Brazilian Journal of Botany

, Volume 39, Issue 3, pp 833–844

Raunkiaerian life-forms in the Atlantic forest and comparisons of life-form spectra among Brazilian main biomes

  • Ana Claudia de Meneses Costa
  • Marcelo Freire Moro
  • Fernando Roberto Martins
Article

DOI: 10.1007/s40415-016-0281-z

Cite this article as:
de Meneses Costa, A.C., Moro, M.F. & Martins, F.R. Braz. J. Bot (2016) 39: 833. doi:10.1007/s40415-016-0281-z

Abstract

Life-form spectra are known to vary greatly among different biomes, being largely defined by macroclimates on a global scale. Benign climates usually have biomes with a large proportion of phanerophytes, whereas harsher climates favor life-forms with greater degrees of bud protection. We sampled the flora of a semideciduous Atlantic forest and classified each recorded species in one of the Raunkiaerian life-forms. We then compared the biological spectrum of this seasonal forest with other published spectra of the main Brazilian biomes. We show that each biome has a clear functional spectrum and that the semideciduous Atlantic forest is to some extent functionally similar to Amazonian and Atlantic ombrophilous forests. The relatively moist semideciduous and the ombrophilous forests have biological spectra dominated by phanerophytes. The fire-prone, seasonal “Cerrado” savannas have spectra dominated by both phanerophytes and hemicryptophytes. However, the spectra of the semiarid “Caatinga” woodlands seem to depend on soil type: woodlands on crystalline terrains are dominated by therophytes, whereas those on sedimentary terrains are dominated by both phanerophytes and therophytes. Sedimentary caatingas have spectra whose features are in an intermediate position between those of crystalline caatingas and moist forests. On a continental scale, macroclimate seems to be the most important environmental factor separating different biological spectra, but wildfires in the “Cerrado” and soil differences between crystalline and sedimentary Caatinga woodlands also play an important role in defining the local spectrum on a regional scale.

Keywords

Floristic survey Tropical forests Tropical woodlands Tropical savannas Vegetation functioning 

Introduction

In the early twentieth century, the Danish botanist C. Raunkiaer developed a system to classify plants’ life-forms (Raunkiaer 1934). He assumed that the degree of protection that plants give to their renewing buds is a key feature, and that this degree of protection reflects an adaptive response of plants to the main environmental conditions, especially climate (Cain 1950; Batalha and Martins 2002).

Raunkiaer (1934) divided the terrestrial plant habits in five main classes of life-forms: (1) Phanerophytes, which keep their vegetative buds high above the soil during the unfavorable season; (2) Chamaephytes, which have their renewing buds close to (but still above) the ground level during the unfavorable season; (3) Hemicryptophytes, which keep their buds at the ground level; (4) Cryptophytes—geophytes, which maintain their vegetative buds protected below ground level; and (5) Therophytes (also called annual plants), which complete their life-cycles within one favorable season, reproduce and die, waiting for the next favorable season protected as resistant seeds. If environmental conditions are very benign, plants do not need to have special protection for their buds, and phanerophytic life-forms may prevail, whereas harsh environmental conditions should favor species with more protection to the buds (Raunkiaer 1934; Cain 1950; Batalha and Martins 2002; Martins and Batalha 2011).

In the eastern side of the South American continent, the Atlantic Forest Phytogeographical Domain (Atlantic forest hereafter) is one of the richest regions in plant species in the world, being recognized as one of the world’s biodiversity hotspots (Myers et al. 2000). Zappi et al. (2015) compiled a general list of angiosperms from the Atlantic forest and registered over 15,000 plant species, 49.5 % of which are endemic. This, summed with thousands of species of bryophytes, lycophytes, and ferns (Costa and Peralta 2015; Prado et al. 2015), justifies the recognition of the Atlantic forest as one of the richest regions in the world. Once covering an area of around 150 million ha, the Atlantic forest is restricted nowadays to fragments, most of which are small sized (Ribeiro et al. 2009). Nevertheless, the Atlantic forest still harbors much of the Brazilian biodiversity and shows complex patterns of species distributions, ruled mainly not only by climatic gradients, but also by latitudinal and altitudinal variables (Oliveira Filho and Fontes 2000; Eisenlohr and Oliveira Filho 2015).

In the Atlantic Forest, there is a climatic gradient from the sea to the inner continent, where the rainfall amount is reduced, and the seasonality increased, when one goes toward inland. According to Oliveira Filho and Fontes (2000), the Atlantic forest may be divided into ombrophilous (rain) and semideciduous (seasonal) forests, with each subtype having a distinguishable flora (Oliveira Filho and Fontes 2000; Eisenlohr and Oliveira Filho 2015). The semideciduous Atlantic forests, being located further from the coast, are subjected to a greater seasonality and less rainfall than the ombrophilous forest, a fact that helps to explain floristic and ecological differences between the Atlantic ombrophilous and semideciduous forests (Oliveira Filho and Fontes 2000). However, what about functional differences between these two forests? Are differences in climate regime influencing only the flora, or is it also detectable in the biological spectrum of the plant communities?

The semideciduous forest is subjected to a stronger seasonality when compared with the ombrophilous forest, but this seasonality is still considered milder when compared to other Brazilian Phytogeographical Domains, such as the semiarid “Caatinga” (in NE Brazil) and the seasonal, fire-prone “Cerrado” (savanna-like vegetation), or even other South American formations, such as those of the Gran Chaco (Cabrera and Willink 1973). How do the biological spectra of the semideciduous Atlantic forests relate to the spectra found in the milder ombrophilous forests and the harsher “Cerrado” and “Caatinga” domains?

Oliveira Filho and Fontes (2000) suggested that the semideciduous seasonal Atlantic forest constitutes a different floristic group when compared with the dense ombrophilous forest. We address this question from a different viewpoint. We aim to verify whether each main Brazilian biome have clear structural differences regarding their life-form spectra and how the spectra of the Atlantic forests relate to other Brazilian biomes.

Materials and methods

Study site

This study was carried out in a riparian seasonal Atlantic forest located inside a vegetation fragment of approximately 100 hectares, in a private area (Clube de Campo Valinhos), in Serra dos Cocais, municipality of Valinhos, State of São Paulo, Brazil (22°57′24″S and 46°55′55″W—Fig. 1). The maximum altitude is 950 m, and the area has varied reliefs, with abundant outcrops of granite boulders (Zampaulo et al. 2007).
Fig. 1

Geographical location of the study site in the Serra dos Cocais, municipality of Valinhos, State of São Paulo, Brazil (22°57′24″S and 46°55′55″W). The Atlantic Forest Domain is colored in gray. Map Design: M.F. Moro

The regional climate is classified as Köppen’s Cwa: megathermic temperate with dry winters (not very harsh, though) and rainy summers. The average annual precipitation is 1425 mm, and average annual temperature is 21 °C (Cepagri/Unicamp). According to Oliveira (1980), the dominant soil is the Yellow-red Latosol (Oxisol).

Data collection

Between October 2008 and November 2009, we collected botanic material monthly by walking in trails all over the studied area, and sampled terricolous angiosperms of all life-forms. Reproductive branches of the species were collected, and the material was processed following usual botanical techniques (Fidalgo and Bononi 1989; Bridson and Forman 1998). The exsiccatae are lodged in the Herbarium of the State University of Campinas (UEC). The taxonomic identification was made with the aid of identification keys, as well as research in literature, herbarium exsiccatae and consults to taxonomists.

Life-form data

Batalha and Martins (2002) compiled biological spectra available in the literature for the “Cerrado” savannas and Atlantic and Amazonian forests. Moro et al. (2016) complemented this dataset adding spectra for the semiarid “Caatinga” woodlands (seasonally dry forests sensu Pennington et al. 2000). We used the spectra of Brazilian biomes presented by these authors to compare with the spectrum that we found in our site of the seasonal semideciduous Atlantic forest. We also present here two extra, inedited life-form spectra. We took floristic data collected by colleagues at the Ribeirão Cachoeira Semideciduous Atlantic Forest, a fragment of semideciduous forest located in the Sousas-Joaquim Egídio Environment Protection Area in Campinas municipality, and floristic data in the ombrophilous Atlantic forest at Serra do Mar State Park, Ubatuba municipality, both areas in São Paulo State. The data for Ribeirão Cachoeira and Ubatuba were kindly provided by Dr. Juliano Van Melis, Dr. Leandro Vieira, Dr. Bruno Aranha, and Dr. André Rochelle. Taking the species lists provided by those authors, we updated species nomenclature using the Plantminer tool (Carvalho et al. 2010), attributed one life-form for each species recorded, and produced a general list of species as well as a life-form spectrum for each of these forests (the data compiled are available as an electronic supplement to this paper at http://dx.doi.org/10.6084/m9.figshare.3114436). To attribute the life-form of each species in Valinhos, Ribeirão Cachoeira, and Ubatuba Atlantic Forests, we used the system proposed by Raunkiaer (1934). We included the epiphytes and woody lianas in the phanerophytic life-form category. The species lists with the respective life-form for each species in Valinhos, Ribeirão Cachoeira, and Ubatuba, as well as the database of life-form spectra used in our analysis are available for download at http://dx.doi.org/10.6084/m9.figshare.3114436.

Data analysis

To compare differences in life-form spectra among the main Brazilian biomes, we constructed a matrix with all life-form spectra compiled from the literature (Table 1) and analyzed the structural relationships among Brazilian biomes using a non-metric multidimensional scaling analysis (NMS) with Euclidian distances (McCune and Grace 2002; Legendre and Legendre 2012) in order to understand how the spectra of different biomes are related to each other. You can download this table at: http://dx.doi.org/10.6084/m9.figshare.3114436.
Table 1

Raunkiaer’s life-form spectra of sites in the main Brazilian biomes

Site

Life-form class (%)

Reference

Ph

Ch

He

Cr

Th

Semideciduous Atlantic forest

 Serra dos Cocais, Valinhos, SP, Brazil

73.3

10.3

11.6

2.1

2.7

This study

 Ribeirão Cachoeira forest, Campinas, SP, Brazil

93.2

2.8

2.6

1.4

0

Inedited data available through our electronic supplementary files

Ombrophilous Atlantic forest

 Ubatuba, SP, Brazil

99

0.3

0

0.8

0

Inedited data available through our electronic supplementary files

 Alto do Palmital, Foz de Iguaçu, PR, Brazil

80

6

11

3

0

Cain et al. (1956)

 Caiobá, Matinhos, PR, Brazil

87

7

3

3

0

Cain et al. (1956)

 Horto Botânico, Pelotas, RS, Brazil

70

4

16

5

5

Cain et al. (1956)

Amazon rain forest

 Mucambo, Belém, PA, Brazil

95

1

3

1

0

Cain et al. (1956)

“Cerrado”

 Brasília, Brazil

39.1

13.5

44.9

1.8

0.7

Ratter (1980)

 Emas National Park, GO, Brazil

31.6

12.8

49.9

2

3.7

Batalha and Martins (2002)

 Lagoa Santa, MG, Brazil

28.8

6.1

55.1

5.4

4.6

Warming and Ferri (1973)

 Mojiguaçu, SP, Brazil

30.9

12.2

47

2.1

7.8

Mantovani (1983)

 Pirassununga, SP, Brazil

40.1

17.1

36.1

1.1

5.6

Batalha et al. (1997)

 Santa Rita do Passa Quatro, SP, Brazil

45.3

17.2

30

0.8

6.7

Batalha and Mantovani (2001)

Crystalline “caatinga”

 Crateús, CE, Brazil

28.5

19

5.1

1.5

46

Araújo et al. (2011)

 Quixadá, CE, Brazil

26.3

15.8

12.8

2.3

42.9

Costa et al. (2007)

 Floresta/Betânia 1, PE, Brazil

28.7

21.8

12.9

1

35.6

Costa et al. (2009)

 Floresta/Betânia 2, PE, Brazil

26.1

19.6

15.2

2.2

37

Rodal et al. (2005)

 Floresta, PE, Brazil

23.4

16.9

16.9

2.6

40.3

Rodal et al. (2005)

Sedimentary caatinga

 Crateús 1, CE, Brazil

57.4

19.1

3.7

2.9

16.9

Araújo et al. (2011)

 Crateús 2, CE, Brazil

58

23.2

2.4

3.2

13.2

Araújo et al. (2011)

 S. José do Piauí, PI, Brazil

71.3

12.5

8.1

3.7

4.4

Mendes and Castro (2010)

Raunkiaer’s life-forms: Ph phanerophytes, Ch Chamaephytes, He Hemicryptophytes, Cr Cryptophytes (geophytes), Th Therophytes. The spectra presented here were generated by our inedited data plus the spectra compiled in Batalha and Martins (2002) and Moro et al. (2016)

Results

We put together for comparison 21 biological spectra from the main Brazilian biomes (Table 1) compiled from Batalha and Martins (2002), Moro et al. (2016), and our three inedited spectra. The database is composed by two studies in the semideciduous Atlantic forests, four in the Ombrophilous Atlantic Forests, one in the Amazon Rainforest, six in the “Cerrado” savannas, and eight in the Caatinga woodlands (five in “Caatinga” woodlands on the crystalline basement and three on sedimentary terrains) (Table 1).

In our study site in Serra dos Cocais, Valinhos, we sampled 146 plant species in 52 families (Table 2). Trees were the most common habit, representing 46 species, followed by shrubs (37 species), herbs (28), climbers (20), subshrubs (12), and three species of climbing shrubs (Table 2). After attributing a life-form to each species, we found 107 species (73 %) of phanerophytes, 17 (12 %) of hemicryptophytes, 15 (10 %) of chamaephytes, 4 (3 %) of therophytes, and 3 (2 %) of cryptophytes—geophytes (Tables 1, 2).
Table 2

Species sampled in the riparian semideciduous Atlantic forest and their habits and Raunkiaer’s life-forms

Family

Species

Habit

Life-form

Collector number

UEC number

Acanthaceae

Justicia lythroides V.A. W. Graham

Subshrub

Ch

99, 113, 150

151,524, 151,538, 151,574

Staurogyne hirsuta (Nees) Kuntze

Herb

He

100, 119

151,525, 151,541

Amaranthaceae

Alternanthera brasiliana (L.) Kuntze

Herb

He

163

151,587

Hebanthe eriantha (Poir.) Pedersen

Climber

Ph

199

151,623

Apocynaceae

Oxypetalum wightianum Hook. & Arn.

Climber

Ph

28, 77, 223

151,502, 151,453, 151,647

Araliaceae

Dendropanax cuneatus (DC.) Decne. & Planch.

Tree

Ph

161, 203

151,585, 151,627

Asparagaceae

Herreria salsaparrilha Mart.

Climber

Ph

160

151,584

Asteraceae

Asteraceae sp.1

Tree

Ph

83

151,508

Baccharis dracunculifolia DC.

Shrub

Ph

86

151,511

Baccharis sp.

Shrub

Ph

94

151,519

Chromolaena odorata (L.) R.M. King & H. Rob.

Subshrub

Ch

177

151,601

Elephantopus mollis Kunth

Herb

He

102, 115

151,527, 15,154

Gochnatia polymorpha (Less.) Cabrera

Tree

Ph

5, 7, 96

151,430, 151,432, 151,521

Heterocondylus alatus (Vell.) R.M. King & H. Rob.

Shrub

Ph

176, 187

151,600, 151,611

Hypochaeris brasiliensis Griseb.

Herb

Th

221

151,645

Mikaniaglomerata Spreng.

Climber

Ph

39

151,464

Mikania sp.

Climber

Ph

186

151,610

Senecio brasiliensis (Spreng.) Less.

Herb

Th

232

151,656

Tilesia baccata (L.) Pruski

Shrub

Ph

98

151,523

Trixis praestans (Vell.) Cabrera

Subshrub

Th

168

151,592

Vernonia herbacea (Vell.) Rusby

Herb

He

93

151,518

Vernonia polyanthes Less.

Shrub

Ph

6

151,431

Vernonia scorpioides (Lam.) Pers.

Climbing shrub

Ph

1, 24, 25

151,426, 151,449, 151,450

Begoniaceae

Begonia fischeri Schrank.

Herb

He

42

151,467

Bignoniaceae

Dolichandra unguis-cati (L.) L.G.Lohmann

Climber

Ph

18, 111, 173

151,443, 151,536, 151,597

Pyrostegia venusta (Ker Gawl.) Miers

Climber

Ph

212

151,212

Boraginaceae

Cordia guazumaefolia (Jacq.) HBK

Shrub

Ph

41, 237

151,466, 151,661

Heliotropium transalpinum Vell.

Herb

Ch

63

151,488

Tournefortiapaniculata Cham.

Climber

Ph

20, 85

151,445, 15,151

Tournefortia villosa Salzm. ex DC.

Climber

Ph

233

151,657

Tournefortia hirsutissima L.

Climber

Ph

145, 178

151,569, 151,602

Bromeliaceae

Billbergia nutans H. Wendland ex Regel

Herb

He

174

151,598

Cactaceae

Cereus hildmannianus Schum.

Shrub

Ph

230

151,654

Cannabaceae

Trema micrantha (L.) Blume

Tree

Ph

73, 90

151,498, 151,515

Commelinaceae

Dichorisandra incurva Mart.

Subshrub

Ch

82

151,507

Curcubitaceae

Melothria aff. pendula.

Climber

Ph

81

151,506

Cyperaceae

Scleria panicoides Kunth

Herb

He

33.34

151,458

Scleria plusiophylla Stend.

Herb

He

195

151,619

Dileniaceae

Davilla elliptica A.St.-Hil.

Climber

Ph

217

151,641

Euphorbiaceae

Acalypha weddelliana Baill..

Herb

Ch

122

151,546

Actinostemon cf. klotzschii (Didr.) Pax

Tree

Ph

210

151,634

Actinostemon concolor (Spreng.) Müll.Arg.

Tree

Ph

190

151,614

Croton floribundus Spreng.

Tree

Ph

46, 47, 55, 59, 95

151,471, 151,472, 151,478, 151,484, 151,520

Dalechampia triphylla Lam.

Climber

Ph

180

151,604

Pachystroma longifolium (Nees) Johnston

Tree

Ph

225

151,649

Fabaceae (Caesalpinioideae)

Chamaecrista pilosa (L.) Greene

Shrub

Ph

148

151,572

Senna cf. villosa (Mill.) H.S. Irwin & Barneby

Shrub

Ph

167

151,591

Fabaceae (Faboideae)

Crotalaria breviflora D.C.

Shrub

Ch

38

151,463

Dalbergia frutescens (Vell.) Britton

Tree

Ph

14

151,439

Galactia striata (Jacq.) Urb.

Climber

Ph

151

151,575

Dioclea bicolor Benth.

Climber

Ph

58

151,483

Fabaceae (Mimosoideae)

Piptadenia gonoacantha (Mart.) J.F. Macbr.

Tree

Ph

189

151,613

Lacistemataceae

Lacistema hasslerianum Choat

Tree

Ph

130, 194

151,554, 151,618

Lamiaceae

Aegiphilaverticillata Vell.

Tree

Ph

8

151,433

Hyptis cf. suaveolens (L.) Poit.

Shrub

Ph

123, 144, 153

151,547, 151,568, 151,577

Peltodon radicans Pohl.

Herb

He

116

151,542

Lauraceae

Endlicheria paniculata (Spreng.) Macbride

Tree

Ph

89, 101, 131, 165

151,514, 151,526, 151,555, 151,589

Nectandra megapotamica (Spreng.) Mez

Tree

Ph

197

151,621

Ocotea odorifera (Vell.) Rohwer

Tree

Ph

209

151,633

Loganiaceae

Strychnos brasiliensis (Spreng.) Mart.

Climbing shrub

Ph

228

151,652

Lythraceae

Cupheamesostemon Koehne

Herb

He

57

151,482

Malvaceae

Byttneria sp.

Tree

Ph

23

151,448

Luehea divaricata Mart.

Tree

Ph

16, 17

151,441, 151,442

Luehea sp.

Shrub

Ph

45

151,470

Pavonia sepium A.St. Hil.

Shrub

Ph

71

151,496

Triumfetta rhomboidea Jacq.

Subshrub

Ph

231

151,655

Triumfetta semitrilobaJacq.

Shrub

Ph

124, 125, 132

151,548, 151,549, 151,556

Wissadula amplissima (L.) R.E. Fr.

Subhrub

He

215

151,640

Marantaceae

Stromanthe tonckat (Aubl.) Eichler

Herb

Ph

156

151,580

Melastomataceae

Leandra sp.1

Shrub

Ph

84

151,509

Leandra sp.2

Shrub

Ph

136

151,560

Leandra sp.3

Shrub

Ph

141

151,565

Leandra sp.4

Tree

Ph

234

151,658

Leandra australis (Cham.) Cogn.

Shrub

Ph

235

151,659

Leandra cf. purpurascens (DC.) Cogn.

Shrub

Ph

80

151,505

Ossaea sanguinea Cogn.

Shrub

Ph

68

151,493

Tibouchina herbacea (DC.) Cogn.

Shrub

Ph

104

151,529

Meliaceae

Cabralea canjerana (Vell.) Mart.

Tree

Ph

56

151,481

Guarea macrophylla Vahl

Tree

Ph

29, 36, 40, 138, 218, 219

151,454, 151,461, 151,465, 151,562, 151,642, 151,643

Guarea macrophylla subsp. tuberculata (Vell.) Penn.

Tree

Ph

53, 60

151,479

Trichilia elegans A. Juss.

Tree

Ph

27, 106, 192, 227

151,452, 151,531, 151,616, 151,651

Trichilia pallida Sw.

Tree

Ph

9, 48, 70, 91, 118, 152

151,434, 151,473, 151,495, 151,516, 151,544, 151,576

Monimiaceae

Mollinedia elegans Tul.

Tree

Ph

226

151,650

Mollinedia triflora (Spreng.) Tul.

Tree

Ph

183, 208

151,607, 151,632

Mollinedia widgrenii A.DC.

Tree

Ph

50, 76, 181, 202

151,475, 151,501, 151,605, 151,626

Moraceae

Ficus guaranitica Chodat

Tree

Ph

88

151,513

Myrtaceae

Eugenialigustrina(Sw.) Willd.

Tree

Ph

121

57,958

Eugenia sp.1

Tree

Ph

15

151,440

Eugenia sp.2

Tree

Ph

159

151,583

Eugenia sp.3

Tree

Ph

213

151,637

Myrcia fallax (Rich.) DC.

Shrub

Ph

4, 21, 22, 52, 79, 198, 201, 214

151,429, 151,446, 151,447, 151,477, 151,504, 151,622, 151,625, 151,638

Psidium guineense Sw.

Shrub

Ph

3, 11, 170

151,428, 151,436, 151,594

Myrtaceae sp.1

Tree

Ph

184

151,608

Nyctaginaceae

Guapira opposita (Vell.) Reitz

Subshrub

Ch

49

151,474

Orchidaceae

Sarcoglottis fasciculata (Vell.) Schltr.

Herb

Cr

135

151,559

Habenaria johannensis Barb.Rodr.

Herb

Cr

149

151,573

Oxalidaceae

Oxalis latifolia H.B.K.

Herb

Cr

229

151,653

Oxalis rhombeo-ovata A.St.Hil.

Herb

He

120, 191

151,545, 151,615

Passifloraceae

Passiflora suberosa subsp. litoralis (Kunth) K. Porter

Climber

Ph

65

151,490

Peraceae

Pera sp.

Tree

Ph

107

151,532

Piperaceae

Ottonia cf. leptostachya

Subshrub

Ph

117, 128, 211

151,552, 151,553, 151,635

Piperamalago L.

Shrub

Ph

74

151,499

Piper aduncum L.

Subshrub

Ph

32, 137

151,561, 151,457

Piper crassinervium Kunth

Shrub

Ph

66, 92

151,491, 151,517

Piper mollicomum Kunth

Subshrub

Ch

26, 30

151,451, 151,455

Piper sp.1

Shrub

Ph

112

151,537

Piper lucaeanum Kunth

Shrub

Ph

207

151,631

Poaceae

Axonopus polystachyus G.A.Black

Herb

He

166

151,590

Chusquea sp.

Herb

He

87, 143

151,512, 151,567

Parodiolyra micrantha (Kunth.) Davidse & Zuloaga

Herb

He

31

151,456

Polygalaceae

Polygala klotzschii Chodat

Subhrub

Ch

206

151,630

PolygalalancifoliaSt. Hil.

Herb

Ch

54

151,480

Primulaceae

Myrsineumbellata Mart.

Tree

Ph

72

151,497

Rhamnaceae

Gouaniainornata Reissek

Climber

Ph

103

151,528

Rosaceae

Prunus myrtifolia (L.) Urb.

Tree

Ph

172

151,596

Rubiaceae

Coccocypselum lanceolatum (Ruiz & Pav.) Pers.

Herb

He

142

151,566

Psychotria carthagenensis Jacq.

Shrub

Ph

35, 44, 105, 110, 129, 164, 169, 179, 216

151,530, 151,553, 151,593, 151,639

Randia armata (Sw.) DC.

Tree

Ph

126

151,550

Spermacoce verticillata L.

Herb

He

171

151,595

Rutaceae

Metrodorea stipularis Mart.

Tree

Ph

127, 193

151,551, 151,617

Salicaceae

Casearia sylvestris Sw.

Tree

Ph

196, 205

151,620, 151,629

Sapindaceae

Allophylus edulis (St. Hil.) Radlk.

Tree

Ph

236

151,660

Cupania vernalis Camb.

Tree

Ph

12, 13, 155

151,437, 151,438, 151,579

Serjania caracasana (Jacq.) Willd.

Climber

Ph

188

151,612

Serjaniafuscifolia Radlk.

Climber

Ph

154

151,578

Serjania mansiana Mart

Climber

Ph

200

151,624

Siparunaceae

Siparuna guianensis Aubl.

Tree

Ph

175

151,599

Solanaceae

Aureliana fasciculata (Vell.) Sendtn.

Tree

Ph

69

151,494

Capsicum flexuosum Sendtn.

Shrub

Ph

108

151,533

Cestrum cf. laevigatum Schltdl.

Tree

Ph

147

151,571

Cestrum schlechtendalii G. Don.

Shrub

Ch

140, 204

151,564, 151,628

Cestrum mariquitense Kunth

Shrub

Ch

43, 185

151,468, 15,169

Solanum argenteum Dun.

Shrub

Ph

146

151,570

Solanum didymum Dunal

Climbing shrub

Ph

139, 157

151,563, 151,581

Solanum megalochiton Mart.

Shrub

Ph

224

151,648

Solanum variabile Mart.

Shrub

Ph

64, 75

151,489, 1515

Solanaceae sp.

Tree

Ph

220

151,644

Urticaceae

Urera baccifera (L.) Gaudich.

Tree

Ph

67

151,492

Verbenaceae

Citharexylum myrianthum Cham.

Tree

Ph

133

151,557

Lantana camara L.

Shrub

Ph

10

151,435

Lantana fucata Lindl.

Shrub

Ph

2

151,427

Lantana hypoleuca Brig.

Shrub

Ph

97

151,522

Stachytarpheta cayennensis (Rich.) Vahl

Herb

Ch

78

151,503

Verbena cf. hirta Spreng.

Herb

Th

62

151,487

Verbena rigida Spreng.

Herb

Ch

61, 222

151,646

Violaceae

Hybanthus atropurpureus (A.St.-Hil.) Taub.

Subshrub

Ch

19, 37, 109, 114, 182

151,444, 151,462, 151,534, 151,539, 151,606

Valinhos municipality, São Paulo State, Brazil (22°57′24″S, 46°55′55″W)

Ph phanerophytes, Ch chamaephytes, He hemicryptophytes, Cr cryptophytes (geophytes), Th therophytes. You can download this table at: http://dx.doi.org/10.6084/m9.figshare.3114436

The biological spectra of the semideciduous forests were positioned within the variation found in spectra of Brazilian ombrophilous forests. The NMS ordination suggested the existence of four groups of biological spectra among the main Brazilian biomes (Fig. 2). Group 1 was formed by Amazonic and Atlantic semideciduous and ombrophilous forests, being characterized by the predominance of phanerophytes. Group 2 was formed by life-form spectra of the Caatinga growing in sedimentary terrains, with an intermediary position between the phanerophytic ombrophilous and semideciduous forests and the therophytic crystalline “Caatinga” woodlands. Group 3 included all woodlands of the “Caatinga” on crystalline terrains, in which the therophytic life-form was strongly predominant. Group 4 assembled all “Cerrado” savanna sites, with hemicryptophytes as the prevailing life-form (Fig. 2).
Fig. 2

Non-metric Multidimensional Scaling ordination of the life-form spectra for the main Brazilian biomes using Euclidian distance to compare the proportion of life-forms in each site. Group 1 includes Ombrophilous and semideciduous Atlantic forests and the Amazonic rainforest; Group 2 represents woodlands of the semiarid sedimentary “caatinga”; Group 3 encompasses woodlands of the semiarid crystalline caatinga; Group 4 includes “Cerrado” savannas of Central and Southeastern Brazil. Our study site in Valinhos is indicated in the plot. Raunkiaerian life-forms: Ph phanerophytes, Cr cryptophytes, He hemicryptophytes, Ch chamaephytes, Th therophytes

Discussion

As stressed by Raunkiaer (1934) and Cain (1950), biological spectra show a functional relationship between plant communities and the regional macroclimate. Benign climates allow most plants to assume phanerophytic life-forms, because plants can expose their buds to environmental circumstances without much protection, as they are not under harsh climatic conditions. When conditions are harsher, as in the “Cerrado” savannas with strong rain seasonality and fire recurrence (Eiten 1972; Ratter 1997), or in the “Caatinga” woodlands with a dry season lasting 6–11 months (Ab’Sáber 1974; Nimer 1989; Moro et al. 2016), life-forms with periodical reduction of aerial shoots become increasingly more important.

The different ecological factors to which each Brazilian biome is subjected were clearly reflected in their biological spectra. Moist forests (both Amazon and Atlantic ombrophilous and semideciduous forests) were strongly dominated by phanerophytes, reflecting a situation of climatically benign environment, which does not require from most plants special protection to their renewing buds (Raunkiaer 1934; Cain 1950). Harsher environments, on the other hand, had plant communities with a greater proportion of life-forms adapted to a periodical reduction of the aerial system. The “Cerrado” savannas, for example, had a high proportion of both phanerophytes and hemicryptophytes, also containing a considerable number of chamaephytes. Many of these plants (representing both phanerophytic, hemicryptophytic, and chamaephytic species), are nested within lineages that evolved locally in the Cerrado with specific attributes to survive the recurrent wildfires (Simon et al. 2009). Phanerophytes in the Cerrados are known to have thick, corky bark, resistant to fire, whereas many hemicryptophytes and chamaephytes have xylopodium and can resprout after fire has passed (Eiten 1972; Batalha and Martins 2002; Ribeiro and Walter 2008; Simon et al. 2009). Tolerating fire as a barky phanerophyte or promptly resprouting after fire as a hemicryptophyte are two contrasting strategies reflected in the biological spectra of the Cerrado savannas (Fig. 2; Batalha and Martins 2002).

Therophytes are expected to abound in environments where conditions are very harsh, and the time and amount of rainfall are unpredictable. These conditions are found in arid and semiarid regions, like in the Caatinga Domain which has a short and unpredictable rainy season and a dry season lasting 6–11 months (Ab’Sáber 1974; Andrade-Lima 1981; Nimer 1989; Pennington et al. 2000; Moro et al. 2016). In Northeastern Brazil, the “Caatinga” woodlands occur mainly in two geologically contrasting environments, namely the crystalline and sedimentary terrains (Velloso et al. 2002; Queiroz 2006, 2009; Moro et al. 2016). In the crystalline terrains, soils can be rich in nutrient, but are usually very shallow and stony (Velloso et al. 2002; Marques et al. 2014). Water is a strongly limiting resource because rainfall amount in the region is very limited as well as the capacity of the soil to retain water. In this environment, the therophytic life-form was predominant (Table 1; Fig. 2) and most species escape the dry season as resistant seeds, with their buds well protected within the seed’s teguments, thus rendering therophytes the most common life-form in this environment (Costa et al. 2007; Moro et al. 2016). “Caatinga” woodlands on sedimentary terrains, on the other hand, occur in areas of deep soils (Velloso et al. 2002; Marques et al. 2014) and were dominated mostly by phanerophytes (Table 1; Fig. 2). This seems to be related to a better soil water supply when compared to the crystalline Caatinga (Moro et al. 2016). However, we also observed relatively high proportions of therophytes and chamaephytes in the sedimentary “Caatinga”, which suggests a synergic actuation of both soil and climate structuring the plant communities (Moro et al. 2015). Therefore, potentially having a better soil water supply, sedimentary “Caatinga” woodlands show intermediate spectra between those of moist forests and crystalline “Caatinga” woodlands.

Although subjected to some degree of rain seasonality, with less rainfall than the Ombrophilous Forest (Eisenlohr and Oliveira Filho 2015), the semideciduous Atlantic forests had a very small number of therophytes. This seems to reflect the high annual rainfall amount, which is in average much greater than in the Caatinga areas (Nimer 1989; Instituto Brasileiro de Geografia e Estatística 2002). Moreover, in the Atlantic forest domain, the rainy season is much more predictable than in the “Caatinga”, as these areas have a higher rainfall amount and greater predictability in rain timing (Nimer 1989; Instituto Brasileiro de Geografia e Estatística [IBGE 2002)]. The spectra we found in the two semideciduous Atlantic forests analyzed are essentially the same found in the Ombrophilous Atlantic and Amazonic Rainforests. The Amazonic and Ubatuba sites had the greatest proportion of phanerophytes; therophytes being virtually absent. It seems that the Amazonic Domain, with plenty of water supply and minimum variation of annual temperature, favors the strong predominance of phanerophytes. The Atlantic forest Domain, on the other hand, stretches over wide latitudinal and altitudinal gradients ranging from equatorial to subtropical latitudes and from sea level to montane cloud forests, and while some areas receive large amounts of rainfall, other have seasonal climates (Oliveira Filho and Fontes 2000; Eisenlohr and Oliveira Filho 2015). Under these conditions, a wider range of biological spectra is expected for the Atlantic forest domain, in which other life-forms acquire importance, although phanerophytes are still predominant. This seems to be the case of our semideciduous forest in Valinhos, which had a spectrum similar to one of the sedimentary “Caatinga” sites. It is important to note that although the semideciduous forests are recognized to depart floristically from ombrophilous forests (Oliveira Filho and Fontes 2000), based on our data, we conclude that the semideciduous Atlantic forest is, to some extent, functionally similar to the wetter ombrophilous forest.

In general, the Raunkiaerian spectra of Brazilian biomes adjusted grossly to the expectation when considering the macroclimate of each biome (Dansereau 1957): moist Amazonic and Atlantic forests are dominated by phanerophytes, strongly seasonal “Caatinga” woodlands are very rich in therophytes, and fire-prone Cerrado savannas are plenty of hemicryptophytes. However, only “Caatinga” woodlands on crystalline basements are according to that scheme, whereas woodlands on sedimentary basements had a smaller proportion of therophytes, even though these areas are subjected to the same regional semiarid macroclimate as the Crystalline “Caatinga” sites. Sedimentary “Caatinga” spectra tended to be closer to the Valinhos spectrum than to other Crystalline “Caatinga” sites, indicating a great heterogeneity of survival strategies in the “Caatinga” domain and suggesting that soil conditions are very different between sedimentary and crystalline environments, harboring different plant communities in each ecosystem (Queiroz 2006; Gomes et al. 2006; Cardoso and Queiroz 2007; Moro et al. 2015, 2016).

According to Batalha and Martins (2002), a noteworthy pattern is that in world scale, a striking difference occurs between the biological spectra of “Cerrado” and other world savannas: “Cerrados” are dominated by phanerophytes and hemicryptophytes, whereas typical savannas are rich in therophytes (Batalha and Martins 2002). When compared to the biological spectra of other Brazilian biomes, the ones in “Cerrado” are distinguished by the predominance of hemicryptophytes. In fact, a high proportion of hemicryptophytes is found in the tundra, cold steppes, and cold-temperate forests, in which the occurrence of snow and frost seasonally kills aerial parts of the plants (Cain 1950; Batalha and Martins 2002). The hemicryptophytic phytoclimate that we found to be predominant in the “Cerrado” may be an indicative that recurrent fires could exert a similar “functional” role to seasonal snow and frost in killing aerial plant parts.

Seddon et al. (2016) has shown in a global analysis using remote sensing that the vegetation productivity of world biomes are modeled by different climatic variables. Alpine and tundra ecosystems showed higher sensitivity to temperature and cloudiness, while “Caatinga” and prairies had productivity strongly related to water availability. As far as rainforests’ productivity (like the Amazon) is concerned, it was found that their productivity were mostly related to cloudiness and water availability. In the same way that ecophysiological processes are modeled differently by regional climates (Seddon et al. 2016), we can expect that those climates also influence the structure of each ecosystem, which was shown here. From a continental scale we can see how climates influence plant communities’ structure by the life-form strategies that abound in each ecosystem. Wetter environments (ombrophilous and semideciduous forests), having larger supply of water and heat, are dominated by phanerophytes and the drier ones (the “caatingas”) are dominated by therophytes. But on a regional scale, within a particular macroclimate, other ecological constraints can be more relevant than variations in rainfall amounts to determine the vegetation physiognomy. The “Cerrado” savanna sites are expected to have their local spectra strongly influenced by local fire regimes while the “Caatinga” spectra were shown here to vary according to the edaphic environment, with “caatingas” in crystalline environments being structurally different from those growing in sedimentary environments. This reinforces the important roles of both climatic gradients, fire regimes, and edaphic features to model not only regional and local floristics (Santos et al. 2012; Oliveira-Filho et al. 2013; Eisenlohr and Oliveira Filho 2015; Neves et al. 2015), but also to model the life-form spectra of local ecosystems, as discussed here.

We present here a synthesis of life-form spectra for the main Brazilian biomes and add data for spectra in the seasonal semideciduous Atlantic forests. Our comparison suggests that macroclimate, wildfires (in the “cerrado”), and soil differences (within the semiarid Caatinga domain) are the main factors to determine the biological spectra. Although the Atlantic forest domain has very heterogeneous climate and relief, the biological spectra of the semideciduous forests presented here are nested within the range of spectra for other ombrophilous forests. Both semideciduous and ombrophilous forests were characterized by a strong predominance of phanerophytes, showing that ecological constrains in both ombrophilous and semideciduous forests do not impose periodical reduction of the plants’ aerial parts. This contrasts clearly with the “Cerrado” and “Caatinga” biomes, which are subjected to harsher conditions and where hemicryptophytes and therophytes, respectively, are much more relevant.

Acknowledgments

We are grateful to the owners of Clube de Campo Valinhos, for kindly allowing us to survey the vegetation in their property. We thank the staff of the Plant Science Department at Unicamp, who provided inestimable help to identify plant species during our survey. We are also appreciative of Daniel B. Umada’s valuable help during fieldwork. M.F. Moro thanks the São Paulo Research Foundation (Fundação de Amparo à Pesquisa do Estado de São Paulo—FAPESP 2013/15280-9) for a post-doctorate Grant awarded to him. We are grateful to Dr. Juliano Van Melis, Dr. Leandro Vieira, Dr. Bruno Aranha, and Dr. André Rochelle for kindly providing their inedited floristic data for Ubatuba and Ribeirão Cachoeira forests, allowing us to produce two extra spectra for our analyses. We also would like to thank Professor João Semir for helping to determine the life-forms of species reported on those sites. We thank Janete Silveira da Silva for reviewing the English of the text.

Compliance with ethical standards

Conflict of Interest

There are no conflicts of interest in this paper.

Funding information

Funder NameGrant NumberFunding Note
Fundação de Amparo à Pesquisa do Estado de São Paulo (BR)
  • FAPESP 2013/15280-9

Copyright information

© Botanical Society of Sao Paulo 2016

Authors and Affiliations

  • Ana Claudia de Meneses Costa
    • 1
  • Marcelo Freire Moro
    • 1
  • Fernando Roberto Martins
    • 1
  1. 1.Departamento de Biologia Vegetal, Instituto de BiologiaUniversidade Estadual de CampinasCampinasBrazil

Personalised recommendations