Without Land and Water in Wallmapu (Ancestral Mapuche Territory): Exploring Causal Relationships and Perceptions in the Chol-Chol Watershed, La Araucanía, Chile

3.1. Models and Principal Components (PCs) One of the models with a poor observed fit is CONADI land (Table 3). This algorithm is associated, in terms of statistical significance, with principal components 2 and 3, particularly the variables “land demanded by Mapuche communities” and “forestry plantations”. This relationship suggests that CONADI acquired part of the Mapuche land in the basin and that this land had forestry plantations when it was acquired, between 1997 and 2013. It can also be inferred that these purchases have been in areas with low arable land availability. The lack of quality farmland and its market value has obligated CONADI to acquire properties with smaller surface areas for agricultural use and greater surface areas suitable for forestry. The arable Mapuche land model is linked to two statistically significant components (Table 3). One of them is associated with the decrease in secondary native vegetation in various parts of the basin between 1997 and 2003, a process that has allowed land to be made available for agricultural production. This model also reflects the decreased forestry plantations in various parts of the basin. This algorithm is associated with components 1 and 2. The main relationship is that fewer water rights have been legally established in areas with a greater concentration of indigenous land. Water in the basin is concentrated among non-Mapuche agricultural and forestry operations located on the best agricultural land toward the west of the basin (Figure 2). However, a water rights concentration is also observed within Mapuche land, especially in agricultural areas and those with arable land located near the cities of Nueva Imperial, Chol-Chol, and Temuco. This model shows that Mapuche land has more soil dedicated to crops compared to other uses such as native vegetation or forestry plantations. Basically, but not exclusively, the Mapuche land in the basin is used for agriculture. The total population model is explained by components 1, 2, and 4. The relationships of interest: as the concentration of the Mapuche and non-Mapuche population increases, the farmland area of the properties increases, as does the intensity of land-use changes from non-agricultural to agricultural uses between 1997 and 2013. The model positively associates population with greater land-use pressure. Inversely, it links population and land-use changes with a decrease in native vegetation, especially secondary vegetation. The clearing of native vegetation to prepare land for agriculture or livestock is a possible explanation. This model also associates the “population” variable with water rights granted; in areas of the basin with a greater Mapuche population, more water rights (number of petitions and flows) have been granted to non-indigenous people, including mid-size and large farmers, as well as forestry companies. This demonstrates a strong process of water resources concentration, especially in the west and south of the Chol-Chol River basin (Figure 2). Table 3 presents the results of the regression model. Table 4, as an example, shows the original variables and the values of the coefficients of the PCA 1 and 2 of the Mapuche Land model (Mapuche Land = 35.56 + 5.85CP1 − 6.40CP2 + 0.22CP3 − 0.15CP4). In Figure 3, component 1 is represented on the ordinate axis and component 2 on the abscissa axis, with the percentage of explained variance. Both components explain more than 50% of the observed variability. The longer vectors provide greater weight in the explanation. The red points show the spatial relationship of the sub-basins with the original variables and allow them to be grouped. It bears mentioning that in the Chol-Chol River basin, there are 25,311 Mapuche farms, of which only 442 have registered water rights (1.7%), with a theoretical availability of 2.8 L/s [21]. In Chile, only 2.2% of water is being used by indigenous populations under the current legal regime of water use ([35]; Water Code, 1981). On average, each Mapuche in the Chol-Chol River basin has 2.8 L per second, while a non-Mapuche has 31.2 L per second, an amount that allows production based on traditional annual crops, as well as market garden products and even conversion to other more profitable uses such as small orchards. In the northern zone, most Mapuche farms do not have water rights, a situation that prevents them from applying for government subsidies such as irrigation projects or Rural Drinking Water (APR) projects. Many of the water rights granted to non-indigenous people (forestry companies) are not used by their owners, generating a speculative process associated with the “water market” in the Chol-Chol watershed. Therefore, it is paradoxical that water is distributed using tanker trucks in areas with established rights. Water rights continue to be established and concentrated among mid-sized agricultural properties and forestry companies, increasingly limiting access by Mapuche people and communities [56]. In order to access water a strategy has been observed. An ecclesiastical institution in the Araucanía Region that has registered rights to water in the Chol-Chol river, was “visionary and protected, so to speak, this area by registering water rights”, adding that this organization “cedes rights to indigenous committees and communities for APRs. It’s an important actor for building APRs, I mean they start looking for all those committees that have water, that take water from the Chol, its practically all of them” (E2-T, 2017). His opinion is categorical: “they are important, they extract 2000 and 3000 liters per second over there at one point in the river, they’re giving it for free, they’re giving it to the communities, for irrigation, to the APRs”. However, the government still has a significant water reserve in the Chol-Chol River watershed, which has been estimated at more than 11,000 L of water per second [21]. These resources could be redistributed among Mapuche people and communities in the northern part of the watershed, thereby reversing the major water concentration process observed and allowing productive development based on land and water as primary sociocultural axes. However, assuming that there will be a more equitable future water distribution as a public policy standard, it asks if Mapuche communities have appropriate land, arable land, for more intensive irrigation. It should be noted that the demand for water has been constantly increasing in the Chol-Chol watershed [41], especially since the 1980s. This is reflected in a continuous increase in requests for rights and the amount of water granted, which peaked in 2016 with more than 500 requests and 440,000 L/s delivered. Only from 2018 onwards has a decrease with 36 requests and 112.2 L/s granted by the DGA. 3.2. Modeling and Water Scarcity Of the set of analyzed variables and defined principal components, a group of three models with the best goodness of fit, statistically significant with R²_ajust values over 0.85 (Table 5), was selected. Figure 4 shows the relationship between principal component 3 and the “forestry plantations” response variable for the For2013 regression model, and Figure 5 shows the principal components of the model For2013, Renob2013, and Agro 2013. The forestry plantations model is explained by the first 3 components. Component 1 is inversely related to the presence of secondary native forest (1997 and 2013) and directly related to the number of people supplied with drinking water by tanker trucks, the total Mapuche population, and Mapuche land (Figure 6). The second component presents a direct relationship with the “Mapuche land acquired by CONADI” variable and an inverse relationship with water rights granted to Mapuche. The interpretation of these relationships indicates a replacement of native vegetation with fast-growing introduced species, mainly pine and eucalyptus. These are forestry plantations of large companies and small areas of Mapuche community members (Figure 4). Other relationships indicate that the properties acquired by CONADI present significant forested areas in zones of the basin where few water rights have been granted to Mapuche, especially in the drier northern part, where there is less availability of arable land and a geographical configuration of low ridges and hills with steep slopes (Figure 5). According to a Traiguén municipal professional, the problem is also related to land purchases. In his judgment, CONADI “buys land and just tosses it there”, referring to Mapuche communities that benefit from government subsidies, adding that: “they should have bought some with water rights, for irrigation at least, they just sold them land” (E3-Tr, 2017). Native vegetation dynamics were measured via the secondary native forests 2013 variable and are explained by all the components. They are related to land used for agriculture (1997 and 2013), land-use changes between agriculture and other uses, and water rights granted to the non-Mapuche population. Between 1997 and 2013, land with secondary native forests in the basin increased in area due to a succession of land-use changes related, in the first stage, to the cleaning of primary or secondary native forests to prepare land for farming or agriculture. Then this land enters a fallow stage, allowing the recovery of native shrub or tree cover. The third component relates the presence of secondary native forests to the locations of Rural Drinking Water (APR) projects. In areas with greater secondary native forest cover, consumption of drinking water provided by APRs is greater compared to consumption from APRs located in zones with other land uses such as agriculture or forestry plantations. Corroborating this aspect, a municipal official of Traiguén, a commune located in the northern part of the basin, states, “that part, that part where the Rehue River rises there and has a lot of native forest, that area doesn’t have water problems” (E3-Tr, 2017). The Agro2013 model indicates relationships among the presence of secondary native forest, the population supplied with drinking water by tanker trucks, the total Mapuche population, and Mapuche land. Other important variables are arable land per capita, consumption of drinking water by the population supplied by APRs, and arable Mapuche land. The most significant relationship is the increase in farmland associated with the decrease in primary and secondary native forests in different parts of the basin. Meanwhile, the increase in farmland in the Chol-Chol and Nueva Imperial communes is associated with the “total population supplied by tanker trucks” variable; that is, there is more water delivery in homes located in agricultural areas, especially in agricultural areas where Mapuche communities predominate. This model also links lower native vegetation cover in the southeastern part of the basin to agricultural uses on arable land, which is of higher quality than that located in the north, and to lower drinking water consumption. Table 6 shows the original variables and the models with the greatest explanatory weight in the different sub-basins of the Chol-Chol watershed. In sub-basins with larger areas subjected to land-use change, a greater decrease in secondary native forest cover and, in some cases, elimination of primary native forest fragments or remnants located in ravines is observed. In the case of Mapuche farms, especially in the northern area, the native vegetation on the properties was replaced by forestry plantations, decreasing the amount of arable land used for crops. This dynamic has caused better-quality land in the communities to become increasingly scarce, a historical process that has been exacerbated by the lack of water in recent decades. 3.3. The Mapuche Lonko Perception A lonko from a Galvarino community states that climate variations also explain the lack of water: “nature in Chile has changed a lot, because that has also affected the communities a lot”. He remembers the climate events of the past and relates them to the geographic conditions of the area: “it has changed a lot because before the experience I had between now and 20 years ago, man, it rained a for week, 15 days, and today the most it rains in a year, about 20 days, multiplying by 5 days for one month, 5 for another month, it could be less, it’s a failure, also, that has harmed practically in these highland areas, like this one” (E1-G, 2017). A lonko from Traiguén explained another: “gradually, in some places, especially where people planted pine and eucalyptus trees, around there the water problem began”. He adds: “because many times, here they thought up, you see eucalyptus here, one hectare, two hectares, but it was CONAF that came with the big lie, saying, in 12 years you’re going to be, man, almost a millionaire” (E4-Tr, 2017). For this interviewee, whose lonko led an important process of territorial recovery of more than 3000 hectares, a large part of which were planted with radiata pine: “there are certain spaces that we will not be able to touch because they have another meaning for us, a meaning beyond that, a spiritual meaning, because I cannot exploit them there, because there are lawenes there, remedies that serve us for the Machi” (E4-Tr, 2017). He acknowledges that plantations affect the availability of water and that: “if there is no water, we will not have any, we will not be able to produce our traditional crops, for example, today you cannot grow a vegetable garden because, with that amount of water they give you, you cannot produce a vegetable garden”. In his opinion, “the fight against forestry companies begins because spaces begin to shrink, families begin to grow and the land does not grow any further”. In many communities, forestation was carried out with government support to generate medium-term earnings, without considering the effects these plantations could have on future water availability. This situation is acknowledged by a lonko of Galvarino, who states: “there was an error here, a human, personal error, and also by the professionals we have, who were aware, it was known that in the future this would be detrimental to the communities” (E1-G, 2017), adding that: “Now it is extremely dry and that has also affected, I think because of the system of Law 701 that I think the Chilean government, to be able to forest private and non-private land…that has made us lose the groundwater of the menoko [waterhole, spring, or swampy site with abundant vegetation; preferred Mapuche space for the collection of medicinal plants and ceremonies]” (E1-G, 2017). The interviewee justifies the Mapuche territorial recovery processes by pointing out that: “So, I want to go in there to recover to feed ourselves, the only mission we have is to work a little on the agricultural part, work on wheat and oats, pasture because we cannot do part of the vegetables because there is no water” (E1-G, 2017). He also recognizes the need for the State to build hydraulic works, such as rainwater accumulation dams and APRs. For the lonkos, the breakup of the menokos is occurring in the watershed [31] This is an event in which a micro-watershed and the native vegetation remnants that protect them are dried out as a result of the extraction of water by exotic plants, such as pine and eucalyptus trees. Their territory is suffering a disease; it is kutrhan (“sick” in Mapudungung, the Mapuche language). Water and land scarcity, sustained migration, cultural resistance, increasing precariousness of life, forced adaptation, and uprooting are some of the symptoms. 3.4. The Forestry Companies Arguments The forestry companies attribute the lack of water to climate change and the megadrought [57,58,59]. On this topic, CORMA (Chilean Wood Corporation) has expressed that “although plantations use water to grow and produce wood, they are not the cause of the reduction in river and stream levels. The main cause is the precipitation deficit due to climate change, as well as the increase in water demand for other uses” [58]. Meanwhile, the drying up of wells for human consumption is due to the fact that “higher average temperatures in the mountains produce less snow accumulation on the high peaks, affecting aquifer recharge in the spring and summer months” [59]. Lower precipitation due to the long drought (cf. Center for Climate and Resilience Research, 2015) and greater water demand for agricultural uses are the main explanatory factors [58]. Forestry corporations present their arguments and evidence that extenuate or disavow their responsibility for the water and territorial crisis. They use scientific information to justify their policies and actions in the court of public opinion [57,58,59]. 3.5. The Perspective of the Municipalities According to a Chol-Chol official, the lack of water is associated with the advance of forestry plantations and the decrease in native vegetation: “if you talk with an adult, the adult tells you, man, before when I went out to work, I went out with a jug and a little flour and went out to work in the countryside, you found water everywhere, you can’t do that now”. As a time reference for the forestation process the official states that “we’re talking many years ago, more than 30, well the previous settlers arrived many years before, but the forestry company got here like 30, 40 years, yeah 40, about 40 years ago they got here, then they planted and near the trayenko [waterfall, water that emerges from the earth, or place where two or more streams meet, where prayer ceremonies are held; the vegetation includes herbaceous plants, shrubs and ferns], which is what we call it” (E5-CH, 2017). The accounts report a decrease in native forest over the years; however, remnants remain in ravines and along stream and river banks, located on forestry company and Mapuche properties. The advance of forestry plantations, including over land suitable for agriculture, has been an important factor, as he states: “they surrounded us and the flat parts that had all been planted with wheat. Now all forestry” (E8-V, 2017). Providing more background, he mentions that, years ago, “water was at 2 meters, but now they dug the latest wells for water for human consumption in a community, they reached 27 meters in deep wells”. He also says that geological characteristics can have an influence, as “with the earthquake more droughts occurred, the groundwater escaped” (E8-V, 2017). The overexploitation of APRs is another reason for the lack of water, due to new household connections for families that acquire land or are resettled in the area as a result of CONADI purchases. Water demand in some areas (mainly in the north of the watershed) has exceeded the supply, and “the well they have isn’t able to supply its own users, there will come a moment they won’t be enough for the current user because the share has been decreasing” (E9-P, 2017). In this case, he states, the drinking water committee that manages the APR has taken some measures: “now they’re buying 5 liters per second, from an irrigation canal”. According to an official from the municipality of Temuco, the greater demand for water and APR projects is due to the fact that “I don’t know if it is so, so effective, but they’ve bought a lot of land in this area and CONADI has brought Mapuche communities from other areas”, adding an interesting perspective, already expressed by other interviewees: “one look there, and the committees are always there, I mean, the communities always, it’s something that they always have unconsciously now, the forestry companies are the ones stealing the water”. In this person’s opinion, it is not only a physical lack of water or technical problems associated with prospecting for it, but also that Mapuche people and communities do not have registered rights; that is, they do not have ownership of it and control of its use. The geographic configuration has been a factor that has influenced public policy results in terms of drinking water provision in the Traiguén Commune, explaining: “there are lots of families who don’t see its geographic use, I mean they build a house on a hill where they are never going to get water…we, because, because of the geography we have more trucks, because of the geographic aspect ONEMI (National Emergency Office) gave us more trucks”. Some professionals interviewed hold forestry companies 100% responsible for the lack of water “before 1100 fell here and now I think 800, 700 millimeters of water fall”. For this official, there are other social impacts more important that the lack of water, related to land sales to forestry companies: “people came to town, bought pick-up trucks, they hauled freight, the truck broke down two years later, they begin to ask for welfare benefits, they live on a lot”. Furthermore, he adds that there are older people who have not been able to adapt to a new way of life: “there are farmers who have committed suicide because they leave about 20 hectares that they lived on, to a 10-by-20-meter lot, the whole system changes on them” (E7-NI, 2017). His knowledge and experience with Mapuche communities allow him to assert that “I think that where there is a problem is when there is a community, let’s say, there are a thousand hectares of communities against ten thousand of plantations around them, they’re surrounded, that happens in Lumaco, in other areas, but here fortunately not”.