Next Article in Journal
Communicating Sustainable Brand Equity in a High Carbon Footprint and High-Risk Sector: Comparing Malaysia and Indonesia Oil and Gas Industry
Previous Article in Journal
The Characteristics and Seepage Stability Analysis of Toppling-Sliding Failure under Rainfall
 
 
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 10
10.3390/su15107737
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

Predicting Climate Change Impacts on Candelilla (Euphorbia antisyphilitica Zucc.) for Mexico: An Approach for Mexico’s Primary Harvest Area

by
Aldo Rafael Martínez-Sifuentes
1,
Juan Estrada-Ávalos
1,*,
Ramón Trucíos-Caciano
1,
José Villanueva-Díaz
1,
Nuria Aidé López-Hernández
1 and
Juan de Dios López-Favela
2
1
Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias, CENID-RASPA, Gómez Palacio 35150, Mexico
2
Facultad de Agricultura y Zootecnia, Universidad Juarez del Estado de Durango, Ejido Venecia, Gómez Palacio 35150, Mexico
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(10), 7737; https://0-doi-org.brum.beds.ac.uk/10.3390/su15107737
Submission received: 6 January 2023 / Revised: 2 May 2023 / Accepted: 3 May 2023 / Published: 9 May 2023
(This article belongs to the Section Sustainability in Geographic Science)
Browse Figures
Review Reports Versions Notes

Abstract

:
Candelilla (Euphorbia antisyphilitica Zucc.) is a non-timber forest resource of ecological and economic importance in the arid zones of Mexico due to the commercialization of its wax for industrial purposes. The objectives of this study were (i) to delimit areas of current and projected future candelilla habitat suitability in Mexico and in the state of Coahuila, (ii) to determine the most important variables that define candelilla habitat, and (iii) to propose areas for candelilla conservation under climate change conditions in Coahuila. Records of candelilla presence, current and future bioclimatic layers from the MPIESM-LR and HadGEM2-ES models with two scenarios RCP 4.5 and 8.5, were used to create species distribution models with soil and topographical variables. MaxEnt software was used to project current habitat suitability zones under climate change. We estimated the current surface area of candelilla in Mexico to be 79,336.87 km2, and for Coahuila 25,620.75 km2. In Coahuila, using the MPIESM-LR model for 2050, the estimate was 20,177.67 km2 and 17,079.61 km2 for RCP scenarios 4.5 and 8.5; while for 2070, the estimate was 12,487.18 km2 and 9812.94 km2 for RCP scenarios 4.5 and 8.5. For the HadGEM2-ES model for 2050, the estimate was 20,066.40 km2 and 17,079.61 km2; for 2070 it was 17,156.02 km2 and 16,073.70 km2. As proposed areas for conservation of candelilla in the face of climate change, we estimated 5435.06 km2 and 3636.96 km2. The study area was located in the northwest and center of the state of Coahuila, near the natural protected areas of Ocampo and Bajo Rio San Juan, areas that are resilient to climate change. The results obtained provide information on the environmental and site conditions for the establishment of candelilla in Mexico, as well as the geographical areas, such as Sierra y Cañon de Jimulco, Tomás Garrido, 026 Bajo Río San Juan, Zapalinamé, Zapalinamé, and Cumbres de Monterrey Restoration Zones for the conservation of the species under local climate change scenarios. In addition, new areas in the northwest and center of Coahuila could be used to establish new protected areas for this economically important species.

1. Introduction

Mexico is a country rich in forest ecosystems with 137 million ha, which is equivalent to 70% of the national territory [1]. The northern portion of Mexico is an area with large expanses of arid and semi-arid zones, which represent more than half of the Mexican territory; these areas contain various types of shrub communities that are generically called xerophytic scrub, with grasslands and some areas with trees [2]. Arid and semi-arid zones have a great variety of biodiversity, including Agave spp., Larrea tridentata, Prosopis laevigata, Olneya tesota, Carnegiea gigantea, and Lippia gareolens; however, not all of these resources have the potential to be commercially exploited.
Due to the problem of unsustainable harvesting of non-timber forest resources, a large number of the human population has migrated from rural to urban areas. People who remained in their places of origin have used natural resources, such as candelilla (Euphorbia antisyphilitica Zucc.) for their subsistence. Candelilla is a non-timber forest resource of great importance in Mexico since the wax from this plant is one of the ten most economically important forest products in the country, and its collection represents an activity that generates income for many families living in communities in arid and semi-arid zones [3]. Candelilla wax generated profits of slightly over USD $4.9 million as a result of selling 1500 tons in 2018 [4].
Candelilla is a shrub with a distribution from the southern United States to central Mexico. It belongs to the Euphorbiaceae family, which includes five subfamilies, 317 genera, and approximately 2000 species [5]. The species has fibrous and gray-brownish roots, its stems are both aerial and underground, and it can reach a height of up to 1.3 m and a diameter of 5.0 mm. It contains green dye when it is young and turns glaucous green at maturity as a product of the wax layer that covers it. The small oblong leaves are deciduous and only remain on the plants for 15–20 days [6].
This species is characterized by the production of solid wax, which is of great economic value for the communities where it is extracted [7]. Candelilla wax is produced as a secondary metabolite that is part of an adaptation to water stress. This adaptation allows the species to survive the typical climate of the Chihuahuan Desert, which presents dry and cold winter conditions, and extremely hot summers [8].
The extraction of this resource is not sustainable by the local people, who generally sell the wax at a low price that does not make up for the damage caused by harvesting [9]. Intensive exploitation of candelilla have reduced its distribution and the number of populations, both of which have contributed to soil loss, a significant decrease in its productivity, and accelerated the desertification process in the region [10]. The cost of mismanagement of this resource has altered the region’s hydrological cycle, increased soil salinity, decreased productivity, and contributed to the loss of its biological diversity. In light of the unsustainable harvest of candelilla and the regional damage it causes, Mexico will need an investment of more than USD $9 million to recover its arid zones [11].
The state of Coahuila is the leader in candelilla wax harvesting and processing in the world due to its desert climate. Conservation of candelilla is economically and environmentally important. It is included in the list of species protected by the Convention on International Trade in Endangered Species of Wild Fauna and Flora since 1975 [12], and the state of Coahuila generates approximately 60% of the annual national production, followed by Durango (22%), Zacatecas (8%), Chihuahua (6%), and Nuevo León (4%) [4].
Habitat suitability and potential species distribution models allow for the establishment of methodologies for large-scale ecological and geographical studies [13]. Research is required to provide detailed projections of species’ distribution based on presence or absence using continuous or categorical environmental variables, and estimates of changes in habitat suitability over time, through specific environmental change scenarios [14].
There are several methods for modeling species’ distribution and habitat suitability, such as Markov random fields, mixed models, and logistic regression models. One of the most widely used by the scientific community is the maximum entropy approach (MaxEnt) [15]. Modeling with MaxEnt provides a probabilistic interpretation through presence records of the species and important variables for its development [16].
General circulation models (GCMs), use scenarios called representative concentration pathways (RCPs) to project possible changes in global climate [17]. The scenarios of GCMs in combination with ecological niche models provide an index of habitat suitable for the subsistence of species in the future. These maps supply relevant information for conservation and restoration programs [18].
The main objective of this study was to model and determine suitable areas for future conservation of candelilla in Coahuila. The specific objectives of this research were (i) to model the present habitat suitability of candelilla in Mexico, (ii) to identify the most relevant environmental variables according to its current and projected future habitat suitability (2050 and 2070) using the MPIESM-LR and HadGEM2-ES models under RCP 4.5 and 8.5 scenarios, and (iii) to suggest conservation areas for Mexico’s current main harvesting area (state of Coahuila).

2. Materials and Methods

2.1. Study Area

The present study was performed in the Chihuahuan Desert of Mexico. Additionally, other areas were considered, based on the presence of species records and their biological and dispersal characteristics, such as the states of Puebla, Hidalgo, and Oaxaca. This zone represents the model area (M), according to the BAM diagram [19], where the climate is generally dry (total annual precipitation of 400 mm) and mean annual high temperatures range from 18 to 22 °C [20].

2.2. Geographical Records

A database with 1130 records was generated from three different sources: Georeferenced points obtained from field visits, the Global Biodiversity Information Facility database [21], and the national forest inventory [22]. The database was cleaned-up to eliminate duplicate records and present one record per linear km2 using NicheToolBox program of the Comisión Nacional para el Conocimiento y uso de la Biodiversidad [23]. We used the clean-up process to avoid the effect of spatial autocorrelation, and thus an overestimation in the habitat suitability models [24]. Overall, 639 spatial records of candelilla were considered for modeling after the clean-up process (Figure 1).

2.3. Current and Future Climate Variables

The database with climate variable information was generated from the 19 layers in the WorldClim database, version 2.0, at a scale of 30″ × 30″ (~1 km2), which represented average climate information from 1970 to 2000 [25]. For spatial modeling of future habitat suitability, we used the Coupled Model Intercomparison Project Phase 5 (CMIP-5) GCMs MPIESM-LR and HadGEM2-ES [26,27]. These models are two of the most recently used models for Mexico and recommended by the National Institute of Ecology and Climate Change, since they represent more accurately the climate variability in Mexico [28]. Two climate scenarios were employed: RCP 4.5 and 8.5 for the periods 2041–2060 (2050) and 2061–2080 (2070), both with a spatial resolution of 30″ × 30″ (~1 km2).

2.4. Edaphological and Topographical Variables

The database with information on edaphological variables was generated by extracting the data of interest from the SoilGrids website at a scale of 250 m per pixel [29]. The variables considered were volumetric fragments (FRA), density (DEN), pH, and depth to bedrock (DEP), which were rescaled to 1 km2 per pixel. The topographical variables were obtained from the digital elevation model (DEM) of the Mexican elevation continuum 3.0 with 30 × 30 m per pixel [30]. The DEM information was used to obtain the elevation variable (ELE) and transformed from 90 m2 to 30 arc-sec (~1 km2) spatial resolution to standardize raster layers. Then, the slope (SLO) and orientation (ORI) were generated with Arcmap 10.3 software [31].

2.5. Model Calibration and Variable Selection

A calibration was performed to reduce model complexity through minimum convex polygon (MPC) information and species records using the standardized Akaike AICc information criterion coefficient [14]. This process was completed by selecting the lowest AICc value. The calibration was developed with the ENMevalate function of the ENMeval library implemented using the R programming language [32]. The function iteratively creates ecological niche models through a variety of fit configurations and allows for model evaluation using cross-validation or a fully retained test data set [33]. The function returns evaluation statistics for each combination of configuration and cross-validation fold, as well as raster predictions for each model when raster data are entered [33]. The evaluation statistics, which is returned in a table, allows for the identification of a model configuration that balances fit and predictive ability [33].
The selection of the variables with the greatest influence for the species was performed by generating a MCP from all of the cleaned records. Then, 10,000 background points were placed and the climate information was extracted [34]. To avoid multicollinearity between variables, those with a correlation greater than 0.70 (p < 0.05) were eliminated [35] with Pearson bivariate correlation analysis using the NicheToolBox program [23], considering first the environmental importance of the variable in the suitability of the species, and then the correlation. The selected variables were delimited at the same spatial resolution of 30″ × 30″ (~1 km2) in the modeling area M (Chihuahuan Desert, and Hidalgo, Puebla and Oaxaca states), which consists of the geographical space where the species has been recorded and is delimited according to biological knowledge and its dispersal capacity [19].

2.6. Current and Future Suitability Modeling

We used MaxEnt software (v3.4.1), which is based on the maximum entropy approach [36] and is widely used by the scientific community, to determine the distribution of flora and fauna species with presence records [16]. The variables after clean-up were: Minimum temperature of the coldest month (BIO 6), annual temperature range (BIO 7), average temperature of the driest quarter (BIO 9), average temperature of the coldest quarter (BIO 11), annual precipitation (BIO 12), coarse volumetric fragments (FRA), depth to bedrock (DEP), elevation (ELE), and bulk density (DEN). The MaxEnt configuration criteria were internal replication by cross-validation (1000 iterations), logistic type output (100 replicates), and a convergence threshold of 0.0001, in which 75% of the records were selected for training the model and 25% for validation [37].
To generate future models, calibration parameters and statistics of the best model were transferred to the MaxEnt program, version 3.4.1 [38].
To estimate the area (km2) with current and future habitat suitability with the MPIESM-LR and HadGEM2-ES models, a reclassification of the continuous values of both projections across three categories with equal intervals (low, medium, and high), was developed with Arcmap software, version 10.3 [31]. The previous procedure allowed for the determination of cut-off thresholds and, through the high category, the transformation of the models from continuous to binary [39].

2.7. Validation of Habitat Suitability and Variable Contribution

The models were evaluated through the area under the curve (AUC) tests of the receiver operating characteristic (ROC) analysis. Test results range from 0.0 to 1.0, where values from 0.7 to 0.9 indicate an adequate fit and values > 0.9 are classified as excellent [40].
Currently, the development of algorithms with presence data is questioned since it also requires true absence data; therefore, it weighs omission and commission errors similarly [41]. Given this situation, the models were evaluated in parallel with partial ROC analysis using the Niche ToolBox platform [23]. The recommendations of Peterson and Nakazawa [24] were followed (i.e., 1000 bootstrap replicates with an error of omission of 5%). Additionally, models were evaluated with z-tests (p < 0.01) of the model with the highest partial ROC value and lowest standard error. Moreover, the models were evaluated by the True Skill Statistics (TSS = sensitivity + specificity − 1) [23].
The relative contribution (%) of the variables of the habitat suitability models was determined through the Jacknife test [42]. The contribution of each variable allowed for the determination of the degree of importance of each variable in the generation of the current and future habitat suitability models for candelilla in the state of Coahuila [36].

2.8. Conservation Areas in the Primary Harvest Area

Information was extracted from the current and future climate models for the polygon corresponding to the state of Coahuila, as it is the main producer of candelilla wax in Mexico. The areas that could be destined for candelilla conservation were identified through an intersect between the areas of current and future suitability (2070) with RCP 8.5 for both GCMs. The areas can be established, due to the modeling of current and future conditions admissible for in situ conservation, reforestation, and restoration activities [43]. Additionally, we estimated the proposed areas for candelilla conservation within state and federal natural protected areas [44].

3. Results

3.1. Candelilla Current Habitat Suitability Model

The current habitat suitability models, based on 100 replicates, showed AUC values between 0.921 and 0.939 for training and 0.827 and 0.923 for model validation. The best model presented a partial ROC value of 1.85, with AUC of 0.939 and 0.923 for training and model validation (SD = 0.0001), respectively. The TSS values for all the future models were > 0.80. The estimated area for current growth of candelilla in Mexico was 79,336.87 km2 (Figure 2) and the estimated area for the state of Coahuila was 25,620.75 km2 (Figure 3). The distribution in Coahuila is concentrated in the center and northwest portions of the state, with a smaller amount in the southwest. The estimated area of candelilla in the natural protected areas of the state is presented in Table 1.

3.2. Candelilla Habitat Suitability Modeling under Climate Change

The AUC values for the MPIESM-LR and HadGEM2-ES models, for the periods 2050 and 2070 under the RCP 4.5 and 8.5 scenarios, are presented in Table 2. Partial ROC values for both GCMs and RCP scenarios ranged from 1.80 to 1.82 and were significant when we used the z-test (p < 0.01).

3.3. Important Variables in Current and Future Candelilla Distribution

The most important variables for current and future habitat suitability areas for candelilla in Mexico are found in Table 3.

3.4. Estimated Candelilla Area in Mexico and Coahuila for Future Scenarios

For Mexico, under the most pessimistic scenario RCP 8.5 and the most distant climate horizon 2070, the estimated area of candelilla presence was 32,425.78 km2 (Figure 4). According to the projection, there will be a considerable reduction in the state of Coahuila; however, it will increase for the state of Chihuahua. The percentage of area at the national scale is projected for Chihuahua (60.35%), Coahuila (30%), and Nuevo Leon (7.25%). For Baja California Sur, Durango, Guanajuato, Hidalgo, Puebla, Querétaro, San Luis Potosí, Tamaulipas, Veracruz and Zacatecas, a smaller percentage of 0.60% of the total area projected for the future is expected.
The estimated future area of candelilla with the MPIESM-LR model for the period 2050 was 20,177.67 km2 and 17,079.61 km2 considering RCP 4.5 and 8.5 climate scenarios, respectively. These areas represented 78.75% and 66.66% of the current area of candelilla. The estimated area for the period 2070 with RCP 4.5 was 12,487.18 km2 and for the RCP 8.5 scenario was 9812.94 km2, which represented 48.73% and 38.30% of the current total area, respectively. The reduction in the period 2050–2070 for RCP 4.5 was 5910.38 km2, and for RCP 8.5 was 9479.39 km2 (Figure 5).
For the HadGEM2-ES model, the estimated future area of candelilla for the period 2050 was 20,066.40 km2 and 17,079.61 km2 for the RCP 4.5 and 8.5 scenarios, respectively. Future estimates represented 78.32% and 62.74% of the current area. For the period 2070, the estimated area for RCP 4.5 was 17,156.02 km2 and for RCP 8.5 it was 16,073.70 km2, which represented 55.25% and 25.74% of the current total in Coahuila. The estimated area reduction in the period 2050–2070 was 7690.49 km2 and 7266.67 km2 for RCP 4.5 and RCP 8.5, respectively (Figure 6).

3.5. Proposed Areas for Candelilla Conservation in Coahuila

Considering the MPIESM-LR model, the 2070 climate horizon, and the RCP 8.5 scenario as the period and scenario with the least area of candelilla in the future, an area of 5435.06 km2 was estimated, which represented 21.21% of the current total of area suitable for conservation in the state of Coahuila (Figure 7A). The estimated area of candelilla within the federal and state natural protected areas was estimated at 743.3 km2 (i.e., 2.90% of the current total). The area estimated by the protected area is shown in Table 4.
The areas proposed for candelilla conservation in Coahuila generated through the HadGEM2-ES model, the 2070 climate horizon, and the RCP 8.5 scenario, had an estimated area of 3636.96 km2, which represented 14.19% of the current total (Figure 7B). The area proposed for conservation within the natural protected areas was 764.63 km2, which represented 2.98% of the current total. The area estimated by the protected area is shown in Table 5.

4. Discussion

The total records for model development were 639 after the debugging process, which constitutes the number of records that allow for a correct performance of the model and covers a large area of the Chihuahuan Desert [45]. Previous studies with current candelilla distribution models showed AUC values of up to 0.97 [46], higher than those found in this study (0.939); however, it is classified as excellent according to Lobo et al. [41]. Our results are supported by the partial ROC tests that presented values > 1.8 and Z (p < 0.01), and that classify the models as excellent and reliable [40].
Non-timber forest species in semi-arid areas have been exposed to human exploitation over thousands of years and their distribution depends on environmental and anthropogenic factors [47]. It is difficult to predict the distribution of species due to perturbations of natural or human origin, such as changes in climatic patterns and land use, desertification, and wildfires. However, the models, even when subject to errors and biases, represent a useful tool for the implementation of management plans [48].
In Mexico, several studies have been developed to model the distribution of species under current climate conditions and for climate change scenarios [18,39,49,50]. The reduction in area due to the effect of climate change in candelilla in Mexico presents a similar behavior at the national level [51]. Coahuila is one of the states with the greatest decrease in area [52], a situation that highlights the importance of developing predictive models for the presence of candelilla in the state.
Due to the high demand for organic wax for commercial purposes, the harvest of candelilla has reduced its natural distribution [53]. At the national scale, harvesting has not represented a threatening activity for the species due to its wide distribution. However, without controlled management, harvesting candelilla could lead to a considerable decline in the species [54]. Previous models of candelilla distribution in Mexico have been developed using climatic variables [51,55], or only with edaphological and terrain variables [46].
Under the assumption that Coahuila contains almost 60% of the total national area of candelilla [4], the results of previous studies are higher compared to our current estimated area (25,620.75 km2). This difference can be attributed to the fact that in this study, we used climatic, edaphological, and topographical variables, which are important in the spatial modeling of species [50].
Based on the superposition of layers, the current candelilla model of this study is geographically correctly positioned within the polygon of rosetophytic desert scrub in Coahuila (Figure 8), which in contrast to the study by Vargas-Pineda et al. [51] that visually overestimates its distribution.
The most important variable in the current model of habitat suitability for candelilla was FRA, which represents the percentage of rocks in the soil (fragments > 6 cm in diameter). In the study by Hernández-Herrera et al. [46], the coarse volumetric fragments variable represented 22.3% of the total model (i.e., almost half of the relative importance obtained in this study); however, it represented the third most relevant variable. The maximum percentage of coarse fragments in this study was 63%, a favorable condition for the presence of candelilla, since high stoniness promotes good drainage [6]. High content of stones leads to a reduction in the soil evaporation rate and soil erosion due to the fragility of the upper profile, in addition to the complete extraction of the candelilla plant [56].
The annual precipitation (BIO 12) for the current model presents a mean of 322.5 mm per year. This precipitation rate agrees with the annual requirements for the growth of the species (200–250 mm) in the candelilla zone of Mexico [57]. In the studies of Bañuelos-Revilla et al. [47] and Vargas-Pineda et al. [51], the variable BIO 12 showed 4.7% and 4.5% of the total contribution of the model to predict the distribution of candelilla in Mexico, which was compared with the 21.9% obtained in the present study. The results of the previous studies are lower, but it has been confirmed that the annual precipitation is of importance for the presence of candelilla at the national scale [55].
Shallow soils are the ideal habitat for the natural presence of candelilla. Rojas-Molina et al. [52] and Martínez-Salvador et al. [58] stated that candelilla grows on hillsides with slopes of 1–3% that are rich in calcium carbonates (<25 cm). According to the response curve for this variable, the ideal conditions are in the range of DEP < 50 cm, a situation that promotes rain percolation and water availability for the root whose length reaches 30–60 cm [59]. The calcareous soil can support a great diversity of plants, since the parental material works as a reservoir of humidity and nutrients, where homeostatic processes of ecosystem regulation occur, and areas with low precipitation can present an ideal habitat for the growth of candelilla [60].
The mean temperature of the coldest quarter (BIO 11) showed a mean of 12.8 °C and a contribution of 6.9% of the total variability of the model, in comparison to the study of Hernández-Herrera et al. [55], where the contribution was 0.3%. The variable has a crucial role in the establishment and development of plant species in arid regions. Species distribution studies have placed BIO 11 as one of the most important in the determination of habitat suitability zones for vegetation species of semi-dry and dry ecosystems in Mexico [39]. According to Flores and Jurado [61], the optimum temperature for natural germination of candelilla in winter ranges between 18 °C and 28 °C; a similar range was found in the BIO 11 variable (8.9–16.7 °C) in this study.
The mean temperature of the driest quarter (BIO 9) in the habitat suitability model was 15.0 °C, which represents the months from February through April, where precipitation rates of 13–56 mm have been reported [25]. Candelilla under dry environmental conditions is tolerant of long periods with low precipitation due to its ability to produce secondary metabolites, such as latex, as well as lactiferous compounds that provide water reserves and environmental protection [62].
An increase in temperature as a result of climate change could have a negative impact on the areas occupied by different genera of tree and shrub species in Mexico [63]. In this context, species growing in arid or semi-arid regions, such as candelilla, are likely to better withstand to the above effects and most areas of their current distribution to be preserved over time (niche conservatism theory—Soberón and Miller [64]; Peterson [40], such as the areas in the state of Coahuila). The two chosen models (MPIESM-LR and HadGEM2-ES) have been used in habitat suitability analyses of tree species since they adequately represent the climatic conditions of the study area [43]. These models have been used by INECC to regionalize the future climate considering the topographic conditions of Mexico, which is why they represent the variability of the Mexican Republic [28]. Meanwhile, both models are consistent in their projections for the two scenarios analyzed (RCP 4.5 and RCP 8.5). For the 2050 horizon, the HadGEM2-ES model showed less reduction in both scenarios, in comparison to the MPIESM-LR model. For the 2070 horizon, MPIESM-LR showed less reduction in the estimated area of candelilla for both scenarios than HadGEM2-ES.
However, it is a reality that considerable reductions in the candelilla area are expected as a result of climate change (e.g., increases in greenhouse gas emissions in the future) [65]. Studies of habitat suitability and potential distribution of candelilla in Mexico for climate change scenarios are limited. However, in the study by Vargas-Pineda et al. [51], the HadGEM2-ES model presented increases in area from RCP 4.5 to 8.5 scenario, which is contrary to that found in this study for the state of Coahuila. On the other hand, in the study by the same author, the MPIESM-LR model presented a reduction in estimated candelilla area from RCP 4.5 to 8.5, as was obtained in this analysis. Other studies have generated the candelilla distribution with the GFDL-CM3 model, which is considered as a conservative model, in comparison to the HadGEM2-ES and MPIESM-LR models for the RCP 4.5 and 8.5 scenarios [51]. Moreover, other studies have modeled the distribution of shrub species with the HadCM3 and CGCM2 models, with good results for the central region of Mexico [66].
Research on conservation of the ecological niche suggests that in order to preserve their niche, species can adapt to the effects of climate change over time or move to colonize new geographical areas with similar conditions to their original niche [40]. The realized niche and part of the fundamental niche can be measured through species distribution modeling and the use of simulation models to understand how climate change will affect their distribution in the future [67]. Ecological niche models estimate a probabilistic index of environmental suitability over large regions, based on the environmental variables of the sites where the species currently occurs. This index provides information on the components of the niche, which applies the theory of the niche as a multidimensional hypervolume [68] (i.e., a theory that states most environmental components are preserved at different temporal scales; Peterson [40]).
However, these kinds of studies are very limited in Mexico, and the few that have been developed are focused on conifers under their natural distribution [39,63], and which occur in natural protected areas [43,69]. To conserve candelilla populations in the primary harvest areas in Mexico, this study has identified areas that may represent ideal conditions for the subsistence of the species in the context of climate change. Based on the perspective of the two GCMs studied, it is important to monitor in detail the natural protected areas with the presence of candelilla. Under this approach for timber species, Manzanilla-Quiñones et al. [50] and Martínez-Sifuentes et al. [43] proposed niche conservation areas in natural populations of Pinus montezumae in the Sierra Madre del Sur, and Taxodium mucronatum Ten. in Puebla, Chiapas and State of Mexico in the protected areas Cuencas de los Ríos Valle de Bravo, Malacatepec, Tilostco and Temascaltepec, La Sepultura, Presa de Silva and its surrounding areas, among others. Coahuila has thirteen areas classified as natural protected areas, and only six of them, under climate change conditions, will conserve > 50% of their surface, in comparison to the current area under the most conservative projection (MPIESM-LR model horizon 2070, RCP 8.5). These zones are Sierra y Cañon de Jimulco, Tomás Garrido, 026 Bajo Río San Juan, Zapalinamé, Zapalinamé and Cumbres de Monterrey Restoration Zones. However, the protected areas where the highest impacts projected are Mapimí, Maderas del Carmen, Cuatrociénegas, and Ocampo. Comisión Nacional Forestal [59], report that Ocampo, Cuatrocíenegas, and Sierra Mojada in Coahuila are the municipalities with the highest harvests of candelilla by small wax producers. Outside of the protected areas, the northwest and center of the state of Coahuila have areas with ideal characteristics for preservation.

5. Conclusions

The current and future suitability models presented good results, where coarse volumetric fragments (stoniness), annual precipitation, depth to bedrock, average temperature of the coldest quarter, temperature of the driest quarter, and elevation were the most relevant variables that contribute to the presence of candelilla in Mexico. In this study, suitable areas were identified for implementing candelilla conservation strategies in Coahuila. These areas are located in the 13 natural protected areas, where the species is projected to remain and can continue to be conserved due to similar environmental conditions. Two main areas have also been identified in the northwest and center of the state outside of any protected area, where the presence of candelilla will be feasible under climate change conditions and where it will be important to legally protect these areas and implement actions that allow for both the harvest and conservation of candelilla in the future. The maps generated from this study will allow for a more efficient management of candelilla at the regional level by decision-makers of the institutions in charge of addressing environmental risk problems in Mexico. Although the study made it possible to identify suitable areas for candelilla in Mexico, it is important to direct efforts to validate these sites, and thus corroborate that the species is indeed found in order to calibrate the current climate model in situ.

Author Contributions

Conceptualization, J.E.-Á. and A.R.M.-S.; methodology, A.R.M.-S.; software, A.R.M.-S. and J.d.D.L.-F.; validation, A.R.M.-S. and J.d.D.L.-F.; formal analysis, A.R.M.-S. and J.d.D.L.-F.; investigation, A.R.M.-S. and J.d.D.L.-F.; data curation, J.d.D.L.-F.; writing—original draft preparation, A.R.M.-S.; writing—review and editing, J.V.-D., R.T.-C. and N.A.L.-H.; supervision, J.E.-Á. and J.V.-D.; project administration, J.V.-D. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Comisión Nacional Forestal. El Sector Forestal Mexicano en Cifras 2019. Bosques Para el Bienestar Social y Climático. 2020. Available online: http://www.conafor.gob.mx:8080/documentos/docs/1/7749El%20Sector%20Forestal%20Mexicano%20en%20Cifras%202019.pdf (accessed on 12 July 2022).
  2. Rzedowski, J. Diversidad y orígenes de la flora fanerogámica de México. Rev. Cienc. 1978, 6, 47–56. [Google Scholar] [CrossRef]
  3. Cervantes-Ramírez, M.C. Plantas de importancia económica en las zonas áridas y semiáridas de México. In Temas Selectos de Geografía de México; Universidad Nacional Autónoma de México: Ciudad de México, México, 2002. [Google Scholar]
  4. Anuario Estadístico de la Producción Forestal. Secretaría de Medio Ambiente y Recursos Naturales; Anuario Estadístico de la Producción Forestal: Mexico City, Mexico, 2019. [Google Scholar]
  5. Webster, G.L. Synopsis of the genera and suprageneric taxa of Euphorbiaceae. Ann. Mo. Bot. Gard. 1994, 81, 33–144. [Google Scholar] [CrossRef]
  6. Zamora-Martínez, M.; Méndez, C.; Pérez, R.; Cortés, E. Euphorbia antisyphilitica Zucc.: Recurso Forestal no Maderable de Alto Valor. Folleto Técnico N° 12; CENID-COMEF: Mexico City, Mexico, 2013. [Google Scholar]
  7. Tapia-Tapia, E.D.; Reyes-Chilpa, R. Productos forestales no maderables en México: Aspectos económicos para el desarrollo sustentable. Madera Y Bosques 2008, 14, 95–112. [Google Scholar] [CrossRef]
  8. Villarreal-Quintanilla, J.A.; Bartolomé-Hernández, J.A.; Estrada-Castillón, E.; Ramírez-Rodríguez, H.; Martínez-Amador, S.J. El elemento endémico de la flora vascular del Desierto Chihuahuense. Acta BotÁNica Mex. 2017, 118, 65–96. [Google Scholar] [CrossRef]
  9. Villa-Castorena, M.; Catalán-Valencia, E.A.; Inzunza-Ibarra, M.A.; González-López, M.d.L.; Arreola-Ávila, J.G. Producción de plántulas de Candelilla (Euphorbia antisyphilitica Zucc.) mediante estacas. Rev. Chapingo Ser. Cienc. For. Y Ambiente 2010, 16, 37–47. [Google Scholar] [CrossRef]
  10. Peña-Contreras, A.R. Utilidad de algunas metodologías de análisis la interacción genotipo-ambiente en la medición de la tasa de recuperación, en el crecimiento de la candelilla (Euphorbia antisyphillitica Zucc.), bajo diferentes condiciones ecológicas. In Tesis de Maestría en Fitomejoramiento; Universidad Autónoma Agraria Antonio Narro: Buenavista, Saltillo, 1998. [Google Scholar]
  11. Flores-Del Ángel, M. Situación actual de las poblaciones de candelilla (Euphorbia antisyphilitica Zucc.) inventario, su propagación sexual y asexual en el estado de Coahuila, México. In Tesis de Doctorado; Universidad Autónoma de Nuevo León: San Nicolás de los Garza, Mexico, 2013; pp. 1–134. [Google Scholar]
  12. Schneider, E.; Naturschutz, B. Trade survey study on succulent Euphorbia species protected by CITES and used as cosmetic, food and medicine, with special focus on Candelilla wax. In Proceedings of the 18th Meeting of the Plants Committee, Buenos Aires, Argentina, 17–21 March 2009. [Google Scholar]
  13. Grotch, S.L.; MacCracken, M.C. The use of general circulation models to predict regional climatic change. J. Clim. 1991, 4, 286–303. [Google Scholar] [CrossRef]
  14. Warren, D.L.; Seifert, S.N. Ecological niche modeling in Maxent: The importance of model complexity and the performance of model selection criteria. Ecol. Appl. 2011, 21, 335–342. [Google Scholar] [CrossRef]
  15. Ahmed, S.E.; Mcinerny, G.; O’Hara, K.; Harper, R.; Salido, L.; Emmott, S.; Joppa, L. Scientists and Software—Surveying the Species Distribution Modelling Community. Divers. Distrib. 2015, 21, 258–267. [Google Scholar] [CrossRef]
  16. Elith, J.; Graham, C.H.; Anderson, R.P.; Dudík, M.; Ferrier, S.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; Leathwick, J.R.; Lehmann, A.; et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 2006, 29, 129–151. [Google Scholar] [CrossRef]
  17. Baek, H.; Lee, J.; Lee, H.; Hyun, Y.; Cho, C.; Kwon, W.; Marzin, C.; Gan, S.; Kim, M.; Choi, D.; et al. Climate change in the 21st century simulated by HadGEM2-AO under representative concentration pathways. Asia-Pac. J. Atmos. Sci. 2013, 49, 603–618. [Google Scholar] [CrossRef]
  18. Sáenz-Romero, C.; Rehfeldt, G.E.; Ortega-Rodríguez, J.M.; Marín-Togo, M.C.; Madrigal-Sánchez, X. Pinus leiophylla Suitable Habitat For 1961–1990 and Future Climate. Bot. Sci. 2015, 93, 709–718. [Google Scholar] [CrossRef]
  19. Soberón, J.; Peterson, A.T. Interpretation of models of fundamental ecological niches and species distributions areas. Biodivers. Inf. 2005, 2, 1–10. [Google Scholar] [CrossRef]
  20. García, E. Isotermas Medias Anuales. Escala 1:1,000,000, México. 1998. Available online: http://www.conabio.gob.mx/informacion/gis/?vns=gis_root/clima/climas/clima1mgw (accessed on 12 December 2022).
  21. GBIF.org. GBIF Occurrence Download. 2018. Available online: https://www.gbif.org/occurrence/download/0012849-180824113759888 (accessed on 18 September 2018).
  22. Comisión Nacional Forestal. Inventario Nacional Forestal y de Suelos 2013–2014. Registros de Candelilla (Euphorbia antisyphillitica Zucc.). 2014. Available online: https://snmf.cnf.gob.mx/infys/ (accessed on 15 January 2022).
  23. Osorio-Olvera, L.; Vijay, B.; Narayani, B.; Soberón, J.; Falconi, M. Ntbox: From Getting Biodiversity Data to Evaluating Species Distributions Models in a Friendly GUI Environment; R package Version 0.2.5.4; GitHub: San Francisco, CA, USA, 2019. [Google Scholar]
  24. Peterson, A.T.; Nakazawa, Y. Environmental data sets matter in ecological niche modelling: An example with Solenopsis invicta and Solenopsis richteri. Glob. Ecol. Biogeogr. 2008, 17, 135–144. [Google Scholar] [CrossRef]
  25. Fick, S.E.; Hijmans, R.J. Worldclim 2: New 1-km spatial resolution climate surfaces for global land areas. Int. J. Climatol. 2017, 37, 4302–4315. [Google Scholar] [CrossRef]
  26. Martin, G.M.; Bellouin, N.; Collins, W.J.; Culverwell, I.D.; Halloran, P.R.; Hardiman, S.C.; Hinton, T.J.; Jones, C.D.; McDonald, R.E.; McLaren, A.J.; et al. The HadGEM2 family of Met Office Unified Model climate configurations. Geosci. Model Dev. 2011, 4, 723–757. [Google Scholar] [CrossRef]
  27. Giorgetta, M.A.; Jungclaus, J.; Reick, C.H.; Legutke, S.; Bader, J.; Böttinger, M.; Brovkin, V.; Crueger, T.; Esch, M.; Fieg, K.; et al. Climate and carbon cycle changes demo1850 to 2100 in MPI-ESM simulations for the Coupled Model Intercomparison Project phase 5. J. Adv. Model. Earth Syst. 2013, 5, 572–597. [Google Scholar] [CrossRef]
  28. Instituto Nacional de Ecología y Cambio Climático. Guía de Escenarios de Cambio Climático Para Tomadores de Decisiones; Instituto Nacional de Ecología y Cambio Climático: Mexico City, Mexico, 2022; p. 65. [Google Scholar]
  29. Batjes, N.H.; Ribeiro, E.; Oostrum, A.V.; Leenaars, J.; Hengl, T.; Mendes, J. WoSIS: Providing standardised soil p rofile data for the world. Earth Syst. Sci. Data 2017, 9, 1–14. [Google Scholar] [CrossRef]
  30. Instituto Nacional de Estadística y Geografía. Continuo de Elevaciones Mexicano 3.0. Aguascalientes, Aguascalientes, México. 2013. Available online: https://www.inegi.org.mx/app/geo2/elevacionesmex/ (accessed on 10 November 2022).
  31. ESRI. Software ArcGIS Version 10.3. 2014. Available online: https://www.arcgis.com/index.html (accessed on 12 September 2021).
  32. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2015; Available online: https://cran.r-project.org (accessed on 8 June 2022).
  33. Muscarella, R.; Galante, P.J.; Soley-Guardia, M.; Boria, R.A.; Kass, J.M.; Uriarte, M.; Anderson, R.P. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol. Evol. 2014, 5, 1198–1205. [Google Scholar] [CrossRef]
  34. Fitz-Maurice, B.; Sotomayor, M.; Fitz-Maurice, W.A.; Hernández, H.; Smith, M. Astrophytum coahuilense (Bonete de Obispo); Distribución Conocida; Catálogo de Metadatos Geográficos/Comisión Nacional Para el Conocimiento y Uso de la Biodiversidad. 2013. Available online: http://www.conabio.gob.mx/informacion/gis/?vns=gis_root/biodiv/distcon/dcplant/dcpmagcact/astaste_gcagw (accessed on 14 December 2022).
  35. Merow, C.; Smith, M.J.; Silander, J.A. A practical guide to MaxEnt for modeling species’ distributions: What it does, and why inputs and settings matter. Ecography 2013, 36, 1058–1069. [Google Scholar] [CrossRef]
  36. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecological Modelling 2006, 190, 231–259. [Google Scholar] [CrossRef]
  37. Phillips, S.J.; Dudik, M. Modeling of species distributions with Maxent: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  38. Morrone, J.J.; Escalante, T. Introducción a la Biogeografía, 1st ed.; México, D.F., Ed.; Universidad Nacional Autónoma de México: Mexico City, Mexico, 2016; ISBN 968-36-9463-2. [Google Scholar]
  39. Martínez-Sifuentes, A.R.; Villanueva-Díaz, J.; Manzanilla-Quiñones, U.; Becerra-López, J.L.; Hernández-Herrera, J.A.; Estrada-Ávalos, J.; Velázquez-Pérez, A. Spatial modeling of the ecological niche of Pinus greggii Engelm. (Pinaceae): A species conservation proposal in Mexico under climate change scenarios. Iforest Biogeosciences For. 2020, 13, 426–434. [Google Scholar] [CrossRef]
  40. Peterson, A.T. Ecological niche conservatism: A time-structured review of evidence. J. Biogeogr. 2011, 38, 817–827. [Google Scholar] [CrossRef]
  41. Lobo, J.M.; Jiménez-Valverde, A.; Real, R. AUC: A misleading measure of the performance of predictive distribution models. Glob. Ecol. Biogeogr. 2007, 17, 145–151. [Google Scholar] [CrossRef]
  42. Shcheglovitova, M.; Anderson, R. Estimating optimal complexity for ecological niche models: A Jacknife approach for species with small simple sices. Ecol. Model. 2013, 269, 9–17. [Google Scholar] [CrossRef]
  43. Martínez-Sifuentes, A.R.; Villanueva-Díaz, J.; Crisantos, E.; Stahle, D. Current and future spatial modeling of habitat suitability of the Mexican baldcypress (Taxodium mucronatum Ten.): A proposal for conervation in Mexico. Bot. Sci. 2021, 99, 752–770. [Google Scholar] [CrossRef]
  44. Secretaría de Medio Ambiente y Recursos Naturales-Comisión Nacional de Areas Naturales Protegidas. Archivo Vectorial de Áreas Naturales. Aguascalientes, Aguascalientes, México. 2017. Available online: http://www.conabio.gob.mx/informacion/gis/?vns=gis_root/region/biotic/anp2021gw (accessed on 4 November 2022).
  45. Stockwell, B.D.; Peterson, A.T. Effects of sample size on accuracy of species distribution models. Ecol. Model. 2002, 148, 1–13. [Google Scholar] [CrossRef]
  46. Hernández-Herrera, J.A.; Moreno-Reséndez, A.; Valenzuela-Núñez, L.M.; Martínez-Salvador, M. Modelation of the presence of Euphorbia antisyphilitica Zucc with physical and chemical properties of soil. Ecosistemas Y Recur. Agropecu. 2019, 6, 499–511. [Google Scholar] [CrossRef]
  47. Bañuelos-Revilla, J.E.; Palacio-Núñez, J.; Martínez-Montoya, J.F.; Olmos-Oropeza, G.; Flores-Cano, J.A. Potential distribution and abundance of candelilla (Euphorbia antisyphilitica) in northern Zacatecas, Mexico. Madera Y Bosques 2019, 25, 1–14. [Google Scholar] [CrossRef]
  48. Varela, S.; Mateo, R.G.; García-Valdés, R.; Fernández-González, F. Macroecología y ecoinformática: Sesgos, errores y predicciones en el modelado de distribuciones. Rev. Ecosistemas 2014, 23, 46–53. [Google Scholar] [CrossRef]
  49. Cruz-Cárdenas, G.; López-Mata, L.; Silva, J.T.; Bernal-Santana, N.; Estrada-Godoy, F.; López-Sandoval, J.A. Potential distribution model of Pinaceae species under climae change scenarios in Michoacán. Rev. Chapingo Ser. Cienc. For. Y Del Ambiente 2016, 22, 135–145. [Google Scholar] [CrossRef]
  50. Manzanilla-Quiñones, U.; Aguirre-Calderón, Ó.A.; Jiménez-Pérez, J.; Treviño-Garza, E.J.; Yerena-Yamallel, J.I. Current and future distribution of the subalpine forest of Pinus hartwegii Lindl. In the Neovolcanic Transversal Axis. Madera Y Bosques 2019, 25, e2521804. [Google Scholar] [CrossRef]
  51. Vargas-Pineda, G.; Valdez-Cepeda, R.D.; López-Santos, A.; Flores-Hernández, A.; Hernández-Quiroz, N.S.; Martínez-Salvador, M. Current and Future Potential Distribution of the Xerophytic Shrub Candelilla (Euphorbia antisyphilitica) under Two Climate Change Scenarios. Forests 2020, 11, 530. [Google Scholar] [CrossRef]
  52. Rojas-Molina, R.; Saucedo-Pompa, S.; De León-Zapata, M.A.; Jasso-Cantú, D.; Aguilar, C.N. Pasado, presente y futuro de la candelilla. Rev. Mex. Cienc. For. 2011, 2, 7–18. [Google Scholar] [CrossRef]
  53. Martínez-Salvador, M.; Hermosillo-Rojas, D.D.; Mojica-Guerrero, A.; Prieto-Amparan, J.A. Potencial Productivo y Zonificación Para el Uso y Manejo de Especies Forestales de Zonas Áridas; INIFAP: Chihuahua, México, 2015; Volume 1, p. 119. Available online: https://www.researchgate.net/publication/274636131_Potencial_productivo_y_zonificaion_para_el_uso_y_manejo_de_especies_forestales_de_zonas_aridas (accessed on 12 December 2021).
  54. Saucedo-Pompa, S.; Rojas-Molina, R.; Aguilera-Carbó, A.F.; Saenz-Galindo, A.; De La Garza, H.; Jasso-Cantú, D.; Aguilar, C.N. Edible film based on candelilla wax to improve the shelf life and quality of avocado. Food Res. Int. 2009, 42, 511–515. [Google Scholar] [CrossRef]
  55. Hernández-Herrera, J.A.; Moreno-Reséndez, A.; Valenzuela-Núñez, L.M.; Flores-Hernández, A.; Zamora-Martínez, M.C. Distribución potencial de Euphorbia antisyphilitica Zucc. En México. Rev. Chapingo Ser. Zonas Áridas 2021, 19, 1–14. [Google Scholar] [CrossRef]
  56. Martínez-Ballesté, A.; Mandujano, M.C. The consequences of harvesting on regeneration of a non-timber wax producing species (Euphorbia antisyphilitica Zucc.) of the Chihuahuan Desert. Econ. Bot. 2013, 67, 121–136. [Google Scholar] [CrossRef]
  57. Flores-López, C. Viabilidad de las semillas, emergencia de plántulas y plantaciones de Candelilla (Euphorbia antisyphilitica Zucc.) en Ramos Arizpe. In Tesis de Licenciatura; Universidad Autónoma Agraria Antonio Narro División de Agronomía: Buenavista, Mexico, 1995. [Google Scholar]
  58. Martínez-Salvador, S.M.; Hermosillo-Rojas, D.E.; Mojica-Guerrero, A.S. Guía para la Zonificación Forestal de Zonas Áridas de México; Publicación Especial: INIFAP—Campo Experimental La Campana: Chihuahua, México, 2014; p. 29. [Google Scholar]
  59. Comisión Nacional Forestal. La Candelilla. 2005. Available online: https://www.conafor.gob.mx/biblioteca/documentos/PROGRAMA_DE_DESARROLLO_DE_PFC_A_15_ANOS_DE_SU_CREACION.PDF (accessed on 21 January 2022).
  60. Pittarello, M.; Lonati, M.; Gorlier, A.; Probo, M.; Lombardi, G. Species-rich Nardus stricta grasslands host a higher vascular plant diversity on calcareous than on siliceous bedrock. Plant Ecol. Divers. 2017, 10, 343–351. [Google Scholar] [CrossRef]
  61. Flores, J.; Jurado, E. Germination and Early Growth Traits of 14 Plant Species Native to Northern Mexico. Southwest. Nat. 1998, 43, 40–46. Available online: https://www.researchgate.net/publication/236323003_Germination_and_early_growth_traits_of_14_plant_species_native_to_northern_Mexico (accessed on 14 January 2022).
  62. Lan, Y.; Melton, L.; Shahidi, F.; Varelis, P. (Eds.) Waxes; Academic Press: Oxford, UK, 2019; pp. 312–316. [Google Scholar] [CrossRef]
  63. Manzanilla-Quiñones, U.; Delgado-Valerio, P.; Hernández-Ramos, J.; Molina-Sánchez, A.; García-Magaña, J.J.; Rocha-Granados, M.C. Similaridad del nicho ecológico de Pinus montezumae y P. pseudostrobus (Pinaceae) en México: Implicaciones para la selección de áreas productoras de semilla y de conservación. Acta Bot. 2019, 126, e1398. [Google Scholar] [CrossRef]
  64. Soberón, J.; Miller, C.P. Evolution of ecological niches. Miscelánea Matemática 2009, 49, 83–99. Available online: http://www.miscelaneamatematica.org/Misc49/4906.pdf (accessed on 8 October 2022).
  65. Uribe-Botero, E. El Cambio Climático y Sus Efectos en la Biodiversidad en América Latina, 1st ed.; United Nations CEPAL: Santiago, Chile, 2015; Volume 1, p. 84. [Google Scholar]
  66. Gelviz-Gelvez, S.M.; Pavón, N.P.; Illoldi-Rangel, P.; Ballesteros-Barrera, C. Ecological niche modeling under climate change to select shrubs for ecological restoration in Central Mexico. Ecol. Eng. 2015, 74, 302–309. [Google Scholar] [CrossRef]
  67. Booth, T.; Nix, H.; Hutchinson, M.; Jovanovic, T. Niche analysis and tree species introduction. For. Ecol. Manag. 1988, 3, 47–59. [Google Scholar] [CrossRef]
  68. Hutchinson, G.E. Concluding remarks. Cold Spring Harb. Symp. Quant. Biol. 1957, 22, 415–427. [Google Scholar] [CrossRef]
  69. Martínez-Sifuentes, A.R.; Hernández-Herrera, J.A.; Valenzuela-Núñez, L.M.; Briceño-Contreras, E.A.; Manzanilla-Quiñones, U.; Gastélum-Arellánez, A.; Trucíos-Caciano, R.; López Calderón, M.J. Climate Change Impact on the Habitat Suitability of Pseudotsuga menziesii Mirb. Franco in Mexico: An Approach for Its Conservation. Sustainability 2022, 14, 8888. [Google Scholar] [CrossRef]
Sustainability 15 07737 g001 550
Figure 1. Records of candelilla presence in Mexico.
Figure 1. Records of candelilla presence in Mexico.
Sustainability 15 07737 g001
Sustainability 15 07737 g002 550
Figure 2. Geographical distribution of candelilla in Mexico.
Figure 2. Geographical distribution of candelilla in Mexico.
Sustainability 15 07737 g002
Sustainability 15 07737 g003 550
Figure 3. Geographical distribution of candelilla in the state of Coahuila. (The locations’ numbers are in Table 1).
Figure 3. Geographical distribution of candelilla in the state of Coahuila. (The locations’ numbers are in Table 1).
Sustainability 15 07737 g003
Sustainability 15 07737 g004 550
Figure 4. Candelilla habitat suitability for the MPIESM-LR model for 2070 and the RCP 8.5 scenario.
Figure 4. Candelilla habitat suitability for the MPIESM-LR model for 2070 and the RCP 8.5 scenario.
Sustainability 15 07737 g004
Sustainability 15 07737 g005 550
Figure 5. Candelilla habitat suitability (green areas) with MPIESM-LR model for the period 2050 (A) RCP 4.5 and (B) RCP 8.5; zones for the period 2070 (C) RCP 4.5 and (D) RCP 8.5.
Figure 5. Candelilla habitat suitability (green areas) with MPIESM-LR model for the period 2050 (A) RCP 4.5 and (B) RCP 8.5; zones for the period 2070 (C) RCP 4.5 and (D) RCP 8.5.
Sustainability 15 07737 g005
Sustainability 15 07737 g006 550
Figure 6. Candelilla habitat suitability (green areas) with HadGEM2-ES model for the period 2050 (A) RCP 4.5 and (B) RCP 8.5; areas for the period 2070 (C) RCP 4.5 and (D) RCP 8.5.
Figure 6. Candelilla habitat suitability (green areas) with HadGEM2-ES model for the period 2050 (A) RCP 4.5 and (B) RCP 8.5; areas for the period 2070 (C) RCP 4.5 and (D) RCP 8.5.
Sustainability 15 07737 g006
Sustainability 15 07737 g007 550
Figure 7. Proposal of suitable areas for in situ conservation and restoration activities of candelilla in Coahuila by 2070 (A) MPIESM-LR model, and (B) HadGEM2-ES model.
Figure 7. Proposal of suitable areas for in situ conservation and restoration activities of candelilla in Coahuila by 2070 (A) MPIESM-LR model, and (B) HadGEM2-ES model.
Sustainability 15 07737 g007
Sustainability 15 07737 g008 550
Figure 8. Overlapping layers of rosetophytic desert scrub and current presence of candelilla in Coahuila.
Figure 8. Overlapping layers of rosetophytic desert scrub and current presence of candelilla in Coahuila.
Sustainability 15 07737 g008
Table
Table 1. Estimated current area of candelilla in federal and state natural protected areas in Coahuila.
Table 1. Estimated current area of candelilla in federal and state natural protected areas in Coahuila.
NumberNatural Protected AreaArea (km2)Percent of Total Surface Area in Coahuila (%)
1004 Don Martín2534.909.89
2Ocampo1885.447.35
3Maderas del Carmen251.830.98
4Cuatrociénegas183.610.71
5Sierra y Cañón de Jimulco165.370.64
6Tomás Garrido142.150.55
7026 Bajo Río San Juan71.260.27
8Villa de Bilbao28.050.11
9Mapimí23.650.09
10Zapalinamé3.920.01
11Zona de Restauración Zapalinamé1.44<0.00
12Cumbres de Monterrey0.42<0.00
13Río Bravo del Norte0.34<0.00
Table
Table 2. Ranges of AUC values of the models under climate change.
Table 2. Ranges of AUC values of the models under climate change.
Global Circulation ModelTraining AUCValidation AUC
MPIESM-LR (RCP 4.5) 20500.921–0.9490.820–0.953
MPIESM-LR (RCP 8.5) 20500.924–0.9630.859–0.906
MPIESM-LR (RCP 4.5) 20700.923–0.9510.803–0.936
MPIESM-LR (RCP 8.5) 20700.927–0.9500.825–0.902
HadGEM2-ES (RCP 4.5) 20500.921–0.9550.891–0.914
HadGEM2-ES (RCP 8.5) 20500.908–0.9560.867–0.891
HadGEM2-ES (RCP 4.5) 20700.924–0.9630.803–0.921
HadGEM2-ES (RCP 8.5) 20700.918–0.9490.818–0.945
Table
Table 3. Most important environmental variables of candelilla in the models of current habitat suitability and under climate change in Coahuila.
Table 3. Most important environmental variables of candelilla in the models of current habitat suitability and under climate change in Coahuila.
ModelImportant VariablesContribution (%)
CurrentFRA, BIO 12, DEP, BIO 11, BIO 9, and ELE96.7
MPIESM-LR (RCP 4.5) 2050FRA, BIO 12, DEP, BIO 11, BIO 9, and DEN97.1
MPIESM-LR (RCP 8.5) 2050FRA, BIO 12, DEP, BIO 7, ELE, and BIO 992.3
MPIESM-LR (RCP 4.5) 2070FRA, BIO 12, ELE, DEP, BIO 7, and BIO 1194.8
MPIESM-LR (RCP 8.5) 2070FRA, BIO 12, DEP, BIO 11, BIO 7, and DEN91.8
HadGEM2-ES (RCP 4.5) 2050FRA, BIO 12, DEP, ELE, BIO 11, and DEN96.2
HadGEM2-ES (RCP 8.5) 2050FRA, BIO 12, DEP, BIO 11, BIO 9, and BIO 797.1
HadGEM2-ES (RCP 4.5) 2070FRA, BIO 12, DEP, BIO 11, BIO 9, and ELE96.1
HadGEM2-ES (RCP 8.5) 2070FRA, BIO 12, DEP, BIO 11, ELE, and BIO 797.4
Table
Table 4. Estimated candelilla area using the MPIESM-LR model within federal and state natural protected areas in Coahuila.
Table 4. Estimated candelilla area using the MPIESM-LR model within federal and state natural protected areas in Coahuila.
Natural Protected AreaArea (km2)Percentage (%) 1
004 Don Martín327.1412.90
Ocampo154.908.21
Sierra y Cañón de Jimulco89.1153.88
Tomás Garrido82.0657.72
026 Bajío Río San Juan69.1096.96
Cuatrociénegas8.624.69
Villa de Bilbao5.6620.17
Zapalinamé3.92100
Maderas del Carmen1.380.54
Zona de Restauración Zapalinamé0.9968.75
Cumbres de Monterrey0.42100
1 Percentage in relation to the current area of candelilla in the natural protected area.
Table
Table 5. Estimated candelilla area from the HadGEM2-ES model within federal and state natural protected areas in Coahuila.
Table 5. Estimated candelilla area from the HadGEM2-ES model within federal and state natural protected areas in Coahuila.
NumberNatural Protected AreaArea (km2)Percentage (%) 1
1Ocampo481.2025.52
2004 Don Martín175.756.93
3Maderas del Carmen50.1219.90
4026 Bajo Río San Juan40.6257.00
5Sierra y Cañón de Jimulco7.484.52
6Tomás Garrido3.482.44
7Cuatrociénegas3.421.86
8Zapalinamé1.4536.98
9Zona de Restauración Zapalinamé0.6142.36
10Mapimí0.271.14
11Río Bravo Norte0.1338.23
12Cumbres de Monterrey0.0819.04
13Villa de Bilbao0.020.07
1 Percentage in relation to the current area of candelilla in the natural protected area.
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

Martínez-Sifuentes, A.R.; Estrada-Ávalos, J.; Trucíos-Caciano, R.; Villanueva-Díaz, J.; López-Hernández, N.A.; López-Favela, J.d.D. Predicting Climate Change Impacts on Candelilla (Euphorbia antisyphilitica Zucc.) for Mexico: An Approach for Mexico’s Primary Harvest Area. Sustainability 2023, 15, 7737. https://0-doi-org.brum.beds.ac.uk/10.3390/su15107737

AMA Style

Martínez-Sifuentes AR, Estrada-Ávalos J, Trucíos-Caciano R, Villanueva-Díaz J, López-Hernández NA, López-Favela JdD. Predicting Climate Change Impacts on Candelilla (Euphorbia antisyphilitica Zucc.) for Mexico: An Approach for Mexico’s Primary Harvest Area. Sustainability. 2023; 15(10):7737. https://0-doi-org.brum.beds.ac.uk/10.3390/su15107737

Chicago/Turabian Style

Martínez-Sifuentes, Aldo Rafael, Juan Estrada-Ávalos, Ramón Trucíos-Caciano, José Villanueva-Díaz, Nuria Aidé López-Hernández, and Juan de Dios López-Favela. 2023. "Predicting Climate Change Impacts on Candelilla (Euphorbia antisyphilitica Zucc.) for Mexico: An Approach for Mexico’s Primary Harvest Area" Sustainability 15, no. 10: 7737. https://0-doi-org.brum.beds.ac.uk/10.3390/su15107737

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