Carapichea ipecacuanha (Brot.) Stokes (Rubiaceae): The three last Latin American groups at risk of extinction under pressure from deforestation and climate change

ABSTRACT

Global climate change is currently a serious threat to biodiversity. The understanding of the effects of climate changes on the flora is urgent. Here, we analyze the effects of future climate change on the spatial distribution of the three remaining groups of Carapichea ipecacuanha (Poaia) in the Americas. We built niche-based models for current and optimistic and pessimistic climate scenarios projected for 2050 and 2090. Our findings showed that climatically suitable areas for the Atlantic and Amazon Forest groups will be strongly reduced. The greatest risk of extinction was observed for the Amazon group due to the drastic spatial reduction and abrupt spatial displacement of climate suitability. The climate suitability for the Atlantic Forest group will be reduced, comprising a small suitable area in the transition zone between Minas Gerais and Espirito Santo. The climate suitability for the Panama group remained in an continuous region in Panama and Colombia. Our results encourage future studies that develop conservation management plans in order to ensure the continuation of ecological refugia for this species.

Keywords:
Climate suitability; Anthropogenic activities; Poaia; Niche-based models; Future scenarios

Introduction

Climate change is an important threat to biodiversity (Thomas et al. 2004). Assessing the rate of contemporary climate change is a global challenge to ensure the sustainability and subsistence of natural systems on which all life depends (Arthington et al. 2018). It is estimated that human-induced activities have already increased the global temperature by 1.0 °C above pre-industrial levels. Global warming is likely to reach 1.5 °C between 2030 and 2052, posing a risk to ecosystems, societies, and economic sectors (IPCC 2021). Human-induced climate changes have caused important habitat changes that led to alterations in the spatial distribution of species over the past decades (Parmesan 2006; Diffenbaugh & Field 2013). Thus, there is growing concern to understand the consequences that climate changes can have on biodiversity and related ecosystem services (Diniz-Filho et al. 2010; Buckley et al. 2010; Hofmann et al. 2021), especially the risk of local extinction of species (Román-Palacios & Wiens 2019; Williams et al. 2021).

Climate conditions play a key role in the spatial distribution of species that only occur at sites where the environmental conditions are appropriate for their survival (Whittaker et al. 2001). Within this context, climate variables can be used as indicators of environmental tolerance to predictions of species distribution (Kearney et al. 2010). Thus, modeling the ecological niche of species and predicting their spatial distribution in different climate scenarios (current and future) are feasible.

Niche-based models are gaining space in different fields of study, such as determination of the amplitude of species distribution (Franklin 2010; Nori et al. 2011); potential identification of the presence of rare or endangered species (Guisan et al. 2006); selection of priority areas for conservation (Araújo et al. 2004), and analysis of the effect of climate change on biodiversity and its implications for conservation (Inague et al. 2020; Sobral-Souza et al. 2021; Bax et al. 2021).

The use of Niche-based models to predict the potential effects of climate change on the biodiversity of native (Reis et al. 2019) or cultivated plants has proved to be a good strategy (Silva et al. 2018; Gomes et al. 2020). Carapichea ipecacuanha (Brot.) L. Andersson, commonly known as poaia or ipeca, is a plant that deserves attention due to its pharmacological importance and high risk of extinction (Zappi et al. 2013). The medicinal and economical value of C. ipecacuanha in compounds such as the alkaloids emetine and cephaeline are extracted from the roots and used to treat amoebic dysentery, as well as expectorant, anti-inflammatory, and emetic agents in cases of poisoning and intoxication. Astringent properties have also been reported (Bertolucci et al. 2008; Giraldo et al. 2015).

During the 1960s and 1970s, poaia was commonly found particularly in the state of Mato Grosso, Brazil. For a long time, “poaieiros” (people who collect poaia) have guaranteed their livelihoods during the rainy months and have moved the state’s economy by selling poaia roots (Governo de Mato Grosso 2017; Silva et al. 2019).

Poaia is a distylous shrub measuring up to 50 cm in height (Rossi et al. 2005) that occurs exclusively in the understory of shady and humid forests (Oliveira & Martins 2002). It has a wide geographic distribution, with records of native populations in Brazil, Colombia, and Central America (Assis & Giulietti 1999; Rossi et al. 2005). However, the species shows a disjunct spatial distribution that is considered to be relict and can be explained by the effects of past climate changes (Oliveira et al. 2010).

The isolation of populations over large geographic distances, as observed for C. ipecacuanha, can be explained by cycles of expansion and retraction of forest areas and areas with dry vegetation formation (related to glacial-interglacial cycles) that have occurred over the last 2 million years in the Neotropical region (Prado & Gibbs 1993; Pennington et al. 2000). These cycles, particularly those that occurred in the late period (the last 100,000 years), caused the expansion and retraction of the South American rainforests, which led to several events of isolation and reconnection of these forest biomes (Sobral-Souza et al. 2015; Ledo & Colli 2017). In the case of C. ipecacuanha, three large population groups have been recognized for their extensive disjunct spatial distribution, one in Central America that predominantly occurs in Panama, another in the southeastern region of the Brazilian Amazon, and the Atlantic Forest population (Oliveira et al. 2010).

Carapichea ipecacuanha is a threatened species (Lameira 2002; Vieira et al. 2002) and is classified as “vulnerable” by the National Center for Flora Conservation (CNCFlora 2012). The main threats to C. ipecacuanha are related to commercial extraction, especially for export due to the pharmacological properties of its roots (Garcia et al. 2005). Concern about the maintenance of natural populations of C. ipecacuanha was reported at the first meeting of the committee of Conservation of Genetic Resources of Medicinal and Aromatic Plants, where C. ipecacuanha was recognized as a priority for conservation studies (Vieira et al. 2002).

In 2013, C. ipecacuanha was classified as vulnerable to extinction due to the effects of predatory exploitation, trade, and export, the fragmentation and deforestation of its forest habitat, and the requirement for cross-pollination, increasing the rareness of subpopulations and rendering them prone to deleterious stochastic, genetic, population, and environmental effects (Zappi et al. 2013).

Situations such as distance from the fragment to the largest population center and in situ and ex situ conservation status are major risk factors for the C. ipecacuanha populations in the Atlantic Forest (Oliveira & Martins 2002; Oliveira et al. 2010). Anthropogenic factors such as a reduced cluster size, the small number of reproductive plants, habitat loss, and advance of the agricultural frontier jeopardize the maintenance of natural populations of C. ipecacuanha in the state of Mato Grosso, Brazil (Silva et al. 2019).

These landscape changes can further reduce the populations, rendering them isolated and rare and prone to stochastic deleterious effects (Oliveira & Martins 1998; Zappi et al. 2013). In addition, distyly in C. ipecacuanha is a limiting morphological factor for reproductive success because it requires the interaction of floral morphs that consists of the reciprocal positioning of stigmas and anthers between two floral morphs that belong to different individuals (Barrett 1992).

Apart from a balanced ratio (1:1) of floral morphs, reciprocal herkogamy of the reproductive organs between the alternative morphs (long-styled and short-styled) is desirable for cross-pollination. This floral polymorphism is usually accompanied by an intramorphic self-incompatibility system and a series of accessory floral features that differ between floral morphs (size of flowers, lobe stigma, number and size of pollen grains) (Ganders 1979). Due to the intramorphic self-incompatibility system, only legitimate pollen from the opposite morph is able to fertilize the ovules and produce viable seeds (Barrett 2002). This reproductive limitation renders this species more vulnerable to the effects of fragmentation and degradation of its native habitat, which is a major complicating factor since in Brazil deforestation for the implementation of agricultural matrices and urban expansion have intensified over the years (TerraBrasilis 2020), increasing the risk of extinction of C. ipecacuanha.

Given the eminent scenarios of global climate changes and the medicinal importance and high risk of extinction of C. ipecacuanha due to the different reasons mentioned above, studies are needed to assess the implications of climate changes for the potential distribution of this species. Therefore, here we aimed (i) to provide new information on the conservation of C. ipecacuanha by building niche-based models to analyze future optimistic and pessimistic climate scenarios of CO₂ emission in 2050 and 2090, and (ii) to apply niche-based models using data from three different poaia groups (Amazon, Atlantic and Panama) separately to analyze and describe the implications of climate change and to quantify the future availability of suitable habitat for these populations with different levels of climate suitability.

Material and methods

Species description

Carapichea ipecacuanha (Brot.) L. Andersson, commonly known as poaia (de Boer & Thulin 2005). Its fresh roots have a ring-like and segmented shape and a yellow or white color (Bertolucci et al. 2008). Poaia is distributed throughout Latin America (Skorupa & Assis 1998; Assis & Giulietti 1999). In its natural environment, the species rarely occurs as an individual plant but tends to grow in population clusters. In Brazil, poaia is found in the states of Mato Grosso, Rondônia, Rio de Janeiro, Minas Gerais, Espírito Santo and Bahia, and has been recognized for its sociocultural and economic importance (Lameira 2002; Zappi et al. 2013).

Species occurrence data

Occurrence records of poaia were obtained from 10 virtual herbaria (IB - UFMT, RFA, INPA, MBM, JPB, BAH, VIC, NYBG, TANG and IAN. Available at Plataform Species (http://inct.splink.org.br) and by review of the literature (Skorupa & Assis 1998; Oliveira et al. 2010). These records were divided into three spatial groups: Amazon, Atlantic Forest, and Panama. After selection and confirmation of the points of occurrence, a filter was applied in order to select only unique occurrences for each environmental cell (2.5 min cell size, approximately 4.5 x 4.5 km resolution at the Equator). This approach prevents inflation of the models, which can cause instability between the number of occurrences and the number of pseudo-absences (Diniz-Filho et al. 2003). We also emphasize that some regions such as Pará, Pernambuco and São Paulo were not included in the data since they were temporal decay samplings (1925) and poaia populations could no longer be found in several subsequent expeditions (Assis & Giulietti 1999). Models were built for all points of occurrence of poaia (all data). However, because the three geographical poaia groups (Amazon, Atlantic, and Panama) are spatially isolated, we also built separate models for each poaia group. This was necessary to better understand the effects of future climate change on each spatial group.

Climate and environmental data

We first analyzed the independence of the bioclimatic variables for the current scenario with a resolution of 2.5 min cell size, available from the WorldClim v. 2.1 database (Fick & Hijmans 2017), using a factorial analysis. The first five factor axes that explained 91% of the data variation were selected. The variables with the greatest contribution to each axis were used in the models. The selected variables were: Bio10 = Mean Temperature of Warmest Quarter (31%); Bio17 = Precipitation of Driest Quarter (16%); Bio07 = Annual Temperature Range (22%); Bio18 = Precipitation of Warmest Quarter (7%), and Bio16 = Precipitation of Wettest Quarter (15%).

To predict the models for future climate conditions, we used Four Atmosphere-Ocean General Circulation Models (AOGCMs): Miroc6, CSM2_MR, CNRM_CM6, and CanESMS. For each AOGCM, we predicted the species distribution considering four simulations: future 2050 optimistic (SSP2-4.5) and pessimistic (SSP5-8.5), and future 2090 optimistic (SSP2-4.5) and pessimistic (SSP5-8.5). The 2050 data are the mean from 2041 to 2060 and those for 2090 are the mean from 2081 to 2100. These variables were extracted from the Coupled Model Intercomparison Project (CMIP6), available from WorldClim v.2.1 (http://worldclim.org/data/cmip6/).

The Shared Socioeconomic Pathways (SSPs) are based on a matrix that uses shared socioeconomic pathways. In short, the SSPs describe alternative evolutions of future society in the absence of climate change or climate policy (O’Neill et al. 2016). Here, we adopted SSP2-4.5 as optimistic scenarios and SSP5-8.5 as pessimistic scenarios.

The SSP2-4.5 (lower energy demand) represents a “middle-of-the-road” scenario in which trends broadly follow their historical patterns. For these scenarios, global warming will increase the earth’s mean temperature by approximately 2.0 °C by 2050 and by 2.7 °C by 2090. This pathway updates the Representative Concentration Pathway (RCP) 4.5. The SSP5-8.5 (maximum power requirement and balanced emissions from fossil fuels and non-fossil fuels) represents the high end of the range of future pathways and updates the RCP 8.5. For these scenarios, global warming will increase the earth’s mean temperature by approximately 2.4 °C by 2050 and by 4.4 °C by 2090 (Rogelj et al. 2018, IPCC 2021).

Niche-based model building

There are various methods to build niche-based models, which can show variable performance depending on the modeling conditions (Zhu & Peterson 2017). The consensus method or ensemble approach based on several algorithms is more acceptable and indicated (Araújo & New 2007). Here, we used five algorithms for niche-model predictions, including two based only on presence data (1) - Bioclim (Nix 1986) and (2) - Gower’s distance (Carpenter et al. 1993) and three using presence/absence data or a pseudo-absence (3) - maximum entropy (MaxEnt v3.4.1; Phillips et al. 2006); (4) - Support Vector Machine (SVM; Tax & Duin 2004), and (5) - Random Forest (Breiman 2001). All model were built using the “dismo” and “kernlab” R packages (Karatzoglou et al. 2004; Hijmans et al. 2015).

We performed 20 repetitions per algorithm and scenario, resulting in 2,000 model runs (20 repetitions x 5 algorithms x 4 scenarios x 4 all data, Amazon, Atlantic and Panama groups). For model evaluation, we randomized the occurrence points into training and test data at a proportion of 75-25%, respectively. This proportion was used to calculate the true skill statistic (TSS). This weighting factor maximizes the point of highest specificity and sensitivity, returning a rank ranging from -1 to 1 for the generated models. This value was used to classify and validate the model. In this study, we only used niche-based models with TSS > 0.5 (Allouche et al. 2006)

After construction of the niche-based models, were built using the frequency ensemble approach (Araújo & New 2007). Thus, the higher the frequency of occurrence of the cell in the different models, the higher the habitat suitability of the cell. The ensemble approach was used for present and future (2050 and 2090) data under the optimistic and pessimistic scenarios. This procedure was carried out independently for the four AOGCMs. Finally, an average model was built with the four ensemble approaches for each scenario. This approach is indicated to reduce the effects of data variation resulting from different AOGCMs (Diniz-Filho et al. 2009; Diniz-Filho et al. 2010). This procedure was adopted for the three groups separately and the ensemble average was used as the final model for the study.

To quantify the implications of climate change for the distribution of poaia, we generated binary maps using the ensemble average model. The following criteria of habitat suitability (threshold) were defined for this purpose: >25, >50 and >75. These binary maps returned areas with 75%, 50% and 25% of climate suitability. Next, the percentage of the number of cells for each binary map was quantified for all scenarios studied. We decided to use the >50% and >75% thresholds for description of the data.

As an exploratory analysis of the climate data, separately for each point of occurrence of the three groups, we built a map for the present scenario using the ensemble average model with a threshold >50% (Fig. 1). All analyses were performed using the R (R Core Team 2021) and Qgis (QGIS 2021) programs.

Figure 1
Occurrence of three groups of Carapichea ipecacuanha (Poaia) in Latin America, showing the present scenario of climate suitability (threshold >50%).

Results

We compiled 66 points of unique occurrences of Poaia, including 31 points for the Amazon group, 16 points for the Atlantic group and 19 points for the Panama group (Fig. 1). The models for all data and for the individual groups demonstrate that climate change will cause drastic changes in the spatial distributions of this species. The models for 2050 including all data demonstrated a strong reduction in suitable areas for both the optimistic scenario (threshold >50 = -38%; threshold >75 = -65%) and the pessimistic scenario (>50 = -53%; >75 = -76%). The results predicted for 2090 were similar, revealing more alarming rates for both the optimistic scenario (>50 = -60%; >75 = -79%) and the pessimistic scenario (>50 = -87%; >75 = -97%) (Fig. 2 and ^{Figure S1}).

Figure 2
Variation in habitat availability for Carapichea ipecacuanha (Poaia) under the present and 2050 and 2090 (optimistic and pessimistic) scenarios using all data and data for the three groups separately. Grey = climate suitability >25%; yellow = climate suitability >50%; green = climate suitability >75%.

The climate data for the Amazon group exhibited the largest variations, with a continuous increase in the “Annual Temperature Range” and “Mean Temperature of Warmest Quarter”, reaching a more than 30% increase by 2090 under the pessimistic scenario, as well as a gradual and continuous reduction in “Precipitation of Driest Quarter” and “Precipitation of Warmest Quarter”. On the other hand, the smallest differences between the present and future scenarios were observed for the Atlantic group (^{Figure S2}).

Amazon group

The current spatial distribution revealed approximately 67,045,600 ha (for the >50 threshold) and 33,069,150 ha (for the >75 threshold) of suitable areas (Fig. 3). The predictions for 2050 showed a broad reduction in suitable areas for both the optimistic scenario (>50 = -61%; >75 = -83%) and the pessimistic scenario (>50 = -72%; >75 = -90%) (Fig. 2). Similar results were obtained for predictions for 2090; however, the reduction rates were even higher for both the optimistic scenario (>50 = -76%; >75 = -93%) and the pessimistic scenario (>50 = -96%; >75 = -99%) (Fig. 2). In addition, the predictions showed a large displacement to the southeastern region of Brazil, particularly for the 2090 scenarios, with complete disappearance of suitable areas in locations of origin of this group (Fig. 3).

Figure 3
Separate distribution of the three groups of Carapichea ipecacuanha (Poaia) under the present and 2050 and 2090 scenarios (optimistic and pessimistic). Grey = climate suitability >25%; yellow = climate suitability >50%; green = climate suitability >75%.

Atlantic group

For the Atlantic group, the present scenario revealed 74,660,900 ha of suitable areas for the >50 threshold and 15,593,950 ha for the >75 threshold (Fig. 3). For 2050, there was a reduction in the areas of climate suitability for both the optimistic scenario (>50 = -56%; >75 = -62%) and the pessimistic scenario (>50 = -75%; >75 = -83%). The results obtained with the 2090 models were similar to the previous scenario but revealed more alarming rates for both the optimistic scenario (>50 = -74%; >75 = -81%) and the pessimistic scenario (>50 = -97%; >75 = -99.4%) (Fig. 2). The models demonstrate that suitable areas are restricted to transition regions in the states of Minas Gerais, Espirito Santo, and a small band in Rio de Janeiro, particularly for the 2090 scenarios (Fig. 3).

Panama group

For the Panama group, the consensus models of the present scenario revealed 269,605,700 ha of suitable areas for the >50 threshold and 33,069,150 ha for the >75 threshold (Fig. 1). The models for 2050 showed a reduction in suitable areas for both the optimistic scenario (>50 = -51%; >75 = -76%) and the pessimistic scenario (>50 = -66%; >75 = -84%) (Fig. 2). The distribution of this group was restricted to a small band in northern Colombia and the isthmus of Panama and Costa Rica. We also highlight the permanence of small suitable patches along the coast of Suriname and the region of Belém-PR (Fig. 3). Similar results were obtained with the 2090 models for both the optimistic scenario (>50 = -67%; >75 = -86%) and the pessimistic scenario (>50 = -89%; >75 = -97%) (Fig. 2). The distribution of suitable areas obtained for the 2090 optimistic scenario was similar to that observed for the 2050 pessimistic scenario. However, under the pessimistic scenario in 2090, there is a reduction in suitable areas concentrated in southern Panama and the northernmost part of Colombia. A reduction was also observed in southern Nicaragua and throughout Costa Rica (Fig. 3).

Discussion

Our findings suggest that future climate change scenarios will lead to drastic spatial reductions in the three geographic poaia groups. Furthermore, there will be a spatial rearrangement of the suitable climate conditions for the Amazon group to Atlantic Forest areas in southeastern Brazil, which are already suffering strong anthropic pressure (Ribeiro et al. 2009). This fact will exert strong extinction pressures on the native populations of poaia. Apparently, climate change related to global warming, especially under the pessimistic scenarios, will have major effects on the local vegetation structure. Regarding the Brazilian vegetation, few studies have addressed these dynamics and there is wide variation in the literature in terms of the origin of climate data used for modeling the scenarios of CO₂ emission and the algorithms used. However, large losses of climatically suitable areas have generally been observed for species and/or groups of species, for example, Atlantic Forest trees (Colombo & Joly 2010), coffee (Gomes et al. 2020), Ilex paraguariensis (Silva et al. 2018), Connarus suberosus (Reis et al. 2019), and 1,553 Cerrado plant species in Bolivia, Brazil and Paraguay (Velazco et al. 2018).

Amazon and atlantic groups

For the Amazon group, the combination of drastic reduction and the broad displacement of climatically suitable areas classifies this group as the one with the highest risk of local extinction. Regarding the displacement of the predicted area, similar results have been reported in a study determining the geographic distribution of tree species of the Atlantic Forest. The authors observed displacement of the potential area to southern Brazil, in agreement with the present study in which the potential area was found to be reduced and displaced to southeastern Brazil (Colombo & Joly 2010). A similar pattern has been reported for Connarus suberosus, with large reduction and displacement of climate suitability to the Atlantic Forest (Reis et al. 2019).

Here, the 2050 and 2090 pessimistic scenarios demonstrate that climate suitability for the Amazon and Atlantic Forest groups of poaia will only occur in areas with current remnants of the Atlantic Forest in São Paulo, Minas Gerais, Rio de Janeiro, and Espirito Santo. Currently, more than 148 million people live in the region of influence of the Brazilian Atlantic Forest, corresponding to approximately 67% of the Brazilian population (WWF 2017). The Atlantic Forest is therefore the biome with the highest anthropic pressure in Brazil and only 11.6% of its original forest cover remains (Ribeiro et al. 2009). Furthermore, it is one of the three biodiversity hotspots most vulnerable to the impacts of climate change (Bellard et al. 2014). This landscape is already under strong anthropic pressure. According to estimates for 2050, these areas will suffer the strongest anthropic pressure in the world, mainly because of the expansion of agricultural areas (Williams et al. 2021).

The increase in human-induced pressures on landscapes will possibly lead to a reduction in the area of different remnants and/or fragmentation of continuous areas. This fragmentation and creation of a landscape with small habitat patches promote biotic homogenization, mainly caused by local extinction and the inability to recolonize specialized species in the forested area (Solar et al. 2015; Bilyaminu et al. 2020). Since poaia is found exclusively in the understory of shady and humid forests, it becomes highly susceptible to the effects of fragmentation and its populations tend to decline rapidly when exposed to forest edge environments (Oliveira & Martins 2002). Fragmented areas are characterized by various disturbance factors that can aggravate the structure of poaia populations, including an increase in luminosity, temperature and wind speed that cause drying and mortality of plants not adapted to these conditions (Didham 1998). Poaia does not tolerate solar radiation and is found only in locations with >70% of shade (Lameira 2002).

Another factor impairing the maintenance of populations in fragmented landscapes formed by small remnant patches is that poaia is a distylous species. This is an important aggravating factor for reproductive success because it requires the interaction of floral morphs belonging to different individuals (Barrett 1992; Silva et al. 2019). In the absence of occurrence of the two morphs at the same location, populations tend to collapse, causing local extinction. Furthermore, small fragments only support a small number of individuals, a fact that can reduce the genetic diversity of the population. Thus, the isolation of poaia populations in these patches that are embedded in an inhospitable matrix tends to increase the rates of long-term local extinction since the decrease in the size of native areas reduces the size of the resident populations. Isolation and population decline, in turn, reduce gene flow and increase the likelihood of genetic drift and inbreeding depression (Hedrick & Kalinowski 2000; Fahrig 2003). The loss of genetic variability as a result of population isolation renders the population more susceptible to environmental stochasticity (Matthies et al. 2004), which can increase the risks of local and regional extinction (Hanski et al. 2004). These characteristics of poaia, associated with its low tolerance to climatic variations, high habitat specialization and low dispersal ability (dependent on birds), increase the vulnerability of the species to future climate change (Trew & Maclean 2021).

Panama group

Although our results demonstrate a future decline in the suitable area for the Panama group, this reduction may be understood as an artifice caused by inflation of the model for the present scenario. The models for this group showed the greatest extrapolation to areas without occurrence of the species. This fact can be explained by the almost nonexistence of strict endemism of the species in Panama (Powell et al. 2018). The diversity of the fauna and flora in the Panama region originates from and is influenced by the biodiversity of South America, while the biodiversity of the Nicaragua and Costa Rica region is of North American origin (Duivenvoorden et al. 2002). However, this range of diversity eventually fuses at some point, characterizing the Panama environment (Stiles 1985; Wilson 1990).

This strong South American influence on Panama’s biodiversity may be linked to the shared environmental and climate characteristics (ANAM 2010), which may have influenced the models to large part of northern South America. Few large areas of primary forest remain intact, which occur only in large reserves such as in eastern Panama along the Internacional La Amistad Park, and particularly in the Biológica La Selva Reserve (~1,500 ha) along northeastern Costa Rica that is connected to the Braulio Carrillo National Park (Powell et al. 2018). This connection is the last intact gradient of primary forest from sea level to 2,900 m (Lieberman et al. 1996). These forest blocks retain almost all vertebrate species in this ecoregion, including most large predators (Stiles 1985). These mountain forest refugia are the sites where climatically suitable areas, in almost all scenarios, remain in future climate scenarios (Fig. 3).

In addition to its small size, most of Panama’s territory is anthropized due to the construction of cities, highways, power plants, and cultivated areas. Consequently, the remaining natural areas have been reduced and fragmented by rapid deforestation throughout the 20th century and extending to the 21st century (Kapos et al. 2015). The parks themselves face enforcement problems of environmental protection, including the Tortuguero National Park and the Barra de Colorado Wildlife Refuge, both in Costa Rica. Deforestation for logging, banana monocultures and opening up lands for grazing are some of the anthropic factors that threaten the biodiversity in the Panama region (Powell et al. 2018). Although large areas of intact habitat still exist, lowland forests have been seriously fragmented (Sánchez-Azofeifa et al. 1999). The evergreen forest areas are the least protected in Costa Rica (Stiles 1985), representing an additional threat to the poaia populations that occur in those areas. Another factor that does not provide hope for the expansion of already protected areas is the difficult economic situation in Panama for investments in the environmental sector (Powell et al. 1992). The climate models for the Colombian poaia population tend to maintain climate suitability under all scenarios, decreasing after 2090.

Landscape roughness and climate refugia

The topographic complexity of the terrain is a local factor that can mitigate the effects of global warming by acting as a buffer for climate variables (Trew & Maclean 2021). Regions with a more heterogeneous relief, including the presence of mountains and valleys, provide a greater variety of habitats because a wide range of micro- and mesoclimatic conditions is available on a small spatial scale (Badgley et al. 2017). Thus, the southeastern region of Brazil, particularly the state of Minas Gerais due to its higher terrain roughness indices (Amatulli et al. 2018), can be considered the most resistant to climate change among the regions of the three populations studied here.

On the other hand, the Amazon region shows the lowest terrain roughness indices (Amatulli et al. 2018) and can therefore be considered the region with the highest susceptibility to climate change. This topographic homogeneity limits habitat availability; thus, the species do not have climate refugia, increasing the risk of local extinction (Trew & Maclean 2021). Furthermore, locations with less topographic complexity, such as the Amazon region, exhibit a higher climate velocity, which is the rate of climate movement in the landscape, calculated as the rate of climate change over time (°C per year or mm of precipitation per year) divided by the rate of climate change in space (IPCC 2014). Thus, the climate velocity is generally high in flat areas because the rate of temperature variation in relation to space is low (Chen et al. 2011). Species that occupy extensive flat landscapes are particularly vulnerable because they must disperse over longer distances than species in mountainous regions in order to find climate refugia. Therefore, the Amazon group is at high risk of extinction since climate change occurs at a speed greater than its migration capacity. Although there is no migration rate for poaia, we can infer that this rate of displacement is slow, as observed for all herbaceous plants (Kinlan & Gaines 2003), and will not be sufficient to ensure the migration of these species to areas with a favorable climate. This fact contributes to the prospect of future extinction of the Amazon population.

Conclusion

Our findings showed that the suitable areas for the three geographic poaia groups will be spatially reduced in the future. However, the three groups exhibited different geographic rearrangements in future scenarios. We can consider the Amazon and Atlantic Forest groups to be at higher risk of extinction because of the drastic reduction and geographic displacement of the areas predicted for the future to locations distant from the location of origin. In addition, future suitable areas will be located in the Atlantic Forest in southeastern Brazil, an area that currently suffers strong anthropic pressure.

An important result was the relationship between lower landscape roughness in the Amazon group and high rates of climate change. This is a matter of great concern since extensive forest and Cerrado areas have a low topographic complexity, a fact that renders these landscapes highly susceptible to the effects of climate change. We therefore encourage future studies to investigate the relationship between landscape roughness and the creation of climate refugia. We also emphasize the importance of future studies that develop conservation management plans in order to ensure the continuation of native forests, especially for groups in Brazil.

Acknowledgments

We thank the Fundação de Amparo à Pesquisa do Estado de Mato Grosso (FAPEMAT, grant numbers: 0586230/2016 and 0235228/2017) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, grant number: 306183/2018) for financial support.

References

• Allouche O, Tsoar A, Kadmon R. 2006. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). Journal of Applied Ecology 43: 1223-1232. doi: 10.1111/j.1365-2664.2006.01214.x
  » https://doi.org/10.1111/j.1365-2664.2006.01214.x
• Amatulli G, Domisch S, Tuanmu MN et al 2018. A suite of global, cross-scale topographic variables for environmental and biodiversity modeling. Scientific Data 5: 180040. doi: 10.1038/sdata.2018.40
  » https://doi.org/10.1038/sdata.2018.40
• ANAM - Autoridad Nacional del Ambiente. 2010. Atlás ambiental de la República de Panamá. Panamá: Editora Novo Art.
• Araújo MB, Cabezas M, Thuiller W, Hannnah L, Williams PH. 2004. Would climate change drive species out of reserves? An assessment of existing reserve-selection methods. Global Change Biology 10: 1618-1626. doi: 10.1111/j.1365-2486.2004.00828.x
  » https://doi.org/10.1111/j.1365-2486.2004.00828.x
• Araújo MB, New M. 2007. Ensemble forecasting of species’ distributions. Trends in Ecology & Evolution 22: 42-47. doi: 10.1016/j.tree.2006.09.010
  » https://doi.org/10.1016/j.tree.2006.09.010
• Arthington AH, Bhaduri A, Bunn SE et al 2018. The Brisbane declaration and global action agenda on environmental flows. Frontiers in Environmental Science 6: 45. doi: 10.3389/fenvs.2018.00045
  » https://doi.org/10.3389/fenvs.2018.00045
• Assis MC, Giulietti AM. 1999. Diferenciação morfológica e anatômica em populações de “ipecacuanha” - Carapichea ipecacuanha (Brot.) Stokes (Rubiaceae). Revista Brasileira de Botânica 22: 205-216. doi: 10.1590/S0100-84041999000200011
  » https://doi.org/10.1590/S0100-84041999000200011
• Badgley C, Smiley TM, Terry R et al 2017. Biodiversity and topographic complexity: Modern and geohistorical perspectives. Trends in Ecology and Evolution 32: 211-226. doi: 10.1016/j.tree.2016.12.010
  » https://doi.org/10.1016/j.tree.2016.12.010
• Barrett SCH. 1992. Heterostylous Genetic Polymorphisms: Model Systems for Evolutionary Analysis. In: Barrett SCH. (eds.) Evolution and Function of Heterostyly. Monographs on Theoretical and Applied Genetics,. Vol. 15. Berlin, Heidelberg, Springer. doi: 10.1007/978-3-642-86656-2_1
  » https://doi.org/10.1007/978-3-642-86656-2_1
• Barrett SCH. 2002. The evolution of plant sexual diversity: Evolution of sex. Nature Reviews Genetics 3: 274-284. doi: 10.1038/nrg776.
  » https://doi.org/10.1038/nrg776
• Bax V, Castro-Nunez A, Francesconi W. 2021. Assessment of potential climate change impacts on montane forests in the Peruvian Andes: Implications for conservation prioritization. Forests 12: 375. doi: 10.3390/f12030375
  » https://doi.org/10.3390/f12030375
• Bellard C, Leclerc C, Leroy B et al 2014. Vulnerability of biodiversity hotspots to global change. Global Ecology and Biogeography 23: 1376-1386. doi: 10.1111/geb.12228
  » https://doi.org/10.1111/geb.12228
• Bertolucci SKV, Lameira OA, Pinto JEBP. 2008. Guia das plantas medicinais. In: Lameira OA, Pinto JEBP (eds.). Plantas medicinais: do cultivo, manipulação e uso à recomendação popular. Belém, Embrapa Amazônia Oriental. p. 159-245
• Bilyaminu H, Abdulrashid I, Wada AF, Gambo MS. 2020. Biotic homogenization and its potential drivers: A review. International Research Journal of Biological Sciences 2: 50-59
• Breiman L. 2001. Random Forest. Machine Learning 45: 5-32
• Buckley LB, Urban MC, Angilletta MJ, Crozier LG, Rissler LJ, Sears MW. 2010. Can mechanism inform species distribution models? Ecology Letters 13: 1041-1054. doi: 10.1111/j.1461-0248.2010.01479.x
  » https://doi.org/10.1111/j.1461-0248.2010.01479.x
• Carpenter G, Gillison AN, Winter J. 1993. DOMAIN: a flexible modeling procedure for mapping potential distributions of animals and plants. Biodiversity and Conservation 2: 667-680. doi: 10.1007/BF00051966
  » https://doi.org/10.1007/BF00051966
• Chen IC, Hill JK, Ohlemuller R, Roy DB, Thomas CD. 2011. Rapid range shifts of species associated with high levels of climate warming. Science 333: 1024-1026. doi: 10.1126/science.1206432
  » https://doi.org/10.1126/science.1206432
• CNCFlora. 2012. Carapichea ipecacuanha In: Lista Vermelha da flora brasileira. Centro Nacional de Conservação da Flora. http://cncflora.jbrj.gov.br/portal/pt-br/profile/Carapichea ipecacuanha 13 Feb. 2021.
  » http://cncflora.jbrj.gov.br/portal/pt-br/profile/Carapichea ipecacuanha
• Colombo AF, Joly CA. 2010. Brazilian Atlantic Forest lato sensu: the most ancient Brazilian forest, and a biodiversity hotspot, is highly threatened by climate change. Brazilian Journal of Biology 70: 697-708. doi: 10.1590/S1519-69842010000400002
  » https://doi.org/10.1590/S1519-69842010000400002
• de Boer HJ, Thulin M. 2005. Lectotypification of Callicocca ipecacuanha Brot. and neotypification of Cephaelis acuminata H. Karst., with reference to the drug ipecac. Taxon 54: 1080-1082. doi: 10.2307/25065496
  » https://doi.org/10.2307/25065496
• Didham RK. 1998. Altered leaf-litter decomposition rates in tropical forest fragments. Oecologia 116: 397-406.
• Diffenbaugh NS, Field CB. 2013. Changes in ecologically critical terrestrial climate conditions. Science 341: 486-492. doi: 10.1126/science.1237123
  » https://doi.org/10.1126/science.1237123
• Diniz-Filho JAF, Bini LM, Hawkins BA. 2003. Spatial autocorrelation and red herrings in geographical ecology. Global Ecology and Biogeography 12: 53-64. doi: 10.1046/j.1466-822X.2003.00322.x
  » https://doi.org/10.1046/j.1466-822X.2003.00322.x
• Diniz-Filho JAF, Bini LM, Rangel TFLB et al 2009. Partitioning and mapping uncertainties in ensembles of forecasts of species turnover under climate changes. Ecography 32: 897-906. doi: 10.1111/j.1600-0587.2009.06196.x
  » https://doi.org/10.1111/j.1600-0587.2009.06196.x
• Diniz-Filho JAF, Nabout JC, Bini LM et al 2010. Ensemble forecasting and geographic range of Tropidacris cristata (Orthoptera: Acridoidea: Romaleidae) under climate change. Insect Conservation Diversity 3: 213-221. doi: 10.1111/j.1752-4598.2010.00090.x
  » https://doi.org/10.1111/j.1752-4598.2010.00090.x
• Duivenvoorden JF, Svenning JC, Wright SJ. 2002. Ecology. Beta diversity in tropical forests. Science 25: 636-7. doi: 10.1126/science.295.5555.636.
  » https://doi.org/10.1126/science.295.5555.636
• Fahrig L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34: 487-515. doi: 10.1146/annurev.ecolsys.34.011802.132419
  » https://doi.org/10.1146/annurev.ecolsys.34.011802.132419
• Fick SE and Hijmans RJ. 2017. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology 37: 4302-4315. doi: 10.1002/joc.5086
  » https://doi.org/10.1002/joc.5086
• Franklin J. 2010. Moving beyond static species distribution models in support of conservation biogeography. Diversity and Distributions 16: 321-330. doi: 10.1111/j.1472-4642.2010.00641.x
  » https://doi.org/10.1111/j.1472-4642.2010.00641.x
• Ganders FR. 1979. The biology of heterostyly. New Zealand Journal of Botany 17: 607-635.
• Garcia RMA, Oliveira LO, Moreira MA, Barros WC. 2005. Variation in emetine and cephaeline contents in roots of wild ipeca (Carapichea ipecacuanha). Biochemical and Systematics Ecology 33: 233-243. doi: 10.1016/j.bse.2004.08.005
  » https://doi.org/10.1016/j.bse.2004.08.005
• Giraldo CB, Trujillo AIU, Gomez EJN. 2015. Potencial de regeneración de Carapichea ipecacuanha (Rubiaceae) a partir de capas delgadas de células. Acta Biológica Colombiana 20: 181-192. doi: 10.15446/abc.v20n3.47354.
  » https://doi.org/10.15446/abc.v20n3.47354
• Gomes LC, Bianchi FJJA, Cardoso IM, Fernandes RBA, Fernandes Filho EI, Schulte RPO. 2020. Agroforestry systems can mitigate the impacts of climate change on coffee production: A spatially explicit assessment in Brazil. Agriculture, Ecosystems and Environment 294: 106858. doi: 10.1016/j.agee.2020.106858
  » https://doi.org/10.1016/j.agee.2020.106858
• Governo de Mato Grosso. 2017. De abundante a quase extinta, a poaia é pesquisada em Mato Grosso. http://www.mt.gov.br/-/5502462-de-abundante-a-quase-extinta-a-poaia-e-pesquisada-em-mato-grosso 30 Jan. 2022.
  » http://www.mt.gov.br/-/5502462-de-abundante-a-quase-extinta-a-poaia-e-pesquisada-em-mato-grosso
• Guisan A, Lehmann A, Ferrier S et al 2006. Making better biogeographical predictions of species’ distributions. The Journal of Applied Ecology 43: 386-392. doi: 10.1111/j.1365-2664.2006.01164.x
  » https://doi.org/10.1111/j.1365-2664.2006.01164.x
• Hanski I, Erãlahti C, Kankare M, Ovaskainen O, Sirén H. 2004. Variation in migration propensity among individuals maintained by landscape structure. Ecology Letters 7: 958-966. doi: 10.1111/j.1461-0248.2004.00654.x
  » https://doi.org/10.1111/j.1461-0248.2004.00654.x
• Hedrick PW, Kalinowski ST. 2000. Inbreeding Depression in Conservation Biology. Annual Review of Ecology and Systematics 31: 139-162. doi: 10.1146/annurev.ecolsys.31.1.139
  » https://doi.org/10.1146/annurev.ecolsys.31.1.139
• Hijmans RJ, van Etten J, Sumner M et al 2015. Raster: Geographic Data Analysis and Modeling. R Package Version 2.4-15. http://CRAN.R-project.org/package=raster
  » http://CRAN.R-project.org/package=raster
• Hofmann GS, Cardoso MF, Alves RJV et al 2021. The Brazilian Cerrado is becoming hotter and drier. Global Change Biology. 27: 4060-4073. doi: 10.1111/gcb.15712
  » https://doi.org/10.1111/gcb.15712
• Inague GM, Zwiener VP, Marques MCM. 2020. Climate change threatens the woody plant taxonomic and functional diversities of the Restinga vegetation in Brazil. Perspectives in Ecology and Conservation 19: 53-60. doi: 10.1016/j.pecon.2020.12.006
  » https://doi.org/10.1016/j.pecon.2020.12.006
• IPCC - Intergovernmental Panel on Climate Change. 2014. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, Cambridge University Press. https://www.ipcc.ch/site/assets/uploads/2018/02/ar5_wgII_spm_en.pdf
  » https://www.ipcc.ch/site/assets/uploads/2018/02/ar5_wgII_spm_en.pdf
• IPCC - Intergovernmental Panel on Climate Change. 2021. Climate Change 2021: The Physical Science Basis. https://www.ipcc.ch/report/ar6/wg1/
  » https://www.ipcc.ch/report/ar6/wg1/
• Kapos V, Walcott J, Thorley J. 2015. Planning for REDD+ in Panama: Securing Social and Environmental Benefits. Cambridge, UNEP-WCMC. http://www.un-redd.org/tabid/5954/Default.asp
  » http://www.un-redd.org/tabid/5954/Default.asp
• Karatzoglou A,Smola A, Hornik K, Zeileis A. 2004. kernlab - an S4 package for kernel methods in R. Journal of Statistical Software 11: 1-20. doi: 10.18637/jss.v011.i09
  » https://doi.org/10.18637/jss.v011.i09
• Kearney MR, Wintle BA, Porter WP. 2010. Correlative and mechanistic models of species distribution provide congruent forecasts under climate change. Conservation Letters 3: 203-213. doi: 10.1111/j.1755-263X.2010.00097.x
  » https://doi.org/10.1111/j.1755-263X.2010.00097.x
• Kinlan BP, Gaines SD. 2003. Propagule dispersal in marine and terrestrial environments: A community perspective. Ecology 84: 2007-2020. doi: 10.1890/01-0622
  » https://doi.org/10.1890/01-0622
• Lameira OA. 2002. Cultivo da ipecacuanha [Psychotria ipecacuanha (Brot.) Stokes]. Cultura Técnica 28. Belém, Embrapa Amazônia Oriental .
• Ledo RMD, Colli GR. 2017. The historical connections between the Amazon and the Atlantic Forest revisited. Journal of Biogeography 44: 2551-2563. doi: 10.1111/jbi.13049
  » https://doi.org/10.1111/jbi.13049
• Lieberman D, Lieberman M, Peralta R, Hartshorn GS. 1996. Tropical forest structure and composition on a large-scale altitudinal gradient in Costa Rica. Journal of Ecology 84: 137-152. doi: 10.2307/2261350
  » https://doi.org/10.2307/2261350
• Matthies D, Bräuer I, Maibom W, Tscharntke T. 2004. Population size and the risk of local 773 extinction : Empirical evidence from rare plants. Oikos 105: 481-488. doi: 10.1111/j.0030-1299.2004.12800.x
  » https://doi.org/10.1111/j.0030-1299.2004.12800.x
• Nix HA. 1986 A biogeographic analysis of Australian elapid snakes. In: Longmore R. (ed.) Atlas of Elapid Snakes of Australia. Canberra, Australian Government Publishing Service. p. 4-15.
• Nori J, Urbina-Cardona JN, Loyola RD, Lescano JN, Leynaud GC. 2011. Climate Change and American Bullfrog Invasion: What Could We Expect in South America? PLoS One 6: e25718. doi: 10.1371/journal.pone.0025718
  » https://doi.org/10.1371/journal.pone.0025718
• Oliveira LO, Martins ER. 1998. O desafio das plantas medicinais brasileiras: 1 - O caso da poaia (Cephaelis ipecacuanha). Campos dos Goytacazes, Fundação Estadual do Norte Fluminense.
• Oliveira LO, Martins ER. 2002. A quantitative assessment of genetic erosion in ipecac (Psychotria ipecacuanha). Genetic Resources and Crop Evolution 49: 607-617. doi: 10.1023/A:1021275006915
  » https://doi.org/10.1023/A:1021275006915
• Oliveira LO, Rossi AAB, Martins ER, Batista FRC, Silva RS. 2010. Molecular phylogeography of Carapichea ipecacuanha, an amphitropical shrub that occurs in the understory of both semideciduous and evergreen forests. Molecular Ecology 19: 1410-1422. doi: 10.1111/j.1365-294X.2010.04591.x
  » https://doi.org/10.1111/j.1365-294X.2010.04591.x
• Parmesan C. 2006. Ecological and Evolutionary Responses to Recent Climate Change. Annual Review of Ecology, Evolution, and Systematics 37: 637-669. doi: 10.2307/annurev.ecolsys.37.091305.30
  » https://doi.org/10.2307/annurev.ecolsys.37.091305.30
• O'Neill BC, Tebaldi C, van Vuuren DP et al 2016. The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6, Geoscientific Model Development 9: 3461-3482. doi: 10.5194/gmd-9-3461-2016, 2016.
  » https://doi.org/10.5194/gmd-9-3461-2016, 2016
• Pennington RT, Prado DE, Pendry CA. 2000. Neotropical seasonally dry forests and Quaternary vegetation changes. Journal of Biogeography 27: 261-273. doi: 10.1046/j.1365-2699.2000.00397.x
  » https://doi.org/10.1046/j.1365-2699.2000.00397.x
• Phillips SJ, Anderson RP, Schapire RE. 2006. Maximum entropy modeling of species geographic distribution. Ecological Modelling 190: 231-259. doi: 10.1016/j.ecolmodel.2005.03.026
  » https://doi.org/10.1016/j.ecolmodel.2005.03.026
• Powell G, Palminteri S, Schipper J. 2018. Ecoregions. Central America: Southern Nicaragua into Costa Rica and Panama. http://www.worldwildlife.org/ecoregions/nt0129 12 Jun. 2021.
  » http://www.worldwildlife.org/ecoregions/nt0129
• Powell G, Rappole JH, Sader SA. 1992. Neotropical migrant landbird use of lowland Atlantic habitats in Costa Rica : a test of remote sensing for identification of habitat. In: Hagan III JM, Johnston DW (eds.). Ecologia e conservação de aves terrestres migrantes neotropicais. Washington, D.C.: Smithsonian Institution Press. p. 287-298.
• Prado DE, Gibbs PE. 1993. Patterns of species distributions in the dry seasonal forests of South America. Annals of the Missouri Botanical Garden 80: 902-927. doi: 10.2307/2399937
  » https://doi.org/10.2307/2399937
• QGIS - QGIS Association. 2021. QGIS Geographic Information System. Version 3.18.2 - Zürich. http://www.qgis.org
  » http://www.qgis.org
• R Core Team. 2021. R: A language and environment for statistical computing. Vienna, R Foundation for Statistical Computing, https://www.R-project.org/
  » https://www.R-project.org/
• Rogelj JD, Shindell K, Jiang K. et al 2018. Mitigation Pathways Compatible with 1.5 °C in the Context of Sustainable Development. In: IPCC - Intergovernmental Panel on Climate Change. Global Warming of 1.5 °C. Cambridge, Cambridge University Press . p. 93-174. https://www.ipcc.ch/site/assets/uploads/sites/2/2022/06/SR15_Chapter_2_LR.pdf
  » https://www.ipcc.ch/site/assets/uploads/sites/2/2022/06/SR15_Chapter_2_LR.pdf
• Ribeiro MC, Metzger JP, Martensen AC, Ponzoni FJ, Hirota MM. 2009. The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biological Conservation 142: 1141-1153. doi: 10.1016/j.biocon.2009.02.021
  » https://doi.org/10.1016/j.biocon.2009.02.021
• Reis CR, Bueno ML, Rocha LF, Santos LG, Gorgens EB. 2019. Climate Change Influencing the Potential Distribution of a Brazilian Savanna Indicator Species. Floresta e Ambiente 26: e20180085. doi: 10.1590/2179-8087.008518
  » https://doi.org/10.1590/2179-8087.008518
• Román-Palacios C, Wiens JJ. 2019. Recent responses to climate change reveal the drivers of species extinction and survival. PNAS 117: 4211-4217. doi: 10.1073/pnas.1913007117
  » https://doi.org/10.1073/pnas.1913007117
• Rossi AAB, Oliveira LO, Vieira MF. 2005. Distyly and variation in floral traits in natural populations of Psychotria ipecacuanha (Brot.) Stockes (Rubiaceae). Brazilian Journal of Botany 28: 285-294. doi: 10.1590/S0100-84042005000200009
  » https://doi.org/10.1590/S0100-84042005000200009
• Sánchez-Azofeifa GA, Quesada-Mateo C, Gonzalez-Quesada P, Dayanandan S, Bawa KS. 1999. Protected areas and conservation of biodiversity in the tropics. Conservation Biology 13: 407-411.
• Silva MAF, Higuchi P, Silva AC. 2018. Impacto de mudanças climáticas sobre a distribuição geográfica potencial de Ilex paraguariensis Rodriguésia 69: 4, 2069-2079. doi: 10.1590/2175-7860201869437
  » https://doi.org/10.1590/2175-7860201869437
• Silva PC, Nascimento TO, Krause W, Chaves DSA, Silva CA. 2019. Risk of genetic vulnerability and aspects of the reproductive biology of Psychotria ipecacuanha (Rubiaceae), a threatened medicinal plant species of Brazilian forest. Acta Botanica Brasilica 33: 548-557. doi: 10.1590/0102-33062019abb0116
  » https://doi.org/10.1590/0102-33062019abb0116
• Skorupa LA, Assis MC. 1998. Collection and conserving Ipecac (Psychotria ipecacuanha, Rubiaceae) germplasm in Brazil. Economic Botany 52: 209-210. doi: 10.1007/BF02861214
  » https://doi.org/10.1007/BF02861214
• Sobral-Souza T, Lima-Ribeiro MS, Solferini VN. 2015. Biogeography of Neotropical Rainforests: past connections between Amazon and Atlantic Forest detected by ecological niche modeling. Evolutionary Ecology 29: 643-655. doi: 10.1007/s10682-015-9780-9
  » https://doi.org/10.1007/s10682-015-9780-9
• Sobral-Souza T, Santos JP, Maldaner ME, Lima-Ribeiro M, Ribeiro MC. 2021. Ecoland: A multiscale niche modelling framework to improve predictions on biodiversity and conservation. Perspectives in Ecology and Conservation 19: 362-368. doi: 10.1016/j.pecon.2021.03.008
  » https://doi.org/10.1016/j.pecon.2021.03.008
• Solar RRC, Barlow J, Ferreira JN et al 2015. How pervasive is biotic homogenization in human-modified tropical forest landscapes? Ecology Letter 18: 1108-1118. doi: 10.1111/ele.12494
  » https://doi.org/10.1111/ele.12494
• Stiles FG. 1985. Conservation of forest birds in Costa Rica: problems and perspectives. In: Diamond AW, Lovejoy TE (eds.). Conservation of Tropical Forest Birds. ICBP Technical Publication No. 4. Cambridge, International Council for Bird Preservation. p. 141-168
• Tax DMJ, Duin RPW. 2004. Support Vector data description. Machine Learning 54: 45-66. doi: 10.1023/b:mach.0000008084.60811.49
  » https://doi.org/10.1023/b:mach.0000008084.60811.49
• TerraBrasilis. 2020. Prodes: Desmatamento. http://terrabrasilis.dpi.inpe.br/app/dashboard/deforestation/biomes/amazon/increments 19 Apr. 2021
  » http://terrabrasilis.dpi.inpe.br/app/dashboard/deforestation/biomes/amazon/increments
• Thomas CD, Cameron A, Green RE et al 2004. Extinction risk from climate change. Nature 427: 145-148. doi: 10.1038/nature02121
  » https://doi.org/10.1038/nature02121
• Trew BT, Maclean IMD. 2021. Vulnerability of global biodiversity hotspots to climate change. Global Ecology Biogeography 30: 768-783. doi: 10.1111/ geb.13272
  » https://doi.org/10.1111/ geb.13272
• Velazco SJE, Villalobos F, Galvão F, De-Marco-Júnior P. 2018. A dark scenario for Cerrado plant species: Effects of future climate, land use and protected areas ineffectiveness. Diversity and Distributions 25: 660-673. doi: 10.1111/ddi.12886
  » https://doi.org/10.1111/ddi.12886
• Vieira R, Ribeiro-Silva S, Alves RBN et al 2002. Estratégias para conservação e manejo de recursos genéticos de plantas medicinais e aromáticas: resultados da 1a reunião técnica. Brasília, Embrapa Recursos Genéticos e Biotecnologia.
• Whittaker R, Willis KJ, Field R. 2001. Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography 28: 453-470. doi: 10.1046/j.1365-2699.2001.00563.x
  » https://doi.org/10.1046/j.1365-2699.2001.00563.x
• Williams DR, Clark M, Buchanan GM, Ficetola GF, Rondinini C, Tilman D. 2021. Proactive conservation to prevent habitat losses to agricultural expansion. Nature Sustainability 4: 314-322. doi: 10.1038/s41893-020-00656-5
  » https://doi.org/10.1038/s41893-020-00656-5
• Wilson DE. 1990. Mammals of La Selva, Costa Rica. In: Gentry H (ed.). Quatro florestas tropicais neotropicais. New Haven, Yale University Press.
• Zappi D, Jardim J, Souza EB et al 2013. Rubiaceae. In: Martinelli G, Moraes MA (eds.). Livro vermelho da flora do Brasil. Rio de Janeiro, Instituto de Pesquisas Jardim Botânico do Rio de Janeiro. p. 922-941
• Zhu GP, Peterson AT. 2017. Do consensus models outperform individual models? Transferability evaluations of diverse modeling approaches for an invasive moth. Biological Invasions 19: 2519-2532. doi: 10.1007/s10530-017-1460-y
  » https://doi.org/10.1007/s10530-017-1460-y
• WWF - World Wide Fund for Nature. 2017. Condição da Floresta Atlântica: Três países, 148 milhões de pessoas, uma das florestas mais ricas da Terra. Puerto Iguazú, Argentina. https://wwfbr.awsassets.panda.org/downloads/pub_estadodamataatlantica_web_25maio17.pdf
  » https://wwfbr.awsassets.panda.org/downloads/pub_estadodamataatlantica_web_25maio17.pdf

Supplementary material

The following online material is available for this article:

Figure S1 - Distribution of Carapichea ipecacuanha (Poaia) obtained for the present scenario and for the 2050 and 2090 optimistic and pessimistic scenarios using all data.

Figure S2 - Variation in bioclimatic variables used for individual modeling of the three Carapichea ipecacuanha (Poaia)groups between the present scenario and the 2050 and 2090 optimistic and pessimistic scenarios.

Publication Dates

• Publication in this collection
  12 May 2023
• Date of issue
  2023

History

• Received
  31 Mar 2022
• Accepted
  30 Mar 2023