Environmental factors affect population structure of tree ferns in the Brazilian subtropical Atlantic Forest

ABSTRACT

Tree ferns are important elements of tropical forests, mainly because they are common and provide microhabitats for epiphytic plants. Due to their ecological importance, the aim of this study was to evaluate population structure, distribution, and influence of environmental variables on tree ferns in the state of Santa Catarina, southern Brazil. All tree ferns with a diameter at breast height ≥ 10 cm on 418 sampling units (SUs) systematically distributed throughout the study area were measured (total sampled area of ​​153.4 ha). Population structure was evaluated through classical phytosociological parameters and the relationships among dominance and environmental variables were evaluated through multiple linear regression models. Dicksonia sellowiana presented the greatest importance value among all species (IV = 13.19 %), followed by Alsophila setosa (IV = 4.37 %) and Cyathea phalerata (IV = 2.71 %). Altitude and mean rainfall of the driest quarter were significantly related to the dominance of D. sellowiana in most of the SUs. The mean temperature of the driest quarter and aspect were significantly related to the dominance of Cyatheaceae. Our study demonstrates that tree ferns are important elements of forest communities in the state of Santa Catarina.

Keywords:
Cyatheaceae; Dicksoniaceae; environmental variables; population structure; regression analysis; tree ferns

Introduction

Tree ferns are important elements of diverse plant formations, especially of tropical forests (Tryon & Tryon 1982). Ferns affect the dynamics of the ecosystem in which they occur, influencing the regeneration of woody species and nutrients cycling (Brock et al. 2016); they also participate in the ecological succession process (Arens & Baracaldo 1998). Furthermore, they contribute to forest biomass stocks, accounting for more than 6 % of the total aboveground biomass in sites with a great density of individuals (Medeiros & Aidar 2011). Tree ferns are also important because many epiphytic plants (e.g., Asplenium mucronatum, Pecluma truncorum, and Trichomanes anadromum) use their caudices as exclusive support (Moran et al. 2003; Schmitt & Windisch 2005; 2010; Fraga et al. 2008).

In the neotropics, most tree ferns species belong to Cyatheaceae or Dicksoniaceae (Pteridophyte Phylogeny Group 1 2016). Cyatheaceae is the richest family, with approximately 640 species worldwide (Pteridophyte Phylogeny Group 1 2016) and 43 species occurring in Brazil (Weigand & Lehnert 2016). This family is distributed over tropical and temperate areas in America, Australia, New Zealand, and Malaysia (Large & Braggins 2004). Dicksoniaceae has 35 species worldwide (Pteridophyte Phylogeny Group 1 2016) and two species in Brazil (Della & Vasques 2017). It is distributed in tropical and southern temperate regions (Large & Braggins 2004).

In the state of Santa Catarina, southern Brazil, Cyatheaceae is the second family with the largest number of individuals in the subtropical evergreen rainforest. Two species of this family, namely Alsophila setosa and Cyathea phalerata, are among the ten most abundant arborescent species in this forest type (Lingner et al. 2013). In turn, the most abundant species in the Araucaria forest is Dicksonia sellowiana, followed by Araucaria angustifolia (Meyer et al. 2013). According to Bystriakova et al. (2011), Cyatheaceae species prefer hot and humid sites. Dicksoniaceae species, in turn, prefer sites with higher altitude and lower temperatures, and tolerate frosts (Mantovani 2004; Gasper et al. 2011).

In the last decades, tree ferns were intensively exploited due to their fibers and ornamental value, causing the depletion of natural populations (Windisch 2002; Santiago et al. 2013). Continuous exploitation can lead to the elimination of these individuals in nature due to their slow growth, as well as the loss of specific epiphytes occurring on them. Although, the ecological importance of tree ferns has been highlighted in the literature (Arens & Baracaldo 1998; Moran et al. 2003; Schmitt & Windisch 2005; 2010; Fraga et al. 2008; Medeiros & Aidar 2011), studies about this plant group are still scarce. In Brazil, most studies regarding tree ferns were conducted in the Atlantic Forest domain, focusing mainly on phenology (Schmitt & Windisch 2005; 2007; Schmitt et al. 2009; Neumann et al. 2014) and population structure (Mantovani 2004; Schmitt & Windisch 2007; Gasper et al. 2011). Relationships among environmental variables and the distribution and population structure of ferns were investigated by Tuomisto et al. (2019) in the Amazon; notwithstanding, the authors did not focus exclusively on tree ferns, but rather on all types of ferns. This demonstrates that there is a lack of information about the influence of environmental factors on tree fern populations.

Therefore, we analyzed population structure, distribution, and influence of environmental variables on tree ferns, aiming to answer the following questions: (1) Are tree ferns a relevant element of subtropical Atlantic Forest communities? (2) Which environmental variables are significantly related to the dominance of tree fern species? According to previous studies, tree ferns occur abundantly in the Atlantic Forest (Gasper et al. 2011; Lingner et al. 2013; Meyer et al. 2013) and their occurrence is influenced by temperature and rainfall-related variables (Mantovani 2004; Bystriakova et al. 2011). Hence, we expect to find similar results in the present study.

Materials and methods

Study area

The study area was defined as the state of Santa Catarina State (Fig. 1), which has ~29 % of its territory covered by native forests (Vibrans et al. 2013). According to the Köppen-Geiger climate classification, two climate types can be found in the state: Cfa - fully humid temperate climate with a hot summer, and Cfb - fully humid temperate climate with a warm summer (Alvares et al. 2013). In the evergreen rainforest, high temperatures, humidity and precipitation are found, and biologically dry periods are absent. The Araucaria forest has moderately hot summers and a long winter period (Leite 2002). Precipitation is abundant in the western region of the state, reaching over 2,000 mm.year^-1 (Wrege et al. 2012). The semi-deciduous forest, located in the west of Santa Catarina, is characterized by temperature seasonality (Leite 2002).

Figure 1
Study area (state of Santa Catarina, southern Brazil) and sampling units of the IFFSC.

Data collection

We obtained tree fern community data from the Forest and Floristic Inventory of Santa Catarina (IFFSC; see more details in Vibrans et al. 2010). Between 2007 and 2010, the IFFSC gathered data using a systematic sampling design; each sampling unit (SU) was located at the intersections of a 10 km × 10 km grid. The SF required a 5 km × 5 km grid to guarantee representativeness (Fig. 1). Each SU was composed of a cluster with a nominal area of ​​4,000 m², where the diameter at breast height (DBH) ≥ 10 cm and total height of each individual were measured.

We selected eight tree ferns species: Alsophila setosa Kaulf, Cyathea atrovirens (Langsd. & Fisch.) Domin, Cyathea corcovadensis (Raddi) Domin, Cyathea delgadii Sternb., Cyathea phalerata Mart., Cyathea hirsuta C.Presl, Sphaeropteris gardneri Hook., and Dicksonia sellowiana Hook. The climatic data were obtained from WorldClim v2.0 (Fick & Hijmans 2017). In addition, altitude, slope and aspect data were obtained from the Brazilian Geomorphometric Database (INPE 2017). In total, 24 environmental variables with approximately 1 km of spatial resolution were used (Tab. 1).

Table 1
Environmental variables used in the Regression Analyses for Santa Catarina State, southern Brazil. Source: 1: Worldclim; 2: INPE.

Code	Description	Unit	Code	Description	Unit
Bio1	Annual Mean Temperature1	ºC	Bio13	Precipitation of Wettest Month1	mm
Bio2	Mean Diurnal Range1	ºC	Bio14	Precipitation of Driest Month1	mm
Bio3	Isothermality1	%	Bio15	Precipitation Seasonality1	mm
Bio4	Temperature Seasonality1	ºC	Bio16	Precipitation of Wettest Quarter1	mm
Bio5	Max Temperature of Warmest Month1	ºC	Bio17	Precipitation of Driest Quarter1	mm
Bio6	Min Temperature of Coldest Month1	ºC	Bio18	Precipitation of Warmest Quarter1	mm
Bio7	Temperature Annual Range1	ºC	Bio19	Precipitation of Coldest Quarter1	mm
Bio8	Mean Temperature of Wettest Quarter1	ºC	Rad_win	Solar radiation in winter1	KJ.m-².day-¹
Bio9	Mean Temperature of Driest Quarter1	ºC	Rad_sum	Solar radiation in summer1	KJ.m-².day-¹
Bio10	Mean Temperature of Warmest Quarter1	ºC	-	Altitude2	m
Bio11	Mean Temperature of Coldest Quarter1	ºC	-	Slope2	%
Bio12	Annual Precipitation1	mm	-	Aspect2	%

Data analysis

To assess the importance of tree ferns in the forest types of Santa Catarina, phytosociological parameters were calculated (dominance, density, frequency and importance value; Müeller-Dombois & Ellenberg 1974). Although the parameters for all the species occurring on the SUs were computed, only parameters related to tree fern species were considered in this study. We also assigned the individuals to height and diameter classes.

The influence of environmental variables on the tree ferns was analyzed using multiple linear regression models, using dominance (m².ha^-1) as the response variable and environmental variables as predictor variables. GWR (Geographically Weighted Regression) and OLS (Ordinary Least Squares) models were fitted and compared. Collinear variables with variance inflation factor > 10 were removed from the models using the ‘vif’ function of the ‘usdm’ R package (Naimi et al. 2014). Models with the smallest Corrected Akaike Information Criterion (AICc) (Burnham & Anderson 2002) were selected. In addition, the global significance of the models was evaluated (α = 0.05), and F tests were performed to verify whether there was a significant improvement of the GWR residuals over the OLS residuals (Fotheringham et al. 2002). The regression models were fitted only for species that occurred at least on 30 SUs. Cyathea atrovirens, Cyathea hirsuta and Sphaeropteris gardneri were disregarded.

Moran’s I correlograms were built for the OLS model residuals aiming to search for significant spatial structure in them, which would suggest that spatial variables could be useful predictors. Default parameters were used to select the number of distance classes in the correlogram. A significance test for each distance class was performed using 999 Monte Carlo permutations. The global significance of the correlograms was tested using α = 0.05 corrected using Bonferroni’s approach (Fortin & Dale 2005). The residuals of the OLS models did not present any significant spatial structure according to this procedure. The normality of the residuals was tested using the D’Agostino-Pearson test (Zar 1999). The assumptions of linearity, homoscedasticity and residuals’ independence were checked through residual plots (Hair et al. 2009). For the GWR models, the adaptive Gaussian kernel function for geographical weighting based on the minimization of AICc was applied (Fotheringham et al. 2002). These analyses were performed using SAM v4.0 (Rangel et al. 2010).

Moreover, we performed a variance partition analysis to assess how much of the variability of the dominance of tree ferns was explained by the (a) “pure” spatial structure; (b) spatially-structured environmental variables; and (c) “pure” environmental variables (Peres-Neto & Legendre 2010). The statistical significance of fractions (a) and (c) was tested using ANOVA models based on 999 permutations and α = 0.05 (Dray et al. 2006). The variance partition was conducted as per Clappe et al. (2018) using the ade4 R package (Dray & Dufour 2007; R Development Core Team 2011).

Results

Population structure

In total, 10,632 individuals (Cyatheaceae = 5,129; Dicksoniaceae = 5,503) were observed. Dicksonia sellowiana presented the greatest importance value among all species (IV = 13.19 %). This species also had the largest Relative Dominance (6.96 %) and the greatest Relative Density (5.59 %) (Tab. 2, Fig. 2).

Table 2
Phytosociological parameters of tree fern species in the IFFSC, considering all species sampled. N: number of individuals; SU: number of sampling units in which the x species occurs; AD: Absolute Density (n/ha); RD: Relative Density (%); AF: Absolute Frequency (%); RF: Relative Frequency (%); ADo: Absolute Dominance (m²/ha); RDo: Relative Dominance (%); VI: Value of importance; SR: Subtropical Rainforest; SMF: Subtropical Mixed Forest; SDF: Subtropical Deciduous Forest.

Region	Species	Ranking	N	SU	AD	RD	AF	RF	ADo	RDo	VI
All study area	Dicksonia sellowiana	1	5503	128	36.0079	5.5953	30.622	0.6416	1.7512	6.9607	13.1975
	Alsophila setosa	6	2890	120	18.9102	2.9385	28.7081	0.6015	0.2087	0.8297	4.3696
	Cyathea phalerata	21	1596	103	10.4431	1.6228	24.6411	0.5163	0.1443	0.5735	2.7126
	Cyathea delgadii	99	303	63	1.9826	0.3081	15.0718	0.3158	0.0305	0.1211	0.745
	Cyathea corcovadensis	106	310	55	2.0284	0.3152	13.1579	0.2757	0.0275	0.1094	0.7003
	Cyathea atrovirens	394	22	10	0.144	0.0224	2.3923	0.0501	0.0021	0.0082	0.0807
	Sphaeropteris gardneri	544	6	3	0.0393	0.0061	0.7177	0.015	0.0011	0.0043	0.0254
	Cyathea hirsuta	671	2	1	0.0131	0.002	0.2392	0.005	0.0001	0.0005	0.0076
	TOTAL	-	10632	-	69.5686	10.8104	115.5501	2.421	2.1655	8.6074	21.8387
SR	Alsophila setosa	2	2689	108	38.2673	5.2738	54.8223	0.9214	0.4239	1.6862	7.8814
	Cyathea phalerata	5	1576	101	22.4282	3.0909	51.269	0.8617	0.3092	1.23	5.1826
	Cyathea delgadii	53	301	62	4.2836	0.5903	31.4721	0.529	0.0652	0.2592	1.3785
	Dicksonia sellowiana	58	247	36	3.5151	0.4844	18.2741	0.3071	0.1075	0.4277	1.2192
	Cyathea corcovadensis	93	189	40	2.6897	0.3707	20.3046	0.3413	0.0374	0.1488	0.8608
	Cyathea atrovirens	289	21	10	0.2989	0.0412	5.0761	0.0853	0.0043	0.017	0.1435
	Sphaeropteris gardneri	434	6	3	0.0854	0.0118	1.5228	0.0256	0.0023	0.0093	0.0467
	Cyathea hirsuta	573	2	1	0.0285	0.0039	0.5076	0.0085	0.0003	0.0011	0.0136
	TOTAL	-	5031	361	71.5967	9.867	183.2486	3.0799	0.9501	3.7793	16.7263
SMF	Dicksonia sellowiana	1	5252	89	95.268	15.6138	62.2378	1.7027	4.7154	17.5955	34.912
	Alsophila setosa	69	167	8	3.0293	0.4965	5.5944	0.1531	0.0333	0.1242	0.7737
	Cyathea corcovadensis	113	70	11	1.2698	0.2081	7.6923	0.2104	0.0196	0.0732	0.4918
	Cyathea phalerata	214	18	3	0.3265	0.0535	2.0979	0.0574	0.0054	0.0201	0.131
	Cyathea delgadii	279	5	2	0.0907	0.0149	1.3986	0.0383	0.0022	0.0081	0.0612
	TOTAL	-	5512	113	99.9843	16.3868	79.021	2.1619	4.7759	17.8211	36.3697
SDF	Cyathea corcovadensis	88	51	4	1.8593	0.3716	5.1282	0.1328	0.018	0.0823	0.5867
	Alsophila setosa	105	34	4	1.2395	0.2477	5.1282	0.1328	0.0119	0.0543	0.4349
	Dicksonia sellowiana	152	4	3	0.1458	0.0291	3.8462	0.0996	0.0047	0.0216	0.1503
	TOTAL	-	89	11	3.2446	0.6484	14.1026	0.3652	0.0346	0.1582	1.1719

Figure 2
Distribution of tree fern species in the state of Santa Catarina, southern Brazil.

The diameter distributions of the five species with the greatest density followed a right asymmetry distribution (Fig. 3), with exception of Alsophila setosa, whose individuals were predominantly in the class of 10-15 cm (96.88 %). Dicksonia sellowiana showed the largest diametric amplitude, with most of the individuals in the class of 20-25 cm (33.58 %). The mean diameter over all species was 16.22 cm (± 7.66 cm); D. sellowiana presented the largest mean diameter (23.02 cm ± 7.2 cm).

Figure 3
Diametric distribution of tree fern species in the state of Santa Catarina, southern Brazil.

Most individuals reached up to 5 m of height (Fig. 4), with a mean of 3.94 m (± 2.16 m). Cyathea delgadii had a less common distribution, because it has a distribution close to the normal distribution, while the others have right asymmetry, with a peak in the class of 6-7 m (20.79 %). For this reason, this species presented a greater mean height (6.28 m ± 2.54 m).

Figure 4
Hypsometric distribution of tree fern species in the state of Santa Catarina, southern Brazil.

Regression models

According to the AICc, OLS models presented better performance for all species (Tab. 3). No spatial structure in the dominance data of this species was found. For D. sellowiana, temperature and rainfall variables were significantly related to the dominance of the individuals. Regarding Cyatheaceae, temperature variables, solar radiation and aspect variables were significant (Tab. 4).

Table 3
Statistical parameters for OLS and GWR models for tree ferns in Santa Catarina State. AICc: corrected Akaike Information Criterion; r: correlation coefficient; r²: coefficient of determination; r² aj: coefficient of determination adjusted; F: F-test value; p: p-value.

Species	Dicksonia sellowiana		Alsophila setosa		Cyathea corcovadensis		Cyathea delgadii		Cyathea phalerata		Cyatheaceae
Parameters	GWR	OLS	GWR	OLS	GWR	OLS	GWR	OLS	GWR	OLS	GWR	OLS
AICc	-589.09	-593.38	-1,149.03	-1,130.74	-656.65	-663.52	-758.31	-768.58	-1,022.62	-1,031.52	-1,627.09	-1,647.96
r	0.693	0.627	0.63	0.467	0.532	0.429	0.616	0.403	0.53	0.437	0.553	0.454
r²	0.48	0.393	0.397	0.218	0.283	0.184	0.38	0.163	0.281	0.191	0.306	0.206
r² aj	0.417	0.373	0.25	0.169	0.2	0.184	0.246	0.135	0.197	0.174	0.2	0.179
F (r²)	7.005	15.809	2.55	4.463	2.793	11.039	2.551	3.883	3.031	7.613	2.757	6.441
p	<0.001	<0.001	<0.001	<0.001	0.016	0.002	0.007	0.013	<0.001	<0.001	<0.001	<0.001

Table 4
Standard coefficients for OLS model variables for tree ferns in Santa Catarina State. VIF: Variance Inflation Factor; t: t-value; p: p-value

Species	Variable	Standard Coefficient	VIF	t	p
Alsophila setosa	Constant	0	0	1.611	0.11
	Aspect	-0.162	1.145	-1.811	0.073
	Slope	-0.219	1.218	-2.379	0.019
	Bio4	0.201	1.61	1.892	0.061
	Bio8	-0.316	1.928	-2.722	0.008
	Bio9	-0.267	1.54	-2.58	0.011
	Bio18	0.209	1.541	2.019	0.046
	Rad_win	-0.323	2.012	-2.725	0.007
Cyathea corcovadensis	Constant	0	0	4.068	<0.001
	Bio9	-0.429	1	-3.323	0.002
Cyathea delgadii	Constant	0	0	-1.078	0.286
	Aspect	-0.269	1.024	-2.253	0.028
	Bio12	-0.53	3.309	-2.465	0.017
	Bio18	0.456	3.313	2.119	0.038
Cyathea phalerata	Constant	0	0	3.71	<0.001
	Bio8	-0.39	1.288	-3.763	<0.001
	Rad_win	-0.185	1.217	-1.832	0.07
	Rad_sum	-0.187	1.078	-1.976	0.051
Cyatheaceae	Constant	0	0	2.987	0.003
	Slope	-0,113	1.165	-1.552	0.123
	Aspect	-0,117	1.225	-2.371	0.019
	Bio4	0,123	1.047	1.776	0.077
	Bio8	-0,261	2.593	-2.4	0.017
	Bio9	-0,213	2.142	-2.155	0.033
	Rad_win	-0,2	1.858	-2.177	0.031
	Rad_sum	-0,154	1.584	-1.807	0.072
Dicksonia sellowiana	Constant	0	0	5.202	<0.001
	Bio4	-0.629	2.376	-5.786	<0.001
	Bio8	-0.676	3.704	-4.98	<0.001
	Bio9	0.678	4.005	4.801	<0.001
	Bio12	-0.293	1.503	-3.386	<0.001
	Bio18	0.387	3.627	2.884	0.005

Variance partitioning

The variance partition indicated that the dominance of tree ferns is significantly influenced by “pure” environmental variables (c) (F_{(df = 8)} = 13.41; p < 0.01) and “pure” space (a) (F_{(df = 8)} = 22.18; p < 0.01). Pure environmental factors explained the largest portion of the data variance (6.99 %), followed by the “pure” space (4.94 %) and by the spatially structured environmental variables (2.5 %). Unexplained factors (residuals) accounted for 85.57 % of the data variance.

Discussion

The increased density of tree ferns found in the subtropical Atlantic Forest indicate that the current environmental conditions in this domain are adequate for the occurrence of such plants, as observed by other authors (Catharino et al. 2006; Reginato & Goldenberg 2007; Conoletti et al. 2009; Klauberg et al. 2010; Ferreira et al. 2012; Lima et al. 2011; Silva et al. 2013). This result answered our first question, indicating that tree ferns species, such as Dicksonia sellowiana, Alsophila setosa and Cyathea phalerata, are very common in the studied forest types.

Inasmuch as Araucaria angustifolia is often regarded as the most abundant species in the Araucaria forest (Seger et al. 2005; Cordeiro & Rodrigues 2007), Dicksonia sellowiana proved to be the most abundant species in this forest type. The increased values of the phytosociological parameters of D. sellowiana were a consequence of the great number of observed individuals, which sometimes assemble as monodominant clusters on some SUs (Gasper et al. 2011). According to the last authors, the abundance of D. sellowiana on some SUs is related to temperature variation conditioned by high altitude (> 1,000 m) and, in fact, it proved to be a factor strongly associated with the occurrence of individuals of this species.

The small diameter and low height observed in most individuals are characteristic of tree ferns because they present slow growth, with growth rates of a few centimeters per year (Schmitt & Windisch 2006; 2012). From an ecological perspective, these features indicate that the populations tend to be stable and growing (Condit et al. 1998), and most likely the species found suitable sites for their development (Schmitt & Windisch 2007). According to Young & León (1989), these features are also common because young individuals present greater mortality rates than mature individuals.

Tree ferns can be found in all forest types in Santa Catarina. Nevertheless, Cyatheaceae has preference for the evergreen rainforest, where temperatures are higher compared to the Araucaria forest, where more intense frost events occur and Dicksonia sellowiana prevails (Mantovani 2004; Bystriakova et al. 2011; Gasper et al. 2011). These preferences are manifestations of the ecological demands of the two families, as reported by Tryon & Tryon (1982) and Gasper et al. (2011). The small abundance of tree ferns observed in the state’s extreme west, especially in the semi-deciduous forest, may be influenced by precipitation seasonality driven by drier winters than other regions of the state (Wrege et al. 2012). The drought period may be restrictive for the tree ferns species considered in this study because they (1) are generally associated with sites with abundant and constant humidity (Mantovani 2004); (2) require water for reproduction; and (3) do not have adaptations for drought periods, such as loss of leaves (Sharpe & Mehltreter 2010).

Aspect, temperature and precipitation are correlated with tree ferns dominance (Bystriakova et al. 2011). Indeed, we found a significant correlation between tree fern dominance and these variables. The negative correlation of Cyatheaceae dominance with aspect indicates that this family prefers less sloped sites (Jones et al. 2007). The influence of temperature and precipitation variables on Cyatheaceae species was reported by other authors (Tryon & Tryon 1982; Fernandes 1997). The positive correlation of Dicksonia sellowiana with precipitation (Bio18) indicates that it prefers humid sites, as observed by Mantovani (2004) and Mallmann et al. (2018). Moreover, the positive correlation with the mean temperature of the driest quarter (Bio8) indicates that sites with higher temperatures do not completely inhibit the occurrence of Dicksonia sellowiana. This may be explained by the fact that the species grows under the forest canopy.

The percentage of variance explained by the predictor variables we selected was smaller than other studies (e.g., Gasper et al. 2015; Saiter et al. 2015). The large unexplained fraction (residuals) suggests that we did not address other relevant variables that may influence the distribution patterns of the species, such as distance from the ocean (Saiter et al. 2015), biotic interactions such as herbivory (Aide 1988), and soil chemical properties that influence germination, such as pH (Marcon et al. 2014; 2017) and fertility (Jones et al. 2007).

Tree ferns are abundant in Santa Catarina, mainly in the evergreen rainforest where Alsophila setosa is expressive, and in the Araucaria forest where Dicksonia sellowiana predominates. Cyatheaceae and Dicksoniaceae occupy distinct habitats in Santa Catarina. The dominance of Dicksonia sellowiana was related to temperature and rainfall variables. For Cyatheaceae, different environmental variables were related to each species; some presented relationships with slope (Alsophila setosa), and others with aspect (Cyathea delgadii), temperature (Alsophila setosa, Cyathea corcovadensis and Cyathea phalerata), precipitation (Alsophila setosa and Cyathea delgadii) and solar radiation (Alsophila setosa and Cyathea phalerata). These results extend our knowledge on ecological patterns of this plant group, and can be used to elaborate conservation actions for each species according to their ecological preferences.

Acknowledgements

The authors thank the IFFSC for sharing its data, as well as to Fundação de Apoio à Pesquisa Científica e Tecnológica de Santa Catarina (FAPESC) for supporting the IFFSC. This study was in part financed by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brazil (CAPES) - Finance Code 001. We are also thankful to Guilherme Salgado Grittz, Laio Zimermann Oliveira, and Marta Helena Caetano for helping with the revision and translation of this manuscript.

References

• Aide TM. 1988. Herbivory as a selective agent on the timing of leaf production in a tropical understory community. Nature 336: 574-575.
• Alvares CA, Stape JL, Sentelhas PC, Gonçalves JLM, Sparovek G. 2013. Köppen’s climate classification map for Brazil. Meteorologische Zeitschrift 22: 711-728.
• Arens NC, Baracaldo PS. 1998. Distribution of tree ferns (Cyatheaceae) across the successional mosaic in an Andean Cloud Forest, Narino, Colombia. American Fern Journal 88: 60-71.
• Brock JMR, Perry GLW, Lee WG, Burns BR. 2016. Tree fern ecology in New Zealand: A model for southern temperate rainforests. Forest Ecology and Management 375: 112-126.
• Burnham KP, Anderson DR. 2002. Model selection and multimodel inference: A practical information-theoretic approach. 2nd. edn. New York, Spring.
• Bystriakova N, Schneider H, Coomes D. 2011. Evolution of the climatic niche in scaly tree ferns (Cyatheaceae, Polypodiopsida). Botanical Journal of the Linnean Society 165: 1-19.
• Catharino ELM, Bernacci LC, Franco GADC, Durigan G, Metzger JP. 2006. Aspectos da composição e diversidade do componente arbóreo das florestas da Reserva Florestal do Morro Grande, Cotia, SP. Biota Neotropica 6: 0-28.
• Clappe S, Dray S, Peres-Neto P. 2018. Beyond neutrality: disentangling the effects of species sorting and spurious correlations in community analysis. Ecology 99: 1737-1747.
• Condit R, Sukumar R, Hubbell SP, Foster RB. 1998. Predicting population trends from size distributions: a direct test in a tropical tree community. The American Naturalist 152: 495-509.
• Conoletti S, Citadini-Zanette V, Martins R, Santos R, Rocha E, Jarenkow JA. 2009. Florística e estrutura fitossociológica em Floresta Ombrófila Densa Submontana na barragem do rio São Bento, Siderópolis, estado de Santa Catarina. Acta Scientiarum Biological Sciences 31: 397-405.
• Cordeiro J, Rodrigues WA. 2007. Caracterização fitossociológica de um remanescente de Floresta Ombrófila mista em Guarapuava, PR. Revista Árvore 31: 545-554.
• Della AP, Vasques DT. 2017. Dicksoniaceae in Flora do Brasil em Construção 2020. http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/FB90945 20 Dec. 2017.
  » http://floradobrasil.jbrj.gov.br/reflora/floradobrasil/FB90945
• Dray S, Dufour AB. 2007. The ade4 package: implementing the duality diagram for ecologists. Journal of Statistical Software 22: 1-20.
• Dray S, Legendre P, Peres-Neto PR. 2006. Spatial modelling: a comprehensive framework for principal coordinate analysis of neighbour matrices (PCNM). Ecological Modelling 196: 483-493.
• Fernandes I. 1997. Taxonomia e fitogeografia de Cyatheaceae e Dicksoniaceae nas regiões sul e sudeste do Brasil. PhD Thesis, Universidade de São Paulo, São Paulo.
• Ferreira PI, Paludo GF, Chaves CL, Bortoluzzi RLC, Mantovani A. 2012. Florística e fitossociologia arbórea de remanescentes florestais em uma fazenda produtora de Pinus spp. Floresta 42: 783-794.
• Fick SE, Hijmans RJ. 2017. Worldclim 2: New 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology 37: 4302-4315.
• Fortin MJ, Dale MRT. 2005. Spatial analysis: a guide for ecologists. New York, Cambridge University Press.
• Fotheringham AS, Brunsdon C, Charlton M. 2002. Geographically Weighted Regression: the analysis of spatially varying relationships. Chichester, Wiley.
• Fraga LL, Silva LB, Schmitt JL. 2008. Composição e distribuição vertical de pteridófitas epifíticas sobre Dicksonia sellowiana Hook. (Dicksoniaceae), em floresta ombrófila mista no sul do Brasil. Biota Neotropica 8: 123-129.
• Gasper AL, Sevegnani L, Vibrans AC, et al 2011. Inventory of Dicksonia sellowiana Hook. in Santa Catarina. Acta Botanica Brasilica 25: 776-784.
• Gasper AL, Eisenlohr PV, Salino A. 2015. Climate-related variables and geographic distance affect fern species composition across a vegetation gradient in a shrinking hotspot. Plant Ecology & Diversity 8: 25-35.
• Hair JF, Black WC, Babin BJ, Anderson RE, Tatham RL. 2009. Análise multivariada de dados. 6st. edn. Porto Alegre, Bookman.
• INPE. 2017. TOPODATA: Banco de dados geomorfométricos do Brasil. http://www.dsr.inpe.br/topodata/index.php 5 Oct. 2017.
  » http://www.dsr.inpe.br/topodata/index.php
• Jones MM, Rojas PO, Tuomisto H, Clark DB. 2007. Environmental and neighbourhood effects on tree fern distributions in a neotropical lowland rain forest. Journal of Vegetation Science 18: 13-24.
• Klauberg C, Paludo GF, Bortoluzzi RLC, Mantovani A. 2010. Florística e estrutura de um fragmento de Floresta Ombrófila Mista no Planalto Catarinense. Biotemas 23: 35-47.
• Large MF, Braggins JE. 2004. Tree ferns. Cambridge, Timber Press.
• Leite PF. 2002. Contribuição ao conhecimento fitoecológico do sul do Brasil. Ciência & Ambiente 24: 51-73.
• Lima MEL, Cordeiro I, Moreno PRH. 2011. Estrutura do componente arbóreo em Floresta Ombrófila Densa Montana no Parque Natural Municipal Nascentes de Paranapiacaba (PNMNP), Santo André, SP, Brasil. Hoehnea 38: 73-96.
• Lingner DV, Schorn LA, Vibrans AC. 2013. Fitossociologia do componente arbóreo/arbustivo da Floresta Ombrófila Mista em Santa Catarina. In: Vibrans AC, Sevegnani L, Gasper AL, Lingner DV. (eds.) Inventário florístico florestal de Santa Catarina: floresta ombrófila mista. Blumenau, Edifurb. p. 157-189.
• Mallmann IT, Silva VL, Port RK, Oliveira FB, Schmitt JL. 2018. Spatial distribution analysis of Dicksonia sellowiana Hook. in Araucaria forest fragments with different sizes. Brazilian Journal of Biology 79: 337-344.
• Mantovani M. 2004. Caracterização de populações naturais de xaxim (Dicksonia sellowiana (Presl.) Hooker), em diferentes condições edafo-climáticas no estado de Santa Catarina. MsC. Thesis, Universidade Federal de Santa Catarina, Florianópolis.
• Marcon C, Silveira T, Droste A. 2014. Germination and gametophyte development of Cyathea corcovadensis (Raddi) Domin (Cyatheaceae) from spores stored at low temperatures. Acta Scientiarum Biological Sciences 36: 403-410.
• Marcon C, Silveira T, Schmitt JL, Droste A. 2017. Abiotic environmental conditions for germination and development of gametophytes of Cyathea phalerata Mart. (Cyatheaceae). Acta Botanica Brasilica 31: 58-67.
• Medeiros MCMP, Aidar MPM. 2011. Structural variation and content of aboveground living biomass in an area of Atlantic Forest in the State of São Paulo, Brazil. Hoehnea 38: 413-428.
• Meyer L, Sevegnani L, Gasper AL. 2013. Fitossociologia do componente arbóreo/arbustivo da Floresta Ombrófila Mista em Santa Catarina. In: Vibrans AC, Sevegnani L, Gasper AL, Lingner DV. (eds.) Inventário florístico florestal de Santa Catarina: floresta ombrófila mista . Blumenau, Edifurb . p. 157-189.
• Moran RC, Klimas S, Carlsen M. 2003. Low-trunk epiphytic ferns on tree ferns versus angiosperms in Costa Rica. Biotropica 35: 48-56.
• Müeller-Dombois D, Ellenberg H. 1974. Aims and methods of vegetation ecology. New York, John Wiley and Sons.
• Naimi B, Hamm NAS, Groen TA, Skidmore AK, Toxopeus AG. 2014. Where is positional uncertainty a problem for species distribution modelling? Ecography 37: 191-203.
• Neumann MK, Schneider PH, Schmitt JL. 2014. Phenology, caudex growth and age estimation of Cyathea corcovadensis (Raddi) Domin (Cyatheaceae) in a subtropical forest in southern Brazil. Acta Botanica Brasilica 28: 274-280.
• Peres-Neto PR, Legendre P. 2010. Estimating and controlling for spatial structure in the study of ecological communities. Global Ecology and Biogeography 19: 174-184.
• Pteridophyte Phylogeny Group 1. 2016. A community-derived classification for extant lycophytes and ferns. Journal of Systematics and Evolution 54: 563-603.
• R Development Core team. 2011. R: A language and environment for statistical computing. Vienna, R Foundation for Statistical Computing.
• Rangel TF, Diniz-Filho JAF, Bini LM. 2010. SAM: A comprehensive application for Spatial Analysis in Macroecology. Ecography 33: 46-50.
• Reginato M, Goldenberg R. 2007. Análise florística, estrutural e fitogeográfica da vegetação em região de transição entre as florestas ombrófilas mista e densa montana, Piraquara, Paraná, Brasil. Hoehnea 34: 349-360.
• Saiter FZ, Eisenlohr PV, França GS, Stehmann JR, Thomas WW, Oliveira-Filho AT. 2015. Floristic units and their predictors unveiled in part of the Atlantic Forest hotspot: Implications for conservation planning. Anais da Academia Brasileira de Ciencias 87: 2031-2046.
• Santiago ACP, Mynssen CM, Maurenza D, Penedo TSA, Sfair JC. 2013. Dicksoniaceae. In: Martinelli G, Moraes MA. (eds.) Livro vermelho da flora do Brasil. Rio de Janeiro, Andrea Jakobsson, Instituto de Pesquisas Jardim Botânico do Rio de Janeiro. p. 475-476.
• Schmitt JL, Schneider PH, Windisch PG. 2009. Crescimento do cáudice e fenologia de Dicksonia sellowiana Hook. (Dicksoniaceae) no sul do Brasil. Acta Botanica Brasilica 23: 283-291.
• Schmitt JL, Windisch PG. 2005. Aspectos ecológicos de Alsophila setosa Kaulf. (Cyatheaceae, Pteridophyta) no Rio Grande do Sul, Brasil. Acta Botanica Brasilica 19: 859-865.
• Schmitt JL, Windisch PG. 2006. Growth Rates and Age Estimates of Alsophila setosa Kaulf. in Southern Brazil. American Fern Journal 96: 103-111.
• Schmitt JL, Windisch PG. 2007. Estrutura populacional e desenvolvimento da fase esporofítica de Cyathea delgadii Sternb. (Cyatheaceae, Monilophyta) no sul do Brasil. Acta Botanica Brasilica 21: 731-740.
• Schmitt JL, Windisch PG. 2010. Biodiversity and spatial distribution of epiphytic ferns on Alsophila setosa Kaulf. (Cyatheaceae) caudices in Rio Grande do Sul, Brazil. Revista Brasileira de Biologia 70: 521-528.
• Schmitt JL, Windisch PG. 2012. Caudex growth and phenology of Cyathea atrovirens (Langsd. & Fisch.) Domin (Cyatheaceae) in secondary forest, southern Brazil. Brazilian Journal of Biology 72: 397-405.
• Seger CD, Dlugosz FL, Kurasz G, et al 2005. Levantamento florístico e análise fitossociológica de um remanescente de Floresta Ombrófila Mista localizado no município de Pinhais, Paraná, Brasil. Floresta 35: 291-302.
• Sharpe JM, Mehltreter K. 2010. Ecological insights from fern population dynamics. In: Mehltreter K, Walker LR, Sharpe JM. (eds.) Fern ecology. New York, Cambridge. p. 61-110.
• Silva AC, Higushi P, Negrini M, Grudtner A, Zech DF. 2013. Caracterização fitossociológica e fitogeográfica de um trecho de floresta ciliar em Alfredo Wagner, SC, como subsídio para restauração ecológica. Ciencia Florestal 23: 579-593.
• Tryon RM, Tryon AF. 1982. Ferns and allied plants with special reference to tropical America. New York, Springer.
• Tuomisto H, Doninki J, Ruokolainen K. 2019. Discovering floristic and geoecological gradients across Amazonia. Journal of Biogeography 48: 1734-1748.
• Vibrans AC, McRoberts RE, Moser P, Nicoletti AL. 2013. Using satellite image-based maps and ground inventory data to estimate the area of the remaining Atlantic forest in the Brazilian state of Santa Catarina. Remote Sensing of Environment 130: 87-95.
• Vibrans AC, Sevegnani L, Lingner DV, Gasper AL, Sabbagh S. 2010. Inventário florístico florestal de Santa Catarina (IFFSC): aspectos metodológicos e operacionais. Pesquisa Florestal Brasileira 30: 291-302.
• Weigand A, Lehnert M. 2016. The scaly tree ferns (Cyatheaceae-Polypodiopsida) of Brazil. Acta Botanica Brasilica 30: 336-350.
• Windisch PG. 2002. Fern Conservation in Brazil. Fern Gazette 16: 295-300.
• Wrege MS, Steinmetz S, Reisser Júnior C, Almeida IR. 2012. Atlas climático da região sul do Brasil: Estados do Paraná, Santa Catarina e Rio Grande do Sul. 2nd. edn. Brasília, Embrapa.
• Young KR, León B. 1989. Pteridophyte species diversity the Central Peruvian Amazon: importance of edaphic specialization. Brittonia 41: 388-395.
• Zar JH. 1999. Bioestatistical analysis. 4th. edn. New York, Prentice Hall.

Publication Dates

• Publication in this collection
  20 Mar 2020
• Date of issue
  Jan-Mar 2020

History

• Received
  10 Oct 2019
• Accepted
  12 Dec 2019