Next Article in Journal
Using Ground-Penetrating Radar and Deep Learning to Rapidly Detect Voids and Rebar Defects in Linings
Next Article in Special Issue
Acute Ecotoxicity Potential of Untreated Tannery Wastewater Release in Arequipa, Southern Peru
Previous Article in Journal
A Study on the Mechanism of Digital Technology’s Impact on the Green Transformation of Enterprises: Based on the Theory of Planned Behavior Approach
Previous Article in Special Issue
Optimization of a Textile Effluent Treatment System and Evaluation of the Feasibility to Be Reused as Influents in Textile Dyeing Processes
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 15
10.3390/su151511856
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

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

Tracking Sediment Provenance Applying a Linear Mixing Model Approach Using R’s FingerPro Package, in the Mining-Influenced Ocoña Watershed, Southern Peru

by
Jorge Crespo
1,2,*,
Elizabeth Holley
1,
Madeleine Guillen
3,
Ivan Lizaga
4,
Sergio Ticona
3,
Isaac Simon
5,
Pablo A. Garcia-Chevesich
6,7 and
Gisella Martínez
3
1
Department of Mining Engineering, Colorado School of Mines, Golden, CO 80401, USA
2
Nevada Bureau of Mining and Geology, University of Nevada Reno, Reno, NV 89509, USA
3
Department of Geology and Geophysics, Universidad Nacional de San Agustín de Arequipa, Av. Independencia and Paucarpata Street s/n, Arequipa 04001, Peru
4
Isotope Bioscience Laboratory—ISOFYS, Department of Green Chemistry and Technology, Ghent University, Coupure Links 653, 9000 Gent, Belgium
5
Department of Geology and Geological Engineering, Colorado School of Mines, Golden, CO 80401, USA
6
Department of Civil and Environmental Engineering, Colorado School of Mines, Golden, CO 80401, USA
7
Intergubernmental Hydrological Programme, UNESCO, Montevideo 11200, Uruguay
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(15), 11856; https://doi.org/10.3390/su151511856
Submission received: 25 January 2023 / Revised: 9 July 2023 / Accepted: 31 July 2023 / Published: 1 August 2023
(This article belongs to the Special Issue Sustainability of Arid Lands in Southern Peru)
Browse Figures
Review Reports Versions Notes

Abstract

:
Stream sediments record water–rock interactions in tributaries followed by fluid mixing in larger downstream catchments, but it can be difficult to determine the relative contributions of each tributary. A good way to analyze this problem is sediment fingerprinting, which allows researchers to identify the source of sediments within a basin and to estimate the contribution of each source to the watershed. Herein, we developed a workflow using the frequentist model FingerPro v1.3 to quantify the sediment source contribution in a semiarid watershed. We applied an unmixing model algorithm to an ICP-MS geochemical database containing information on 32 elements in 362 stream sediment samples. By modeling the source contributions to these mixed samples, we infer that the main sediment contribution comes from the upper portion of the catchment (61–70%), followed by the middle (21–29%) and lower (8–10%) parts, with geochemical anomalies (As and Cu) being closely related to mining sites. Results from this study can be helpful for future management decisions to ensure a better environment in this semiarid watershed.

1. Introduction

The Ocoña watershed is one of the most important drainage areas in southern Peru. Land use in the watershed includes mining, agriculture, industry, and urban areas. All of these activities can impact the geochemical properties of soil or water, either due to inappropriate watershed management practices or by geochemical characteristics of rock–water interaction that produce erosion and sedimentation. One of the problems affecting the watershed is rapid population growth, especially in places where industrial activities, such as artisanal mining, have increased, leading to the deterioration of natural resources and the generation of environmental liabilities (i.e., tailings and waste rock). In addition, the basin is affected by geodynamic processes, such as seismic activity, mass movements, mudflows, etc., [1] that increase sediment transport. Previous studies of the Ocoña watershed have concluded that artisanal mining is a point-source of both noble (Au) and polluting metals (As, Hg), which can be easily transported to the river through its tributaries [2,3]. The gold recovery processes occur in areas close to water sources, using mercury to obtain gold through amalgamation. These sites are located mostly in the middle portion of the watershed, whereas the majority of the agricultural and urban activities are located in lower areas. All these activities can generate substantial amounts of sediments that end up in riverbeds, e.g., [4].
Sediment dynamics provide important information about the distribution of different pollutants, such as heavy metals in a drainage path [5,6,7], which can strongly affect, for example, agricultural fields located downstream [8]. Sediment transport can also lead to changes in water quality by carrying chemical-bound particles and due to the increase in water turbidity [9,10,11,12,13,14,15,16,17]. Whether natural, as a result of anthropic landscape transformations, or a combination of both, sediment provenance within a watershed represents essential information for the integral management of water and land, particularly in vulnerable communities that lack basic services (drinking water, sewage, etc.) and live on what they can locally cultivate [18,19,20]. Similarly, knowledge of watershed processes is necessary to prevent erosion and the subsequent loss of soil nutrients and, also, the transport of possible polluting material downstream. Tracing the source of sediments has been the target of many studies, resulting in the generation of tools that can be applied towards identifying the origins of such contaminants within a basin. Moreover, “sediment fingerprinting” is a technique that allows identification of the sediment provenance and estimation of the amount of contribution of each source to the target mixture, e.g., [21,22,23]. Many authors have focused their research on sediment fingerprinting to determine the provenance of suspended sediment in agricultural catchments [24,25,26], tropical fluvial and marine environments [27,28,29], wind-deposited sediments [30,31,32], and glacier regions [33,34,35], among other landscapes. Tracers used in sediment fingerprinting studies include color properties [36,37,38], major and trace elemental composition [39,40], rare earth elements [33,41,42,43], radionuclide characteristics [38,44,45], and organic matter [46,47]. Furthermore, authors around the world, e.g., [48,49] have used this technique to determine sediment sources; in South America, studies have been carried out in Argentina [50], Brazil [51], and Venezuela [52]. In Peru, however, sediment fingerprinting has been applied only on studies involving marine environments [53], but nothing has been carried out in inland regions of the country.
There are various software packages that enable the user to estimate the sources of the sediments through models (MixSIAR, FingerPro v1.3, SourceTracker, SIFT). The FingerPro package allows unmixing of the sediments of previously-assigned points (mix.sample) by selecting a set of optimal tracers that best fit the reality of databases [22]. Understanding the set of tracers is essential when making linear models to determine the origin of sediments, as poor tracer selection may lead to incorrect models and interpretation. In addition, the graphics and statistical tools included in the FingerPro package provide greater confidence in identifying the best tracers that can be used for the subsequent sediment provenance model [22].
Herein, we took advantage of the Fingerpro frequentist model [22,23,54] to investigate the provenance of sediments in the Ocoña watershed of southern Peru, with the objective of providing a tool for a more sustainable management of land and water in a region with large anthropic activity.

2. Material and Methods

2.1. Study Area

The Ocoña watershed is located in southern Peru, with a land area of 15,913 km2 and a perimeter of 883.34 km2 [55]. The watershed is part of the Andean slope draining to the Pacific Ocean (Figure 1), sharing boundaries with the Alto Apurimac and Pampas catchments to the north, the Pescadores, Atico, Chaparra, Yauca, and Pacific Ocean catchments to the southm the Camana-Majes-Colca catchment to the east, and the Pescadores, Yauca, and Pampas catchments to the west.

2.1.1. Catchment Morphology and Hydrological Characteristics

The Ocoña watershed has a rugged topography that ranges between 6425 m.a.s.l. in the snowy Coropuna [56] to 0 m.a.s.l. in the coast of the Pacific Ocean. There is a flat (0.85% slope) relief called the “Ocoña valley” zone, located between 0 and 1000 m.a.s.l. In the northeast area of the catchment, the relief is very rugged from 1000 to 6425 m.a.s.l., where the 3370-meter-deep Cotahuasi canyon is located, considered one of the deepest canyons in the world. Peru’s National Water Authority carried out a longitudinal profile of the Ocoña catchment and obtained an average slope of 1.64% [56], while the mean slope of the land in the catchment, where surface runoff, infiltration, and dragging of materials occur, is 30.09% (see Figure 1). The total length of the Ocoña riverbed is 286.6 km, with a flow rate that normally varies between 15 and 400 m3/s. A hydrographic delimitation in ten sub-catchments (Parinacochas, Bajo Ocoña, Churunga, Medio Bajo Ocoña, Chichas, Medio Ocoña, Cotahuasi, Medio Alto Ocoña, Mirmaca, and Alto Ocoña) was carried out by the National Water Authority [56].
A series of snow-capped mountains are located in this catchment, among which Solimana, Coropuna, Firura, Sopohuana, Chilluri, and Sarasara stand out. The Ocoña watershed’s main river is supplied mostly from two sources: (a) rainfall in the north (higher areas), and (b) melting of snowpack up in the mountains. The wet season, where most annual precipitation falls, is concentrated in summer months (December through March), while the dry season is normally from May to August [56]. Mean annual rainfall for the Ocoña watershed is only 362.2 mm/year, though the Cotahuasi sub-watershed receives more rainfall (500 mm/year). The mean annual temperature is 15.3 °C, reaching maximum and minimum values of 16.2 and 14.1 °C, respectively [57].
The geodynamic processes in the Ocoña watershed are mainly rockfall (28%), flows (29%), landslides (11%), and creep (1%), among others (26%) [56]. The identified sites where major geodynamic processes occur are the Cotahuasi rivers (Cotahuasi sub-watershed) and the Maran River, which belongs to the Alto Ocoña sub-watershed.
Nearly 62 thousand people live in the Ocoña watershed, with most of them located in Alto Ocoña (25.6%) and Cotahuasi (22.2%) sub-watersheds [56]. The main economic activities in the area are mining, agriculture, pisciculture and fishing, and tourism. According to the Geological and Mining Cadastral Information System of INGEMMET (GEOCATMIN) database, there are around 60 locations within the watershed where mining is performed. The largest agricultural area is located in the lower valleys, where rice, beans, corn, alfalfa, and fruit trees are cultivated. Around 11 artisanal fishing organizations are dedicated to capturing river shrimp. Other activities, such as tourism generate an additional burden on water sources. Tourism is especially prevalent in the upper catchment, as 60 archaeological zones, 21 hot springs, and 8 other touristic areas are distributed within this portion of the watershed. The best-known touristic areas are the Cotahuasi River Canon, the Sipia Waterfall, and the Luicho thermal baths [1].

2.1.2. Geological Setting

The Ocoña watershed is part of the Nazca–Ocoña metallogenic belt, which hosts around 70 Au-Ag and base metal deposits in a zone that extends more than 350 km along the central part of Peru [2,3,58,59,60,61,62]. The region’s basement rock is constituted by the Yamano and Yura groups, with ages from the late Paleozoic to Jurassic periods [63]. Gold and base metal deposits are spatially related to the Cretaceous intrusive rocks of the Coastal Batholith, with gold occurring as native gold, gold tellurides, and electrum. The Au-Ag veins are hosted by granodioritic to dioritic bodies, which were intruded by andesitic bodies [3]. Gold in these deposits is mainly hosted in fault-controlled quartz veins, and the vein orientations usually vary among deposits. Quartz veins are dominantly composed of quartz, carbonate, and sulfide minerals. The most dominant sulfide minerals are pyrite, chalcopyrite, galena, and sphalerite [2,3].

2.2. Methodology

2.2.1. Statistical Analysis of the ICP-MS Stream Sediment Samples Database

Data from a total of 362 stream sediment samples collected at the Ocoña watershed ICP-MS were drawn from the GEOCATMIN-INGEMMET database (see Supplementary Materials File S1). This database is composed of the following elements: Al, Ca, Na, K, Mg, Fe, Ti, V, Au, Ag, As, Cu, Pb, Zn, Co, Ni, Cr, Mn, P, Sr, Zr, Hg, La, Mo, Sc, W, Y, Be, Bi, Cd, Sn, and Sb. Elemental correlation at the watershed scale were analyzed using the ioGASTM 6.2 Software, emphasizing trace elements, such as As, Au, Cu, Pb, Zn, and Hg, among others. Additionally, anomaly maps of As and Cu were developed using ArcGIS, employing several interpolation tools to create a surface grid. The interpolation maps were performed using an inverse distance weighted (IDW) algorithm to improve the visualization of possible anomalies.
Details on spatial interpolation methods are provided in Liu and Yan [64] and are used to discuss methods in this section. The inverse distance weighting (IDW) interpolation method was applied to the dataset as opposed to the Kriging method. The Kriging method is a statistical approach used when the behavior or trend of the variables in the study area is known. The trend of the stream sediment values in this study area is not known, so there is risk in overestimating interpolated values through Kriging. The IDW interpolation method is calculated as follows:
Z 0 = ∑ i = 0 n ( Z i , Q i )  
where Z 0 represents the estimated value at an observation point, Z i is the value of variable Z at interpolation point i , and Q i is the weight coefficient assigned to the interpolation point i calculated as a function of the distance between the point of observation and the interpolation point i . Variable n represents the number of interpolation points. The IDW interpolation calculation works off the assumption that the observation point has a local effect on interpolation points. The influence of the observation point value on interpolation points decreases with increasing distance.

2.2.2. Source Selection

The Ocoña basin has ten sub-watersheds as defined by ANA (National Water Authority) [56]. In order to group the sub-watersheds into more detailed source groups, the LDAPlot() analysis function in the Fingerpro package was carried out, which performs a linear discriminant analysis and displays data in two dimensions. Once sources were well defined, one or more unmixing samples were selected. For this study, the Medio Ocoña sub-basin, which represents 0.1% of the watershed area [56], does not have stream sediment data, and was not taken into account for the modeling. For the purpose of this study, we defined unmixed samples located at the lower portion of the Bajo Ocoña sub-watershed, where sediments originating from the sub-watersheds are well combined (Figure 2). Once the number of sources was known (n), the next step was to identify which set of tracers was the best able to carry out the unmixing model.

2.3. Tracer Selection Methods

The main goal of tracer selection methods is to eliminate the introduction of non-conservative tracers into models that might otherwise affect their output [23]. Scholars have debated the reliability of the most widespread methods, such as the three-step technique or the mixing polygon [65]. Recent studies [23,54] propose another option to solve this debate, which is based on working out the new conservativeness index (CI), consensus ranking (CR), and consistent tracer selection (CTS) methods that generate similar outputs in unmixing with frequentist or Bayesian models. The CI, CR, and CTS rankings identify the non-conservative, non-consensual, and non-consistent tracers, respectively, show and report the effect of each tracer on the fingerprinting models, and identify whether there are multiple solutions in a dataset [65]. With this information, the user has the ability to decide, preventing the automatic inclusion of a tracer.
In this study, the CI, CR, and CTS methods were applied to investigate and select the best tracers, a procedure that needs at least n − 1 tracers to solve the sediment source uncertainty (where n is the number of sources), as reviewed in [22,23,54] and the GitHub repository at https://github.com/eead-csic-eesa/fingerPro, accessed on 24 January 2023.
The process of unmixing sediment mixtures to their sources was achieved by applying a frequentist model [22]. The open-source FingerPro 1.3 model is implemented in an R-package and estimates the relative contribution of each sediment source by exploring the complete space of all possible solutions [22]. The model uses a standard linear multivariate mixing process to estimate the source proportions, preserving the mass balance for all tracers. All possible combinations for each source contribution (0–100%) are evaluated in small increments using Latin hypercube sampling.

3. Results

IDW interpolation of the stream sediment geochemical data showed that elements, such as As and Cu, are anomalous in the areas of the watershed influenced by mining activities and geological exploration (Figure 3), such as Pecoy, Arihua, Tororume, and Chalhuane. In other words, human activities (in this case, mining) seem to have a strong influence in river pollution in the Ocoña watershed. Moreover, Figure 4 illustrates a box plot analysis showing As, Au, Cu, Pb, Zn, and Hg concentrations in the stream sediment database found in the Ocoña watershed. The highest As and Cu grades originated from sources in the middle part of the basin. Furthermore, copper concentrations in the entire Ocoña catchment range from 4.4 to 504 ppm, with a median of 29.1, whereas the middle part of the catchment has a median of 64 ppm.
Based on the linear discriminant analysis (LDAPlot, Figure 5), it was possible to visually identify the three potential sources for the Ocoña watershed: (1) upper (Cotahuasi, Alto Ocoña, Parinacochas, and Mirmaca); (2) middle (Medio Alto Ocoña, Medio Bajo Ocoña, and Chichas); and (3) lower (Churunga and Bajo Ocoña). Two samples were selected (mix-sample 1 and mix-sample 2) to quantify the sediment source contribution in the basin through the frequentist model Fingerpro, with both samples being located in the lower part of the Ocoña basin. The new CI, CR, and CTS methodologies were applied for mix-samples 1 and 2 to extract the information of each individual tracer [23,54]. Results of the tracer’s selection for mixture samples 1 and 2, carried on with the CI and CR methods, are detailed in Table 1. Tracers, such as Co, Sc, As, Cu, Be, V, and La indicate high CR values (up to 92%) for mix-sample 1 (Table 1A). For mix-sample 2 (Table 1B), Cu, Be, Pb, V, K, and Al reached values up to 93%. A consensus ranking provides the consensus among tracers, but this does not mean that the first ones are better than using the second and the third tracer. They still need to be consistent.
Ternary diagrams were created to illustrate CI results for tracer selection, representing all possible predictions from each tracer for mix-samples 1 and 2, with results being illustrated in Figure 6 and Figure 7, respectively. For example, visual inspection of mix-sample 1 (Figure 6) indicates that among 32 tracers, Co, Sc, Cu, As, Be, V, and La discriminate sources 1, 2, and 3. Similarly, visual inspection of ternary diagrams applied to mix-sample 2 (Figure 7) suggests that Cu, Pb, Be, and V discriminate the three sources.
Although preliminarily defined tracers for mix-samples 1 and 2 are available with the CR and CI methodology, there is still a possibility that using each of these tracers in the unmixing model will generate inconsistent results. For this reason, the CTS method was implemented, ordering the possible consistent solutions according to their discriminant capacity. The CTS results are presented in the Supplementary Material (File S1). For mix-sample 1, As and Co were the most discriminating tracer pair for the Ocoña dataset, with CR values of 97.5 and 97.9%, respectively. Arsenic and Co are the pair of elements that also have the best CR and CI values (Table 1A). When CTS was applied to mix-sample 2, V and Cu were the most discriminating tracer pair of the dataset, with the best CR values (greater than 90%), that is, Cu, Pb, and V, with values of 99.4, 97.7, and 96.6%, respectively (see Table 1B). All results from the CTS methodology with the pairs of tracer’s selections applied to mix-samples 1 and 2 are included in File S1.
According to results from mix-sample 1, the upper, middle, and lower sources contributed 70, 21, and 9% of sediments, respectively (Figure 8). In the case of mix-sample 2, sediment contributions from upper, middle, and lower sources were 61, 29, and 10%, respectively (Figure 9).

4. Discussion

4.1. Sediment Source Fingerprinting

The sediment fingerprinting technique was developed in order to understand pollutant sources within watersheds, and watershed-scale sediment provenance research is considered the most effective tool to assess this research topic [22]. The most traditional approach for applying sediment source tracking techniques is to define the relevant properties of tracers that provide a specific signature between source samples and unequivocally discriminate the different sources [66]. A single tracer was used at the beginning of sediment source estimation [67], but, as the quantification of the models evolved, it became necessary to discriminate between more than two sources, with the subsequent increase in the number of tracers. A variety of chemical and physical tracer techniques have been used to investigate sediment sources in watersheds, so a tracer should be able to discriminate among the studied sources [21].
The methods currently used for tracer selection in sediment fingerprinting have been discussed by other authors, e.g., [68,69]. The most commonly used tracer selection techniques include the (a) range test (RT) or mass conservation test, which exclude the tracer properties of the source mixture with the lowest and highest values in the sediment sources; (b) the Kruskal–Wallis (KW) test, which is a non-parametric test based on ranges used to determine if there are significant differences between the medians of groups or selected sources; and (c) the discriminant functions analysis (DFA) test, which identifies the optimal set of tracers that maximizes discrimination between sediment sources and minimizes the number of tracers to be used, and is based on the Wilk’s Lambda classification criteria. The RT, KW, and DFA tests are statistics that evaluate the ability of individual tracers to differentiate between sources and identify the best combination of tracers to provide maximum discrimination of source classes. One of the limitations of applying RT, KW, and DFA tests is that they do not incorporate the information from the mix-sample in the analysis in two of the three tests [22]. Even though the traditional methods could sometimes select appropriate tracers, the availability of open-source tools, such as CI, CR, and CTS that display novel and valuable information, prevent possible issues generated by the previously cited ones.
Although frequentist and Bayesian models can be affected by the selection of erroneous or non-informative tracers, the consistent tracer selection (CTS) method is necessary to assist during the selection of the optimal tracers, avoiding non-conservative and dissenting tracers present in a dataset for the analysis of each mix-sample. Moreover, the CTS methodology in the FingerPro package was used to avoid the inclusion of tracers pointing to different solutions that create overdetermined systems, misleading both Bayesian and Frequentist models.

4.2. Global Perspectives

Comparing sediment fingerprinting results with those obtained by other authors elsewhere is complicated because they vary depending on sampling efficiency, modeling approach, climate and seasonality, geology, soil types, particle size, land use, and watershed topography and size, among other variables, e.g., [70,71,72,73,74,75]. Our results indicate that most sediments come from the upper portion of the Ocoña watershed, as described in other sections of this manuscript. Assuming high mining activities in those areas, these results agree with the findings by Fang and Fan [76], who investigated sediment sources in the Shouchang watershed (southeastern China), concluding that most sediments were originated from human activities. Similarly, one of the main sediment sources found by Garcias et al. [77] in Argentina was exotic forest plantations, normally characterized for yielding sediments after clearcutting and road construction, e.g., [78,79], while Tiecher et al. [80] concluded that agricultural fields in Brazil are a significant source of sediments, even under no-tillage conditions, similar to the findings by Zhang et al. [81] in Hampshire, UK. However, Liu et al. [82] reported that agriculture represented less than 10% of sediment sources in the Lower Little Bow River (Alberta), with the majority of sources unknown. Moreover, Patault et al. [83] concluded that land management policies strongly affected sediment sources in the Canche River (France). All those authors used fingerprinting techniques to determine the source of sediments in watersheds around the world, but, in short, even though our results are robust and the methodology applied is established, we cannot easily compare them with other studies; under the unique site conditions that characterize the Ocoña watershed, our findings still provide a god guidance for better land management.

4.3. Environmental Implications, Economic Impact, Novelty, and Practical Applicability

Erosion and sediment contribution in a watershed affected by mining operations and, therefore, mining liabilities with high contents of metal pollution (e.g., As, Hg, and Pb, among others) in semiarid climates are an issue of concern to ecosystems, local communities, the economy of the area, and national and local authorities [2,3,11,16,26].
There are many artisanal and informal mining operations in the Ocoña watershed, which provide income to thousands of families. The vast majority of the waste rock that originates from the mining activity is stored in “dumps” on the margins of the river network. In addition to waste rock, there are environmental liabilities, such as mining tailings and cyanidation wells. For example, the waste rock and the mining tailings at Natividad and Alpacay [84] are sources of metals (i.e., As, Cu, Pb, Fe, and Hg) contained in sediments that are ultimately deposited in Ocoña’s main riverbed. The interaction between mining activity and geological processes (erosion, rockfall, flows, landslides, and creep), result in the transport of sediments through the channel in tributaries and towards the main river. There is an extensive artisanal and informal mining activity in the Secocha sector (the middle portion of the Ocoña watershed). The mercury concentration in this portion of the river has been studied [85], and it was concluded that it exceeds Peruvian environmental quality standards (MPLs).
One of the main concerns of the local population, economy, and ecosystems is Hg pollution from the gold-extracting process at artisanal and/or informal mining operations, as the contaminant is released into the environment in liquid form and directly into mining tailings, or in gaseous form during the amalgamation process [3]. Palacios et al. [86] conducted a study on Hg contamination in the Ocoña watershed, identifying three areas with values greater than 1 μg/L, all in the Misky region, near the area where gold is extracted using “quimbaletes”, which are large mortars composed of a hand-made stone with an oval base, allowing a back-and-forth movement with minimal effort (see [87] for more details).
According to Sabino et al. [88], the Ocoña watershed has moderate erosion rates (1.53 Mg ha/yr), and our sediment provenance study suggests that the largest contributions of sediments in this specific watershed come from the headwaters (61–70%), followed by the middle (21–29%), and lower (9–10%) portions. Geochemical anomaly maps (see Figure 3) suggest that the highest metal concentrations originate from the middle part of the basin, which may be linked to a natural process (release of metals due to the erosion of mineralized veins) or to anthropogenic activity represented by artisanal mining, although more sediments come from the upper parts of the basin.
Currently, water demand in the Ocoña watershed is estimated to be 552.86 hm3/year, though water consumption is expected to be more than 800 hm3/year by 2035 [57]. Current agricultural water demands represent nearly 90% of the annual volume, while mining and domestic uses are barely 1%. According to Medina et al. [1], the water quality of Ocoña’s main river exceeds MPLs for analytes, such as Al, Cu, Fe, Mn, and Hg; those authors concluded that Cl, SO4, Al, As, Fe, and Mn are present in concentrations below MPLs during the rainy season.
The middle Ocoña sub-watershed has the highest gross hydroelectric potential, reaching a maximum of 1000 MW/km2, whereas the gross potential is 23,917 MW/km2 at the upper portion of the watershed, specifically downstream of the Aguaguina stream’s headwaters [56]. Since results from the present study suggest that most sediment originates from the upper and the middle portion of the drainage area, it is imperative to control erosion and sediment release if these areas are to be utilized for hydroelectric power, e.g., [4]. Moreover, effective watershed management practices are based on natural resources administration oriented towards producing and protecting water sources, including erosion/sediment control, flood mitigation, and landscape conservation. Since sediment release is mainly a mixed process between natural and human activities in the study area, sediment detachment should be focused on anthropic polluting areas, such as mining operations. Similarly, our fieldwork observations suggest that agriculture and grazing (especially in degraded areas) are also activities that can contribute to sediment detachment and transport toward the riverbeds [89].
Our study is pioneering research applied to determining the impact of mining activities on river pollution through stream sediment fingerprinting techniques. This study is novel in Peru, a country highly impacted by river pollution.

5. Conclusions

For the first time, a fingerprinting technique has been successfully applied to semiarid fluvial systems in Peru to track sediment sources in a watershed with significant mining influence. Results indicate that this technique is highly appropriate for future studies in areas with similar climates where mining activity is present.
Results from this study suggest that most stream sediments originate from the upper portion of the Ocoña watershed, with geochemical anomalies (As and Cu) being closely related to sites with mining activities or geological exploration, all concentrated in the middle portion of the study area.
Results from this study can be helpful for future decision making to ensure a better environment in this semiarid watershed. Most importantly, the results suggest that erosion and sediment control practices are strictly necessary for areas where human activities can potentially contribute pollutants to the watershed.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/su151511856/s1, File S1.

Author Contributions

J.C., M.G. and S.T. collected the samples from the watershed and from the INGEMMET database. J.C., M.G. and S.T. studied the samples applying statistical analyses. All authors (J.C., E.H., M.G., S.T., I.S., I.L., P.A.G.-C. and G.M.) discussed the results and evaluated the data. J.C., E.H., I.L., P.A.G.-C. and G.M. wrote, organized, reviewed, and edited the original draft and reviewed manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

Project 2.7: Sustainability of five watersheds in the Arequipa Region, Project # 470160.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Acknowledgments

This important investigation was possible thanks to the Center for Mining Sustainability, a collaboration between Universidad Nacional de San Agustín de Arequipa (Peru) and the Colorado School of Mines (USA). The contribution of I. Lizaga was supported by the Research Foundation-Flanders (FWO, mandate 12V8622N).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Medina, L.; Gomez, H.; Santos, B.; Moreno, J.; Pari, W. Estudio Geoambiental en la Cuenca del Río Ocoña Boletín Nº2. INGEMMET. 2021, 222p. Available online: https://hdl.handle.net/20.500.12544/3167 (accessed on 24 January 2023).
  2. Alfonso, P.; Anticoi, H.; Yubero, T.; Bascompta, M.; Henao, L.; Garcia-Valles, M.; Yáñez, J. The importance of mineralogical knowledge in the sustainability of artisanal gold mining: A mid-south Perú case. Minerals 2019, 9, 345. [Google Scholar] [CrossRef] [Green Version]
  3. Crespo, J.; Holley, E.; Pfaff, K.; Guillen, M.; Huamani, R. Ore Mineralogy, Trace Element Geochemistry and Geochronological Constraints at the Mollehuaca and San Juan de Chorunga Au-Ag Vein Deposits in the Nazca-Ocoña Metallogenic Belt, Arequipa, Perú. Minerals 2020, 10, 1112. [Google Scholar] [CrossRef]
  4. Garcia-Chevesich, P. Erosion Control and Land Restoration; Outskirts Press: Denver, CO, USA, 2016; 486p, ISBN 9781478765837. [Google Scholar]
  5. Dung, T.T.T.; Cappuyns, V.; Swennen, R.; Phung, N.K. From geochemical background determination to pollution assessment of heavy metals in sediments and soils. Rev. Environ. Sci. Bio/Technol. 2013, 12, 335–353. [Google Scholar] [CrossRef]
  6. Brady, J.; Ayoko, G.; Martens, W.; Goonetilleke, A. Development of a hybrid pollution index for heavy metals in marine and estuarine sediments. Environ. Monit. Assess. 2015, 187, 306. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  7. Ciszewski, D.; Grygar, T. A review of flood-related storage and remobilization of heavy metal pollutants in river systems. Water Air Soil Pollut. 2016, 227, 239. [Google Scholar] [CrossRef] [Green Version]
  8. Alves, L.; Reis, A.; Gratão, P. Heavy metals in agricultural soils: From plants to our daily life (a review). Científica 2016, 44, 346–361. [Google Scholar] [CrossRef] [Green Version]
  9. Salomons, W. Sediments and water quality. Environ. Technol. 1985, 6, 315–326. [Google Scholar] [CrossRef]
  10. Lartiges, B.S.; Deneux-Mustin, S.; Villemin, G.; Mustin, C.; Barrès, O.; Chamerois, M.; Gerard, B.; Babut, M. Composition, structure and size distribution of suspended particulates from the Rhine River. Water Res. 2001, 35, 808–816. [Google Scholar] [CrossRef]
  11. Osman, A.; Kloas, W. Water quality and heavy metal monitoring in water, sediments, and tissues of the African Catfish Clarias gariepinus (Burchell, 1822) from the River Nile, Egypt. J. Environ. Prot. 2010, 1, 389. [Google Scholar] [CrossRef] [Green Version]
  12. Chapman, D. Water Quality Assessments: A Guide to the Use of Biota, Sediments and Water in Environmental Monitoring; CRC Press: Abingdon, UK, 2021; ISBN 0-203-78495-2. [Google Scholar]
  13. Wood, P.; Armitage, P. Biological Effects of Fine Sediment in the Lotic Environment. Environ. Manag. 1997, 21, 203–217. [Google Scholar] [CrossRef] [PubMed]
  14. Davies-Colley, R.; Smith, D. Turbidity suspended sediment, and water clarity: A review 1. J. Am. Water Resour. Assoc. 2001, 37, 1085–1101. [Google Scholar] [CrossRef]
  15. Parra, L.; Rocher, J.; Escrivá, J.; Lloret, J. Design and development of low cost smart turbidity sensor for water quality monitoring in fish farms. Aquac. Eng. 2018, 81, 10–18. [Google Scholar] [CrossRef]
  16. Yang, Q.; Gao, H.; Han, Y.; Li, Z.; Lu, K. Evolution of the Relationship between Runoff and Sediment Transport during Flood Event in the Chabagou Watershed of the Loess Plateau. Sustainability 2022, 14, 11573. [Google Scholar] [CrossRef]
  17. Chen, B.; He, R.; Cai, P.; Huang, G.; Wang, F. Geochemical Speciation, Risk Assessment, and Sources Identification of Heavy Metals in Mangrove Surface Sediments from the Nanliu River Estuary of the Beibu Gulf, China. Sustainability 2022, 14, 9112. [Google Scholar] [CrossRef]
  18. Bach, H.; Clausen, T.; Trang, D.; Emerton, L.; Facon, T.; Hofer, T.; Whiting, L. From Local Watershed Management to Integrated River Basin Management at National and Transboundary Levels; Mekong River Commission: Vientiane, Laos, 2011; ISBN 978-9932-412-04-4. [Google Scholar]
  19. Shi, Z.; Ai, L.; Fang, N.; Zhu, H. Modeling the impacts of integrated small watershed management on soil erosion and sediment delivery: A case study in the Three Gorges Area, China. J. Hydrol. 2012, 438, 156–167. [Google Scholar] [CrossRef]
  20. Bremer, L.; Hamel, P.; Ponette-González, A.; Pompeu, P.; Saad, S.; Brauman, K. Who are we measuring and modeling for? Supporting multilevel decision-making in watershed management. Water Resour. Res. 2020, 56, e2019WR026011. [Google Scholar] [CrossRef] [Green Version]
  21. Haddadchi, A.; Ryder, D.; Evrard, O.; Olley, J. Sediment fingerprinting in fluvial systems: Review of tracers, sediment sources and mixing models. Int. J. Sediment Res. 2013, 28, 560–578. [Google Scholar] [CrossRef] [Green Version]
  22. Lizaga, I.; Latorre, B.; Gaspar, L.; Navas, A. FingerPro: An R package for tracking the provenance of sediment. Water Resour. Manag. 2020, 34, 3879–3894. [Google Scholar] [CrossRef]
  23. Latorre, B.; Lizaga, I.; Gaspar, L.; Navas, A. A novel method for analyzing consistency and unravelling multiple solutions in sediment fingerprinting. Sci. Total Environ. 2021, 789, 147804. [Google Scholar] [CrossRef]
  24. Collins, A.; Walling, D.; Leeks, G. Use of composite fingerprints to determine the provenance of the contemporary suspended sediment load transported by rivers. Earth Surface Processes and Landforms. J. Br. Geomorphol. Group 1998, 23, 31–52. [Google Scholar] [CrossRef]
  25. Walling, D. Tracing suspended sediment sources in catchments and river systems. Sci. Total Environ. 2005, 344, 159–184. [Google Scholar] [CrossRef] [PubMed]
  26. Collins, A.L.; Zhang, Y.; Walling, D.; Grenfell, S.; Smith, P. Tracing sediment loss from eroding farm tracks using a geochemical fingerprinting procedure combining local and genetic algorithm optimization. Sci. Total Environ. 2010, 408, 5461–5471. [Google Scholar] [CrossRef] [PubMed]
  27. Maher, B.; Watkins, S.; Brunskill, G.; Alexander, J.; Fielding, C. Sediment provenance in a tropical fluvial and marine context by magnetic ‘fingerprinting’ of transportable sand fractions. Sedimentology 2009, 56, 841–886. [Google Scholar] [CrossRef]
  28. D’Haen, K.; Verstraeten, G.; Degryse, P. Fingerprinting historical fluvial sediment fluxes. Prog. Phys. Geogr. 2012, 36, 154–186. [Google Scholar] [CrossRef] [Green Version]
  29. Koiter, A.J.; Owens, P.N.; Petticrew, E.L.; Lobb, D.A. The behavioural characteristics of sediment properties and their implications for sediment fingerprinting as an approach for identifying sediment sources in river basins. Earth-Sci. Rev. 2013, 125, 24–42. [Google Scholar] [CrossRef]
  30. Liu, B.; Niu, Q.; Qu, J.; Zu, R. Quantifying the provenance of aeolian sediments using multiple composite fingerprints. Aeolian Res. 2016, 22, 117–122. [Google Scholar] [CrossRef]
  31. Gholami, H.; Telfer, M.W.; Blake, W.H.; Fathabadi, A. Aeolian sediment fingerprinting using a Bayesian mixing model. Earth Surf. Process Landf. 2017, 42, 2365–2376. [Google Scholar] [CrossRef] [Green Version]
  32. Ahmady-Birgani, H.; Agahi, E.; Ahmadi, S.J.; Erfanian, M. Sediment source fingerprinting of the Lake Urmia sand dunes. Sci. Rep. 2018, 8, 206. [Google Scholar] [CrossRef] [Green Version]
  33. Navas, A.; Lizaga, I.; Gaspar, L.; Stott, T.; Mavlyudov, B.; Dercon, G. Modeling the provenance of fine sediments in Parón Glacier Lake (Cordillera Blanca, Perú) by using a new procedure of tracer selection. In Proceedings of the EGU General Assembly Conference Abstracts, Online, 4–8 May 2020; European Geosciences Union: Munich, Germany, 2020; p. 3270. [Google Scholar] [CrossRef]
  34. Tsyplenkov, A.; Vanmaercke, M.; Collins, A.L.; Kharchenko, S.; Golosov, V. Elucidating suspended sediment dynamics in a glacierized catchment after an exceptional erosion event: The Djankuat catchment, Caucasus Mountains, Russia. Catena 2021, 203, 105285. [Google Scholar] [CrossRef]
  35. Navas, A.; Lizaga, I.; Santillán, N.; Gaspar, L.; Latorre, B.; Dercon, G. Targeting the source of fine sediment and associated geochemical elements by using novel fingerprinting methods in proglacial tropical highlands (Cordillera Blanca, Perú). Hydrol. Process. 2022, 36, e14662. [Google Scholar] [CrossRef]
  36. Martínez-Carreras, N.; Krein, A.; Gallart, F.; Iffly, J.F.; Pfister, L.; Hoffmann, L.; Owens, P.N. Assessment of different colour parameters for discriminating potential suspended sediment sources and provenance: A multi-scale study in Luxembourg. Geomorphology 2010, 118, 118–129. [Google Scholar] [CrossRef]
  37. Barthod, L.R.; Liu, K.; Lobb, D.A.; Owens, P.N.; Martínez-Carreras, N.; Koiter, A.J.; Gaspar, L. Selecting color-based tracers and classifying sediment sources in the assessment of sediment dynamics using sediment source fingerprinting. J. Environ. Qual. 2015, 44, 1605–1616. [Google Scholar] [CrossRef] [Green Version]
  38. Boudreault, M.; Koiter, A.J.; Lobb, D.A.; Liu, K.; Benoy, G.; Owens, P.N.; Li, S. Using colour, shape and radionuclide fingerprints to identify sources of sediment in an agricultural watershed in Atlantic Canada. Can. Water Resour. J. Rev. Can. Des. Ressour. Hydr. 2018, 43, 347–365. [Google Scholar] [CrossRef]
  39. Miller, J.R.; Lord, M.; Yurkovich, S.; Mackin, G.; Kolenbrander, L. Historical trends in sedimentation rates and sediment provenance, Fairfield Lake, Western North Carolina1. J. Am. Water Resour. Assoc. 2005, 41, 1053–1075. [Google Scholar] [CrossRef]
  40. Mukundan, R.; Radcliffe, D.E.; Ritchie, J.C.; Risse, L.M.; McKinley, R.A. Sediment fingerprinting to determine the source of suspended sediment in a southern Piedmont stream. J. Environ. Qual. 2010, 39, 1328–1337. [Google Scholar] [CrossRef] [Green Version]
  41. Zhang, Q.; Lei, T.; Zhao, J. Estimation of the detachment rate in eroding rills in flume experiments using an REE tracing method. Geoderma 2008, 147, 8–15. [Google Scholar] [CrossRef]
  42. Zhou, X.; Li, A.; Jiang, F.; Meng, Q. A preliminary study on fingerprinting approach in marine sediment dynamics with the rare earth elements. Acta Oceanol. Sin. 2010, 29, 62–77. [Google Scholar] [CrossRef]
  43. Habibi, S.; Gholami, H.; Fathabadi, A.; Jansen, J.D. Fingerprinting sources of reservoir sediment via two modelling approaches. Sci. Total Environ. 2019, 663, 78–96. [Google Scholar] [CrossRef]
  44. Estrany, J.; Garcia, C.; Walling, D.E. An investigation of soil erosion and redistribution in a Mediterranean lowland agricultural catchment using caesium-137. Int. J. Sediment Res. 2010, 25, 1–16. [Google Scholar] [CrossRef]
  45. Wang, J.; Zhang, W.; Baskaran, M.; Du, J.; Zhou, F.; Wu, H. Fingerprinting Sediment Transport in River-Dominated Margins Using Combined Mineral Magnetic and Radionuclide Methods. J. Geophys. Res. Ocean. 2018, 123, 5360–5374. [Google Scholar] [CrossRef] [Green Version]
  46. Walling, D.E.; Amos, C.M. Source, storage and mobilization of fine sediment in a chalk stream system. Hydrol. Process. 1999, 13, 323–340. [Google Scholar] [CrossRef]
  47. Koiter, A.J.; Owens, P.N.; Petticrew, E.L.; Lobb, D.A. Assessment of particle size and organic matter correction factors in sediment source fingerprinting investigations: An example of two contrasting watersheds in Canada. Geoderma 2018, 325, 195–207. [Google Scholar] [CrossRef]
  48. Lake, N.F.; Martinez-Carreras, N.; Shaw, P.; Collins, A. Using particle size distributions to fingerprint suspended sediment sources—Evaluation at laboratory and catchment scales. Hydrol. Process. 2022, 36, e14726. [Google Scholar] [CrossRef]
  49. Yue, W.; Yue, X.; Zhang, L.; Liu, X.; Song, J. Morphology of Detrital Zircon as a Fingerprint to Trace Sediment Provenance: Case Study of the Yangtze Delta. Minerals 2019, 9, 438. [Google Scholar] [CrossRef] [Green Version]
  50. Chitu, A.; Zijp, M.; Zwaan, J. A novel exploration technique using the microbial fingerprint of shallow sediment to detect hydrocarbon microseepage and predict hydrocarbon charge; an Argentinian case study. Interpretation 2022, 10, SB77–SB92. [Google Scholar] [CrossRef]
  51. Franz, C.; Makeschin, F.; Weiß, H.; Lorz, C. Sediments in urban river basins: Identification of sediment sources within the Lago Paranoá catchment, Brasilia DF, Brazil—Using the fingerprint approach. Sci. Total Environ. 2014, 466–467, 513–523. [Google Scholar] [CrossRef] [PubMed]
  52. Deplazes, G.; Meckler, A.; Peterson, L.; Hamann, Y.; Aeschlimann, B.; Guenther, D.; Martinez-Garcia, A.; Haug, G. Fingerprint of tropical climate variability and sea level in sediments of the Cariaco Basin during the last glacial period. Sedimentology 2018, 66, 1967–1988. [Google Scholar] [CrossRef]
  53. Scholz, F.; Severmann, S.; McManus, J.; Hensen, C. Beyond the Black Sea paradigm: The sedimentary fingerprint of an open-marine iron shuttle. Geochim. Cosmochim. Acta 2014, 127, 368–380. [Google Scholar] [CrossRef]
  54. Lizaga, I.; Latorre, B.; Gaspar, L.; Navas, A. Consensus ranking as a method to identify non-conservative and dissenting tracers in fingerprinting studies. Sci. Total Environ. 2020, 720, 137537. [Google Scholar] [CrossRef]
  55. Autoridad Nacional del Agua SNIRH. Available online: https://snirh.ana.gob.pe/observatorioSNIRH/# (accessed on 17 March 2022).
  56. ANA (Autoridad Nacional del Agua). Evaluación de Recursos Hídricos en la Cuenca de Ocoña, Infome Final. 2015. Available online: https://hdl.handle.net/20.500.12543/8 (accessed on 24 January 2023).
  57. Ministerio de Agricultura; Instituto Nacional de Recursos Naturales; Intendencia de Recursos Hidricos. Evaluación de los Recursos Hidricos de la Cuenca del Rio Ocoña, Estudio Hidrologico. Informe Final. 2007. Available online: https://hdl.handle.net/20.500.12543/3886 (accessed on 24 January 2023).
  58. Vargas, A. Estudio Geológico-Minero de la Faja Aurífera Nazca-Ocoña; Ingemmet Techincal Report; INGEMMET: Lima, Peru, 1978; p. 179. [Google Scholar]
  59. Vidal, C. Metallogenesis associated with the Coastal Batholith of Perú: A review. In Magmatism at a Plate Edge. The Peruvian Andes; Blckie & Son Ltd.: Glasgow, UK, 1985; pp. 243–249. [Google Scholar]
  60. Acosta, J. Características metalogenéticas de los yacimientos asociados a los arcos magmáticos mesozoicos y cenozoicos del sur del Perú. In Informe Técnico Interno DGEPM; INGEMMET: Lima, Peru, 2006; pp. 20–26. [Google Scholar]
  61. Acosta, J.; Quispe, J.; Santiesteban, A.; Acosta, H. Épocas metalogenéticas y tipos de yacimientos metálicos en la margen occidental del Sur del Perú: Latitudes 14° S–18° S. In Proceedings of the Resúmenes Extendidos XIV Congreso Peruano de Geología, Lima, Peru, 20 September–3 October 2008; Available online: https://hdl.handle.net/20.500.12544/424 (accessed on 24 January 2023).
  62. Acosta, J.; Rivera, R.; Valencia, M.; Chirif, H.; Huancacuni, D.; Rodriguez, I.; Villareal, E.; Paico, D.; Santiesteban, A. Mapa Metalogenético del Perú. Informe Técnico; INGEMMET: Lima, Peru, 2009; p. 17. Available online: https://hdl.handle.net/20.500.12544/1947 (accessed on 24 January 2023).
  63. Boekhout, F.; Sempéré, T.; Spikings, R.; Schaltegger, U. Late Paleozoic to Jurassic chronostratigraphy of coastal southern Perú: Temporal evolution of sedimentation along an active margin. J. S. Am. Earth Sci 2013, 47, 179–200. [Google Scholar] [CrossRef]
  64. Liu, Z.; Yan, T. Comparison of spatial interpolation methods based on ArcGIS. J. Phys. 2021, 1961, 012050. [Google Scholar] [CrossRef]
  65. Evrard, O.; Batista, P.; Dabrin, A.; Foucher, A.; Frankl, A.; García-Comendador, J.; Sellier, V. Improving the design and implementation of sediment fingerprinting studies: Summary and outcomes of the TRACING 2021 Scientific School. J. Soils Sediments 2021, 22, 1648–1661. [Google Scholar] [CrossRef] [PubMed]
  66. Collins, A.L.; Walling, D.E. Selecting fingerprint properties for discriminating potential suspended sediment sources in river basins. J. Hydrol. 2002, 261, 218–244. [Google Scholar] [CrossRef]
  67. Walling, D.E.; Peart, M.R.; Oldfield, F.; Thompson, R. Suspended sediment sources identified by magnetic measurements. Nature 1979, 281, 110–113. [Google Scholar] [CrossRef]
  68. Pulley, S.; Foster, I.; Antunes, P. The uncertainties associated with sediment fingerprinting suspended and recently deposited fluvial sediment in the Nene river basin. Geomorphology 2015, 228, 303–319. [Google Scholar] [CrossRef] [Green Version]
  69. Palazón, L.; Gaspar, L.; Torre, B.; Blake, W.; Navas, A. Identifying sediment sources by applying a fingerprinting mixing model in a Pyrenean drainage catchment. J. Soils Sediments 2015, 15, 2067–2085. [Google Scholar] [CrossRef] [Green Version]
  70. Nazari Samani, A.; Wasson, R.; Malekian, A. Application of multiple sediment fingerprinting techniques to determine the sediment source contribution of gully erosion: Review and case study from Boushehr province, southwestern Iran. Prog. Phys. Geogr. 2011, 35, 375–391. [Google Scholar] [CrossRef]
  71. Laceby, J.P.; Batista, P.; Taube, N.; Kruk, M.; Chung, C.; Evrard, O.; Orwin, J.; Kerr, J. Tracing total and dissolved material in a western Canadian basin using quality control samples to guide the selection of fingerprinting parameters for modelling. Catena 2021, 200, 105095. [Google Scholar] [CrossRef]
  72. Munroe, J.S.; Norris, E.; Carling, G.; Beard, B.; Satkoski, A.; Lianwen, L. Isotope fingerprinting reveals western North American sources of modern dust in the Uinta Mountains, Utah, USA. Aeolian Res. 2019, 38, 39–47. [Google Scholar] [CrossRef]
  73. Clarke, R.T.; Minella, J. Evaluating sampling efficiency when estimating sediment source contributions to suspended sediment in rivers by fingerprinting. Hydrol. Process. 2016, 30, 3408–3419. [Google Scholar] [CrossRef]
  74. Vale, S.; Smith, H.; Matthews, A.; Boyte, S. Determining sediment source contributions to overbank deposits within stopbanks in the Oroua river, New Zealand, using sediment fingerprinting. J. Hydrol. 2020, 59, 147–172. [Google Scholar]
  75. Navratil, O.; Evrard, O.; Esteves, M.; Legout, C.; Ayrault, S.; Némery, J.; Mate-Marin, A.; Ahmadi, M.; Lefevre, I.; Poirel, A.; et al. Temporal variability of suspended sediment sources in an alpine catchment combining river/rainfall monitoring and sediment fingerprinting: Temporal Variability of Suspended Sediment Sources in Mountains. Earth Surf. Process. Landf. 2012, 37, 828–846. [Google Scholar] [CrossRef]
  76. Fang, H.; Fan, Z. Fingerprinting sediment sources of the Shouchang River in the southeastern China. Environ. Earth Sci. 2021, 80, 33. [Google Scholar] [CrossRef]
  77. Garcias, Y.; Torres, R.; Ojeda, G.; de los Santos Villalobos, S.; Tejeda, S.; Velasco, H. Using Geochemical Fingerprints for Assessing Sediment Source Apportionment in an Agricultural Catchment in Central Argentina. Water 2021, 13, 3632. [Google Scholar] [CrossRef]
  78. Abdullah, A.; Erdas, O.; Mahmut, R.; Alaaddin, Y. Estimating sediment yield from a forest road network by using a sediment prediction model and GIS techniques. Build. Environ. 2008, 43, 687–695. [Google Scholar] [CrossRef]
  79. Khanal, S.; Parajuli, P. Evaluating the Impacts of Forest Clear Cutting on Water and Sediment Yields Using SWAT in Mississippi. J. Water Resour. Prot. 2013, 05, 474–483. [Google Scholar] [CrossRef] [Green Version]
  80. Tiecher, T.; Gomez, J.; Evrard, O.; Caner, L.; Merten, G.; Capoane, V.; Didone, E.; Santos, D. Fingerprinting sediment sources in a large agricultural catchment under no-tillage in southern Brazil (Conceicao River). Land Degrad. Dev. 2018, 29, 939–951. [Google Scholar] [CrossRef]
  81. Zhang, Y.; Collins, A.; McMillan, S.; Dixon, E.; Cancer-Berroya, E.; Poiret, C.; Stringfellow, A. Fingerprinting source contributions to bed sediment-associated organic matter in the headwater subcatchments of the River Itchen SAC, Hampshire, UK. River Res. Appl. 2017, 33, 1515–1526. [Google Scholar] [CrossRef]
  82. Liu, K.; Lobb, D.; Miller, J.; Owens, P.; Caron, M. Determining sources of fine-grained sediment for a reach of the Lower Little Bow River, Alberta, using a colour-based sediment fingerprinting approach. Can. J. Soil Sci. 2018, 98, 55–69. [Google Scholar] [CrossRef] [Green Version]
  83. Patault, E.; Alary, C.; Franke, C.; Abriak, N. Quantification of tributaries contributions using a confluence-based sediment fingerprinting approach in the Canche river watershed (France). Sci. Total Environ. 2019, 668, 457–469. [Google Scholar] [CrossRef] [PubMed]
  84. Garcia, B.V. Contaminación del Agua por Metales Pesados As, B, Cu, Pb, Cd y CN-en las Cuencas de los Ríos Tambo, Quilca, Camaná y Ocoña de la Región Arequipa. Ph.D. Thesis, Universidad Nacional de San Agustín de Arequipa, Arequia, Peru, 2019. [Google Scholar]
  85. Medina, D.E. Determinación de los Aspectos Ambientales que Afectan a la Población de Camarón de rio (Cryphiops caementarius) en el Rio Ocoña-Región Arequipa. Master’s Thesis, Universidad Nacional de San Agustín de Arequipa, Arequia, Peru, 2019. [Google Scholar]
  86. Palacios, U.S.; Abella, M.P.; Yáñez, J.A.; Higueras, P.L. Evaluación de la contaminación de mercurio en La Cuenca del rio Ocoña, Perú. In Proceedings of the III Congreso Internacional de Geología y Minería Ambiental para el Ordenamiento Territorial y el Desarrollo, Cardona, Spain, 6–9 November 2013. [Google Scholar]
  87. Florsch, N.; Téreygeol, F.; Cruz, P. The Ore-Dressing Grindstone Called a ‘Quimbalete’: A Mechanics-Based Approach. Archaeometry 2015, 58, 881–898. [Google Scholar] [CrossRef]
  88. Sabino Rojas, E.; Felipe-Obando, O.; Lavado-Casimiro, W. Atlas de Erosión de Suelos por Regiones Hidrológicas del Perú. Nota Técnica Nº 002 SENAMHI-DHI-2017. Available online: https://hdl.handle.net/20.500.12542/261 (accessed on 24 January 2023).
  89. Moreira, B.; Mirandola, P.; Pinto, A.; Salinas, E.; Oliveira, I. Manejo integrado de cuencas hidrográficas: Posibilidades y avances en los análisis de uso y cobertura de la tierra. Cuadernos de Geografía. Rev. Colomb. Geogr. 2020, 29, 69–85. [Google Scholar] [CrossRef] [Green Version]
Sustainability 15 11856 g001
Figure 1. Slope map of the Ocoña watershed using data from INGEMMET.
Figure 1. Slope map of the Ocoña watershed using data from INGEMMET.
Sustainability 15 11856 g001
Sustainability 15 11856 g002
Figure 2. Ocoña basin map with source selection and unmixing samples (own work).
Figure 2. Ocoña basin map with source selection and unmixing samples (own work).
Sustainability 15 11856 g002
Sustainability 15 11856 g003
Figure 3. Geochemical maps resulting from IDW interpolation of stream sediment data from the Ocoña catchment for (A) arsenic and (B) copper.
Figure 3. Geochemical maps resulting from IDW interpolation of stream sediment data from the Ocoña catchment for (A) arsenic and (B) copper.
Sustainability 15 11856 g003
Sustainability 15 11856 g004
Figure 4. Box plot analysis of the stream sediment concentration database, showing As, Au, Cu, Pb, Zn, and Hg. In each box plot, minimum, median, and maximum concentrations are indicated, and the number of analyses above the detection limit for each element is displayed inside each box. Outliers are indicated by circles (closer) and triangles (farther).
Figure 4. Box plot analysis of the stream sediment concentration database, showing As, Au, Cu, Pb, Zn, and Hg. In each box plot, minimum, median, and maximum concentrations are indicated, and the number of analyses above the detection limit for each element is displayed inside each box. Outliers are indicated by circles (closer) and triangles (farther).
Sustainability 15 11856 g004
Sustainability 15 11856 g005
Figure 5. LDAPlot of the Ocoña watershed from different sources: Source 1 in red (upper, circles), Source 2 in blue (middle, squares), and Source 3 in green (lower, triangles). Circled areas indicate 95% coverage.
Figure 5. LDAPlot of the Ocoña watershed from different sources: Source 1 in red (upper, circles), Source 2 in blue (middle, squares), and Source 3 in green (lower, triangles). Circled areas indicate 95% coverage.
Sustainability 15 11856 g005
Sustainability 15 11856 g006
Figure 6. Ternary diagram showing all possible contributions of each tracer mix-sample 1, with blue dots representing results of the simple tracer model. Ternary diagrams represent all possible predictions from each tracer for the selected mixture (see more details about how ternary diagrams work in Lizaga et al. [54]).
Figure 6. Ternary diagram showing all possible contributions of each tracer mix-sample 1, with blue dots representing results of the simple tracer model. Ternary diagrams represent all possible predictions from each tracer for the selected mixture (see more details about how ternary diagrams work in Lizaga et al. [54]).
Sustainability 15 11856 g006
Sustainability 15 11856 g007
Figure 7. Ternary diagram showing all possible contributions of each tracer mix-sample 2, with blue dots representing results of the simple tracer model. Ternary diagrams represent all possible predictions from each tracer for the selected mixture (see more details about how ternary diagrams work in Lizaga et al. [54]).
Figure 7. Ternary diagram showing all possible contributions of each tracer mix-sample 2, with blue dots representing results of the simple tracer model. Ternary diagrams represent all possible predictions from each tracer for the selected mixture (see more details about how ternary diagrams work in Lizaga et al. [54]).
Sustainability 15 11856 g007
Sustainability 15 11856 g008
Figure 8. (A) Results from the sediment provenance model applied to the Ocoña watershed using FingerPro, an R package, with the consistent tracer selection (CTS) method for mix-sample 1. (B) Violin plot showing sediment provenance results, including upper (red), middle (blue), and lower (green) sources.
Figure 8. (A) Results from the sediment provenance model applied to the Ocoña watershed using FingerPro, an R package, with the consistent tracer selection (CTS) method for mix-sample 1. (B) Violin plot showing sediment provenance results, including upper (red), middle (blue), and lower (green) sources.
Sustainability 15 11856 g008
Sustainability 15 11856 g009
Figure 9. (A) Results from the sediment provenance model applied to the Ocoña watershed using FingerPro, an R package, with the consistent tracer selection (CTS) method for mix-sample 2. (B) Violin plot showing sediment provenance results, including upper (red), middle (blue), and lower (green) sources.
Figure 9. (A) Results from the sediment provenance model applied to the Ocoña watershed using FingerPro, an R package, with the consistent tracer selection (CTS) method for mix-sample 2. (B) Violin plot showing sediment provenance results, including upper (red), middle (blue), and lower (green) sources.
Sustainability 15 11856 g009
Table 1. Tracer selection showing the main tracers of the database and indicating results from (A) mix-sample 1 and (B) mix-sample 2, with their respective consensus ranking (CR, %) and conservativeness index (CI). Orange color indicates that the parameters were met, whereas blue color means that parameters were not met.
Table 1. Tracer selection showing the main tracers of the database and indicating results from (A) mix-sample 1 and (B) mix-sample 2, with their respective consensus ranking (CR, %) and conservativeness index (CI). Orange color indicates that the parameters were met, whereas blue color means that parameters were not met.
(A) Mix-Sample 1 (B) Mix-Sample 2
AnalyteCRCIAnalyteCRCI
Co97.90−0.03Cu99.400.00
Sc97.600.00Be98.600.00
As97.500.00Pb97.70−0.02
Cu97.20−0.02V96.600.00
Be97.100.00K94.40−0.08
V94.50−0.10Al93.400.00
La92.30−0.04Fe81.300.00
K88.20−0.20Ag80.60−0.66
Mg87.20−0.12Cd69.00−0.09
Ag82.90−0.66Co68.00−0.31
Cr82.00−0.18As58.00−0.70
Ti78.30−0.22Au57.80−0.10
Cd73.30−0.09Tl47.70−0.68
Fe71.80−0.61Mn45.50−0.63
Al66.90−0.58W42.80−0.41
Au63.50−0.10Sr39.80−0.29
Zr54.90−0.39Hg35.10−0.18
Pb51.20−0.81Zr35.00−0.48
Mn51.20−0.59Mo33.90−0.81
W46.20−0.41Cr30.00−1.49
Hg36.50−0.18Zn26.90−1.55
Zn31.60−2.02Sc21.00−1.05
Mo31.50−0.81Ca17.30−0.93
Sn19.70−2.85Sn15.30−2.85
Sr3.30−6.80La7.20−1.58
Ni3.00−2.28Mg4.50−0.83
Sb1.90−11.46Sb3.90−11.46
Ca1.70−1.25Bi1.70−10.51
Y1.60−3.51Ni0.20−6.13
Bi0.20−10.51Na0.00−8.66
Na0.00−7.70P0.00−2.60
P0.00−1.93Y0.00−10.08
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Crespo, J.; Holley, E.; Guillen, M.; Lizaga, I.; Ticona, S.; Simon, I.; Garcia-Chevesich, P.A.; Martínez, G. Tracking Sediment Provenance Applying a Linear Mixing Model Approach Using R’s FingerPro Package, in the Mining-Influenced Ocoña Watershed, Southern Peru. Sustainability 2023, 15, 11856. https://doi.org/10.3390/su151511856

AMA Style

Crespo J, Holley E, Guillen M, Lizaga I, Ticona S, Simon I, Garcia-Chevesich PA, Martínez G. Tracking Sediment Provenance Applying a Linear Mixing Model Approach Using R’s FingerPro Package, in the Mining-Influenced Ocoña Watershed, Southern Peru. Sustainability. 2023; 15(15):11856. https://doi.org/10.3390/su151511856

Chicago/Turabian Style

Crespo, Jorge, Elizabeth Holley, Madeleine Guillen, Ivan Lizaga, Sergio Ticona, Isaac Simon, Pablo A. Garcia-Chevesich, and Gisella Martínez. 2023. "Tracking Sediment Provenance Applying a Linear Mixing Model Approach Using R’s FingerPro Package, in the Mining-Influenced Ocoña Watershed, Southern Peru" Sustainability 15, no. 15: 11856. https://doi.org/10.3390/su151511856

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

Article Metrics

Back to TopTop