Next Article in Journal
Essential Oil and Phylogenetic Positions of Five Medicinal Litsea Species (Lauraceae)
Previous Article in Journal
Foraging Resource Partitioning in the California Sea Lion (Zalophus californianus) from the Southwestern Gulf of California
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 17
Issue 3
10.3390/d17030167
diversity-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
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Modeling the Potential Distribution of Aulonemia queko: Historical, Current, and Future Scenarios in Ecuador and Other Andean Countries

by
Hugo Cedillo
1,2,*,
Luis G. García-Montero
1,
Omar Cabrera
3,
Mélida Rocano
2,
Andrés Arciniegas
2 and
Oswaldo Jadán
2,*
1
Centro para la Conservación de la Biodiversidad y el Desarrollo Sostenible (CBDS), E.T.S.I. Montes, Forestal y del Medio Natural, Universidad Politécnica de Madrid (UPM), 28040 Madrid, Spain
2
Grupo de Ecología Forestal, Agroecosistemas y Silvopasturas en Sistemas Ganaderos, Facultad de Ciencias Agropecuarias, Universidad de Cuenca, Cuenca 010114, Ecuador
3
Department of Biological Sciences, Private Technical University of Loja, San Cayetano Alto s/n, Loja 1101608, Ecuador
*
Authors to whom correspondence should be addressed.
Diversity 2025, 17(3), 167; https://doi.org/10.3390/d17030167
Submission received: 9 January 2025 / Revised: 13 February 2025 / Accepted: 14 February 2025 / Published: 26 February 2025
(This article belongs to the Topic Responses of Trees and Forests to Climate Change)
Browse Figures
Review Reports Versions Notes

Abstract

:
Aulonemia queko Goudot (Poaceae, Bambusoideae) is a species of great cultural importance that has been used as a non-timber forest product in Andean forests for centuries. Despite inhabiting montane forests vulnerable to deforestation, its distribution has not been thoroughly assessed for conservation. This study analyzes its potential distribution at the regional scale (the four countries where it is distributed) and locally (in greater detail within Ecuador), using presence records and climatic and land-use data. Maxent was identified as the best algorithm, achieving high values of AUC, TSS, sensitivity, and specificity. At a regional level, A. queko is estimated to occupy approximately 264,540 km2, mostly in Peru, with small areas in Bolivia. In Ecuador, the historical scenario showed the widest distribution, while the current–near-future scenario (20–40–SSP126) presented a more stable model. Temperature and rainfall represented critical factors in defining suitable habitats, as A. queko is highly sensitive to seasonal moisture availability. Land-use changes have reduced potential habitats by more than 35%, underscoring an intensified threat of habitat loss in these biodiversity-rich regions. However, projected climate changes pose an even greater impact, significantly reducing potential distribution. Our findings highlight the compelling effects of both climate-change-driven and human-driven land-use change on the future persistence of A. queko and emphasize the urgent need for targeted conservation strategies to protect its core habitats.

1. Introduction

Species distribution models reveal the areas with appropriate environmental conditions required by species or plant formation of interest, based on a set of their presence data as well as environmental variables [1]. These models take into account different topics related to ecology, biogeography, evolution, and climate change [2]. In the current context, there is a growing interest in species modeling, which supports the design of actions for the management, control, and conservation of species within a given area [3,4]. When including future scenarios, spatial ecological modeling plays a crucial role in predicting the impacts of climate change on species distribution [5], considering future variations of climatic factors, but also the pressure of deforestation and biodiversity loss on native species, resulting in fragmented habitats [6]. In these scenarios, along with other key species, certain bamboo species must be considered due to their ecological and cultural significance at both regional and local levels.
Aulonemia queko Goudot (Poaceae, Bambusoideae) is a neotropical bamboo species found in the Andean regions of Colombia, Ecuador, Peru, and Bolivia [7]. Few studies have addressed its taxonomy, anatomy, and uses in these countries [8]. It thrives in forests facing deforestation pressures. A. queko holds ethnobotanical significance in Ecuador as a non-timber product [9]. Its long stems are used to craft musical instruments like flutes and blowpipes [10,11]. These stems also serve as structural elements for rural house roofs and are employed in making baskets and crafts [10,12,13,14]. Additionally, A. queko is used in floral arrangements, kitchen utensils, lamps, curtains, fishing rods, and hunting instruments [15]. Its leaves provide forage for livestock, including cattle, sheep, guinea pigs, and rabbits [16,17,18]. However, tropical habitats of the Aulonemia Goudot genus face threats from deforestation and human activities [10,19]. The future of A. queko is uncertain due to overexploitation and habitat loss from deforestation, mining, and land burning [20,21].
The ecology and potential distribution of Andean bamboo species, particularly of A. queko, are poorly known. Some noteworthy local studies are modeling the distribution of certain Aulonemia species for management and conservation purposes, such as those regarding the ecological niche of A. herzogiana (Henrard) McClure, A. hirtula (Pilg.) McClure, and A. tremula Renvoize, endemic to the Yungueño páramo of La Paz in Bolivia [19]. In northwestern Argentina, the distribution of native and introduced woody bamboo species was modeled, predicting the potential distribution into higher altitudes based on mean temperatures and precipitation [22]. However, the low research interest for the Aulonemia genus is likely due to its predominantly artisanal use in rural areas, where it serves as raw material for small-scale local production. Consequently, its reproduction, ecology, and current and future distribution for management and conservation purposes have not been studied, as has been conducted with other bamboo genera such as Guadua and Bambusa [16,23].
In this context, predictive spatial ecological modeling of A. queko could play a pivotal role in assessing this species’ distribution, providing essential insights to guide effective management and conservation efforts. Various modeling methods are available to assess past, present, and future species distributions [24]. Maxent stands out for its effectiveness with limited distribution data, which is a typical situation for tropical ecosystems, as it integrates multiple environmental variables to improve predictive accuracy and identify potential conservation areas under climate change scenarios [25,26]. Generalized linear models (GLMs) are also widely used for their flexibility in handling diverse response data and multiple environmental factors, making them ideal for complex species–habitat relationships [27,28]. In tropical ecosystems, GLMs well capture the effects of variable environmental conditions on species distribution [29,30]. Similarly, random forest (RF) excels at managing large datasets with numerous variables, providing accurate predictions in complex contexts [31]. Random forest has proven effective in tropical biodiversity studies, aiding in the design of strategies to mitigate the effects of climate change [32,33].
As a pioneer study on a poorly known species and its contribution to its management and conservation, the primary aim of the present research was to evaluate the potential distribution and temporal changes of A. queko. The specific objectives were as follows: (1) Model the potential distribution of A. queko at a regional scale using presence data from the Global Biodiversity Information Facility (GBIF) digital platform. (2) Model its distribution using historical and future climate data, focusing on the distribution of the species in natural habitats within Ecuador; we expected a significant decrease in areas of potential distribution from historical to future scenarios, with steeper declines under more severe climate change conditions. (3) Determine whether recent land-use changes have a more significant impact on the potential distribution of A. queko than projections based on future climate scenarios; we expected severe future climate change projections to have a more decisive impact on the species’ potential distribution area. Additionally, we wanted to determine the environmental variables that most strongly influence the potential distribution across the different models. To achieve this, we tested various ecological modeling methods and metrics to identify the best-performing model.

2. Materials and Methods

2.1. Study Area

At a regional level, the study area spans across Colombia, Ecuador, Peru, and Bolivia (Figure 1A). The distribution of A. queko in these countries was documented using presence records from the GBIF, with an altitudinal range from 1100 m to 4004 m. To delineate this species’ potential distribution with more precision, we defined a geographical area with an elevation range between 650 and 4500 m, incorporating a 500 m buffer beyond the species’ recorded lower and upper elevation limits (Figure 1B). At a local level, we focused on continental Ecuador, where A. queko is particularly abundant in secondary cloud forests in the Western and Eastern Cordilleras, especially in the provinces of Azuay, Imbabura, Loja, and Napo, as documented by Aguirre [12].

2.2. Data Collection

To model the potential distribution of A. queko at the regional level, geographical presence data (x, y) were acquired from the Global Biodiversity Information Facility (GBIF) platform using the R package ‘rgbif’ [34]. These records of A. queko presence were concentrated in the inter-Andean zone across Colombia, Ecuador, Peru, and Bolivia. Subsequently, data cleaning was performed using the ‘CoordinateCleaner’ R package [35], which removed records lacking coordinates and addressed incomplete or inaccurate data. The resulting matrix depicting the species distribution in the four countries was based on their presence. The dataset included records reported between 1945 and 2010.
Nineteen historical climatic variables (climate averages for the years 1970–2000) with their continuous data were incorporated into the matrix, derived from georeferenced raster layers from the global climatic database WorldClim [36] (Appendix A). For this purpose, the presence records were used to extract the environmental variable values using the ‘extract’ function from the ‘raster’ package in R [37]. This comprehensive set of climatic variables facilitated a thorough analysis of the ecological niche and potential distribution of A. queko across the specified region (Table 1).
This presence data matrix was also used to model Ecuador’s historical climate scenario (average from 1970–2000), including country-specific records. Additional models were created for present and future scenarios using shared socioeconomic pathways (SSPs), which provide projections of emissions and land-use changes [38]. We selected SSP126 (optimistic), SSP245 (moderate), and SSP585 (pessimistic) across two time periods: the present and near future (2020–2040), represented by 21–40–SSP126, 21–40–SSP245, and 21–40–SSP585, and the longer-term future (2041–2060), with 41–60–SSP126, 41–60–SSP245, and 41–60–SSP585. To assess the distribution of A. queko in Ecuador under these scenarios (historical, the near future, and longer-term future), we applied the same climate variables used at the regional level but also included slope data derived from digital models created using 40 m elevation contour lines.

2.3. Data Analysis

To analyze the impact of land-use change on the distribution of A. queko, we focused on the scenario showing the largest area of potential distribution across all analyzed scenarios. We used the 2020 vegetation cover and land use layer for continental Ecuador, provided by the Ministry of Environment and Ecological Transition of Ecuador [39] on a shapefile format. This layer was intersected with the distribution areas identified through the optimal modeling approach (Maxent) and reclassified in ArcGIS. This analysis allowed us to assess whether land-use change has a more decisive influence in predicting the species’ future distribution than the ecological modeling based on climate factors. Following these intersections, we further examined the areas with the highest potential for A. queko distribution, divided by province, within its natural distribution range in Ecuador.

2.4. Generation of Models

Using spatial distribution models, we generated maps depicting the distribution of A. queko based on presence records collected regionally and within continental Ecuador. At both regional and Ecuadorian scales, a correlation analysis was conducted to select non-correlated climatic variables (r < 0.7) (to avoid collinearity), using the ‘caret’ package in R [40].
We used three algorithms to model the potential distribution of A. queko: maximum entropy models (MaxEnt) [26,41], general linear models (GLMs) [42,43], and random forest (RF) [44,45], implemented in R using several specialized packages. The MaxEnt algorithm was executed via the ‘dismo’ package in R [46], which allows maximum entropy modeling based on presence data, using the maxent function. For the generalized linear model (GLM), the GLM function from R’s base package [38] was employed, fitting the model with a binomial distribution suited for presence/absence data. The random forest model was implemented with the ‘randomForest’ package [47], using the ‘randomForest’ function to generate the model from training data. In addition to using the available presence points for each species, we generated random reference points by increasing the number of presence data fivefold, at both regional scale and within Ecuador. Although no strict rule exists for determining the number of random points, this proportion was selected to achieve a reasonable balance between the presence and absence of data. The dataset was then randomly divided into 70% training and 30% testing subsets, with 10 validation replicates to minimize uncertainty.

2.5. Accuracy and Predictive Ability of the Models

To assess the accuracy and predictive capacity of the models, three validation methods were applied, including metrics such as the area under the ROC curve (AUC), true skill statistic (TSS), sensitivity, and specificity. The AUC was calculated using the roc function from the ‘pROC’ package in R [48], which evaluates the models’ ability to distinguish between species presence and absence; AUC values less than 0.7 were considered poor, values between 0.7 and 0.9 were considered moderate, and values greater than 0.9 were considered excellent [27,49]. True skill statistic was computed with the TSS function from the ‘PresenceAbsence’ package in R [50], a metric that accounts for both sensitivity and specificity. True skill statistic values less than 0.4 were considered “poor”, values between 0.4 and 0.6 were classified as ‘moderate’, and values between 0.6 and 1 were classified as ‘good’ [27]. The TSS value is calculated as sensitivity plus specificity minus 1. Sensitivity (true positives) and specificity (true negatives) were determined using the sensitivity and specificity functions from the ‘caret’ package in R. Sensitivity reflects the probability of correctly predicting a presence, while specificity reflects the probability of correctly predicting an absence [27,51].

2.6. Assessment and Selection of the Best Model

The model with the highest AUC, TSS, sensitivity, and specificity values was selected as the best predictive model for the regional level and within Ecuador’s historical scenario. A higher weight of 3 was assigned to the AUC metric due to its reliability in evaluating model performance, particularly in cases of class imbalance. As the authors of [51] highlight, AUC is a robust measure of a model’s ability to distinguish between presences and absences across various thresholds, making it especially useful for imbalanced datasets common in ecological modeling. Furthermore, Jiménez-Valverde [52] emphasizes that AUC is independent of species prevalence, providing a more universally applicable metric for model comparison. Although TSS is also valuable, its sensitivity to class prevalence, particularly when absence classes dominate, can lead to biased results, as noted by Allouche [53]. Therefore, assigning a weight of 2 to TSS acknowledges its importance, while prioritizing AUC reflects its superior ability to assess discriminative power without being influenced by class imbalance. At the Ecuadorian level, the same selected model (historical) was applied to other scenarios, which allowed us to homogenize the comparison with the same model.

2.7. Model Classification and Area Calculation

The best models selected for both the regional and historical levels in Ecuador, in addition to the models for the current and future scenarios, were divided into two classifications: (1) two categories of distribution and absence (binary classification); and (2) four categories: absence, low distribution, medium distribution, and high distribution (detailed classification). These two classifications enable us to conduct general and exhaustive analyses, respectively, of the potential distribution of A. queko in various ecological modeling scenarios. The classification was performed using the ArcGIS Pro 10.8 software, applying the Jenks natural breaks classification method. To further explore the effects of environmental factors on species distribution, we calculated the percentage of contribution of each predictor variable to the best-fitting distribution model.

3. Results

3.1. Selection of the Best Model

Maxent recorded the highest AUC values, with 0.99 at the regional level and 0.95 at the Ecuadorian level, both considered excellent (Table 2). Similarly, TSS showed the highest values for Maxent, with 0.74 and 0.69 for the regional and Ecuadorian levels, classified as good and moderate, respectively. At the regional level, Maxent outperformed in both sensitivity and specificity, whereas at the Ecuadorian level, GLM had higher sensitivity and RF had higher specificity. Therefore, based on the final rating values for the three models (taking into account the previous weighting), Maxent achieved the highest overall scores and was selected as the best model for both the regional and Ecuadorian levels.

3.2. Potential Distribution of A. queko at the Regional Level

At the regional level, A. queko was found to have a potential distribution area of 264,540 km2 across the four Andean countries (Figure 2A). Peru exhibited the largest area of potential distribution, while Bolivia recorded the smallest. A similar trend was observed in the areas of high potential distribution (Figure 2B), with Bolivia’s high-potential area being notably small, measuring just 693 km2. Furthermore, when comparing the overall potential distribution to the area of the highest potential, there was a decrease of 41%, reducing the area to 110,381 km2.

3.3. Potential Distribution of A. queko at the Ecuadorian Level

Considering the minimum distribution threshold and its binary classification, the historical scenario exhibited the largest potential distribution area at 78,658 km2, compared to the other scenarios (Figure 3A,H). The future scenarios for 2021–2040 and 2014–2060 showed minimal changes in potential distribution (Figure 3B–G,H). Among these, the 21–40–SPP126 scenario demonstrated the greatest potential for future distribution, covering 24,001 km2 (Figure 3C,H). A similar pattern was evident in the detailed classification (Figure 4), where the area of high distribution was greater in the historical scenario, measuring 56,355 km2 (Figure 4A,H). Over the studied period, no significant changes were observed in areas of high potential distribution (Figure 4B–G,H). Notably, the 41-60-SPP245 scenario exhibited the highest potential for future distribution (Figure 4F,H).

3.4. Potential Distribution with Land-Use Changes in Ecuador

The historical scenario, which showed the largest potential distribution area (Figure 5A–C), was intersected with the land-use change layer. After this intersection, the areas of distribution (binary classification) and the high distribution category (detailed classification) decreased by 37% and 35%, respectively (Figure 5A–C). Furthermore, the binary classification scenario indicated a larger potential surface area than the high distribution category of the detailed classification (Figure 5C). Nevertheless, the intersected results revealed that the potential distribution areas are larger than any of the analyzed future modeling scenarios (Figure 5C). Additionally, these results demonstrated that future climate variables have a more definitive impact on reducing potential distribution areas than land-use changes recorded up to 2020.
In the historical scenario overlapped with land-use change, the provinces of Morona Santiago and Zamora Chinchipe showed a larger area of distribution in the binary classification, while Orellana and Los Ríos exhibited smaller surface areas (Figure 5A). Considering the widespread distribution of A. queko (Figure 5B), the provinces of Zamora Chinchipe and Morona Santiago exhibited larger areas of high distribution, in contrast to Pastaza and Guayas.

3.5. Environmental Variables and Their Influence on Potential Distribution Models

At the regional level, the uncorrelated predictor variables used in the model were B1, B2, B4, B15, and B18. Among these, only B4 had a >50% contribution to the model, while the remaining variables each contributed less than 50%. At the Ecuadorian level, the uncorrelated variables B2, B3, B4, B5, B14, B15, B18, and slope were used across all scenarios. Of these, B5 and B14 were the most significant in explaining the models for all scenarios of A. queko distribution. Additionally, in scenarios with a larger predicted area, such as the historical one (Figure 6A), B5, B3, and B14 were the most influential. In the current and near-future scenario named 21–40–SPP126 (Figure 6B), B5, B3, and B18 were the key contributors, while in the future scenario 41-60-SPP245 (Figure 6G), B18, B14, and B5 played the most important roles in the model.

4. Discussion

Before performing the analyses at regional and Ecuadorian scales, it is important to evaluate the data used in selecting the best model. The AUC values for GLM and RF highlight the influence of spatial scale on predictive accuracy. At the regional level, AUC values are lower (GLM: 0.59, RF: 0.49) compared to the national level (0.87 and 0.76, respectively), likely due to greater geographic extension and increased environmental heterogeneity, which complicates species distribution predictions [54]. In contrast, at the national scale, more homogeneous and detailed data improve model performance [55]. Additionally, environmental complexity particularly affects RF, which, despite its flexibility, requires well-structured data to achieve accurate predictions [56].

4.1. Potential Distribution of A. queko at a Regional Scale

In our research, we analyzed at two levels the potential distribution of A. queko, an emblematic species used for handicraft production. At the regional level, we assessed its distribution across four Andean countries in South America, using various species distribution algorithms and finding Maxent to be the best fit. This algorithm combines pre-processed environmental variables and presence records to model the species’ ecological niche and create potential distribution maps. Previous studies confirm that a species’ ecological niche can be quantitatively represented by integrating presence records with biotic and abiotic variables [57]. Zhang [58] used the Maxent model to predict the suitable habitats of Cinnamomum camphora at China under climate change scenarios, demonstrating its effectiveness in diverse environmental conditions. Their findings align with our results, where Maxent also proved to be the best algorithm for modeling the potential distribution of A. queko across in four countries at the Andean region.
At the regional level, the presence of A. queko showed potential distribution areas greater than 50% in Colombia and Peru, both in general and high distribution, compared to smaller areas at Ecuador and, especially, Bolivia. This reflects the broad natural range of the species across the Andean territories. Seasonal temperature played a key role in defining the species’ distribution limits by impacting its survival, reproduction, and resource availability, thus shaping the habitats where A. queko can thrive [59]. This variable may also influence certain ecological interactions and migration patterns [60]. Similar results were found by Perez-Suarez [61], who used the Maxent model to predict the distribution of Pinus hartwegii in Mexico, identifying temperature and precipitation as key factors shaping its habitat. Their findings, particularly regarding the impact of seasonal temperature variations, align with our results, reinforcing the critical role of climatic variables in defining the ecological niches of species across the Andean region.

4.2. Distribution of A. queko in Natural Habitats Within Ecuador

The study used both historical and projected climate scenarios to model the potential distribution of A. queko across Ecuador. We also analyzed distribution changes under three intensity scenarios in SSP—SSP126, SSP245, and SSP585—over two periods: the near future (2020–2040) and the longer-term future (2041–2060). In addition to climate data, the slope was incorporated to minimize uncertainty in distribution models of A. queko in Ecuador. Findings from the historical scenario indicated the largest potential distribution area, suggesting that A. queko currently occupies an optimal range in stable historical climates. The present and near-future projections showed slight reductions in the species’ range, with SSP126 (optimistic) offering the greatest stability. This aligns with studies indicating that lower-emission scenarios are more favorable in maintaining species distribution [62,63]. In line with this, a study by Gonzales-Trujillo [64] investigated the distribution of Andean plant species under climate change and found that regions with lower emission scenarios (SSP126) were less prone to habitat loss, similar to our results for A. queko. Although high-potential areas for A. queko remained relatively stable across scenarios, the regional climate variability presents challenges for persistence. Core habitats for A. queko likely include the regions less affected by climate- and human-driven changes, providing stable conditions that support survival despite environmental shifts.
At Ecuador’s historical context, the distribution of A. queko is primarily driven by three climate factors: the maximum temperature of the warmest month, isothermality, and precipitation during the driest month. High temperatures set an ecological limit, beyond which the species may experience stress or mortality, particularly in warmer regions. This is consistent with findings for other Andean species, such as Puya raimondii, where temperature increases have been linked to habitat contraction and higher mortality rates in the warmer parts of its range [65]. Isothermality plays a fundamental role in maintaining stable growth conditions, being particularly important for mountain species that depend on consistent thermal patterns to regulate their physiological processes. This concept aligns with the statement by Quintero [66] who argues that seasonal variation is a key factor explaining much of the differences in the climatic niche amplitudes among species, as greater thermal stability facilitates more efficient adaptation to environmental changes. Additionally, sufficient rainfall during the driest month is crucial for mitigating drought stress, highlighting water availability as a key factor in the persistence of species across diverse landscapes. The findings of Wang [67] reinforce this notion, demonstrating that precipitation during this critical period plays a fundamental role in determining the suitable areas for species distribution. The species in their study exhibited a strong correlation with the amount of rainfall received during the driest month, with limited precipitation during this period restricting their expansion into areas with inadequate water availability.
Under the most stable present and future scenario, i.e., 21–40 SSP126, the precipitation during the warmest quarter becomes an even more decisive factor for A. queko’s potential distribution, as this coincides with the period of highest environmental demand for moisture. During these months, A. queko likely experiences elevated evaporation and transpiration due to warmer temperatures, making water availability essential. Precipitation during the warmest quarter has been identified as a crucial factor for species distributions, particularly during times of high moisture demand, as observed in studies on Magnolia sieboldii and Haloxylon ammodendron [68,69]. These species rely heavily on water availability during this period, with insufficient rainfall restricting their distribution, particularly under changing climatic conditions. Similarly, for A. queko, the importance of precipitation during the warmest quarter is expected to intensify in future scenarios, as it coincides with the species’ peak moisture demand, highlighting the critical role of water availability in shaping its potential range.

4.3. Distribution of A. queko in the Historical Scenario, Intersected with a Land-Use Change Layer for Ecuador (2020)

This study reveals a significant reduction in the species’ potential distribution due to land-use changes and climate variability. While the historical scenario exhibited the largest distribution area, land-use changes led to a 37% decline in the binary classification, and the high-distribution category decreased by 35%. These results are consistent with findings from other Andean countries, where habitat loss and climate change are primary drivers of species distribution shifts [70,71]. Furthermore, the interaction between climate change and land-use transformations exacerbates habitat degradation, complicating conservation planning [72].
Although climate variables have a more significant impact on habitat reduction than land-use changes, driving shifts in species distributions and altering habitat suitability [72], some suggest that land-use change could ultimately have a more significant impact than climate change, given its long-standing nature and its likely limited connection to climate modification [73,74]. However, recent studies have documented a strong relationship between land-use change and climate variation, both contributing significantly to future habitat loss [75]. In Colombia, deforestation and agricultural expansion have dramatically reduced suitable habitats for endemic species [76], while in Peru, montane forests are threatened by climate-induced range shifts [77]. In Ecuador, regional variability across provinces emphasizes the need for targeted conservation efforts, with some areas acting as refuges and others requiring urgent intervention. Similarly, studies in Argentina project severe range reductions for historically widespread species in the future [78]. These findings highlight the need to integrate land-use policies with climate adaptation strategies for effective conservation planning.

5. Conclusions

Aulonemia queko occupies a substantial range across the Andes, predominantly in Peru, with smaller, less extensive areas in Bolivia. In Ecuador, however, it has notable coverage relative to its overall distribution in the Andes, with its range showing significant shifts from historical to current and projected future scenarios. The potential distribution of A. queko is heavily influenced by climatic variables, particularly temperature and precipitation, underscoring the species’ sensitivity to seasonal changes in moisture availability. This highlights the critical need to incorporate climate projections into habitat conservation planning, especially in regions where the species naturally thrives. Land-use changes have reduced suitable habitats for A. queko by over 35%, directly threatening the species’ ability to find resources, reproduce, and maintain viable populations, underscoring the severe impact of human activities on its long-term survival. Combined with climate-driven environmental shifts that further decrease potential distribution, these findings stress the urgent need for conservation strategies to mitigate habitat loss in these biodiversity-rich areas.
Incentives for the conservation of natural cover, environmental education and awareness, the promotion of sustainable agricultural practices, economic alternatives for local communities, and the restoration of degraded areas represent challenging but viable alternatives to halt land-use change in Ecuador. These actions would contribute to reducing the loss of natural habitats, thus protecting the flora and fauna of a megadiverse country. By integrating these approaches, it is possible to create a more sustainable balance between development and conservation, ensuring the long-term survival of Ecuador’s unique biodiversity.

Author Contributions

H.C.: Conceptualization, data curation, methodology, software, visualization, writing—original draft, writing—review and editing. L.G.G.-M.: Visualization, writing—original draft, writing—review and editing. O.C.: Methodology, software, supervision, validation. M.R. and A.A.: Investigation, methodology. O.J.: Funding acquisition, investigation, project administration, resources, conceptualization, data curation, methodology, software, visualization, writing—original draft, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data will be made available on request.

Acknowledgments

The authors express their gratitude to the Vice-Rectorate for Research of the University of Cuenca (Vicerrectorado de Investigación de la Universidad de Cuenca, VIUC, https://www.ucuenca.edu.ec/investigacion-innovacion/), Ecuador, accessed on 4 January 2024, for providing financial support for the research project “El Rol de los bosques andinos frente al cambio climático con base a la relación taxonómica y funcional de la vegetación leñosa con los stocks de carbono, Azuay—Ecuador”. The authors also thank VIUC for allocating part of our workload to the production of this paper. Omar Cabrera thanks the UTPL for their support in the culmination of the manuscript. The authors thank Diana Szekely for her help in the language revision. This research was conducted under permit No. 214-19-IC-FLO-DPAA/MA, granted by the Ministry of the Environment of Ecuador.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Abbreviations

AUCArea under the curve
TSSTrue skill statistic
GLMsGeneralized linear models
RFRandom forest
GBIFGlobal Biodiversity Information Facility
SSPsShared socioeconomic pathways
MAATEMinistry of Environment and Water and the Ecological Transition of Ecuador

Appendix A. Climatic Variables and Elevation Used to Generate Potential Distribution Models. The Meaning of the Variables and Their Units Can Be Found in Table 1 of the Main Text

SpecieLongitudLatitudeB1B2B3B4B5B6B7B8B9B10B11B12B13B14B15B16B17B18B19Elevation (m)
Aulonemia queko−77.01.213.79.387.846.118.78.110.613.413.914.113.02088235.6129.519.9661.6418.0572.0533.42483
Aulonemia queko−78.2−1.011.610.083.153.317.95.912.011.311.912.110.91114130.753.522.8352.7191.0254.5341.52925
Aulonemia queko−75.0−11.59.613.279.449.017.20.616.69.69.010.29.01020160.516.060.9451.066.0309.366.04004
Aulonemia queko−73.6−11.618.812.482.061.626.010.915.118.718.119.618.11705236.745.048.8693.0179.7407.4184.52039
Aulonemia queko−79.4−2.89.39.882.052.215.33.411.99.78.69.78.5974137.732.843.3389.5118.3342.3123.33310
Aulonemia queko−79.5−2.812.910.084.034.719.07.012.013.212.413.212.4767151.24.083.2418.025.0317.028.72649
Aulonemia queko−79.4−2.810.110.182.549.716.34.112.210.59.410.59.4917138.626.450.1388.997.2331.297.93155
Aulonemia queko−78.1−0.29.310.688.246.015.23.212.09.39.49.68.61212118.778.513.5353.0263.0273.5320.23282
Aulonemia queko−79.2−3.916.910.381.540.423.711.012.717.117.117.316.3834113.046.028.8282.0146.4158.8203.22046
Aulonemia queko−79.1−1.814.110.182.732.920.48.212.214.413.714.413.7856177.35.588.0495.337.0399.037.02328
Aulonemia queko−72.5−12.518.313.181.664.726.210.116.118.317.519.117.51543239.335.557.2676.3141.8385.3143.02248
Aulonemia queko−79.2−1.619.49.088.134.424.414.210.219.819.119.919.11674342.58.792.5944.851.5771.365.51144
Aulonemia queko−76.91.112.98.786.251.517.77.610.112.613.213.312.22474338.0135.032.1900.7424.0664.0782.52615
Aulonemia queko−75.6−10.414.312.379.962.421.15.815.414.413.414.913.4858127.818.858.5376.168.7269.568.72860
Aulonemia queko−67.8−16.313.712.675.9107.121.24.616.614.612.314.712.3889183.316.071.6454.563.0373.563.02937
Aulonemia queko−67.9−16.314.612.576.399.421.95.516.415.413.215.513.2934192.015.072.5479.861.3393.361.32736
Aulonemia queko−76.1−9.817.513.281.850.125.18.916.217.317.018.217.0643102.811.364.7287.340.0161.340.02449
Aulonemia queko−76.71.911.110.188.343.417.05.511.511.210.511.510.51837219.399.929.7626.8321.8522.9324.72904
Aulonemia queko−75.8−9.519.812.483.847.726.812.014.819.519.420.419.4828116.227.350.0330.583.5203.283.52144
Aulonemia queko−78.8−2.915.611.685.347.622.79.213.615.915.916.014.9873101.857.119.4276.6178.5244.0215.72280

References

  1. Perosa, M.; Rojas, F.; Villagra, P.; Tognelli, M.F.; Carrara, R.; Alvarez, J.A. Distribución potencial de los bosques de Prosopis fexuosa en la Provincia Biogeográfica del Monte (Argentina). Ecol. Austral 2014, 24, 238–248. [Google Scholar] [CrossRef]
  2. Guisan, A.; Thuiller, W. Predicting species distribution: Offering more than simple habitat models. Ecol. Lett. 2005, 8, 993–1009. [Google Scholar] [CrossRef] [PubMed]
  3. Palma-Ordaz, S.; Delgadillo-Rodríguez, J. Distribución potencial de ocho especies exóticas de carácter invasor en el estado de Baja California, México. Bot. Sci. 2014, 92, 587–597. [Google Scholar] [CrossRef]
  4. Sánchez, C.; Townsend, A.; Escalante, P. El modelado de la distribución de especies y la conservación de la diversidad biológica. In Enfoques Contemporáneos en el Estudio de la Diversidad Biológica; Instituto de Biología, UNAM y Academia Mexicana de Ciencias: Mexico city, Mexico, 2001; pp. 359–379. Available online: https://www.researchgate.net/publication/230709853 (accessed on 1 October 2024).
  5. Maguire, K.C.; Nieto-Lugilde, D.; Fitzpatrick, M.C.; Williams, J.W.; Blois, J.L. Modeling species and community responses to past, present, and future episodes of climatic and ecological change. Annu. Rev. Ecol. Evol. Syst. 2015, 46, 343–368. [Google Scholar] [CrossRef]
  6. Haddad, N.M.; Brudvig, L.A.; Clobert, J.; Davies, K.F.; Gonzalez, A.; Holt, R.D.; Lovejoy, T.E.; Sexton, J.O.; Austin, M.P.; Collins, C.D. Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci. Adv. 2015, 1, e1500052. [Google Scholar] [CrossRef]
  7. Viana, P.L. O gênero Aulonemia Goudot (Poaceae: Bambusoideae: Bambuseae) no Brasil. Ph.D. Thesis, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil, 2010. Available online: http://hdl.handle.net/1843/BUOS-8XUNDD (accessed on 14 November 2024).
  8. Londoño, X. El bambú en Colombia. Biotecnología Vegetal. 2011, 11 (3). Available online: https://revista.ibp.co.cu/index.php/BV/article/view/485/881 (accessed on 15 November 2024).
  9. Camarillo Cuenca, A.Y.; León-Merino, A.; Sangerman-Jarquín, D.M.; Hernández-Juárez, M.; Zamora-Martínez, M.C. Aspectos socioeconómicos del aprovechamiento del bambú en una comunidad rural de Veracruz, México. Rev. Mex. Cienc. For. 2020, 11, 33–54. [Google Scholar] [CrossRef]
  10. Linares, E.L. Inventario preliminar de las plantas utilizadas para elaborar artesanías en Colombia. Univ. Sci. 1994, 2, 7–43. Available online: https://bit.ly/3Bu0TN1 (accessed on 14 November 2024).
  11. Londoño, X. Distribución, morfología, taxonomía, anatomía, silvicultura y usos de los bambúes del Nuevo Mundo. In Cátedra Maestria en Construccion. Módulo Guadua, Arquitectura; Universidad Nacional de Colombia: Bogotá, Colombia, 2002; Available online: http://www.maderinsa.com/guadua/taller.html (accessed on 18 November 2024).
  12. Aguirre, Z.; Aguirre, L. Estado actual e importancia de los Productos Forestales No Maderables. Bosques Latid. Cero 2021, 11, 71–82. [Google Scholar]
  13. Bystriakova, N.; Kapos, V.; Lysenko, I. Bamboo Biodiversity: Africa, Madagascar and the Americas; UNEP/Earthprint: Nairobi, Kenya, 2004; Available online: https://bit.ly/3ZxfGi9 (accessed on 16 November 2024).
  14. Vidaurre, P.J.; Paniagua, N.; Moraes, M. Etnobotánica en los Andes de Bolivia. Botánica Económica Los Andes Cent. 2006, 1, 224–238. [Google Scholar]
  15. Giraldo Cañas, D.A. Las Gramíneas en Colombia: Riqueza, Distribución, Endemismo, Invasión, Migración, Usos y Taxonomías Populares; Instituto de Ciencias Naturales, Universidad Nacional de Colombia: Bogotá, Colombia, 2013; Available online: https://bit.ly/4exkJ7r (accessed on 18 November 2024).
  16. Añazco, M. Estudio de Vulnerabilidad del Bambú (Guadua angustifolia) al Cambio Climático; Red Internacional del Bambú y Ratán INBAR: Quito, Ecuador, 2013; Available online: https://bit.ly/3ZuvP7K (accessed on 18 November 2024).
  17. Goudot, J. Aulonemia, nouveau genre de la tribu des Bambusées. In Annales des Sciences Naturelles; Fortin, Masson: Paris, France, 1846; Volume 3, pp. 75–77. Available online: https://www.biodiversitylibrary.org/page/41542486 (accessed on 20 November 2024).
  18. Valdez Cancinos, D. Manual Para el Cultivo de Bambú Experiencias en Guatemala; Ministerio de Agricultura, Ganadería y Alimentación: Guatemala City, Guatemala, 2013; Available online: https://bit.ly/41yTnLv (accessed on 22 November 2024).
  19. Hachmeyer, S. Modelando el nicho ecológico de bambúes leñosos (Bambuseae, Poaceae) en Bolivia: Implicaciones para la construcción de aerófonos autóctonos y la conservación de bambúes musicales. In la rebelión de los objetos Cestería y Maderas: Anales de la reunión anual de etnología; MUSEF: La Paz, Bolivia, 2018; Volume XXXI, pp. 183–210. Available online: https://bit.ly/412SgC0 (accessed on 25 September 2024).
  20. Jadán, O.; Donoso, D.A.; Cedillo, H.; Bermúdez, F.; Cabrera, O. Floristic groups, and changes in diversity and structure of trees, in tropical montane forests in the Southern Andes of Ecuador. Diversity 2021, 13, 400. [Google Scholar] [CrossRef]
  21. Londono, C.; Cleef, A.; Madriñán, S. Angiosperm flora and biogeography of the páramo region of Colombia, Northern Andes. Flora-Morphol. Distrib. Funct. Ecol. Plants 2014, 209, 81–87. [Google Scholar] [CrossRef]
  22. Lizarazu, M.A. Bambúes Leñosos (Poaceae: Bambusoideae: Bambuseae) del Noreste Argentino y Regiones Limítrofes: Estudios Taxonómicos, Morfológicos, Anatómicos y Biogeográficos. Ph.D. Thesis, Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales, Buenos Aires, Argentina, 2013. Available online: https://bit.ly/3Bj4YDV (accessed on 25 November 2024).
  23. Lobovikov, M.; Ball, L.; Guardia, M. World Bamboo Resources: A Thematic Study Prepared in the Framework of the Global Forest Resources Assessment 2005; Food & Agriculture Organization Of The United Nations: Rome, Italy, 2007; Available online: https://bit.ly/4iyMaBb (accessed on 2 December 2024).
  24. Navarro, M.; Franco, P.; Oyague, E.; Garitano, A.; Abarca, J. Importancia del Modelamiento del Nicho Ecológico del Suri Rhea Tarapacensis (Chubb 1913) Para su Conservación en los Andes Sudamericanos. In Tópicos en Biodiversidad Transfronteriza Chile, Perú y Bolivia; RIL Editores: Valencia, Spain, 2021; pp. 183–202. Available online: https://bit.ly/4izYf98 (accessed on 4 December 2024).
  25. Elith, J.; Phillips, S.J.; Hastie, T.; Dudík, M.; Chee, Y.E.; Yates, C.J. A statistical explanation of MaxEnt for ecologists. Divers. Distrib. 2011, 17, 43–57. [Google Scholar] [CrossRef]
  26. Phillips, S.J.; Anderson, R.P.; Schapire, R.E. Maximum entropy modeling of species geographic distributions. Ecol. Model. 2006, 190, 231–259. [Google Scholar] [CrossRef]
  27. Luo, W.; Han, S.; Yu, T.; Wang, P.; Ma, Y.; Wan, M.; Liu, J.; Li, Z.; Tao, J. Assessing the suitability and dynamics of three medicinal Sambucus species in China under current and future climate scenarios. Front. Plant Sci. 2023, 14, 1194444. [Google Scholar] [CrossRef]
  28. McCullagh, P.; Nelder, J. Generalized Linear Models, 2nd ed.; Hall, C.A., Ed.; Routledge: London, UK, 2019. [Google Scholar]
  29. De Araújo, C.B.; Marcondes-Machado, L.O.; Costa, G.C. The importance of biotic interactions in species distribution models: A test of the Eltonian noise hypothesis using parrots. J. Biogeogr. 2014, 41, 513–523. [Google Scholar] [CrossRef]
  30. Guisan, A.; Zimmermann, N.E. Predictive habitat distribution models in ecology. Ecol. Model. 2000, 135, 147–186. [Google Scholar] [CrossRef]
  31. Cutler, D.R.; Edwards, T.C., Jr.; Beard, K.H.; Cutler, A.; Hess, K.T.; Gibson, J.; Lawler, J.J. Random forests for classification in ecology. Ecology 2007, 88, 2783–2792. [Google Scholar] [CrossRef] [PubMed]
  32. Blanco, J.A.; Ameztegui, A.; Rodríguez, F. Modelling Forest Ecosystems: A crossroad between scales, techniques and applications. Ecol. Model. 2020, 425, 109030. [Google Scholar] [CrossRef]
  33. Mascaro, J.; Asner, G.P.; Knapp, D.E.; Kennedy-Bowdoin, T.; Martin, R.E.; Anderson, C.; Higgins, M.; Chadwick, K.D. A tale of two “forests”: Random Forest machine learning aids tropical forest carbon mapping. PLoS ONE 2014, 9, e85993. [Google Scholar] [CrossRef]
  34. Chamberlain, S.; Oldoni, D.; Waller, J. rgbif: Interface to the Global Biodiversity Information Facility API, version 3.8.1; 2022. Available online: https://cran.r-project.org/package=rgbif (accessed on 12 September 2024).
  35. Zizka, A.; Silvestro, D.; Andermann, T.; Azevedo, J.; Duarte Ritter, C.; Edler, D.; Farooq, H.; Herdean, A.; Ariza, M.; Scharn, R. CoordinateCleaner: Standardized cleaning of occurrence records from biological collection databases. Methods Ecol. Evol. 2019, 10, 744–751. [Google Scholar] [CrossRef]
  36. 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]
  37. Hijmans, R.J.; Van Etten, J.; Cheng, J.; Mattiuzzi, M.; Sumner, M.; Greenberg, J.A.; Lamigueiro, O.P.; Bevan, A.; Racine, E.B.; Shortridge, A. Package ‘raster’. R Package 2015, 734, 473. [Google Scholar]
  38. Riahi, K.; Van Vuuren, D.P.; Kriegler, E.; Edmonds, J.; O’neill, B.C.; Fujimori, S.; Bauer, N.; Calvin, K.; Dellink, R.; Fricko, O. The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: An overview. Glob. Environ. Chang. 2017, 42, 153–168. [Google Scholar] [CrossRef]
  39. Ministerio de Ambiente y Transición Ecológica de Ecuador. Capa de Cobertura Vegetal y uso del Suelo de 2020 [Shapefile]; Ministerio de Ambiente y Transición Ecológica de Ecuador: Ibarra, Ecuador, 2024. Available online: https://www.ambiente.gob.ec (accessed on 26 September 2024).
  40. Kuhn, M.; Wing, J.; Weston, S.; Williams, A.; Keefer, C.; Engelhardt, A.; Cooper, T.; Mayer, Z.; Kenkel, B.; Team, R.C. Package ‘caret’. R J. 2020, 223, 48. [Google Scholar]
  41. Cotrina Sanchez, A.; Rojas Briceno, N.B.; Bandopadhyay, S.; Ghosh, S.; Torres Guzman, C.; Oliva, M.; Guzman, B.K.; Salas Lopez, R. Biogeographic distribution of Cedrela spp. genus in Peru using MaxEnt modeling: A conservation and restoration approach. Diversity 2021, 13, 261. [Google Scholar] [CrossRef]
  42. Nelder, J.A.; Wedderburn, R.W. Generalized linear models. J. R. Stat. Soc. Ser. A Stat. Soc. 1972, 135, 370–384. [Google Scholar] [CrossRef]
  43. Tshwene-Mauchaza, B.; Aguirre-Gutiérrez, J. Climatic drivers of plant species distributions across spatial grains in Southern Africa tropical forests. Front. For. Glob. Change 2019, 2, 69. [Google Scholar] [CrossRef]
  44. Burns, P.A.; Clemann, N.; White, M. Testing the utility of species distribution modelling using Random Forests for a species in decline. Austral Ecol. 2020, 45, 706–716. [Google Scholar] [CrossRef]
  45. Valavi, R.; Elith, J.; Lahoz-Monfort, J.J.; Guillera-Arroita, G. Modelling species presence-only data with random forests. Ecography 2021, 44, 1731–1742. [Google Scholar] [CrossRef]
  46. Hijmans, R.J.; Phillips, S.; Leathwick, J.; Elith, J.; Hijmans, M.R.J. Package ‘dismo’. Circles 2017, 9, 1–68. [Google Scholar]
  47. Breiman, L.; Cutler, A.; Liaw, A.; Wiener, M. Package ‘Randomforest’, version 4.7-1.2; University of California: Berkeley, CA, USA, 2018; Available online: https://bit.ly/4gC8fwQ (accessed on 23 September 2024).
  48. Robin, X.; Turck, N.; Hainard, A.; Tiberti, N.; Lisacek, F.; Sanchez, J.-C.; Müller, M.; Siegert, S.; Doering, M.; Billings, Z. Package ‘pROC’. 2021. Available online: https://cran.r-project.org/web/packages/pROC/pROC.pdf (accessed on 26 September 2024).
  49. Li, J.; Fan, G.; He, Y. Predicting the current and future distribution of three Coptis herbs in China under climate change conditions, using the MaxEnt model and chemical analysis. Sci. Total Environ. 2020, 698, 134141. [Google Scholar] [CrossRef] [PubMed]
  50. Freeman, E.A.; Moisen, G. PresenceAbsence: An R package for presence absence analysis. J. Stat. Softw. 2008, 23, 31p. [Google Scholar] [CrossRef]
  51. Liu, C.; White, M.; Newell, G. Measuring and comparing the accuracy of species distribution models with presence–absence data. Ecography 2011, 34, 232–243. [Google Scholar] [CrossRef]
  52. Jiménez-Valverde, A. Insights into the area under the receiver operating characteristic curve (AUC) as a discrimination measure in species distribution modelling. Glob. Ecol. Biogeogr. 2012, 21, 498–507. [Google Scholar] [CrossRef]
  53. Allouche, O.; Tsoar, A.; Kadmon, R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 2006, 43, 1223–1232. [Google Scholar] [CrossRef]
  54. Willis, K.J.; Whittaker, R.J. Species diversity—Scale matters. Science 2002, 295, 1245–1248. [Google Scholar] [CrossRef]
  55. Abrego, N.; Norberg, A.; Ovaskainen, O. Measuring and predicting the influence of traits on the assembly processes of wood-inhabiting fungi. J. Ecol. 2017, 105, 1070–1081. [Google Scholar] [CrossRef]
  56. Garcia-Carretero, R.; Holgado-Cuadrado, R.; Barquero-Pérez, Ó. Assessment of classification models and relevant features on nonalcoholic steatohepatitis using random forest. Entropy 2021, 23, 763. [Google Scholar] [CrossRef]
  57. Anderson, R.P.; Lew, D.; Peterson, A.T. Evaluating predictive models of species’ distributions: Criteria for selecting optimal models. Ecol. Model. 2003, 162, 211–232. [Google Scholar] [CrossRef]
  58. Zhang, L.; Jing, Z.; Li, Z.; Liu, Y.; Fang, S. Predictive modeling of suitable habitats for Cinnamomum camphora (L.) presl using maxent model under climate change in China. Int. J. Environ. Res. Public Health 2019, 16, 3185. [Google Scholar] [CrossRef] [PubMed]
  59. Kearney, M.; Porter, W. Mechanistic niche modelling: Combining physiological and spatial data to predict species’ ranges. Ecol. Lett. 2009, 12, 334–350. [Google Scholar] [CrossRef] [PubMed]
  60. Forrest, J.; Miller-Rushing, A.J. Toward a synthetic understanding of the role of phenology in ecology and evolution. Philosophical Transactions of the Royal Society B. Biol. Sci. 2010, 365, 3101–3112. [Google Scholar] [CrossRef] [PubMed]
  61. Pérez-Suárez, M.; Ramírez-Albores, J.E.; Vargas-Hernández, J.J.; Alfaro-Ramírez, F.U. A review of the knowledge of Hartweg’s Pine (Pinus hartwegii Lindl.): Current situation and the need for improved future projections. Trees 2022, 36, 25–37. [Google Scholar] [CrossRef]
  62. Hannah, L.; Ikegami, M.; Hole, D.G.; Seo, C.; Butchart, S.H.; Peterson, A.T.; Roehrdanz, P.R. Global climate change adaptation priorities for biodiversity and food security. PLoS ONE 2013, 8, e72590. [Google Scholar] [CrossRef] [PubMed]
  63. Warren, R.; Price, J.; Fischlin, A.; de la Nava Santos, S.; Midgley, G. Increasing impacts of climate change upon ecosystems with increasing global mean temperature rise. Clim. Change 2011, 106, 141–177. [Google Scholar] [CrossRef]
  64. González-Trujillo, J.D.; Escobar-Alba, M.R.; Lara, D.E.; Carvajal-C, J.E. Mapping the threat: Projecting invasive plant distribution in the tropical Andes under climate change. Perspect. Ecol. Conserv. 2024, 22, 348–357. [Google Scholar] [CrossRef]
  65. Mansfield, N. A Biologically Meaningful Evaluation of Phenological Responses to Climate Change. Ph.D. Thesis, University of Plymouth, Plymouth, UK, 2022. Available online: https://pearl.plymouth.ac.uk/cgi/viewcontent.cgi?article=1415&context=bms-theses (accessed on 29 November 2024).
  66. Quintero, I.; Wiens, J.J. What determines the climatic niche width of species? The role of spatial and temporal climatic variation in three vertebrate clades. Glob. Ecol. Biogeogr. 2013, 22, 422–432. [Google Scholar] [CrossRef]
  67. Lan, Z.; Huiliang, L.; Hongxiang, Z.; Yanfeng, C.; Lingwei, Z.; Kudusi, K.; Taxmamat, D.; Yuanming, Z. Potential distribution of three types of ephemeral plants under climate changes. Front. Plant Sci. 2022, 13, 1035684. [Google Scholar] [CrossRef] [PubMed]
  68. Li, M.; Zheng, C.-F.; Gao, X.-Q.; Li, C.-H.; Li, Y.-X.; Xia, X.-H.; Yang, J.; Zheng, Y.-Q.; Huang, P. Distinct Ecological Habits and Habitat Responses to Future Climate Change in Two Subspecies of Magnolia sieboldii K. Koch, a Tree Endemic to East Asia. Plants 2024, 13, 3097. [Google Scholar] [CrossRef] [PubMed]
  69. Xiao, F.; Liu, Q.; Qin, Y. Predicting the Potential Distribution of Haloxylon ammodendron under Climate Change Scenarios Using Machine Learning of a Maximum Entropy Model. Biology 2023, 13, 3. [Google Scholar] [CrossRef]
  70. Manchego, C.E.; Hildebrandt, P.; Cueva, J.; Espinosa, C.I.; Stimm, B.; Günter, S. Climate change versus deforestation: Implications for tree species distribution in the dry forests of southern Ecuador. PLoS ONE 2017, 12, e0190092. [Google Scholar] [CrossRef]
  71. Larsen, T.H.; Brehm, G.; Navarrete, H.; Franco, P.; Gomez, H.; Mena, J.L.; Morales, V.; Argollo, J.; Blacutt, L.; Canhos, V. Range shifts and extinctions driven by climate change in the tropical Andes: Synthesis and directions. In Climate Change and Biodiversity in the Tropical Andes; Inter-American Institute for Global Change Research (IAI); Scientific Committee on Problems of the Environment (SCOPE): Paris, France, 2011; Volume 1, pp. 47–67. Available online: https://core.ac.uk/download/pdf/48017894.pdf#page=60 (accessed on 11 December 2024).
  72. Prokopová, M.; Salvati, L.; Egidi, G.; Cudlín, O.; Včeláková, R.; Plch, R.; Cudlín, P. Envisioning present and future land-use change under varying ecological regimes and their influence on landscape stability. Sustainability 2019, 11, 4654. [Google Scholar] [CrossRef]
  73. Pereira, H.M.; Leadley, P.W.; Proença, V.; Alkemade, R.; Scharlemann, J.P.; Fernandez-Manjarrés, J.F.; Araújo, M.B.; Balvanera, P.; Biggs, R.; Cheung, W.W. Scenarios for global biodiversity in the 21st century. Science 2010, 330, 1496–1501. [Google Scholar] [CrossRef]
  74. Jantz, S.M.; Barker, B.; Brooks, T.M.; Chini, L.P.; Huang, Q.; Moore, R.M.; Noel, J.; Hurtt, G.C. Future habitat loss and extinctions driven by land-use change in biodiversity hotspots under four scenarios of climate-change mitigation. Conserv. Biol. 2015, 29, 1122–1131. [Google Scholar] [CrossRef]
  75. Dale, V.H. The relationship between land-use change and climate change. Ecol. Appl. 1997, 7, 753–769. [Google Scholar] [CrossRef]
  76. Bailey, T.E. Investigating the Effect of Multitemporal Land Use Changes to the Middle Magdalena River Valley on the Abundance and Distribution of Important Plant Species. Master’s Thesis, University of Exeter, Exeter, UK, 2022. Available online: https://bit.ly/40B1Vzo (accessed on 11 December 2024).
  77. Bax, V.; Castro-Nunez, A.; Francesconi, W. Assessment of potential climate change impacts on montane forests in the peruvian andes: Implications for conservation prioritization. Forests 2021, 12, 375. [Google Scholar] [CrossRef]
  78. Giudicelli, G.C.; Turchetto, C.; Silva-Arias, G.A.; Freitas, L.B. Influence of climate changes on the potential distribution of a widespread grassland species in South America. Perspect. Plant Ecol. Evol. Syst. 2019, 41, 125496. [Google Scholar] [CrossRef]
Diversity 17 00167 g001
Figure 1. (A) Regional-scale map centered on Colombia, Ecuador, Peru, and Bolivia, the countries where there are presence records of A. queko. (B) Local-scale map of continental Ecuador showing its political division (provinces in orange) and its geographical area outlined in red.
Figure 1. (A) Regional-scale map centered on Colombia, Ecuador, Peru, and Bolivia, the countries where there are presence records of A. queko. (B) Local-scale map of continental Ecuador showing its political division (provinces in orange) and its geographical area outlined in red.
Diversity 17 00167 g001
Diversity 17 00167 g002
Figure 2. Potential distribution of A. queko at the regional level across Colombia, Ecuador, Peru, and Bolivia. (A) Binary classification. (B) Detailed classification.
Figure 2. Potential distribution of A. queko at the regional level across Colombia, Ecuador, Peru, and Bolivia. (A) Binary classification. (B) Detailed classification.
Diversity 17 00167 g002
Diversity 17 00167 g003
Figure 3. Binary classification maps illustrating the potential changes in the distribution of A. queko. (A) Historical scenario. (B–D) Current and near-future scenarios 2020–2040. (E–G) Future scenarios 2041–2060. (H) Area comparisons between scenarios. The letters up to G have the same meaning in Figure (H).
Figure 3. Binary classification maps illustrating the potential changes in the distribution of A. queko. (A) Historical scenario. (B–D) Current and near-future scenarios 2020–2040. (E–G) Future scenarios 2041–2060. (H) Area comparisons between scenarios. The letters up to G have the same meaning in Figure (H).
Diversity 17 00167 g003
Diversity 17 00167 g004
Figure 4. Detailed classification maps illustrating potential changes in the distribution of A. queko. (A) Historical scenario. (B–D) Current and near-future scenarios (2020–2040). (E–G) Future scenarios (2041–2060). (H) Area comparisons between scenarios. The letters up to G have the same meaning in Figure (H).
Figure 4. Detailed classification maps illustrating potential changes in the distribution of A. queko. (A) Historical scenario. (B–D) Current and near-future scenarios (2020–2040). (E–G) Future scenarios (2041–2060). (H) Area comparisons between scenarios. The letters up to G have the same meaning in Figure (H).
Diversity 17 00167 g004
Diversity 17 00167 g005
Figure 5. Distribution maps of A. queko, considering the historical scenario and changes in land use in Ecuador until 2020, illustrated through binary (A) and detailed (B) classifications. In (C), the distribution of A. queko across two categories is shown and the letters represent the following: (A)—historical scenario; (B–D)—current and near-future scenarios (2020–2040); (E–G)—future scenarios (2041–2060); H—historical scenario intercepted with a land-use change layer of Ecuador (2020).
Figure 5. Distribution maps of A. queko, considering the historical scenario and changes in land use in Ecuador until 2020, illustrated through binary (A) and detailed (B) classifications. In (C), the distribution of A. queko across two categories is shown and the letters represent the following: (A)—historical scenario; (B–D)—current and near-future scenarios (2020–2040); (E–G)—future scenarios (2041–2060); H—historical scenario intercepted with a land-use change layer of Ecuador (2020).
Diversity 17 00167 g005
Diversity 17 00167 g006
Figure 6. Uncorrelated environmental variables and their respective percentage of explanation in the models for the different scenarios (A–H) of the potential distribution of A. queko at the regional level and within Ecuador. The meaning of the constant climatic variables is given in Table 1.
Figure 6. Uncorrelated environmental variables and their respective percentage of explanation in the models for the different scenarios (A–H) of the potential distribution of A. queko at the regional level and within Ecuador. The meaning of the constant climatic variables is given in Table 1.
Diversity 17 00167 g006
Table 1. Climatic variables used for modeling the potential distribution of A. queko at the regional (four South American countries) and local (Ecuador) levels.
Table 1. Climatic variables used for modeling the potential distribution of A. queko at the regional (four South American countries) and local (Ecuador) levels.
CodeVariable—UnitRangeAverage
B1Annual mean temperature (°C)9.4–19.912.5
B2Mean diurnal range (mean of monthly (max temp–min temp)) (°C)9.0–11.610.3
B3Isothermality (BIO2/BIO7) (×100) (%)81.5–8883.9
B4Temperature seasonality (standard deviation ×100) (°C)20.4–65.342.9
B5Max temperature of warmest month (°C)15.5–24.918.7
B6Min temperature of coldest month (°C)3.5–14.66.5
B7Temperature annual range (BIO5-BIO6) (°C)10.3–13.612.2
B8Mean temperature of wettest quarter (°C)9.7–20.412.7
B9Mean temperature of driest quarter (°C)8.7–19.412.1
B10Mean temperature of warmest quarter (°C)9.9–20.412.9
B11Mean temperature of coldest quarter (°C)8.6–19.511.9
B12Annual precipitation (mm)717–1687903.1
B13Precipitation of wettest month (mm)103–351143
B14Precipitation of driest month (mm)1–7927.1
B15Precipitation seasonality (coefficient of variation) (%)13.8–96.156.2
B16Precipitation of wettest quarter (mm)277–962386.4
B17Precipitation of driest quarter (mm)12–26397.5
B18Precipitation of warmest quarter (mm)147–782309.5
B19Precipitation of coldest quarter (mm)12–346120.4
Table 2. Accuracy evaluation of the GLM, Maxent, and RF models to predict the potential distribution of A. queko at the regional and Ecuadorian levels.
Table 2. Accuracy evaluation of the GLM, Maxent, and RF models to predict the potential distribution of A. queko at the regional and Ecuadorian levels.
LevelMethodsAUCTSSSensitivitySpecificityScore
RegionalMaxent0.990.740.660.775.91
GLM0.59−0.030.620.332.66
RF0.490012.47
Ecuador (Historical)Maxent0.950.690.200.294.3
GLM0.87−0.340.600.061.38
RF0.760.410.450.963.89
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

Cedillo, H.; García-Montero, L.G.; Cabrera, O.; Rocano, M.; Arciniegas, A.; Jadán, O. Modeling the Potential Distribution of Aulonemia queko: Historical, Current, and Future Scenarios in Ecuador and Other Andean Countries. Diversity 2025, 17, 167. https://doi.org/10.3390/d17030167

AMA Style

Cedillo H, García-Montero LG, Cabrera O, Rocano M, Arciniegas A, Jadán O. Modeling the Potential Distribution of Aulonemia queko: Historical, Current, and Future Scenarios in Ecuador and Other Andean Countries. Diversity. 2025; 17(3):167. https://doi.org/10.3390/d17030167

Chicago/Turabian Style

Cedillo, Hugo, Luis G. García-Montero, Omar Cabrera, Mélida Rocano, Andrés Arciniegas, and Oswaldo Jadán. 2025. "Modeling the Potential Distribution of Aulonemia queko: Historical, Current, and Future Scenarios in Ecuador and Other Andean Countries" Diversity 17, no. 3: 167. https://doi.org/10.3390/d17030167

APA Style

Cedillo, H., García-Montero, L. G., Cabrera, O., Rocano, M., Arciniegas, A., & Jadán, O. (2025). Modeling the Potential Distribution of Aulonemia queko: Historical, Current, and Future Scenarios in Ecuador and Other Andean Countries. Diversity, 17(3), 167. https://doi.org/10.3390/d17030167

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