RESEARCH ARTICLE Structural Controls on Crustal Fluid Circulation
10.1029/2020GC008919 and Hot Spring Geochemistry Above a
Key Points: Flat‐Slab Subduction Zone, Peru
• The Cordillera Blanca detachment
hosts a hydrothermal system and B. E. Scott1,2, D. L. Newell1 , M. J. Jessup3 , T. A. Grambling3 , and C. A. Shaw4
deep flow paths
• Hot spring aqueous and isotope 1
geochemistry reveal crustal uid Department of Geosciences, Utah State University, Logan, UT, USA,
2Now at Hess Corp., Houston, TX, USA,
fl
3
sources, temperatures, and mixing Department of Earth and Planetary Sciences, University of Tennessee, Knoxville, Knoxville, TN, USA,
4Department of
• Contrasting structural styles in the Earth Sciences, Montana State University, Bozeman, MT, USA
Cordillera Blanca and Cordillera
Huayhuash influence water‐rock
interaction and groundwater Abstract Hot spring geochemistry from the Cordillera Blanca and Cordillera Huayhuash, Peru, reveal
chemistry
the influence of crustal‐scale structures on geothermal fluid circulation in an amagmatic region located
Supporting Information: above a flat‐slab subduction zone. To test the influence of contrasting modes of faulting in these regions,
• Supporting Information S1 springs were targeted along the Cordillera Blanca detachment fault, within its hanging wall, in the footwall
of the detachment, and in the Cordillera Huayhuash. Hot springs along the Cordillera Blanca detachment
fault zone are associated with recent extension and normal faulting, and those in its footwall and the
Correspondence to: Cordillera Huayhuash are located in the Marañon fold and thrust belt where compressional structures
D. L. Newell,
dennis.newell@usu.edu dominate. Springs along and in the hanging wall of the Cordillera Blanca detachment fault yield
brackish‐saline, alkaline‐chloride waters, with oxygen, hydrogen, carbon, and chlorine stable isotope values
that suggest mixing between meteoric groundwater and saline brine affected by high water‐rock interaction.
Citation:
Scott, B. E., Newell, D. L., Jessup, M. J., Geothermometry reservoir temperature estimates (RTEs) of 91–226°C indicate maximum flow path
Grambling, T. A., &Shaw, C. A. (2020). depths of 8.7 or 11 km, depending on geothermal gradient, associated with the Cordillera Blanca detachment
Structural Controls on Crustal Fluid fault. In contrast, springs in the footwall and in the Cordillera Huayhuash exhibit a wide range of water
Circulation and Hot Spring
Geochemistry Above a Flat Slab types with an isotopic affinity to meteoric water, suggesting a greater influence from shallow groundwater‐
Subduction Zone, Peru. Geochemistry, and less water‐rock interaction. For these springs, RTEs of 40–98°C correspond to much shallower
Geophysics, Geosystems, 21, circulation (1.6–4 km). Results indicate that the Cordillera Blanca detachment system accommodates
e2020GC008919. https://doi.org/
10.1029/2020GC008919 significantly deeper circulation of crustal fluids compared to other regional compressional structures.
Received 15 JAN 2020
Accepted 17 JUN 2020
Accepted article online 22 JUN 2020
1. Introduction
Spatially and temporally, fault zones act as conduits, barriers, or combined conduit‐barrier systems to flow,
thereby affecting the distribution, circulation depth, and overall geochemistry of aqueous fluids in the con-
tinental lithosphere (e.g., Bense & Person, 2006; Caine et al., 1996; Hooper, 1991). Hot spring emanations are
commonly located along fault systems, suggesting these structures, at least locally, control the upward flow
of fluids through the crust. Thus, geochemical and isotopic data from thermal springs provide windows into
the provenance and fluid‐rock interaction history of deeply circulated fluids in tectonically active regions.
Prior studies use water and gas composition, and various stable isotopic systems (e.g., H, O, C, and He) to
demonstrate that faults act as deep conduits into the continental lithosphere with connections to the mantle
in some cases. For example, these natural tracers in springs have been investigated at a variety of tectonic
settings, such as major strike‐slip fault systems (San Andreas, Anatolia, and Karakorum faults) (e.g., de
Leeuw et al., 2010; Kennedy et al., 1997; Klemperer et al., 2013; Kulongoski et al., 2013; Mutlu et al., 2008),
extensional settings (Rio Grande rift, Basin and Range, and East African rift) (e.g., Barry et al., 2013; Crossey
et al., 2009; Darling et al., 1995; Kennedy & van Soest, 2007; Newell et al., 2005), and faults along convergent
margins (Himalaya‐Tibet and Andean orogens) (e.g., Evans et al., 2008; Hoke et al., 2000; Hoke &
Lamb, 2007; Newell et al., 2008; Newell et al., 2015). These studies highlight that many active strike‐slip
and normal faults aid in the transfer of deep crustal and mantle‐derived volatiles to the surface in the
absence of recent magmatism. These studies also illustrate that spring geochemistry informs the
©2020. American Geophysical Union. temperature of thermal fluid reservoirs, the potential depth of groundwater infiltration along faults, and
All Rights Reserved. the influence of deeply circulated fluids on groundwater quality.
SCOTT ET AL. 1 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 1. Thermal springs investigated along the Cordillera Blanca detachment (yellow triangles; CBD), the hanging
wall (red triangles; HW), the footwall of the Cordillera Blanca (purple triangles; FW), and the region near the
Cordillera Huayhuash (blue triangles; CHY). Spring surface temperature is indicated. These data are combined with
previously published geochemistry data from the Cordillera Blanca (Newell et al., 2015). Generalized surface geology
modified from Giovanni et al. (2010), and the Peru Carta Geológica Nacional Escala 1:100,000 (see supporting
information Table S5 for citations to maps).
Here we explore the variations in thermal spring geochemistry in the central Peruvian Andes, a region
located above the flatly subducting Nazca plate. We are motivated by recent work on hot springs along
the Cordillera Blanca detachment, a ~200‐km‐long southwest‐dipping normal fault system, that identified
a CO2‐rich, saline thermal system carrying up to 25% mantle‐helium (Newell et al., 2015). Such levels of
mantle helium are enigmatic in this tectonic environment, where the last phase of magmatism was ~9–
4.5 Ma (Giovanni et al., 2010; Petford & Atherton, 1992), but are interpreted to require a present‐day volatile
flux facilitated by the ascent of slab‐ or mantle‐derived fluids (Newell et al., 2015). The Cordillera Blanca
SCOTT ET AL. 2 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
detachment fault exhibits major structural control on the westernmargin of the Cordillera Blanca, providing
a likely conduit to move these deeply sourced fluids to the surface. However, it is not known if these fluids
are unique to the Cordillera Blanca detachment fault or are more widespread in the subsurface away from
major fault zones. This work tests the hypothesis that different fault‐controlled fluid pathways play a
primary role in controlling variations in the geochemistry of crustal fluids. Specifically, we target the greater
Cordillera Blanca and Cordillera Huayhuash, adjacent mountain ranges hosting numerous hot spring
emanations with distinctly different styles of faulting.
In this contribution we present new aqueous and stable isotope geochemistry data from thermal spring loca-
tions in the Cordillera Blanca and Cordillera Huayhuash regions of Peru (Figure 1). We expand prior work
that focused specifically on the Cordillera Blanca detachment fault‐related hot springs (Newell et al., 2015)
to include locations east of the Cordillera Blanca massif, and southeastward into the Cordillera Huayhaush,
where hot springs emanate from Marañon fold and thrust belt‐related thrust faults. We frame the data as a
function of geographic location, highlight notable geochemical differences and similarities between hot
springs, and explore the influence of crustal‐scale structures, or lack of such features, on deeply circulating
fluids in the Peruvian Andes.
2. Geological Setting
2.1. Tectonic Setting
The Cordillera Blanca and Cordillera Huayhuash represent contrasting structural and lithologic domains
within the modern amagmatic, Peruvian flat‐slab segment of the Andean subduction zone. Both ranges
are notable for their high topography with numerous peaks exceeding 6,000 m, including Peru's highest
peak, Huascaran (6,655 m) in the Cordillera Blanca, and the second highest peak, Nevado Yerupaja
(6,634 m) in the Cordillera Huayhuash. The early development of the Cordillera Blanca and Cordillera
Huayhuash followed broadly similar tectonic histories from Paleozoic and Mesozoic basin development
through the late Cretaceous to Paleogene thin‐skinned deformation that produced the Marañon fold thrust
belt (MFTB) (Mégard, 1984; Scherrenberg et al., 2012). The tectonic evolution of the ranges diverged in the
Neogene with batholith emplacement starting ca. 14 Ma, large‐scale extensional faulting, and footwall basin
development in the Cordillera Blanca (Giovanni et al., 2010). As discussed below, these events led to the
development of a distinct deep crustal architecture in the Cordillera Blanca that is not expressed in the
Cordillera Huayhuash. This study examines the role that this deep crustal architecture plays in controlling
the evolution of hydrothermal systems that have developed over the Peruvian flat slab.
The MFTB belt dominates the structural style of the central part of the Andes in Peru (Mégard, 1984;
Scherrenberg et al., 2012; Scherrenberg et al., 2016). Subsequent to MFTB deformation, Tertiary arc magma-
tism produced a thick sequence of dominantly intermediate volcanic rocks in the Cordillera Negra (Figure 1)
and the later emplacement of the leucogranodioritic Cordillera Blanca batholith in the Miocene (Mukasa,
1984; Petford & Atherton, 1992). During this time, shallowing of the subducting Nazca plate is recorded
by eruption of the adakitic Yungay and Fortaleza ignimbrites (Giovanni et al., 2010; Petford &
Atherton, 1992), and the eastward migration of arc volcanism before the cessation of magmatism in the
Pliocene as the mantle wedge closed (Gutscher, 2002; Margirier et al., 2017; Ramos & Folguera, 2009).
2.2. Cordillera Blanca
The Cordillera Blanca is a northwest‐southeast‐trending massif hosting some of the highest elevations in the
Andes. The range is cored by the Cordillera Blanca batholith and has been exhumed by the
southwest‐dipping Cordillera Blanca detachment fault (Figure 1). Granodiorite of the Cordillera Blanca
batholith that was emplaced between ~14 and 5 Ma is the main rock type of the footwall (Giovanni, 2007;
Mukasa, 1984; Petford & Atherton, 1992). In addition to these intrusive rocks, siliceous volcanic units (ca.
9–4.5 Ma Yungay and Fortaleza ignimbrites and 5.4 Ma Lloclla tuff) in the area represent the most recent
arc‐related magmatic events (Cobbing, 1981; Coldwell et al., 2011; Giovanni et al., 2010). Sedimentary rock
units exposed in footwall of the Cordillera Blanca detachment fault include Jurassic shales and Cretaceous
carbonates and quartzites that were folded and faulted in the MFTB prior to Cordillera Blanca batholith
emplacement (Figure 1) (Giovanni et al., 2010). These Mesozoic units are also present in the hanging wall
and are overlain by Eocene‐Miocene volcanic rocks (Calipuy Formation), the Yungay ignimbrites, and
Miocene‐Pliocene basin‐fill sediments (Giovanni et al., 2010). The Jurassic sedimentary rocks locally
SCOTT ET AL. 3 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
intruded by the Cordillera Blanca batholith exhibit low‐grade contact metamorphism to phyllites (Atherton
& Sanderson, 1987).
The Cordillera Blanca detachment fault is a prominent feature that trends NE‐SW for ~200 kmwith a 20–45°
westerly dip (Figure 1) (McNulty & Farber, 2002). The detachment is well exposed near the mouth of
glacially cut valleys (quebradas) along the western slopes of the Cordillera Blanca. North of Huaraz, the
structure is characterized by an extensional shear zone with a gradient from mylonitic fabrics to brittle
overprinting toward to the detachment surface (Giovanni et al., 2010; Hughes et al., 2019). South of
Huaraz the mylonitic zones are rare to absent, and the fault breaks into segments, terminating ~80 km to
the SE (Figure 1). 40Ar/39Ar biotite dates from the Lloclla tuff in the hanging wall constrain the onset of nor-
mal faulting at 5.4 ± 0.1 Ma (Giovanni et al., 2010), followed by rapid batholith exhumation from ~5 to 2 Ma
(Margirier et al., 2015; Margirier et al., 2016). Recent fault movement is apparent from extensive Quaternary
scarps, and paleoseismic investigations suggest that the most recent activity was 2,400–750 kyr ago (Sébrier
et al., 1988). Additionally there are multiple hot springs emanations from the Cordillera Blanca detachment
fault and other normal faults located in the hanging wall (Figure 1) (Newell et al., 2015).
2.3. Cordillera Huayhuash
The Cordillera Huayhuash is ~60 km southeast of the Cordillera Blanca and exhibits a contrasting structural
style (Figure. 1). The Cordillera Huayhuash comprises Cretaceous carbonates and lesser Jurassic siliciclastic
rocks, generally correlated with units in the Cordillera Blanca. Igneous units include Oligocene‐Miocene
volcanic rocks and Miocene granitic plutons (Coney, 1971; Garver et al., 2005). A chain of plutons can be
traced from the Cordillera Huayhuash north into the Cordillera Blanca, yet the Huayhuash lacks intrusions
on the scale of the Cordillera Blanca batholith. Siliceous volcanism in the Cordillera Huayhuash at ~6 Ma
(Puscanturpa volcanics) is similar in age to the Yungay and Fortaleza ignimbrites in the Cordillera Blanca
region (Garver et al., 2005), but it is unknown if these are correlated in terms of source melt. Structurally
the Cordillera Huayhuash is dominated by exhumed MFTB folds and thrust faults. The recent extension
and exhumation that characterizes the Cordillera Blanca is not apparent in this region. These prominent dif-
ferences and the presence of thermal springs in the Cordillera Huayhuash provide a natural laboratory to
interrogate structural controls on fluid circulation in central Peru.
3. Materials and Methods
3.1. Thermal Spring Sampling
New aqueous and stable isotope geochemistry data were acquired from 18 thermal springs located along the
hanging wall of the Cordillera Blanca detachment, in the footwall of this detachment and further to east of
the Cordillera Blanca, and in the Cordillera Huayhuash (Figure 1 and Table S1). These data are combined
with previously published data from seven thermal springs along the trace and hanging wall of the
Cordillera Blanca detachment fault (Table S1) (Newell et al., 2015). To frame the data analysis and discus-
sion, we subdivide the springs into four groups based on geographic location: CBD, HW, FW, and CHY,
defined as follows. CBD group springs include springs emanating along the trace of the Cordillera Blanca
detachment fault, and HW springs are located in the hanging wall and within 10 km (west) of the fault trace
(Figure 1). FW comprises springs within the footwall to the north and east of the Cordillera Blanca massif
(~10–60 km from the fault trace), and CHY is springs located southeastward into the Cordillera
Huayhuash (Figure 1).
All springs were sampled as close to the source as possible to reduce the effect of cooling, degassing, atmo-
spheric exchange, and evaporation at the surface. At several locations, we observed discharge as a group of
spring emanations that may be controlled by a combination of localized structures and/or permeability var-
iations. In these cases, field parameters (T, pH, and conductivity) were used to screen sample locations, and
generally, the location with the highest temperature and/or conductivity was chosen for sampling
(Table S1). In addition to hot spring samples, nearby cold meteoric water (e.g., streams and lakes) was col-
lected, if present, for oxygen and hydrogen stable isotope analysis (Table S4).
Standard techniques for surface and groundwater sampling were employed in this study (e.g., USGS, 2006).
Temperature, pH, and conductivity of spring waters were measured in the field using an Oakton
pH/conductivity/temperature portable meter. Samples for water major and trace element chemistry, and
SCOTT ET AL. 4 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
stable isotope ratios of chlorine were collected in 60 and 125 ml high‐density polyethylene bottles (cation
samples filtered with 0.45 μmdisposable syringe membrane filters; anions and alkalinity collected unfiltered
with no headspace). Cation samples were acidified with trace‐metal grade HNO3. Hot spring samples for car-
bon, oxygen, and hydrogen stable isotope analyses were collected in 30 ml amber glass vials with no head-
space. Cold meteoric water was collected in 12 ml glass septa vials with no headspace for oxygen and
hydrogen stable isotope analysis.
3.2. Analytical Techniques
Major and trace element concentrations in spring water were determined at the Utah State University (USU)
Water Research Laboratory using a Dionex Ion Chromatograph (anions) and Agilent ICP‐MS (cations). The
minimum reporting limits for major ions and trace elements are reported in Tables S2 and S3. The uncer-
tainty on major cations and anions is ~ ±2 mg/L and for trace elements is ±1.8 μg/L. Total alkalinity was
measured by manual colorimetric titration (±0.05 mg/L) (USGS, 2006). Geochemist's Workbench
(Bethke, 2006) was used to calculate ion (charge) balances and total dissolved solids (TDS) from the water
analytical results (Table S2).
Oxygen, hydrogen, and carbon stable isotope ratios were measured at the USU Stable Isotope Laboratory by
continuous‐flow isotope ratio mass spectrometry (CF‐IRMS) using a Thermo Scientific Delta V Advantage
IRMS and Gasbench II. CO2 equilibration and H2 equilibration with Pt reduction methods were used to
acquire oxygen and hydrogen stable isotope ratios, respectively. Results are reported using delta notation
(δ18O and δ2H values) in per mil (‰) relative to Vienna Standard Mean Ocean Water (VSMOW) based on
internal laboratory standards, calibrated using VSMOW and Vienna Standard Light Antarctic
Precipitation (VSLAP). Based on replicate analyses of internal water standards, δ18O and δ2H precisions
were ±0.10‰ and ±2.0‰, respectively.
Carbon stable isotope (δ13C) values were measured from dissolved inorganic carbon (DIC) in spring water
samples using a modified phosphoric acid method (Salata et al., 2000). In brief, water samples were injected
into helium ®‐flushed 12 ml Exetainer vials containing ~103% phosphoric acid and equilibrated for 24 hr at
25°C. The liberated CO2 was analyzed using CF‐IRMS. The calibration standards for this method are solid
carbonates, and in order to follow the principle of identical treatment, a DIC standard using
reagent‐grade NaHCO3 was also prepared, calibrated, and analyzed along with the samples. The difference
between the dissolved and solid NaHCO3 was 0.6‰. Samples were calibrated using international standards
(NBS 19, LSVEC), corrected for the isotopic difference between solid and dissolved carbonate, and are
reported in per mil (‰) versus Vienna Peedee belemnite (VPDB), with a precision of ±0.1‰ determined
by repeat analysis of in‐house calcite standards.
Chlorine stable isotope ratios (δ37Cl values) were measured by IRMS equipped with a CH3Cl purification
line at the University of Texas at Austin (e.g., Barnes et al., 2009). δ37Cl values are reported in per mil (‰)
relative to Standard Mean Ocean Chlorine (SMOC) based on internal laboratory standards. Analytical pre-
cision of the δ37Cl values is ±0.2‰, based on long‐term analyses of three seawater standards and one rock
standard.
Additional methodological information on creating the binary and ternary mixing models and computing
the geothermometry estimates are provided in the supporting information.
4. Results
4.1. Aqueous Geochemistry
Spring water temperature, pH, and conductivity range from 16.9–88.9°C, 5.0–7.9, and 170–23,000 μS/cm,
respectively (Table S1). Except for one HW spring (Shangol, Ca+Na‐SO4 type), the CBD and HW springs
are best described as Na‐Cl type waters (Figure 2). In contrast, FW and CHY springs display a wider range
of water types (Na‐HCO3, Ca‐SO4, Ca‐Cl, Na‐Cl, and Ca‐HCO3) (Figure 2). CBD and HW Na‐Cl waters also
have relatively high concentrations of TDS (up to 15,510 mg/L) compared to FW and CHY springs with TDS
< ~1,000 mg/L (excluding Baños Jocos Peinado). Baños Jocos Peinado is a Ca‐SO4 water with 2,840 mg/L
TDS, and the highest SO 2−4 concentration (1,452 mg/L) was observed. Major element chemistry is provided
in Table S2.
SCOTT ET AL. 5 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 2. Piper diagram (Piper, 1944) illustrating the major ion chemistry (Table S2) in thermal springs divided into
groups based on geographic location. In the middle parallelogram, the total dissolved solids (TDS, mg/L) are scaled
by circle size. Note that the CBD and HW springs are dominated by Na‐Cl water types and have much higher TDS than
FW and CHY springs.
Trace element concentrations vary between springs, but notable constituents found in all groups include As,
Ba, Be, Fe, Mn, Sb, Sr, Tl, and Zn (Table S3). CBD and HW springs generally yield higher concentrations of
trace elements than FW and CHY springs. For example, dissolved arsenic in CBD and HW springs is present
up to 10,800 and 2,600 ppb, respectively, as compared to < 600 ppb in FW and CHY springs.
4.2. Stable Isotope Geochemistry
New oxygen (δ18O) and hydrogen (δ2H) stable isotope measurements from the investigated hot springs are
reported with previously published spring data (Newell et al., 2015) (Table 1). By spring group, the δ18O and
δ2H values range from −14.4‰ to −12.6‰ and −112.9‰ to −94.4‰ (CBD), −14.7‰ to −4.9‰ and −
111.0‰ to −74.3‰ (HW), −15.3‰ to −13.1‰ and −116.1‰ to −93.9‰ (FW), and −16.8‰ to −
14.4‰ and −129.9‰ to −114.5‰ (CHY), respectively. A majority of springs from CBD, FW, and CHY have
isotopic values that fall close to the Global Meteoric Water Line (GMWL) (Craig, 1961) (Figure 3a). More
specifically, CHY springs have the lowest isotopic values falling along the GMWL. In contrast, most HW
springs form a trend with a slope less than meteoric water and are positively correlated (slope of 3.1, r2
= 0.9) (Figure 3a).
We also report δ18O and δ2H values from sources representative of local meteoric water, and these are com-
bined with values reported by Newell et al. (2015) and Mark and McKenzie (2007) to assess the composition
of local waters in the Cordillera Blanca and Cordillera Huayhuash regions (Figure 3a and Table S4). Note
that two samples of meteoric water have δ18O and δ2H values shifted significantly to the right of the
GMWL (Figure 3a). One of these samples is sourced from a small pond with no outlet near the town of
SCOTT ET AL. 6 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Table 1
Stable Isotope Values (O, H, C, and Cl) for Thermal Springs
δ18O VSMOW δ2H VSMOW δ13C VPDB δ37‰ ‰ ‰ Cl ‰ SMOC
Spring Sample ID (±0.10) (±2.0) (±0.1) (±0.2)
CBD springs
Huancarhuaz DN13CB11a −13.3 −103.7 −10.7 —
Aquilina DN13CB14a −14.2 −107.3 −10.7 —
Huancarhuaz DNCB15‐07 −13.6 −105.9 −10.8 0.1
Huancarhuaz DNCB15‐08 −12.6 −94.4 — −0.3
Aquilina DNCB15‐9b −14.4 −112.9 — −0.5
Pacatque DNCB15‐10 −14.3 −102.6 — 0.4
Pumapampa DNCB15‐12 −14.7 −111.0 — 0.1
HW springs
Olleros 1 DN13CB04a −7.8 −81.1 — —
Olleros 2b DN13CB05a −6.4 −74.3 −7.8 —
Olleros 4 DN13CB06a −4.9 −79.8 −10.3 —
Merced DN13CB07a −12.6 −94.4 −6.3 —
Chancos DN13CB08aa — — −9.0 —
Chancos DN13CB08ba −10.5 −85.9 −9.7 —
Monterrey DN13CB09a −10.2 −87.7 −10.0 —
Shangol DN13CB13a −13.0 −98.5 — —
Recuay DNCB15‐13 −6.3 −78.8 — −0.4
FW springs
Aticara DNCB15‐11 −13.1 −93.9 — —
Jacobs DNCB17‐01 −14.3 −104.2 −3.7 —
(Pomabamba)
Pomabamba DNCB17‐02 −14.5 −110.1 −2.8 0.2
laundry
Chilhuan DNCB17‐06 −14.9 −109.7 −9.6 −0.1
Rupac DNCB17‐7b −15.2 −107.3 −8.7 —
Rupac DNCB17‐09 −15.3 −112.6 −7.6 0.7
Jocos Peinado DNCB17‐13 −13.8 −103.3 −1.6 −0.6
Chavin DNCB17‐21 −14.6 −116.1 −2.1 0.2
CHY springs
Conococha DNCB17‐24 −14.9 −117.8 −12.0 0.3
Azulmina DNCB17‐25 −16.0 −123.9 −1.2 0.1
Taurimpampa DNCB17‐30 −14.4 −114.5 −1.3 0.2
Banos (Batan) DNCB17‐32 −16.8 −129.9 −2.9 0
Conoc (La DNCB17‐36 −15.3 −118.2 −5.4 −0.1
Union)
Machaycancha DNCB17‐37 −16.1 −120.3 −7.3 −0.2
Janac DNCB17‐38 −16.6 −125.4 −7.0 0.1
Note. Em dashes denote not determined values.
aReported in Newell et al. (2015).
Quiches (this study), and the other is sourced from Laguna Conococha, a closed basin lake between the
Cordillera Blanca and Cordillera Huayhuash (Mark & Mckenzie, 2007). Evaporative processes in these
closed basin water bodies result in isotopic evolution along a slope less than the GMWL (Craig &
Gordon, 1965). The remainder of the meteoric water samples fall close to the GMWL trend.
The δ13C values of DIC were measured from a subset of thermal spring water samples (Table 1). CBD and
HW springs yield a relatively narrow range in δ13C values (−10.8‰ to −6.3‰) and wide range in δ18O
values (−14.7‰ to −4.9‰), compared to FW and CHY springs that yield a much wider range in δ13C values
(−12.0‰ to −1.2 ) and narrower range in δ18‰ O values (−16.8‰ to −13.8‰) (Figure 3b and Table 1).
4.3. Halogen Geochemistry
To aid in evaluating the source of salinity in the springs, additional data on halogens including Br− concen-
tration (Table S2) and δ37Cl values (Table 1) weremeasured in a subset of thermal springs. Based on this ana-
lysis, spring waters exhibit a range of Cl/Br molar ratios, with CBD (626–1,148) and HW (1,038–1,410)
SCOTT ET AL. 7 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 3. (a) δ18O versus δ2H values for thermal springs and local meteoric surface water (streams, lakes, and cold
springs) in the Cordillera Blanca region. Mark and McKenzie (2007) data comprise glaciated and nonglaciated
meteoric waters strictly from the Cordillera Blanca (Table S4). HW springs form a linear trend to values higher than local
meteoric water and most other springs cluster along GMWL. (b) δ13C (dissolved inorganic carbon) and δ18O values
of springs in the Cordillera Blanca and Huayhuash.
springs yielding ratios on average higher than FW (228–1,095) and CHY (331–586) springs (Figures 4a and
4b). Compared to the seawater molar ratio of 647 (Walter et al., 1990), CBD and HW springs yield higher
ratios, and FW and CHY ratios are lower (excluding Baños Chihuan). The published halogen
concentration data from thermal springs emanating along other representative active volcanic arcs are
included for comparison (Figures 4a–4d), including the Taupo Volcanic Zone, New Zealand (Bernal
et al., 2014), and Cascadia (Cullen et al., 2015). Cl/Br ratios from these arc‐related springs are similar to
the ratios observed in CBD and HW springs (Figure 4b).
Additionally, we note that the δ18O values from all the investigated hot springs exhibit a positive linear cor-
relation (r2 = 0.8) with Cl− concentration (Figure 4c). Note that omitting the three HW samples with the
highest δ18O and Cl− values from the linear regression results in the same linear relationship but a weaker
correlation (r2 = 0.3). Similarly, δ18O values and Cl− concentrations from other volcanic arc‐related hot
springs are positively correlated, and for reference, Cascadia hot springs overlap with CBD and HW springs
from this study (Bernal et al., 2014; Cullen et al., 2015) (Figure 4c).
Chlorine stable isotope ratios (δ37Cl values) fall in a relatively narrow range from −0.6 to 0.7 (±0.2)‰ and
reveal little differences between groups (Table 1). δ37Cl values and Cl− data from this study are compared to
a chlorine‐source ternary mixing model (Li et al., 2015), alongside data from other arc‐related hot springs
with similar data sets for comparison (Bernal et al., 2014; Cullen et al., 2015; Li et al., 2015) (Figure 4d).
SCOTT ET AL. 8 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 4. Evaluation of salinity sources in hot springs along the Cordillera Blanca and Huayhuash. Other arc‐related systems are shown for reference to aid in
interpreting data (Bernal et al., 2014; Cullen et al., 2015; Li et al., 2015). (a) Cl− versus Br− , with the molar ratio of seawater (Cullen et al., 2015) and Midwestern
U.S. precipitation for reference (Panno et al., 2006) denoted by dashed lines. CBD and HW springs generally exhibit Cl/Br molar ratios greater than seawater,
consistent with other arc‐related springs. (b) CBD springs can be described by a local mixing trend between a saline and dilute endmember. Cl− and Cl/Br
endmembers that define the binary mixing curve: 3.5 and 50 mM (dilute) and 200 and 1,200 mM (saline), with the percent of the saline endmember indicated.
(c) Data from this study exhibit a positive correlation between δ18O values and Cl− content with a linear best fit (r2 = 0.8). Note that by omitting the three
highest values from this trend results in a nearly identical linear fit, but with a weaker correlation (r2 = 0.3). Also note that the Cascadia arc springs have a similar
trend, but the Taupo arc samples have much higher δ18O. (d) δ37Cl versus Cl− ternary mixing model (after Li et al., 2015) showing results from this study
compared to the Cascade (Cullen et al., 2015), Taupo (Bernal et al., 2014), and Lesser Antilles (Li et al., 2015) arcs. Some Cordillera Blanca and Huayhuash springs
fall in the ternary mixing field between meteoric (0.05 mM, 0‰), seawater (500 mM, 0‰), and magmatic (20 mM, −0.65‰) endmembers. However, the
scatter above the meteoric water‐seawater mixing curve may be controlled by geothermal water‐rock iteration with igneous rocks (star). Other processes such as
boiling and devolatilization are needed to capture to full range of δ37Cl seen at Cascadia (Cullen et al., 2015).
In our data set, the primary difference observed is in Cl− content, with far higher concentrations in CBD and
HW than the FW and CHY springs. Also, only some of the data fall within the ternary mixing field. Springs
from the Lesser Antilles and several from the Taupo arc conformwell to the ternarymixingmodel. However,
similar to data in our study, many Cascadia and Taupo arc springs fall outside of the mixing field in terms of
δ37Cl values. Some of these data patterns are attributed to water‐rock interaction with basalts or another
process such as boiling and devolatilization (Cullen et al., 2015).
4.4. Geothermometry
Conventional silica, Na‐K, Na‐K‐Ca, and K‐Mg geoindicators and geothermometers were utilized to esti-
mate subsurface reservoir temperature estimates (RTEs) for each thermal spring. These thermometers are
SCOTT ET AL. 9 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Table 2 theoretical and empirical relationships based on the assumption of
Reservoir Temperatures (°C) Calculated Using Silica and Cation water‐rock chemical equilibrium at depth, and the preservation of
Geothermometers for Thermal Springs
equilibrium during migration to the surface (e.g., Arnorsson
Na‐ et al., 1983; Fournier, 1977; Fournier & Truesdell, 1973;
Surface Quartz Chalcedony Na‐ K‐ K‐
Giggenbach, 1988). Silica geothermometers are based on the solubi-
Spring T (°C) conductive conductive K Ca Mg
lity of silica species (i.e., quartz and chalcedony) in water as a func-
CBD springs
a tion of mainly temperature (Fournier, 1977). Thus, relatively rapidHuanacarhuaz 73.3 169 141 260 226 167
a reequilibration and silica precipitation during cooling along flowAquilina 78.3 140 112 200 152 121
Pacatque 88.9 220 168 132 paths can cause problems in the accuracy of this geothermometer— —
Pumapampa 19.2 — — 313 176 118 (Rimstidt & Barnes, 1980). Assuming no steam loss, the conductive
HW springs quartz geothermometer is best suited for reservoir temperatures
Ollerosa 47.8 — — 265 245 172
a >150°C, whereas the conductive chalcedony geothermometer is aMerced 38.4 85 57 263 91 71
a better approximation for temperatures < 150°C. The Na‐K cationChancos 47.6 140 112 275 209 137
Monterreya 46.4 257 229 147 geothermometer is based on the equilibrium between hydrothermal— —
Shangola 39.6 87 58 234 108 83 fluids and feldspars, and the Na‐K‐Ca system assumes that calcite is
Recuay 16.9 — — 242 217 136 present in the system (Giggenbach, 1988). The advantage to the
FW springs Na‐K geothermometer is that it reequilibrates more slowly than the
Aticara 36.6 — — 120 43 55 silica geothermometer and can preserve temperatures from deeper
Pomabamba 50.5 113 84 273 79 58
Chihuan 36.5 89 60 189 44 50 in the system. The Na‐K geothermometer is most applicable to waters
Rupac 61.5 101 73 348 48 45 with reservoir temperatures >100°C and Ca content that meets the
Jocos Peinado 40.6 67 38 326 66 65 criteria: log[Ca1/2/Na] + 2.06 < 1 (Fournier & Truesdell, 1973).
Chavin 41.0 65 36 248 42 46 The Na‐K‐Ca geothermometer may be more appropriate for higher
CHY springs
Conococha 28.0 67 38 226 44 51 concentrations of Ca that meet the criteria: log[Ca
1/2/Na] + 2.06
Azulmina 70.7 110 81 285 78 69 > 1, and assumes the conversion of Ca‐plagioclase to calcite
Tauripampa 45.3 106 77 288 80 76 (Fournier & Truesdell, 1973). Unlike the Na‐K and Na‐K‐Ca
Baños Batan 64.7 91 62 284 98 77 geothermometers, the K‐Mg geothermometer reequilibrates rapidly
Conoc 42.2 78 49 294 42 44 at cooler temperatures, thereby preserving a cooler and likely
Machayacancha 44.6 98 69 333 43 57
Janac 51.5 127 99 351 40 56 shallower fluid‐rock interaction signal (Giggenbach, 1988). The
K‐Mg geothermometer is also more influenced by mixing with
Note. Em dashes denote Si not measured.
aReported in Newell et al. (2015). shallow groundwaters containing dissolved Mg. Acknowledging the
applicability and limitations of these thermometers, differences and
similarities between geothermometry temperatures can provide valu-
able information about the circulation depth and degree of fluid‐rock interaction in thermal spring fluids.
Reservoir temperatures for the four groups of hot springs are estimated using silica and various cation
geothermometers (Table 2). Several distinctions between geothermometers are observed across all spring
groups. Springs, excluding Baños Huancarhuaz, yield quartz geothermometer temperatures < 150°C, indi-
cating that the chalcedony geothermometer is more appropriate (Fournier, 1977). Temperatures yielded by
the chalcedony geothermometer (36–141°C) are similar to K‐Mg temperatures (44–172°C) and generally
lower than corresponding Na‐K (120–351°C) and Na‐K‐Ca (40–245°C) temperatures. Note that dissolved
silica was not measured on the 2015 samples (Table S2), and thus, silica temperatures are not calculated
for these springs. A ternary diagram comparing Na‐K and K‐Mg temperatures (Giggenbach, 1988) shows
that several CBD and CH springs and one FW spring plot in the “partial equilibrium field”, yielding Na‐K
temperatures from 200–275°C and K‐Mg temperatures from 71–172°C (Figure 5). The majority of FW and
all CHY springs plot as “immature waters” with RTE and proportionally higher amounts of dissolved Mg
(Figure 5).
Determining the reliability of these estimates is challenging, but we observe some patterns. Surface tempera-
tures from all CBD springs (73–89°C, with one outlier of 19°C at Pumapampa spring) are notably higher than
HW (17–48°C), FW (37–62°C), and CHY (28–71°C) springs (Table 1). Reservoir temperatures calculated
using any of the geothermometers are generally higher in CBD and HW than in FW and CHY (Table 2).
Note that four springs from FW and CHY (Rupac, Conoc, Machayacancha, and Janac) yield Na‐K‐Ca tem-
peratures that are lower than their corresponding surface temperature. We observe this same contradicting
relationship from the chalcedony geothermometer (Jocos Peinado, Chavin, and Batan) and K‐Mg geotherm-
ometer (Rupac and Azulmina), strictly among the FW and CHY springs. RTEs that are erroneously low are a
SCOTT ET AL. 10 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 5. Geothermometry ternary diagram of the Na‐K and K‐Mg geothermometers (modified from Giggenbach, 1988).
FW and CHY springs plot as immature waters in the Mg corner, and CBD and HW springs exhibit a range of trends,
including along the 200–260°C Na‐K isotherms in the partial equilibrium field.
common result of thermal fluid disequilibrium due to cooling along flow paths and mixing with shallow
groundwater (Karingithi, 2007); thus, these estimates are unreliable.
5. Discussion
Aqueous and stable isotope geochemistry data are compared from spring groups (CBD, HW, FW, and CHY)
to discern the role of major geologic structures like the Cordillera Blanca detachment fault as a conduit for
deeply circulated fluids. CBD springs issue directly along strike of the fault, and springs located in the hang-
ing wall (HW springs) are interpreted to issue from steep normal faults that likely intersect with the detach-
ment fault at depth (Newell et al., 2015). Compared to FW and CHY springs, CBD and HW springs yield
predominantly Na‐Cl water types, high TDS, elevated trace metals, similar stable isotope geochemistry,
and a higher range of RTE based on geothermometry. Geochemical and stable isotope similarities between
CBD and HW springs support an interconnected fracture network promoting deep fluid pathways for these
two groups. Data from CBD and HW also suggest mixing between meteoric groundwater and saline, rela-
tively high‐temperature geothermal fluids, facilitated by these interrelated flow paths. In contrast, data from
FW and CHY springs suggest much shallower circulation and fluid‐rock equilibration with the sedimentary
host rock. Sections 5.1–5.3 examine these observations in detail to reveal how differences in regional faulting
styles influence depths of fluid migration, thereby affecting fluid‐rock interactions and resulting fluid
geochemistry.
5.1. Geochemistry of CBD and HW Springs
These springs are dominated by Na‐Cl water types with TDS concentrations up to 15,510 mg/L (Figure 2).
These data suggest a relatively high TDS, Na‐Cl rich source, and/or fluid‐rock reactions controlling their
water chemistry. Na‐Cl water types can result from meteoric recharge mixing with deeply sourced hydro-
thermal brine, similar to processes observed at active volcanic arcs (Fournier, 1987; Giggenbach, 1990).
Volatiles derived from emplaced magma can produce brine; however, there is no evidence of recent magma-
tism in this setting. Alternatively, fluid interaction with sodic plagioclase (albite) and alteration minerals
(e.g., chlorite) can yield a Na‐Cl brine characteristic of deeply circulating fluids in a granite reservoir
(Bucher & Stober, 2010; Kühn, 2004; Pepin et al., 2015). Possible origins for Cl− in this setting include
SCOTT ET AL. 11 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
residual Cl‐rich fluids partitioned from earlier magmatism (~14 to 5 Ma) and/or prolonged circulation of
dilute fluids through hot rocks with a history of magma injections (Fournier, 1987).
Leaching of evaporites in host rocks along the flow paths can also result in Na‐Cl‐dominated waters. Halide
content and Cl/Br ratios are informative as geochemical tracers for sources of chloride and thus crustal
brines (Bernal et al., 2014; Hanor, 1987; Kesler et al., 1996; Leisen et al., 2012; Panno et al., 2006; Walter
et al., 1990). Leaching of evaporite deposits by circulating fluids would result in Cl/Br molar ratios that
are substantially higher than the seawater ratio of 647. Dissolution of halite results in Cl/Br molar ratios
>10,000 because Br− does not readily substitute into halite (Banks et al., 2000; Leisen et al., 2012; Walter
et al., 1990). The CBD and HW spring Cl/Br ratios are about 2× the seawater ratio (Figure 4a), implying that
if evaporite dissolution is contributing to the salinity of these springs, it is a minor source. This is consistent
with the lack of reported evaporite units in the sedimentary rock types and supradetachment basin‐fill sedi-
ments in the hanging wall of the Cordillera Blanca detachment (Giovanni et al., 2010).
Serpentinites (0.2–0.5 wt% Cl) are another potential source of Cl− in subduction zones (e.g., Kendrick
et al., 2011; Sharp & Barnes, 2004). Dehydration of a subducting flat slab and transfer of seawater‐derived,
Cl‐rich fluids to the overriding lithosphere may be a source for the Na‐Cl water types (Butcher et al., 2017;
Hoke & Lamb, 2007; Humphreys et al., 2003). If this process contributes to the localized distribution of Na‐
Cl‐type thermal groundwater along the trace and hanging wall of the Cordillera Blanca detachment fault, it
suggests deep flow paths that are connected to the movement of slab‐derived fluids through the lithosphere.
Water O, H, and C stable isotope ratios are useful for discerning fluid phase changes, fluid‐rock interactions,
and fluid mixing. Unlike CBD springs that exhibit a strong affinity to meteoric water, HW springs have posi-
tively correlated (slope 3.1, r2 = 0.9) δ18O and δ2H values that are shifted to values higher than the GMWL
(Figure 3a). The observed trend in the HW springs could result from several processes including surface eva-
poration (Craig, 1963), steam separation in the geothermal system (Giggenbach & Stewart, 1982), or meteo-
ric groundwater mixing with geothermal brines derived from water‐rock interaction and equilibration with
metamorphic or igneous rocks at depth (Sheppard, 1986). Newell et al. (2015) argue that near‐surface eva-
poration is unlikely given that each spring sample is collected from the discharge point and that positive cor-
relations between δ18O and δ2H and dissolved CO2 cannot be explained by evaporation or steam separation.
These authors favor mixing between meteoric groundwater and geothermal brine at depth to explain the
trend in O and H isotopes in these springs. Mixing between a deep brine and shallow groundwater is also
consistent with the high TDS and trends in other major ion concentrations (e.g., Na, Cl, and HCO3) observed
in the HW springs (Figure 2).
Carbon stable isotopes of the DIC fall in a narrow range (−10.8‰ to −6.3‰) in CBD and HW springs and
were interpreted by Newell et al. (2015), in concert with helium isotope ratios (3He/4He) and CO /32 He ratios
to suggest a mixture of small to modest amounts of mantle‐derived CO2 (−6 ± 3‰) with dominantly upper
crustal fluids that derive carbon from sedimentary organic matter (low δ13C, ~ −20‰) and marine carbo-
nate sources (δ13C ~ 0‰). However, trends in the carbon isotopic and gas data suggest that overprinting
by near‐surface degassing and boiling processes in hot springs makes it difficult to quantity the relative con-
tribution of these sources (Newell et al., 2015).
Trends observed in halogen and isotopic data (Cl− , Cl/Br, δ18O, and δ37Cl) from this study also support mix-
ing between brine and shallower groundwater sources (Figure 4b). Several studies that use halogen data to
assess mixing between groundwater and higher‐salinity sources such as seawater, brine, halite dissolution,
and magmatic fluids are a useful template to evaluate results from this study (e.g., Katz & Bullen, 1996;
Leisen et al., 2012; Li et al., 2015). Binary mixing between a low salinity meteoric endmember and Cl‐rich
sources can explain some of the trends in our CBD and HW spring data (Figure 4b). The CBD springs can
be fit relatively well with a binary mixing model between a low dissolved Cl− (low salinity), low Cl/Br
meteoric groundwater with a high Cl− , and high Cl/Br brine (Figure 4b). However, we only have Br− result
for two HW springs, and these samples do not fall on the CBD binary Cl/Br versus Cl− mixing trend and
suggest that the composition of the saline endmember is more variable, or perhaps other processes such
as evaporation or boiling are impacting some spring water Cl− concentrations.
The positive linear correlation between chloride concentrations and δ18O values also corroborates the inter-
pretation of mixing between a saline endmember and fresher groundwater (Figure 4c). We note that the
SCOTT ET AL. 12 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
correlation on Figure 4c also includes the data from FW and CHY springs that anchor the low Cl− end of the
trend. The low δ18O values associated with the low salinity endmember are consistent with meteoric
recharge (Figure 3a). As discussed above, the higher δ18O values are associated with the saline geothermal
fluids.
The range of δ37Cl values in these springs (Figure 4d), although narrow,may also be a useful provenance tool
for identifying the contributions and mixing of different Cl sources (e.g., magmatic, seawater, subducted
marine sediments, and fluid‐volcanic rock interaction) (Chiaradia et al., 2014; Cullen et al., 2015). Our data
fall generally within the spread of δ37Cl values observed at other arc‐related hot springs (Bernal et al., 2014;
Cullen et al., 2015) and, alongside Cl− values, may indicate a ternary mixture between a “seawater” source
and meteoric water with a magmatic component in some of the springs (Figure 4d). We caution, however,
the overinterpretation of this mixing model. For example, some spring compositions from the Taupo and
Cascade arcs fall within this ternary mixing field, but most samples actually cannot be explained by this pro-
cess. Cullen et al. (2015) attribute some of the data scatter in δ37Cl values and Cl− concentrations as result-
ing from hydrothermal fluid‐rock interactions with volcanic rocks (basalts) along flow paths (Figure 4d).
Similarly, processes such as equilibration with igneous rocks along flow paths may be influencing the
observed chlorine isotope and concentration data, suggesting that these data may better reflect fluid‐rock
interaction, rather than mixing of primary sources.
In summary, the aqueous chemistry and stable isotope composition from CBD and HW springs support mix-
ing between a thermal brine and shallow groundwater that has evolved due to fluid‐rock interaction along
flow paths. We propose that hydrothermal brine is located at depth along the western flank of the Cordillera
Blanca massif, and flow to the surface is focused along hanging‐wall normal faults and the Cordillera Blanca
detachment fault. The origin of the deep hydrothermal brine remains enigmatic, but it is possible that it is
the product of prolonged, high‐temperature fluid‐rock interaction with the Cordillera Blanca batholith,
and perhaps with inputs from flat‐slab‐derived fluids from the subduction zone.
5.2. Geochemistry of FW and CHY Springs
The first major difference from CBD and HW springs is the diversity in FW and CHY water types with rela-
tively low TDS (Figure 2). We suggest that these springs do not share a common fluid‐rock interaction his-
tory as those along the Cordillera Blanca detachment fault and rather act as separate emanations with
geochemistry reflecting local water‐rock interaction. A comparison of Cl/Br ratios and Cl− concentrations
suggests that FW and CHY springs are more strongly influenced by shallowly circulated meteoric water with
limited water‐rock interaction than CBD and HW spring, likely explaining their overall low TDS (Figure 4b).
In addition to lower Cl− concentrations, the Cl/Br molar ratios from most springs in FW and CHY are also
distinctly lower than observed CBD and HW springs. Although a few springs can be explained by the mixing
curve, most fall to the left of this mixing relationship with lower Cl− concentrations (Figure 4b). This suggest
that the FW and CHY springs are less influenced by deep saline brines, require multiple mixing models to
explain their compositions, and are generally more indicative of meteoric‐sourced groundwater. The rela-
tionship between δ18O values and Cl− content also supports these interpretations (Figure 4c), and the scat-
ter in δ18O values is likely due to isotope effects that impact the composition of meteoric water as described
below. Similar to CBD and HW springs, δ37Cl values are not particularly diagnostic due to the narrow range
of values and lack of trends. When combined with Cl− concentration, a mixture between meteoric water
and a minor saline component may explain the data (Figure 4d), or as with the CBD and HW spring data,
these results may also reflect fluid‐rock interaction along flow paths.
Water stable isotope data measured from FW and CHY spring waters are consistent with a meteoric source
in this part of Peru. δ18O and δ2H values from FW springs are slightly higher than CHY springs, and values
from both groups fall very close to the GMWL (Figure 3a). Most CHY springs issue from higher elevations
(and have higher recharge elevations) and slightly higher latitudes (further south) than FW springs. The
overall low delta values from FW and CHY springs are consistent with meteoric water recharge from high
elevation sources (Dansgaard, 1964). The differences in recharge altitude and latitude between FW and
CHY springs explain the distribution in δ18O and δ2H values (Figure 3a). Increasing altitude and latitude
is generally associated with lower temperatures, which correspond to larger isotopic fractionation factors
between water vapor and precipitation and, in combination with increased amounts of precipitation,
SCOTT ET AL. 13 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
result in lower δ18O and δ2H values of meteoric water (e.g., Sharp, 2007). Also, FW and CHY springs do not
show the shift to higher δ18O and δ2H values, interpreted as mixing with a geothermal fluid, as seen in the
HW springs. The range of δ13C values of DIC in these springs (−12.0 to −1.2‰) is much wider than CBD
and HW springs (Figure 3b). This wide range is consistent with carbon derived from marine carbonates and
organic‐rich shales (or soil organic matter), which are present in the upper few kilometers of crust at all these
springs, suggesting that local water‐rock interaction controls their δ13C values. Without additional helium
isotope ratios, the presence of a mantle‐derived CO2 component cannot be discerned from these data.
Collectively, these data indicate that thermal groundwaters emanating from FW and CHY springs are geo-
chemically distinct from thermal waters observed at CBD and HW springs. These graphical and qualitative
observations are also supported by simple statistical comparisons. Here a Mann‐Whitney U test (nonpara-
metric, two‐tailed distribution) of several key parameters between the two groups of thermal springs (FW
+ CHY vs. CBD + HW) shows significance to the 0.01 confidence level. FW and CHY springs have lower
total dissolve solids (average 599 vs. 5,527 ppm; p = 0.0001), lower Cl/Br molar ratios (average 465 vs.
1,032; p = 0.0024), lower dissolved HCO −3 (average 209 vs. 1,003; p = 0.0002), lower δ
18O values (average
−15‰ vs. −11‰; p = 0.00006), and a wider range of δ13C values (average −5.2‰ vs. −9.5‰; p = 0.0041)
than the CBD and HW hot springs.
There is one interesting outlier in the FW springs. Baños Jocos Peinado has a distinct Ca‐SO4 chemistry,
yielding a TDS of 2,840 mg/L and the highest overall concentration of SO 2−4 (1,452 mg/L) in this investiga-
tion. Jocos Peinado also has the lowest δ37Cl value (−0.6 ± 0.2) of FW and CHY springs, more similar to
some CBD and HW springs (Figure 4d). This spring emanates from a region ~50 km northeast of the
Cordillera Blanca massif, where some recent extensional faulting is documented. In 1946, a MW 6.8 earth-
quake ruptured a previously unmapped normal fault near Quiches, Peru, that is nearby this hot spring
(Bellier et al., 1991; Doser, 1987). Pleistocene to recent extension in this region is hypothesized to have reac-
tivated Eocene thrusts (Bellier et al., 1991), and we suggest that young extensional features in this region
may play a prominent role in influencing deep flow paths and fluid properties, similar to processes hypothe-
sized for CBD and HW springs.
5.3. Insights Into Fluid Circulation Depths
RTEs provided by geothermometry are useful for evaluating temperatures at depth, fluid‐rock interaction
along flow paths, and estimates of fluid circulation depth. Processes such as boiling, mineral precipitation
during ascent, reequilibration with rocks along flow paths, and mixing with shallowly circulated meteoric
waters can promote disequilibrium and alter RTE (Fournier, 1977; Fournier & Truesdell, 1973;
Giggenbach, 1988). For the springs investigated, the Na‐K‐Ca or Na‐K geothermometry estimates are likely
most representative of the geothermal fluid conditions at depth (maximum equilibration T) and thus for
evaluating depth of fluid circulation. Baños Olleros, Monterrey, Huanacarhuaz, and Recuay are the only
springs that meet the criteria for the Na‐K geothermometer, and most others are best estimated using the
Na‐K‐Ca geothermometer (see section 4.4). As presented earlier, a number of FW and CHY springs have
Na‐K‐Ca temperatures that are lower than measured spring temperature. In these cases, another geotherm-
ometer may be more informative (e.g., K‐Mg or chalcedony) and must be considered on a case‐by‐case basis.
Geothermometry data provide an additional means for explaining the geochemical differences between
springs near the Cordillera Blanca detachment fault (CBD and HW springs) and springs east of the
Cordillera Blanca massif and in the Cordillera Huayhuash (FW and CHY springs). By applying an appropri-
ate geothermal gradient, the depth of fluid reservoirs, and thus circulation depth, can be estimated. Due to a
lack of direct constraints on the geotherm in these regions, we explore a range of values. First, assuming a
nominal geothermal gradient of 25°C/km and an average surface temperature of 0°C, RTEs (Na‐K‐Ca or
Na‐K) from CBD springs (152–260°C) and HW springs (91–265°C) correspond to circulation depths from
~4 to 11 km.However, given the recent and active exhumation of the Cordillera Blanca (Margirier et al., 2015)
and the distribution of relatively high‐temperature hot springs along the Cordillera Blanca detachment fault,
the geothermmay be elevated above nominal conditions in this region. Microseismicity along and within the
hanging wall of the Cordillera Blanca detachment (Deverchere et al., 1989) suggests that the base of the seis-
mogenic crust is at ~10–12 km, with expected temperatures of 300–350°C (Sibson, 1982). Therefore, a gradi-
ent of 30°C/km, or perhaps as high as 35°C/km, may be more appropriate here. This reduces circulation
SCOTT ET AL. 14 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Figure 6. Schematic cross section across the Cordillera Blanca massif and conceptual model for structural controls on fluid circulation. Cross section highlights
hypothesized fluid sources and circulation depth and pathway differences between CBD and HW springs west of the Cordillera Blanca detachment fault and FW
springs to the east. CHY springs are not depicted in this cross section but are analogous to the FW systems. Two representative hot springs are depicted
(Huancarhuaz and Pomabamba). Cross section is not to scale, and the general geology, structure, and deformation style are inspired by Giovanni et al. (2010) and
Scherrenberg et al. (2016).
depth estimates to ~2.6–8.7 km. In contrast, the nominal 25°C/km gradient is likely appropriate for the FY
and CHY springs. Thus, Na‐K‐Ca RTEs from FW springs (42–79°C) and CHY springs (40–98°C) correspond
to substantially shallower circulation depths (~1.6–4 km) than for the CBD and HW hot springs. It is
important to note that given the FW and CHY spring waters are “immature” (Figure 5), these RTEs and
subsequent fluid circulation depth estimates have more uncertainty than the CBD and HW springs.
In addition to estimates on the depth of circulation, the various geothermometers provide information on
the extent of fluid‐rock interaction that has occurred. Ideally, all of the geothermometers should yield simi-
lar temperatures for a given spring if a fluid is in equilibrium with the reservoir minerals. In hot springs this
is often not the case because the geochemistry of the spring water has evolved during fluid ascent due to ret-
rograde water‐rock interaction and mixing. For example, comparison of the Na‐K and K‐Mg systems
(Giggenbach, 1988) shows that the CBD and HW springs form a trend between the “partial equilibrium
field” and “immature waters” near the K‐Mg apex (Figure 5). In the context of the aqueous and isotope geo-
chemistry, Newell et al. (2015) interpreted this as due to the ascent and mixing of deep thermal fluids with
cooler shallow groundwater. Also, along this flow path the partial reequilibration between water and rock at
lower temperatures will be reflected by the K‐Mg thermometer (Fournier, 1977). The FW and CHY springs
all fall in the immature field, clustering near the K‐Mg corner of (Figure 5). Thus, the geothemometry indi-
cates that not only are the FW and CHY spring waters much more shallowly circulated, they have experi-
enced less fluid‐rock equilibration at depth.
Collectively, we propose that waters emanating from CBD and HW springs are deeply circulated (up to 9–
11 km), compared to the much shallower flow paths ( < 4 km) interpreted for FW and CHY springs
(Figure 6). Consistent with earlier results from Newell et al. (2015), the CBD and HW springs appear related
to a similar geothermal system that is hosted at depth in the hanging wall of the Cordillera Blanca detach-
ment fault. Also, the maximum fluid circulation depth estimates of ~11 km based on geochemistry are con-
sistent with the interpreted depth of the brittle crust based on hypocenters in the region ( < 12 km)
(Deverchere et al., 1989). Ascent of these brines along brittle features related to the detachment fault and
in its hanging wall has facilitated mixing and some retrograde water‐rock interaction along these pathways.
In contrast, the diverse geochemistry from hot springs located to the east of the Cordillera Blanca massif and
SCOTT ET AL. 15 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
on the flanks of the Cordillera Huayhuash to the south suggests more isolated hydrothermal systems. Many
of these springs appear spatially associated with thrust faults related to theMFTB, and these fracture systems
must facilitate relatively shallower fluid circulation ( < 4 km), as suggested by the geothermometry. Distinct
fluid chemistries to the east of the Cordillera Blanca at Baños Jocos Peinado may be linked to documented
normal faulting and extension in the Quiches area (Bellier et al., 1991). Although geothermometry estimates
still indicate relatively shallow circulation (Table 2), we hypothesize that the Quiches region may be a snap-
shot of the early stages of extension as now expressed along the Cordillera Blanca detachment fault.
We suggest that it is the difference in structural setting between the Cordillera Blanca detachment fault and
surrounding systems that controls the depth of fluid circulation. In all cases, the hot springs are surrounded
by catchments with high elevation headwaters; thus, a similar hydraulic drive is available at all these sys-
tems. If this was the primary factor, then we would expect similar depths of circulation. Furthermore, the
calculated depths are consistent with the two styles of faulting in the area. The Cordillera Blanca detachment
system is extensional with penetration depths at least 10 km (Giovanni et al., 2010; Hughes et al., 2019;
McNulty & Farber, 2002). In contrast, most of the MFTB structures are more shallowly penetrating and sole
out into décollements within 5 km (e.g., Scherrenberg et al., 2016), consistent with thin‐skinned fold and
thrust belt deformation.
6. Summary and Conclusions
This study evaluates the influence of the Cordillera Blanca detachment fault on hot spring geochemistry in
the Cordillera Blanca, Peru, and how this differs from fold‐ and thrust‐dominated systems to the east of the
Cordillera Blanca and in the Cordillera Huayhuash. Hot springs from a variety of locations in this part of the
Peruvian Andes were sampled and analyzed for their aqueous and stable isotope geochemistry. Spring loca-
tions are geographically categorized into four groups to discern the role of the Cordillera Blanca detachment
fault as a conduit for fluid flow. CBD and HW springs issue along the trace and hanging wall of the
Cordillera Blanca detachment fault, respectively, FW springs are located in the footwall of the detachment
fault and to the east of the Cordillera Blanca massif, and CHY springs comprise springs southeast of the
Cordillera Blanca and in the Cordillera Huayhuash. Geochemistry results from CBD and HW springs show
similar Na‐Cl‐dominated waters, high TDS, and stable isotopic compositions that collectively suggest inter-
related flow paths mixing with deep‐seated thermal brine. Geothermometry constraints from these CBD
springs correspond to maximum fluid circulation depths up to 9 or possibly 11 km. These data also support
the inference of HW springs issuing along steep hanging‐wall normal faults that intersect the Cordillera
Blanca detachment fault at depth. In contrast, FW and CHY springs yield a wide range in water types,
low TDS, and stable isotope results and geothermometry estimates that imply a greater degree of influence
from shallower groundwater flow paths ( < 4 km).
We propose that the Cordillera Blanca detachment fault is acting as a primary structural control on fluid dis-
tribution, facilitating deep flow paths and the migration of Na‐Cl brine to springs along the trace of the fault
and those cutting its hanging wall. Away from this major extensional system, hot springs appear to be con-
trolled by faults and structures associated with the older MFTB. These appear to facilitate flow to modest but
much shallower depths due to the style of deformation. Similar to other studies of fault‐related fluids, this
investigation demonstrates that chemical and isotopic tracers from fault‐bound hot springs are excellent
Acknowledgments tools for discerning flow paths and determining the influence of regional tectonics on fluid distribution.
Research was made possible through
funding from the National Science Significantly, however, this study also shows a clear difference in fluid flow path and extent of fluid‐rock
Foundation through Grant interaction between deeply penetrating extensional structures and shallowly penetrating structures asso-
EAR‐1623034 to Newell and Jessup. We ciated with compression, all within the same orogenic system.
thank Dr. Jamie Barnes (UT Austin) for
chlorine stable isotope analysis,
Andrew Lonero (USU) for general Data Availability Statement
geochemistry and stable isotope
analytical support, the USU Water Access to the full data set is archived in the EarthChem Library (https://doi.org/10.26022/IEDA/111569)
Research Lab for assistance with water (Newell & Scott, 2020). All data are also tabulated in the paper and in Tables S1–S5.
chemistry, and Alberto Cafferata
(Caraz, Peru) for field logistics and
guide support. We thank Randy References
Williams and an anonymous reviewer
for their insightful and constructive Arnorsson, S., Gunnlanugsson, E., & Svavarsson, H. (1983). The chemistry of geothermal waters in Iceland. III. Chemical geothermometry
reviews that improved this paper. in geothermal investigations. Geochimica et Cosmochimicha Acta, 47, 567–577.
SCOTT ET AL. 16 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Atherton, M. P., & Sanderson, L. M. (1987). The Cordillera Blanca Batholith: A study of granite intrusion and the relation of crustal
thickening to peraluminosity. Geologische Rundschau, 76, 213–232.
Banks, D., Giuliani, G., Yardley, B., & Cheilletz, A. (2000). Emerald mineralisation in Colombia: Fluid chemistry and the role of brine
mixing. Mineralium Deposita, 35(8), 699–713.
Barnes, J. D., Sharp, Z. D., Fischer, T. P., Hilton, D. R., & Carr, M. J. (2009). Chlorine isotope variations along the Central American volcanic
front and back arc. Geochemisty, Geophysics, and Geosystems, 10, Q11S17. https://doi.org/10.1029/2009GC002587
Barry, P., Hilton, D., Fischer, T., DeMoor, J., Mangasini, F., & Ramirez, C. (2013). Helium and carbon isotope systematics of cold “mazuku”
CO2 vents and hydrothermal gases and fluids from Rungwe Volcanic Province, southern Tanzania. Chemical Geology, 339, 141–156.
Bellier, O., Dumont, J. F., Sébrier, M., & Mercier, J. L. (1991). Geological constraints on the kinematics and fault‐plane solution of the
Quiches fault zone reactivated during the 10 November 1946 Ancash earthquake, northern Peru. Bulletin of the Seismological Society of
America, 81(2), 468–490.
Bense, V. F., & Person, M. (2006). Faults as conduit‐barrier systems to fluid flow in siliciclastic sedimentary aquifers. Water Resources
Research, 42, W05421. https://doi.org/10.1029/2005WR004480
Bernal, N. F., Gleeson, S. A., Dean, A. S., Liu, X.‐M., & Hoskin, P. (2014). The source of halogens in geothermal fluids from the Taupo
Volcanic Zone, North Island, New Zealand. Geochimica et Cosmochimica Acta, 126, 265–283.
Bethke, C. M. (2006), The Geochemist's Workbench Release 6.0, Hydrogeology Program, Dept. of Geology, University of Illinois.
Bucher, K., & Stober, I. (2010). Fluids in the upper continental crust. Geofluids, 10(1–2), 241–253.
Butcher, L., Mahan, K., & Allaz, J. (2017). Late Cretaceous crustal hydration in the Colorado Plateau, USA, from xenolith petrology and
monazite geochronology. Lithosphere, 9(4), 561–578.
Caine, J. S., Evans, J. P., & Forster, C. B. (1996). Fault zone architecture and permeability structure. Geology, 24(11), 1025–1028.
Chiaradia, M., Barnes, J. D., & Cadet‐Voisin, S. (2014). Chorine stable isotope variations across the Quaternary volcanic arc of Ecuador.
Earth and Planetary Science Letters, 396, 22–33.
Cobbing, E. J. (1981), The geology of the Western Cordillera of northern Peru, 144 pp., Inst. of Geol. Sci., London, UK.
Coldwell, B., Clemens, J., & Petford, N. (2011). Deep crustal melting in the Peruvian Andes: Felsic magma generation during delamination
and uplift. Lithos, 125(1–2), 272–286.
Coney, P. J. (1971). Structural evolution of the Cordillera Huayhuash, Andes of Peru. Geological Society of America Bulletin, 82, 1863–1884.
Craig, H. (1961). Isotopic variations in meteoric waters. Science, 133(3465), 1702–1703.
Craig, H. (1963). The isotopic geochemistry of water and carbon in geothermal areas. Nuclear geology on geothermal areas. Spoleto, 1963,
17–53.
Craig, H., and L. Gordon (1965),Deuterium and oxygen 18 variations in the ocean and marine atmosphere, Stable Isotopes in Oceanographic
Studies, 122 pp., Consiglio Nazional delle Richerche, Pisa, Spoleto, Italy.
Crossey, L. J., Karlstrom, K. E., Springer, A., Newell, D. L., Hilton, D. R., & Fischer, T. P. (2009). Degassing of mantle‐derived CO2 and He
from springs in the southern Colorado Plateau region—Neotectonic connections and implications for groundwater systems. GSA
Bulletin, 121, 1034–1053.
Cullen, J. T., Barnes, J. D., Hurwitz, S., & Leeman, W. P. (2015). Tracing chlorine sources of thermal and mineral springs along and across
the Cascade Range using halogen concentrations and chlorine isotope compositions. Earth and Planetary Science Letters, 426, 225–234.
Dansgaard, W. (1964). Stable isotopes in precipitation. Tellus, 16(4), 436–468.
Darling, W., Griesshaber, E., Andrews, J., Armannsson, H., & O'Nions, R. (1995). The origin of hydrothermal and other gases in the Kenya
Rift Valley. Geochimica et Cosmochimica Acta, 59(12), 2501–2512.
de Leeuw, G. A. M., Hilton, D. R., Güleç, N., &Mutlu, H. (2010). Regional and temporal variations in CO2/
3He, 3He/4He and δ13C along the
North Anatolian Fault Zone, Turkey. Applied Geochemistry, 25, 524–539.
Deverchere, J., Dorbath, C., & Dorbath, L. (1989). Extension related high topography: Results from a microearthquake survey in the Andes
of Peru and tectonic implications. Geophysical Journal International, 98, 281–292.
Doser, D. I. (1987). The Ancash, Peru, earthquake of 1946 November 10: Evidence for low‐angle normal faulting in the high Andes of
northern Peru. Geophysical Journal of the Royal Astronomical Society, 91, 57–71.
Evans, M. J., Derry, L. A., & France‐Lanord, C. (2008). Degassing of metamorphic carbon dioxide from the Nepal Himalaya. Geochemistry,
Geophysics, Geosystems, 9, Q04021. https://doi.org/10.1029/2007GC001796
Fournier, R. (1977). Chemical geothermometers and mixing models for geothermal systems. Geothermics, 5(1–4), 41–50.
Fournier, R. O. (1987), Conceptual models of brine evolution in magmatic‐hydrothermal systems, Volcanism in Hawaii, U.S. Geological
Survey Professional Paper 1350, 1487–1506.
Fournier, R. O., & Truesdell, A. H. (1973). An empirical Na‐K‐Ca chemical geothermometer for natural waters. Geochimica et
Cosmochimica Acta, 37, 1255–1275.
Garver, J. I., Reiners, P. W., Walker, L. J., Ramage, J. M., & Perry, S. E. (2005). Implications for timing of Andean uplift from thermal
resetting of radiation‐damaged zircon in the Cordillera Huayhuash, northern Peru. The Journal of Geology, 113, 117–138.
Giggenbach, W. (1990). The chemistry of fumarolic vapor and thermal‐spring discharges from the Nevado del Ruiz volcanic‐magmatic‐
hydrothermal system, Colombia. Journal of Volcanology and Geothermal Research, 42(1–2), 13–39.
Giggenbach, W., & Stewart, M. (1982). Processes controlling the isotopic composition of steam and water discharges from steam vents and
steam‐heated pools in geothermal areas. Geothermics, 11(2), 71–80.
Giggenbach, W. F. (1988). Geothermal solute equilibria. Derivation of Na‐K‐Mg‐Ca geoindicators. Geochimica et Cosmochimica Acta, 52,
2749–2765.
Giovanni, M. (2007), Tectonic and thermal evolution of the Cordillera Blanca detachment system, Peruvian Andes: Implications for normal
faulting in a contractional orogen, 237 pp, University of California Los Angeles.
Giovanni, M. K., Horton, B. K., Garzione, C. N., McNulty, B., & Grove, M. (2010). Extensional basin evolution in the Cordillera Blanca,
Peru: Stratigraphic and isotopic records of detachment faulting and orogenic collapse in the Andean hinterland. Tectonics, 29, TC6007.
https://doi.org/10.1029/2010TC002666
Gutscher, M.‐A. (2002). Andean subduction styles and their effect on thermal structure and interplate coupling. Journal of Asian Earth
Sciences, 15, 3–10.
Hanor, J. S. (1987), Origin and migration of subsurface sedimentary brines, SEPM Society for Sedimentary.
Hoke, L., & Lamb, S. (2007). Cenozoic behind‐arc volcanism in the Bolivian Andes, South America: Implications for mantle‐melt gen-
eration and lithospheric structure. Journal of the Geological Society, 164, 795–814.
Hoke, L., Lamb, S., Hilton, D. R., & Poreda, R. J. (2000). Southern limit of mantle‐derived geothermal helium emissions in Tibet:
Implications for lithospheric structure. Earth and Planetary Science Letters, 180, 297–308.
SCOTT ET AL. 17 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Hooper, E. (1991). Fluid migration along growth faults in compacting sediments. Journal of Petroleum Geology, 14, 161–180.
Hughes, C. A., Jessup, M. J., Shaw, C. A., & Newell, D. L. (2019). Deformation conditions during syn‐convergent extension along the
Cordillera Blanca shear zone, Peru. Geosphere, 15.
Humphreys, E., Hessler, E., Dueker, K., Farmer, G. L., Erslev, E., & Atwater, T. (2003). How Laramide‐age hydration of North American
lithosphere by the Farallon slab controlled subsequent activity in the western United States. International Geology Review, 45(7),
575–595.
Karingithi, C. W. (2007), Chemical geothermometers for geothermal exploration, 001045504.
Katz, B. G., & Bullen, T. D. (1996). The combined use of 87Sr/86Sr and carbon and water isotopes to study the hydrochemical interaction
between groundwater and lakewater in mantled karst. Geochimica et Cosmochimica Acta, 60(24), 5075–5087.
Kendrick, M. A., Scambelluri, M., Honda, M., & Phillips, D. (2011). High abundances of noble gas and chlorine delivered to the mantle by
serpentinite subduction. Nature Geoscience, 4(11), 807.
Kennedy, B. M., Kharaka, Y. K., Evans, W. C., Ellwood, A., DePaolo, D. J., Thordsen, J., et al. (1997). Mantle fluids in the San Andreas fault
system, California. Science, 278, 1278–1281.
Kennedy, B. M., & van Soest, M. C. (2007). Flow of mantle fluids through the ductile lower crust: Helium isotope trends. Science, 318,
1433–1436.
Kesler, S. E., Martini, A. M., Appold, M. S., & Walter, L. M. (1996). Na‐Cl‐Br systematics of fluid inclusions from Mississippi Valley‐type
deposits, Appalachian Basin: Constraints on solute origin and migration paths. Geochimica et Cosmochimica Acta, 60(2), 225–233.
Klemperer, S. L., Kennedy, B. M., Sasty, S. R., Makovsky, Y., Harinarayana, T., & Leech, M. L. (2013). Mantle fluids in the Karakoram fault:
Helium isotope evidence. Earth and Planetary Science Letters, 366, 59–70.
Kühn, M. (2004). Reactive flow modeling of hydrothermal systems (Vol. 103). Springer Science+Business Media.
Kulongoski, J. T., Hilton, D. R., Barry, P. H., Esser, B. K., Hillegonds, D., & Belitz, K. (2013). Volatile fluxes through the Big Bend section of
the San Andreas Fault, California: Helium and carbon‐dioxide systematics. Chemical Geology, 339, 92–102.
Leisen, M., Boiron, M.‐C., Richard, A., & Dubessy, J. (2012). Determination of Cl and Br concentrations in individual fluid inclusions by
combining microthermometry and LA‐ICPMS analysis: Implications for the origin of salinity in crustal fluids. Chemical Geology, 330,
197–206.
Li, L., Bonifacie, M., Aubaud, C., Crispi, O., Dessert, C., & Agrinier, P. (2015). Chlorine isotopes of thermal springs in arc volcanoes for
tracing shallow magmatic activity. Earth and Planetary Science Letters, 413, 101–110.
Margirier, A., Audin, L., Robert, X., Herman, F., Ganne, J., & Schwartz, S. (2016). Time and mode of exhumation of the Cordillera Blanca
batholith (Peruvian Andes). Journal of Geophysical Research: Solid Earth, 121, 6235–6249. https://doi.org/10.1002/2016JB013055
Margirier, A., Audin, L., Robert, X., Pêcher, A., & Schwartz, S. (2017). Stress field evolution above the Peruvian flat‐slab (Cordillera Blanca,
northern Peru). Journal of South American Earth Sciences, 77, 58–69.
Margirier, A., Robert, X., Audin, L., Gautheron, C., Bernet, M., Hall, S., & Simon‐Labric, T. (2015). Slab flattening, magmatism, and surface
uplift in the Cordillera Occidental (northern Peru). Geology, 43(11), 1031–1034.
Mark, B. G., &Mckenzie, J. M. (2007). Tracing increasing tropical Andean glacier melt with stable isotopes in water. Environmental Science
& Technology, 41(20), 6955–6960. https://doi.org/10.1021/es071099d
McNulty, B., & Farber, D. (2002). Active detachment faulting above the Peruvian flat slab. Geology, 30, 567–570.
Mégard, F. (1984). The Andean orogenic period and its major structures in central and northern Peru. Journal of the Geological Society,
141(5), 893–900.
Mukasa, S. (1984). Comparative Pb isotope systematics and zircon U‐Pb geochronology for the Coastal, San Nicolas and Cordillera Blanca
batholiths, Peru (PhD thesis, pp. 148). Santa Barbara: University of California–Santa Barbara.
Mutlu, H., Güleç, N., & Hilton, D. R. (2008). Helium–carbon relationships in geothermal fluids of western Anatolia, Turkey. Chemical
Geology, 247(1–2), 305–321.
Newell, D. L., Crossey, L. J., Karlstrom, K. E., Fischer, T. P., & Hilton, D. R. (2005). Continental‐scale links between the mantle and
groundwater systems of the western United States: Evidence from travertine springs and regional He isotope data. GSA Today, 15(12),
4–10.
Newell, D. L., Jessup, M. J., Cottle, J. M., Hilton, D., Sharp, Z., and Fischer, T. (2008). Aqueous and isotope geochemistry of mineral springs
along the southern margin of the Tibetan plateau: Implications for fluid sources and regional degassing of CO2, Geochemistry,
Geophysics, and Geosystems, 9, Q08014. https://doi.org/10.01029/02008GC002021
Newell, D. L., Jessup, M. J., Hilton, D. R., Shaw, C. A., & Hughes, C. A. (2015). Mantle‐derived helium in hot springs of the Cordillera
Blanca, Peru: Implications for mantle‐to‐crust fluid transfer in a flat‐slab subduction setting. Chemical Geology, 417, 200–209.
Newell, D. L., & Scott, B. E. (2020). Hot spring stable isotope and aqueous geochemistry of the Cordillera Blanca and Cordillera Huayhuash,
Peru. Edited, Interdisciplinary Earth Data Alliance (IEDA).
Panno, S., Hackley, K. C., Hwang, H., Greenberg, S., Krapac, I., Landsberger, S., & O'Kelly, D. (2006). Characterization and identification of
Na‐Cl sources in ground water. Groundwater, 44(2), 176–187.
Pepin, J., Person, M., Phillips, F., Kelley, S., Timmons, S., Owens, L., et al. (2015). Deep fluid circulation within crystalline basement rocks
and the role of hydrologic windows in the formation of the Truth or Consequences, New Mexico low‐temperature geothermal system.
Geofluids, 15(1–2), 139–160.
Petford, N., & Atherton, M. P. (1992). Granitoid emplacement and deformation along a major crustal lineament: The Cordillera Blanca,
Peru. Tectonophysics, 205, 171–185.
Piper, A. M. (1944). A graphic procedure in the geochemical interpretation of water‐analyses. Transactions, American Geophysical Union,
25(6), 914. https://doi.org/10.1029/tr025i006p00914
Ramos, V. A., & Folguera, A. (2009). Andean flat‐slab subduction through time, in Ancient orogens and modern analogues, Geological
Society, London, Special Publications, edited by J. B. Murphy, J. D. Keppie and A. D. Hynes, pp. 31–54.
Rimstidt, J. D., & Barnes, H. (1980). The kinetics of silica‐water reactions. Geochimica et Cosmochimica Acta, 44(11), 1683–1699.
Salata, G. G., Reoelke, L. A., & Cifuentes, L. A. (2000). A rapid and precise method for measuring stable carbon isotope ratios of dissolved
inorganic carbon. Marine Chemistry, 69, 153–161.
Scherrenberg, A. F., Jacay, J., Holcombe, R. J., & Rosenbaum, G. (2012). Stratigraphic variations across the Marañón fold‐thrust belt, Peru:
Implications for the basin architecture of the West Peruvian Trough. Journal of South American Earth Sciences, 38, 147–158.
Scherrenberg, A. F., Kohn, B. P., Holcombe, R. J., & Rosenbaum, G. (2016). Thermotectonic history of the Marañón Fold–Thrust Belt, Peru:
Insights into mineralisation in an evolving orogen. Tectonophysics, 667, 16–36.
Sébrier, M., Alain, L., Michel, F., & Soulas, J. P. (1988). Tectonics and uplift in Central Andes (Peru, Bolivia and Northern Chile) from
Eocene to present. Géodynamique, 3, 85–106.
SCOTT ET AL. 18 of 19
Geochemistry, Geophysics, Geosystems 10.1029/2020GC008919
Sharp, Z., & Barnes, J. (2004). Water‐soluble chlorides in massive seafloor serpentinites: A source of chloride in subduction zones. Earth
and Planetary Science Letters, 226(1–2), 243–254.
Sheppard, S. M. F. (1986). Characterization and isotopic variations in natural waters. In J. W. Valley & H. P. Taylor (Eds.), Stable isotopes in
high temperature geological processes (pp. 165–183). Chantilly, VA: Mineralogical Society of America.
Sharp, Z. D. (2007). Principles of Stable Isotope Geochemistry (pp. 344). Upper Saddle River, NJ: Pearson/Prentice Hall.
Sibson, R. H. (1982). Fault zone models, heat flow, and the depth distribution of earthquakes in the continental crust of the United States.
Bulletin of the Seismological Society of America, 72(1), 151–163.
USGS (2006), National field manual for the collection of water‐quality data, in U.S. Geological Survey Techniques of Water‐Resources
Investigations: Book 9, A1‐A9, edited, USGS.
Walter, L. M., Stueber, A. M., & Huston, T. J. (1990). Br‐Cl‐Na systematics in Illinois basin fluids: Constraints on fluid origin and evolution.
Geology, 18(4), 315–318.
SCOTT ET AL. 19 of 19