NATURAL REGENERATION AND CONSERVATION STATUS OF THE TREE COMMUNITY OF FOREST REMNANTS IN URBAN PARKS IN SOUTHERN BRAZIL

Purpose: The study characterised the dynamics of the natural regeneration process of urban remnants in four Curitiba (PR, Brazil) urban parks, aiming to contribute to the green infrastructure management process in Araucaria Forest areas. Theoretical framework: Urban forests offer significant socio-environmental benefits, being relevant for maintaining ecosystem services for cities and their inhabitants. Studying the regeneration of forest remnants in urban environments is essential for analysing their conservation status. Method/design/approach: The floristic diversity, structure, and population dynamics of the shrub-regenerating component of urban forest remnants were studied through the analysis of phytosociological parameters and the similarity between the shrub-regenerating, and arboreal components was verified. Results and conclusion: Significant dissimilarity was found between the study areas' regenerating shrub and tree components. The forest remnant of the Municipal Zoo, located in a locality with greater integration of different green areas, showed the best conservation status and a more advanced process of forest succession. Research implications: The analysis of the natural regeneration of urban remnants serves as an essential diagnosis regarding their conservation status and the capacity of species to perpetuate in the place, in addition to inferring about the connectivity with other urban remnants, being of great importance for the planning of actions aimed at the management of urban forests.


INTRODUCTION
In 2018, approximately 55% of the world's population resided in urban areas, with projections indicating that this proportion is expected to increase to 68% by 2050 (United Nations, 2019). In addition, cities are responsible for 60 to 80% of energy consumption and generate up to 70% of man-induced greenhouse gas emissions and land use changes (Un-Habitat, 2019). Future urban growth and the subsequent use of land and natural resources will play a crucial role in achieving an environmentally sustainable future.
Thus, urban forest remnants play a primary role in promoting ecosystem services, as well as in mitigating and adapting to the effects of climate change (Fernández;Wu;Simonetti, 2018;Steenberg;Duinker;Nitoslawski, 2019;Esperon-Rodrigues et al. 2022), as well as improving human health conditions by promoting well-being, as well as preventing the emergence of epidemic diseases (Azevedo et at., 2020;Pamukcu-Albers et al., 2021).
The United Nations (UN) in 2015 launched the 17 Sustainable Development Goals (SDGs), targeting inclusive, safe, resilient and sustainable cities. Among them, SDGs 3, 13, 15 and 11, related to health, climate change, terrestrial life and sustainable cities, are directly related to the importance of biodiversity preservation and plant cover in urban areas.
In this sense, the understanding of the ecological quality of the remaining forests that are part of urban forests can contribute to the improvement of urban green infrastructure planning, guiding economic and environmental zoning through municipal master plans, as well as other integrated actions from different spheres of government (Macedo;Jacobi, 2019). In addition, it can stimulate the support of civil society for the maintenance of the private green areas (Macedo;Jacobi, 2019) by encouraging the creation of Private Natural Heritage Reserves (RPPN), In addition, urban environmental sustainability is closely related to the correct management of green infrastructure (Steenberg;Duinker;Nitoslawski, 2019;Barona et al., 2020) and the analysis of the natural regeneration of urban forest remnants can help planning actions for better integration and maintenance of urban forest, as it provides managers with information regarding the floristic diversity, estimation of species population parameters and the conservation status of these remnants.
In the South of Brazil, the Mixed Ombrophilous Forest (FOM), also known as Araucaria Forest, presents remnants in an advanced stage of ecological succession that total little more than 7% of its original area in Brazil (Wrege et al., 2017) and do not reach 1% in the state of Paraná, where it was once predominant (FUPEF, 2001), which makes this forest phytophysiognomy one of the most threatened in the Atlantic Forest Biome.
With almost 2 million inhabitants, Curitiba is the largest city in the southern region of Brazil and from 2000 to 2022 showed a population increase of approximately 12% (IBGE, 2023). It presents a history of urban planning that values sustainability, being a reference among Brazilian capitals from the 70s, by public policies of urban planning and implementation of projects for the conservation of green areas, besides the establishment of specific, preventive and innovative legislation (Rechia, 2004). Although its forest cover is significant in comparison to other Brazilian capitals, Vieira and Bionde (2008) observed that between 1986 and 2004 there was a reduction from 39% to 30% as a result of urban expansion.
Given the current scenario, it becomes of great importance to investigate the ecological quality of FOM forest remnants in the urban ecosystem, aiming at a better understanding as to their resilience to anthropic impacts. In this sense, this study analyzed the structure, diversity and natural regeneration of urban remnants of FOM present in four urban parks of Curitiba, with the objective of characterizing the population dynamics in these remnants and diagnose the conservation status, in order to contribute to the design of actions aimed at the management and management of urban forest remnants with areas of this phytophysiognomy.
The forest remains surveyed are located in four municipal public parks located in the urban perimeter, among them: Municipal Zoo of Curitiba/Regional Park of Iguaçu (ZOOM), Municipal Natural Park Barigui (BARI), Municipal Park Barreirinha (BARE) and Municipal Natural Park of Leisure "General Iberê de Mattos" (BACA).
For the study of the shrub-regenerating component, henceforth called arb-reg, the field data survey was performed by simple random sampling using temporary plots of 5 m x 5 m (25 m 2 ), allocated in the upper right quadrant of the plots used by Santos et. al (2023), and all individuals with diameter at chest height (DAP) < 15 cm were measured, including individuals killed standing. The number of plots varied in each park according to the total area of its remnants: 20 plots in ZOOM (0.05 ha), 20 plots in BARI (0.05 ha), 16 plots in BARE (0.04 ha) and 10 plots in BACA (0.03 ha). The taxonomic classification followed the system of Angiosperm Phylogeny Group IV (2023) and the valid nomenclature was verified in Flora e Funga do Brasil (REFLORA, 2023). Classification was also done regarding the form of life (bush, tree, palm, liana and arborescent fetus), ecological group (pioneer and non-pioneer) and dispersion syndrome (anemocoric, autocoric and zoochronic).
The analysis of the sample sufficiency of species richness occurred by means of rarefaction curves by the method of interpolation and extrapolation, adopting the number of individuals as a criterion of standardization. For this, the iNEXT package (Hsieh et al., 2016), R programming language (R Core Team, 2022) was used.
Phytosociological parameters were calculated according to Ellenberg and Mueller-Dombois (1974), with absolute and relative values for density, dominance and frequency, as well as the Value of Importance (IV). In order to compare the remainder, the Shannon-Wiener (H') diversity index (Shannon;Weaver, 1963), the Simpson dominance index (D') (Simpson, 1949) and the Pielou (J') uniformity index (Pielou, 1966) were calculated. Phytosociological analysis considered dead individuals for their relevance in the study of secondary forest succession.
For each area, individuals in arb-reg were classified into three size classes: class 1height between 1.30 m and 2.0 m; class 2 -height between 2.01 m and 3.0 m; and class 3height above 3.0 m (Silva et al. 2010;Passos et al. 2021), with the Natural Regeneration Rate of Species per Size Class (RNC) estimated, using the equation RNC = RD + RF / 2 where: RD is the relative density of species; RF is the relative frequency of species, values size classes considered (Volpato, 1994;Silva et al., 2010;Passos et al. 2021). Then, the Total Natural Regeneration Rate of the Sampled Population (TRN) for each species was obtained by the equation TRN = Σ (RNC) / Z, where Z = number of size classes (Volpato, 1994;Passos et al. 2021).
The phytosociological data of the tree component collected by Santos et al. (2023) were used for the analysis of floristic similarity with the arb-reg component. The Bray-Curtis method was adopted as an index of similarity, using abundance data of different species, at the level of plots, using Principal Coordinate Analysis (PCoA/ACoP) and Multivariate Permutational Variance Analysis (PERMANOVA), based on 9999 permutations. The similarity percentage test (SIMPER) indicated the species responsible for the difference between the two components. The program PAST v.3.25 (Hammer, 2001) was used for similarity analyzes.

RESULTS AND DISCUSSIONS
Considering the four areas of study, a total of 1,522 live individuals belonging to 115 species distributed in 39 families, one unidentified and 152 individuals standing dead, were obtained in the sampling. It was observed that a total of 82.5% of the species identified are of arboreal habit and 84.2% of the species show zoographic dispersion. Four invasive alien species have been identified: Ligustrum lucidum W.T.Aiton, Pittosporum undulatum Vent., Hovenia dulcisThunb. andEriobotrya japonica (Thunb.) Hey, Lindl. Figure 01 presents the rarefied species curve, revealing that the rarefied species richness of BARE and BARI form a distinct group of BACA and ZOOM. It is possible to verify that the median curves present a difference between the parks, however the confidence intervals overlap indicating the absence of a significant difference between the remaining ones. The species richness in the arb-reg component of urban remnants ranged between 59 (ZOOM) and 46 (BACA) species (Table 01), while Bardal et al (2004) observed a total of 39 species and 23 botanical families in a remnant of FOM in the metropolitan region, in a total sample area of 0.25ha.
Myrtaceae (21 species), Lauraceae (9 spp), Fabaceae (8 spp), Solanaceae (8 spp) and Rubiaceae (8 spp) were the families with the greatest richness of species in arb-reg and together accounted for 47% of the total of the species identified. The expressiveness of the Myrtaceae and Lauraceae families in species richness is consistent with floral patterns from other studies in FOM remnants (Souza et al. 2012;Cordeiro et al. 2013;Fiorentin et al. 2015;Higuchi et al. 2015;Santos et al. 2015;Passos et al. 2021a;Passos et al. 2021b).
The diversity and uniformity indices for the four parks are shown in table 01, it being observed that ZOOM presented the highest values for H' and for J', being the highest value for the index of D' observed in BACA (Table 01).
In relation to the alpha diversity of the arboreal-regenerating component, the values of H' varied from 2.56 to 3.43, indicating a diversity of species similar to other remnants of Mixed Ombrophilous Forest located outside urban areas. H' values were found between 2.49 nats.ind -1 and 3.76 nats.ind -1 in other areas of Mixed Ombrophilous Forest (Bardal et al., 2004;Passos et al., 2021;Gonçalves et al., 2022). Values close to 3 nats.ind -1 are expected for anthropized remnants in a secondary succession stage in this domain, characterizing a median diversity (Nascimento et al., 2001).
Considering the H' values of the tree component of the study areas that ranged between 3.06 and 3.52 nats.ind -1 (Santos et al. 2023), no increase in diversity was observed in relation to arb-reg. A similar situation was observed in the Araucaria State Park (PEAR), Santa Catarina, when comparing the values of this index for the tree component (H'=3.96 nats ind -1 ) and for the regenerating component (H'= 3.76 nats ind -1 ) (Passos et al. 2021a;Passos et al. 2021b). However, Bardal et al. (2004) reported higher H' value in the lower diametric value component compared to the tree component, attributing the increase in H' value to the entry of new regenerating species (Bardal et al. 2002;Bardal et al. 2004). Table 01. Comparison of the general phytosociological parameters from the size classes established for the analysis of the arb-reg component and arboreal component (Santos et. al., 2023) of the studied remnants, Curitiba, PR, Brazil.

Phytosociology of the Shrub-Regenerating Component of Curitiba Urban Parks
The phytosociological parameters for the species with the 15 highest values for the Total Natural Regeneration of the Sampled Population (TNR) rate, as well as for the standing dead individuals, are presented in table 02.
The remainder of ZOOM presented the greatest species richness, with 59 species distributed in 25 families, and no exotic species were recorded. With the largest NRTs, Mollinedia elegans and Psychotria suterella, both of shrub habit, stood out. It was observed that this remnant recorded the largest number of shrub species (11 species in total). Among the arboreal species, Sorocea bonplandii, Myrceugenia miersiana andCabralea canjerana were the most representative. Only one pioneer species, Seguieria aculeata Jacq. (TRN = 0.3%), was identified in arb-reg, which indicates low representativeness of this group in this remnant, since, according to Santos et al. (2023), the pioneer species are in the canopy, making up only 5.3% of the species in the tree component.
In BARI, 53 species from 23 families were identified, with emphasis on the shrub species Rudgea jasminoides and Psychotria suterella with the largest TRNs. These two species accounted for 49% of the total absolute density. In the group of tree species, Ocotea silvestris and Alophylus edulis were the most representative in TRN. Although not prominent about TRN,  In the shrub-regenerating component of BARE, there were 52 species belonging to 25 families, all 15 species with the highest TRN of arboreal habit, besides being the only remaining to include an exotic species in this ranking: Pittosporum undulatum, popularly known as incense wood. This exotic species was also observed earlier by Mielk et al. (2015), who suggested a possible structural change and impact on ecological balance due to the high incidence of this species, which highlights the need to implement population control measures to mitigate its impact on the ecological succession of native species of this remnant. In addition to this invasive alien species, Hovenia dulcis Thunb. (grape from japan) has also been observed in the shrub-regenerating component, although it is not among the species with the highest TRN. As for the native species, four are pioneers: Miconia sellowiana, Cordyline spectabilis, Solanum swartzianum and Clethra sgoba. This was the only remnant where the shrub species Mollinedia clavigera, commonly found in soft slopes and fom undergroves (Peixoto, 2002), did not stand out among the 15 species with the largest TRN.
In BACA, 46 species and 23 families were recorded. The shrub species Mollinegia clavigera (24.5%), which alone accounted for 31% of the total absolute density, showed also the largest TRN. Among the arboreal species, the highlights were Casearia decandra, Eugenia uniflora andCupanea vernalis. This was the only remnant that recorded two pioneer species, Myrsine parvula and Solanun sanctae-catarinae, among the largest NRTs. Although not prominent for TRN, exotic species Ligustrum lucidum and Eriobotrya japonica (TRN of 0.6% and 0.5%, respectively) were recorded.
Santana et. al. (2019)  highlighting the influence of the historical process of using the area and the composition of species in the secondary succession in urban remnants. In addition, the absence of a closed canopy favored the development of pioneer species, by allowing a greater entry of light into the undergrove (Santana et al., 2019).
Other factors related to anthropic intervention, such as proximity to public roads and soil degradation, may also influence the floristic composition of forest remnants (Silva et al. 2010;Aguiar et al. 2017;Oliva et al., 2018;). Furthermore, seeking to understand the landscape as a mosaic facilitates the analysis of structural modifications of anthropic scope, incorporating the complexity of spatial relations between natural and cultural elements. This approach provides a comprehensive and detailed view of transformations over time, revealing the delicate interaction between the environment and human activities. (Fontgalland;Oliveira, 2023).
In the case of this study, although the remnants comprise areas protected by the municipal government, the presence of trails inside the remnants, accumulation of garbage and the intensity of unmonitored visitation, suggest the possibility of increased depredation, being points that may in a way influence the composition of the arb-reg.
As for A. angustifolia, symbol of FOM, no individuals were registered in the arb-reg component, although it was generally distinguished in the tree component of the areas of study (Santos et al., 2023). Although it is a common species in FOM areas, Santos et al. (2018) reported that different studies on regeneration in the FOM undergrove point to a low representativeness of A. angustifolia in this component.
The regeneration of the araucaria depends on the formation of clearings in the canopy to establish itself and develop in its habitat (Narvaes et al., 2005;Beckert et al., 2014;Longhi et al., 2018;Vefago et al., 2019). Studies suggest that the araucaria may face difficulties of establishment and low capacity of regeneration in the sub-woods of well-conserved remnants of the FOM. Furthermore, climate change has a negative impact on the conservation of this species, affecting processes such as reproduction, germination and seed dispersion (Marchioro, Santos, Siminski, 2020;Wrege et al., 2017;Taglieri et al., 2021). These factors highlight the importance of implementing adequate management and monitoring of regeneration, especially in protected areas such as urban parks. It is essential to guarantee the protection of these environments so that the araucaria and the species associated with it are conserved for the maintenance of biodiversity, especially in the urban environment.
In the floristic composition of arb-reg of all areas, the predominance of zoo and nonpioneer ecological species was observed. Similarly, Passos et al. (2021) also reported the predominance of species this dispersion syndrome and non-pioneers in the regenerating and arboreal components in a remnant of FOM in Santa Catarina (Passos et al., 2021 a and b).
The predominance of species with zoo dispersion syndrome may indicate the substantial presence of frugivorous animals in the urban environment, contributing to the dispersion of propagules of these species. Other authors (Miretzki, 2003;Almeida et al., 2011;Straube et al., 2009;Gadda et al., 2021) have already reported the presence of different species of mammals and birds in the urban remnants of Curitiba, which characterizes the importance of these areas in providing shelter and food for these animals, reiterating the relevance of urban remnants for the maintenance of local fauna. In the light of the above, the expansion of studies referring to the knowledge of biodiversity in urban remnants can raise awareness of the importance of environmental sustainability and stimulate pro-environmental behavior (Naparin;Helmi, 2023) among visitors and the population around the parks.
As for pioneer species, the highest percentage of species in arb-reg was in BARE (13.5%), which also showed the highest TRN of species in this group (8.73%), while ZOOM recorded the lowest percentage of species and consequently the lowest TRN for this group. The decrease in the pioneer species indicates an advanced stage of successional development of the forest, with more restrictive conditions for these initial groups (Paula et al., 2004). However, pioneer species can occur in intermediate or advanced stages of succession due to the formation of clearings (Higuchi et al., 2006). The opening of clearings in the stages of forest maturation is related to the death of pioneer trees in the canopy, restarting regeneration in the undergrove with the presence of pioneer and non-pioneer species (Chazdon, 2012).

Horizontal structure of natural regeneration
The mean density of the arb-rege component of the four areas was 10,049 ind.ha -1 . With respect to regeneration classes, there was a reduction in the absolute density of individuals between the smallest (RNC1) and the largest class (RNC3), with a decrease of 35.8% in BARE,30.7% in ZOOM and 16.1% in BARI. This reduction may be related to the gradual replacement of pioneer species by nonpioneer species indicating a successional advance (Machado et al., 2017). This trend of substitution occurs due to faster growth and earlier competition of pioneer species in relation to non-pioneer species, which are progressively replaced by species with less need for light and slower growth (Whitmore, 1989). Duarte et al. (2019), when analyzing environmental and spatial variables that influence the distribution of species in natural regeneration, observed that the opening of the canopy was the most influential factor.
In addition, the reduction in species richness and D' was observed, while uniformity increased in the remaining studied. Diversity, as measured by H', was smaller only in ZOOM. In the remaining BACA, there was a 19.5% increase in absolute density between the RNC1 and RNC3 classes, but there were no significant differences in species richness and H', J' and D' indices between these classes.
The increase in absolute density between these classes of regeneration, as well as the lower values of richness, diversity and uniformity, were also reported by Souza et al. (2018), concluding that changes in the pattern of size in regeneration, in which species of the tree community are best represented in the smaller class, affect the population but not the species. In fact, Higuchi et al. (2015) reported a considerable reduction in the abundance of the smallest class to the largest class in a fragment of FOM in the Santa Catarina Plateau. Similarly, ecological filters limit population size, resulting in higher uniformity and lower plant density, which may indicate that despite the good initial establishment in the area, there is a decrease in species that maintains its development until the adult phase (Souza et al., 2018).
Differences in the density of the regenerating classes can also be attributed to the initial period of establishment of individuals from the plantule bank in the forest (Valladares;Niinemets, 2008). In addition, biotic and abiotic phenomena can reduce the population, resulting in a smaller number of individuals, these being larger (WEBB et al., 2006;Poorter, 2007;Sccoti et al., 2011).
The highest density of standing dead was observed in BARE (1,300 ind.ha -1 ), as well as the highest absolute frequency (93.8%), which characterizes a good distribution of this group among sample units. ZOOM and BARI presented values close to each other, and in BARI this group presented the highest IV (29.1%), indicating a high representativeness in the remainder. Relative dominance for both BARI and BARE was about 12%. The high density of standing dead individuals in BARE, coupled with the greater number of individuals of pioneer species, may indicate that this remnant is in a more dynamic succession process, compared to the other areas of study.
A high tree density may portray a limited availability of resources such as light and nutrients, especially in the lower strata of the remainder, which may lead to lower growth rates and high mortality rates (Longhi et al., 2018). Mortality rates are related to self-regulation processes in natural tropical forests, and are associated with growth and regeneration, leading to the loss and continuous substitution of trees, until reaching steady-state equilibrium (Swaine et al., 1987).

Similarity between tree and shrub-regenerating component
The ordering analysis (PCoA) for the remnants pointed to the formation of groups, evidencing differences in the floristic composition between the shrub-regenerating component and the arboreal of the four parks ( Figure 02). For the remaining ZOOM, BARI and BACA it was observed the overlap of the formed groups, and in the latter there is a significant sharing of species. In BARE, there was no overlap of groups, and it is possible to observe a lesser floristic similarity between the tree and shrub-regenerating components when compared to the other remnants.
In ZOOM, a total of 80 species were identified, considering the data of the tree component (Santos et al., 2023) and arb-reg. The remainder recorded the highest percentage of species shared according to the SIMPER analysis, corresponding to 45%, all being non-pioneer and tree habit. Although this percentage is higher than that observed by Rodrigues and Magalhães (2011) in the National Forest (FLONA) Mário Xavier, RJ, was lower than the percentage presented by Aguiar et al. (2017) and Seki et al. (2022), both in FOM areas, with approximate percentage of 57%. ZOOM presented 23 species exclusive to the arb-reg component, 48% of which were shrubs. The SIMPER analysis of ZOOM, pointed out a general average dissimilarity of 82%, pointing out the species of shrub habit Mollinedia elegans (8.4%) and Psychotria suterella (6.2%), registered exclusively in the arb-reg component, with greater contributions to the difference between the components.
It is important to point out that ZOOM integrates a large urban park area, the Municipal Park of Iguaçu, in the southeastern portion of the city, located in the Administrative Regional Boqueirão, which according to Monteiro (2015) holds the largest concentration of urban remnants of Curitiba (37.35%), which may have contributed to the registration of a greater number of species in arb-reg (Table 01) compared to the other remnants.
In BARI, of a total of 88 recorded species, 35.2% are shared between the arboreal and arb-reg communities, among them the invasive exotic species Eriobotrya japonica. The arbreg component showed 22 exclusive species, of which 36% are pioneer arboreal. The overall mean dissimilarity was 84%, with Rudgea jasminoides (22.3%), Palicourea sessilis (6.3%) andPsychotria suterella (6%) being indicated by SIMPER with the largest contributions to the difference between the components.
A total of 74 species were found in BARE, of which 39.2% are shared between both components, including Hovenia dulcis and Pittosporum undulatum. It was observed that 31.1% of the species are exclusive to arb-reg, being 8.1% shrubs. The SIMPER analysis for this remainder indicated overall average dissimilarity of 76%. The tree species Ocotea bicolor (10.3%) and Palicourea sessilis (9%), although shared, showed the largest contributions to dissimilarity, in addition to the higher Importance Values (IV) in arb-reg, indicating a high ecological value in this component.
In turn, BACA presented a total of 63 species, with the lowest percentage of shared species (33.3%) among the remaining analyzed. The exclusive species in arb-reg accounted for 25%, of which 4.5% were shrubs. The overall mean dissimilarity was 75%, with Mollinedia clavigera (17%) and Allophylus edulis (5.2%) the largest contributions to dissimilarity. BACA, despite being the smallest remnant among the parks studied, connects directly with the Natural Heritage Reserve of the Bacacheri (Municipal Decree No. 0464/2011) and private remnants, expanding the total area of forest, which contributes to a better state of conservation and flow of dispersing animals. Different factors may influence the dissimilarity between the two components analyzed, with the natural dynamics of species substitution along the ecological succession being one of the most relevant factors (Seki et al. 2022;Magalhães, 2011), since it influences the floristic composition (Oliva et al., 2018). Nevertheless, the presence of smaller tree species and shrub species, the presence of which is predominant in the undergrowth, should also be considered in the analysis of the dissimilarity between arb-reg and tree components, for an adequate management of the remaining forest.

FINAL CONSIDERATIONS
The analysis of the regenerative bush component recorded representative rate of Total Natural Regeneration for species that stood out in the tree component of the urban forest remnants studied, in particular representative species of the canopy and intermediate stratum in the tree component.
The majority presence of zoo species, which demand the circulation of seed dispersers in the area, may provide additional relevance to the establishment and conservation of ecological corridors, important in urban ecosystems subject to a great fragmentation of the landscape. In addition, the high representativeness of zoo species in the forest fragments may explain the introduction of new species into the remnants.
The proximity of the urban fragments sampled to other green areas probably had a positive effect on the richness and functional composition of species. The remaining forest located in the Municipal Zoo stood out for the state of conservation among the other areas, for presenting a more advanced stage of ecological succession, which may be related to its location, integrated to a more extensive forest cover (Iguaçu Regional Park). This, added to other green areas located adjacent to the park, composing vegetation corridors, may have positively influenced the higher percentage of regenerating species observed in this fragment compared to the other areas, with emphasis on the large percentage of non-pioneer and zoo species.
The characterization of regeneration in the remaining studies reiterates the importance of these areas for the maintenance of the species characteristic of FOM phytophysiognomy in urban environment, as well as for local biodiversity, as it demonstrates the resilience of the remaining areas in maintaining the ecological balance and, consequently, in providing the ecosystem services essential for urban environmental sustainability.
The large presence of zoo individuals in regeneration indicates that the remains studied actively contribute to the integration and maintenance of the fauna in the urban environment, serving as a source of food and shelter and natural processes of urban forest renewal. Thus, in view of the peculiarities of the urban environment, the analysis of the floristic composition and structure of the natural regeneration in a continuous way, would make up a good strategy for monitoring the state of conservation of these environments, measuring anthropic impacts and their effects over time, configuring an important tool for managing the urban natural heritage.
The metrics calculated in this study for the botanical species of the regenerating shrub component indicate that ecological succession is occurring adequately, so as to contribute to the environmental quality of urban parks, since the diversity of the floristic composition contributes to the maintenance of a well-structured community of arboreal and shrub species, as well as its dispersing agents, making it more resilient to future disturbances. Thus, in view of the peculiarities of urban ecosystems, the analysis of the floristic composition and structure of regeneration over time would represent a good instrument for monitoring the conservation status of these forest stands, considering their importance in providing numerous ecosystem services essential to the quality of life of the population living in large urban centers.