Applied Geochemistry 95 (2018) 97–109

Contents lists available at ScienceDirect

Applied Geochemistry
journal homepage: www.elsevier.com/locate/apgeochem

Modeling of groundwater processes in a karstic aquifer of Sierra Madre T

Oriental, Mexico
J. Moran-Ramíreza, J.A. Ramos-Leala,∗, J. Mahlknechtb, G. Santacruz-DeLeónc,
F. Martín-Romerod, R. Fuentes Rivase, A. Morab
a
Applied Geosciences Division, Potosin Institute of Scientific and Technological Research, C.A. (IPICYT), Camino a la Presa San José # 2055, Lomas 4a, Sección, C.P.
78216, San Luís Potosí, SLP, Mexico
b
Escuela de Ingenieria y Ciencias, Tecnologico de Monterrey, Av. Eugenio Garza Sada Sur No. 2501,Monterrey CP 64849, Nuevo León, Mexico
c
Water and Society Program, College of San Luis C. A. (COLSAN), Parque de Macul 155, Colinas del Parque, C.P. 78294 San Luis Potosí, Mexico
d
Department of Geochemistry, Institute of Geology, National Autonomous University of Mexico University City, Circuito de la investigación Científica s/n, C.P. 04510,
Ciudad Universitaria, Delegación Coyoacán, Mexico City, Mexico
e
Facultad de Geografía, Universidad Autónoma del Estado de México, Cerro de Coatepec, Ciudad Universitaria, Toluca estado de México C. P. 50110, Mexico

A R T I C LE I N FO A B S T R A C T

Editorial handling by Philippe Negrel The Sierra Madre Oriental (SMO) is composed of ranges with important regional geological structures in highly
Keywords: fractured carbonate rocks (Abra Formation), which facilitate the recharge of aquifers. At the regional scale,
VISHMOD recharge occurs in the western SMO and locally, in ranges such as Palmillas, La Colmena and Abra, while the
Ternary mixing regional discharge is to the east in the Huasteca zone. Intermountain valleys, such as El Salto, are located near
Water-rock interaction the regional discharge zone. Three end members were identified in these regions, one related to local flow,
Forward modeling another to intermediate and a third to regional flow. These flows define ternary mixing processes in ground-
Inverse modeling water. In this paper, the evolution of groundwater is studied, which is important for learning the system per-
formance and hydrogeochemical processes through VISHMOD (Virtual Samples in Hydrochemical Modeling)
methodology. The mixing model for the intermountain valleys indicates that many recharges occur, for which
local flow contributes 68.3%, intermediate flow 12.3% and regional flow 19.4%. The inverse modeling shows
that the water-rock interaction, precipitation and/or dissolution of calcite, gypsum and dolomite are the main
processes occurring in the aquifer system of the intermountain valleys.

1. Introduction
The chemical composition of groundwater is mainly controlled by
factors such as residence time, distance, mixture, mineralogy and
geology of the aquifer. The evolution of groundwater begins when
rainwater seeps into the land and is enriched with ions due to inter-
action with the geological environment during its path. This interaction
produces different chemical compositions in the groundwater, based on
which different types of groundwater families can be identified.
Evolution of groundwater can be studied using tools such as hydro-
geochemistry, particularly conservative elements, which are considered
tracers. This evolution can be handled as mixing end members, where
the starting point is the recharge and discharge is the end (Morán-
Ramírez et al., 2013, 2016; Moran-Ramirez and Ramos-Leal, 2014;
Ledesma-Ruiz et al., 2015). In numerous publications, the amount of
mixing between freshwater and seawater or between groundwater from
different sources has been evaluated (Dixon and Chiswell, 1992;
Genereux et al., 1996; Laaksoharju et al., 1999; Lee and Krothe, 2001;
Ramos-Leal et al., 2007; Gómez et al., 2008; Petitta et al., 2010;
Hernández-Antonio et al., 2015). These mixing models are identified by
end members (EM) containing the extreme (maximum and minimum)
chemical concentrations in the hydrogeological system. This identifi-
cation using EM is based on conservative elements that do not interact
with the environment. To assess water rock interaction, the tool used is
chemical modeling, in which physical and chemical principles are ap-
plied to the simulation of the system. The main objective of modeling is
to create or devise theoretical reaction models that can provide ele-
ments to explain that which is observed in nature. These models are
constructed using chemical composition in the aqueous phase, isotope
geology and the mineralogy of the system (Lee and Krothe, 2001;
Hidalgo and Cruz, 2001; Abu-Jaber, 2001; André et al., 2004; Hereford
et al., 2007). Chemical modeling can be performed in two ways: 1)
direct modeling, which involves predicting the water composition and
the mass transfer that could result from hypothetical reactions, and 2)

∗
Corresponding author.
E-mail address: jalfredo@ipicyt.edu.mx (J.A. Ramos-Leal).

https://doi.org/10.1016/j.apgeochem.2018.05.011
Received 17 October 2017; Received in revised form 23 February 2018; Accepted 14 May 2018
0883-2927/ © 2018 Elsevier Ltd. All rights reserved.
J. Moran-Ramírez et al.
inverse modeling, which applies to inverse modeling of field samples
along a flow line. However, no work to date has used only hydro-
chemical modeling of EM. Nevertheless, in these studies, mixing pro-
cesses do not explain the final composition of groundwater. In addition,
few studies have included the application of direct modeling of water-
rock interactions and inverse modeling to explain and understand the
physicochemical processes occurring in groundwater (Helgeson et al.,
1970; Plummer and Back, 1980; Plummer et al., 1988; Parkhurst and
Apello, 1999; Hidalgo and Cruz, 2001; André et al., 2004; Hereford
et al., 2007).
In this paper, VISHMOD (Virtual Samples in Hydrochemical
Modeling) methodology (Moran-Ramirez and Ramos-Leal, 2014) is
used to understand the evolution of groundwater, which can be re-
presented as a mixture of two or more EM, to identify water-rock in-
teraction processes that occur in the system which give rise to the
chemical signature of water at each site. The so-called VISHMOD
methodology, which unlike other models uses virtual samples con-
structed from the sum of the mixing fractions of each EM (Moran-
Ramirez and Ramos-Leal, 2014).
In general, VISHMOD methodology models are based on mixtures
with a small number of representative samples (EM) of the overall
system that represent the behavior of groundwater during its evolution.
In this methodology, the processes occurring in the hydrogeochemical
evolution, such as mixing, water-rock interaction are identified.
In some cases, it is not possible to apply inverse modeling in the first
instance, so VISHMOD methodology is useful to identify the processes
that occur before applying the inverse modeling. The hydrogeochemical
models in hydrogeological systems need to be calibrated. The
VISHMOD methodology makes standardization and control possible to
demonstrate the degree to which a model is able to reproduce field
measurements (Moran-Ramirez and Ramos-Leal, 2014).
The study area is located in San Luis Potosi State, Mexico, in the
area known as Huasteca. Located in the central part of the Sierra Madre
Oriental (SMO), bordering the Gulf of Mexico, this area is mainly
composed of limestone karst, with north-south folds forming inter-
mountain valleys (Fig. 1). This region is considered to be part of the
discharge area of the SMO (Moran-Ramirez et al., 2012, 2013). In
general, the knowledge on the hydrodynamics of karstic system is poor
because this area is difficult to study due to steep terrain, excessive
Fig. 1. Flowchart of modeling strategy of VISHMOD methodology.
98
Applied Geochemistry 95 (2018) 97–109
vegetation and the absence of extraction wells. The availability of
groundwater in the area depends on climatic conditions and the topo-
graphy of the intermountain region, as well as the physical character-
istics of the subsurface. In this work, inverse modeling is performed
between a virtual sample (Composition 1) obtained from the EM and
the actual sample (Composition 2).
With this methodology, work arises from a mixing model for pos-
sible hydrogeochemical processes (dissolution and precipitation of mi-
nerals, ion exchange, terms of acidity and basicity, redox, etc.) leading
to the chemical composition of the system. The main objective of
modeling is to create or devise theoretical reaction models that can
provide elements to explain that which is observed in the actual system.
2. Methodology
2.1. Field and laboratory work
Hydrogeochemical data were determined in March 2010 from
samples of 27 wells, springs and hand dug wells in the southern study
area, in the municipality of Naranjo (Fig. 1 and Table 1). The springs
are located in the limestone of El Abra Formation, while the dug wells
are in shales and the thinly layered argillaceous limestone of the San
Felipe Formation. The wells are located in both formations.
Water samples to determine anions were collected in 60 ml poly-
ethylene bottles (HDPE) and washed and rinsed 7 times with deionized
water, whereas the bottles for samples used to determine cations were
washed with 10% HCl and rinsed 7 times with deionized water. Water
samples to determine cations and Sr were acidified with concentrated
HNO3 (pH < 2), whereas the samples to determine anions were not
acidified. All samples were preserved at a temperature below 4 °C until
laboratory analysis. Field measurements included pH, electrical con-
ductivity, alkalinity by titration, redox potential (RP), dissolved oxygen
(DO), total dissolved solids (TDS) and temperature (T).
The chemical composition of the water samples was determined at
the Geoscience Laboratory Center, National Autonomous University of
Mexico. Cations (Ca2+, Mg2+, Na+ and K+) and Sr were determined by
Inductively Coupled Plasma-Atomic Emission Spectroscopy (Model
Thermo ICP-AES 6500Duo) and anions (Cl−and SO42−) were de-
termined by colorimetricat the Institute for Scientific and Technological
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Table 1
Physicochemical data of the samples.
Sample T pH Eh EC STD Alcalinidad Cl− SO42- Ca2+ K+ Mg2+ Na+ Sr2+

°C mV μS/cm mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L

1 26.2 6.83 30.3 943 472 285 0.9 516 159.1 0.829 17.84 2.02 1.193
2 24.32 6.86 136.2 806 403 372 7.3 26 135.9 3.89 6.52 7.5 0.318
3 25.31 7.02 174.1 724 361 375 3.9 6 129.1 1.227 4.66 7.68 0.504
4 25.52 7.06 101.2 719 360 405 4.2 5 109.1 1.583 5.09 31.56 0.737
5 24.54 6.85 12.4 1056 528 285 1.2 100 150.5 1.204 25.2 2.5 1.327
6 25.17 8.86 −53.7 1164 582 300 2 294 180.8 1.396 27.28 6.89 1.91
7 23.05 6.87 73.5 878 435 375 20.6 9 134.5 1.627 5.25 17.16 0.98
8 22.6 6.8 154.5 789 394 360 6 23 134.2 1.468 4.21 16.38 0.983
9 25.29 6.94 84.9 811 405 333 7.7 23 119.8 1.13 10.95 30.8 1.186
10 21.11 7.09 95.6 664 332 300 5.9 46 112.5 4.38 4.6 11.1 0.603
11 22.95 6.54 75.5 930 464 435 3.2 91 168.2 0.98 1.65 11.15 0.915
12 22.24 6.89 −27.1 622 311 300 1 1 109.1 0.91 5.44 11.23 0.567
13 23.34 6.75 −20.1 853 429 420 3.1 66 158.5 1.137 1.26 9.58 0.779
14 22.69 7.04 34.2 783 392 345 23.3 47 140.3 0.86 1.42 11.95 0.657
15 23.76 6.69 −23.4 875 437 360 8.8 30 165.5 0.84 0.98 8.02 0.712
16 23.78 6.59 30.1 933 468 420 18.4 86 173.9 1.046 1.83 11.85 0.823
17 27.36 7 −50 623 311 300 2.5 9 104.7 1.004 2.58 10.37 0.551
18 23.51 6.79 −57.6 668 334 285 3.4 8 123.6 1.752 1.51 11.92 0.786
19 23.26 7.1 −79.2 687 343 360 1.5 6 131.2 1.154 3.15 11.84 0.642
20 24.86 6.64 −39.9 1637 817 330 2.4 276 314.4 3.202 21.36 20.12 4.147
21 25.8 7.05 −114 1559 779 360 2.6 261 316.6 1.479 30.38 15.29 3.951
22 30.12 8.04 −28.2 705 353 210 1.1 758 120.8 1.738 19.9 3.1 1.076
23 24.23 7.02 −63.2 893 447 225 1.2 192 164.2 1.24 25.01 2.34 1.298
24 22.66 7.18 −80.6 416 208 225 1.4 5 86.6 0.337 1.63 7.16 0.175
25 21.68 7.02 47.6 616 308 204 0.05 70 103.3 0.804 15.15 0.94 0.403
26 23.76 7.52 −101.4 1692 846 291 2.5 332 246.6 3.791 55.65 20.55 5.433
27 23.34 7.21 −119 893 447 375 27.1 51 137.5 2.405 9.96 29.15 1.081

Research. The ionic balance of the chemical data had an error of less
than 10%.
2.2. Interpretation using VISHMOD methodology
The VISHMOD methodology has been used for the study of hydro-
geochemical processes in an aquifer formed in limestone (Moran-
Ramirez and Ramos-Leal, 2014). The sequence of steps taken using the
VISHMOD methodology is described as follows (Fig. 1):
2.2.1. Hydrogeological conceptual model
Any study area must have a conceptual hydrogeological model in
which the different flows, geological materials and structures (folds and
faults), mixing zones, recharge and discharge zones are described and
displayed. The conceptual model should be consistent with the poten-
tial hydrogeochemical processes that occur in the study area. This
model will also provide guidelines to establish a hydrogeochemical
model to explain the evolution of groundwater chemistry.
2.2.2. Hydrogeochemical characterization
Before starting the analysis and interpretation of data, it is very
important to check the data quality. The ionic balance of the chemical
data had an error of less than 10%. Using conservative and trace ele-
ments is very important, because when only major ions is used, there is
much uncertainty in the models produced by geochemical codes such as
PHREEQC (Parkhurst and Apello, 1999). Chemical data is used to
generate Piper diagrams, distribution maps of the hydrogeochemical
parameters and Mifflin diagrams to define families, flow systems and
geochemical processes. Scatter diagrams are also generated to identify
EM and define the type of binary or ternary mixture. The hydro-
geochemical characterization provides elements used to build the
conceptual model of the hydrogeological system.
2.2.3. Forward modeling
The chemical composition of groundwater changes during its flow
path and some minerals can precipitate. To determine whether the
minerals in the water precipitate, dissolve or are in balance, saturation
indices (SI = log IAP/K) for the mineral of interest or that might be
present in the water are obtained; this is done through the direct
modeling of chemical data, calculated with the numerical model
PHREEQC (Parkhurst and Apello, 1999).
2.2.4. End member
EMs are identified during this stage. The identification can be based
on scatter plots of conservative elements or tracers (Cl, Sr, B, Li, Br, F,
Ba, isotopes), taking into account the minimum and maximum con-
centrations in the system. Other studies apply multivariate analysis
techniques (Laaksoharju et al., 1999; Hernández-Antonio et al., 2015).
Sometimes, these are representative of flow systems.
2.2.5. Identification of mixtures and hydrochemical model
Based on the scatter plot of tracers or conservative elements (i.e.,
those that do not react chemically during the evolution of ground-
water), the EM are identified, in the case of a binary mixture are two
members and a ternary are three water samples (f1, f2 and f3); the
remaining cases are among the lines of the mixture are considered as a
fraction of the EM. Any water sample system can be generated using the
three EM. Once the type of mixture has been identified, the conceptual
hydrogeochemical model is generated. This model must be consistent
with the hydrogeological model. At this point, the consideration of a
simultaneous or sequential ternary mixture is justified according to the
hydrogeological model.
2.2.6. Mixing fractions
The chemical composition of each sample (Equation (1)) represents
the mixing percentage of the three end members in the system. In
Equation (1), ft is the total sum of the three mixing fractions of the EM,
obtained using the mass balance equation (Ramos-Leal et al., 2007).
The total is set equal to 1:
ft = f1 + f2 + f3 = 1 (1)
where, f1 is the EM with lower concentrations in its conservative

99
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

elements; f2 is the EM with high concentrations in one of the con-

servative elements; f3 is the EM with the highest concentration of the
conservative elements or a high concentration in one of them.
To solve Equation (1) with three unknown variables, a system of 3
equations is needed, for which the concentrations of the conservative
elements from the ternary mixture model of the EM are used (Equations
(2) and (3)).
ft∗Clw = f1 ∗Cl1 + f2 ∗ Cl2 + f3 ∗Cl3 (2)

ft∗Srw = f1 ∗Sr 1 + f2 ∗ Sr 2 + f3 ∗Sr 3 (3)

This results in a system of 3 equations with three unknown vari-
ables.
Solving for f1 in Equation (2):
ft∗Clw − f 2∗Cl2 − f 3∗Cl3
f1 = V1, V2, V3 and Vn are vectors with the chemical composition of the
Cl1 (4)
three EM.
Substituting f1 in Equation (1): In general, the chemical composition of the overall system can be
ft∗Clw − f 2∗Cl2 − f 3∗Cl3 represented in the condensed equation (17).
ft = + f2 + f3
Cl1 (5) m n 3

Solving f3 and simplifying:

Vkj = ∑ ∑ ∑ (fmi ani)
k=1 j=1 i=1 (17)
ft ( Cl1 − Clw ) + f 2 (Cl2 − Cl1)
f3 = Where Vkj, the virtual matrix represents the composition of the
Cl1 − Cl3 (6)
system, The mixing fraction of each of the EM is multiplied by the
In Equation (3) we solve for f3: chemical composition of each EM and their sum represents the virtual
ft∗Srw − f 1∗Sr − f 2∗Sr 2 chemical composition of each sample of system.
f3 = The chemical composition of each sample in the system is obtained;
Sr 3 (7)
this resulting composition represents a virtual chemical composition
Substituting the value of f3 in Equation (1) and simplifying: that does not consider the water-rock interaction. This composition is
f 1(Sr 3 − Sr 1) + f 2(Sr 3 − Sr 2) + ftSrw used in the chemical modeling of the system.
ft =
Sr 3 (8)
2.2.8. First calibration (actual versus virtual chemical composition data 1)
Solving for f1 and simplifying:
Calibration can be performed both qualitatively and quantitatively.
ft (Sr 3 − Srw ) + f 2 (Sr 2 − Sr 3) The former can be conducted using hydrogeochemical diagrams such as
f1 =
(Sr 3 − Sr 1) (9) Durov, Piper and Stiff, among others. Piper diagrams are used by the
methodology presented herein, with which the modeled results are
The EMs are calculated using Equation (6) for f3 and Equation (9)
compared to the actual data corresponding to the different stages. The
for f1.
quantitative calibration of the model with actual data is performed by
The only solution for f2 when ft = 1 can be obtained by substituting
obtaining an ionic delta, which is the difference between its parameters.
f1 and f3 in Equation (1):
This calibration enables us to observe whether there are differences
f2 = 1 − f1 − f3 (10) between the virtual chemical compositions and the actual data. In
particular cases, the ionic delta can be close to zero, indicating that
Using Equations (7), (9) and (10), the mixing fractions of each
mixing is the dominant process in the chemical evolution of ground-
sample of the system are obtained.
water in the system. Inverse modeling or water-rock interactions are
needed for some cases. This study used Piper diagrams; the modeled
2.2.7. Virtual samples used in chemical modeling results were compared to the real data during the different stages.
Given a system with m samples, equation (1) is written:
3
2.2.9. Modeling with water-rock interaction of virtual samples
∑ fmi =1
Taking into account the saturation indices (SI) obtained from the
i=1 (11)
actual data and the fact that the mineral phases are consistent with the
Where m is the sample number, and i is the end member mixture. geology of the hydrogeological system, the rock-water interaction is
The mixing fraction of each of the EM (fmi) is multiplied by each modeled with the data from virtual composition 1 obtained from the
chemical parameter (ani) of that EM, and their sum represents the vir- mixing model. Using this modeling, virtual composition 2 is obtained,
tual chemical composition (pH, Cl, Na, HCO3, etc.) of each parameter, which considers the interaction with the major minerals. The stability
of the m-th sample of the system. of the mineral phases depends on the physicochemical conditions of the
3 system, for which phase diagrams are used.
Vm = ∑ fmi ani
i=1 (12)
2.2.10. Second calibration (actual versus virtual chemical composition data
As an example, for the vector V1 (virtual sample 1), each parameter 2)
is obtained as shown in equation (13). V2 and V3 for the samples are The Piper diagram is used again to compare actual data to virtual
shown in equations (14) and (15). For the n-th sample (Vn) is obtained chemical composition data 2. The presence of ionic delta would be
by equation (16); i is the EM (f1, f2 and f3). indicative of other processes occurring, such as ion exchange, dissolu-
fmi, is the matrix fractions mixing system for each EM, iWhere m is tion and/or precipitation of the main mineral phases. In these cases,
the sample number of the system, and n are the physical and chemical inverse modeling is applied as a next step to identify and quantify these
parameters of the water of the EM. processes. If the ionic delta is close to zero, this would indicate that the

100
J. Moran-Ramírez et al.
Fig. 2. Location of samples, section and regional geology of the study area.
mixing and the water-rock interaction were the dominant processes in
the system. Lastly, calibration is performed.
2.2.11. Inverse modeling
Normally, the inverse modeling is performed between two samples
along flow paths of the system, with a different ionic delta zero.
However, with this method, inverse modeling is performed between
real and virtual samples for the same site. The results of this modeling
with water-rock interactions were compared to the actual data. Inverse
modeling shows other processes such as ion exchange, dissolution and/
or precipitation of the primary species, along with redox processes that
occur in the groundwater.
3. Results and discussion
3.1. Geological and hydrogeological framework
The region known as Valles-San Luis Potosi Platform (Carrillo-
Martinez, 1981) is located in the western part of Mexico on the edge of
the SMO. This region consists of sedimentary rocks from the Mesozoic
to Recent ages (Fig. 2). The valleys of the area consist of a system of
close anticline and syncline folds with a north-south orientation. Salto
Valley and Chantol are the main valleys in the study area, which are
bordered by three major mountain ranges namely, the Palmillas, La
Colmena and Cerro Alto (Fig. 2).
The oldest rocks are associated with the Guaxcama Formation from
the Lower Cretaceous, which consist of deposits of gypsum (CaSO4) and
dolomite (CaMgCO3). This unit is distributed toward the northwest
Rioverde and it is not exposed in the study area (Fig. 2). Covering the
Guaxcama Formation is El Abra Formation, which consists of a lime-
stone platform and reefs with thick to massive stratification (López-
Doncel, 2003). This unit is the main aquifer in the San Luis Valley
Platform, with a thickness that exceeds 1000 m; its main recharge zone
is located in the SMO. Nonetheless, given its characteristics of frac-
turing and high primary and secondary permeability, its recharge is
101
Applied Geochemistry 95 (2018) 97–109
significant in the western part of the study area and throughout the
entire SMO.
The regional groundwater flow dynamics can be adapted as pro-
posed by Tóth (1963), where the recharge areas are in the raised por-
tion of the region and unloading areas in the lower zones (valleys).
Regional recharge starts from the Sierra de Alvarez (Moran-Ramirez
and Ramos-Leal, 2012; Morán-Ramírez et al., 2013) and flows into the
region of the Huasteca, leaving behind evidence of their presence
through springs in the middle zone (Media Luna), Tamasopo and Nar-
anjo. The flow systems can be classified as local, intermediate and re-
gional flows. The first flows come from the mountains close to the
nearest valleys; in the Huasteca area, there are many uses of this type
that are characterized by small springs. The seconds have a larger travel
distance and are from sizeable sources such as Minas Viejas and El
Salto, among others. Regional systems have flows traveling at greater
depths and at greater distances from the Sierra de Alvarez to Ciudad
Valles and are represented by large springs such as El Nacimiento,
Taninul, El Mante and El Coy.
Fig. 2 shows an east-west section, indicating the distribution in the
valleys of the San Felipe and Mendez formations and the great thickness
of El Abra Formation. The valleys are formed by synclines and the
mountains by anticlines. Some sinkholes are present in the axes of the
anticlines, such as the Sierra del Toro. This figure shows the flow sys-
tems, the regional flow reaches to flow through the Guaxcama For-
mation; the intermediate flow circulates in El Abra Formation, while
the local flow is represented by local recharge throughout the region.
The valleys of El Abra Formation are covered by the San Felipe and
Mendez formations, the former consisting of a package of 100 m thick
argillaceous limestone and the latter of 80 m thick shale. Evidence of
local and regional recharge is located in Agua Fria, La Laguna, Minas
Viejas and Los Gatos. Locally, the behavior is similar to that of an un-
confined aquifer in the hills and a semi-confined aquifer in the valleys.
The relationships between flow systems and their physicochemical
behavior during their evolution were proposed by Toth (1963,1999);
which describes the characteristics of groundwater flow for each
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Fig. 3. Hydrogeological section showing the hydrodynamic behavior of groundwater flows, water groups and ternary mixing in the intermountain valleys.

102
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Fig. 4. a) Relationships between Ca2+ + Mg2+ vs HCO− 2-

3 + SO4 b) (Ca
2+
+ Mg2+) - (HCO− 2-
3 + SO4 ) vs (Na
+
+ K+) – Cl− and c) HCO−
3 /Na
+
vs Ca2+/Na+.

system. In the study area, it is possible to distinguish different types of influence of evaporation. In the study area, there is interaction between
flow, based on their chemical composition. rock and water; this finding is congruent with the geology of the site,
because it is a karstic zone, and water has a rapid infiltration. However,
the evaporation processes are also present in the study area, as the TDS
3.2. Conceptual model and hydrogeochemistry
are constant but the amount of dissolved ions increases in a zone of
carbonates, which interacts with the water and increases the amount of
In general, the wells and springs are located in the discharge area of
Ca2+ and HCO− 3 in the medium (Fig. 5a and b).
the study area, where it is possible that the regional fracturing con-
The Piper (1953) diagram (Fig. 6) identifies 3 water families: cal-
tributes to the mixing of the different flows of the system (Fig. 3). The
cium bicarbonate (Ca2+-HCO3), sulfate calcium bicarbonate (Ca-HCO3-
alkalinity in the recharge zone of the Palmillas Range is 300 mg/L and it
SO4) and sulfate calcium (Ca-SO2- 4 ). The flow types were identified with
increases to values close to 400 mg/L in the Naranjo Valley and
scatterplots of cation vs anions in Mifflin (1988) diagrams (Fig. 7).
360 mg/L in the Los Gatos Valley (Fig. 3). The electrical conductivity is
These can be used to identify three main flows: a) Local flow with low
close to 500 μS/cm in the recharge zone of the Palmillas Range, in the
ionic content, due to its small circulation and/or short residence time
Naranjo Valley it increases and in the Los Gatos Valley it reaches
(local recharge); b) Intermediate flow with more water-rock interac-
1400 μS/cm. Chlorides are low in the recharge zone (1 mg/L), and in
tion, perhaps because of its higher travel and/or longer residence time;
discharge areas in the Naranjo Valley they are as high as 9 mg/L
and c) Regional flow with the highest ionic content due to greater
(Fig. 3). The pH in the recharge zone is 6.9 and increases in the Naranjo
movement and residence time (Fig. 7).
Valley up to 7.6 and decreases in some springs in the same valley. The
The Ca-HCO3-family corresponds to a local system of shallow water
sulfates are low in the Palmilla Range (20 mg/L) and show changes in
created by the circulation of water in the limestone of El Abra
the Naranjo Valley, reaching the highest values near the La Colmena
Formation, since the CaCO3 contained in the limestone dissolves during
Range (450 mg/L) and decreasing again towards the Los Gatos Valley.
infiltration, which explains its calcium bicarbonate (Ca-HCO3) char-
The temperature is low in the recharge zone (22 °C) and increases in the
acter (Figs. 6 and 7). In the deeper flow (regional groundwater), water
Naranjo Valley, reaching the highest value in the La Colmena Range
circulates through gypsum (CaSO4.2H2O) in the Guaxcama Formation,
(27 °C), and in the Los Gatos Valley it falls to 23 °C (Fig. 3).
producing a calcium sulfate water type (Figs. 6 and 7). Furthermore,
Fig. 4a shows the relationship between Ca 2++ Mg2+ and
because El Abra Formation is highly fractured, mixtures of deep and
HCO− 2-
3 + SO4 . The 1:1 correlation line divides two regions, a) with
shallower water flows may be produced through deep fractures,
alteration of silicates and b) with alteration of carbonate rocks. As can
creating the mixed group Ca-HCO3-SO4 (Figs. 6 and 7).
be observed, the samples fall on the line and are located in the region of
This phenomenon can be seen in Fig. 3, which shows the distribu-
carbonate alteration, which is congruent with the distribution of
tion of water families, flow system and the rock-water interaction
limestones of El Abra Formation in the region. Only two samples are
throughout the flow. Water flows from W to E through the El Abra
found in silicate alteration; these two samples are found in two small
Formation and the presence of faults and fractures can produce ternary
springs that rise under a basaltic rock.
mixtures. Because the karstic characteristics of the region, local re-
Fig. 4b shows the relationship of (Ca2+ + Mg2+) - (HCO− 2-
3 + SO4 )
charge (f1) is significant, and it mixes with the groundwater flow (f2). In
vs (Na+ + K+) - Cl−. The marked black line is related to the cation
addition, the presence of faults and fractures that cut the folds in the
exchange line. As seen, the samples are not distributed with this line,
SMO provides a hydraulic connection that leads to the mixing of re-
which is interpreted as the ion exchange; as such, it is not an important
gional flows (f3) with shallow flows (Fig. 3).
process in the evolution of groundwater.
Fig. 4c shows the relationship between HCO− 3 /Na
+
and Ca2+/Na+,
which illustrates the different materials (evaporites, rainwater, silicates 3.3. Mixing model
and carbonates) through which water circulates and interacts. As seen
in the figure, water in the region is related to local recharge water that The three EM present in the study area and the two main evolu-
infiltrates and interacts with calcareous rocks of the region. tionary processes in groundwater are identified on the scatter plot of Cl
Fig. 5 shows the Gibbs diagram, in which TDS are compared to the and Sr (Fig. 8). Both elements can behave as conservatives or tracers. In
a) anions Cl−/(Cl-+HCO− +
3 ) and b) cations Na /(Na +Ca
+ 2+
). The this work, Sr is used as a tracer, considering that the saturation indexes
Gibbs diagram is composed of three regions bounded by a boomerang; of Sr, indicates that this is not interacting with the medium. In addition,
at the bottom of this water is finded with low chemical concentrations. the main aquifer is a carbonate formation, which contributes to having
Considering that it is meteoric water, in the central part, there is water high Ca concentrations, approximately 8 orders of magnitude with re-
that has interacted with rock; the upper part contains waters with the spect to Sr. The pH conditions favor the formation of calcium

103
J. Moran-Ramírez et al.
Fig. 5. Gibbs diagram, a) where TDS are compared to the anions (Cl−/(Cl-+HCO−
Fig. 6. Piper diagram of water sampled in the intermountain Salto Valley.
carbonates. In the region, groundwater has a pH of 6.6–8.8, which
makes the formation of SrCO3 unlikely (Table 1). The former has a low
concentration of Sr and high Cl, whereas the latter has high values of Sr
and low Cl.
f1 is associated with local recharge and is located in the western
study area. This EM has low values for Cl and Sr and a temperature of
22.0 °C.The lowest values were found in sample 8, which showed 0.003
meq/L of Cl and 0.009 meq/L of Sr. These values were used for the
mixing model, which was of the Ca-HCO3 type (Figs. 3 and 8). f2 is
represented by sample 19, located in the south western study area. This
sample represents an intermediate flow from the SMO. f2 has a tem-
perature of 23.3 °C, Cl concentrations of the order of 1.55 meq/L Cl and
0.025 meq/L of Sr. The type of water for this EM is Ca-HCO3-SO4
(Figs. 3 and 8). f3 is represented by sample 11 in the northern Los Gatos
River. This EM represents regional flow, has a temperature of 23.8 °C
and Cl concentrations (0.15 meq/L), as well as a high concentration of
104
Applied Geochemistry 95 (2018) 97–109
+ +
3 ) b) where TDS are compared to the cations Na /(Na +Ca
2+
).
Sr (0.124 meq/L). Water of this type is Ca-SO4 (Figs. 3 and 8).
Cl and Sr were used to obtain mixing fractions (f1, f2, f3) through
equations (4), (7) and (10). The Cl ranges for groundwater in the Salto
Valley are 0.003–1.55 meq/L and 0.009–0.124 meq/L of Sr. The per-
centages of these mixing fractions obtained for each sample considered
are presented in Table 2.
According to the model for the ternary mixture, the water in the
study area, f1 (local recharge) contributes 68.3%, f2 (intermediate flow)
contributes 12.3% and f3 (regional) contributes 19.4%. Since this model
represents a binary mixture between f1 and f3, it is identified only in
samples 2, 14, 16 and 24 (Table 2).
With mixing fractions (Table 2) and applying Equation (12), the
theoretical chemical composition of each component of the water
samples was obtained. For the chemistry of each of the water samples
composition, Equations (13)–(16) are used, or for the entire system,
Equation (17) applies (Table 3). This table represents the first virtual
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Table 2
Mixing fractions for groundwater in the study area.
Sample C1 C2 C3

1 0.83 0.16 0.02

2 0.73 0 0.27
3 0.86 0 0.14
4 0.8 0.05 0.15
5 0.79 0.18 0.03
6 0.66 0.29 0.05
7 0.23 0.01 0.76
8 0.7 0.09 0.21
9 0.61 0.12 0.27
10 0.77 0.01 0.22
11 0.8 0.09 0.11
12 0.94 0.03 0.03
13 0.83 0.06 0.11
14 0.14 0 0.86
15 0.66 0.02 0.32
16 0.32 0 0.68
17 0.89 0.02 0.09
18 0.82 0.06 0.12
19 0.91 0.04 0.05
20 0.24 0.74 0.02
21 0.27 0.7 0.03
22 0.84 0.13 0.03
23 0.8 0.17 0.03
Fig. 7. Evolution and groundwater flow systems in the study area, according to 24 0.95 0 0.05
Mifflin (1988).

Table 3
Theoretical chemical composition of rock samples without the water mixing
interaction model.
Sample ALK Cl SO4 Ca K Mg (mEq) Na (mEq)

(mEq) (mEq) (mEq) (mEq)

1T 3.61 0.05 6.3 6.31 0.03 0.88 0.19

2T 4.1 0.42 3.71 5.62 0.03 0.57 0.37
3T 3.79 0.22 4.07 5.55 0.03 0.62 0.23
4T 3.74 0.24 4.38 5.6 0.03 0.65 0.26
5T 3.68 0.07 6.68 6.5 0.04 0.92 0.23
6T 3.89 0.11 8.35 7.35 0.04 1.1 0.35
7T 5.49 1.18 3.36 6.55 0.05 0.48 0.98
8T 4.06 0.34 5.07 6.15 0.04 0.72 0.38
9T 4.28 0.44 5.49 6.49 0.04 0.76 0.48
10T 3.96 0.34 3.93 5.61 0.03 0.6 0.31
11T 3.77 0.18 5.19 5.98 0.03 0.75 0.25
12T 3.48 0.06 4.4 5.43 0.02 0.66 0.11
13T 3.73 0.18 4.79 5.78 0.03 0.7 0.22
4T 5.76 1.33 3.07 6.64 0.06 0.44 1.1
15T 4.27 0.5 3.93 5.85 0.04 0.59 0.45
16T 5.24 1.05 3.26 6.32 0.05 0.48 0.87
17T 3.62 0.14 4.17 5.44 0.03 0.63 0.17
18T 3.76 0.19 4.78 5.8 0.03 0.7 0.24
19T 3.54 0.09 4.56 5.55 0.03 0.68 0.14
20T 4.46 0.14 15.08 10.51 0.08 1.85 0.7
21T 4.43 0.15 14.47 10.24 0.08 1.78 0.68
22T 3.61 0.06 5.93 6.15 0.03 0.84 0.19
Fig. 8. Ternary mixing model using Cl and Sr concentrations from the inter- 23T 3.67 0.07 6.59 6.46 0.04 0.91 0.22
mountain valley, where f1, f2 and f3 are end members. 24T 3.48 0.08 3.95 5.25 0.02 0.61 0.1

model obtained from the chemical composition of the EM f1, f2 and f3.
real processes occurring in the aquifer and reflects their interaction
The Piper diagram was used to compare the theoretical chemical
with the carbonate rocks in the region. This result suggests an inter-
composition with the actual water in the study area; a difference can be
action with the limestone rocks of El Abra Formation, the marl of San
seen between the two compositions (Fig. 9). This difference is largely
Felipe Formation and the shale of Mendez Formation, since the most
because the water-rock interaction process was not considered in the
important mineral phases are calcite, dolomite and gypsum. In the
theoretical composition. In Fig. 9, we see that in the actual data, the
study area, direct modeling shows that 76% of the samples are over-
dissolution of calcite is the dominant process; while the virtual data is
saturated with respect to calcite while under saturated with respect to
more strongly influenced by the gypsum.
gypsum (92%), barite (55%) and dolomite (81%), indicating the dis-
In response to the above, direct modeling of the actual samples was
solution of these minerals when these are in contact with water as it
performed to determine the mineral phases present in the groundwater
circulates through these rocks. In some instances, water can be satu-
(Table 4).
rated with respect to dolomite due to contributions of Mg. All samples
The resulting hydrogeochemical model is an approximation of the
are under saturated with respect to celestite (100%). The Barite is very

105
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Fig. 9. Piper diagram representing the difference between the real data and the mixing model. Note: Circles represent the theoretical samples and crosses the field
data).

Table 4
Saturation indices for the main mineral phases in the groundwater of the study area.
Sample Barite Calcite Chrysotile Stroncionite Dolomite Gypsum Halite

1 0.14 0.08 −9.13 −1.18 −0.4 0.77 −10.38

2 −0.47 0.14 −10.48 −2.89 −0.69 −1.96 −8.85
3 −0.98 0.31 −9.46 −3.31 −0.49 −2.6 −9.13
4 −1.67 0.29 −8.94 −3.21 −0.37 −2.76 −8.47
5 −0.37 0.05 −8.59 −1.75 −0.27 −1.4 −10.15
6 0.23 2 3.68 −1.2 3.57 0.94 −9.47
7 −0.62 0.14 −10.12 −2.86 −0.81 −2.42 −8.03
8 0.19 0.04 −10.82 −2.45 −1.09 −2.02 −8.59
9 −0.24 0.13 −8.72 −2.36 −0.41 −2.07 −8.19
10 0.11 0.17 −9.33 −2.33 −0.73 −1.77 −8.74
11 0.41 −0.03 −13.34 −1.95 −1.69 −1.36 −9.02
12 −1.96 −0.02 −10.14 −4 −1.03 −3.41 −9.51
13 0.29 0.13 −12.37 −2.13 −1.41 −1.52 −9.1
14 0.2 0.28 −10.66 −2.32 −1.04 −1.7 −8.11
15 0.04 0.13 −13.09 −2.55 −1.61 −1.77 −8.72
16 0.48 0.04 −12.97 −2.04 −1.55 −1.37 −8.23
17 −0.8 0.17 −9.82 −3.06 −0.88 −2.46 −9.16
18 −0.71 −0.09 −11.73 −2.96 −1.65 −2.49 −8.94
19 −0.9 0.34 −9.42 −3.2 −0.59 −2.6 −9.32
20 0.79 0.19 −9.72 −0.98 −0.45 −0.78 −8.93
21 −0.19 0.62 −6.7 −1.03 0.61 −0.83 −9.01
22 0.32 1.09 −2.44 −1.09 1.89 −0.83 −10.07
23 −0.04 0.12 −7.62 −1.49 −0.17 −1.13 −10.13
24 −1.37 0.01 −9.95 −3.74 −1.2 −2.81 −9.56
25 −0.74 −0.14 −9.17 −2.31 −0.73 −0.67 −12.02
26 0.16 0.9 −3.22 −0.79 1.52 −0.8 −8.91
27 −0.08 0.51 −7.45 −2.11 0.16 −1.65 −7.66

106
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Table 5 close to equilibrium and thus can sometimes find under saturated or
Results of the mixing model with water-rock interaction. saturated with respect to this mineral (Table 4).
Sample ALK Cl S Ca K Mg Na
3.4. Inverse modeling of water-rock interaction
(mEq) (mEq) (mEq) (mEq) (mEq) (mEq)

1M 5.643 0.051 9.008 10.36 0.033 4.37 0.08 Modeling was performed using the saturation rates and the results
2M 6.671 0.412 0.571 6.1 0.032 0.49 0.369 from the theoretical mixture (Table 3) and by taking into account the
3M 5.758 0.22 0.058 5.576 0.028 0.315 0.226 water-rock interaction (Table 5). As shown in the Piper diagram
4M 5.54 0.237 0.095 5.388 0.029 0.437 0.26 (Fig. 10), the chemical composition of the mixture obtained from the
5M 5.964 0.068 2.262 6.57 0.035 2.104 0.227
6M 4.268 0.113 230.8 184.6 0.045 52.86 0.347
theoretical mixing modeled with the water-rock interaction has a slight
7M 8.645 1.163 0.195 6.4 0.053 0.403 0.983 deviation from actual samples. This is partly due to different processes
8M 6.702 0.339 0.415 7.898 0.036 0.42 0.375 that release ions into the water-such as dissolution or precipitation of
9M 6.513 0.434 0.468 5.848 0.041 0.912 0.477 minerals - causing an increase or decrease in its chemical composition
10M 5.723 0.333 0.882 5.992 0.03 0.001 0.314
which is not considered in the mixing process (Table 5).
11M 7.89 0.181 1.889 8.52 0.032 0.156 0.248
12M 5.548 0.056 0.021 5.166 0.024 0.421 0.105 With the rock-water interaction process, there is a better estimation
13M 0.768 0.175 1.327 6.948 0.03 0.115 0.224 of, or a smaller difference between actual samples and those modeled
14M 8.295 1.315 0.961 7.214 0.056 0.142 1.097 with water-rock interactions. This result can also be seen in Fig. 10,
15M 8.479 0.496 1.254 3.898 0.01 0.001 0.016 which shows the ionic deltas of the major ions in groundwater in the
16M 1.221 1.038 1.663 9.164 0.048 0.165 0.873
17M 5.379 0.1411 0.185 5.436 0.025 0.226 0.161
study area.
18M 6.222 0.192 0.165 5.756 0.03 0.15 0.238 The inverse modeling (Table 6) explains the ionic deltas of the
19M 5.23 0.085 0.13 5.604 0.026 0.232 0.137 chemical composition of mineral phases and the saturation indices of
20M 8.168 0.135 6.324 14.3 0.078 1.584 0.698 the initial (modeled) and end (current) solutions for the minerals cal-
21M 6.582 0.147 6.082 12.98 0.075 2.226 0.677
cite, dolomite and gypsum.
22M 4.342 0.062 7.794 9.99 0.032 3.43 0.186
23M 5.3 0.068 4.018 7.58 0.032 2.246 0.185 Table 6 Shows that the dissolution of calcite was the dominant
24M 4.734 0.079 0.088 4.74 0.023 0.225 0.102 process in 73% of samples, in 15% of the samples the dissolution of
gypsum was dominant and in only one sample the dissolution of do-
lomite was the dominant process.

Fig. 10. Piper diagram representing the difference between real and modeled mixtures with water-rock interaction. Note: Triangles represent mixed samples with
water-rock interaction, while circles represent the real samples.

107
J. Moran-Ramírez et al. Applied Geochemistry 95 (2018) 97–109

Table 6
Inverse modeling of the results from the mixing model with water-rock interaction.
Sample Calcite % Gypsum % Dolomite % NaX MgX2
(mmol) Total (mmol) Total (mmol) Total (mmol) (mmol)

Δ1 2.85E-03 46.2 −1.91E-03 31 −1.41E-03 22.8 −1.70E-06 8.49E-07

Δ6 5.03E-02 26.7 −1.13E-01 59.9 −2.52E-02 13.4 – –
Δ22 1.69E-03 33.2 −2.54E-03 50 −8.54E-04 16.8 – –

Δ3 −3.86E-04 85.5 3.34E-05 7.4 3.19E-05 7.1 – –

Δ9 −1.95E-04 95 5.31E-06 2.6 5.06E-06 2.5 – –
Δ11 −7.27E-04 98.9 3.65E-06 0.5 4.12E-06 0.6 – –
Δ13 −1.85E-04 85.2 2.40E-05 11.1 8.19E-06 3.8 – –
Δ16 −2.80E-04 80.3 6.45E-05 18.5 4.14E-06 1.2 – –
Δ17 −1.85E-04 98.3 1.04E-06 0.6 2.19E-06 1.2 – –
Δ18 −1.95E-04 97.8 8.60E-07 3.43E-06 – –

Δ4 −2.56E-04 96.5 4.60E-06 1.7 −4.63E-06 1.7 – –

Δ10 −7.27E-05 58.6 3.83E-05 30.9 −1.31E-05 10.6 – –
Δ14 −4.13E-06 29.4 8.71E-06 61.9 −1.23E-06 8.7 – –

Δ2 −4.00E-04 90.6 −1.49E-05 3.4 2.67E-05 6 – –

Δ5 −5.69E-04 71.8 −8.99E-05 11.3 1.34E-04 16.9 – –
Δ8 −5.69E-04 81 −8.98E-05 0.1 1.33E-04 18.9 – –
Δ19 −3.25E-04 93.6 −2.63E-06 0.8 9.70E-06 5.7 – –
Δ20 −7.19E-04 63.6 −2.86E-04 25.3 1.26E-04 11.1 – –
Δ21 −7.79E-04 59.7 −3.21E-04 24.6 2.05E-04 15.7 – –

Δ7 −2.04E-04 92.5 0 1.65E-05 7.5 – –

Δ15 −3.15E-04 89.4 0 3.75E-05 10.6 – –
Δ23 −6.68E-06 14.2 −8.59E-06 18.2 −3.19E-05 67.6 – –
Δ24 1.15E-04 77.1 8.26E-06 5.5 −2.59E-05 17.4 – –

The mass balance quantitatively provides the mole transfer that
occurs in the water-rock interaction process, for which the main mi-
nerals are calcite, dolomite and gypsum. Table 6 shows main groups in
which the main processes occur that explain the chemical composition
of the water in the study area.
As mentioned above, the system has two major evolutionary pro-
cesses, where in starting from the EM f1 (local flow), this is mixed with
f2 (intermediate flow) or f3 (regional flow), and depending on mixing
ratios, it is possible to group samples with similar processes.
In the first group, the values for calcite are positive (gain) and for
gypsum and dolomite negative (loss), which indicates that calcite is
dissolved and thereby provided the means for the formation of gypsum
and dolomite. In addition, only the sample 1 presented ion exchange
resulting from circulation through clay materials. In the second group,
the values for calcite are negative (loss) and for gypsum and dolomite
positive (gain), indicating that calcite was formed with ions resulting
from the dissolution of gypsum and dolomite. In the third group, the
values for calcite and dolomite are negative (loss) and for gypsum po-
sitive (gain), indicating that calcite and dolomite were formed and
gypsum was dissolved. In the fourth group, the values of calcite and
gypsum are negative (loss) and the values of dolomite are positive
(gain), indicating that calcite and gypsum were formed and dolomite
was dissolved.
In other groups, the values for calcite are negative (loss) and for
dolomite positive (gain), indicating that calcite was formed and dolo-
mite was dissolved. In sample 23, values for calcite, dolomite and
gypsum are negative, which indicates the formation of these minerals.
In sample 24, values for calcite and gypsum are positive and for dolo-
mite negative, suggesting the dissolution of calcite and gypsum and the
formation of dolomite. No model was produced for samples 8 and 12
because the initial and final composition showed an ionic delta close to
zero, which implies that mixing with water-rock interaction, occurred
as the predominating process.
4. Conclusions
The regional recharge zone is located in the Sierra Madre Oriental,
the intermediate flow is produced in the mountainous borders of the
Sierra Madre Oriental and local recharge occurs in the anticlines that
make up the elongated mountains where El Abra formation is exposed.
The main groundwater families in the intermountain valleys are
calcium bicarbonate (Ca-HCO3-), sulfate calcium bicarbonate (Ca-
HCO3-SO4) and calcium sulfate (Ca-SO4). The water in the Salto Valley
aquifer represents the ternary mixture of the three end members, which
are related to local, intermediate and regional flows. The mixing model
for the intermountain valleys shows that local recharge contributes
68.3%, intermediate flow 12.3% and regional recharge 19.4%.
The inverse modeling of the ionic delta indicates that the pre-
cipitation and/or dissolution of calcite, dolomite and gypsum, the main
processes occurring in the aquifer in the study area's intermountain
valleys.
The dissolution of calcite is the dominant process in system (73%),
followed by the dissolution of gypsum (15%).
With VISHMOD methodology, was identified in the first instance
qualitatively evolutionary tendencies, the end of the methodology
could be identified in more detail the groups with similar processes in
the evolution of groundwater.
According to these results, VISHMOD methodology approach
proved to be an excellent tool to understand the evolution of ground-
water, with this methodology the evolution could be represented as a
mixture of three end members, to identify water-rock interaction pro-
cesses that occur in the carbonated system which give rise the chemical

108
J. Moran-Ramírez et al.
signature of water at each site.
Acknowledgments
The authors thank the Potosin Institute of Scientific and
Technological Research that supported part of the study.
References
Abu-Jaber, N., 2001. Geochemical evolution and recharge of the shallow aquifers at Tulul
al Ashiqif NE, Jordan. Environ. Geol. 41, 372–383.
André, M., Franceschi, P., Pouchan, O., Atteia, 2004. Using geochemical data and mod-
elling to enchance the understanding of groundwater flow in a regional deep aquifer,
Aquitaine Basin. J. Hydrol. 305, 40–60.
Carrillo, M., 1981. Contribución al estudio geológico del Macizo Calcáreo el Doctor,
Estado de Querétaro. Univ. Nal. Auton. México, Rev. Inst. Geol. 5 (1), 25–29.
Dixon, W., Chiswell, B., 1992. The use of hydrolochemical section to identify recharge
areas and saline intrusions in alluvial aquifers, southeast Queensland: Australia. J.
Hidrol. 130, 299–338.
Genereux, D., Wood, S.J., Pringle, C.M., 1996. Chemical mixing model of streamflow
generation at La Selva Biological Station, Costa Rica. J. Hidrol. 199, 319–330.
Gómez, J.B., Aunqué, L.F., Gimeno, M.J., 2008. Sensivity and uncertainty analysis of
mixing and mass balance calculations whit Standard and PCA-based geochemical
codes. Appl. Geochem. 23, 1941–1956.
Helgeson, H.C., Brown, T.H., Nigrini, A., Jones, T.A., 1970. Calculations of mass transfer
in geochemical processes involving aqueous solutions. Geochem. Cosmochim. Acta
34, 569–592.
Hereford, G., Keating, H., Guthrie Jr., D., Chen, hu, 2007. Reactions and reaction rates in
the regional aquifer beneath the Pajarito Plateau, north-central New Mexico: USA.
Environ. Geol. 52, 965–977.
Hernández-Antonio, A., Mahlknecht, J., Tamez-Meléndez, C., Ramos-Leal, J., Ramírez-
Orozco, A., Parra, R., Eastoe, C.J., 2015. Groundwater flow processes and mixing in
active volcanic systems: the case of Guadalajara (Mexico). Hydrol. Earth Syst. Sci. 19
(9), 3937–3950.
Hidalgo, C., Cruz, J., 2001. Groundwater composition, hydrochemical evolution and mass
transfer in a regional detrital aquifer (Baza Basin, Southern Spain). Appl. Geochem.
16, 745–758.
Laaksoharju, M., Skarman, C., Skarman, E., 1999. Multivariate Mixing and mass balance
(M3) calculation a new tool for decudinghydrogeochemical information. Appl.
Geochem. 14, 861–871.
Ledesma-Ruiz, R., Pastén-Zapata, E., Parra, R., Harter, T., Mahlknecht, J., 2015.
Investigation of the geochemical evolution of groundwater under agricultural land: a
case study in northeastern Mexico. J. Hydrol. 521, 410–423.
109
Applied Geochemistry 95 (2018) 97–109
Lee, E.S., Krothe, N.C., 2001. A four-component mixing model for water in a karst terrain
in south-central Indiana, USA. Using solute concentration and stable isotopes as
tracers. Chem. Geol. 179, 129–143.
López-Doncel, R., 2003. La Formación Tamabra del Cretácico medio en la porción central
del margen Occidental de la Plataforma Valles-San Luis Potosí, centro-noreste de
México. Rev. Mex. Ciencias Geol. 20 (1), 1–19.
Mifflin, M.D., 1988. Region 5, great basin. In: Back, W., Rosenshein, J.S., Seaber, P.R.
(Eds.), Hydrogeology. Geological Society of America, pp. 69–78.
Moran-Ramirez, J., Ramos-Leal, J.A., 2012. Conexión de Cuencas Hidrogeológicas con
Modelos Hidrogeoquímicos. Aplicaciones en la Sierra Madre Oriental. Editorial
Académica Española, Mexico, pp. 68.
Moran-Ramirez, J., Ramos-Leal, J.A., 2014. The VISHMOD methodology with hydro-
chemical modeling in intermountain (karstic) aquifers: case of the Sierra Madre
Oriental, Mexico. J. Geogr. Geol. 6 (2), 132–144.
Morán-Ramírez, J., Ramos-Leal, J.A., López-Álvarez, B., Carranco Lozada, S., Santacruz-
De León, G., 2013. Comportamiento Hidrogeoquímico de Flujos Subterráneos en
Acuíferos Cársticos Fracturados, Aplicando Modelación Inversa: caso Huasteca
Potosina. Bol. Soc. Geol. Mex. 65 (1), 71–82 2013.
Morán-Ramírez, J., Ledesma-Ruiz, R., Mahlknecht, J., Ramos-Leal, J.A., 2016.
Rock–water interactions and pollution processes in the volcanic aquifer system of
Guadalajara, Mexico, using inverse geochemical modeling. Appl. Geochem. 68,
79–94.
Parkhurst, D.L., Apello, C.A.J., 1999. User Guide to PHREEQC (Version 2)—a Computer
Program for Speciation, Batch Reaction, One Dimensional Transport, and Inverse
Geochemical Calculations: U.S. Geological Survey Water Resources Investigations
Report. 99–4259.
Petitta, M., Primavera, P., Tuccimei, P., Aravena, R., 2010. Interaction between deep and
shallow groundwater systems in areas affected by Quaternary tectonic (Central Italy):
a geochemical and isotope approach. Environ. Earth Scí. http://dx.doi.org/10.1007/
s12665-010-0663-7.
Piper, A.M., 1953. A Graphic Procedure in the Geochemical Interpretation of Water
Analysis. United States Geological Survey, Washington D.C. OCLC 37707555. ASIN
B0007HRZ36.
Plummer, L.N., Back, W., 1980. The mass balance approach: application to interpreting
the chemical evolution of hydrologic systems. Am. J. Sci. 280, 130–142.
Plummer, L.N., Parkhurst, D.L., Fleming, G.W., Dunkle, S.A., 1988. PHRQPITZ, a com-
puter program incorporating Pitzer's equations for calculating of geochemical reac-
tions in brines. U.S.G.S. Water Resour. Invest. 88–4153, 310.
Ramos-Leal, J., Martínez-Ruiz, V., Rangel-Méndez, J., Alfaro de la Torre, M., 2007.
Hydrogeological and mixing process of waters in aquifers in arid regions: a case study
in San Luis Potosí Valley, Mexico. Environ. Geol. 53, 325–337.
Tóth, J., 1963. A theoretical analysis of groundwater flow in small drainage basins. J.
Geophys. Res. 68 (16), 4795–4812.
Tóth, J., 1999. Groundwater as a geologic agent: an overview of the causes, processes,
and manifestations. Hydrogeol. J. 7, 1–14.