Ethnobotany, Availability, and Use of Lianas by the Kaingang People in Suburban Forests in Southern Brazil

Ethnobotany, Availability, and Use of Lianas by the KaingangPeople in Suburban Forests in Southern Brazil1

DEMETRIO LUIS GUADAGNIN*,2 AND ISABEL CRISTINA GRAVATO3

2Department of Ecology, Universidade Federal do Rio Grande do Sul, P.O. 15007, 91501-970, PortoAlegre, RS, Brazil

3Faculdade Anhanguera, Pelotas, RS, Brazil*Corresponding author; e-mail: [email protected]

Ethnobotany, Availability, and Use of Lianas by the Kaingang People in Suburban Forests inSouthern Brazil. Lianas are important components of tropical forests and also non-timberforest products (NTFPs). In this study we 1) record the species of lianas exploited byindigenous Kaingang people and their availability in the remnant forest patches around thecity of Porto Alegre, Southern Brazil, 2) describe the harvesting and crafting techniques, and3) estimate the amount of lianas sold as handicrafts in the main local market and the incomeobtained. Lianas are exploited exclusively for the handicrafts trade. Seven liana species, out of31 found in forest remnants, are used by Kaingangs. Two of them account for 90% of allhandicrafts sold and 40.5% of all liana stems in secondary growth remnants—Forsteroniaglabrescens Müll.Arg. and Amphilophium paniculatum (L.) Kunth. Another three species arescarce and are exploited less. Raw material must be used within a few weeks and becontinuously harvested. Kaingang harvesting strategy includes mechanisms to avoidoverexploitation: the division of exploitation areas into plots that are left to rest for oneyear before a new harvesting campaign, and the family’s control of information aboutpotential harvesting sites. The entire process, from collection to sale, is carried out by theKaingang families without intermediaries, contrary to the most common pattern of NTFPmarket chains. Trading occurs only locally. Each family earns from USD 98.15 to USD 371.12per month (30% to 115% of the minimum Brazilian salary). In one year each familyconsumed on average 10,328.40 meters of lianas in the manufacture of handicrafts for sale inthe major fair in the city, which represents about 50% of the demand for lianas. Weestimated that one hectare of unexploited forest can have on average a standing crop of10,165.40 m of the most desired species. Market pressure regulates exploitation, a commonpattern in NTFP use, but in this case leading to a mix of usually oppositetrends—manufacture with increasingly sophisticated techniques, but using a few lianaspecies and an increasing number of alternative materials.

Etnobotânica, Disponibilidade e Uso de Lianas por Indígenas Kaingang no Sul do Brazil.Lianas são componentes importantes das florestas tropicais e também Produtos FlorestaisNão-Madeireiros (PFNM). Neste trabalho registramos as espécies de lianas exploradas porindígenas Kaingang e sua disponibilidade em remanescentes florestais nos arredores dacidade de Porto Alegre, Sul do Brasil; descrevemos as técnicas de coleta e manufatura; eestimamos a quantidade de lianas vendidas como artesanato no principal mercado local e osganhos obtidos. Lianas são exploradas exclusivamente para o comércio de artesanato. Seteespécies de lianas, dentre 31 encontradas nos remanescentes florestais, são utilizadas pelosKaingangs. Duas delas respondem por 90% de todo o artesanato vendido e 40,5% daabundância de lianas em floretas secundárias—Forsteronia glabrescens Müll.Arg. eAmphilophium paniculatum (L.) Kunth. Outras três espécies são pouco exploradas etambém raras. A matéria prima precisa ser processada dentro de poucas semanas e sercontinuamente obtida das florestas. A estratégia de coleta dos Kaingang inclui mecanismospara evitar a sobre-exploração: a divisão das áreas de exploração em parcelas deixadas em

1 Received 18 November 2012; accepted 15October 2013; published online 8 November 2013.

Economic Botany, 67(4), 2013, pp. 350–362© 2013, by The New York Botanical Garden Press, Bronx, NY 10458-5126 U.S.A.

descanso por um ano entre campanhas de coleta e o controle familiar das informações sobreáreas potenciais de coleta. Todo o processo, desde a coleta até o comércio, é feito pelasfamílias Kaingang sem envolvimento de intermediários, contrariamente ao padrão maiscomum das cadeias de mercado de PFNM. O comércio é apenas local. Cada família obtém deUSD 98,15 a USD 371,12 por mês (30% to 115% do salário-mínimo brasileiro) com a vendade artesanato de lianas. Em um ano cada família consumiu em média 10.328,40 metros delianas na confecção de artesanato para venda na principal feira da cidade, o queprovavelmente representa cerca de 50% da demanda por lianas. Nós estimamos que umhectare de floresta não explorada pode apresentar um estoque de 10.165,40 m das espéciesde lianas mais desejadas. O mercado regula a exploração, um padrão comum no uso dePFNM, mas neste caso combinando duas tendencies normalmente opostas—a manufaturacom técnicas progressivamente sofisticadas, mas empregando poucas espécies e um númerocrescente de materiais alternativos.

Key Words: Lianas, vines, extractivism, exploitation, harvesting, traditional uses, indigenouspeople, handicrafts, non-timber forest products.

IntroductionLianas, or woody vines, are important structural

components of tropical forests (Campanello et al.2007; Schnitzer and Bongers 2002), as well asnon-timber forest products (NTFP) used formultiple purposes around the world. While someliana species grow fast and can even becomeoverabundant in secondary or degraded forests(Schnitzer and Bongers 2002; Tabanez and Viana2000), other species may be overexploited(Escalante et al. 2004; Martinez-Romero et al.2004; McGeoch et al. 2008; Muhwezi et al. 2009;Plowden et al. 2003). Abundance, assemblagecomposition and richness, and dominance ofclimbing mechanisms vary across tropical forestsaccording to the characteristics of tree stands, suchas the density of adequate climbing trees(Campanello et al. 2007; Van der Heijden andPhillips 2008). Tendril climbing lianas tend to bethe most common climbing mechanism insecondary forests (Boom and Mori 1982; Putz1984; Putz and Chai 1987), while scramblers aremore adapted to trees of greater height (Engel et al.1998).

The harvesting of NTFPs has been a matter ofconcern in economic and biodiversity conservationarenas (Belcher et al. 2005; Kusters et al. 2006;Stanley et al. 2012; Ticktin 2004). NTFPexploitation can either pose a threat to biodiversity,due to overexploitation or cascade ecological effectson biological communities and ecosystems, or canfoster biodiversity conservation, while providingincome from forests without clearing them.Communities usually exploit resources accordingto their local abundance and accessibility (Martinez-Romero et al. 2004; Thomas et al. 2009a, b).However, when pressured either by scarcity due to

deforestation (Arnold and Ruiz-Pérez 2001) or bymarket demands (Stanley et al. 2012; Ticktin2004) this may lead to overexploitation.

NTFP exploitation can either represent a safetynet or a poverty trap to poor people (Angelsenand Wunder 2003; Stanley et al. 2012).However, there is scarce information about theimportance of NTFP to the household economyand security (Jensen and Meilby 2010) and theoptions to improve economic viability and gains.The degree of families’ dependence on NTFPincome (“degree of forest dependence,” Stanley etal. 2012) is highly variable, ranging from 10% to65% of the total household income according toBelcher et al. (2005). Moreover, differentcommercial chains, production options, andmarket value of products can have importanteffects on the ecological and economicsustainability of exploitation (Angelsen andWunder 2003; Belcher and Schreckenberg2007; Boot and Gullison 1995). Lianas areamong the least studied NTFP resources on bothecological and economic grounds (Stanley et al.2012).

In Southern Brazil, Kaingang indigenouspeople depend on the sale of handicrafts madeof lianas and other materials as an importantsource of income (Bloemer and Nacke 2008;Freitas 2005). Plastic materials bought in cityshops are also used. Some indigenous families areliving today in major cities, such as Porto Alegre(1.5 million inhabitants), where they exploit lianastems and seeds in private and public suburbanforest fragments and in protected areas (Freitas2005; Garlet and Bellini 2009). Around 63Kaingang families and 400 individuals arecurrently living in Porto Alegre. Handicrafts are

351GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

sold in two open markets, one during the week andthe other on Sundays. The difficulties of access toforests and the existence of fewer and smallerremnants in a large, growing city lead toexploitation being concentrated in a few areas. Asa consequence, conflicts with conservationauthorities and NGOs have arisen in recent yearsdue to concerns about the sustainability ofexploitation and rights of access to collecting areas(COMIN 2008; Santa Maria 2008).In this study we (1) record the liana species

used by Kaingang people in Porto Alegre, (2)describe the harvesting of lianas and craftingtechnique, (3) estimate the amount of lianas usedin handicrafts sold in the main local open market,(4) estimate the income obtained, (5) quantifythe abundance and distribution of lianas insuburban forest fragments, and (6) associate theabundance of lianas with characteristics of treestands.

MethodsETHNOBOTANY OF LIANA HARVESTING AND

HANDICRAFT MANUFACTURING

We approached Kaingang families through theintermediation of the Nucleus of Public Politicsfor Indigenous People of the MunicipalPrefecture of Porto Alegre (NPPPI/PMPA), andidentified seven informants based on the timethey have been living in the city, leadership in thecommunity, and knowledge of harvesting andmanufacturing of liana artifacts. After a period ofapproach, the informants indicated an area theywere currently harvesting, a suburban forestremnant in Porto Alegre (30º03'S, 51º07'W).In this area we carried out participant observationof the harvesting and manufacturing processes,exploring all aspects related to this activity. Weobserved and asked questions about the speciesused, their usefulness, criteria for selecting plantsto harvest, and methods of cutting, stocking,preparing raw material, and assembling pieces.Voucher specimens of all species collected ormentioned as usable were deposited in theherbarium of the Centro Universitário Metodistaof the Instituto Porto Alegre, in Porto Alegre.

MARKETING OF LIANA HANDICRAFTS AND CASH

INCOME

Kaingang families sell handicrafts directly tobuyers in flea markets or streets. We monitoredone of the two marketplaces in Porto Alegre—an

open-air fair of handicrafts and antiquities thattakes place on Sundays, locally known as “Briqueda Redenção” (Freitas 2005; Fig. 1). The othermarketplace is a downtown walkway run by thesame families during the week. We identified 13families that sell liana handicrafts in this marketand randomly selected three of them each Sundayto record all the sales of that day. We observedfrom a short distance and with caution in ordernot to interfere in the trading dynamics. Werecorded the type and number of articles sold. Wealso bought samples of all pieces that wereweighed and disassembled to measure total lengthof lianas used and to identify the species used. Wemonitored selling twice a month for one year,from September 2008 to August 2009. In 2011we checked current prices of pieces sold toestimate the income obtained. We used monetaryvalues applicable in March 2011 in all analyses:minimum Brazilian wage of BRL 545.00,corresponding to USD 328.31. With these datawe estimated the annual demand for lianas andthe average annual income obtained by eachfamily.

LIANA AVAILABILITY AND STAND

CHARACTERISTICS

In order to understand the abundance of theliana species exploited by Kaingangs, we carried outa phytosociological study in four private forestpatches in different areas around the city. Allpatches lack recent historical records of lianaexploitation due to access restrictions by thelandowners, but are near and representative of theareas currently exploited. All patches are located inhill slopes and correspond to a single foresttype—semideciduous Atlantic Forest. The climateis subtropical, humid, with clearly defined seasonsand temperatures ranging on average from 25°C to35°C in summer and 2°C to 15°C in winter.Litholic and red-yellow podzolic soils predominate.At each site we randomly placed five plotsmeasuring 10x10 m. We excluded areas <50 mfrom forest edges and areas with >45º of declivitybecause the indigenous people avoid these areas (seebelow). We followed the protocol for liana censusesproposed by Schnitzer et al. (2008), recording thenumber and species of all lianas stems withdiameter > 1 cm at 130 cm from the roots. Werecorded all tree trunks with diameter >10 cm at130 cm above ground. We measured DBH withpachymeters and dentrometers. Height of trees was

352 ECONOMIC BOTANY [VOL 67

estimated visually by an experienced botanist. Weclassified the climbing mechanisms of lianas intoone of the following categories, adapted from Putz(1984): (a) tendril climbers, (b) scramblers,sprawlers, and hooked or forked climbers, and (c)twiners and branch twiners. We plotted thenumber of species and number of individuals ineach climbing category.

Because Kaingangs exploit stems with less than1.0 cm of diameter, we recorded the abundanceof all liana stems of the three most exploitedspecies in 20 plots of 5x5 m. In these plots wealso collected all stems of the most used lianaspecies, simulating the harvest strategy employedby Kaingangs (see Results). We measured thelength of all harvested stems to estimate the totalstanding crop of lianas that can be produced perunit area. We excluded from counting andharvesting young sprouts with length <1.0 m ordiameter <0.2 mm, because they are not suitablefor making handicrafts.

We estimated the density of liana species as thenumber of stems/hectare, pooling the data of thefive plots at each forest patch. In order to test thehypothesis that the species used by indigenouspeople are widespread in the forest remnants, wetested whether the spatial distribution fits aPoisson distribution. We assessed the pattern of

spatial distribution at plot level by calculating theindex of dispersion (I = variance/mean) of thenumber of stems/plot and the significance ofdeviation from a random pattern by computingthe T-statistic = (2I)0.5 – (2n-1)0.5, where n =number of plots (Pielou 1974). This statistic fallsbetween -1.96 and +1.96 when the distribution israndom with a 95% confidence interval.

In order to analyze whether the totalabundance of liana species and the abundance ofeach of the three most exploited liana speciesdiffer according to the forest characteristics, we fita linear mixed effect model with forestcharacteristics as fixed effects and forest sites asrandom effect. We ran models with allcombinations of the three fixed effects, and thebest structure was chosen based on the AkaikeInformation Criterion. Forest descriptors were logtransformed. We used the lme4 package (Bates2013) in the software R 2.10.1 (R Core Team2013). Species names and authorship werechecked and validated according to The PlantList (2013).

ResultsSPECIES, STRATEGY OF HARVESTING, AND USES

Seven species were collected by Kaingangs onthe trips in which we took part (Table 1), four of

Fig. 1. Flea market (“Brick da Redenção”) where liana handcrafts are sold by Kaingang people in Porto Alegre,Southern Brazil.

353GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

Bignoniaceae and one each from another threefamilies. Five species were tendril climbers, onewas a scrambler, and one was a twiner.The strategy used by the Kaingang people to

harvest lianas includes the following activities: (1)choice of an area, based on previous knowledge ofinformation from family members and relatives;(2) division of the area into plots, based onlandmarks; (3) scrutiny of plots in search oflianas; and (4) harvest. Forest edges are avoidedbecause they usually are too harsh and difficult tomove in. All activities are family based and eacharea is exploited by a single group, according totheir social, basically family relationships. Familieschoose areas for harvesting considering thedistance from villages, accessibility, andavailability of desired species. The division intoplots helps them to remember which areas wereexploited or left for future harvesting, thuspreventing a single area from being exploitedbefore it has been “rested” for at least one year.Some plants may be left unharvested, accordingto perceived demand for different species andhandicrafts at the moment. The harvestingactivities stop either when the family considersthat they have enough raw materials formanufacturing before it decays, according toperceived demand, or when the area has beentotally exploited. New harvests occur when thefamily runs out of raw material.Once a liana is found, harvesters drag it by

parts on the ground and cut the stems at about 20cm from the base (Fig. 2a). They pull the lianasfrom trees or from the ground until they rip, or

cut them at the maximum possible distancewithout climbing trees. Lianas of any diametermay be collected depending on the intended useand provided they have sufficient elastic resilienceto resist tensions for making handicrafts. Lianastems with lengths >1 m and diameters >0.2 mmare usually worth harvesting. Pieces are arrangedinto bundles about 120 cm in diameter (Fig. 2b)to facilitate transport, stocking, and drying.Handicrafts are produced only for the sale of

decorative items. The manufacturing ofhandicrafts starts after a short period of drying,either in the shade in hot or dry periods, or in thesun in cold or wet periods. There is a correct levelof drying, enough to prevent decay, but beforethe material becomes too hard to handle. Whenthey are too dry, the pieces are left outsideovernight to dampen with dew. The raw materialcan be workable for roughly a week. Lianas can beused either intact or splintered into thin lashes,depending on the intended use. Thin stems, orthe thin distal parts of the stems, usually ofForsteronia glabrescens Müll.Arg., are used formanufacturing small, delicate pieces, such asChristmas trees, sideboard dishes, and fruit bowls(Fig. 1). Coarse pieces, such as baskets, use stems>1.0 cm and are mainly made of A. paniculatum(L.) Kunth, Pithecoctenium crucigerum (L.) A.H.Gentry, and Serjania meridionalis Cambess.(Fig. 1). When a certain species is not availablein a family extraction area, the family groupsexchange raw material. Branches of trees, trunksof saplings, or small pieces of wooden boardscomplete the structure of some large pieces.

Table 1. LIANA SPECIES EXPLOITED IN PERI-URBAN FORESTS AROUND PORTO ALEGRE (SOUTHERN BRAZIL) BYKAINGANG PEOPLE FOR HANDICRAFT MANUFACTURING.

Kaingang name Brazilian common name Scientific name FamilyClimbingmechanism

Mrũr-kuxum Cipó marronzinho Forsteronia glabrescens Müll.Arg. Apocynaceae TendrilMrũr-tar Cipó-de-cesto Amphilophium paniculatum

(L.) KunthBignoniaceae Tendril

Mrũr-ga Cipó batata-de-morcego

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

Bignoniaceae Tendril

Mrũr-roj; Mrũr-mãréro

Cipó pente-de-macaco

Pithecoctenium crucigerum(L.) A.H. Gentry

Bignoniaceae Tendril

Mrũr-ger; Mrũrtãig

Cipó cravo Tynanthus elegans Miers Bignoniaceae Tendril

Mrũr-monh Cipó-olho-de-boi Dioclea violacea Benth. Fabaceae TwinerMrũr-gr Cipó casca grossa Serjania meridionalis Cambess Sapindaceae Tendril

354 ECONOMIC BOTANY [VOL 67

MARKETING OF LIANA HANDICRAFTS AND CASH

INCOME

Consumption of lianas, as well as manufacturingand selling of handicrafts, varied greatly frommonth to month. In one year each family consumedon average 10,328.40 meters of lianas in themanufacture and sale of handicrafts (Fig. 3),ranging from 524 m in April to 1,820 m inNovember. F. glabrescens was the material mostemployed, accounting for 62% of total length oflianas, followed by A. paniculatum (27.9%) and S.meridionalis (5.3 %). In one year each familyearned an average income of BRL 3,197.50 (USD1,926.20). Liana consumption per family variedfrom 624 meters (Jan.) to 1,820 meters (Nov.),and income from BRL 156.00 to BRL 589.00(USD 93.98 to USD 354.82) in the same months.Peaks of extraction and sales were near Christmasand Easter, and lows during the summer and

winter vacation periods. The average lianaconsumption per family/month was 860.7 m andthe average income BRL 266.5 (USD 160.54),corresponding to a daily household income fromNTFP of BRL 8.88 (USD 5.35).

LIANA AVAILABILITY AND STAND

CHARACTERISTICS

We recorded 380 stems of 31 liana species inthe studied plots, including six of the sevenspecies exploited by Kaingang people (Fig. 4).The density of lianas in the forest stands averaged1,900 stems/hectare (standard deviation ± 332).Seven species dominated the liana community,with over 10 stems in the set of 20 plots. Three ofthe exploited species were among the mostabundant and frequent types: F. glabrescens (600stems/ha; 80% frequency), S. meridionalis (165st./ha; 50% freq.), and A. paniculatum (140 st./

Fig. 2. (a) Kaingang harvester Mr. João Padilha cutting a liana stem at the base. Note the dragged liana aroundhis body. (b) Beam of liana ready for transporting.

Fig. 3. Total length of lianas used monthly on average by one Kaingang family in the handcrafts sold at the“Brick da Redenção” flea market in Porto Alegre, Southern Brazil, from September 2008 to August 2009.

355GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

ha; 40% freq.). The other three exploited species aresomewhat rare: Tynanthus elegans Miers (25 st./ha;15% freq.), Dolichandra unguis-cati (L.)L.G.Lohmann (15 st./ha; 15% freq.), and Diocleaviolacea Benth. (10 st./ha; 10% freq.). We did notrecord P. crucigerum A.H. Gentry in the plots. OnlyF. glabrescens, A. paniculatum, and S. meridionaliswere found in sufficient abundance for analysis.Most species were tendril climbers (54.8%),

including the exploited types (66.7%). Twinersrepresented 25.0% and 16.7%, respectively, ofthe total and exploited species, while scramblerscomprised the least represented group (19.4%and 16.7%, respectively). When we take intoaccount the abundance of stems, 75.8% of allstems are tendril climbers, and 49.2% were fromexploited species, a figure influenced by the highlevel of abundance of F. glabrescens.

F. glabrescens exhibited a random distribution(ID = 10.1; t = -1.8; n = 20), while the other sixspecies with at least 10 recorded stems allexhibited an aggregated pattern, including S.meridionalis and A. paniculatum.We recorded 224 individual trees of 64 species

in the studied plots, 48 of them used by lianas assupport for growing. The density of trees in theforest stands averaged 11.2 individuals/100m2

(standard deviation ± 6.25) and a mean basalarea of 85.4 cm2 (± 45.75). Mean height of treeswas 10.3 meters (± 8.70).Mixed model analysis showed that the

abundance of lianas varied according to standcharacteristics (Table 2). Stands with lowernumber of trees and lower mean height of treeshad a higher total number of lianas and highernumber of F. glabrescens and S. meridionalis. The

Fig. 4. Total number of stems (N) of liana species accumulated over 20 plots of 10x10 m in subtropical, peri-urban forests around Porto Alegre City, Southern Brazil. Black bars denote species exploited by Kaingang people inthe area. ni = unidentified species.

356 ECONOMIC BOTANY [VOL 67

mean basal area of trees was also negativelyassociated with the total number of lianas andnumber of F. glabrescens. A. paniculatum showeda distinct pattern, with abundance positivelyassociated with the mean height of trees.Variation among the forest sites explained 40.5% and 39.7%, respectively, of the variation in thenumber of A. paniculatum and S. meridionalis,and had negligible effect on the number of F.glabrescens and the total number of lianas.

The standing crop of stems of the three mostexploited species in the plots suitable for use inhandicrafts ranged from 3,496 m to 17,244 mper hectare. The standing crop of stems of F.glabrescens reached a total of 3,800 m per hectare,

while the other two most used species showedconsiderable lower availability—a total of around500 m of S. meridionalis and 770 m of A.paniculatum per hectare (Table 3).

DiscussionThe Kaingang exploit a few liana species, 7

according to our collaborators, out of 31 speciesin the studied plots. Up to 47 species can befound in forest remnants in the region (Backes1981; Knob 1978). Two of the exploited speciestogether accounted for 90% of all handicraftssold—F. glabrescens (62%) and A. paniculatum(27.9%). These two and S. meridionalis are themost abundant and frequent types in the studied

Table 2. RELATIONSHIP BETWEEN CHARACTERISTICS OF FOREST STANDS (FIXED EFFECTS) AND THE NUMBER OFLIANAS IN SUBTROPICAL, PERI-URBAN FORESTS AROUND PORTO ALEGRE. THE SPECIES MENTIONED ARE EXPLOITED

BY KAINGANG PEOPLE IN THE REGION. SE = STANDARD ERROR; T = T VALUE (MIXED LINEAR MODEL).

Estimate SE t

Total number of liana stemsIntercept 342.47 86.40 3.96log(total number of trees) -138.10 46.03 -3.00log(mean height of trees) -139.06 45.77 -3.04log(mean basal area of trees) -31.45 8.44 -3.73

Number of stems of Forsteronia glabrescensIntercept 219.14 45.26 4.842log(total number of trees) -91.46 24.13 -3.790log(mean height of trees) -98.96 23.73 -4.169log(mean basal area of trees) -17.58 5.12 -3.432

Number of stems of Serjania meridionalisIntercept 53.78 27.35 1.966log(total number of trees) -26.38 14.23 -1.854log(mean height of trees) -28.39 14.47 -1.962

Number of stems of Amphilophium paniculatumIntercept -1.01 1.52 -0.668log(mean height of trees) 2.50 1.45 1.721

Table 3. NUMBER AND LENGTH OF LIANA STEMS OF THE THREE MOST EXPLOITED SPECIES BY KAINGANGS IN

SUBTROPICAL, PERI-URBAN FORESTS AROUND PORTO ALEGRE, SOUTHERN BRAZIL, IN 20 PLOTS OF 5X5 M (500 M2),

EXCLUDING SPROUTS WITH LENGTH <1 M OR DIAMETER <0.2 MM.

Species Total no. of stems

Length of stems

Mean ± EP Minimum Maximum Total

Forsteronia glabrescens 66 8.0 ± 0.54 2.3 24.7 380.2Serjania meridionalis 17 5.6 ± 0.45 2.3 8.4 50.8Amphylophium paniculatum 12 6.4 ± 0.54 4.1 10.0 77.2

357GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

areas and are also common in other forestremnants in the region. The other four speciesare less used and less abundant. Therefore,exploitation tends to follow the availability ofresources. Indigenous people in other countrieswho also use NTFP for basketry also tend toexploit resources according to their availability inforests (Martínez-Romero et al. 2004).The Kaingang harvesting strategy includes

mechanisms to avoid overexploitation, such asthe division of exploitation areas into plots, restperiods between harvest campaigns, and familycontrol of information about profitable harvestingstands. According to Freitas (2005), the mostprofitable stands inside forest remnants, usuallyup to about three hectares, are kept under controlby a single family group by keeping secret theinformation about their location only among thefamily members and avoiding been seen by othergroups when going to those stands. Accessibilityof sites to people seems to affect both theperception of usefulness of sites and theanthropogenic pressure on resources (Thomas etal. 2009a, b). In the Porto Alegre area, theaccessibility of protected areas was at the root ofthe conflicts. Although most extraction occurs inprivate lands, a few indigenous forests and publiclands not designated as protected areas are alsoexploited (Freitas 2005).Collection of lianas as well as production and

marketing of handicrafts are all carried out by theKaingang families. No intermediaries areinvolved, contrary to the most common patternof other marketing chains (Arnold and Ruiz-Pérez2001; Martinez-Romero et al. 2004; Ticktin2004). Trading is only done locally, in fairs andstreets. Handicrafts made of lianas sold in the fairthat was studied guarantee each family an incomeof between USD 93.97 to USD 354.82 permonth, or roughly 28.62% to 108.07% of theBrazilian minimum wage. In comparison, theexploitation of scapes of Syngonanthus nitens(Bong.) Ruhland in Brazil, a more complex,organized chain focused on tourism, results inearnings of USD 1,800/year per artisan, or 1.5times the Brazilian minimum wage at the time(Schmidt et al. 2007). The daily return of BRL8.8 (USD 5.35) to the household is above theinternational threshold standard for absolutepoverty of USD 2/per day/ per capita (Stanleyet al. 2012). Poor families in Brazil receivegovernment support (“Bolsa família”) rangingfrom BRL 32.00 to BRL 242.00 (USD 19.28

to USD 145.78), depending on the total familyincome and number of children. The averagevalue is about BRL 66.51, or USD 40.07 (seehttp://www.mds.gov.br/bolsafamilia). A lownumber of Kaingangs have formal employmentand are not involved in handicraft trade. Othernon-monetary assistance may be received fromdifferent sources, such as food, clothes, medicines,educational material, and construction material.These figures would suggest a degree of economicdependence on forest resources of roughly 80%.It should be stressed that we are providing

minimum figures of extraction, income, andeconomic forest dependence, since we monitoredonly the most important fair. According toKaingang informants, this fair accounts for abouthalf of their incomes from handicrafts, but wewere unable to check this information. Thesefigures suggest that the exploitation of lianas is animportant economic supplement for Kaingangpeople. Poorer people tend to be more dependentupon the NTFP extraction, and also morevulnerable to scarcity of resources, regulation(Ticktin 2004), and the effects of deforestation(Arnold and Ruiz-Pérez 2001).T h e r e a r e t w o m o d e l s o f l i a n a

exploitation—traditional, for self-consumption,and commercial (Dufour 1990; Martinez-Romero et al. 2004). The exploitation of NFTPdoes not need to be commercial in order to havenegative impacts on plant population viability(Thomas et al. 2009a), but the greatest concernabout overexploitation is the increasing demandfor trade (Ticktin 2004). Martínez-Romero et al.(2004) found signs of overharvesting of lianas inMexico, leading to the local scarcity of resourcesin communities where extraction was mainly forcommerce, in opposition to those that exploited itbasically for self-consumption. La Torre-Cuadrosand Islebe (2003) observed that the mostexploited forests for extraction of NTFP werethose that provided more resources for externaldemands. Muhwezi et al. (2009) recorded athreefold expansion in the use of lianas in Africain tea production systems when compared withthe amount needed for traditional uses. Weobserved only one purpose for the extraction oflianas by Kaingangs—the production ofhandicrafts for sale. Unlike other reported cases,however, Kaingang trading is small and local. Theseasonality of exploitation depends on marketdemands, as also noted by Martinez-Romero etal. (2004). We recorded the use of alternative

358 ECONOMIC BOTANY [VOL 67

materials in handicrafts, such as wooden boardsand other vegetable materials, as well as the sale ofhandicrafts made of plastic beads. Peasantcommercial exploitation of NTFP pressured bythe market can induce changes in extraction andmanufacturing techniques (Arnold and Ruiz-Pérez 2001; Dufour 1990). Muhwezi et al.(2009) recorded the substitution of vines bybamboo due to local scarcity.

Another aspect of the exploitation model wediscovered was the use of only a few species.Kaingang informants say they use all speciesadequate to make handcrafts. According toMartínez-Romero et al. (2004), indigenouscommunities with fewer resources tend to exploitmore species, use more sophisticated techniques,and make more elaborate crafts, increasing theirdemand and prices. Ruiz-Pérez et al. (2004), onthe other hand, maintain that commercialexploitation leads to specialization and the useof a lower number of species exploited in largerquantities. Our observations suggest that in thecase of the Kaingang, market pressure has led to amix of the trends observed—manufacture withincreasingly sophisticated techniques, but using alimited number of liana species and an increasingnumber of alternative materials.

The availability of lianas is associated withyoung, secondary forests. The abundance oflianas, including the exploited types, was greaterin the studied plots with a lower number, meanheight, and mean basal area of trees. Theexception was A. paniculatum, whose numberswere greater in forest stands of greater height. F.glabrescens, the most abundant species, showed arandom pattern of distribution in the studiedplots and its numbers were not influenced bydifferences among forests. Differences amongforest sites, however, influenced the numbers ofthe other two exploited species, whosedistribution is aggregated. These figures suggestthat whereas F. glabrescens is a widespread andcommon resource, the other exploited species arefound scattered in small clusters.

Six out of the seven exploited species are tendril-climbers. The dominance of the tendril climbingmechanism coincides with the pattern found inremnants in South Brazil (Tibiriçá et al. 2006).Tendril-climbers are most successful in theunderstory of intermediate successional stages andadapted to forest interior and forest gaps with alarge number of small diameter supports andsaplings of small trees (Gentry 1991), a condition

found in the studied plots that seems to becommon to forest remnants in the region. Lianaabundance tends to be higher in forests disturbedboth by natural or anthropogenic forces (Schnitzerand Bongers 2002; Van der Heijden and Phillips2008). The total number of lianas of these threemost used species estimated from samplesincluding diameters above 0.2 mm was at aboutthe same range of that estimated from samplesconsidering only diameters above 1.0 cm. This isprobably because the studied plots have no recentrecords of exploitation and so the liana communityis developed to maturity, with low numbers ofyoung, thin, and short stems.

In one year each family consumed on average10,328.40 meters of lianas in the manufacture ofhandicrafts for sale in the major fair in the city,which represents about 50% of the demand forlianas. We estimated that one hectare ofunexploited forest can have on average a standingcrop of 10,165.40 m of the most desired species,ranging from 3,496 m to 17,224 m. This suggeststhat a few hectares of secondary, previouslyunexploited forest stands around Porto Alegrecan have the standing crop needed to sustain thehandicraft production of one Kaingang family. Itis important to note that currently there is noinformation about the standing crop in forestspreviously exploited or the productivity of lianasin the forest stands around the city.

Severa l key aspects of l iana harvestsustainability and commerce by Kaingangsdeserve further investigation. The first key issueis the harvesters’ knowledge of plant ecology. Inthis case, only the above-ground parts areremoved during harvest, leaving roots and baseparts to resprout. Collectors argue thatresprouting is common, but this has not yet beeninvestigated scientifically. Exploitation byexperienced harvesters may support greaterharvest pressure than by less experiencedcollectors (Ticktin 2004). Second, there arecurrently no data about the time needed for aforest to recover the standing crop of lianas afterextraction and the standing crop in exploitedforest stands. If recovery is low, large areas may beneeded to sustain the demand for lianas, althougha few hectares may hold the abundance needed bya family. Accessibility to more areas would becritical in this case. Third, currently there are nodata about trends in the abundance anddistribution of the most desired species, marketdemands, and manufacturing techniques.

359GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

Muhwezi et al. (2009) categorized species usedin basketry according their availability and needof management. Common, widespread species,usually those that are common in disturbed sites,can be used sustainably without externallyimposed resource management. The same maybe true of species that are locally uncommon, butof limited use and subject to low-impactharvesting; this seems to be the situation for F.glabrescens. Exploitation of resilient species,usually associated with secondary forests, maylead to conflicts that can be managed—probablythe situation for the other two most heavilyexploited species, S. meridionalis and A.paniculatum. Concerns are critical in the case ofslow-growing species for which there is highdemand, usually from secondary and matureforests. All these classes can be found in theKaingang list of exploited species, all of themexploited from the forest interior in smallremnants. Ndangalasi et al. (2007) suggestedseveral management strategies for NTFPexploitation in Africa, including red-listing andenforcement of limitation on harvesting, marketapproaches through improvement of producerrevenue, alternatives to needs of local people,and plantations. Market approaches wereconsidered risky by the same authors, as theycould lead to overexploitation. In the case ofKaingang liana exploitation, several options couldbe put forward in case of scarcity, including aparticipatory approach that builds on theharvesters’ knowledge and existing managementpractices, community monitoring, and controlover less abundant and slow-growing species. Theuse of alternative materials could also beencouraged. In situ propagation and forestmanagement could also be options, but, atpresent, there is not enough traditional orscientific information to assist such strategies.Forest conservation through economic gains by

NTFP extraction has become a paradigm, as haveconcerns about overexploitation (Ticktin 2004).Our study shows that the harvest of lianas for themanufacture and marketing of handicraftsprovides a safety net of income for Kaingangpeople in South Brazil. Exploitation of lianas byKaingangs around Porto Alegre is associated withspecies abundance, and indigenous people havedeveloped their own strategies to regulateextraction. Only a few hectares of unexploitedforest might have the standing crop of lianasneeded to supply a family of extractors per year,

but information about the regeneration of stocksafter exploitation and the standing crop ofexploited forests is needed to best estimate theamount of forest required to withstand demand.In this study: 1) restrictions on forest accessconcentrates exploitation in a few areas; 2)exploitation is made only for marketing and isregulated by market demand; 3) exploited speciesare confined to the interior of small, secondaryforest remnants; and 4) several species occur inlow abundance and are clustered. These featurescan help guide the debate around the lianaexploitation by Kaingangs in Southern Braziland improve our knowledge concerning thecurrent forms of NTFP exploitation around theworld.

AcknowledgmentsThanks to Kaingangs Sr. João Padilha, Sra.

Iracema Rã Ga, and their families for teachingsand access to exploitation areas. Thanks to MariaCarmen Sestren Bastos and the Nucleus of PublicPolitics for Indigenous People of the Municipalityof Porto Alegre (NPPPI/PMPA) for stimulatingthe research and facilitating access to Kaingangs.Thanks to the research assistants Ilka Marques,Camila Delanhese, Maria Noeci Jung, EmersonCoutinho, Camila Condessa Wildner Patines,Carolina Sampaio, Caroline Moraes, Luiz FelipeForgiarini, Marcelo Paiva, and Marília Bravo forhelping in the data collection. This paper wasfully supported by the Conselho Nacional deDesenvolvimento Científico e Tecnológico(CNPq, Edital Verde-Amarelo, Grant 560226/2008-0). CNPq also provided research fellowshipto DLG (309298/2009-1).

Literature CitedAngelsen, A. and S. Wunder. 2003. Exploring the

forest—Poverty link: Key concepts, issues andresearch implications. Center for InternationalForestry Research, Jacarta. CIFOR OccasionalPaper No. 40.

Arnold, J. E. M. and M. Ruiz-Pérez. 2001. Cannon-timber forest products match tropicalforest conservation and developmentobjectives? Ecological Economics 39:437–447.

Backes, A. 1981. A flora do Morro do Coco,Viamão, RS. Iheringia. Série Botânica 27:27–40.

Bates, D. 2013. Linear mixedmodel implementationin lme4. R package version 3.1-10. http://

360 ECONOMIC BOTANY [VOL 67

cran.r-project.org/web/packages/lme4/vignettes/Implementation.pdf (June 2013).

Belcher, B. and K. Schreckenberg. 2007.Commercialisation of non-timber forestproducts: A reality check. Development PolicyReview 25:355–377.

———, M. Ruíz-Pérez, and R. Achdiawan.2005. Global patterns and trends in the useand management of commercial NTFPs:Implications for livelihoods and conservation.World Development 33:1435–1452.

Bloemer, N. M. S. and A. Nacke. 2008. Aprecarização de políticas públicas e suasrepercussões na auto-sustentação dos Kaingangdo Oeste Catarinense. Ilha–Revista deAntropologia 10:82–102.

Boom, B. M. and S. A. Mori. 1982. Falsificationof two hypotheses on liana exclusion fromtropical trees possessing buttresses and smoothbark. Bulletin of the Torrey Botanical Club109:447–450.

Boot, R. G. A. and R. E. Gullison. 1995.Approaches to developing sustainableextraction systems for tropical forest products.Ecological Applications 5:896–903.

Campanello, P. I., J. F. Garibaldi, M. G. Gatti, andG. Goldstein. 2007. Lianas in a subtropicalAtlantic forest: Host preference and tree growth.Forest Ecology and Management 242:250–259.

COMIN (Conselho de Missão entre Índios).2008. Entrevistas com Kaingang de PortoAlegre e São Leopoldo. In: Caderno daSemana dos Povos Indígenas do COMIN.Igreja Evangélica de Confissão Luterana noBrasil (IECLB), São Leopoldo.

Dufour, D. L. 1990. Use of tropical rainforest bynative Amazonians. BioScience 40:652–59.

Engel, V. L., R. C. B. Fonseca, and R. E.Oliveira. 1998. Ecologia de lianas e o manejode fragmentos florestais. Série Técnica IPEF12:43–64.

Escalante, S., C. Montaña, and R. Orellana. 2004.Demography and potential extractive use of theliana palm, Desmoncus orthacanthos Martius(Arecaceae), in Southern Quintana Roo, Mexico.Forest Ecology and Management 187:3–18.

Freitas, A. E. C. 2005. Mrur Jykre—A cultura docipó: Territorialidade Kaingang na margemleste do Lago Guaíba, Porto Alegre, RS. Ph.D.thesis in Social Anthropology, UniversidadeFederal do Rio Grande do Sul, Porto Alegre.

Garlet, M. and M. I. B. Bellini. 2009. FamíliasKaingang num espaço urbano. Pages 104–114

in Anais do 1o. simpósio sobre família:Políticas de atendimento. Universidade doSul de Santa Catarina, Tubarão.

Gentry, A. H. 1991. The distribution and evolutionof climbing plants. Pages 3–53 in F. E. Putz andH. A. Mooney, eds., The biology of vines.Cambridge University Press, Cambridge.

Jensen, A. and H. Meilby. 2010. Returns fromharvesting a commercial non-timber forestproduct and particular characteristics ofharvesters and their strategies: Aquilaria crassnaand agarwood in Lao PDR. Economic Botany64:34–45.

Knob, A. 1978. Levantamento fitossociológico daformação mata do Morro do Coco, Viamão,Rio Grande do Sul. Iheringia Série Botânica23:65–108.

Kusters, K., R. Achdiawan, B. Belcher, and M.Ruíz-Pérez. 2006. Balancing development andconservation? An assessment of livelihood andenvironmental outcomes of nontimber forestproduct trade in Asia, Africa, and LatinAmerica. Ecology and Society 11:art20.

La Torre-Cuadros, M. D. and G. A. Islebe. 2003.Traditional ecological knowledge and use ofvegetation in southeastern Mexico: A casestudy From Solferino, Quintana Roo.Biodiversity and Conservation 12:2455–2476.

Martinez-Romero, M. M., A. E. Castro-Ramirez,P. Macario, and J. C. Fernandez. 2004. Useand availability of craft vines in the influencezone of the biosphere reserve Sian Ka'an,Quintana Roo, Mexico. Economic Botany58:83–97.

McGeoch, L., I. Gordon, and J. Schmitt. 2008.Impacts of land use, anthropogenicdisturbance, and harvesting on an Africanmedicinal liana. Biological Conservation141:2218–2229.

Muhwezi, O., A. B. Cunningham, and R.Bukenya-Ziraba. 2009. Lianas and livelihoods:The role of fibrous forest plants in foodsecurity and society around BwindiImpenetrable National Park, Uganda.Economic Botany 63:340–352.

Ndangalasi, H. J., R. Bitariho, and D. B. K.Dovie. 2007. Harvesting of non-timber forestproducts and implications for conservation intwo montane forests of East Africa. BiologicalConservation 134:242–250.

Pielou, E. C. 1974. Population and communityecology—Principles and methods. Gordonand Breach, New York.

361GUADAGNIN AND GRAVATO: USE OF LIANAS BY KAINGANG IN BRAZIL2013]

The Plant List. 2013. The plant l ist .www.theplantlist.org (July 2013).

Plowden, C., C. Uhl, and F. A. Oliveira. 2003.The ecology and harvest potential of titica vineroots (Heteropsis flexuosa: Araceae) in theeastern Brazilian Amazon. Forest Ecology andManagement 182:59–73.

Putz, F. E. 1984. The natural history of lianas onBarro Colorado Island, Panama. Ecology68:1713–1724.

——— and P. Chai. 1987. Ecological studies oflianas in Lambir National Park, Sarawak,Malaysia. Journal of Ecology 75:523–531.

R Core Team. 2013. R: A language andenvironment for statistical computing. RFoundation for Statistical Computing, Vienna.Version 3.0.1. http://www.r-project.org/.

Ruiz-Pérez, M., B. Belcher, R. Achdiawan, M.Alexiades, C. Aubertin, J. Caballero, B.Campbell, C. Clement, T. Cunningham, A.Fantini, H. de Foresta, C. García Fernández,K. H. Gautam, P. Hersch Martínez, W. deJong, K. Kusters, M. G. Kutty, C. López, M.Fu, M. A. Martínez Alfaro, T. R. Nair, O.Ndoye, R. Ocampo, N. Rai, M. Ricker, K.Schreckenberg, S. Shackleton, P. Shanley, T.Sunderland, and Y. Youn. 2004. Marketsdrive the specialization strategies of forestpeoples. Ecology and Society 9:art4.

Santa Maria, R. M. 2008. O conflito entredireitos fundamentais: As dimensões doconflito entre o direito à moradia adequada eo direito ao meio ambiente ecologicamenteequilibrado no espaço urbano Brasileiro.Master’s thesis on Advocacy, Universidade doVale do Rio dos Sinos, São Leopoldo.

Schmidt, I. B., I. B. Figueiredo, and A. Scariot.2007. Ethnobotany and effects of harvestingon the population ecology of Syngonanthusnitens (Bong.) Ruhland (Eriocaulaceae), a

NTFP from Jalapao Region, Central Brazil.Economic Botany 61:73–85.

Schnitzer, S. A. and F. Bongers. 2002. The ecologyof lianas and their role in forests. Trends inEcology and Evolution 17:223–230.

———, S. Rutishauser, and S. Aguilar. 2008.Supplemental protocol for liana censuses. ForestEcology and Management 255:1044–1049.

Stanley, D., R. Voeks, and L. Short. 2012. Is non-timber forest product harvest sustainable in theless developed world? A systematic review of therecent economic and ecological literature.Ethnobiology and Conservation 1:1–39.

Tabanez, A. A. J. and V. M. Viana. 2000. Patchstructure within Brazilian Atlantic Forestfragments and implications for conservation.Biotropica 32:925–933.

Thomas, E., I. Vandebroek, and P. Van Damme.2009. Valuation of forests and plant species inindigenous territory and National Park Isiboro-Sécure, Bolivia. Economic Botany 63:229–241.

———, ———, ———, P. Goetghebeur, D.Douterlungne, S. Sanca, and S. Arrazola.2009b. The relation between accessibility,diversity and indigenous valuation ofvegetation in the Bolivian Andes. Journal ofArid Environments 73:854–861.

Tibiriçá, Y. J. A., L. F. M. Coelho, and L. C.Moura. 2006. Florística de lianas em umfragmento de floresta estacional semidecidual,Parque Estadual de Vassununga, Santa Rita doPassa Quatro, SP, Brasil. Acta BotanicaBrasilica 20:339–346.

Ticktin, T. 2004. The ecological implications ofharvesting non-timber forest products. Journalof Applied Ecology 41:11–21.

Van der Heijden, G. M. F. and O. L. Phillips.2008. What controls liana success inneotropical forests? Global Ecology andBiogeography 17:372–383.

362 ECONOMIC BOTANY [VOL 67