Next Article in Journal
Full Orthotropic Mechanical Characterization of Pinus radiata Plywood Through Tensile, Compression and Shear Testing with Miniaturized Specimens
Previous Article in Journal
Phylogenetic Structure Analysis Based on the Blue-Light Receptor Cryptochrome: Insights into How Light Shapes the Vertical Structure of Subtropical Forest Community
Previous Article in Special Issue
Climate-Change Impacts on Distribution of Amazonian Woody Plant Species Key to Conservation, Restoration and Sustainable Use in the Colombian Amazon
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Forests
Volume 16
Issue 11
10.3390/f16111674
forests-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

Formation- and Species-Level Responses of the Atlantic Forest to Climate Change

by
Eduardo Vinícius S. Oliveira
1,
Carla Diele Cabral Vieira
1,
Jhonatan Rafael Zárate-Salazar
1,
Wadson de Jesus Correia
1,
Alexandre de Siqueira Pinto
2 and
Sidney F. Gouveia
2,*
1
Graduate Program in Ecology and Conservation, Federal University of Sergipe, São Cristóvão 49107-230, Sergipe, Brazil
2
Department of Ecology, Federal University of Sergipe, São Cristóvão 49107-230, Sergipe, Brazil
*
Author to whom correspondence should be addressed.
Forests 2025, 16(11), 1674; https://doi.org/10.3390/f16111674
Submission received: 8 September 2025 / Revised: 25 October 2025 / Accepted: 31 October 2025 / Published: 2 November 2025
(This article belongs to the Special Issue Modeling of Forest Dynamics and Species Distribution)
Browse Figures
Versions Notes

Abstract

The hyper-diverse Atlantic Rainforest on the eastern coast of South America comprises deciduous, semideciduous, and evergreen forest formations. How these formations, both as communities and through their individual species, are responding to climate change remains elusive. Using habitat suitability modeling, we examine the effects of climate change on the distribution of the Atlantic Rainforest assessed both at the species level and the formation level. Additionally, we investigated whether mismatches between species- and formation-level trends are linked to the climatic affinities of species at the formations where they occur. We predicted a decrease in habitat suitability for all deciduous, semideciduous, and evergreen formations, based on individual species models, up to 2100. However, when considering species together as formations, we predicted expansions of deciduous and semideciduous formations and contractions of evergreen formations for the same period. The divergence between the synchronous and individual suitability models for deciduous and semideciduous formations suggests that climate-tolerant species will likely expand their range, replacing those with narrower climate tolerances. This shift may alter the structure and composition of these communities as currently known. Our findings provide valuable insights that can inform strategies for conserving the Atlantic Rainforest, including the development of new regulatory measures, the establishment of protected areas, and the formulation of effective forest management policies.

1. Introduction

The question of whether plant formations are driven by climate as a single ecosystem-level entity or through the individual responses of their species dates back to the early days of ecology [1,2]. On the one hand, the distribution of plant formations has been largely defined by climatic limits [3]. On the other hand, they are collections of individual species that are distributed according to their niche requirements [4]. The idea of a collective response of plant communities to environmental conditions has led to the concept of alternative stable states [5]. Accordingly, assembling plant communities converge to a state of species composition with characteristic traits (i.e., physiognomy) when submitted to certain environmental conditions. However, alternative models claim that such a description is more of a simplification for data handling than a true end of the assembling process, and the observed physiognomies are post hoc perceptions of the communities [6].
Beyond the theoretical interest, this question is now pressing in the context of human-induced climate change [7]. Will plant formations shift as a single entity, or will species shift individually, if any, and create new physiognomies? According to the shared response hypothesis, similar niche requirements or positive interactions among species (e.g., facilitation) may lead the whole community to respond in tandem [8]. Conversely, the individual response hypothesis claims that species will track their niche, causing communities to reshuffle and eventually form novel assembling patterns [7]. Evidence suggests both collective [9,10] and individual [10,11] responses to climate change. For instance, when studying communities distributed across an elevational gradient, Pucko et al. [10] found both outcomes, depending on the climatic position of the species and the communities along the environmental gradient they occupy.
However, existing evidence is largely biased towards temperate environments [12]. In the tropics, some studies have indicated that the distributions of species are shifting, especially towards colder conditions [13]. However, at the community level, there is evidence for both composition stability [14] and shift [15]. Still, little is known about whether the shift in species distribution will match the predicted distribution of different formations or will be independent, potentially creating novel assemblages. Addressing this question for tropical forests is critical as they play a crucial role in regulating the global greenhouse effect by acting as major carbon sinks [16]. In South America, tropical forests harbour almost half of the Earth’s tree species [17]. Changes in South American forest formations are predicted to cause significant ecosystem disruptions, with alterations to climatic regimes across the continent, affecting biodiversity, agriculture, and human well-being [18,19].
The Brazilian Atlantic Rainforest is a biodiversity hotspot that encompasses tropical and subtropical regions. Its environmental heterogeneity determines the distribution of different plant formations, including Evergreen and Seasonal forests [20]. Evergreen forests experience 0–4 dry months and are dominated by hygrophytic plants, exhibiting a range of physiognomies, including open, dense, and mixed types. In turn, seasonal forests undergo 4–6 dry months and are composed of both hygrophytic and xerophytic plants, with physiognomies ranging from deciduous to semideciduous forests [21]. In the Atlantic Rainforest, the effect of climate change on plant distribution has been fairly investigated. Predicted effects include changes in species range size and position [22,23], including upward migration of species to higher altitudes [15,24], and a favouring of generalist species [25]. Changes in the distributions of formation types were also investigated, but not in comparison to a large dataset of individual species within a single framework [26].
Therefore, it remains to be determined whether the species’ distributional responses to climate change will match the predicted distribution of forest formation types where they occur. Here, we investigate whether the shifts in the distribution of tree species due to climate change, by 2050 and 2090, involve individual or shared responses of entire forest formations. We also tested whether mismatches between individual and shared responses, if any, are related to the climatic affinities of species at the formations where they occur. We anticipate that climate change will impact each vegetation type differently, with more pronounced effects on wetter formations than on drier ones [26]. Drier formations, such as seasonal forests, are more resilient to rising temperatures and reduced precipitation, while wetter formations are more vulnerable to these conditions [27]. We also expect tree species to respond differently to climate change, where species associated with a wide climatic range are likely to expand their geographic limits. In contrast, species with narrow climatic ranges may experience spatial contractions [28]. Finally, we foresee individual species responding more strongly than vegetation types, since species often have narrower climatic ranges than entire formations [15,29].

2. Materials and Methods

2.1. Study Area and Classification of Forest Formations

The Brazilian Atlantic Rainforest, on the east coast of South America, comprises a diverse range of forest formations. They are subject to rainfall loads ranging from 1000 to 4500 mm per year (mean range: 1500–2000 mm per year) and temperatures from 1 to 35 °C (mean range: 14–21 °C) [30].
We first classified the forest formations of the Atlantic Rainforest as evergreen, seasonal semideciduous, and seasonal deciduous [31,32]. Additionally, due to the significant differences in precipitation load and regime, as well as temperature levels, between the northern and southern Atlantic Rainforest, we further subdivided the evergreen formation into northern evergreen and southern evergreen [33,34]. Because the mixed forest is classified as a type of evergreen vegetation, similar to an Ombrophilous Forest [21], we integrated the mixed forest from southern Brazil into the southern evergreen formation. As such, we considered four forest formations for habitat suitability modelling: semideciduous, deciduous, northern evergreen, and southern evergreen (Figure 1).

2.2. Occurrence Records of the Forest Formations

To characterize the edaphoclimatic conditions of the forest formations, we used indicator tree species groups for each formation. Based on phytosociological studies in the Atlantic Rainforest, we selected the three most ecologically dominant species from each site (Table S1). Occurrence records for the indicator species were obtained from the Botanical Information and Ecology Network (BIEN) platform using the R package ‘BIEN’ v1.2.7 [35]. Since this database provides standardized occurrence records, filtering was limited to removing duplicate coordinates.
Using a randomization technique, we carried out a spatial thinning procedure on the occurrence records. In this step, we eliminated sampling bias by enforcing a minimum distance of 25 km between points [36]. This allowed us to create two spatially independent datasets, enabling an external evaluation using distinct datasets for training and testing points (see Text S1 for further details) [37]. Then, we sampled 1000 background points distributed throughout the entire Atlantic Rainforest as pseudo-absence points [38]. The predictor values were extracted from the coordinates of both the presence and pseudoabsence points. For this step, we used the R packages ‘spThin’ (v0.2.0) [39] and ‘dismo’ (v1.3-16) [40].

2.3. Environmental Layers

As climatic predictors, we used the 19 bioclimatic variables derived from temperature (°C) and precipitation (mm) data sourced from the WorldClim database (Table S2) [41]. Edaphic variables encompassed bulk density (Mg m−3), pH (soil acidity, logarithmic scale), and soil texture (%) (sand, silt, and clay), which were obtained from the SoilGRIDS database [42]. Furthermore, data on solar radiation (kWh m−2) and altitude (m) were acquired from WorldClim [41].

2.4. Habitat Suitability Models

Multicollinearity among explanatory variables can influence the importance values of predictors and restrict the simulated distribution of a species or ecological entity [37]. Therefore, we assessed the multicollinearity among the 26 selected environmental predictors using Pearson’s correlation. We excluded from the analysis those predictors that displayed correlation coefficients with r values greater than 0.70. As a result, we retained ten predictor variables for model generation. These included climatic variables (temperature and precipitation seasonality, maximum temperature of the warmest month, annual precipitation, precipitation of the wettest month, and solar radiation), edaphic factors (bulk density, clay content, and pH), and altitude.
Habitat suitability models characterize environmental conditions that are suitable for species occurrence [37]. Here, we employed this approach to describe the environmental hyperspace that characterizes the distribution of each formation [43]. For each forest formation, we used two categories of habitat suitability models. In the first category, the occurrence records of indicator species were aggregated to model the distribution of each formation (for a similar approach, see Casalegno et al. [44] and Bergamin et al. [45]). In the second category, indicator species were modelled individually based on their occurrence records. Then, models were generated for each formation by summing the suitability of their indicator species (for a similar approach, see Esser et al. [26]). Both approaches were run using the Maxent algorithm [46]. This method estimates the target distribution by integrating presence records with pseudo-absences, based on the principle of maximum entropy [47]. While other techniques, such as joint species distribution models [48], can also project the distributions of forest formations, habitat suitability models remain appropriate for our purposes [49].
In the model’s adjustment, we set up the Maxent parameters using an automatic routine that defines the feature classes (e.g., linear, quadratic, hinge, product, and threshold) from the occurrence points of the training data [46]. In constructing the habitat suitability models, the data underwent 100 repetitions and a cross-validation procedure (k = 5) using previously divided independent training and testing sets [37]. We evaluated model discrimination using the area under the receiver operating characteristic curve (AUC). The AUC ranges from 0 to 1, with values closer to 1 indicating superior model performance [50].
We projected each forest formation into future scenarios, considering two periods: 2050 (average for 2041–2060) and 2090 (average for 2081–2100). For both periods, we used the MIROC-6 global circulation model (GCM) available in WorldClim [41]. This model has been recognized as one of the most accurate for Central and South America [51]. We generated bioclimatic layers identical to those used in the model fitting, incorporating monthly temperature (maximum and minimum) and precipitation values. Other environmental predictors (i.e., edaphic and solar radiation) were included in the projections without modifications (for example, Zhang et al. [52]). We considered two future scenarios based on the 6th Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), using shared socioeconomic pathways (SSPs): a medium-emission scenario (SSP245) and a high-emission scenario (SSP585). The first scenario represents intermediate greenhouse gas emissions that remain at current levels until the middle of the century, while the second scenario considers an unrestricted increase in greenhouse gas emissions, roughly doubling current levels by 2050 [53].
The model projections describe the environmental suitability for the modelled entities. The suitability values were binarized (0–1) based on a threshold criterion that maximizes sensitivity (i.e., the proportion of correctly predicted presence) and specificity (i.e., the proportion of correctly predicted pseudo-absences) [54]. This allowed us to assess expansions, contractions, and/or shifts in suitable areas for each forest formation. The steps described above were carried out with the R packages ‘sdm’ (v1.2-59) and ‘dismo’ (v1.3-16) [40,55].

2.5. Climatic Similarity of Species and Formation Models

Using the same variables as in the habitat suitability models, we extracted climatic data from 10,000 points randomly distributed within the boundaries of the Atlantic Rainforest. From the binary models, we created an incidence matrix by combining climatic variables with the occurrences of each forest formation and individual species. We then performed a principal coordinates analysis, using the new variables to create histograms of density distributions for each formation and species [56]. We evaluated the climatic mismatches between species and formation-level from the cosine distance of their histograms [57]. This method proved effective for histogram comparison, as it measures the similarity in the direction of density vectors by the cosine angle [58]. We transformed histograms into vectors from bin frequency counts and computed the similarity of their shapes, including peaks and slopes [59]. The cosine similarity index ranges from 0 (different shapes) to 1 (identical shapes) [60]. Because this index is efficient in checking whether histogram peaks align, it performs better in detecting climatic affinities than tests that only consider data ranges [57]. Finally, we calculated the mean and dispersion of the cosine index between each formation and its indicator species. For these procedures, we used the R package ‘ade4’ v1.7-23 [61].

3. Results

3.1. Performance of Habitat Suitability Models

We identified 329 affiliated species of forest formations of the Atlantic Rainforest, totalling 112,932 occurrence records. Habitat suitability models were developed for 308 affiliated species, excluding those with fewer than 15 occurrence records.
Habitat suitability models for species with weak performance (i.e., AUC ≤ 0.75) were excluded from the final projection. Consequently, 294 species were retained, most of which exhibited excellent performance (approximately 53%; AUC > 0.90) or good performance (approximately 91%; AUC > 0.80). The model for Casearia sylvestris displayed lower accuracy (AUC = 0.75), whereas the model for Licania belemii demonstrated the highest accuracy (AUC = 0.99).
Habitat suitability models for forest formation varied from excellent to acceptable. The deciduous formation model showed the best performance (AUC = 0.97), followed by northern evergreen (AUC = 0.94), southern evergreen (AUC = 0.82), and semideciduous (AUC = 0.81). Detailed information on the models’ performance is available in the Supplementary Materials (Tables S3 and S4).

3.2. Future Formation Suitability from Separate Models

Models predicted that 252 tree species (approximately 86%) would experience a decrease in suitability within their current range by 2100, regardless of the climate change scenario. Fewer species (42 or 14%) were predicted to increase their suitability area by 2100 in both scenarios.
When comparing current habitat suitability with projections under climate change scenarios, we observed a marked reduction in suitable areas for the Atlantic Rainforest (Figure 2 and Figure 3; Figure S1). The semideciduous formation was predicted to undergo the most severe decline, with 98.6% and 99.1% of its area classified as unsuitable under the intermediate scenario (SSP245) by 2050 and 2090, respectively. Under the pessimistic scenario (SSP585), this reduction was predicted to reach 99% by 2050 and 98.3% by 2090. The northern evergreen formation ranked second in the percentage of unsuitable areas, including for both scenarios, SSP245 (2050 = 93% and 2090 = 95.5%) and SSP585 (2050 = 95.2% and 2090 = 99.7%). In the deciduous formation, 84% of the current range is projected to become unsuitable by 2050, increasing to 88% by 2090 under SSP245. The SSP585 scenario yielded similar results, with 87% unsuitable in 2050 and 88% in 2090. The southern evergreen formation exhibited the lowest loss of suitability, ranging from 77.7% to 80.8% under SSP245, and from 79.9% to 88.9% under SSP585, by 2050 and 2090, respectively (Figure 2 and Figure 3).
Interestingly, the models projected slight increases in suitability across all formations for 2050 and 2090. The southern evergreen formation retained the largest extent of suitable gains, ranging from 18% to 15% under SSP245, and from 17% to 8% under SSP585, in 2050 and 2090, respectively. In the deciduous formation, the suitable gains range from 7% (2090-SSP245, 2050-SSP585, and 2090-SSP585) to 10% (2050-SSP245). The northern evergreen formation exhibited minimal suitability gains, with increases of 4–6% under SSP245 scenarios (both 2050 and 2090) but no gains under the more extreme 2090-SSP585 scenario. The semideciduous formation exhibited the lowest gains, with only 1% of suitable area observed in both scenarios and timeframes (Figure 2 and Figure 3).

3.3. Future Formation Suitability from Joint Models

Based on the joint models that assess the response of individual species, the suitability of deciduous formation is expected to increase by 3.7% to 10.7% by 2090, according to SSP245 and SSP585, respectively (Table 1; Figure 4). Similarly, the semideciduous formation was predicted to increase in suitability by 2090 in both the SSP245 (4.8%) and SSP585 (6.6% from 2050 onward) (Table 1; Figure 4).
Conversely, the northern evergreen formation loses suitability, varying from 14.0% to 30.4% in SSP245 and SSP585 by 2090, respectively (Table 1; Figure 4). From 2050 onwards, some of these areas in northern evergreen are replaced either by deciduous or semideciduous formations that were predicted to increase (Figure 5). The southern evergreen formation was also predicted to reduce its suitable areas from 2050 onwards. This decline was more pronounced in the SSP585 (12.3% by 2090) than in SSP245 (7.5% by 2090) (Table 1; Figure 4). From 2050 onward, some of the areas currently occupied by southern evergreen will be replaced by semideciduous formations in both scenarios (Figure 5).

3.4. Climatic Similarity from Joint and Separated Models

We observed that Evergreen formations generally exhibited higher mean and lower standard deviation of the cosine index than deciduous formations, with only one instance diverging from this pattern (Figure 6). Accordingly, Evergreen formations tend to exhibit, on average, species-level models that are more climatically similar to their respective formations, coupled with lower internal variability (Figure 6).

4. Discussion

We found that formation- and species-level model projections showed partial agreement. Both approaches predicted a predominant reduction in southern and northern evergreen formations. However, the models for deciduous and semideciduous formations diverged. While the formation-level model predicted a decrease in suitability, the species-level model suggested a predominance of an increase in suitability. We also found that the difference in the prediction based on intermediate and pessimistic climate change scenarios was slight, as were the differences between the projections made for 2050 and 2090. Overall, these results align with previous findings that suggest partial agreement between shared- and species-level responses to climate change [7,10]. For the Atlantic Rainforest, these results indicate a consistent decline in the suitability of evergreen formations in the second half of this century, regardless of greenhouse gas emissions, and a more complex response from seasonal (deciduous and semideciduous) formations.
The divergence between the shared- and individual-species models for the decidual and semideciduous formations may be indicative of a general trend in the Atlantic Rainforest. On the one hand, the overall climatic conditions currently found in these regions are expected to change, as indicated by the reduction in suitability for all formations in the shared-species models. On the other hand, many individual species that have adapted to more seasonal conditions should benefit from these novel conditions and expand or shift their ranges into colder areas [13]. These areas are mainly occupied by evergreen formations that are, in turn, predicted to lose suitability. This scenario suggests a turnover of currently dominant species in evergreen formations by species characteristic of more deciduous physiognomies, and a possible dominance of these species at deciduous and semideciduous formations [14,15]. Such species exhibit greater resilience and tolerance to water stress, likely thriving over less tolerant species at both seasonal and evergreen formations under increasingly drier conditions [27,62].
The lower climatic affinities between deciduous formations and species-level models suggest that indicator species are distributed across a broader spectrum of environmental conditions, extending beyond the climatic envelope that defines these formations. This broader tolerance may explain the divergence between species- and formation-level projections, indicating that many deciduous species persist at the margins of their climatic niche [63,64]. Indeed, several of these species are typical of other Atlantic Rainforest formations, including evergreen, and therefore reach their distributional limits within deciduous habitats [28]. As a result, species-level models projected contraction under future climates, reflecting their already marginal position within these drier formations [65]. By contrast, the formation-level model can be interpreted as a representation of a widespread species adapted to the climatic regime that characterizes the deciduous and semideciduous formations [66,67]. From this perspective, the expansion projected at the formation level represents the suitability of the whole system, regardless of the marginality of the individual species that currently occur within it.
The shift of the Atlantic Rainforest towards drier physiognomies can have profound effects on the ecosystem and its services. As lower precipitation rates reduce carbon stock and exchange, the Atlantic Forest may experience reduced biomass production, thereby having a negative feedback effect on atmospheric emissions [68,69] and leading to cascading effects on trophic networks [70]. Furthermore, critical increases in temperature induce stomatal closure in response to water deficits, which negatively impacts photosynthesis rates, reduces net primary productivity, and consequently lowers growth rates [71,72]. The projected predominance of more tolerant, generalist species is likely to decrease ecological specialization, erode specific functional roles, and reduce both stability and functional resilience within the Atlantic Rainforest [73]. These consequences at the ecosystem level may be linked to the temporal trend of the predicted changes that we found. Significant changes are predicted at the formation level by 2050, with only minor changes by 2090, including recoveries. Such recoveries may be linked to the distinct effects that temperature and precipitation have on the climatic hyperspace of forest formations [69], as well as the fact that precipitation in some regions appears to be increasing [74]. In any case, these predicted recoveries are minimal, and the broader picture is one of stabilization from the mid-century onwards. Because most currently living individuals will persist until then, the observation of these changes may be subtle to the eye but pronounced in the recruitment, establishment, and growth dynamics of the species’ individuals [71,72].
On the other hand, the contraction of evergreen formations in the synchronous modelling approach may indicate habitat-specialist trees [26,75]. Tropical forests, recognized for their high diversity and intense niche partitioning, contain numerous tree species with a limited range of climatic tolerance [29,68]. The climatic variations expected could reduce the number of suitable areas for these species by 2100, affecting their reproduction, growth, and survival [13,76]. The decline in suitability for habitat-specialist trees could jeopardize community functioning [77]. Changes in species composition within Atlantic Rainforest formations have resulted from the replacement of ecologically important species by habitat generalists [25]. Habitat generalists have a high tolerance for disturbance and demonstrate significant phenotypic plasticity [22], which facilitates their expansion within Atlantic Rainforest formations. As they become predominant, they can modify species composition and structure, impacting the resilience and stability of ecosystems [15,68,78].
The projections indicate major distributional shifts until 2050, followed by a tendency of stability or attenuation after 2050. This pattern reflects the saturation of potential range shifts, where most climatically suitable areas are already occupied and marginal habitats are lost, limiting further redistribution [79,80]. Until 2050, formations are expected to undergo high instability, with compositional turnover, mortality, and disturbance-driven carbon release exceeding regrowth [77,81,82]. After 2050, alternative stable states may emerge, characterized by altered composition, structure, and carbon pools [82,83]. Notably, the differences between intermediate and pessimistic climate scenarios were slight, since for many species the loss of suitable area already reaches a threshold under intermediate conditions [84,85]. Consequently, more extreme scenarios do not necessarily translate into proportional additional losses, and the impact of aggravated climate change becomes limited once ecological thresholds are crossed [83,86].
Many species exhibited a migration trend toward the Southern Evergreen formations until 2100. In addition to responding to climate change through phenotypic plasticity and/or adaptive evolution, species may avoid extinction if they can migrate to a suitable site [29,87]. Several studies have documented species migrating toward higher latitudes and altitudes, serving as refuges [87,88]. Consequently, these regions have experienced an increase in species richness at the expense of hotter and drier areas [89]. This phenomenon is also reported for the Atlantic Rainforest, where numerous species have migrated to higher-altitude regions in search of more favourable climates [15,24,89]. Although our results suggest a potential shift in the distribution of tree species, some of these species may have limited dispersal capacity [90]. The impact of barriers and life history factors may hinder the colonization of certain areas [91]. Additionally, forest fragmentation and the occurrence of fires may obstruct species migrations [87].
The projected contraction of suitable habitats under climate change will likely drive complex ecological responses across different forest formations. Rather than a complete vegetation loss, effects are likely to manifest as a partial decline in forest cover and gradual shifts in community composition and structure [14,82,92]. In addition to the reduction in aboveground biomass and increases in mortality rates [69,71], resilient tree species may become dominant in regions that currently harbour sensitive species, leading to a reorganization of the community [25]. These climate-driven changes will require strategies designed to ensure the maintenance of ecological functions and services. For example, climatic refuges must be protected, supporting the establishment of conservation corridors and permeable matrices [93]. Ecological restoration should be strategic and based on landscape-scale planning, prioritizing the selection of species adapted to future climates [94,95]. Finally, integrating climate-based forecasts into forest management planning may help to mitigate the expected impacts, promoting resilience and adaptation in a rapidly changing environment [96].
Regardless of the climate-driven expansion or reduction of suitable areas in the Atlantic Rainforest until 2100, its formations remain highly vulnerable to anthropogenic global change processes. In addition to the potential loss of suitable areas by 2100, the Evergreen formations are also threatened by coastal erosion and rising sea levels [26,74]. Furthermore, habitat fragmentation and land-use changes compromise the conservation of all the formations investigated [20,97]. Except for many forest remnants in southern Bahia, the northern evergreen formation is in the worst conservation status [20]. Our results indicate that this formation is likely to be one of the most impacted by climate change, underscoring the need for intensified governmental actions to protect it. Moreover, the other Atlantic Rainforest formations should not be neglected, and long-term conservation actions and plans must be strengthened. This is essential for the stability of ecosystem services [98] and in addressing climate change, safeguarding the ability of these formations to act as carbon sinks in aboveground biomass [78].

5. Conclusions

Using a synchronous approach where the models consider species collectively, we found potential changes in the suitable areas of forest formations in the Brazilian Atlantic Forest, including expansions and contractions. Climate suitability decreased for the Evergreen formations, while it increased for the deciduous and semideciduous formations under both intermediate and pessimistic scenarios. The divergence between synchronous and individual suitability models for the deciduous and semideciduous formations suggests that some species with broad climate tolerance will expand their distribution, replacing those with restricted climate tolerance. This process is likely to alter the structure and composition of communities, potentially destabilizing ecological processes. The future distribution of species will be influenced not only by climate but also by their dispersal ability, biotic interactions, genetic adaptation, and other abiotic factors. Given this evidence, it becomes essential to strengthen conservation actions aimed at each type of formation. Our data provide insights that can be used to define strategies for protecting the Atlantic Rainforest, including new regulatory instruments, the creation of protected areas, and guidance for forest management policies. Furthermore, efforts focused on ecological restoration should consider anticipated future climate changes, aiming to ensure the resilience of the formations and the maintenance of their ecosystem services.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/f16111674/s1, Text S1: Notes on habitat suitability models procedures, Table S1: List of indicator tree species for each forest formation, with their dominant bioclimatic zone in the Brazilian Atlantic Rainforest, Table S2: List of codes and descriptions of the 19 bioclimatic variables obtained from the WorldClim database used in the habitat suitability modeling, Table S3: Performance of the habitat suitability models for each forest formation in the Brazilian Atlantic Rainforest, Table S4: Performance of the habitat suitability models for each tree species in the Brazilian Atlantic Rainforest, Figure S1: Habitat suitability of indicator species in each forest formation in the present (2021), intermediate scenario (SSP245), and pessimistic scenario (SSP585) of climate change.

Author Contributions

Conceptualization, E.V.S.O., C.D.C.V., A.d.S.P. and S.F.G.; Methodology, E.V.S.O., C.D.C.V. and S.F.G.; Software, E.V.S.O. and C.D.C.V.; Validation, E.V.S.O. and S.F.G.; Formal Analysis, E.V.S.O. and C.D.C.V.; Investigation, E.V.S.O. and C.D.C.V.; Resources, E.V.S.O.; Data Curation, E.V.S.O., C.D.C.V., J.R.Z.-S. and W.d.J.C.; Writing—Original Draft Preparation, E.V.S.O. and S.F.G.; Writing—Review and Editing, E.V.S.O., C.D.C.V., J.R.Z.-S., W.d.J.C., A.d.S.P. and S.F.G.; Visualization, E.V.S.O. and S.F.G.; Supervision, S.F.G. and A.d.S.P. All authors have read and agreed to the published version of the manuscript.

Funding

E.V.S.O. and J.R.Z.R. were financially supported by a post-doctoral fellowship provided by CAPES (PDPG-POSDOC; Proc. 88881.691646/2022-01). C.D.C.V. was supported by a master’s scholarship from CAPES (Finance Code 001). S.F.G. was supported by a FAPITEC PRONEM grant (Prof. 432/2023), a CNPq Universal grant (Proc. 405967/2023-3), and has been continuously supported by CNPq productivity grants (Proc. 302552/2022-7). This work was also supported by the National Institute of Science and Technology (INCT) in Ecology, Evolution, and Biodiversity Conservation, funded by CNPq (grant 409197/2024-6) and FAPEG (grant 201810267000023).

Data Availability Statement

Habitat suitability maps for Atlantic Rainforest tree species and other additional information can be found online at https://doi.org/10.6084/m9.figshare.30032254. Further datasets can be made available by the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Clements, F.E. Plant Succession: An Analysis of the Development of Vegetation; Carnegie institution of Washington: Washington, DC, USA, 1916; p. 658. [Google Scholar]
  2. Gleason, H.A. The individualistic concept of the plant association. Bull. Torrey Bot. Club 1926, 53, 7–26. [Google Scholar] [CrossRef]
  3. Woodward, F.; Williams, B. Climate and plant distribution at global and local scales. Vegetatio 1987, 69, 189–197. [Google Scholar] [CrossRef]
  4. Wiens, J.J. The niche, biogeography and species interactions. Philos. Trans. R. Soc. B Biol. Sci. 2011, 366, 2336–2350. [Google Scholar] [CrossRef]
  5. Sutherland, J.P. Multiple stable points in natural communities. Am. Nat. 1974, 108, 859–873. [Google Scholar] [CrossRef]
  6. Hastings, A. Time scales, dynamics, and ecological understanding. Ecology 2010, 91, 3471–3480. [Google Scholar] [CrossRef]
  7. Walther, G.R. Plants in a warmer world. Perspect. Plant Ecol. Evol. Syst. 2003, 6, 169–185. [Google Scholar] [CrossRef]
  8. Koffel, T.; Daufresne, T.; Klausmeier, C.A. From competition to facilitation and mutualism: A general theory of the niche. Ecol. Monogr. 2021, 91, e01458. [Google Scholar] [CrossRef]
  9. Walther, G.R. Community ecosystem responses to recent climate change. Philos. Trans. R. Soc. B Biol. Sci. 2010, 365, 2019–2024. [Google Scholar] [CrossRef] [PubMed]
  10. Pucko, C.; Beckage, B.; Perkins, T.; Keeton, W.S. Species shifts in response to climate change: Individual or shared responses? J. Torrey Bot. Soc. 2011, 138, 156–176. [Google Scholar] [CrossRef]
  11. Beckage, B.; Osborne, B.; Gavin, D.G.; Pucko, C.; Siccama, T.; Perkins, T. A rapid upward shift of a forest ecotone during 40 years of warming in the Green Mountains of Vermont. Proc. Natl. Acad. Sci. USA 2008, 105, 4197–4202. [Google Scholar] [CrossRef] [PubMed]
  12. Felton, A.J.; Smith, M.D. Integrating plant ecological responses to climate extremes from individual to ecosystem levels. Philos. Trans. R. Soc. B Biol. Sci. 2017, 372, 20160142. [Google Scholar] [CrossRef]
  13. Feeley, K.J. Distributional migrations, expansions, and contractions of tropical plant species as revealed in dated herbarium records. Glob. Change Biol. 2012, 18, 1335–1341. [Google Scholar] [CrossRef]
  14. Esquivel-Muelbert, A.; Baker, T.R.; Dexter, K.G.; Lewis, S.L.; Brienen, R.J.; Feldpausch, T.R.; Lloyd, J.; Monteagudo-Mendoza, A.; Arroyo, L.; Álvarez-Dávila, E.; et al. Compositional response of Amazon forests to climate change. Glob. Change Biol. 2019, 25, 39–56. [Google Scholar] [CrossRef]
  15. Bergamin, R.S.; Bastazini, V.A.G.; Esquivel-Muelbert, A.; Bordin, K.M.; Klipel, J.; Debastiani, V.J.; Vibrans, A.C.; Loyola, R.; Müller, S.C. Elevational shifts in tree community composition in the Brazilian Atlantic Forest related to climate change. J. Veg. Sci. 2024, 35, e13289. [Google Scholar] [CrossRef]
  16. D’Amato, G.; Vitale, C.; Rosario, N.; Neto, H.J.C.; Chong-Silva, D.C.; Mendonça, F.; Perini, J.; Landgraf, L.; Solé, D.; Sánchez-Borges, M.; et al. Climate change, allergy and asthma, and the role of tropical forests. WAO J. 2017, 10, 11. [Google Scholar] [CrossRef]
  17. Gatti, R.C.; Reich, P.B.; Gamarra, J.G.; Crowther, T.; Hui, C.; Morera, A.; Bastin, J.-F.; de-Miguel, S.; Nabuurs, G.-J.; Svenning, J.C.; et al. The number of tree species on Earth. Proc. Natl. Acad. Sci. USA 2022, 119, e2202784119. [Google Scholar]
  18. Brandão, D.O.; Barata, L.E.S.; Nobre, C.A. The effects of environmental changes on plant species and forest dependent communities in the Amazon region. Forests 2022, 13, 466. [Google Scholar] [CrossRef]
  19. Bennett, A.C.; Sousa, T.R.; Monteagudo-Mendoza, A.; Esquivel-Muelbert, A.; Morandi, P.S.; Souza, F.C.; Castro, W.; Duque, L.F.; Llampazo, G.F.; Santos, R.M. Sensitivity of South American tropical forests to an extreme climate anomaly. Nat. Clim. Chang. 2023, 13, 967–974. [Google Scholar] [CrossRef]
  20. Ribeiro, M.C.; Metzger, J.P.; Martensen, A.C.; Ponzoni, F.J.; Hirota, M.M. The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biol. Conserv. 2009, 142, 1141–1153. [Google Scholar] [CrossRef]
  21. IBGE–Instituto Brasileiro de Geografia e Estatística. Províncias Estruturais, Compartimentos de Relevo, Tipos de Solos e Regiões Fitoecológicas; IBGE: Rio de Janeiro, Brazil, 2019; p. 179. [Google Scholar]
  22. Colombo, A.F.; Joly, C.A. Brazilian Atlantic Forest lato sensu: The most ancient Brazilian forest, and a biodiversity hotspot, is highly threatened by climate change. Braz. J. Biol. 2010, 70, 697–708. [Google Scholar] [CrossRef]
  23. Leão, T.C.; Reinhardt, J.R.; Nic Lughadha, E.; Reich, P.B. Projected impacts of climate and land use changes on the habitat of Atlantic Forest plants in Brazil. Glob. Ecol. Biogeogr. 2021, 30, 2016–2028. [Google Scholar] [CrossRef]
  24. Mata-Guel, E.O.; Soh, M.C.; Butler, C.W.; Morris, R.J.; Razgour, O.; Peh, K.S.H. Impacts of anthropogenic climate change on tropical montane forests: An appraisal of the evidence. Biol. Rev. 2023, 98, 1200–1224. [Google Scholar] [CrossRef]
  25. Zwiener, V.P.; Lira--Noriega, A.; Grady, C.J.; Padial, A.A.; Vitule, J.R. Climate change as a driver of biotic homogenization of woody plants in the Atlantic Forest. Glob. Ecol. Biogeogr. 2018, 27, 298–309. [Google Scholar] [CrossRef]
  26. Esser, L.F.; Neves, D.M.; Jarenkow, J.A. Habitat—Specific impacts of climate change in the Mata Atlântica biodiversity hotspot. Divers. Distrib. 2019, 25, 1846–1856. [Google Scholar] [CrossRef]
  27. Stan, K.D.; Sanchez-Azofeifa, A.; Duran, S.M.; Hesketh, M.; Laakso, K.; Portillo-Quintero, C.; Rankine, C.; Doetterl, S. Tropical dry forest resilience and water use efficiency: An analysis of productivity under climate change. Environ. Res. Lett. 2021, 16, 054027. [Google Scholar] [CrossRef]
  28. Klipel, J.; Bergamin, R.S.; Esquivel-Muelbert, A.; Lima, R.A.; Oliveira, A.A.; Prado, P.I.; Müller, S.C. Climatic distribution of tree species in the Atlantic Forest. Biotropica 2022, 54, 1170–1181. [Google Scholar] [CrossRef]
  29. Bawa, K.S.; Dayanandan, S. Global climate change and tropical forest genetic resources. Clim. Change 1998, 39, 473–485. [Google Scholar] [CrossRef]
  30. Guedes, M.L.; Batista, M.D.A.; Ramalho, M.; Freitas, H.D.B.; Silva, E.D. Breve incursão sobre a biodiversidade da Mata Atlântica. In Mata Atlântica e Biodiversidade; Franke, C.R., Rocha, P.L.B., Klein, W., Gomes, S.L., Eds.; Edufba: Salvador, Brazil, 2005; pp. 39–92. [Google Scholar]
  31. Joly, C.A.; Metzger, J.P.; Tabarelli, M. Experiences from the Brazilian Atlantic Forest: Ecological findings and conservation initiatives. New Phytol. 2014, 204, 459–473. [Google Scholar] [CrossRef] [PubMed]
  32. Arruda, D.M.; Fernandes-Filho, E.I.; Solar, R.R.C.; Schaefer, C.E.G.R. Combining climatic and soil properties better predicts covers of Brazilian biomes. Sci. Nat. 2017, 104, 32. [Google Scholar] [CrossRef] [PubMed]
  33. Silva, J.M.C.; Casteleti, C.H.M. Status of the biodiversity of the Atlantic Forest of Brazil. In The Atlantic Forest of South America: Biodiversity Status, Threats, and Outlook; Galindo-Leal, C., Câmara, I.G., Eds.; CABS and Island Press: Washington, DC, USA, 2003; pp. 43–59. [Google Scholar]
  34. Sobral-Souza, T.; Lima-Ribeiro, M.S.; Solferini, V.N. Biogeography of Neotropical Rainforests: Past connections between Amazon and Atlantic Forest detected by ecological niche modeling. Evol. Ecol. 2015, 29, 643–655. [Google Scholar] [CrossRef]
  35. Maitner, B.S.; Boyle, B.; Casler, N.; Condit, R.; Donoghue, J.; Durán, S.M.; Guaderrama, D.; Hinchliff, C.R.; Jørgensen, P.M.; Kraft, N.J.B.; et al. The bien r package: A tool to access the Botanical Information and Ecology Network (BIEN) database. Methods Ecol. Evol. 2018, 9, 373–379. [Google Scholar] [CrossRef]
  36. Boria, R.A.; Olson, L.E.; Goodman, S.M.; Anderson, R.P. Spatial filtering to reduce sampling bias can improve the performance of ecological niche models. Ecol. Model. 2014, 275, 73–77. [Google Scholar] [CrossRef]
  37. Guisan, A.; Thuiller, W.; Zimmermann, N.E. Habitat Suitability and Distribution Models: With Applications in R; Cambridge University Press: Cambridge, UK, 2017; p. 478. [Google Scholar]
  38. Hysen, L.; Nayeri, D.; Cushman, S.; Wan, H.Y. Background sampling for multi-scale ensemble habitat selection modeling: Does the number of points matter? Ecol. Inform. 2022, 72, 101914. [Google Scholar] [CrossRef]
  39. Aiello-Lammens, M.E.; Boria, R.A.; Radosavljevic, A.; Vilela, B.; Anderson, R.P. spThin: An R package for spatial thinning of species occurrence records for use in ecological niche models. Ecography 2015, 38, 541–545. [Google Scholar] [CrossRef]
  40. Hijmans, R.J.; Phillips, S.; Leathwick, J.; Elith, J.; Hijmans, M.R.J. Package ‘dismo’ v1.3-16. CRAN: Contributed Packages. 2024. Available online: https://cran.r-project.org/web/packages/dismo/index.html (accessed on 31 August 2025).
  41. 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]
  42. ISRIC SoilGrids250m version 2.0. Available online: https://soilgrids.org/ (accessed on 13 October 2024).
  43. Del Río, S.; Canas, R.; Cano, E.; Cano-Ortiz, A.; Musarella, C.; Pinto-Gomes, C.; Penas, Á. Modelling the impacts of climate change on habitat suitability and vulnerability in deciduous forests in Spain. Ecol. Indic. 2021, 131, 108202. [Google Scholar] [CrossRef]
  44. Casalegno, S.; Amatulli, G.; Bastrup-Birk, A.; Durrant, T.; Pekkarinen, A. Modelling and mapping the suitability of European forest formations at 1-km resolution. Eur. J. For. Res. 2011, 130, 971–981. [Google Scholar] [CrossRef]
  45. Bergamin, R.S.; Debastiani, V.; Joner, D.C.; Lemes, P.; Guimarães, T.; Loyola, R.D.; Müller, S.C. Loss of suitable climatic areas for Araucaria forests over time. Plant Ecol. Divers. 2019, 12, 115–126. [Google Scholar] [CrossRef]
  46. Phillips, S.J.; Dudík, M. Modeling of species distributions with Maxent: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  47. Araújo, M.B.; New, M. Ensemble forecasting of species distributions. Trends Ecol. Evol. 2007, 22, 42–47. [Google Scholar] [CrossRef]
  48. Ovaskainen, O.; Abrego, N. Joint Species Distribution Modelling: With Applications in R; Cambridge University Press: Cambridge, UK, 2020; p. 372. [Google Scholar]
  49. Norberg, A.; Abrego, N.; Blanchet, F.G.; Adler, F.R.; Anderson, B.J.; Anttila, J.; Araújo, M.B.; Dallas, T.; Dunson, D.; Elith, J.; et al. A comprehensive evaluation of predictive performance of 33 species distribution models at species and community levels. Ecol. Monogr. 2019, 89, e01370. [Google Scholar] [CrossRef]
  50. Elith, J.; Graham, C.H.; Anderson, R.P.; Dudík, M.; Ferrier, S.; Guisan, A.; Hijmans, R.J.; Huettmann, F.; Leathwick, J.R.; Lehmann, A.; et al. Novel methods improve prediction of species’ distributions from occurrence data. Ecography 2006, 29, 129–151. [Google Scholar] [CrossRef]
  51. Ortega, G.; Arias, P.A.; Villegas, J.C.; Marquet, P.A.; Nobre, P. Present--day and future climate over central and South America according to CMIP5/CMIP6 models. Int. J. Clim. 2021, 41, 6713–6735. [Google Scholar] [CrossRef]
  52. Zhang, K.; Yao, L.; Meng, J.; Tao, J. Maxent modeling for predicting the potential geographical distribution of two peony species under climate change. Sci. Total Environ. 2018, 634, 1326–1334. [Google Scholar] [CrossRef]
  53. Van Vuuren, D.P.; Riahi, K.; Calvin, K.; Dellink, R.; Emmerling, J.; Fujimori, S.; KC, S.; Kriegler, E.; O’Neill, B. The Shared Socio-economic Pathways: Trajectories for human development and global environmental change. Glob. Environ. Change 2017, 42, 148–152. [Google Scholar] [CrossRef]
  54. Araújo, M.B.; Peterson, A.T. Uses and misuses of bioclimatic envelope modeling. Ecology 2012, 93, 1527–1539. [Google Scholar] [CrossRef]
  55. Naimi, B.; Araújo, M.B. sdm: A reproducible and extensible R platform for species distribution modelling. Ecography 2016, 39, 368–375. [Google Scholar] [CrossRef]
  56. Zuur, A.F.; Leno, E.N.; Smith, G.M. Analysing Ecological Data; Springer: New York, NY, USA, 2007; p. 698. [Google Scholar]
  57. Zhang, Q.; Canosa, R.L. A comparison of histogram distance metrics for content-based image retrieval. SPIE 2014, 9027, 154–162. [Google Scholar]
  58. Wada, A.; Sumiyoshi, C.; Yoshimura, N.; Hashimoto, R.; Matsumoto, J.; Stickley, A.; Yamada, Y.; Kikuchi, A.; Kubota, R.; Matsui, M.; et al. Semantic memory disorganization linked to social functioning in patients with schizophrenia. Schizophrenia 2025, 11, 61. [Google Scholar] [CrossRef]
  59. Irimatsugawa, T.; Shimizu, Y.; Okubo, S.; Inaba, H. Cosine similarity for quantitatively evaluating the degree of change in an optical frequency comb spectra. Opt. Express 2021, 29, 35613–35622. [Google Scholar] [CrossRef]
  60. Fan, Y.; Rui, X.; Poslad, S.; Zhang, G.; Yu, T.; Xu, X.; Song, X. A better way to monitor haze through image based upon the adjusted LeNet-5 CNN model. Signal Image Video P 2020, 14, 455–463. [Google Scholar] [CrossRef]
  61. Dray, S.; Dufour, A.B. The ade4 package: Implementing the duality diagram for ecologists. J. Stat. Softw. 2007, 22, 1–20. [Google Scholar] [CrossRef]
  62. Stan, K.; Sanchez-Azofeifa, A. Tropical dry forest diversity, climatic response, and resilience in a changing climate. Forests 2019, 10, 443. [Google Scholar] [CrossRef]
  63. Chevalier, M.; Broennimann, O.; Guisan, A. Climate change may reveal currently unavailable parts of species’ ecological niches. Nat. Ecol. Evol. 2024, 8, 1298–1310. [Google Scholar] [CrossRef] [PubMed]
  64. Das, S.; Ossola, A.; Beaumont, L. Records of urban occurrences expand estimates of the climate niches in tree species. Glob. Ecol. Biogeogr. 2024, 33, e13809. [Google Scholar] [CrossRef]
  65. Grinder, R.; Wiens, J. Niche width predicts extinction from climate change and vulnerability of tropical species. Glob. Change Biol. 2022, 29, 618–630. [Google Scholar] [CrossRef]
  66. Vincent, H.; Bornand, C.; Kempel, A.; Fischer, M. Rare species perform worse than widespread species under changed climate. Biol. Conserv. 2020, 246, 108586. [Google Scholar] [CrossRef]
  67. Esperón-Rodríguez, M.; Tjoelker, M.; Lenoir, J.; Laugier, B.; Gallagher, R. Wide climatic niche breadth and traits associated with climatic tolerance facilitate eucalypt occurrence in cities worldwide. Glob. Ecol. Biogeogr. 2024, 33, e13833. [Google Scholar] [CrossRef]
  68. Bazzaz, F.A. Tropical forests in a future climate: Changes in biological diversity and impact on the global carbon cycle. Clim. Change 1998, 39, 317–336. [Google Scholar] [CrossRef]
  69. Taylor, P.; Cleveland, C.; Wieder, W.; Sullivan, B.; Doughty, C.; Dobrowski, S.; Townsend, A. Temperature and rainfall interact to control carbon cycling in tropical forests. Ecol. Lett. 2017, 206, 779–788. [Google Scholar] [CrossRef]
  70. Rocha, S.; Torres, C.; Villanova, P.; Schettini, B.; Jacovine, L.; Leite, H.; Gelcer, E.; Reis, L.; Neves, K.; Comini, I.; et al. Drought effects on carbon dynamics of trees in a secondary Atlantic Forest. For. Ecol. Manag. 2020, 465, 118097. [Google Scholar] [CrossRef]
  71. Lyra, A.; Imbach, P.; Rodriguez, D.; Chou, S.C.; Georgiou, S.; Garofolo, L. Projections of climate change impacts on central America tropical rainforest. Clim. Change 2017, 141, 093–105. [Google Scholar] [CrossRef]
  72. Crous, K.Y. Plant responses to climate warming: Physiological adjustments and implications for plant functioning in a future, warmer world. Am. J. Bot. 2019, 106, 1049. [Google Scholar] [CrossRef]
  73. Ferreira, P.; Boscolo, D.; Lopes, L.; Carvalheiro, L.; Biesmeijer, J.; Da Rocha, P.; Viana, B. Forest and connectivity loss simplify tropical pollination networks. Oecologia 2020, 192, 577–590. [Google Scholar] [CrossRef]
  74. IPCC-Intergovernmental Panel on Climate Change. Climate Change: The Physical Science Basis (Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Cambridge University Press: Cambridge, UK, 2021; p. 1300. [Google Scholar]
  75. Cubino, J.P.; Vilà-Cabrera, A.; Retana, J. Tree species abundance changes at the edges of their climatic distribution: An interplay between climate change, plant traits and forest management. J. Ecol. 2024, 112, 2785–2797. [Google Scholar] [CrossRef]
  76. Loarie, S.R.; Carter, B.E.; Hayhoe, K.; McMahon, S.; Moe, R.; Knight, C.A.; Ackerly, D.D. Climate change and the future of California’s endemic flora. PLoS ONE 2008, 3, e2502. [Google Scholar] [CrossRef]
  77. Kramer, R.D.; Ishii, H.R.; Carter, K.R.; Miyazaki, Y.; Cavaleri, M.A.; Araki, M.G.; Azuma, W.A.; Inoue, Y.; Hara, C. Predicting effects of climate change on productivity and persistence of forest trees. Ecol. Res. 2020, 35, 562–574. [Google Scholar] [CrossRef]
  78. Feng, X.; Uriarte, M.; González, G.; Reed, S.; Thompson, J.; Zimmerman, J.K.; Murphy, L. Improving predictions of tropical forest response to climate change through integration of field studies and ecosystem modeling. Glob. Change Biol. 2018, 24, 213–232. [Google Scholar] [CrossRef] [PubMed]
  79. Colwell, R.K.; Brehm, G.; Cardelús, C.L.; Gilman, A.C.; Longino, J.T. Global warming, elevational range shifts, and lowland biotic attrition in the wet tropics. Science 2008, 322, 258–261. [Google Scholar] [CrossRef]
  80. Sharma, M.; Ram, B.; Chawla, A. Ensemble modelling under multiple climate change scenarios predicts reduction in highly suitable range of habitats of Dactylorhiza hatagirea (D.Don) Soo in Himachal Pradesh, western Himalaya. S. Afr. J. 2023, 154, 203–218. [Google Scholar] [CrossRef]
  81. Bakkenes, M.; Alkemade, J.R.M.; Ihle, F.; Leemans, R.; Latour, J.B. Assessing effects of forecasted climate change on the diversity and distribution of European higher plants for 2050. Glob. Change Biol. 2002, 8, 390–407. [Google Scholar] [CrossRef]
  82. Price, D.T.; Alfaro, R.I.; Brown, K.J.; Flannigan, M.D.; Fleming, R.A.; Hogg, E.H.; Girardin, M.P.; Lakusta, T.; Johnston, M.; McKenney, D.W.; et al. Anticipating the consequences of climate change for Canada’s boreal forest ecosystems. Environ. Rev. 2013, 21, 322–365. [Google Scholar] [CrossRef]
  83. Anjos, L.; De Toledo, P. Measuring resilience and assessing vulnerability of terrestrial ecosystems to climate change in South America. PLoS ONE 2018, 13, e0194654. [Google Scholar] [CrossRef]
  84. Dyderski, M.; Paź, S.; Frelich, L.; Jagodziński, A. How much does climate change threaten European forest tree species distributions? Glob. Change Biol. 2018, 24, 1150–1163. [Google Scholar] [CrossRef]
  85. Ali, F.; Khan, N.; Khan, A.; Ali, K.; Abbas, F. Species distribution modelling of Monotheca buxifolia (Falc.) A. DC.: Present distribution and impacts of potential climate change. Heliyon 2023, 9, e13417. [Google Scholar] [CrossRef] [PubMed]
  86. Salazar, L.F.; Nobre, C.A. Climate change and thresholds of biome shifts in Amazonia. Geophys. Res. Lett. 2010, 37, 1–5. [Google Scholar] [CrossRef]
  87. Bussotti, F.; Pollastrini, M.; Holland, V.; Brüggemann, W. Functional traits and adaptive capacity of European forests to climate change. Environ. Exp. Bot. 2015, 111, 91–113. [Google Scholar] [CrossRef]
  88. Shen, Y.; Tu, Z.; Zhang, Y.; Zhong, W.; Xia, H.; Hao, Z.; Zhang, C.; Li, H. Predicting the impact of climate change on the distribution of two relict Liriodendron species by coupling the MaxEnt model and actual physiological indicators in relation to stress tolerance. J. Environ. Manag. 2022, 322, 116024. [Google Scholar] [CrossRef]
  89. Fadrique, B.; Báez, S.; Duque, Á.; Malizia, A.; Blundo, C.; Carilla, J.; Osinaga-Acosta, O.; Malizia, L.; Silman, M.; Farfán-Ríos, W.; et al. Widespread but heterogeneous responses of Andean forests to climate change. Nature 2018, 564, 207–212. [Google Scholar] [CrossRef] [PubMed]
  90. Wang, W.J.; He, H.S.; Thompson, F.R., III; Spetich, M.A.; Fraser, J.S. Effects of species biological traits and environmental heterogeneity on simulated tree species distribution shifts under climate change. Sci. Total Environ. 2018, 634, 1214–1221. [Google Scholar] [CrossRef]
  91. Bond, N.; Thomson, J.; Reich, P.; Stein, J. Using species distribution models to infer potential climate change-induced range shifts of freshwater fish in south-eastern Australia. Mar. Freshw. Res. 2011, 62, 1043–1061. [Google Scholar] [CrossRef]
  92. Dollinger, C.; Rammer, W.; Suzuki, K.F.; Braziunas, K.H.; Keller, T.T.; Kobayashi, Y.; Mohr, J.; Mori, A.S.; Turner, M.G.; Seidl, R. Beyond resilience: Responses to changing climate and disturbance regimes in temperate forest landscapes across the Northern Hemisphere. Glob. Change Biol. 2024, 30, e17468. [Google Scholar] [CrossRef]
  93. Morelli, T.L.; Daly, C.; Dobrowski, S.Z.; Dulen, D.M.; Ebersole, J.L.; Jackson, S.T.; Lundquist, J.D.; Millar, C.I.; Maher, S.P.; Monahan, W.B.; et al. Managing climate change refugia for climate adaptation. PLoS ONE 2016, 11, e0159909. [Google Scholar] [CrossRef]
  94. Fremout, T.; Thomas, E.; Taedoumg, H.; Briers, S.; Gutiérrez-Miranda, C.E.; Alcázar-Caicedo, C.; Lindau, A.; Kpoumie, H.M.; Vincet, B.; Kettle, C.; et al. Diversity for Restoration (D4R): Guiding the selection of tree species and seed sources for climate—Resilient restoration of tropical forest landscapes. J. Appl. Ecol. 2022, 59, 664–679. [Google Scholar] [CrossRef]
  95. Antongiovanni, M.; Venticinque, E.M.; Tambosi, L.R.; Matsumoto, M.; Metzger, J.P.; Fonseca, C.R. Restoration priorities for Caatinga dry forests: Landscape resilience, connectivity and biodiversity value. J. Appl. Ecol. 2022, 59, 2287–2298. [Google Scholar] [CrossRef]
  96. Law, B.E.; Hudiburg, T.W.; Berner, L.T.; Kent, J.J.; Buotte, P.C.; Harmon, M.E. Land use strategies to mitigate climate change in carbon dense temperate forests. Proc. Natl. Acad. Sci. USA 2018, 115, 3663–3668. [Google Scholar] [CrossRef]
  97. Ferreira, I.J.M.; Campanharo, W.A.; Fonseca, M.G.; Escada, M.I.S.; Nascimento, M.T.; Villela, D.M.; Brancalion, P.; Magnago, L.F.S.; Anderson, L.O.; Nagy, L.; et al. Potential aboveground biomass increase in Brazilian Atlantic Forest fragments with climate change. Glob. Change Biol. 2023, 29, 3098–3113. [Google Scholar] [CrossRef]
  98. Allen, C.D.; Macalady, A.K.; Chenchouni, H.; Bachelet, D.; McDowell, N.; Vennetier, M.; Kitzberger, T.; Rigling, A.; Breshears, D.D.; Hogg, E.H.; et al. A global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. For. Ecol. Manag. 2010, 259, 660–684. [Google Scholar] [CrossRef]
Forests 16 01674 g001
Figure 1. Forest formations of the Atlantic Rainforest considered for habitat suitability modeling.
Figure 1. Forest formations of the Atlantic Rainforest considered for habitat suitability modeling.
Forests 16 01674 g001
Forests 16 01674 g002
Figure 2. Differences in habitat suitability of indicator species for each forest formation in 2050 and 2090 under intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Figure 2. Differences in habitat suitability of indicator species for each forest formation in 2050 and 2090 under intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Forests 16 01674 g002
Forests 16 01674 g003
Figure 3. Potential areas (in km2) classified as stable, suitable, and unsuitable according to habitat suitability in each forest formation under the intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Figure 3. Potential areas (in km2) classified as stable, suitable, and unsuitable according to habitat suitability in each forest formation under the intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Forests 16 01674 g003
Forests 16 01674 g004
Figure 4. Spatial dynamics of species habitat suitability for each forest formation in 2050 and 2090 under intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Figure 4. Spatial dynamics of species habitat suitability for each forest formation in 2050 and 2090 under intermediate (SSP245) and pessimistic (SSP585) climate change scenarios.
Forests 16 01674 g004
Forests 16 01674 g005
Figure 5. Pixel transition matrix (%) illustrating changes in habitat suitability for each forest formation from 2050 onwards, in response to climate projections under the intermediate (SSP245) and pessimistic (SSP585) scenarios.
Figure 5. Pixel transition matrix (%) illustrating changes in habitat suitability for each forest formation from 2050 onwards, in response to climate projections under the intermediate (SSP245) and pessimistic (SSP585) scenarios.
Forests 16 01674 g005
Forests 16 01674 g006
Figure 6. Mean and standard deviation of the cosine similarity between each indicator species and its corresponding forest formation. The cosine index was computed from histograms of the density distributions of principal components (PC1 and PC2) derived from climatic variables. Points indicate mean values, and error bars represent standard deviation.
Figure 6. Mean and standard deviation of the cosine similarity between each indicator species and its corresponding forest formation. The cosine index was computed from histograms of the density distributions of principal components (PC1 and PC2) derived from climatic variables. Points indicate mean values, and error bars represent standard deviation.
Forests 16 01674 g006
Table 1. Changes (%) in habitat suitability for each forest formation, indicating gains (+) and losses (−) by 2050, from 2050 to 2090, and cumulative changes by 2090 in response to climate projections under the intermediate (SSP245) and pessimistic (SSP585) scenarios.
Table 1. Changes (%) in habitat suitability for each forest formation, indicating gains (+) and losses (−) by 2050, from 2050 to 2090, and cumulative changes by 2090 in response to climate projections under the intermediate (SSP245) and pessimistic (SSP585) scenarios.
SSP245SSP585
Forest Formation20502050–2090209020502050–20902090
Deciduous+1.99+1.71+3.70+3.74+6.98+10.72
Northern Evergreen−10.49−3.52−14.01−17.94−12.47−30.41
Semideciduous+4.37+0.46+4.83+3.22+3.41+6.63
Southern Evergreen−5.32−2.20−7.52−5.35−7.01−12.36
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

Oliveira, E.V.S.; Vieira, C.D.C.; Zárate-Salazar, J.R.; Correia, W.d.J.; Pinto, A.d.S.; Gouveia, S.F. Formation- and Species-Level Responses of the Atlantic Forest to Climate Change. Forests 2025, 16, 1674. https://doi.org/10.3390/f16111674

AMA Style

Oliveira EVS, Vieira CDC, Zárate-Salazar JR, Correia WdJ, Pinto AdS, Gouveia SF. Formation- and Species-Level Responses of the Atlantic Forest to Climate Change. Forests. 2025; 16(11):1674. https://doi.org/10.3390/f16111674

Chicago/Turabian Style

Oliveira, Eduardo Vinícius S., Carla Diele Cabral Vieira, Jhonatan Rafael Zárate-Salazar, Wadson de Jesus Correia, Alexandre de Siqueira Pinto, and Sidney F. Gouveia. 2025. "Formation- and Species-Level Responses of the Atlantic Forest to Climate Change" Forests 16, no. 11: 1674. https://doi.org/10.3390/f16111674

APA Style

Oliveira, E. V. S., Vieira, C. D. C., Zárate-Salazar, J. R., Correia, W. d. J., Pinto, A. d. S., & Gouveia, S. F. (2025). Formation- and Species-Level Responses of the Atlantic Forest to Climate Change. Forests, 16(11), 1674. https://doi.org/10.3390/f16111674

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