Creek 350995 5924804 1,301 21.2 8.47 49 nd 4.9 2.2 3.10 7.9 na 0.8 50 1.11 0.07 Ca-HCO
3
dor12 Rio Varvarco Rio 350368 5921396 1,184 18.0 8.04 875 12,700
§
97 9.1 1.72 67.1 0.78 149 56 133 0.51 Na-Cl
dos1 El Humazo Cold spring 361866 5942377 2,198 12.9 6.80 80 nd 3.1 1.3 0.85 4.5 nd 0.4 28 1.39 0.04 Ca-HCO
3
dos2 Los Tachos Cold spring 361906 5938262 2,050 8.1 na 78 nd 13.7 1.2 0.34 3.9 0.10 19.1 15 3.1 0.06 Na-Cl
dot3 El Humazo Thermal spring 361002 5942747 1,991 97.2 8.32 9,600 nd 1280 129.0 0.07 37.5 12.37 2020 78 219 3.78 Na-Cl
dot4 Los Tachos Thermal spring 361901 5938273 2,043 96.2 na na nd 1160 67.5 0.60 58.7 11.16 1820 119 187 3.42 Na-Cl
dot7 Las Olletas Thermal spring 358400 5941715 1,870 97.0 8.00 7,860 nd 1020 57.0 0.51 36.1 9.91 1590 102 160 2.98 Na-Cl
dot9 Agua Caliente Thermal spring 356511 5939451 1,762 67.5 6.81 4,600 nd 630 40.3 1.04 26.7 5.54 945 103 96 1.85 Na-Cl
na = not analyzed; nd = not determined; EC = electrical conductivity.
§
This value represents the mean flow rate recorded during the measurements (see text for further explanations).
73G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77

3. Material and methods
A survey of the thermal features of the SW part of Domuyo volcano,
where the 2003 hydrothermal explosions occurred, was performed in
March 2013 by an Argentine-Italian team. This sector of the volcano
was previously recognized (Palacio and Llambias, 1978) as the site of
an important hydrothermal activity characterized by many thermal dis-
charges, among which we selected the main ones. In particular, four
thermal springs from El Humazo (dot3), Los Tachos (dot4), Las Olletas
(dot7) and Agua Caliente (dot7) were sampled and analyzed for the
main chemical compounds, as was Rio Varvarco upstream (dor6) and
downstream (dor12) of the hot water inflow and the main creeks
(from north to the south, Ahilinco: dor7; Manchana Covunco: dor1,
dor2, dor8; Agua Caliente: dor9, Covunco: dor4, dor5; Atreuco: dor10;
Matancilla: dor11) discharging into the Rio Varvarco. Two cold springs
discharging close to El Humazo (dos1) and Los Tachos (dos2) (Fig. 2)
were also sampled.
Temperature, pH and electrical conductivity were measured in the
field. Water samples were filtered using 0.45 μmfilters and collected
in 125 and 50 mL polyethylene bottles for the determination of anions
(HCO
3
,Cl,SO
4
) and cations (Na, K, Ca, Mg, NH
4
and Li). The latter aliquot
was acidified with 0.5 mL of ultrapure HCl (30%). Bicarbonate was
determined by acidimetric titration with 0.01 N HCl, using methyl-
orange as indicator. Anions and cations were determined by ion
chromatography using a Dionex ICS-3000 at the Laboratory of Istituto
Nazionale di Geofisica e Vulcanologia (Naples, Italy). All determinations
were referred to standard solutions calibrated versus NIST Standard
Reference Materials. All the laboratory analytical methods and the alka-
linity determinations have accuracy better than 3%.
Flow rate measurements were made in Rio Varvarco, upstream
(dor6) and downstream (dor7) of the thermal inflow, and in the ther-
mally heated creeks (dor5, dor8 and dor9). Owing to the unavailability
of standard stream gaging equipment, flow rates were calculated in
suitable sections using the float method. Selected sections of each
river were measured in detail, and the water velocity was then comput-
ed as the average of 10 or more values obtained by measuring the travel
time of a float across a fixed distance (generally, 10 to 20 m). The max-
imum uncertainty of the flow rate measurement of dor12 (Rio Varvarco
downstream) deserves further explanations because this measurement
is used to estimate the total energy release of Domuyo geothermal
system. The cross section of dor12 is shown in Fig. 4 where the mean
superficial velocity of the entire section (1.54 m s
−1
), of the central
fastest zone (1.89 m s
−1
) and of the two lateral zones of lower velocity
(1.33 m s
−1
) is shown. In our computation, we assumed the flow rates
computed with the lower and higher velocity as reasonable limits of the
measurement uncertainty (11,000 and 15,700 kg s
−1
). This range will
be used to quantify the uncertainty of the thermal energy release
Fig. 3. Imagesof thermal dischargefrom the SW slopeof Domuyo volcano:(a and b) El Humazo;(c) Los Tachos; (d) LasOlletas. (For interpretation of the references to colour in this figure,
the reader is referred to the web version of this article.)
Fig. 4. The outflow area (A in m
2
) of the Rio Varvarco(dot12 in Table 1) and the velocities
(v in m/s) used to calculate the flow rate(Q)anditsuncertaintiesbyQ=A×v(see textfor
explanation).
74 G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77

estimated by the Cl-inventory method. The basis of the Cl-inventory
method is simultaneous measurement of flow rate and chemical com-
position of streams and rivers at specific sampling locations. Most
high-chloride hot springs in non-arid regions occur near perennial
streams that eventually capture most of the thermal fluid (Taran and
Peiffer, 2009). The total discharge from hot-spring areas can be moni-
tored on thebasis of downstream increases in the solute loads of nearby
streams, chloride beingthe most conservative elementwith which ther-
mal waters are usually enriched relative to nearby surface and/or shal-
low ground waters.
4. Results
Dot3, dot4 and dot7 thermal springs discharge between 1,870 and
2,040 m of altitude and have boiling temperatures (up to 97 °C) and al-
kaline pH (slightly higher than 8), while dot9 has a temperature of
67.5 °C and a pH value of 6.8. Salinities (expressed as TDS, Total Dis-
solved Solids) are 3.78, 3.42 and 2.98 g kg-1, respectively. Agua Caliente
(dot9) discharges at 1,792 m and has lower temperature (67.5 °C), pH
(6.81) and salinity (1.85 g kg
−1
) lower than the other thermal waters.
All the thermal waters have a Na-Cl composition (Table 1). A similar
composition is observed for the creeks (dor1, dor2, dor4, dor5, dor8,
dor9 and dor12) into which these thermal waters discharge. Neverthe-
less, their salinity(from 0.06 to 1.8 g kg
−1
) is lower thermal springs due
to dilution processes affecting the thermal component.
Samples dor7, dor10, dor11 and dos1 are characterized by a Ca(Na)-
HCO
3
composition and low salinities (≤0.11 g L
−1
), while only sample
dor6 has a Ca-SO
4
composition (TDS = 0.29 g L
−1
), likely derived from
the dissolution of the gypsum present upstream of the surveyed area
(Palacio and Llambias, 1978).
The calculated flow rates for dor2, dor4, dor6, dor8, dor9 and
dor12 are reported in Table 1. The highest value was recorded at
Rio Varvarco downstream (dor12, mean value: 12,700 kg s
−1
)
where all the tributaries, including those that coll ect the thermal wa-
ters, have coalesced.
5. Discussion
5.1. Water chemistry and geothermometry
Equilibrium reservoir temperatures were calculated for the main
thermal water discharges, i.e., El Humazo (dot3, T=97°C),Los
Tachos (dot4, T= 96 °C), Las Olletas (dot7, T=97°C)andAgua
Caliente (dot9, T= 67.5 °C) (Table 1). These Na-Cl thermal waters
have chloride contents from 950 mg kg
−1
(dot9) to 2,000 mg kg
−1
(dot3). Samples dot3, dot4 and dot7 are at boiling point and are
affected by shallow steam separation, which likely increases their
original Cl content.
This is particularly evident at El Humazo spring where there are
nearby steam vents (Fig. 3a,b). During steam separation, CO
2
is lost,
causing an increase in pH (up to 8.3 in dot3) and precipitation of
CaCO
3
,astestified by the presence of travertinedeposits. At El Humazo,
travertine forms a deposit of 500 ×200 m with a maximum thickness of
50 m (Mas et al., 2009). Steam separation does not seem to affect Agua
Caliente, the temperature of which is below the boiling point and which
is characterized by the lowest chloride content (950 mg kg
−1
)among
the studied thermal waters. Agua Caliente can be considered as derived
by mixing between the original thermal component and cold ground-
water. Despite the possible steam separation or mixing processes, the
four thermal waters are suitable for application of the Na-K-Mg
geothermometer because of their very low Mg contents (from 0.07 to
1mgkg
−1
), which are compatible with a “mature”geothermal liquid
(Giggenbach, 1988). This geothermometric method compares, in a tri-
angular diagram, the relative proportions of the measured values (in
mg kg
−1
) with the theoretical ratios among Na, K and Mg
1/2
evaluated
for different temperatures (Fig. 5;Giggenbach, 1988). The theoretical
values refer to a solution in equilibrium with a typical hydrothermal
mineral assemblage that considers the following reactions:
K−feld þNaþ¼Na−feld þKþð1Þ
2:8K−feld þ1:6H
2OþMg2þ¼0:8K−mica þ0:2Chlorite
þ5:4SiO
2þ2Kþð2Þ
These reactions refer to a mineralogical assemblage when an iso-
chemical equilibrium is achieved. The Na/K and the K
2
/Mg ratios de-
pend on temperature according to
TC

¼1390=1:75 þlog Na=K
ðÞ½
fg–273:15 ð3Þ
TC

¼4410=14:0−log K2=MgðÞ½
fg
–273:15 ð4Þ
The intersection of each Na-K and K-Mg isotherm corresponds to
water compositions in equilibrium with a mineralogical assemblage
that controls both geothermometers and delineates the so-called “full
equilibrium”curve. In Fig. 5,thefields of partial equilibrium and imma-
ture waters are also delineated. Because of the fast kinetics of the reac-
tions involving Mg, Eq. (4) can quickly equilibrate at the decreasing
temperatures encountered by a geothermal liquid ascending toward
the surface, displacing the points in Fig. 5 toward the Mg corner. The ef-
fect is similar to that caused by mixing of the thermal component with
shallow cold groundwater which, relative to the geothermal liquids, is
enrichedin Mg. However, suchsecondary processes do not substantially
change the temperature estimates based on the Na-K geothermometer,
the latter providing reliable estimation of the temperature of the origi-
nal thermal liquid. The Domuyo thermal waters lie either along the
“full equilibrium”curve (dot3) or in the “partial equilibrium”field
(dot4, dot7, dot9), indicating Na/K temperatures of 233 °C (dot3),
192 °C (dot4), 190 °C (dot7) and 199 °C (dot9). As suggested by the
relatively low Cl content and emergence temperature, dot9 (Agua
Caliente) is likely sourced by a thermal system that mixes with shallow
cold groundwater of composition similar to samples dos1 and dos2
(Fig. 5). The other thermal springs, which emerge at the boiling point,
plot close to the full equilibrium curve. In these cases, the significant
Fig. 5. The Na-K-Mg geothermometric triangular diagram (Giggenbach, 1988) and the
Domuyo thermal waters. Cold waters dos1 and dos2 are reported for comparison. (For
interpretation of the references to colour in this figure, the reader is referred to the web
version of this article.)
75G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77

secondary process is loss of steam during boiling, which does not affect
the Na/K ratio, but increases the absolute concentrations of ions.
5.2. Flux of the geothermal liquid in the Rio Varvarco and its tributaries and
computation of the total thermal release
The relatively alkaline pH values and high Cl concentrations suggest
that the Domuyo thermal waters may be affected by steam loss during
boiling. Such a process is particularly evident for dot3 (El Humazo)
where vigorous vapor vents accompany the water discharge (Fig. 3a,
b). According to Mas et al. (2009), the emitted gas consists of 99% of
steam and 1% of other gases, among which CO
2
is the most abundant
(JICA, 1983).
Assuming an iso-enthalpic boiling process, the fractions of separated
steam (from 0.17 to 0.26) were subsequently used to restore the origi-
nal composition of the un-boiled liquid by referring to the mobile spe-
cies (Cl ~ 1,500 mg kg
−1
, Na ~ 950 mg kg
−1
, Li ~ 9.2 mg kg
−1
for dot3
and dot4, and Cl ~1,300 mg kg
−1
, Na ~ 850 mg kg
−1
, Li ~ 8.2 mg kg
−1
for dot2). The restored compositions of the three samples are similar
and in particular those of sample dot3 and dot4 are practically the
same. The restored concentration of Cl from dot3 and dot4, which are
characterized by the highest flow rate, is our reference concentration
for the un-boiled liquid and is assumed as the minimum concentration
in computation of thethermal energy. The measured Cl concentration of
dot3 (2,000 mg kg
−1
) was considered as the maximum concentration
of the original geothermal liquid, assuming a non-isenthalpic process
where the cooling of the water is governed by heat exchange with the
rocks, without any steam separation.
The conservative behavior of Na, Cl and Li may allow use of these
solutes as tracers of the thermal component flux in Rio Varvarco and
its tributaries. We assumed that Cl, Na and Li contents increase from
Rio Varvarco upstream (dor6) to Rio Varvarco downstream (dor12) is
due to the hot waters in the tributaries, i.e., Manchana Covunco (dor8:
29.7 °C, 4,160 μS/cm), Agua Caliente (dor9: 47.8 °C, 4,800 μS/cm) and
Covunco (dor5: 23.1 °C, 2,510 μS/cm). The other tributaries have
ambient temperature and lower salinity as suggested by their low elec-
trical conductivity (dor7: Ahilinco 11.5 °C, 147 μS/cm; dor10, Atreuco
16.4 °C, 32 μS/cm; dor11, Matansilla: 21.1 °C, 49 μS/cm).
Qualitatively the mixing between the thermal component and the
low salinity waters is indicated by the very high positive correlations
(R
2
= 0.999) among Na, Cl and Li, as shown in Fig. 6a,b. In these figures,
the computed Cl concentration of the pre-boiled geothermal liquid is
also shown.
The fluxes of Na, Cl and Li were computed by multiplying their
concentrations by the respective flow rates measured at Rio Varvarco
(upstream and downstream) and Manchana Covunco, Agua Caliente
and Covunco hot creeks. For each species, the sum of the creek fluxes
(Na: 960 g s
−1
; Cl: 1,450 g s
−1
; Li: 8.5 g s
−1
) is systematically lower
than those computed in the Rio Varvarco as the difference between the
downstream and upstream values (Na: 1,190 g s
−1
; Cl: 1,880 g s
−1
;Li:
9.9 g s
−1
). This systematic difference, though less than 25%, could be
due to some uncertainties in the flow rate computation. Alternatively,
part of the creek water may be lost as infiltration into the permeable al-
luvial deposits but later seeps into the river downstream from the con-
fluence or some groundwater inflow to the Rio that includes a thermal
component. Nevertheless, we think that the observed differences are ac-
ceptable. Assuming that the flux calculated for Rio Varvarco is the most
reliable, the computation of the thermal energy release (QH) and related
uncertainties were performed using the simple relation:
QH¼QCl=Clt
ðÞHtð5Þ
where Q
Cl
is the flux of Cl in the river (g s
−1
), Cl
t
is the Cl concentration
in the original thermal liquid (g g
−1
, the unity has been converted
from mg kg
−1
)andH
t
is the enthalpy (J g
−1
) of the liquid at the estimat-
ed reservoir temperature.
We solved Eq. (5) with a Monte Carlo procedure considering the un-
certainties of the flow rate of Rio Varvarco (11,000 and 15,700 kg s
−1
)
and of Cl
t
(1.5–2.0 g kg
−1
). The enthalpy H
t
was computed for an
original temperature of 220 °C ± 10 °C (enthalpy = 944 J g
−1
±
46 J g
−1
), i.e., the temperature resulting from the average Na/K
ratio of the thermal tributaries which enter Rio Varvarco as reported
in the Na-K diagram of Fig. 6c. The Monte Carlo simulation gives a
thermal water total discharge of 1,150 ± 150 kg s
−1
and a thermal
energy release of 1.1 × 10
9
±0.2×10
9
W (1.1 ± 0.2 GW).
6. Conclusion
A geochemical survey of the main thermal waters discharging in the
southwestern part of the Domuyovolcanic complex has highlighted the
strikingly high thermal energy release from this remote site in Patago-
nia. A simple hydrogeochemical approach shows that ~1,150 kg s
−1
of
boiling water is conveyed to the main drainage system (Rio Varvarco)
through its tributaries. Referring to a liquid originally at 220 °C, the calcu-
lated thermal energy release is as large as ~ 1.1 GW, a value that, to the
best of our knowledge, is the second largest advective heat flux ever
measured from a single onland volcanic center (other than Yellowstone).
This result suggests an enormous geothermal potential of the Domuyo
system, which has yet to be exploited.
A previous geochemical survey of the same thermal waters, per-
formed in 1976 (Llambías et al., 1978), reports chemical compositions,
temperatures and thermal features similar to those measured in our
work, suggesting a persistent heat flux. In addition to local sealing of hy-
drothermal channels, this large flux of energy may be the main reason
a)
b)
c)
Fig. 6. Binary diagrams (a) Na vs. Cl, (b) Li vs.Cl and (c) Na vs. K. Symbols as inFi g. 2.The
unlabelled samples are diluted waters close to the axis origin in each diagram. (For inter-
pretationof the references to colourin this figure, the readeris referred to the web version
of this article.)
76 G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77

for the 2003 hydrothermal explosions at El Humazo. Finally, this very
large heat flux is somewhat hard to explain in terms of either the
cooling of mapped magmatic intrusions, which seem to have occurred
at 0.72 Ma, or the most recent volcanic activity at 0.11 Ma. In our opin-
ion, the presence of an undocumented, younger volcanic activity cannot
be excluded. Alternatively, the high heat flux may originate from a
magma intrusion at shallow depth, possibly related to a recent reactiva-
tion of the Domuyo volcanic system.
Acknowledgements
The authorsare gratefully to the personnel of the Domuyo Provincial
Park and to our guide Paulo Fanti. Many thanks are due to the four
Brazilianstudents (Vanessa Araujo, Cristiane Schmidt, Marcello Cappelli
and Hugo Vidal) who accompanied and helped us during the fieldwork.
The authors thank William C. Evans and an anonymous reviewer for
their useful comments and suggestions on the early version of the
manuscript.
References
Bodmer, Ph., Rybach, L., 1985. Heat flow maps and deep ground water circulation: exam-
ples from Switzerland Original. J. Geodyn. 4, 24–233.
Bodri, B., Rybach, L., 1998. Influence of topographically driven convection on heat flow in
the Swiss Alps: a model study. Tectonophysics 291, 19–27.
Brousse, R., Pesc e, A.H., 1982. Cerro Domo: un volcán cuartario con posibilidades
geotérmicas, Provincia del Neuquén. 9° Congreso Latinoamericano de Geología
(Buenos Aires). Actas, 4, pp. 197–208 (Buenos Aires (Argentina)).
Brumm, M., Wang, C.Y., Manga, M., 2009. Spring temperatures in the Sagehen Basin,
Sierra Nevada, CA: implications for heat flow and groundwater circulation. Geofluids
9, 195–207.
Cembrano, J., Lara, L.,2009. The link betweenvolcanism and tectonics in the southernvol-
canic zone of Chilean Andes: A review. Tectonophysics 471, 96–113.
Čermak, V., Jetel, J., 1985. Heat flow and ground water movement in the Bohemian Creta-
ceous basin (Czechoslovakia) Original. J. Geodyn. 4, 285–303.
Chiodini, G., Cardellini, C., Caliro, S., Chiarabba, C., Frondini, F., 2013. Advective heat trans-
port associated to regional Earth degassing in central Apennine (Italy). Earth Planet.
Sci. Lett. 373, 65–74.
Ellis, A.J., Wilson, S.H., 1955. The heat from Wairakei-Taupo thermal region calculated
from the chloride output. N. Z. J. Sci. Technol. Sect. B 36, 622–631.
Facca, G., Tonani, F., 1971. The self-sealing geothermal field. Bull. Volcanol. 30, 271–273.
Folguera, A., Ramos, V.A., Zapata, T., Spagnuolo, M.G., 2007. Andean evolution at the Gua-
ñacos and Chos Malal fold and thrust belts (36°30′–37°S). J. Geodyn. 44, 129–148.
Fournier, R.O., 1989.Geochemistry and dynamics of theYellowstone National Park hydro-
thermal system. Ann. Rev. Earth Planet. Sci. 17, 13–53.
Giggenbach, W.F., 1988. Geothermal solute equilibria. Derivation of Na–K–Mg–Ca
geoindicators. Geochim. Cosmochim. Acta 52, 2749–2765.
Gosnold,W.D., 1990. Heat flow in theGreat Plains of the United States. J. Geophys.Res. 95,
353–374.
Groeber, P., 1947. Hojas Domuyo, Mari Mahuida, Huahuar Co y parte de Epu Lauquen.
Observaciones Geológicas a lo largo del Meridiano 70. Asociación Ge ológica
Argentina. Serie Reimpresiones 1, 75–136 (1980).
Ingebritsen, S.E., Mariner, R.H., 2010. Hydrothermal heat discharge in the Cascade Range,
northwestern United States. J. Volcanol. Geotherm. Res. 196, 208–218.
JICA, 1983.Argentine Republic.Final Report on the NorthernNeuquen Geothermal Devel-
opment Project. First-Second Phase Survey. No 25.
JICA, 1984.Argentine Republic.Final Report on the NorthernNeuquen Geothermal Devel-
opment Project. Third Phase Survey. No 25.
Kay, S., Burns, W., Copeland, P., Mancilla, O., 2006. Upper Cretaceous to Holocene
magmatism and evidence for transient Miocene shallowing of the Andean subduc-
tion zone under the northern Neuquen basin. In: Kay, S.M., Ramos, V.A. (Eds.), Late
Cretaceous to Recent Magmatism and Tectonism of the Southern Andean Margin at
the Latitude of th e Neuquen Basin (36-39 S). Geol. Soc. Am. Special paper, 407,
pp. 19–60.
Lawless, J.V., Browne, P.R.L., 2001. Hydrothermal eruptions: mechanisms implica-
tions for Prediction. Proceedings 23rd New Zealand Geothermal Workshop
2001 51–56.
Llambías, E., Palacios, M., Danderfer, J.C., Brogioni, N., 1978. Las rocas ígneas Cenozoicas
del Volcán Domuyo y áreas adyacentes. Actas VII Congreso Geológico Argentino
Neuquén, 2, pp. 569–584.
Manga, M., 1998. Advective heat discharge by low-temperature discharge in the Oregon
Cascades. Geology 26, 1,813–1,816.
Mas, G.R., Bengochea,L., Mas, L.C., Lopez, N.,2009. Hydrothermalexplosion due to seal ef-
fect in el Humazo geothermal manifestation, Domuyo volcano, Neuquén, Argentina.
Proceeding, Thirty-Fourth Workshop on Geothermal Reservoir Engineering Stanford
University, Stanford, California, February 9-11, 2009 5.
Miranda, F.J., 1996. Caracterización petrográfica y geoquímica del Cerro Domuyo, Pcia. de
Neuquén - Argentina. [Licenciatura Thesis]. Buenos Aires. Universidad de Buenos
Aires, Buenos Aires, Facultad de Ciencias Exactas y Naturales p. 118.
Miranda, F., Folguera,A., Leal, P., Naranjo, J., Pesce, A., 2006. Upper Plioceneto lower Pleis-
tocene volcanic complexes and Upper Neogene deformation in the south-central
Andes (36°30″–38°S). Geol. Soc. Am. Spec. Pap. 407 (specialpapers.gsapubs.org).
Palacio, M., Llambia, E.J., 1978. Las fuentes termales del Volcan Domuyo, Provincia del
Neuquen. 7th Argentine Geol. Congress, 9-15 April, 1978, Neuquen (Argentina), 2,
pp. 145–159 (In Spanish with English abstract).
Sorey, M.L., 1986. Hot spring monitoring at Lassen Volcanic National Park, California
1983–1985. Proceedings of the 11th Workshop on Geothermal Reservoir Engineer-
ing. Stanford University, Stanford, CA, USA, pp. 141–149.
Spagnuolo, M.G., Litvak, V.D., Folguera, A., Bottesi, G., Ramos, V.A., 2012. Neogene mag-
matic expansion and mountain building processes in the southern Central Andes,
36–37 S, Argentina. J. Geodyn. 53, 81–94.
Suárez, M.,Emparan, C., 1995. Thestratigraphy, geochronology andpaleophysiographyof
a Miocene fresh-water interarc basin, southern Chile.J. South Am. Earth Sci. 8, 17–31.
Suárez, M., Emparán, C., 1997. Hoja Curacautín. Regiones de la Araucanía y del Bío Bío.
Carta Geológica de Chile: Servicio Nacional de Geología y Minería de Chile, Santiago,
v. 71, pp. 105, scale 1:250,000, 1 sheet.
Taran, Y.A., Peiffer, L., 2009. Hydrology, hydrochemistry and geothermal potential of El
Chichón volcano- hydrothermal system, Mexico. Geothermics 38 (4), 370–378.
http://dx.doi.org/10.1016/j.geothermics.2009.09.002 (Cited 4 times).
Vakin, E.A., Pilipenko, G.F., 1986. The Mutnovsky geothermal area in Kamchatka. Explora-
tion and Exploitation of Geothermal Resources in Volcanic Areas. Nauka, Moscow,
Russia (36–45 pp. (in Russian)).
Völker, D., Kutterolf, S., Wehrmann, H., 2011. Comparative mass balance of volcanic edi-
fices at the southern volcanic zone of the Andes between 33°S and 46°S. J. Volcanol.
Geotherm. Res. 205, 114–129.
77G. Chiodini et al. / Journal of Volcanology and Geothermal Research 274 (2014) 71–77