remote sensing
Article
Framework for Remote Sensing and Modelling of
Lithium-Brine Deposit Formation
Cristian Rossi 1, * , Luke Bateson 2 , Maral Bayaraa 1 , Andrew Butcher 2 , Jonathan Ford 2 and Andrew Hughes 2






Citation: Rossi, C.; Bateson, L.;
Bayaraa, M.; Butcher, A.; Ford, J.;
Hughes, A. Framework for Remote
Sensing and Modelling of
Lithium-Brine Deposit Formation.
Remote Sens. 2022, 14, 1383. https://
doi.org/10.3390/rs14061383
Academic Editors: Ana C. Teodoro,
Joana Cardoso-Fernandes
and Alexandre Lima
Received: 7 February 2022
Accepted: 10 March 2022
Published: 12 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Remote Sens. 2022, 14, 1383. https://doi.org/10.3390/rs14061383
1 Satellite Applications Catapult, Harwell Campus, Didcot OX11 0QR, UK; maral.bayaraa@sa.catapult.org.uk
2 British Geological Survey, Keyworth, Nottingham NG12 5GG, UK; lbateson@bgs.ac.uk (L.B.);
asb@bgs.ac.uk (A.B.); jford@bgs.ac.uk (J.F.); aghug@bgs.ac.uk (A.H.)
* Correspondence: cristian.rossi@sa.catapult.org.uk
Abstract: The demand for “green” metals such as lithium is increasing as the world works to
reduce its reliance on fossil fuels. More than half of the world’s lithium resources are contained in
lithium-brine deposits, including the salt flats, or “salars”, of the Andean region of South America,
also known as the Lithium Triangle. The genesis of lithium-brine deposits is largely driven by the
leaching of lithium from source rocks in watersheds, transport via groundwater systems to salars, and
evaporative concentration in salars. The goal of this research is to create a consistent and seamless
methodology for tracking lithium mass from its source in the watershed to its greatest concentration
in the nucleus. The area of interest is in and around Bolivia’s Salar de Uyuni, the world’s largest
salt flat. We explore how Li-brine deposits form, where the water and solute come from, how the
brines are formed, and how abstraction affects the mass balance inside the salar. To support the
entire system, open-source Earth observation (EO) data are analysed. We found that by constructing
a flexible and repeatable workflow, the question of how lithium reaches the Salar de Uyuni can be
addressed. The work demonstrated the importance of groundwater flow to the river network and
highlighted the need for flow data for the main river supplying the salar with both water inflow and
lithium mass.
Keywords: Earth observation; geological remote sensing; lithium; brine; water balance; Salar de
Uyuni; Bolivia; Lithium Triangle
1. Introduction
Net-zero policies necessitate a shift to cleaner transportation and renewable energy
storage, but energy and mineral supply chains face numerous challenges. From a mineral
standpoint, low-carbon technology is demanding. The World Bank predicts that more
than 3 billion tonnes of minerals and metals will be required to deploy solar, wind, and
geothermal power, as well as energy storage, if our planet is to stay within the COP21 Paris
Agreement target of a global average temperature increase of well below 2 ◦ C. In light of
this, demand for battery metals such as cobalt and lithium is expected to rise 500% from
current levels by 2050 [1]. The speed and scale with which green energy technologies can
reduce greenhouse gas emissions and enable climate-resilient growth will be determined
by mineral availability.
Electric vehicles, which use Li-ion rechargeable batteries as a key component, are at
the forefront of green technology. Lithium is also used for the production of other batteries,
such as cell phones and laptops. The battery market currently takes 71% of global lithium
production. Lithium is also employed in the ceramics and glass industries, accounting
for 14% of the production [2]. Various other uses, including lubricating greases and air
purification, take the remaining 15%. Rapid growth in demand is driven by the battery
market, which increased by 48% in 10 years (2010–2020). Worldwide lithium production
rose more than 300% from 25,300 tonnes in 2010 to 82,000 tonnes in 2020 [2,3].
https://www.mdpi.com/journal/remotesensing
Remote Sens. 2022, 14, 1383 2 of 22

Within this context, the search for and extraction of lithium are becoming an important
revenue source for many world economies. In particular, the U.S. Geological Survey [2]
estimates that the first three producers lie within the South American “Lithium Triangle”,
i.e., Bolivia, with 21 million tonnes resources; Argentina, with 19.3 Mt; and Chile, with
9.6 Mt. Other important producers are the United States, with 7.9 Mt, and Australia, with
6.4 Mt. Security of supply of lithium to the global markets and increasing expectations
by consumers for responsibly sourced raw materials result in a growing global interest in
lithium resources and their extraction.
Lithium is extracted from three types of deposits: “hard-rock” pegmatites, continental
brines in “salars”, and hydrothermally altered clays. Brine deposits account for around
75% of the world’s lithium, due to lower costs of extraction [4,5]. Li-rich brines are found
predominantly in basins of internal drainage or salars which are mainly located in the Andes
in the Lithium Triangle countries. They are endorheic basins, meaning that the only exit for
the inflowing water is via evaporation. This means that water, rich in lithium, flows to the
salars and is then evaporated, concentrating the minerals. Over long time periods (~Ma),
high-density brines collect in a nucleus that contains economic concentrations of lithium
which is then pumped out. Whilst extraction from rocks requires a hydrometallurgical
process (crushing and heating), extraction from brines, i.e., saline aquifers, involves the
pumping from the shallow subsurface and a set of solar evaporation ponds where the
liquid solution is moved from one to another until a desired concentration, or “purity”, is
reached. In this process, other components are removed by either precipitation or chemical
reactions through the introduction of reagents. This is necessary since lithium coexists
in brines with other elements, such as magnesium, that are noneconomic. To be noted,
the evaporation/precipitation process is a slow one, taking one to two years from the
water pumping to the processing of the concentrated brine in the processing plant, where a
further processing is employed to produce lithium carbonate [5].
The Salar de Uyuni in Bolivia (Figure 1) is being used as a study area for developing a
repeatable and seamless workflow for tracking lithium from its source in the watershed
to the salar nucleus at its maximum concentration. It gives a systems-based knowledge
of the genesis of lithium-brine deposits, which can help with broader issues such as
resource reporting and establishing uncertainty bounds to resource estimations. It is
outwith the purpose of the study to develop sociopolitical considerations regarding the
extraction, and this work is a purely technological one, making use of a combination of
several scientific disciplines, i.e., remote sensing, geology, and hydrology. Nevertheless,
given the study area, it is worth summarising the work of Barandiarán, who analysed
sociopolitical debates in the Lithium Triangle [6]. Just focusing on Bolivia, which has
yet to start large-scale commercialisation, Barandiarán concluded that lithium is used to
build a sociotechnical imagery where the government supports scientific and technological
research and development, achieved by Bolivian-trained scientists, to embody political
values linked to the modernisation of the nation.
Remote sensing is being widely used for geological applications [7]. Some Earth
observation missions are specifically built with geology as the principal application area,
such as the multispectral Advanced Spaceborne Thermal Emission and Reflection Ra-
diometer (ASTER) [8]. Indeed, geological applications are particularly effective with data
capturing short-wave infrared (SWIR) and thermal infrared (TIR) electromagnetic bands,
where the largest spectral variations are taking place, thus allowing accurate land cover
discrimination. Among geological applications, examples useful to this study are around
(1) fault and lineament mapping [9,10], where digital elevation models (DEMs) and mul-
tispectral data are used in relation to geomorphological variations and colour/spectral
changes; (2) alteration unit mapping [11,12], where multispectral data and techniques such
as principal component analysis (PCA) and spectral angle mapping are used to derive iron
oxide minerals, hydrothermal alteration zones, and other units; (3) mineral quantification
mapping and lithological discrimination [13,14], where multi- and hyperspectral data are
used to detect specific mineral spectral features, e.g., absorption bands, for sedimentary,
 Remote Sens. 2022, 14, x FOR PEER REVIEW 3 of 21

Remote Sens. 2022, 14, 1383 3 of 22

rive iron oxide minerals, hydrothermal alteration zones, and other units; (3) mineral quan-
tification mapping and lithological discrimination [13,14], where multi- and hyperspectral
data are used to detect specific mineral spectral features, e.g., absorption bands, for sedi-
igneous,
mentary,and metamorphic
igneous, rocks; and (4)
and metamorphic groundwater
rocks; investigationinvestigation
and (4) groundwater [15,16], where[15,16],
several
types
where several types of remote sensing technologies are used to derive indicatorssince
of remote sensing technologies are used to derive indicators of groundwater, of
remote sensingsince
groundwater, cannot penetrate
remote deep
sensing into surface.
cannot penetrate Typical
deep groundwater indicators
into surface. Typical are fea-
ground-
tures
waterassociated
indicatorswith rechargeassociated
are features zones (rivers,
withlakes, etc.),zones
recharge discharge zones
(rivers, (springs,
lakes, etc.), soil
etc.), discharge
moisture, landforms, drainage characteristics, depth of weathering, and all the
zones (springs, etc.), soil moisture, landforms, drainage characteristics, depth of weather- indicators
derived
ing, andfrom the
all the above-described
indicators applications
derived from [7].
the above-described applications [7].

Figure (a)Aerial
Figure 1. (a) Aerialview
viewofoflithium
lithium ponds
ponds at the
at the Salar
Salar de Uyuni,
de Uyuni, Bolivia
Bolivia (© contains
(© contains modified
modified Co-
pernicus Sentinel
Copernicus data
Sentinel (2017),
data (2017),processed byby
processed ESA,
ESA,CCCCBY-SA 3.03.0
BY-SA IGO);
IGO);(b)
(b)view
viewacross
acrossthe
theponds,
ponds,ofof
about 11 km
about km width
width (©
(© UK
UK Research
Research andand Innovation
Innovation (UKRI)/British
(UKRI)/BritishGeological
GeologicalSurvey
Survey(BGS)).
(BGS)).

Remote sensing of lithium

Remote lithium deposits
deposits isis aa relatively
relativelynew newapplication
applicationand andonlyonlya afewfew
studies have been published, also reviewed by Cardoso et al.
studies have been published, also reviewed by Cardoso et al. [17], with a few papers [17], with a few papers pub-
lished by the
published by same
the sameresearch groupgroup
research in scientific journals
in scientific (e.g., [18])
journals and
(e.g., a few
[18]) and other
a fewconfer-
other
ence proceedings.
conference Specifically,
proceedings. [17–20] refs
Specifically, cover the extraction
[17–20] cover the from rocks and
extraction from address
rocks theand
detection
address theofdetection
Li-bearingofpegmatites
Li-bearing throughpegmatitesspectral
throughanalysis.
spectralThere are several
analysis. Therelimitations
are several
for an accurate
limitations for an discrimination, linked to the
accurate discrimination, degree
linked of weathering
to the of the host of
degree of weathering rock,
the the
host
vegetation
rock, covering,covering,
the vegetation and the potential need for need
and the potential thermal for bands
thermal [17]. Rossi
bands et al.
[17]. [21] et
Rossi instead
al. [21]
covers the
instead extraction
covers from subterranean
the extraction from subterraneanbrines with an approach
brines that combines
with an approach several
that combines
geological and geobotanical proxies. The study presented in
several geological and geobotanical proxies. The study presented in this paper is the this paper is the first study
first
covering
study the genesis
covering of lithium
the genesis brines in
of lithium salt lakes;
brines in saltitlakes;
does not focus
it does notonfocus
the mineral explo-
on the mineral
ration of extraction
exploration itself,itself,
of extraction i.e., the
i.e.,derivation of a prospectivity
the derivation of a prospectivitymap, map,
but provides
but providesan evi- an
dence-based model
evidence-based model andandmoremore general
general hydrogeological
hydrogeological framework
frameworkthat thatemploys
employssatellite
satellite
remotelysensed
remotely senseddatadatatotodescribe
describethe thesource
source ofof lithium
lithium andanditsits movement
movement in the
in the environ-
environment.
ment.
To To be the
be noted, noted, the derivation
derivation of hydrogeological
of hydrogeological frameworks frameworks with ground
with ground data is adata commonis a
common
and and oldan
old practice, practice,
example an being
example thebeing
studythe studywhich
in [22], in [22], which the
defined defined the hydraulic
hydraulic character
character
and and subsurface
subsurface distribution distribution
of the major of the major aquifers
aquifers and aquitards,
and aquitards, including including theirof
their areas
areas of recharge
recharge and discharge
and discharge and the rate and andthe rate and the direction
the direction of groundwater
of groundwater movement movement
for a test
for aintest
site site in United
Nevada, Nevada, United States.
States.
The paper is
The is structured
structuredasasfollows:
follows:Section
Section 2 describes
2 describes the the
test test
site and
site presents
and presentsthe
datadata
the usedused
in the study
in the alongside
study alongside its processing.
its processing.Section 3 presents
Section the geological
3 presents pro-
the geological
cessing andand
processing related
related workflow,
workflow, andand Sections
Sections 4 and
4 and 5 present
5 presentthe thehydrogeology
hydrogeologyand andmass
mass
balance workflows,
balance workflows,respectively.
respectively.Finally,Finally,Section
Section 6 discusses
6 discusses thethe generic
generic lithium
lithium move-
movement
ment framework
framework and traces
and traces the investigation
the investigation conclusions.
conclusions.

Remote Sens. 2022, 14, 1383
of elements such as lithium, sodium, potassium, magnesium, and boron. The largest of
these salt lakes is the Salar de Uyuni, with an area of about 9600 km2 and an elevation of
approximately 3600 m, at the southwestern corner of Bolivia (Figure 2). The Salar de
Uyuni is extremely flat, with altitude variations of less than a meter, such that it is the
reference test site to calibrate altimetric satellite missions [23,24]. The surrounding water-
shed covers an area of approximately 55,000 km2, rising to elevations of 6550 m ASL. Lith- 4 of 22

ium “mining” at the Salar de Uyuni is operated by the Bolivian state lithium concern,
Yacimientos de Litio Bolivianos (YLB), which reports a resource estimate of 21 million
tonnes and annual
2. Materials production of lithium carbonate of 250 tonnes in 2020 [25].
and Methods
2.1. Six
Testgeo-environmental
Site criteria or processes are required for salar formation (Figure
3) [4]. Firstly, there need to
Salars are essentially inlandbe lithium-bearing source
salt flats that form in rocks in the catchment
arid conditions, for the
and notable basin:
examples
southwest Bolivia is dominated by Cenozoic volcanics, including lithium-enriched
are those that are found in the Lithium Triangle in the central Andes of Bolivia, Chile, and lithol-
ogies. Secondly,
Argentina. Theythere needs to
are created by betheanatural
closed weathering
drainage basin: the with
of rocks Bolivian Altiplano
elevated levelsisof
flanked
elements by such
high asmountain
lithium, ranges,
sodium,the Eastern and
potassium, the Western
magnesium, andCordillera
boron. The (Figure
largest2).of
Thirdly,
these saltthelakes
climate
is theneeds
Salartodebe Uyuni,
arid: in with
southwest
an area Bolivia,
of about evaporation
9600 km2 and exceeds precipita-of
an elevation
tion
approximately 3600 m, at the southwestern corner of Bolivia (Figure 2). The Salar de grow
by about 500 mm. Fourthly, there needs to be accumulation space as the basins Uyuni
and the sediments
is extremely are deposited:
flat, with regional of
altitude variations tectonics
less thanensures
a meter,that thisthat
such is the
it iscase. Fifthly,
the reference
there need
test site toto be sediments
calibrate to fill
altimetric these missions
satellite basins: thick sandThe
[23,24]. andsurrounding
gravels provide both acovers
watershed host
toanthe brines
area and also contribute
of approximately 55,000 to 2 , rising
kmtheir formation. Finally,
to elevations of these
6550 m conditions
ASL. Lithium need“mining”
to per-
sist
at for
the sufficient time (thousands
Salar de Uyuni is operated to by
millions of years)
the Bolivian to allow
state lithiumtheconcern,
accumulation and con-
Yacimientos de
centrate the brines,
Litio Bolivianos and this
(YLB), is very
which much
reports the case also
a resource givenofthat
estimate in Bolivia
21 million thereand
tonnes are annual
more
than 40 salars.
production of lithium carbonate of 250 tonnes in 2020 [25].

Figure
Figure2.2.Location
Location(a)
(a)and
andextension
extension of
of the
the area of
of interest
interest(AoI)
(AoI)over
overthe
theSalar
SalardedeUyuni
Uyuniinin Bolivia
Bolivia (b).
(b). Basemap images copyright © 1995–2020
Basemap images copyright © 1995–2020 Esri. Esri.

Six geo-environmental criteria or processes are required for salar formation

(Figure 3) [4]. Firstly, there need to be lithium-bearing source rocks in the catchment
for the basin: southwest Bolivia is dominated by Cenozoic volcanics, including lithium-
enriched lithologies. Secondly, there needs to be a closed drainage basin: the Bolivian
Altiplano is flanked by high mountain ranges, the Eastern and the Western Cordillera
(Figure 2). Thirdly, the climate needs to be arid: in southwest Bolivia, evaporation exceeds
precipitation by about 500 mm. Fourthly, there needs to be accumulation space as the
basins grow and the sediments are deposited: regional tectonics ensures that this is the case.
Fifthly, there need to be sediments to fill these basins: thick sand and gravels provide both
a host to the brines and also contribute to their formation. Finally, these conditions need to
persist for sufficient time (thousands to millions of years) to allow the accumulation and
concentrate the brines, and this is very much the case also given that in Bolivia there are
more than 40 salars.
Remote
RemoteSens. 2022,14,
Sens.2022, 14,1383
x FOR PEER REVIEW 55 of
of 22
21

Figure3.3.Schematic
Figure Schematic illustration
illustration of
of the
thesalar
salarand
andcatchment,
catchment,showing
showingpotential
potentialsources
sourcesof oflithium
lithium
(weathering of volcanic bedrock and derived sediments) in the upper part of the drainage basin,
(weathering of volcanic bedrock and derived sediments) in the upper part of the drainage basin,
coinciding with areas of higher precipitation; pathways for surface water and groundwater flow
coinciding with areas of higher precipitation; pathways for surface water and groundwater flow that
that transports lithium from source areas to the salar (blue lines); and the salar where lower precip-
transports
itation and lithium
higherfrom source areas
evaporation resultto in
thethe
salar (blue lines); of
concentration and the salarwater
inflowing wheretolower precipitation
create lithium-en-
and higher evaporation result in the concentration of inflowing water to create lithium-enriched
riched brines. Other potential sources of lithium exist, including direct input from hydrothermal
brines.
sourcesOther potential sources of lithium exist, including direct input from hydrothermal sources
(red lines).
(red lines).
Previous studies on the source of lithium showed how high concentrations of lithium
Previous
are present atstudies onend
the south the source of lithium
of the salar, showed
near the mouthhow high
of the concentrations
Rio of lithium
Grande de Lipez, where
are present at the south end of the salar, near the mouth of the Rio Grande de
the current mining operations and ponds are located [26]. Based on geochemical studies Lipez, where
the
fromcurrent
salarsmining
within operations and ponds
similar geological are located
settings, [26]. Based
the authors assumeon that
geochemical studies
major sources of
from salars
lithium arewithin similar
the recent geological
volcanic rocks,settings, therhyolitic
including authors assume that (pyroclastic
ignimbrites major sources of
rocks
lithium are the recent volcanic rocks, including rhyolitic ignimbrites (pyroclastic rocks
made up from pumice, rock fragments, volcanic ash, and glass shards), and derived un-
made up from pumice, rock fragments, volcanic ash, and glass shards), and derived uncon-
consolidated sediments prevalent in the drainage area of the Rio Grande [27–30]. Accord-
solidated sediments prevalent in the drainage area of the Rio Grande [27–30]. According to
ing to these findings, we selected a study area covering the southern catchment area of
these findings, we selected a study area covering the southern catchment area of the salar,
the salar, marked in yellow in Figure 2b.
marked in yellow in Figure 2b.
2.2.Remote
2.2. RemoteSensing
SensingData
DataandandInitial
InitialBaseline
Baseline
The framework
The framework presented
presented in in this
this study
study isis making
making use use of
ofremote
remotesensing
sensingdata
dataand
and
auxiliary information taken from existing literature. The data sources
auxiliary information taken from existing literature. The data sources and related work- and related work-
flow(geology,
flow (geology,hydrogeology,
hydrogeology,and andmass
massbalance)
balance)arearepresented
presentedin inTable
Table1.1.Optical
Opticalremote
remote
sensing data from the ASTER [8] and Landsat-8 [31] missions are
sensing data from the ASTER [8] and Landsat-8 [31] missions are employed for the geo- employed for the geo-
logicalprocessing.
logical processing. A A digital
digital elevation
elevation model
model from
from thethe SRTM
SRTMmission
mission[32][32]isisused
usedfor
forthe
the
derivationof
derivation ofthe
theriver
rivernetwork
networkand andthethegroundwater
groundwatermodelling,
modelling,in inconjunction
conjunctionwithwithWWF
WWF
HydroSHEDSdata
HydroSHEDS data[33].
[33].AAglobal
globalland
landcover
covermapmapfrom
fromthe theClimate
ClimateChange
ChangeInitiative
Initiative[34]
[34]
is employed for the recharge modelling, alongside meteorological
is employed for the recharge modelling, alongside meteorological data from the ERA5 data from the ERA5 re-
analysis model [35]. Digital geological map data at a scale of 1:500 000
re-analysis model [35]. Digital geological map data at a scale of 1:500,000 were obtained were obtained from
the United
from States
the United Geological
States GeologicalSurvey, generalised
Survey, generalisedfromfromprimary
primary1:250,000 scale
1:250,000 mapping
scale map-
by geologists
ping fromfrom
by geologists the Servicio Geolgico
the Servicio de Bolivia
Geolgico (GEOBOL)
de Bolivia (GEOBOL)[36]. [36].
The digital data data
The digital were
complemented
were complemented by the byuse
theofuse
a subset of theof
of a subset printed 1:250,000
the printed scale GEOBOL
1:250,000 scale GEOBOLmaps, maps,
which
include
which a morea detailed
include classification
more detailed of the geological
classification succession
of the geological and geological
succession profiles
and geological
showing the vertical relationships and thicknesses [37]. Finally, several
profiles showing the vertical relationships and thicknesses [37]. Finally, several parameters parameters neces-
sary to properly
necessary develop
to properly the workflows
develop the workflowsare derived from BGS
are derived fromdatabases and theand
BGS databases existing
the
literature
existing [38–40]. [38–40].
literature To be noted,
To beall the input
noted, products
all the and parameters
input products are openare
and parameters access
open in
access
order in
to order
preserveto preserve the sustainability
the sustainability of theframework.
of the overall overall framework.
 Remote Sens. 2022, 14, x FOR PEER REVIEW 6 of 21
Remote Sens. 2022, 14, 1383 6 of 22

Table 1. Datasets used for the study (G = geology, H = hydrogeology, MB = mass balance).
Table 1. Datasets used for the study (G = geology, H = hydrogeology, MB = mass balance).
Dataset Workflow
Dataset Workflow
ASTER G
ASTER Landsat-8 G G
Landsat-8 G
SRTM G
SRTM G
USGS 1:500 k geological
USGS 1:500 k geological map map G, MB MB
G,
GEOBOL 1:250 k geological maps maps
GEOBOL 1:250 k geological G G
ERA5 ERA5 MB MB
CCI Land
CCICover
Land Cover MB MB
WWF WWFHydroSHEDS
HydroSHEDS MB MB
Parameters from available literature (hydrogeology,
Parameters from available literature (hydrogeology, river H, MB
river flow, Li concentration) H, MB
flow, Li concentration)

To
Tocreate
createaaseamless
seamless mosaic
mosaic over the whole
over the whole study
studyarea,
area,ASTER
ASTERscenes
sceneshave
have been
been
composited together within the Google Earth Engine (GEE) platform.
composited together within the Google Earth Engine (GEE) platform. ASTER SWIR bands ASTER SWIR bands
malfunction
malfunctionfrom from April
April 2008;
2008; therefore, combinationsofofimages
therefore, combinations imagesbetween
between 2000
2000 andand 2007
2007
have been used. Figure 4 illustrates the challenges in mosaicking ASTER
have been used. Figure 4 illustrates the challenges in mosaicking ASTER scenes covering scenes covering
the
thewhole
wholestudystudy area.
area. In
In this study, 2000–2001
this study, 2000–2001 AprilApriltotoSeptember
Septembermedian medianimagery
imagery has
has
been used for the short-wave Infrared (SWIR) bands (Figure 4e) given
been used for the short-wave Infrared (SWIR) bands (Figure 4e) given the least amount of the least amount
ofdistortions.
distortions. The The combined
combined distortions
distortions fromfrom
clouds clouds and scene-related
and other other scene-related
issues areissues
mostare
most clearly
clearly illustrated
illustrated by theby the alternating
alternating light and light
darkand dark
bands bands
across theacross
west ofthethewest
image ofinthe
image
Figurein4a.
Figure 4a. The individual
The individual scene edge scene edge
effects areeffects
most are most in
obvious obvious in the
the black blackofpatches
patches Fig-
ofure
Figure
4b,c. 4b,c.
AlthoughAlthough more subtle,
more subtle, the mean thecomposite
mean composite
image inimage
Figurein 4dFigure
contains4d signifi-
contains
significantly
cantly moremore distortions
distortions compared
compared to Figure
to Figure 4e. the
4e. For Forthermal
the thermal infrared
infrared (TIR)(TIR)
bands, bands,
a
a composite
compositeofofthe themean
mean reflectance
reflectance of winter
of winter months
months (May(May to Sep)
to Sep) fromfrom
2003 2003
to 2007 to of-
2007
offered themost
fered the mostsuitable
suitabledataset
dataset with
with thethe least
least amount
amount of distortions
of distortions in the
in the TIR.TIR.

Figure 4. Different approaches to compositing ASTER scenes: (a) 2000–2006 May–September mean;
Figure 4. Different approaches to compositing ASTER scenes: (a) 2000–2006 May–September mean;
(b) 2000–2006 May–September mosaic; (c) 2000–2006 May–September median; (d) 2000–2001 April–
(b) 2000–2006 May–September mosaic; (c) 2000–2006 May–September median; (d) 2000–2001 April–
September mean, scene footprints visible; (e) 2000–2001 April–September median.
September mean, scene footprints visible; (e) 2000–2001 April–September median.
Minerals exhibit unique absorption and reflection properties in the infrared region of
Minerals exhibit unique absorption and reflection properties in the infrared region
the electromagnetic (EM) spectrum that allow them to be identified, similar to human fin-
of the electromagnetic (EM) spectrum that allow them to be identified, similar to human
gerprints. Whilst iron minerals show characteristic spectral features in the very-near part
fingerprints. Whilst iron minerals show characteristic spectral features in the very-near part
of the infrared (VNIR), the Al-OH and Fe/Mg-OH bonds of clay minerals show pro-
of the infrared (VNIR), the Al-OH and Fe/Mg-OH bonds of clay minerals show pronounced
absorption features in the short-wave infrared. One of the techniques for highlighting the
relevant spectral features is through band ratios. Specifically, the Abrams ratio is one of the
most widely utilised band ratio for geological mapping in the VNIR–SWIR [41–44]. Using
 Abrams ratio composition in Equation (1). Due to the broad carbonate absorption band at
B14 the Mafic Index is sensitive to carbonates, so an adjusted Mafic Index (MIc) has been
developed to correct this effect [7]. Figure 5b displays the TIR indices over the study area.
∗
𝑄𝐼 𝐶𝐼 𝑀𝐼 𝑀𝐼𝑐 (2)
∗ ³
Remote Sens. 2022, 14, 1383 7 of 22
Finally, the spectral information in the study area is analysed through comparison
on a ternary diagram as illustrated in Figure 6. The study area is divided into square pol-
ygons of 500 m by 500 m; these polygons are then used to extract the pixel statistics of the
specific ASTER band naming, the Abrams ratio can be expressed as in Equation (1), where
overlapping EO data
the three channels (SWIR and
represent TIR green,
the red, imagesand
described in Figure
blue image 5) for each
composition. polygon.
To be Then,
noted, the
pixel
three channels can be seen as proxies for clay, iron oxide, and vegetation. Figure 5a showster-
statistics within the polygons are computed and analysed on a ternary plot. The
nary plot created
the Abrams ratio is
forthrough
the studythearea.
open source library “plotly” in python. This is an interac-
tive tool, which allows hovering over the  a point of interest
 on the graph and understand-
ing which polygon it is through the B04 B03
shapefile ID B02its spectral position on the ternary
and
AI ( R, G, B) = , , (1)
plot. B07 B04 B01

Figure 5. Spectral index maps developed over the study area. (a) SWIR Abrams ratio. RGB image
channels are clay (R), iron oxide (G), and vegetation (B). (b) TIR index combination. RGB image
channels are Quartz Index (QI) (R), Carbonate Index (CI) (G), and Mafic Index corrected (MIcor) (B).

The longer wavelengths in the thermal infrared (TIR) are used for the detection of
quartz (SiO2 ) and carbonate minerals. Several other indices have been developed in the
literature. In this study, we also make use of the Quartz Index (QI), Carbonate Index (CI),
and Mafic Index (MI) [7] and combine them in a colour composite image similarly to the
Abrams ratio composition in Equation (1). Due to the broad carbonate absorption band at
B14 the Mafic Index is sensitive to carbonates, so an adjusted Mafic Index (MIc) has been
developed to correct this effect [7]. Figure 5b displays the TIR indices over the study area.

B11 ∗ B11 B13 B2 MI

QI = CI = MI = MIc = (2)
B10 ∗ B12 B14 B13 CI 3
Finally, the spectral information in the study area is analysed through comparison
on a ternary diagram as illustrated in Figure 6. The study area is divided into square
polygons of 500 m by 500 m; these polygons are then used to extract the pixel statistics of
the overlapping EO data (SWIR and TIR images described in Figure 5) for each polygon.
Then, pixel statistics within the polygons are computed and analysed on a ternary plot.
The ternary plot created is through the open source library “plotly” in python. This is
an interactive tool, which allows hovering over the a point of interest on the graph and
Remote Sens. 2022, 14, x FOR PEER REVIEW 8 of 21
Remote Sens. 2022, 14, 1383 8 of 22

Figure 5. Spectral index maps developed over the study area. (a) SWIR Abrams ratio. RGB image
channels are claywhich
understanding (R), iron oxide (G),
polygon it is and vegetation
through (B). (b) TIR
the shapefile index
ID and its combination. RGB on
spectral position image
the
channels are Quartz
ternary plot. Index (QI) (R), Carbonate Index (CI) (G), and Mafic Index corrected (MIcor) (B).

Figure 6. Polygon spectra of “IRG” units on ternary plot. Spectra from Abrams ratio image with
3 bands6.representing
Figure clay, of
Polygon spectra iron oxide,
“IRG” andon
units vegetation.
ternary plot. Spectra from Abrams ratio image with 3
bands representing clay, iron oxide, and vegetation.
3. Geological Processing
The first Processing
3. Geological stage of the technical framework is the derivation of a conceptual 3D geo-
logical model. Determining the three-dimensionality of the geology is important, as the
The first stage of the technical framework is the derivation of a conceptual 3D geo-
fluid flow and liberation of lithium into the groundwater system takes place both at the
logical model. Determining the three-dimensionality of the geology is important, as the
ground surface and in the subsurface. Hence, capturing the spatial relationships between
fluid flow and liberation of lithium into the groundwater system takes place both at the
the geological units and with the topography allows the vertical ordering and thicknesses
ground surface and in the subsurface. Hence, capturing the spatial relationships between
of individual units to be inferred of the geology. The model provides a framework that
the geological units and with the topography allows the vertical ordering and thicknesses
can be used to consider the evolution and movement of lithium-bearing water from source
of individual units to be inferred of the geology. The model provides a framework that
to salar. This was achieved through the integration and interpretation of complementary
can be used to consider the evolution and movement of lithium-bearing water from source
input datasets including the following: existing digital geological map data and accompa-
to salar. This was achieved through the integration and interpretation of complementary
nying written explanations, printed geological map data, digital elevation data, and related
input
studies datasets
on the including the following:
local and regional geology existing
[45,46].digital geological mapworkflow
The corresponding data andisaccompa-
depicted
nying written explanations, printed geological map
in Figure 7. Optical remote sensing data from ASTER and Landsat-8 are used data, digital elevation data,toand re-
derive
lated
indications of mineral assemblages and underlying lithology as described in Section 2.2. is
studies on the local and regional geology [45,46]. The corresponding workflow A
depicted in Figure 7.interpretation
manual geological Optical remote of sensing data from
the EO imagery andASTER
derived andproducts
Landsat-8 and areusing
usedtheto
derive indicationsmap
input geological of mineral assemblages
as a starting point is and underlying
carried lithology
out to derive as described
a refined geological in map.
Sec-
tion
Geological interpretation in this context uses the spectral and textural information inand
2.2. A manual geological interpretation of the EO imagery and derived products the
using
imagery the and
input geological
DEMs and the map as a starting
geologist’s point isof
knowledge carried out to derive
the geological a refined
succession geo-
to draw
logical
and refinemap. Geological
polygons interpretation
around areas containing in thisthecontext
same uses the spectral
lithology and textural
[7]. Spectral infor-
characteristics,
mation in the imagery and DEMs and the geologist’s knowledge
as identified in imagery discussed in Section 2, allow areas to be grouped together or of the geological succes-
sion to draw
separated and on
based refine polygons
certain around areas
characteristics. containing
Ternary the same
plots (Figure lithology
6) are [7]. Spectral
a valuable tool for
characteristics,
understanding as identified
such in imagery
similarities; in these, discussed
we plot ainpixel’s
Sectionposition
2, allowaccording
areas to be to grouped
the three
together or separated
bands. Therefore, based on certain
a predominantly red characteristics.
area in Figure 5a Ternary plots (Figure
has a higher 6) are aofvalu-
concentration clay
able
(red) than iron oxide (green) or vegetation (blue), so it would plot close to theaccording
tool for understanding such similarities; in these, we plot a pixel’s position top apex
to
in the
thethree
ternarybands.
plot Therefore,
(Figure 6).aIn predominantly
this way we are redable
areatoinuse
Figure 5a has aplots
the ternary higher to concen-
identify
tration
the areas of clay (red) thansediment
of superficial iron oxide (green)
which have or a
vegetation (blue), signature
similar spectral so it would toplot close to
the lavas or
the top apex in
ignimbrites; theisternary
that plotpixel
to say, the (Figure
over6).theInsediment
this way will
we areplotable
in atosimilar
use theposition
ternaryon plots
the
to identify
ternary plotthe
toareas of superficial
that from sediment
the ignimbrite. which
Those with have a similar
a similar spectral
signature tosignature to the
the ignimbrites
lavas or ignimbrites;
are thought that is derived
to be primarily to say, the from pixel
the over the sediment
ignimbrites and those willwith
plotain a similar
similar posi-
signature
tion
to theonlavas
the ternary plot totothat
are thought from thefrom
be derived ignimbrite.
the lavas. Those
These with a similar
three signature
components allowto the
the
ignimbrites
assessment are thought
of the to bedepositional
geological primarily derived order (as from the ignimbrites
described below) whichand those
bringswith a
to the
final enhanced
similar signature geological
to the lavasmapare that coversto
thought thebeentire study
derived fromareatheand whose
lavas. attributes
These three com-split
mappedallow
ponents ignimbrites accordingoftothe
the assessment their relativedepositional
geological age and subdivides
order (assuperficial
describedgeologybelow)
Remote Sens. 2022, 14, x FOR PEER REVIEW 9 of 21

Remote Sens. 2022, 14, 1383 9 of 22

which brings to the final enhanced geological map that covers the entire study area and
whose attributes split mapped ignimbrites according to their relative age and subdivides
superficial geology based on parent material and hence likely lithium concentrations. In
based onthe
parallel, parent
rivermaterial
networkand hence likely
is derived throughlithium concentrations.
geomorphological In parallel,
analysis the river
of the available
network is derived through geomorphological analysis of the available elevation
elevation model [7]. This framework has been exploited for the current test case as de-model [7].
This framework has been
scribed in the following. exploited for the current test case as described in the following.

Figure 7.
Figure 7. Geological
Geological processing
processing workflow.
workflow.

The drainage
drainage basin
basinforforthe
theSalar
SalardedeUyuni
Uyuniextends
extendsbeyondbeyond the coverage
the coverage of of
thethe
existing
exist-
geological
ing geologicalmapmap data. Therefore,
data. in the
Therefore, firstfirst
in the instance,
instance,it isitnecessary
is necessaryto extend
to extend thethe
existing
exist-
1:500,000
ing 1:500,000scalescale
geological
geologicalmapmapto ensure
to ensurecomprehensive
comprehensive coverage
coverage of the
of thedrainage
drainage basin.
ba-
Specifically, for the hydrological workflow described in the next section,
sin. Specifically, for the hydrological workflow described in the next section, it is im- it is important to
know all potential sources of lithium that might be mobilised and
portant to know all potential sources of lithium that might be mobilised and deposited in deposited in the salar.
This is accomplished
the salar. via visualvia
This is accomplished interpretation of a combination
visual interpretation of the satellite
of a combination of theimagery
satellite
and derived
imagery and elevation models. models.
derived elevation Spectral Spectral
responses and textural
responses featuresfeatures
and textural in the visible
in the and
vis-
near-infrared parts of the
ible and near-infrared electromagnetic
parts spectrum spectrum
of the electromagnetic allow lithological units to beunits
allow lithological identified
to be
on false colour
identified composites
on false of the Landsat-8
colour composites and ASTER
of the Landsat-8 andimagery (Figures (Figures
ASTER imagery 4, 5 and 4,8a).
5,
Existing polygons of the USGS geological map, based on
and 8a). Existing polygons of the USGS geological map, based on primary surveys primary surveys that predate
that
the widespread
predate availability
the widespread of EO data,
availability of EO aredata,
thenareextended using the
then extended imagery
using as a guide.
the imagery as a
New
guide. New polygons are created where necessary, and consistent attribution is across
polygons are created where necessary, and consistent attribution is maintained main-
the existing
tained acrossandthe new polygons.
existing and new Whilst carrying
polygons. Whilstout carrying
this remote out geological
this remotemappinggeological it
becomes apparent that some of the existing polygons in the USGS
mapping it becomes apparent that some of the existing polygons in the USGS dataset dataset could be modified
considering the newconsidering
could be modified satellite imagery
the new andsatellite
conceptual 3D geological
imagery understanding
and conceptual of the
3D geological
project geologists. In these cases, boundaries are adjusted accordingly
understanding of the project geologists. In these cases, boundaries are adjusted accord- to create the best
geological dataset possible, without visiting the field.
ingly to create the best geological dataset possible, without visiting the field.
Understanding
Understanding the the geological
geologicalsuccession
successionisisimportant
importanttotoestimate
estimate thethe relative
relative ages
ages of
of the rock and their spatial relationships, or their relative vertical order or cross-cutting
the rock and their spatial relationships, or their relative vertical order or cross-cutting re-
relationship. This can then be linked to the phases of magmatic activity and evolution of
lationship. This can then be linked to the phases of magmatic activity and evolution of the
the catchment geology, including the distribution of different rock types, which in turn
catchment geology, including the distribution of different rock types, which in turn can
can be linked to the potential amount of lithium in the magmatic system [47]. In the
be linked to the potential amount of lithium in the magmatic system [47]. In the existing
existing geological data, some of the ignimbrites are grouped into a single age. From
geological data, some of the ignimbrites are grouped into a single age. From the relation-
the relationships observed in the imagery, it is possible to separate the ignimbrites into
ships observed in the imagery, it is possible to separate the ignimbrites into different rel-
different relative ages depending on their geometrical relationship to the other ignimbrites
ative ages depending on their geometrical relationship to the other ignimbrites and vol-
and volcanic deposits, including lavas, related to stratovolcanic activity. Figure 8 shows the
canic deposits, including lavas, related to stratovolcanic activity. Figure 8 shows the pub-
published geological map (Figure 8b) and a Landsat-8 image (Figure 8a) over a portion of
lished geological map (Figure 8b) and a Landsat-8 image (Figure 8a) over a portion of a
a study area to demonstrate the geological processing workflow. The mapped polygons
study area to demonstrate the geological processing workflow. The mapped polygons in-
include two orange polygons representing the same ignimbrite deposit separated by a
central brown polygon representing lava flows. Examination of the Landsat-8 image and
Abrams ASTER ratio (as described in Section 2.2, Figure 8c) shows that the ignimbrites
on to the east appear different in both composition and texture to those in the west. We
Remote Sens. 2022, 14, x FOR PEER REVIEW 10 of 21

clude two orange polygons representing the same ignimbrite deposit separated by a cen-
Remote Sens. 2022, 14, 1383 tral brown polygon representing lava flows. Examination of the Landsat-8 image 10 of 22
and
Abrams ASTER ratio (as described in Section 2.2, Figure 8c) shows that the ignimbrites on
to the east appear different in both composition and texture to those in the west. We can
alsoalso
can see that the lava
see that flowsflows
the lava over over
the ignimbrites to theto
the ignimbrites east;
thetherefore, the eastern
east; therefore, ignim-
the eastern
ignimbrites
brites were were in place
in place beforebefore the lava
the lava was extruded
was extruded and and are therefore
are therefore older.
older. However,
However, the
the ignimbrites
ignimbrites in west
in the the west
appear appear to overlap
to overlap and beandonbetoponoftop
the of the lavas;
lavas; therefore,
therefore, the
the ignim-
ignimbrites
brites to theto the are
west west are younger
younger than thethan theand
lava lavaasand as cannot
such such cannot
be thebe the same
same unit asunit as
those
those in the east. Using this manual interpretation, the geological map
in the east. Using this manual interpretation, the geological map was enhanced via the was enhanced via
the addition
addition andand updating
updating of of polygon
polygon attributestotoclearly
attributes clearlydefine
definethe
the ignimbrites;
ignimbrites; this was
was
completed
completed over
over the
the area
area of
of interest
interestand
andpassed
passedtotothe
thefollowing
followingworkflows.
workflows.

Figure 8.
Figure 8. Geological
Geological interpretation
interpretation aided
aided with
with EO
EO data
data over
over aa portion
portion of
of the
the study
study area.
area. The
The general
general
temporal relationship is indicated by the legend (Qs, youngest, to Qtig, oldest; see
temporal relationship is indicated by the legend (Qs, youngest, to Qtig, oldest; see Appendix Appendix A for
A
details); however, the geological interpretation shows that multiple instances and ages of Qtig (ig-
for details); however, the geological interpretation shows that multiple instances and ages of Qtig
nimbrite) occur, including relatively young ignimbrite deposits (western part of the map) and rela-
(ignimbrite) occur, including relatively young ignimbrite deposits (western part of the map) and
tively old ignimbrite deposits (eastern part of the image), separated by an intervening unit of stra-
relatively
tovolcanoold ignimbrite
deposits. deposits (eastern
(a) Landsat-8 part of
RGB colour the image),
composite separated by shaded;
topographically an intervening unit of
(b) geological
stratovolcano
map; (c) Abramsdeposits.
ASTER(a)ratio.
Landsat-8 RGB colour composite topographically shaded; (b) geological
map; (c) Abrams ASTER ratio.
4. Hydrogeological Processing
4. Hydrogeological Processing
The second stage of the technical framework is the development of a workflow
The second stage of the technical framework is the development of a workflow
whereby studied ranges of aquifer properties that control groundwater flow and storage,
whereby studied ranges of aquifer properties that control groundwater flow and storage,
such as porosity, specific yield, intrinsic and field scale, permeability, hydraulic conduc-
such as porosity, specific yield, intrinsic and field scale, permeability, hydraulic conduc-
tivity, and hence, transmissivity, are constrained. These are attributed to the geological
tivity, and hence, transmissivity, are constrained. These are attributed to the geological
units for
units for the
the formations
formations encountered
encountered in in the
the study
study area
area in
in order
order to
to better
better understand
understand how how
mineralised water will move. In each case, the data can be refined through
mineralised water will move. In each case, the data can be refined through ground truthingground truth-
ing these
these properties
properties by and/or
by field field and/or drilltesting
drill core core testing either
either on on the
the salar salar
itself or itself
on theormargins
on the
margins or elsewhere in the watershed, and the attributed geology model
or elsewhere in the watershed, and the attributed geology model can then form a hydro- can then form
a hydrogeological attribution model or map. By better understanding the
geological attribution model or map. By better understanding the aquifer properties of aquifer proper-
tiesformations,
the of the formations,
one canone can investigate
investigate the hydrochemical
the hydrochemical processes processes
that can that
leadcan lead to
to making
deposits of particular dissolved elements become valuable resources. Figure 9 shows the
relative processing framework. To demonstrate how the workflow can be completed, the
initial hydrogeological parameterisation has been derived from BGS data holdings and
 drogeological units (HUs). These enable the spatial distribution of the geology either at
outcrop for 2D maps or in terms of volumes (3D models) to be attributed. The table in
Appendix A illustrates how the 16 lithologies are parametrised based on experience de-
rived from working on other salars and literature values [48]. To be noted, the Geological
Description in the table is derived from geo-nsdi.er.usgs.gov/metadata/open-file/95-
Remote Sens. 2022, 14, 1383 11 of 22
494/metadata.html (accessed 01 February 2022). In particular, the geology is parametrised
using 16 hydrogeological units (HUs; see Appendix A). Each geological unit is given a
hydraulic conductivity along with a porosity value. Both are based on experience or back-
existing
ground literature,
knowledgeassupplemented
performed forby theliterature
current test caseand
values [38–40]. Thisas
are used was refined
base as the
variables to
modelling proceeded to keep the relative values, but the absolute values were adjusted
feed into the modelling process. These values are then passed to the next and final frame- to
ensure
work. credible modelling results.

Figure9.9.Hydrogeological
Figure Hydrogeologicalprocessing
processingframework.
framework.

ThisBalance
5. Mass framework has been exploited for the current test case. The mapping between the
Derivation
geological units and their hydrogeological properties is undertaken by identifying hydroge-
The final stage of the technical framework has the aim of quantifying the amount of
ological units (HUs). These enable the spatial distribution of the geology either at outcrop
lithium reaching the salar and how long it takes to accumulate. The two methods by which
for 2D maps or in terms of volumes (3D models) to be attributed. The table in Appendix A
lithium is transported through the watershed are surface water and groundwater. The
illustrates how the 16 lithologies are parametrised based on experience derived from work-
twoon
ing are interrelated,
other salars andwith groundwater
literature valuessupplying
[48]. To besurface
noted, water in the dryDescription
the Geological season. There-
in
fore, on a long-term basis (~100,000 years [28]), it is necessary to consider
the table is derived from geo-nsdi.er.usgs.gov/metadata/open-file/95-494/metadata.html how groundwa-
ter moves1 through
(accessed Februarythe system
2022). and interacts
In particular, with theisrock
the geology mass to leach
parametrised usinglithium. To un-
16 hydrogeo-
derstand the groundwater component, a groundwater flow and particle
logical units (HUs; see Appendix A). Each geological unit is given a hydraulic conductivity tracking model
of the Uyuni watershed have been developed. These are built on
along with a porosity value. Both are based on experience or background knowledge the hydrogeological un-
derstanding of the catchment which in turn is built on the geological characterisation.
supplemented by literature values and are used as base variables to feed into the modelling The
framework
process. is shown
These in Figure
values are 10, andtoitthe
then passed is built on afinal
next and sequence of processing steps as fol-
framework.
lows: recharge calculation, e.g., ZOODRM [49]; groundwater flow and velocities, e.g.,
5.MODFLOW
Mass Balance [50]; and particle tracking, e.g., MODPATH [51]. They take as input several
Derivation
EO products
The final stage of in
derived theprevious stages.
technical framework has the aim of quantifying the amount of
lithium reaching the salar and how long it takes to accumulate. The two methods by which
lithium is transported through the watershed are surface water and groundwater. The two
are interrelated, with groundwater supplying surface water in the dry season. Therefore,
on a long-term basis (~100,000 years [28]), it is necessary to consider how groundwater
moves through the system and interacts with the rock mass to leach lithium. To understand
the groundwater component, a groundwater flow and particle tracking model of the Uyuni
watershed have been developed. These are built on the hydrogeological understanding of
the catchment which in turn is built on the geological characterisation. The framework is
shown in Figure 10, and it is built on a sequence of processing steps as follows: recharge
calculation, e.g., ZOODRM [49]; groundwater flow and velocities, e.g., MODFLOW [50];
and particle tracking, e.g., MODPATH [51]. They take as input several EO products derived
in previous stages.
Fluxes from groundwater and surface water masses are computed with the output
of the modelling and ground data, respectively. They are required to compute the final
mass accumulation in the salt lake. This framework has been exploited for the current
test case as described in the following. It is to be noted that the ZOODRM model has not
been applied in this case due to the availability of gridded rainfall for Bolivia providing a
suitable first guestimate of recharge. It is thought that this method of estimating recharge
is commensurate with the complexity of the model and confidence of the other data
sources. Once the groundwater model has been developed further and ground-truthed
Remote Sens. 2022, 14, 1383 12 of 22

Remote Sens. 2022, 14, x FOR PEER REVIEW suitable data such as groundwater heads and baseflow in the rivers, then a12more
against of 21
sophisticated approach to calculating recharge can be adopted.

Figure10.
Figure 10.Mass
Massbalance
balanceframework.
framework.

5.1. MODFLOW
Fluxes fromModelgroundwater and surface water masses are computed with the output of
A steady-state
the modelling groundwater
and ground flow model They
data, respectively. is developed using
are required tothe USGSthe
compute MODFLOW
final mass
code [50]. The in
accumulation model takes
the salt boundary
lake. conditions
This framework hasand
been includes
exploited transmissivity
for the current (saturated
test case
thickness
as described multiplied by hydraulic
in the following. conductivity),
It is to be noted with distributed
that the ZOODRM recharge
modelproviding
has not beenthe
inflow and outflows based on the rivers and evaporation from the salar itself.
applied in this case due to the availability of gridded rainfall for Bolivia providing a suit- The ground-
water model
able first and particle
guestimate tracking model
of recharge. are constructed
It is thought that this using
method a model grid of 1recharge
of estimating km in theis
Xcommensurate
and Y directions. The groundwater flow model simulates the movement
with the complexity of the model and confidence of the other data sources. of water in the
subsurface and is based on
Once the groundwater undertaking
model has been a mass balanced
developed for each
further andofground-truthed
the grid cells based on
against
Darcy’s
suitablelaw.dataThe
suchparticle tracking model
as groundwater heads uses
andmathematical
baseflow in representations
the rivers, then of “packets”
a more of
sophis-
water, the particles, to trace the path of groundwater
ticated approach to calculating recharge can be adopted. through the system. Alongside this, it
enables the time of travel to be calculated, which enables the residence time of the waters
in theMODFLOW
5.1. system to be estimated.
Model
Data inputs are detailed as follows:
A steady-state groundwater flow model is developed using the USGS MODFLOW
•codeTransmissivity
[50]. The model based on boundary
takes a constant conditions
100 m thickness combined
and includes with hydraulic
transmissivity conduc-
(saturated
tivity distribution;
thickness multiplied by hydraulic conductivity), with distributed recharge providing the
•inflowRecharge set to 10%
and outflows based[52]
onofthe
long-term
rivers and average rainfallfrom
evaporation [40];the salar itself. The ground-
•water Rivers—drain cells based
model and particle on river
tracking network
model are derived fromusing
constructed national datasetgrid
a model (Worldof 1Bank);
km in
•the XEvaporation based on salar outline.
and Y directions. The groundwater flow model simulates the movement of water in
These model
the subsurface andfeatures
is based areonoutlined
undertakingin Figure
a mass11a,b, and the
balanced forresulting
each of the steady-state
grid cells
groundwater
based on Darcy’s heads areThe
law. provided
particleintracking
Figure 12. The uses
model groundwater
mathematicalheads, which are fluid
representations of
pressures
“packets”expressed
of water, as theheights
particles,combined
to trace with datum
the path of levels, show that
groundwater the salars
through in the
the system.
system
Alongside act as lowitpoints
this, enables attracting
the timegroundwater
of travel to be outflow. The distribution
calculated, which enables of groundwater
the residence
heads then approximates the topography,
time of the waters in the system to be estimated. with higher heads in the hinterland surrounding
the river valleys and the salars. The
Data inputs are detailed as follows: initial set of parameters were adjusted so as to ensure
that the distribution of groundwater heads was sensible, i.e., not above the ground surface.
· Transmissivity based on a constant 100 m thickness combined with hydraulic con-
Further calibration can be undertaken once observed groundwater heads are available.
ductivity distribution;
Groundwater flow occurs to the rivers in the base of the valleys and then onto the Salar de
· Recharge set to 10% [52] of long-term average rainfall [40];
Uyuni, reaching it via the main river channel Rio Grande de Lipez.
· Rivers—drain cells based on river network derived from national dataset (World
Bank);
· Evaporation based on salar outline.
These model features are outlined in Figure 11a,b, and the resulting steady-state
groundwater heads are provided in Figure 12. The groundwater heads, which are fluid
pressures expressed as heights combined with datum levels, show that the salars in the
 system act as low points attracting groundwater outflow. The distribution of groundwater
heads then approximates the topography, with higher heads in the hinterland surround-
ing the river valleys and the salars. The initial set of parameters were adjusted so as to
ensure that the distribution of groundwater heads was sensible, i.e., not above the ground
surface. Further calibration can be undertaken once observed groundwater heads are
Remote Sens. 2022, 14, 1383
available. Groundwater flow occurs to the rivers in the base of the valleys and then13onto
of 22

the Salar de Uyuni, reaching it via the main river channel Rio Grande de Lipez.

Remote Sens. 2022, 14, x FOR PEER REVIEW 14 of 21

Figure 11. MODFLOW
MODFLOW modelling.
modelling. (a)
(a) Model
Model setup
setup with
with HUs
HUs (see
(see Appendix
Appendix A),
A), model
model extent, and
rivers; (b) long-term average rainfall.

Figure 12.
Figure 12. MODFLOW
MODFLOW modelling
modelling output.
output. Steady-state groundwater
groundwater heads.
heads.

5.2.
5.2. MODPATH
MODPATHModelModel
The MODPATH
The MODPATH model model(Version
(Version7.2.001)
7.2.001)was
was developed
developed by by inheriting
inheriting most
most of itsofdata
its
data
fromfrom the MODFLOW
the MODFLOW modelmodel described
described above.above. The extra
The main main dataset
extra dataset was porosity
was porosity based
based on the zonation from the geology distribution (see Appendix A). The particle tracks
on the zonation from the geology distribution (see Appendix A). The particle tracks were
produced for the southern part of the Uyuni watershed developed for the prototype work-
flow and are shown in Figure 13a. They mainly enter the river system or find their way to
the salar itself. The latter is a low point in the system from a groundwater head perspective
(see Figure 12) and so will attract any particles that are not captured by the river network.
Remote Sens. 2022, 14, 1383 14 of 22

were produced for the southern part of the Uyuni watershed developed for the prototype
workflow and are shown in Figure 13a. They mainly enter the river system or find their
way to the salar itself. The latter is a low point in the system from a groundwater head
perspective (see Figure 12) and so will attract any particles that are not captured by the river
network. The time distribution of the particles (longest in the system first) is illustrated
in Figure 13b. The maximum time of travel is of the order of 1 Ma, consistent with times
observed in other salar watersheds, e.g., Atacama [38]. However, the shorter-lived particles
Remote Sens. 2022, 14, x FOR PEER REVIEW
(<10,000 a), shown in light blue in Figure 11b, migrate to the rivers where they can15flowof 21
to
the salar via the Rio Grande de Lipez delta [39].

Figure
Figure13.
13.MODPATH
MODPATHand andmass
massbalance
balanceoutputs:
outputs:(a)(a)particle
particletracks (104 4years
tracks(10 yearsand
andyounger
youngerare
are
highlighted
highlightedininlight
lightblue)
blue)and
and(b)
(b)time
timeofoftravel
traveldistribution
distributionfor
forarea
areaof
ofinterest.
interest.

5.3.Mass
5.3. MassBalance
Balance
Thefinal
The final part
part of
of the
the workflow
workflowisisthe thecalculation
calculationofof thethemass
mass balance to help
balance to helpunderstand
under-
how the lithium may have accumulated in the salar. This is undertaken
stand how the lithium may have accumulated in the salar. This is undertaken by a com- by a combination
of determining
bination the inflow
of determining theofinflow
lithium mass along
of lithium mass with
alongthewith
timethegroundwater takes to takes
time groundwater arrive
at the salar. The typical concentration of lithium in the main river
to arrive at the salar. The typical concentration of lithium in the main river feeding the feeding the salar, Rio
Grande de Lipez, is of the order of 1–5 mg/L. Whilst this is low compared
salar, Rio Grande de Lipez, is of the order of 1–5 mg/L. Whilst this is low compared to the to the maximum
lithium concentration
maximum in the salarinofthe
lithium concentration 5–10 g/L,
salar ofit5–10
is relatively
g/L, it ishigh for surface
relatively high water.
for surface
water. Lithium concentrations in the rock mass within the Uyuni watershed are of the order
16–65 mg/L concentrations
Lithium (see Table 4 in in [53]). The groundwaters
the rock mass within the can thenwatershed
Uyuni leach lithiumare ofandtheprovide
order
input to the river system before it flows into the salar itself in the delta of Rio Grande de
16–65 mg/L (see Table 4 in [53]). The groundwaters can then leach lithium and provide
Lipez. Previous calculations shown by Risacher and Fritz [39] demonstrate that streamflow
input to the river system before it flows into the salar itself in the delta of Rio Grande de
inputs estimated at 2 cumecs and lithium concentrations of 3 mg/L can “fill” the salar
Lipez. Previous calculations shown by Risacher and Fritz [39] demonstrate that stream-
every 100,000 years. This is a relatively short timescale compared to the time to fill other
flow inputs estimated at 2 cumecs and lithium concentrations of 3 mg/L can “fill” the salar
salars with lithium, e.g., 1.9 Ma for Atacama [38]. However, the particle tracks demonstrate
every 100,000 years. This is a relatively short timescale compared to the time to fill other
that in the time of travel (~104 years) sufficient lithium can be leached from the rock mass
salars with lithium, e.g., 1.9 Ma for Atacama [38]. However, the particle tracks demon-
to provide the likely lithium concentrations in groundwaters of the order of 10 mg/L.
strate that in the time of travel (~104 years) sufficient lithium can be leached from the rock
Given that the particles mainly go to the river system, the concentration observed in the
mass to provide the likely lithium concentrations in groundwaters of the order of 10 mg/L.
Rio Grande de Lipez can be produced by leaching from rocks in the watershed. Therefore,
Given that the particles mainly go to the river system, the concentration observed in the
the workflow enables the current working assumption of a relatively short time to fill
Rio
theGrande de Lipez
salar with lithiumcantobe
beproduced by leaching
realistic. However, from rocks in
considerable the watershed.
uncertainty remains Therefore,
both on
the
theworkflow
processesenables
as theythe current
exist now working assumption
and the dynamic natureof a of
relatively
the systemshortwith
timechanges
to fill theto
salar
rockwith
masslithium to be realistic.
and climate. The former However, considerable
can only be addressed uncertainty
with more remains both such
field data on theas
processes as they exist now and the dynamic nature of the system
groundwater heads and river flows. The latter requires an understanding of geological with changes to rock
mass and climate. The former can only be addressed with more field
processes which relies on dating approaches such as geochemical analysis. Nonetheless, data such as ground-
water heads and river flows. The latter requires an understanding of geological processes
which relies on dating approaches such as geochemical analysis. Nonetheless, the flexibil-
ity and repeatability of the workflow enable the calculation to be revisited once more data
and understanding become available.
Remote Sens. 2022, 14, 1383 15 of 22

the flexibility and repeatability of the workflow enable the calculation to be revisited once
more data and understanding become available.

6. Conclusions
In this study, we have developed a complete workflow to characterise the movement
of lithium from the watershed into the salt lake. Figure 14 shows the overall conceptual
framework, where the groundwater flow and particle tracking models integrate the data
workflows. The detailed processing frameworks are shown in Figures 7, 9 and 10, which
run sequentially and where the interconnection between frameworks is illustrated by em-
ploying the “intermediate output” block which represents an output from a previously run
framework. At their hearts is the use of remote sensing data to provide the underpinning
datasets. Firstly, remote sensing data feed into the geological modelling (Section 3), which
ensures that the geology is as accurate as possible. The extents of geological units are encap-
sulated in a shapefile which can be ingested in the groundwater flow model. This shapefile
is parameterised with hydrogeological properties (Section 4) based on data previously col-
lected on related studies in the region and for similar geological settings. The groundwater
flow model is relatively data-hungry (Section 5) and amongst other data requires input from
a recharge model. The recharge model, which is itself data-intense, relies on rainfall and
potential evaporation data, along with land use, DEM, and the drainage or river network.
The extent of the models is defined by the basin boundary. The outflow from the recharge
model is fed into the groundwater flow model which along with the parametrised geology
and the river network can be used to set up the groundwater flow model. The output from
the groundwater flow model is used to drive the particle tracking model in conjunction
with the porosity distribution, which is again derived from the geology. Appendix B details
the data requirements to build the framework. The models are then used to determine
both surface and groundwater flow paths in the basin and how water–rock interaction can
allow the waters to pick up lithium and then bring it into the salar. Whilst lithium has
accumulated in the salar over the long timescales, this is difficult to simulate as both the
rock mass and the climate vary over this time period. For example, the region is subject
to wetter and drier periods on the cyclicity of 10,000 years [39]. Further, the groundwater
inflows to the salar may not be recharged contemporaneously; rather, they may have been
built up during wetter periods and are only now finding their way to the salar [52]. This is
borne out by the timescales predicted by the modelling, albeit with uncertainties associated
with simulating processes over such long timescales.
The workflow is flexible and can be used with different model codes as well as simpler
approaches, such as distributed rainfall rather than a fully featured recharge model. Whilst
the workflow presented here is an example based on open data sources, measurements
can and should be used to validate the approach. However, given the extent of open
data sources available, the workflow demonstrates how far it is possible to go with freely
available data. Given the open approach, uncertainty can be addressed in a number of
different ways. Firstly, choosing and swapping out different components of the workflow
can be used to assess how differing data sources affect the outcome. Secondly, the ease
by which the process can be reproduced can enable sensitivity analysis to be undertaken.
Finally, the availability of the workflow enables independent assessment and scrutiny of
the process and its outcome. The outputs from this study have been benchmarked against
previous work undertaken in the Uyuni watershed as well as other studies in salars in the
Lithium Triangle countries, e.g., Atacama.
Remote Sens.
Remote 2022,
Sens. 14,14,
2022, 1383
x FOR PEER REVIEW 17 of1621of 22

Figure 14. Conceptual framework.

Figure 14. Conceptual framework.
Author Contributions: “Conceptualization, C.R., L.B, M.B, A.B, J.F., A.H; methodology, C.R., L.B,
Author Contributions:
M.B, A.B, Conceptualization,
J.F., A.H; software, M.B and A.H.; C.R., L.B., M.B.,
validation, A.B.,
C.R., L.B, J.F. A.B,
M.B, and J.F.,
A.H.; methodology,
A.H; formal analy-C.R.,
L.B.,
sis, M.B., A.B.,M.B,
C.R., L.B, J.F. and
J.F., A.H.;
A.H; software, M.B.C.R.,
investigation, and A.H.; validation,
L.B, M.B, C.R.,
A.B, J.F., A.H;L.B.,
dataM.B., A.B., J.F.
curation, C.R.,and A.H.;
L.B,
formal analysis,
M.B, J.F., A.H;; C.R., L.B., M.B., J.F.
writing—original andpreparation,
draft A.H.; investigation,
C.R., L.B,C.R.,
M.B,L.B.,
J.F., M.B., A.B., J.F. and A.H.;
A.H; writing—review anddata
curation, C.R.,L.B,
editing, C.R., L.B.,M.B,
M.B., J.F.A.H;
J.F., andvisualization,
A.H.; writing—original draft
C.R., L.B, M.B, J.F.,preparation,
A.H; projectC.R., L.B., M.B., C.R.
administration, J.F. and
A.H.; writing—review
and A.H.; and editing,
funding acquisition, C.R. C.R., L.B., All
and A.H. M.B., J.F. and
authors A.H.;
have visualization,
read and agreed to C.R.,
the L.B., M.B., J.F.
published
version
and A.H.;ofproject
the manuscript.”
administration, C.R. and A.H.; funding acquisition, C.R. and A.H. All authors have
read and agreed to
Funding: This research the published
was funded version of the manuscript.
by InnovateUK grant number 134012
Funding: ThisReview
Institutional researchBoard
was funded by InnovateUK
Statement: grant number 134012.
Not applicable
Institutional Review
Informed Consent Board Statement:
Statement: Not applicable.
Not applicable
Informed Statement: Not
Consent Statement:
Data Availability applicable.
Materials to support the workflow development are found here:
github.com/HughesAG/LiBol_Workflow
Data Availability Statement: Materials to support the workflow development are found here: github.
Acknowledgments: This work has (accessed
com/HughesAG/LiBol_Workflow been madeon possible by InnovateUK
1 February 2022). through the grant “Use of
innovative techniques to ensure Li brine supply for the low-cost battery market” No. 134012. Peter
Schweitzer of USGS This
Acknowledgments: work has
is thanked been made
for assistance possible
with by InnovateUK
the digital through
geological map thethe
data of grant “Use of
Bolivian
innovative
Altiplano. techniques to ensure
Bateson, Butcher, Ford,Liand
brine supply
Hughes for thewith
publish low-cost battery market”
the permission No. 134012.
of the Executive Peter
Direc-
Schweitzer
tor, British of USGS is Survey
Geological thanked for assistance
(UKRI). withwould
The authors the digital
like togeological
acknowledgemapthe
data of the Bolivian
anonymous re-
Altiplano.
viewers forBateson, Butcher, Ford,
their comments whichandhaveHughes publish
improved with the permission of the Executive Director,
the paper.
British Geological Survey (UKRI). The authors would like to acknowledge the anonymous reviewers
Conflicts of Interest: The authors declare no conflict of interest.
for their comments which have improved the paper.
Appendix
Conflicts A. Hydrogeological
of Interest: Units Used
The authors declare for the
no conflict of Groundwater
interest. Flow Model Develop-
ment
Remote Sens. 2022, 14, 1383 17 of 22

Appendix A. Hydrogeological Units Used for the Groundwater Flow Model Development

Geological Unit Hydrogeological Hydraulic Conductivity

Unit ID Parent Rock Type Geological Definition Porosity (%)
Code Code (m/d)
Pzs: Sedimentary rocks (Palaeozoic) Chiefly marine sandstone
1 Pzs Pzs Sedimentary rocks and shale of Devonian to Ordovician age. Rocks are generally 0.1 20
highly folded and locally penetratively deformed.
Ks: Sedimentary rocks (Cretaceous) Marine and nonmarine
2, 3 Ks FRSC Sedimentary rocks 0.5 20
sandstone, shale, marl, and limestone.
Ts1: Sedimentary rocks (Oligocene to Paleocene) Nonmarine,
4 Ts1 FRSC Sedimentary rocks mostly reddish coloured conglomerate, sandstone, shale, and 0.5 20
mudstone.
Tig: Pyroclastic rocks (Miocene and Oligocene) Chiefly
5 Tig CLIC Pyroclastic rocks welded to nonwelded ash-flow tuffs, but includes air-fall tuffs 0.5 25
and thin, volcaniclastic beds. Mostly dacitic in composition.
Tdp: Gypsum diapirs (Miocene to Eocene?) May include
6 Tdp FRSC Gypsum diapirs 0.5 1
halite and other evaporite minerals.
Tvnd: Volcanic rocks, undifferentiated (Miocene and
Oligocene) Chiefly lava flows, but includes extensive
pyroclastic deposits and intrusive rocks in some areas where
7 Tvnd CLIC Volcanic rocks not mapped separately as units Tig and Ti, and locally may 5 25
include interbedded nonmarine sedimentry rocks. Mostly of
andesitic and dacitic composition; sources are poorly defined
volcanic eruptive centres, now deeply eroded.
8 Tmf LAIF Pyroclastic rocks Tmf: Los Frailes and Morococala Ignimbrites (Miocene). 0.5 25
Ti: Intrusive rocks (Pliocene to Oligocene) Chiefly subvolcanic
9 Ti CLIC Intrusive rocks stocks, plugs, and dikes of dacitic composition in vent 0.5 3
complex of eroded volcanic eruptive centres.
Ts2: Sedimentary rocks (Pliocene to Oligocene) Nonmarine
10 Ts2 FRSC Sedimentary rocks 0.5 20
sandstone, conglomerate, shale, marl, and evaporites.
Remote Sens. 2022, 14, 1383 18 of 22

Geological Unit Hydrogeological Hydraulic Conductivity

Unit ID Parent Rock Type Geological Definition Porosity (%)
Code Code (m/d)
QTs: Sedimentary rocks (Pleistocene and Pliocene)
11 QTs CCS Sedimentary rocks Nonmarine sandstone, conglomerate, and shale. May include 5 25
minor interlayered volcanic rocks.
QTig: Ignimbrite (Pleistocene to Miocene) Welded and
12 Qtig IRG Pyroclastic rocks nonwelded ash-flow tuffs, chiefly in extensive outflow sheets. 5 25
Mostly of dacitic composition.
QTev: Stratovolcano deposits (Holocene to Miocene) Lava
13 Qtev FEV Volcanic rocks flows, flow breccias, lahars, and minor pyroclastic deposits 0.5 10
chiefly of andesitic to dacitic composition.
Ql: Lacustrine deposits (Holocene and Pleistocene) Chiefly
14 Ql LL L Lacustrine sediments calcareous tufa in ancient lake shorelines and lacustrine mud 5 30
and silt deposits.
Qsu: Surficial deposits, undifferentiated (Holocene and
Unconsolidated Pleistocene) Includes unconsolidated alluvial, aeolian,
15 Qsu Qsu 5 30
sediments colluvial, and glacial deposits. Locally may include lacustrine
and salt deposits that are not shown separately.
Qs: Salt deposits (Holocene and Pleistocene) Salar /
16 Qs Qs Evaporites playa-lake evaporites. May include interbedded fine-grained 20 20
lacustrine deposits.
Remote Sens. 2022, 14, 1383
Appendix B. Data Requirements for the Groundwater Flow Model Development
Materials to support the workflow development are found here: github.com/
HughesAG/LiBol_Workflow (accessed on 1 February 2022).
Task Activity
Develop understanding of
occurrence of groundwater
Conceptual model of
flow along with rock mass
groundwater flow and solute
that contributes Li and other
transport
solutes to the inflow to the
salar at the watershed scale
Water balance: surface and
groundwater
Recharge model (incl. runoff)
Riverflow data
Abstraction: surface water
and groundwater
Springs
Inflow to the salars
Groundwater flow model
Recharge
Abstraction
River baseflow
Springflow
Groundwater head time series
Geometry of aquifer units
19 of 22
Data Requirements Issues
-3D geological understanding
-Distribution of parameters:
transmissivity, storage coefficients,
porosity, and exchange coefficients for
leaching Li from host
Need to establish
rocks-Groundwater flowpaths
depth of groundwater
(piezometric surface)
flow in watershed
-Groundwater hydrographs (30 years)
-Groundwater geochemistry to inform
-Output: 3D conceptual
understanding of groundwater flow
at the watershed scale
Static data:
-DEM—25 m resolution (ASCII
gridded)
-River network (shapefile)
-Land cover map—vector/1 km
gridded
-Soil map—vector/1 km gridded
-Geology map—vector/1 km gridded
Driving data:
-Rainfall—1 km gridded/daily
(30 years)
-Potential evaporation—1 km
gridded/monthly (30 years)
-Temperature—1 km gridded/daily
(30 years))
Output: gridded monthly recharge
values/gridded monthly runoff/time
series river flows
Daily riverflow (30 years)
Monthly actual abstraction (30 years)
Location (X, Y) and outflows
(30 years)
Calculated output
Provide by recharge model
Collated for water balance
Calculated from riverflow data Calibration data
Collated for water balance Calibration data
Collated for conceptual modelling Calibration data
Top and base for each unit
(gridded ASCII)
Remote Sens. 2022, 14, 1383 20 of 22

Task Activity Data Requirements Issues

Derived from conceptual modelling
Hydraulic conductivity
(gridded ASCII)
Derived from conceptual modelling
Storage coefficient
(gridded ASCII)
Model output at required times
Groundwater surface(s)
(gridded ASCII)
Groundwater
pathways/particle tracking
modelling
Derived from conceptual modelling
Porosity distribution
(gridded ASCII)
Driving data Same as for groundwater flow model
Groundwater flow
Model output
pathways/particle tracks
Lithium mass balance
Derived from particle tracking
Groundwater flow paths
modelling
Distribution of Derived from conceptual modelling
lithium-bearing rocks (gridded ASCII)
Exchange coefficients to
Literature review/derived from
determine water–rock
conceptual modelling
interaction
Derived from combining
Lithium mass arriving at salar groundwater path lines and
water–rock interaction

References
1. Hund, K.; La Porta, D.; Thao, P.; Tim, L.; John, D. Minerals for Climate Action: The Mineral Intensity of the Clean Energy Transition;
World Bank Group Report; World Bank: Washington, DC, USA, 2020.
2. USGS. Lithium: Mineral Commodity Summary; USGS: Reston, VA, USA, 2021.
3. USGS. Lithium: Mineral Commodity Summary; USGS: Reston, VA, USA, 2011.
4. Munk, L.A.; Hynek, S.A.; Bradley, D.C.; Boutt, D.; Labay, K.; Jochens, H. Lithium Brines: A Global Perspective. Rev. Econ. Geol.
2016, 18, 339–365.
5. Flexer, V.; Celso Fernando, B.; Claudia, I. Lithium recovery from brines: A vital raw material for green energies with a potential
environmental impact in its mining and processing. Sci. Total Environ. 2018, 639, 1188–1204. [CrossRef] [PubMed]
6. Barandiarán, J. Lithium and development imaginaries in Chile, Argentina and Bolivia. World Dev. 2019, 113, 381–391. [CrossRef]
7. Gupta, R. Remote Sensing Geology; Springer: Berlin/Heidleberg, Germany, 2017.
8. Yamaguchi, Y.; Kahle, A.; Tsu, H.; Kawakami, T.; Pniel, M. Overview of advanced spaceborne thermal emission and reflection
radiometer (ASTER). EEE Trans. Geosci. Remote Sens. 1998, 36, 1062–1071. [CrossRef]
9. Marghany, M.; Mazlan, H. Lineament mapping using multispectral remote sensing satellite data. Int. J. Phys. Sci. 2010, 5,
1501–1507. [CrossRef]
10. Masoud, A.; Katsuaki, K. Auto-detection and integration of tectonically significant lineaments from SRTM DEM and remotely-
sensed geophysical data. ISPRS J. Photogramm. Remote Sens. 2011, 66, 818–832. [CrossRef]
11. Rajan, G.; Sundararajan, M. Mapping of mineral resources and lithological units: A review of remote sensing techniques. Int. J.
Image Data Fusion 2019, 10, 79–106. [CrossRef]
12. Pour, A.; Hashim, M.; Park, Y.; Hong, J.K. Mapping alteration mineral zones and lithological units in Antarctic regions using
spectral bands of ASTER remote sensing data. Geocarto Int. 2018, 33, 1281–1306. [CrossRef]
13. Sabins, F. Remote sensing for mineral exploration. Ore Geol. Rev. 1999, 14, 157–183. [CrossRef]
14. Van der Meer, F.; van Ruitenbeek, F.J.A.; van der Werff, H.; Noomen, M.F.; van der Meijde, M.; Hecker, C.A.; Carranza, E.J.M.; de
Smeth, B.; Woldai, T. Multi-and hyperspectral geologic remote sensing: A review. Int. J. Appl. Earth Obs. Geoinf. 2012, 14, 112–128.
[CrossRef]
15. Waters, P.; Greenbaum, D.; Smart, P.L.; Osmaston, H. Applications of remote sensing to groundwater hydrology. Remote Sens. Rev.
1990, 4, 223–264. [CrossRef]
Remote Sens. 2022, 14, 1383 21 of 22

16. Singhal, B.; Ravi, P. Applied Hydrogeology of Fractured Rocks; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2010.
17. Cardoso-Fernandes, J.; Teodoro, A.C.; Lima, A.; Perrotta, M.; Roda-Robles, E. Detecting Lithium (Li) mineralizations from space:
Current research and future perspectives. Appl. Sci. 2020, 10, 1785. [CrossRef]
18. Cardoso-Fernandes, J.; Teodoro, A.C.; Lima, A. Remote sensing data in lithium (Li) exploration: A new approach for the detection
of Li-bearing pegmatites. Int. J. Appl. Earth Obs. Geoinf. 2019, 76, 10–25. [CrossRef]
19. Santos, D.; Teodoro, A.; Lima, A.; Cardoso-Fernandes, J. Remote Sensing Techniques to Detect Areas with Potential for Lithium
Exploration in Minas Gerais, Brazil. In Proceedings of the SPIE, SPIE Remote Sensing, Strasbourg, France, 9–12 September 2019.
20. Gemusse, U.; Lima, A.; Teodoro, A. Comparing Different Techniques of Satellite Imagery Classification to Mineral Mapping
Pegmatite of Muiane and Naipa: Mozambique. In Proceedings of the SPIE, SPIE Remote Sensing, Strasbourg, France, 9–12
September 2019.
21. Rossi, C.; Spittle, S.; Bayaraa, M.; Pandey, A.; Henry, N. An Earth Observation Framework for the Lithium Exploration. In
Proceedings of the IGARSS 2018—2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 22–27
July 2018.
22. Winograd, I.; William, T. Hydrogeologic and Hydrochemical Framework, South-Central Great Basin, Nevada-California, with Special
Reference to the Nevada Test Site; USGS: Washington, DC, USA, 1975.
23. Fricker, H.A.; Minster, B.; Carabajal, C.; Quinn, K.; Bills, B.; Borsa, A. Assessment of ICESat performance at the salar de Uyuni,
Bolivia. Geophys. Res. Lett. 2005, 32. [CrossRef]
24. Rossi, C.; Rodriguez Gonzales, F.; Fritz, T.; Yague-Martinez, N.; Eineder, M. TanDEM-X calibrated raw DEM generation. ISPRS J.
Photogramm. Remote Sens. 2012, 73, 12–20. [CrossRef]
25. Brown, T.J.; Idoine, N.E.; Wrighton, C.E.; Raycraft, E.R.; Hobbs, S.F.; Shaw, R.A.; Everett, P.; Deady, E.A.; Kresse, C. World Mineral
Production 2015–2019, BGS. 2019. Available online: https://www2.bgs.ac.uk/mineralsuk/download/world_statistics/2010s/
WMP_2015_2019.pdf (accessed on 1 February 2022).
26. Rettig, S.L.; Jones, B.F.; François, R. Geochemical evolution of brines in the Salar of Uyuni, Bolivia. Chem. Geol. 1980, 30, 57–79.
[CrossRef]
27. Risacher, F.; Hugo, A.; Carlos, S. The origin of brines and salts in Chilean salars: A hydrochemical review. Earth-Sci. Rev. 2003, 63,
249–293. [CrossRef]
28. Risacher, F.; Fritz, B. Origin of Salts and Brine Evolution of Bolivian and Chilean Salars. Aquat. Geochem. 2008, 15, 123–157.
[CrossRef]
29. Hofstra, A.H.; Todorov, T.I.; Mercer, C.N.; Adams, D.T.; Marsh, E.E. Silicate melt inclusion evidence for extreme pre-eruptive
enrichment and post-eruptive depletion of lithium in silicic volcanic rocks of the western United States: Implications for the
origin of lithium-rich brines. Econ. Geol. 2003, 105, 1691–1701. [CrossRef]
30. Godfrey, L.; Álvarez-Amado, F. Volcanic and saline lithium inputs to the Salar de Atacama. Minerals 2020, 10, 201. [CrossRef]
31. Roy, D.P.; Wulder, M.A.; Loveland, T.R.; Woodcock, C.E.; Allen, R.G.; Anderson, M.C.; Helder, D.; Irons, J.R.; Johnson, D.M.;
Kennedy, R.; et al. Landsat-8: Science and product vision for terrestrial global change research. Remote Sens. Environ. 2014, 145,
154–172. [CrossRef]
32. Van Zyl, J.J. The Shuttle Radar Topography Mission (SRTM): A breakthrough in remote sensing of topography. Acta Astronaut.
2001, 48, 559–565. [CrossRef]
33. Wickel, B.A.; Lehner, B.; Sindorf, N. HydroSHEDS: A Global Comprehensive Hydrographic Dataset. In AGU Fall Meeting
Abstracts; Center for Astrophysics: Cambridge, MA, USA, 2007; Volume 2007.
34. Li, W.; MacBean, N.; Ciais, P.; Defourny, P.; Lamarche, C.; Bontemps, S.; Houghton, R.A.; Peng, S. Gross and net land cover
changes in the main plant functional types derived from the annual ESA CCI land cover maps (1992–2015). Earth Syst. Sci. Data
2018, 10, 219–234. [CrossRef]
35. Hoffmann, L.; Günther. G.; Li, D.; Stein, O.; Wu, X.; Griessbach, S.; Heng, Y.; Konopka, P.; Muller, R.; Vogel, B.; et al. From
ERA-Interim to ERA5: The considerable impact of ECMWF’s next-generation reanalysis on Lagrangian transport simulations.
Atmos. Chem. Phys. 2019, 19, 3097–3124. [CrossRef]
36. Sherman, P.; Richter, D.H.; Ludington, S.; Soria-Escalante, E.; Escobar-Diaz, A. Digital Geologic Map of the Altiplano and Cordillera
Occidental, Bolivia, United States Geological Survey; Open-File Report 95–494; U.S. Department of the Interior: Washington, DC,
USA, 1995.
37. Departamento Nacional de Geologia (DNG). Sheet 6234 Rio Mulatos; Departamento Nacional de Geologia, Ministerio de Minas y
Petroleo: La Paz, Bolivia, 1962.
38. Munk, L.A.; Boutt, D.F.; Hynek, S.A.; Moran, B.J. Hydrogeochemical fluxes and processes contributing to the formation of
lithium-enriched brines in a hyper-arid continental basin. Chem. Geol. 2018, 493, 37–57. [CrossRef]
39. Risacher, F.; Fritz, B. Geochemistry of Bolivian salars, Lipez, southern Altiplano: Origin of solutes and brine evolution. Geochim.
Cosmochim. Acta 1991, 55, 687–705. [CrossRef]
40. VVicente-Serrano, S.; Kenawy, A.; Azorin-Molina, C.; Chura, O.; Trujillo, F.; Aguilar, E.; Martín, N.; Lopez-Moreno, I.;
Sanchez-Lorenzo, A.; Morán-Tejeda, E.; et al. Average monthly and annual climate maps for Bolivia. J. Maps 2015, 12, 295–310.
[CrossRef]
41. Abrams, M.J.; Rothery, D.A.; Pontual, A. Mapping in the Oman ophiolite using enhanced Landsat Thematic Mapper images.
Tectonophysics 1988, 151, 387–401. [CrossRef]
Remote Sens. 2022, 14, 1383 22 of 22

42. Abrams, M.; Yamaguchi, Y. Twenty years of ASTER contributions to lithologic mapping and mineral exploration. Remote Sens.
2019, 11, 1394. [CrossRef]
43. Amer, R.; Kusky, T.; Ghulam, A. Lithological mapping in the Central Eastern Desert of Egypt using ASTER data. J. Afr. Earth Sci.
2010, 56, 75–82. [CrossRef]
44. Mia, B.; Fujimitsu, Y. Mapping hydrothermal altered mineral deposits using Landsat 7 ETM+ image in and around Kuju volcano,
Kyushu, Japan. J. Earth Syst. Sci. 2012, 121, 1049–1057. [CrossRef]
45. Baker, M.C.W. The nature and distribution of upper Cenozoic ignimbrite centres in the Central Andes. J. Volcanol. Geotherm. Res.
1981, 11, 293–315. [CrossRef]
46. Salisbury, M.; Jicha, B.R.; de Silva, S.L.; Singer, B.S.; Jimenez, N.C.; Ort, M.H. 40Ar/39Ar chronostratigraphy of Altiplano-Puna
volcanic complex ignimbrites reveals the development of a major Magmatic Province. Geol. Soc. Am. Bull. 2011, 123, 821–840.
[CrossRef]
47. Ort, M.H.; De Silva, S.L.; Jiménez, C.N.; Jicha, B.R.; Singer, B.S. Correlation of ignimbrites using characteristic remanent
magnetization and anisotropy of magnetic susceptibility, Central Andes, Bolivia. Geochem. Geophys. Geosystems 2013, 14, 141–157.
[CrossRef]
48. Houston, J.; Butcher, A.; Ehren, P.; Evans, K.; Godfrey, L. The evaluation of brine prospects and the requirement for modifications
to filing standards. Econ. Geol. 2011, 106, 1225–1239. [CrossRef]
49. Mansour, M.M.; Wang, L.; Whiteman, M.; Hughes, A.G. Estimation of spatially distributed groundwater potential recharge for
the United Kingdom. Q. J. Eng. Geol. Hydrogeol. 2018, 51, 247–263. [CrossRef]
50. Harbaugh, A.W. MODFLOW-2005, the US Geological Survey Modular Ground-Water Model: The Ground-Water Flow Process;
US Department of the Interior, US Geological Survey: Reston, VA, USA, 2005; pp. 6–16.
51. Pollock, D.W. User Guide for MODPATH Version 6: A Particle Tracking Model for MODFLOW; US Department of the Interior,
US Geological Survey: Reston, VA, USA, 2012; p. 58.
52. Boutt, D.F.; Corenthal, L.G.; Moran, B.J.; Munk, L.; Hynek, S.A. Imbalance in the modern hydrologic budget of topographic
catchments along the western slope of the Andes (21–25◦ S): Implications for groundwater recharge assessment. Hydrogeol. J.
2021, 29, 985–1007. [CrossRef]
53. Davis, J.R.; Howard, K.A.; Rettig, S.L.; Smith, R.L.; Ericksen, G.E.; Risacher, F.; Morales, H.A.R. Progress Report on Lithium-Related
Geologic Investigations in Bolivia; USGS: Washington, DC, USA, 1982.