Abstract

The highly biodiverse Atlantic Forest has been heavily exploited and its remnant areas, which are mostly composed of secondary forests, cover less than 20% of its original extension. This study analyzed the structure, aboveground biomass (AGB) and species diversity of a secondary evergreen rainforest in the state of Santa Catarina, southern Brazil. Individuals with diameter at breast height ≥ 5 cm on 22 sample plots were measured, totalizing 3.49 hectares of sampled area. A mean AGB of 201.6 Mg ha^-1 (standard deviation = 35.5) was found; Hyeronima alchorneoides and Miconia cinnamomifolia represented 29.1% and 7.6% of the community AGB ha^-1, respectively. A total of 153 species distributed in 107 genera and 55 families was observed; it was estimated that 121 species (95% confidence interval: [117, 125]) occurred in one hectare. These results highlight the importance of secondary forests for biodiversity restoration/conservation and carbon sequestration.

Keywords:
secondary forest; forest carbon; Mata Atlântica; hotspot

1. INTRODUCTION AND OBJECTIVES

The Atlantic Forest is among the Earth’s hotspots for biodiversity conservation, having great levels of tree species endemism (Myers et al., 2000; Lima et al., 2020Lima RAF, Souza VC, Siqueira MF, ter Steege H. Defining endemism levels for biodiversity conservation: tree species in the Atlantic Forest hotspot. Biological Conservation 2020; 252:108825.). Nevertheless, historical and ongoing deforestation reduced its area to less than 20% of its original cover (Ribeiro et al., 2009Ribeiro MC, Metzger JP, Martensen AC, Ponzoni FJ, Hirota MM. The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biological Conservation 2009; 142:1141-1153.), and most of its remnants consist of fragmented forests often disturbed by human activities such as selective logging (Schneider et al., 2018Schneider LCA, Silva MT, Agostinetto L, Siegloch AE. Deforestation in Mixed Ombrophilous Forest in the serrana region of Santa Catarina. Revista Árvore 2018; 42:e420206. ), cattle grazing within stands (Pereira et al., 2015Pereira JAA, Oliveira-Filho AT, Eisenlohr PV, Miranda PLS, Lemos Filho JP. Human impacts affect tree community features of 20 forest fragments of a vanishing neotropical hotspot. Environmental Management 2015; 55:296-307.), and hunting (Bragagnolo et al., 2019Bragagnolo C, Gama GM, Vieira FA, Campos-Silva JV, Bernard E, Malhado AC et al. Hunting in Brazil: What are the options? Perspectives in Ecology and Conservation 2019; 17:71-79. ). However, the Atlantic Forest still holds a surprising number of arborescent species (Lima et al., 2020Lima RAF, Souza VC, Siqueira MF, ter Steege H. Defining endemism levels for biodiversity conservation: tree species in the Atlantic Forest hotspot. Biological Conservation 2020; 252:108825.), many of them with some degree of rarity in nature (Caiafa & Martins, 2010Caiafa AN, Martins FR. Forms of rarity of tree species in the Southern Brazilian Atlantic rainforest. Biodiversity and Conservation 2010; 19:2597-2618.; Oliveira et al., 2019aOliveira LZ, Gasper AL, Lingner DV, Sevegnani L, Vibrans AC. Secondary subtropical Atlantic forests shelter a surprising number of rare tree species: outcomes of an assessment using spatially unbiased data. Biodiversity and Conservation 2019a; 28:751-768.).

Among the multiple forest types occurring in the Atlantic Forest domain, the evergreen rainforest - ‘Floresta Ombrófila Densa’ (IBGE, 2012IBGE - Instituto Brasileiro de Geografia e Estatística. Manual técnico da vegetação brasileira. Brasília: IBGE; 2012.) - stands out as the one with the greatest tree/shrub species richness and longitudinal range (Oliveira-Filho & Fontes, 2000Oliveira-Filho AT, Fontes MAL. Patterns of floristic differentiation among Atlantic forests in southeastern Brazil, and the influence of climate. Biotropica 2000; 32:793-810.). This forest type is characterized by a canopy composed of numerous tree species, multiple vertical layers, and an abundance of climbers and epiphytes (Roderjan et al., 2002Roderjan CV, Galvão F, Kuniyoshi YS, Hatschbach GG. As unidades fitogeográficas do estado do Paraná, Brasil. Ciência & Ambiente 2002; 24:75-92.; IBGE, 2012IBGE - Instituto Brasileiro de Geografia e Estatística. Manual técnico da vegetação brasileira. Brasília: IBGE; 2012.). Regarding tree/shrub species composition, evergreen rainforests have the botanical families Myrtaceae, Fabaceae, Lauraceae, Rubiaceae and Melastomataceae as the most speciose ones (Oliveira-Filho & Fontes, 2000Oliveira-Filho AT, Fontes MAL. Patterns of floristic differentiation among Atlantic forests in southeastern Brazil, and the influence of climate. Biotropica 2000; 32:793-810.; Lingner et al., 2015; Brotto et al., 2019Brotto ML, Lozano ED, Marinero FEC, Uhlmann A, Blum CT, Roderjan CV. Lauraceae along an altitudinal gradient in southern Brazil. Floresta e Ambiente 2019; 26:e20170637. ).

The state of Santa Catarina is totally inserted in the Atlantic Forest and most likely has the best studied flora among the Brazilian states. A large body of knowledge about its flora was developed based on field surveys carried out since the 1950s (e.g., Veloso & Klein, 1951Veloso HP, Klein RM. O problema ecológico: Vegetação - Bromeliáceas - Anofelinos. I - A presença relativa das formas aquáticas A. (Kertezia) spp. Como índice de positividade das espécies de Bromeliáceas. Sellowia 1951; 1:187-270., 1957Veloso HP, Klein RM. As comunidades e associações vegetais da mata pluvial do sul do Brasil. I. As comunidades do município de Brusque, Estado de Santa Catarina. Sellowia 1957; 1:81-235., 1959, 1961, 1968; Reitz, 1965Reitz R. Plano de Coleção. In: Reitz R, editor. Flora Ilustrada Catarinense., Itajaí: Herbário Barbosa Rodrigues ; 1965.; Klein, 1978Klein RM. Mapa fitogeográfico do Estado de Santa Catarina. In: Reitz R, editor. Flora ilustrada Catarinense. Itajaí: Herbário Barbosa Rodrigues; 1978., 1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389., 1984Klein RM. Aspectos dinâmicos da vegetação do Sul do Brasil. Sellowia 1984; 36:5-54. ), and more recently by the Forest and Floristic Inventory of Santa Catarina (Vibrans et al., 2020Vibrans AC, Gasper AL, Moser P, Oliveira LZ, Lingner D, Sevegnani L. Insights from a large-scale systematic inventory in the southern Brazilian Atlantic Forest. Scientia Agricola 2020; 77:e20180036.). However, classical studies on structure and diversity of Atlantic Forest remnants are still needed, as stressed by Lima et al. (2015Lima RAF, Mori DP, Pitta G, Melito MO, Bello C, Magnago LF et al. How much do we know about the endangered Atlantic Forest? Reviewing nearly 70 years of information on tree community surveys. Biodiversity and Conservation 2015; 24:2135-2148.). These authors demonstrated that the existing knowledge on structure of the Atlantic Forest is based solely on sampled areas that altogether represent merely 0.01% of its total remaining area. Moreover, a substantial portion of the results of studies on this topic are registered in non-indexed vehicles, hampering the retrieval of information to carry out ecological analyses at larger scales (Caiafa & Martins, 2007Caiafa AN, Martins FR. Taxonomic identification, sampling methods, and minimum size of the tree sampled: implications and perspectives for studies in the Brazilian Atlantic Rainforest. Functional Ecosystems and Communities 2007; 1:95-104.; Lima et al., 2015Lima RAF, Mori DP, Pitta G, Melito MO, Bello C, Magnago LF et al. How much do we know about the endangered Atlantic Forest? Reviewing nearly 70 years of information on tree community surveys. Biodiversity and Conservation 2015; 24:2135-2148.). Local information on forest structure and diversity is also essential to underpin the sustainable use and restoration of forest resources (Zambiazi et al., 2021Zambiazi DC, Fantini AC, Piotto D, Siminski A, Vibrans AC, Oller DC et al. Timber stock recovery in a chronosequence of secondary forests in Southern Brazil: Adding value to restored landscapes. Forest Ecology and Management 2021; 495:119352.), regional biodiversity conservation and monitoring (Gardner, 2010Gardner T. Monitoring forest biodiversity: improving conservation through ecologically responsible management. London: Earthscan; 2010. ), and payment for ecosystem services, such as carbon sequestration (Alarcon et al., 2017Alarcon GG, Fantini AC, Salvador CH, Farley J. Additionality is in detail: Farmers’ choices regarding payment for ecosystem services programs in the Atlantic forest, Brazil. Journal of Rural Studies 2017; 54:177-186.).

In this context, this study sought to analyze the structure, biomass and timber stocks, and species composition and diversity of a secondary evergreen rainforest stand in the state of Santa Catarina, southern Brazil. Data were gathered over a sampled area of 3.49 ha, which is nearly 4.5 times larger than the average sampled area of studies carried out in the Atlantic Forest (Lima et al., 2015Lima RAF, Mori DP, Pitta G, Melito MO, Bello C, Magnago LF et al. How much do we know about the endangered Atlantic Forest? Reviewing nearly 70 years of information on tree community surveys. Biodiversity and Conservation 2015; 24:2135-2148.).

2. MATERIALS AND METHODS

2.1. Study area

The study area consists of a 42-ha private property in the state of Santa Catarina, southern Brazil (26°32’01”S, 49°02’30”W), located within the evergreen rainforest region (Figure 1) (Klein, 1978Klein RM. Mapa fitogeográfico do Estado de Santa Catarina. In: Reitz R, editor. Flora ilustrada Catarinense. Itajaí: Herbário Barbosa Rodrigues; 1978.; Oliveira-Filho et al., 2015Oliveira-Filho AT, Budke JC, Jarenkow JA, Eisenlohr PV, Neves DRM. Delving into the variations in tree species composition and richness across South American subtropical Atlantic and Pampean forests. Journal of Plant Ecology 2015; 8:242-260.). Based on historical aerial imagery from 1957 and 1978, it is possible to describe the area as a mosaic of secondary forests in different successional stages (Figures 1, 2 and S1), resulting from the abandonment of agricultural patches and pastures. A small portion of study area was enriched in the late 1970s with seedlings of Hyeronima alchorneoides, Miconia cinnamomifolia, and Nectandra spp. - typical species with economic value of secondary forests in the region. The natural regeneration of non-commercial species nearby the planted seedlings was slashed with a machete in the first years.

According to the Köppen classification, the climate in the study area is classified as Cfa - humid subtropical oceanic climate with hot summers, without dry seasons (Alvares et al., 2013Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G. Köppen’s climate classification map for Brazil. Meteorologische Zeitschrift 2013; 22:711-728.). The average annual temperature varies between 20 and 22 °C, and the average annual rainfall between 1,600 and 1,900 mm; average monthly absolute minimum temperatures may reach -3 °C in the coldest months (Wrege et al., 2012Wrege MS, Steinmetz S, Reisser Júnior C, Almeida IR. Atlas climático da região Sul do Brasil: estados do Paraná, Santa Catarina e Rio Grande do Sul. Brasília: Embrapa; 2021.). According to the Brazilian Soil Classification System, two soil types are predominant in the area: Cambissolo Háplico Tb A moderado and Neossolo Litólico A moderado (Embrapa, 2004Embrapa. Solos do Estado de Santa Catarina. Rio de Janeiro: Embrapa; 2004.). The slope varies between 10 to 40%, and the elevation above sea level between 205 to 426 m.

2.2. Forest inventory data

Forest inventory data were gathered on 21 sample plots, each with an area of 1,600 m² (40 m × 40 m), plus a 40 m × 30 m sample plot, totaling 3.49 ha of sampled area (Figure 1). All living and standing dead trees, palms and tree ferns with diameter at breast height (dbh) ≥ 5 cm on the sample plots had their dbh measured. Species identification was mostly made in the field by trained foresters; taxa not identified in the field were collected and later identified by consulting the Dr. Roberto Miguel Klein Herbarium (FURB) and experts. Species names were verified with the ‘flora’ R package (Carvalho, 2021Carvalho G. flora: Tools for Interacting with the Brazilian Flora 2020. R package version 0.3.5. [cited 2021 out. 26]. Available from: Available from: http://www.github.com/gustavobio/flora
http://www.github.com/gustavobio/flora... ), which consults the Flora do Brasil 2020 (2021Flora do Brasil 2020 (2021) Jardim Botânico do Rio de Janeiro. [cited 2021 out. 14]. Available from: Available from: http://floradobrasil.jbrj.gov.br .
http://floradobrasil.jbrj.gov.br... ) database and checks for misspellings or no longer valid names. The sample plot elevation above sea level ranges from 242 to 359 m; it increases from the base of the hillside towards the north direction (Figure 1).

2.3. Data analysis

The aboveground biomass (AGB) for each stem was predicted using the allometric model constructed by Uller et al. (2019Uller HF, Oliveira LZ, Vibrans AC, Klitzke AR, Eleotério JR. Aboveground biomass quantification and tree-level prediction models for the Brazilian subtropical Atlantic Forest. Southern Forests 2019; 81:261-271.) based on 105 individuals harvested in the study area (Eq. 1):

where $\stackrel{⏜}{AGB}$ is the predicted dry aboveground biomass (kg), dbh is the diameter at breast height (cm), h is the total height (m), and ρ is the wood specific gravity (kg m^-3).

The total height for each stem was predicted through a Korf nonlinear model (Flewelling and Jong 1994Flewelling JW, Jong RD. Considerations in simultaneous curve fitting for repeated height-diameter measurements. Canadian Journal of Forest Research 1994; 24:1408-1414.) constructed with data from 420 trees that had both their dbh and h measured in the study area: $\stackrel{⏜}{h}=1.30+7747·exp\left(-4.51·db{h}^{-0.34}\right)$; the model has a pseudo-R² of 0.75 and a root mean square error of 2.46 m. The ρ for each species record in the dataset was retrieved from Oliveira et al. (2019bOliveira LZ, Uller HF, Klitzke AR, Eleotério JR, Vibrans AC. Towards the fulfillment of a knowledge gap: Wood densities for species of the subtropical Atlantic Forest. Data 2019b; 4:104.).

Species-specific double-entry (dbh and h) models constructed by Uller et al. (2021Uller HF, Oliveira LZ, Klitzke AR, Freitas JV, Vibrans AC. Biomass models for three species with different growth forms and geographic distribution in the Brazilian Atlantic Forest. Canadian Journal of Forest Research 2021; 51:1-13.) were used to predict the AGB for Cecropia glaziovii, Cyathea spp. and Euterpe edulis individuals, as well as the single-entry (h) model constructed by Moreira-Burger and Delitti (2010) for Syagrus romanzoffiana. These models were used because Uller et al. (2021Uller HF, Oliveira LZ, Klitzke AR, Freitas JV, Vibrans AC. Biomass models for three species with different growth forms and geographic distribution in the Brazilian Atlantic Forest. Canadian Journal of Forest Research 2021; 51:1-13.) showed that generic models may yield predictions with strong systematic deviances for species with different growth forms and/or with particular morphological features. The total height of individuals of these species were predicted using species-specific nonlinear models (Table S1).

The total volume (VOL; stem plus branches with diameter ≥ 5 cm) for each stem was predicted through an updated version of the model constructed by Oliveira et al. (2018Oliveira LZ, Klitzke AR, Fantini AC, Uller HF, Correia J, Vibrans AC. Robust volumetric models for supporting the management of secondary forest stands in the Southern Brazilian Atlantic Forest. Anais da Academia Brasileira de Ciências 2018; 90:3729-3744. ) based on 419 trees harvested in the study area (Eq. 2):

where $\stackrel{⏜}{VOL}$ is the predicted total volume (m³), dbh is the diameter at breast height (cm), and h is the total height (m). The total height of each stem was predicted by the model mentioned above.

Means of tree density (ind. ha^-1), basal area (m² ha^-1), AGB (Mg ha^-1), and growing stock volume (VOL; m³ ha^-1) were calculated based on plot level predictions scaled to hectare. Non-woody individuals (i.e., tree ferns and palm trees) and C. glaziovii individuals were excluded from the calculation of mean VOL ha^-1. Means were computed independently considering individuals with dbh ≥ 5 cm and individuals with dbh ≥ 10 cm to allow for comparisons with other studies. It should be mentioned that uncertainty in tree-level model predictions of AGB and VOL was not accounted for in estimates per unit area.

Phytosociological parameters were predicted for each species as per Mueller-Dombois & Ellenberg (2002Mueller-Dombois D, Ellenberg H. Aims and methods of vegetation ecology. New Jersey: The Blackburn Press; 2002.), namely, absolute density (AD), relative density (RD), absolute dominance (ADo), relative dominance (RDo), absolute frequency (AFq), and relative frequency (RFq). Additionally, the absolute AGB (AAGB) and relative AGB (RAGB) were predicted. The percentage of importance (PI) was calculated as the mean over RD, RDo, RAGB, and RFq. These parameters were predicted considering individuals with dbh ≥ 5 cm.

Based on the observed abundance of individuals of each species, three measures of Hill numbers of order q were estimated with the aid of the ‘iNEXT’ R package (Hsieh et al., 2020Hsieh TC, Ma KH, Chao A. iNEXT: iNterpolation and EXTrapolation for species diversity. R package version 2.0.20. [cited 2021 out. 14]. Available from: Available from: http://chao.stat.nthu.edu.tw/wordpress/software-download/ .
http://chao.stat.nthu.edu.tw/wordpress/s... ): species richness (q = 0), Shannon diversity (q = 1, exponential of Shannon entropy), and Simpson diversity (q = 2, inverse of Simpson concentration). Rarefaction and extrapolation curves for the Hill numbers were also constructed using the referred package. For the latter curves, a two-fold extrapolation of the reference sample was carried out, as suggested by Chao et al. (2014Chao A, Gotelli NJ, Hsieh TC, Sander EL, Ma KH, Colwell RK et al. Rarefaction and extrapolation with Hill numbers: a framework for sampling and estimation in species diversity studies. Ecological Monographs 2014; 84:45-67.). In addition, using the ‘endpoint’ argument of the ‘iNEXT’ function, Hill numbers of order q = 0, q = 1 and q = 2 were estimated considering 1724 individuals (dbh ≥ 5 cm), which is the estimated mean tree density per hectare for the study area (see below). The estimates were compared to diversity estimates reported by studies in the region that measured one hectare of forest. This analysis provides approximate results because uncertainty in the estimation of mean tree density was not accounted for.

3. RESULTS

3.1. Means of forest attributes per unit area

A mean tree density of 1724 ind. ha^-1 (standard deviation [SD] = 364), mean basal area of 32.2 m² ha^-1 (SD = 4.9), mean AGB of 201.6 Mg ha^-1 (SD = 35.5), and mean VOL of 320.7 m³ ha^-1 (SD = 77.3) were found, considering all living individuals with dbh ≥ 5 cm. Considering all living individuals with dbh ≥ 10 cm, a mean tree density of 773 ind. ha^-1 (SD = 151), mean basal area of 27.8 m² ha^-1 (SD = 4.6), mean AGB of 187.8 Mg ha^-1 (SD = 35.2), and mean VOL of 306.8 m³ ha^-1 (SD = 78.1) were found. Medians and quartiles of sample plot predictions are shown in Figure 3.

3.2. Forest structure

H. alchorneoides, Cyathea delgadii, Euterpe edulis, M. cinnamomifolia, and Clusia criuva were the five species with the greatest percentage of importance (PI) (Table 1). The first three species were found on all sample plots, whereas M. cinnamomifolia was found on 82% of the sample plots. C. criuva was abundant in the less developed parts of the community, occurring on about one-third of the sample plots (Table 1). H. alchorneoides and M. cinnamomifolia represented 21.2% and 6.1% of the total basal area ha^-1, respectively. The two species together represented 36.7% of the total AGB ha^-1 (Table 1). Although Nectandra membranacea − a typical species of subtropical secondary evergreen rainforests − was not among the species with the greatest absolute density, it was observed on every sample plot. The understory shade-tolerant species Psychotria nuda and Psychotria suterella were observed on almost every sample plot, and together represented 7.8% of the total tree density ha^-1. Notably, the climax species Virola bicuhyba and Cabralea canjerana were amidst the 10 species with the largest PI; the former was the fourth tree species with the greatest AGB ha^-1 (Table 1). Phytosociological parameters for all species are found in Supplementary Material (Table S2).

The stem dbh distribution of the entire community followed a negative exponential distribution (Figure 4). M. cinnamomifolia had a unimodal stem dbh distribution, where small individuals (dbh < 10 cm) were scarce. In turn, H. alchorneoides had similar numbers of stems per dbh class up to the intermediate classes, with a major drop in number of stems in the subsequent classes.

3.3. Species diversity

Among all observed native taxa, 153 were identified at the species level, distributed in 107 genera and 55 families. The families Fabaceae (19 spp.), Myrtaceae (17), Lauraceae (11), Rubiaceae (11) and Euphorbiaceae (7) were the most speciose and represented 42.5% of all taxa identified at the species level.

It was estimated that 121 species (95% confidence interval [CI]: [117, 125]) occurred in one hectare (Figure 5). The exponential of the Shannon diversity, exp(H’), was estimated to be 37 species per hectare (95% CI: [36, 38]), whereas the inverse of Simpson concentration, 1/D, was estimated to be 18 species per hectare (95% CI: [17, 19]) (Figure 5). This means that the community has the same diversity as a community composed of 37 or 18 equally-common species per hectare depending on the metric, which gives different weights to lesser common species (Jost, 2006Jost L. Entropy and diversity. Oikos 2006; 113:363-375.).

It was estimated that 163 species (95% CI: [152, 175]) would be observed if the reference sample would be extrapolated two-fold. Therefore, on average, 10 new species would be observed if the sample size would be doubled. The exp(H’) was estimated to be 39 species (95% CI: [37, 40]) based on the extrapolated sample, whereas the 1/D was estimated to 18 species (95% CI: [17, 19]) (Figure 5).

4. DISCUSSION

4.1. Forest attributes

The mean tree density found in this study was similar to that reported by Maçaneiro et al. (2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ) for a mature evergreen rainforest (ERF) remnant in Santa Catarina, where selective logging of Ocotea catharinensis (‘canela-preta’) and other species with economic value was conducted ~85 years ago. On the other hand, they found a greater basal area (40.4 m² ha^-1) than us (32.2 m² ha^-1), a result that could be attributed to the degree of development of the community they studied, in which the climax species Sloanea guianensis, Euterpe edulis, Tapirira guianensis, Ocotea aciphylla and Vantanea compacta together represented one-third of total basal area (Maçaneiro et al., 2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ). Likewise, Guislon (2017Guislon AV. Comunidade arbórea e histórico de ocupação humana em uma Floresta Ombrófila Densa Montana no Sul do Brasil [dissertação]. Criciúma: Universidade do Extremo Sul Catarinense; 2017.) found a basal area of 60.4 m² ha^-1 for a mature ERF within a protected area in Santa Catarina, where O. catharinensis and senescent Alchornea triplinervia individuals represented nearly half of the total basal area.

Vibrans et al. (2022Vibrans AC, Oliveira LZ, Gasper AL, Lingner DV, Schorn LA, Silva DA. Unprecedented large-area turnover estimates for the subtropical Brazilian Atlantic Forest based on systematically-gathered data. Forest Ecology and Management 2022; 505:119902.), based on data (dbh ≥ 10 cm) gathered by the Forest and Floristic Inventory of Santa Catarina (IFFSC), reported a mean basal area of 22.3 m² ha^-1 (95% CI: [21.3, 23.3]) for the ERF in the state, which is a substantially smaller value than that we found (27.8 m² ha^-1). Using the IFFSC data and the same allometric model employed in this study, a mean AGB of 117.6 Mg ha^-1 (95% CI: [109.7, 125.5]) was calculated for the same forest type (unpublished data). Vibrans et al. (2020Vibrans AC, Gasper AL, Moser P, Oliveira LZ, Lingner D, Sevegnani L. Insights from a large-scale systematic inventory in the southern Brazilian Atlantic Forest. Scientia Agricola 2020; 77:e20180036.) asserted that a relevant portion of ERF remnants in Santa Catarina consists of stands with reduced α-diversity and biomass stocks. In fact, the community we studied appears to be in a later succession stage than the average secondary ERF in the state. Although the mean AGB ha^-1 we found was larger than the 88^th percentile of AGB ha^-1 predictions for the IFFSC sample plots in the ERF, Vibrans et al. (2022Vibrans AC, Oliveira LZ, Gasper AL, Lingner DV, Schorn LA, Silva DA. Unprecedented large-area turnover estimates for the subtropical Brazilian Atlantic Forest based on systematically-gathered data. Forest Ecology and Management 2022; 505:119902.) highlighted the existence of mature ERF remnants with AGB stocks greater than 300 Mg ha^-1 in northern Santa Catarina. Therefore, about two-thirds of the AGB accumulation potential of the remnant we studied has been actualized.

The expressive growing stock volume we found merits comment. The management of secondary forests has been highlighted as a win-win activity that would generate income for landowners and promote forest conservation (Fantini & Siminski, 2016Fantini AC, Siminski A. Manejo de florestas secundárias da Mata Atlântica para produção de madeira: possível e desejável. Revista Brasileira de Pós-Graduação 2016; 13:673-698.; Zambiazi et al., 2021Zambiazi DC, Fantini AC, Piotto D, Siminski A, Vibrans AC, Oller DC et al. Timber stock recovery in a chronosequence of secondary forests in Southern Brazil: Adding value to restored landscapes. Forest Ecology and Management 2021; 495:119352.). According to our data, H. alchorneoides and M. cinnamomifolia have, on average, 83.0 m³ ha^-1 and 23.2 m³ ha^-1 of total timber volume in the study area, respectively. Hence, these species have great potential for being managed in secondary forests due to their increased abundance and stocks, good quality wood, among other desirable features (Schuch et al., 2008Schuch C, Siminski A, Fantini AC. Uso e potencial madeireiro do jacatirão-açu (Miconia cinnamomifolia (de Candolle) Naudin) no litoral de Santa Catarina. Floresta 2008; 38:735-741., 2011Schuch C, Alexandre S, Fantini AC. Hieronyma alchorneoides Licurana. In: Coradin L, Siminski A, Reis A, editors. Espécies nativas da flora brasileira de valor econômico atual ou potencial plantas para o futuro - Região Sul. Brasília: Ministério do Meio Ambiente; 2011.; Zambiazi et al., 2021Zambiazi DC, Fantini AC, Piotto D, Siminski A, Vibrans AC, Oller DC et al. Timber stock recovery in a chronosequence of secondary forests in Southern Brazil: Adding value to restored landscapes. Forest Ecology and Management 2021; 495:119352.).

4.2. Forest structure and composition

Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389., 1984Klein RM. Aspectos dinâmicos da vegetação do Sul do Brasil. Sellowia 1984; 36:5-54. ) discussed several possible successional trajectories of the ERF in Santa Catarina, highlighting M. cinnamomifolia as one of the most relevant light-demanding species in early- and mid-successional subtropical ERF stands, often being the most abundant species. Given its requirement for open areas to successfully regenerate and its relatively short life cycle, this species is progressively replaced by H. alchorneoides, Alchornea triplinervia, among others (Klein, 1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.). Accordingly, the distribution of stem dbh we found for M. cinnamomifolia suggests that it is being replaced by shade-tolerant species, as small individuals (dbh < 10 cm) were found to be scarce (Figure 4). This species had the seventh largest RD, but still had the second largest RDo (Table 1), indicating the presence of adult/senescent individuals. In fact, Schorn and Galvão (2009Schorn LA, Galvão F. Dinâmica do estrato arbóreo em três estádios sucessionais de uma Floresta Ombrófila Densa em Blumenau, SC. Cerne 2009; 15:221-235.) showed that the dynamics of mid-successional subtropical ERFs in Santa Catarina is accelerated by the death of M. cinnamomifolia individuals.

Notedly, H. alchorneoides was the most representative species in terms of AGB, alone accounting to nearly one-third of the community stock. Although the enrichment with saplings of this species carried out in the late 1970s may have contributed to its current dominance, it was mentioned by Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.) as one of the most important species in late-successional ERFs, and as one of the shade-tolerant species that benefits from the environmental conditions shaped by mature M. cinnamomifolia individuals. H. alchorneoides, in turn, prepares the environment for S. guianensis, which, along with E. edulis and other shade-tolerant species, prepares the environment for more selective Lauraceae species, such as Ocotea catharinensis, Ocotea odorifera, and Cryptocarya spp. (Klein, 1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.; Maçaneiro et al., 2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ; Brotto et al., 2019Brotto ML, Lozano ED, Marinero FEC, Uhlmann A, Blum CT, Roderjan CV. Lauraceae along an altitudinal gradient in southern Brazil. Floresta e Ambiente 2019; 26:e20170637. ). Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.) observed that at this late successional stage − that is, when H. alchorneoides dominates and S. guianensis begins to appear − several Myrtaceae species join the community, a phenomenon we detected in the study area; this family had the second greatest number of species. However, the number of Myrtaceae species observed by Maçaneiro et al. (2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ) in 1 ha of mature subtropical ERF is greater than the number we observed in 3.49 ha by a factor of 1.6. Similarly, Caglioni et al. (2015Caglioni E, Curcio GR, Uhlmann A, Bonnet A. Estrutura e diversidade do componente arbóreo de Floresta Atlântica no parque nacional da serra do Itajaí, Santa Catarina. Floresta 2015; 45:289-302.) found more Myrtaceae species in 0.33 ha of subtropical ERF within a protected area than we found. Indeed, this family can be regarded as an indicator of overall species diversity and biological conservation value (Lucas & Bünger, 2015Lucas EJ, Bünger M. Myrtaceae in the Atlantic forest: their role as a ‘model’ group. Biodiversity and Conservation 2015; 24:2165-2180.).

Maçaneiro et al. (2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ) did not find an expressive density of H. alchorneoides (12 ind. ha^-1), but rather a great density of S. guianensis (107 ind. ha^-1), E. edulis (442 ind. ha^-1), and Garcinia gardneriana (66 ind. ha^-1) associated with climax Lauraceae species, which together represented 19% of the community total basal area ha^-1. Likewise, Caglioni et al. (2015Caglioni E, Curcio GR, Uhlmann A, Bonnet A. Estrutura e diversidade do componente arbóreo de Floresta Atlântica no parque nacional da serra do Itajaí, Santa Catarina. Floresta 2015; 45:289-302.) found an increased density of S. guianensis associated with E. edulis, but only 15 ind. ha^-1 of H. alchorneoides. Lauraceae species represented only 5.6% of the total basal area ha^-1 of the community we studied, and Nectandra membranacea − a typical soft-wood species of secondary subtropical ERFs − was the most representative (RDo = 2.9%). Overall, these evidences suggest that the community we studied is less developed relative to the communities studied by Veloso & Klein (1968), Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.), Caglioni et al. (2015Caglioni E, Curcio GR, Uhlmann A, Bonnet A. Estrutura e diversidade do componente arbóreo de Floresta Atlântica no parque nacional da serra do Itajaí, Santa Catarina. Floresta 2015; 45:289-302.), Maçaneiro et al. (2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ), and Guislon (2017Guislon AV. Comunidade arbórea e histórico de ocupação humana em uma Floresta Ombrófila Densa Montana no Sul do Brasil [dissertação]. Criciúma: Universidade do Extremo Sul Catarinense; 2017.). According to our data, it was predicted that O. catharinensis − one of the most important climax species in the region (Veloso & Klein, 1968) − has only 2 ind. ha^-1 and represents merely 0.4% of the total AGB ha^-1. Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.) stated that O. catharinensis would often represent up to one-third or even half of the growing stock volume of pristine subtropical ERF stands. Accordingly, Brotto et al. (2019Brotto ML, Lozano ED, Marinero FEC, Uhlmann A, Blum CT, Roderjan CV. Lauraceae along an altitudinal gradient in southern Brazil. Floresta e Ambiente 2019; 26:e20170637. ) found that this species represents ~25% of the total stem volume of a montane ERF remnant within a protected area in the state of Paraná, southern Brazil.

We found that E. edulis, one of the most important species composing the intermediate layer of the subtropical ERF (Klein, 1978Klein RM. Mapa fitogeográfico do Estado de Santa Catarina. In: Reitz R, editor. Flora ilustrada Catarinense. Itajaí: Herbário Barbosa Rodrigues; 1978., 1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.), had a smaller density (222 ind. ha^-1) than those found in other studies (330 to 645 ind. ha^-1; Colonetti et al., 2009Colonetti S, Citadini-Zanette V, Martins R, Santos R, Rocha E, Jarenkow JA. Florística e estrutura fitossociológica em floresta ombrófila densa submontana na barragem do rio São Bento, Siderópolis, Estado de Santa Catarina. Acta Scientiarum 2009; 31:397-405.; Bosa et al., 2015Bosa DM, Pacheco D, Pasetto MR, Santos R. Florística e estrutura do componente arbóreo de uma Floresta Ombrófila Densa Montana em Santa Catarina, Brasil. Revista Árvore 2015; 39:49-58.; Maçaneiro et al., 2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ; Melo Júnior et al., 2017Melo Júnior JCF, Amorim MW, Arriol IA, Canuto KK, Pereira LGS. Flora vascular, estrutura comunitária e conservação de fragmentos da floresta atlântica na Bacia Hidrográfica do Rio Cachoeira, Joinville, SC, Brasil. Acta Biológica Catarinense 2017; 4:41-72.). The smaller density we found is related to historical and recent extractivism without permission of the landowner, given that E. edulis usually occurs abundantly in humid environments over flat or moderately sloped terrains, such as those found in our study area. The understory species Psychotria nuda and Psychotria suterella, which are abundant in mature ERF stands dominated by O. catharinensis, S. guianensis and E. edulis, may also be abundant in late-successional forests, as pointed out by Klein (1980Klein RM. Ecologia da flora e vegetação do Vale do Itajaí. Sellowia 1980; 32:165-389.). The author argued that their short life cycle and fast regeneration dynamics, in addition to more stable microclimatic conditions within the forest, explain why the understory tends to evolve towards an association of shade-tolerant species in a faster pace relative to the canopy layer.

4.3. Species diversity

Our estimates of species richness and exp(H’) per hectare (121 and 37 spp., respectively) are within the expected for late-successional subtropical ERFs; according to the meta-analysis conducted by Maçaneiro et al. (2016Maçaneiro JP, Oliveira LZ, Eisenlohr PV, Schorn LA. Paradox between species diversity and conservation: a subtropical Atlantic Forest reserve in Brazil has similar tree species diversity to unprotected sites in the same region. Tropical Conservation Science 2016; https://doi.org/10.1177/1940082916668011
https://doi.org/10.1177/1940082916668011... ) based on 12 studies carried out in the ERF of Santa Catarina, the former measure varies between 106 and 149 species and the latter between 11 to 64 species. The community we studied has a quite similar exp(H’) than that found by Maçaneiro et al. (2015Maçaneiro JP, Seubert RC, Schorn LA. Fitossociologia de uma floresta pluvial subtropical primária no Sul do Brasil. Floresta 2015; 45:555-566. ) for a mature ERF in Santa Catarina; in terms of species richness though, the community studied by these authors had nearly 20% more species per hectare than we found.

As stressed by Maçaneiro et al. (2016Maçaneiro JP, Oliveira LZ, Eisenlohr PV, Schorn LA. Paradox between species diversity and conservation: a subtropical Atlantic Forest reserve in Brazil has similar tree species diversity to unprotected sites in the same region. Tropical Conservation Science 2016; https://doi.org/10.1177/1940082916668011
https://doi.org/10.1177/1940082916668011... ) and Oliveira & Vibrans (2020Oliveira LZ, Vibrans AC. An approach to illustrate the naturalness of the Brazilian Araucaria forest. Canadian Journal of Forest Research 2020; 50:32-41. ), the abundance of dominant species often found in old-growth subtropical Atlantic forests is negatively correlated with diversity metrics (e.g., Shannon and Simpson diversity metrics). Therefore, late-successional forests may have greater species diversity than old-growth forests, although the structural complexity, overall plant richness and AGB stocks of the latter are usually greater (Liebsch et al., 2008Liebsch D, Marques MC, Goldenberg R. How long does the Atlantic Rain Forest take to recover after a disturbance? Changes in species composition and ecological features during secondary succession. Biological Conservation 2008; 141:1717-1725.; Oliveira & Vibrans, 2020Oliveira LZ, Vibrans AC. An approach to illustrate the naturalness of the Brazilian Araucaria forest. Canadian Journal of Forest Research 2020; 50:32-41. ; Siminski et al., 2021Siminski A, Zambiazi DC, Santos KL, Fantini AC. Dynamics of Natural Regeneration: Implications for Landscape Restoration in the Atlantic Forest, Brazil. Frontiers in Forests and Global Change 2021; 5:1-15.).

5. CONCLUSIONS

Four conclusions may be drawn from this study. First, the studied subtropical ERF remnant can generally be described as a late-successional forest, given the dominance of the shade-tolerant species Hyeronima alchorneoides and the small populations of typical climax Lauraceae species. Second, the remnant has a substantially larger AGB stock than the average for the ERF in Santa Catarina, but still has potential of further accumulation, especially with the recovery of populations of climax species. Third, H. alchorneoides and Miconia cinnamomifolia accumulated more than one-third of the community AGB ha^-1 − an indication of their potential for enrichment plantings aimed to enhance forest productivity. Because of their desirable silvicultural features and market acceptance, these two species represent valuable resources for forest management in the region. Fourth, species richness in the study area is within the expected for secondary forests in the region − that is, more than 100 arborescent species per hectare −, and therefore corroborates the potential of such forests for biodiversity recovery, which is simultaneously achieved with carbon sequestration and other ecosystem services.

REFERENCES

SUPPLEMENTARY MATERIAL

The following online material is available for this article:

Figure S1. Historical aerial imagery from 1957 and 1978 for the study area. Scale: 1:4,500. Datum: WGS-84, UTM Zone 22S.

Table S1. Species-specific nonlinear models to predict total height (ĥ, in meters). The models were constructed using the data collected by the following authors (cited in the manuscript) - Uller et al. (2021) and Moreira-Burger and Delitti (2010). dbh = diameter at breast height (cm).

Table S2. Phytosociological parameters per species. AD = absolute density (ind. ha⁻¹); RD = relative density (%), ADo = absolute dominance (m² ha⁻¹); RDo = relative dominance (%); AAGB = absolute aboveground biomass (Mg ha⁻¹); RAGB = relative aboveground biomass (%); AFq = absolute frequency (%); RFq = relative frequency (%); PI = percentage of importance (%), that is the mean over RD, RDo, RAGB, and RFq.

Edited by

Publication Dates

History