From the following paper, is there a nice statistic I can use to talk about the amount by which the forst has recovered?
Atlantic Forest Ecosystems: Are There Significant Differences When Compared at the Same Stage of Regeneration?
by Adriano Bressane 1,2,*ORCID,João Pedro da Cunha Pinto 1,Julia Villalta da Silva 1,Mirela Beatriz Silva 2,Alexandre Siminski 3ORCID,Líliam César de Castro Medeiros 1 andRogério Galante Negri 1,4ORCID
1
São Paulo State University (UNESP), Institute of Science and Technology, São José dos Campos 12245-000, Brazil
2
Civil and Environmental Engineering Graduate Program, São Paulo State University (UNESP), School of Engineering, Bauru 17033-360, Brazil
3
Agricultural and Natural Ecosystems Graduate Program, Federal University of Santa Catarina (UFSC), Curitibanos 89520-900, Brazil
4
Natural Disasters Graduate Program, National Center for Monitoring and Early Warning of Natural Disasters (CEMADEN)/UNESP, São José dos Campos 12247-016, Brazil
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(8), 6823; https://doi.org/10.3390/su15086823
Submission received: 1 March 2023 / Revised: 9 April 2023 / Accepted: 11 April 2023 / Published: 18 April 2023
Downloadkeyboard_arrow_down Browse Figures Versions Notes
Abstract
For the monitoring and assessment of Atlantic Forest ecosystems, key indicators of the regeneration stage are considered. However, the current classification of these stages by experts does not consider the possible differences among such ecosystems. In order to test the hypothesis of significant differences, Atlantic Forest ecosystems in southern Brazil were compared at the same stage (initial, medium, and advanced stages of regeneration). An extensive database of the floristic forest inventory of Santa Catarina state, southern Brazil, with 460 sampling units, was used, addressing the seasonal deciduous forest (SDF), dense ombrophilous forest (DOF), and mixed ombrophilous forest (MOF). The regeneration stage of each sample unit was previously classified by experts using five key indicators (basal area per hectare—BA/ha; number of individuals per hectare—N/ha; number of species—S; Shannon biodiversity index—H’; and diameter at breast height—DBH). The Kruskal–Wallis method and pairwise multiple comparisons based on Dunn’s procedure were applied, considering two-way tests with 5% significance, and 95% power. The statistical tests confirmed the research hypothesis, namely, that the key indicators had significant differences in the later regeneration stages of Atlantic Forest ecosystems. For instance, S in DOF was statistically higher (p < 0.001) than in MOF and SDF (44 ± 9); N in SDF was significantly lower (394 ± 248 at mid-stage) compared to MOF (475 ± 233) and DOF (545 ± 173); and H’ showed increasing differences towards more advanced stages. Considering that the stage of forest regeneration is the main criterion for decision-making between suppression and conservation of forest ecosystems, the results achieved should support the review of current procedures applied to the classification of successional stages. Given the demonstrated differences, in the next steps of research, we will evaluate and propose specific standards for each Atlantic Forest ecosystem, i.e., intervals of discriminating values for the key indicators applicable to this biome.
Keywords: regeneration stage; key indicators; monitoring and assessment
- Introduction
The scale of loss and degradation of the world’s forests has global implications for energy, climate, water, and carbon cycles, especially in tropical forests, severely impacting all forms of life, including public health and quality of life of humankind [1,2,3,4]. The Atlantic Forest is a tropical forest biome located in Latin America; it represents the second largest forest in South America, covering eastern Paraguay and the province of Misiones in Argentina, but its greatest territorial extension is located on the easter, northeaster, southeaster, and southern coasts of Brazil [5,6]. Although the Atlantic Forest is considered a Natural World Heritage site—a Biosphere Reserve—by UNESCO and a biodiversity hotspot, the forest has been intensely damaged through the centuries following the arrival of Europeans to South America [7]. Thus, the suppression and extreme fragmentation of its original cover has made it one of the most threatened biomes in the world [8,9].
Several strategies are underway to meet the demand for large-scale recovery of the Atlantic Forest [10]. Romanelli et al. [11] carried out a meta-analysis to verify how biodiversity characteristics have responded to forest restoration in the Brazilian Atlantic Forest. The study results found that forests undergoing restoration have a recovery gap of 22% for diversity features and 34% for structure features in relation to the reference ecosystem. Creating legal rules is one of the main public policy strategies for mitigating deforestation and facilitating ecological restoration [12,13]. Mohebalian et al. [5] point out that all three governments (Paraguay, Argentina, and Brazil) have taken important legislative measures to protect the remaining Atlantic forests. However, the authors verified that, between 2000 and 2020, roughly 20% of the Atlantic Forest’s expanse was depleted in Misiones, 13% in Paraná, and 18% in Alto Paraná (southern Brazil).
In Brazil, deforestation has resumed an upward trend, reaching the highest rate in the last decade [14,15]. Environmental problems have been aggravated in recent years, at least partially due to setbacks in the protectionist legislation during the last Brazilian federal government [16]. Humanity has severely transformed the structure and function of forest ecosystems and, consequently, this transformation is impacting human health in several important ways, as well as the environmental health of other life forms [17,18]. Lemgruber et al. [19] argue that, in addition to recovering biodiversity and providing ecosystem services, forest restoration contributes to improving social aspects, such as the creation of jobs and the increase in income.
In accordance with relevant regulations, the assessment of the stages of regeneration serves as a primary directive for evaluating petitions to clear secondary vegetation (forests undergoing regeneration), establish reparative actions, and enforce statutory sanctions [20,21]. Despite the advances in the procedure of classifying the regeneration stages, currently, key indicators are used without taking into account the possible differences among various the Atlantic Forest ecosystems [22,23,24,25]. However, several earlier studies have documented the impact of environmental fluctuations, including weather patterns and topographical factors, on the ecosystems of the Atlantic Forest [26,27,28,29,30].
From the above, the objective of this study was to verify whether there are significant differences when these ecosystems are compared at the same regeneration stage. The research hypothesis that these differences exist is based on the fact that the Atlantic Forest biome is characterized by a certain dissimilarity among ecosystems due to the natural variability of environmental conditions, such as humidity, temperature, soil, and water regime [31,32,33].
Without taking into account these possible differences, the classification of regeneration stages becomes an even more challenging and complex task [15,20,26,34,35,36,37]. In summary, the following reasons motivated the present study:
The successional stage of regeneration is the main criterion that guides the decision between deforestation or protection of native vegetation in the Atlantic Forest biome;
Currently, the normative guidelines employed for monitoring, assessing, and classifying the succession stage disregard the potential particularities of each Atlantic Forest ecosystem;
If the hypothesis under scrutiny is corroborated, this study could lend support to the review of guidelines and, consequently, to the decision-making processes concerning land use and environmental preservation in the regions encompassing the Atlantic Forest biome;
Furthermore, a more comprehensive understanding of the variations among the various ecosystems of this biome can facilitate the accurate monitoring and evaluation of reforestation efforts.
- Materials and Methods
This research was conducted in Southern Brazil’s Santa Catarina State, using the forestry inventory database to explore the Atlantic Forest’s ecosystems prevalent in the study area, namely the dense ombrophilous forest (DOF), mixed ombrophilous forest (MOF), and seasonal deciduous forest (SDF) (Figure 1).
Sustainability 15 06823 g001 550
Figure 1. Location of the study area with indication of sampling unities of Atlantic Forest ecosystems. Source: Modified from Bressane et al. [38].
The dataset comprised 460 sample units, each spanning 4000 m2. To perform a comparative analysis of the ecosystems, five explanatory variables were considered, which were prioritized as key indicators by Bressane et al. [38], using multivariate ordering statistics, considering its greater discriminating power for the classification of forest regeneration stages: number of individuals per hectare (N/ha); basal area per hectare (BA/ha); diameter at breast height (DBH); number of species (S); and Shannon biodiversity index (H’); as shown in Table 1 and Table 2.
Table 1. Descriptive characterization of sample grouped by Atlantic Forest ecosystems (AFE).
Table
Table 2. Descriptive characterization of sample grouped by regeneration successional stages (RSS): I = initial, M = medium, A = advanced stage.
Table
Normality and homoscedasticity assumptions were checked using the Shapiro–Wilks and Levene tests, which were rejected at a 5% significance level. Consequently, non-parametric tests were employed for statistical analysis. To compare the DOF, MOF, and SDF ecosystems based on the key indicators, the Kruskal–Wallis test was used, followed by pairwise multiple comparisons in independent groups, employing Dunn’s procedure.
All Atlantic Forest ecosystems were compared at the same regeneration stage. The classification of the stage of the sample units was previously carried out by a group of experts, including biologists and forest engineers. Moreover, Pythagorean tree-based plane fractals were used for visualizing the hierarchy of the key indicators regarding their discriminant power over the Atlantic Forest ecosystems [39].
All analyses were performed using the R software (version 4.0) [40], Jamovi version 2.3, a statistical computing software [41], and Orange Data Mining software (version 3.34) [42], applying two-way tests at a significance level (α) of 0.05, and a power (1 – β) set at 0.95 [43].
Differences among Atlantic Forest ecosystems based on key indicators of regeneration stage (Δ) were also measured considering the greatest difference between means based on Cohen’s d formula, given by:
𝑑=𝑥̲1−𝑥̲2𝑠𝑑12−𝑠𝑑222−−−−−−−√; Δ=𝑑/((𝐷2)+4)−−−−−−−−−√
where 𝑑
is the 𝐷
Cohen value (standardized mean difference), and 𝑥̲𝑖
and 𝑠𝑑𝑖
are the mean and standard deviation of the ith group, respectively. According to Funder and Ozer [44], a Δ around 5% (ρ ≤ Δ < 7.5%) indicates a very small difference, around 10% a small difference (7.5 ≤ Δ < 15%), 20% a medium difference (15 ≤ Δ < 25%), around 30% a large difference (25 ≤ Δ < 35%), and around 40% or greater (Δ ≥ 35%) a very large difference.
- Results
A descriptive characterization of sample units grouped by Atlantic Forest ecosystem and regeneration of successional stages is shown in Figure 2.
Sustainability 15 06823 g002 550
Figure 2. Atlantic Forest ecosystems grouped by regeneration successional stages (RSS).
As can be seen in Figure 2, the different Atlantic Forest ecosystems appear to visually diverge. It is noted that such ecosystems have a probability distribution of occurrences in intervals with different thresholds that overlap but with a certain variation, mainly in the more advanced stages of regeneration.
Table 3 presents the differences among Atlantic Forest ecosystems based on key indicators of regeneration stage, measured using the Cohen index, which considers the distance between the means of the compared groups weighted by the standard deviation.
Table 3. Difference among Atlantic Forest ecosystems based on key indicators of regeneration stage.
Table
Table 3 shows that the differences among Atlantic Forest ecosystems range from very small to very large. The greatest differences were found at the advanced stage of regeneration, in comparisons based on S, which reached 70.1% (DOF × MOF in advanced stage), N/ha with 55.7% (SDF × DOF in advanced stage), and H’ with 49.8% (DOF × MOF).
The comparative analysis indicated that there were statistically significant variations (p < 0.05) between the different forest ecosystems within the Atlantic Forest biome when evaluated at the same stage of regeneration. This confirmed the research hypothesis, as presented in Table 4. The multiple pairwise comparisons are presented in Figure 3, Figure 4, Figure 5, Figure 6 and Figure 7.
Sustainability 15 06823 g003 550
Figure 3. Comparison among Atlantic Forest ecosystems (SDF, DOF, and MOF) in relation to the number of species (S): (a) initial, (b) medium, and (c) advanced stage. The values with different superscript letters (A or B) are significantly different (p < 0.05).
Sustainability 15 06823 g004 550
Figure 4. Comparison among Atlantic Forest ecosystems (SDF, DOF, and MOF) in relation to the individuals per hectare (N/ha): (a) initial, (b) medium, and (c) advanced stage. The values with different superscript letters (A, B or C) are significantly different (p < 0.05).
Sustainability 15 06823 g005 550
Figure 5. Comparison among Atlantic Forest ecosystems (SDF, DOF, and MOF) in relation to the Shannon index (H’): (a) initial, (b) medium, and (c) advanced stage. The values with different superscript letters (A, B or C) are significantly different (p < 0.05).
Sustainability 15 06823 g006 550
Figure 6. Comparison among Atlantic Forest ecosystems (SDF, DOF, and MOF) in relation to the diameter at breast height (DBH): (a) initial, (b) medium, and (c) advanced stages. The values with different superscript letters (A or B) are significantly different (p < 0.05).
Sustainability 15 06823 g007 550
Figure 7. Comparison among Atlantic Forest ecosystems (SDF, DOF, and MOF) in relation to the basal area per hectare (BA/ha): (a) initial, (b) medium, and (c) advanced stage. The values with different superscript letters (A, B or C) are significantly different (p < 0.05).
Table 4. Comparative analyses of variance among Atlantic Forest ecosystems (AFE) interacting with regeneration successional stages (RSS).
Table
The number of species (S) in DOF (69 ± 16) was statistically higher (p < 0.001) than MOF (38 ± 15) and SDF (44 ± 9) at the same regeneration stage. (Henceforth, the notation M ± D stands for the median ± IQR range) in the advanced stage). However, for such indicators, there was no significant difference between the SDF and the MOF, as can be seen in Figure 3.
At the same stage of regeneration, the number of individuals per hectare (N/ha) in the SDF was significantly lower (394 ± 248) when compared with the MOF (475 ± 233) and DOF (545 ± 173) ecosystems. Additionally, the comparison between the ombrophilous Atlantic forests (dense and mixed forests) was significant (p < 0.001) in terms of N/ha, with DOF sample units exhibiting higher values (Figure 4).
The diversity measured by the Shannon index (H’) demonstrated an increase in variation with the progression of regeneration stages. In general, there were no significant differences among all Atlantic Forest ecosystems at the initial stage (p = 0.911). However, in the medium stage, significant differences were observed only in the comparison of DOF (p < 0.001) with the other ecosystems. Finally, in the advanced stage, there was a significant difference among all ecosystems (Figure 5).
Unlike the previous indicator analyzed (H’), the DOF ecosystem showed significantly lower DBH values (19.2 ± 3 in the advanced stage) compared to the SDF (21.6 ± 4, p < 0.001) and MOF (21.5 ± 5, p < 0.001) at the same stage of regeneration, as depicted in Figure 6.
The comparison of BA/ha among Atlantic Forest ecosystems in the medium regeneration stage indicates that SDF (14.5 ± 6) has a significantly lower value than DOF (15.7 ± 6, p = 0.015) and MOF (16.8 ± 6, p = 0.002). However, there was no significant difference between the ombrophilous forests (DOF × MOF, p = 0.375). On the other hand, all formations showed significant differences in the multiple paired comparisons for the advanced stage (Figure 7).
The hierarchy of the key indicators of the stages of regeneration, based on their discriminant power over the Atlantic Forest ecosystems, are shown in the Pythagorean tree plane fractals below (Figure 8).
Sustainability 15 06823 g008 550
Figure 8. Key indicators hierarchy regarding their discriminant power for the Atlantic Forest ecosystems: (a) initial, (b) medium, and (c) advanced stages of regeneration; (𝑏1
) S, (𝑏2
) H’, (𝑏3
) N/ha, (𝑐1
) S, (𝑐2
) BA/ha, (𝑐3
) N/ha, (𝑐4
) BA/ha, and (𝑐5
) N/ha.
In Figure 8a, there is no significant discrimination among the Atlantic Forest ecosystems (DOF, MOF, and SDF), which can be attributed to the greater homogeneity in the initial stage of forest regeneration. In contrast, a clear split among forest ecosystems occurs in the later stages (Figure 8b,c), which is consistent with the results of the comparative analyses (Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6) based on the key indicators (𝑏1
to 𝑏3
and 𝑐1
to 𝑐5
), as discussed further.
- Discussion
The similarity in the initial stage of regeneration was already expected (Figure 8a). The dominance of a few common pioneer plant species is the main reason for the greater homogeneity observed at the beginning of regeneration [45,46]. However, as seen in this research, as the forest progresses towards greater diversity, biotic interactions result in increased complexity and, consequently, traits diverge among ecosystems [21,31]. Thus, the findings of this study are consistent with those of Pastório et al. [26], who also observed significant variation in various structural and compositional characteristics when comparing distinct forest types within the Atlantic Forest biome.
According to Muscarella et al. [47], topography has an impact on abiotic conditions, which in turn can influence the structure and dynamics of ecological communities [48]. The MOF is situated in the central–western region of the study area, with an altitude range of 500 to 1500 m, and stretches towards the border with Argentina [46]. The DOF is in the coastal region with elevations below 700 m, characterized by high temperatures, humidity levels, and cumulative precipitation [49]. The SDF is found in the Uruguay River Basin and spans altitudes from 200 to 600 m. It is recognized by the shedding of leaves during periods of seasonal changes in climate [46,50].
The DOF and SDF regions have a subtropical climate without a dry season and with a hot summer (Cfa), while the MOF ecosystem experiences a temperate summer and severe winters with extreme frosts, despite also having a subtropical climate without a dry season (Cfb) [51]. Thus, the more challenging environmental conditions in MOF areas give rise to the presence of better adapted species, such as Araucaria angustifolia, Dicksonia sellowiana, and Ocotea porosa [52,53].
As shown in Figure 8b,c, the key indicator (b1 and c1) with the greatest discriminating power between DOF and SDF was the number of species (S), and the Shannon diversity index (H’) was between SDF and MOF (b2). According to Oliveira et al. [54], DOF is one of the main forest types in southern Brazil, exhibiting high species diversity. The higher diversity observed in DOF can be attributed to the fact that this forest ecosystem harbors around 80% of all species found in Santa Catarina State, located in southern Brazil [55,56].
Roderjan et al. [30] attribute the lower diversity in FOM areas, particularly those located at higher altitudes, to the climatic severity that imposes selective pressure on species, as mentioned earlier. The MOF, characterized by a misty environment with high relative humidity, is an ecologically selective ecosystem due to its lower incidence of solar radiation and low temperatures. These factors are often associated with shallow soils and higher concentrations of organic matter [27,28,53,57].
The DOF had a significantly higher number of species (S) compared with other Atlantic Forest ecosystems (p < 0.001). This result supports earlier studies, which also reported a high number of species in the DOF, including endemic and exclusive species, as well as a significant presence of epiphytes and palm trees [56].
Scudeller et al. [57] conducted a study on the distribution and abundance of tree species in the Atlantic DOF in southeastern Brazil and found that the tree flora of this vegetation formation is heterogeneous and is associated with altitude, latitude, temperature, and precipitation. The authors concluded that the occurrence of complex and non-linear gradients suggests the importance of several other abiotic features in the spatial distribution and abundance of species in the studied area, in line with Haq et al. [58].
Around 85% of all arboreal species in southern Brazil are found in the DOF region, as highlighted by Siminski et al. [46]. In this present study, there was no statistical difference (p > 0.05) between MOF and SDF based on this indicator (S). Orihuela et al. [59] discussed that the similarity between MOF and SDF areas may be due to the presence of approximately one-third of the same species in both ecosystems.
Comparing the DBH indicator, lower values were found in DOF than in SDF and MOF, particularly in the advanced stage (p < 0.0001). MOF had the highest DBH and BA/ha values, where the latter was the key indicator with the highest discriminating power between MOF and SDF (𝑐2
) and between MOF and DOF (𝑐4
). The dominant presence of Araucaria angustifolia, a characteristic species of the MOF, was the main factor for this result in a specific area, as it presented one of the most significant volumetric growths [28,52,60,61].
As previously described, the Atlantic Forest biome occurs only in South America, specifically in regions of Brazil, Argentina, and Paraguay. For the conservation of the remaining forest ecosystems, legal instruments were created in these countries. In Paraguay, the Zero Deforestation Law was passed in 2004, reducing annual forest loss by 82%. In Argentina, the National Law for the Protection of Native Forests was created in 2007, prohibiting clearing of 73% of what remains of forest areas [62]. In both countries, there was a significant reduction in the high rates of deforestation prior to legal protection; however, guidelines for classifying the stage of forest regeneration were not adopted.
In Paraguay, after the enactment of the guidelines, legislation prohibited the issuance of authorizations for suppression of forest areas for conversion to other uses in certain regions and periods, even for agricultural use or human settlements, with future authorizations dependent on environmental impact studies and the definition of compensatory measures [63]. In Argentina, the authorization for deforestation considers the classification of native forests in the following categories: (i) red, with high conservation value, which includes areas with connectivity value, presence of biological values, and protection of hydrographic basins, in which deforestation cannot be authorized; (ii) yellow, with medium conservation value, which may be degraded but can reach high conservation value with the implementation of restoration activities, in which deforestation cannot be authorized; (iii) green, with low conservation value, for which partial or full deforestation authorization can be obtained [64].
In Brazil, a 1993 decree stands out, which recognized the existence of different stages of regeneration in the Atlantic Forest biome, which began to be adopted as the main criterion for the conservation of forest areas. Therefore, in the study area (southern Brazil), the protection of the Atlantic Forest ecosystems or their suppression for conversion of the area to another type of land use mainly considers the stage of regeneration of the forest. For this classification, basic parameters were used, both qualitative and quantitative, defined in technical guidelines established by the Brazilian National Council for the Environment, without taking into account the possibility of significant differences between the forest ecosystems in this biome.
As a consequence, several studies have shown that classification based on such parameters is a complex and difficult task [38] due for example, to the clear gradient of variation between stages and types of ecosystems associated with the Atlantic Forest present in the coverage area of this biome. As a result, the guidelines have been revised over the years, but there have been no significant changes regarding the parameters applicable to the classification of this biome’s regeneration stages.
The technical guidelines suggest quantitative parameters such as basal area, number of individuals and species, Shannon diversity index, diameter at breast height, total height, and stem height. On the other hand, qualitative parameters like leaf litter, canopy structure, canopy cover, density of lianas and epiphytes, and indicator species should also be considered. Value intervals (quantitative variables) or frequency quantification (qualitative variables) have been established for each parameter, based on which the regeneration stages are classified. However, due to the variation gradients of these parameters, few of them show discriminating power among regeneration stages, as demonstrated by Bressane et al. [38].
As Andreacci and Marenzi [65] point out, the classification of the stages of regeneration of the Atlantic Forest has direct implications on the possibilities of land use. Therefore, the fragility associated with current limitations, which do not consider possible differences between forest ecosystems in the Atlantic Forest, may result in incorrect classifications, with direct impact on the environmental conservation of forests and their undue exploitation for economic purposes.
Despite being criticized for their low effectiveness, particularly in terms of reference values and limits to properly differentiate regeneration stages, the Brazilian guidelines still offer valuable guidance [20,26,65]; however, previous research has failed to take into account the possible variations among the different ecosystems in the Atlantic Forest. Thus, the current research helps to reinforce the evidence that supports the presence of these variations and emphasizes the significance of taking them into account. It is worth noting that, to the best of our knowledge, this is the first study so far to assess these differences on a large scale, considering a representative number of sample units across Santa Catarina State in southern Brazil.
- Conclusions
This study analyzed Atlantic Forest ecosystems in southern Brazil in order to verify whether there is a significant difference when compared at the same stage of regeneration. Statistical procedures were applied to an extensive database from the Santa Catarina State forest inventory, which contains a comprehensive number of sampling units representative of the southern region of Brazil. The statistical significance of key indicators was observed in at least one of the multiple paired comparisons, especially in the advanced stages of forest succession, thereby confirming our research hypothesis.
Considering that logging permits, penalties for illegal suppression of native vegetation, and corresponding compensation measures mainly depend on the stage of forest regeneration, the obtained results can significantly influence the decisions of experts, stakeholders, and authorities. Moreover, our results can contribute to the improvement of guidelines for forest regeneration monitoring and assessment.