Forest Inventory and Quantification of Stored Carbon in the Bolivian Chaco

Ellen Arnstein, MEM 2013¹

Abstract

This research aimed to quantify forest carbon stocks for a payment for watershed services project in the Integrated Management Natural Area Rio Grande – Valle Cruceños, Bolivia. A tree inventory was conducted in Yumao, Bolivia, a transitional area between dry and Bolivian-Tucuman forest types, using eight 0.25 ha permanent plots to evaluate forest structure and floristic patterns and generate basic data about forest biomass and carbon sequestration. All trees with a diameter at breast height of at least 10 cm were sampled to provide information on their size (diameter and height), species, health, and coverage with woody vines; understory species were also measured and identified in sixteen subplots. This data was then used to calculate frequency and importance values at both the species and family levels. A total of 986 individuals, representing 52 tree species belonging to 24 botanical families, were identified. The forest was dominated by relatively few species, particularly Phyllostylon rhamnoides, Anadenanthera colubrina, Pisonia zapallo, Calycophyllum multiflorum, and Ruprechtia apetala. The estimation of biomass volume is essential to measure storage and flows of carbon in forest ecosystems. Since destructive measures to calculate biomass were not possible, the generalized allometric equations for dry forest developed by Brown (1997) and Chavé et al. (2005) were used with the result of biomass values of 48.12 t/ha and 61.93 t/ha respectively.

Introduction

Deforestation in the lowland plains of Bolivia is continuous and increasing primarily as a result of the expansion of agricultural activities which contribute not only to the loss of valuable plant and animal species but also to climate change; 80% of the greenhouse gas emissions in Bolivia are the result of land use and land cover change (UNFCC 2006). As a result, the incidence of extreme weather events that cause flooding or, conversely, reduced stream flow occur more frequently—a particular threat in the Rio Grande watershed. Given this problem, Fundación Natura Bolivia, which specializes in the development of financial mechanisms for conservation, has entered the voluntary carbon market to consolidate stocks of carbon in community forests. My research, therefore, aimed to take the necessary step of supplying the Foundation with biomass measurements with an application consistent with the Bolivian context. While various factors prevented measurements consistent with methodologies developed by organizations such as Winrock International, my research contributes to an improved understanding of the floristic patterns, composition, and forest structure of the transitional dry forest found in the study area of Yumao and offers a good starting point for further quantification of stored carbon.

Background

The mission of Fundación Natura Bolivia, formed in 1999, is to conserve critical ecosystems and improve the quality of life of Bolivian people through alternative financial mechanisms. Their central initiative is reciprocal water agreements where annual direct payments are made to farmers in return for forest conservation, ensuring the provision of water to downstream communities. The Foundation also offers compensation to farmers in the form of beehives and beekeeping training, fencing for their land, or fruit trees for every 10 hectares of protected forest. To date, more than 600 farmers have signed agreements to conserve more than 30,000 hectares of forest (Huayrana 2012).

Figure 1. The study area of Yumao. Yumao is located within the Rio Grande Integrated Management Natural Area (AMNI) of Santa Cruz, Bolivia. The AMNI is highlighted in grey and Yumao is in red.

The town of Yumao and the study area fall within the Integrated Management Natural Area Rio Grande – Valle Cruceños (Spanish acronym ANMI RG-VC). The ANMI RG-VC, located in the center of the Rio Grande watershed, was formed with the objective of conserving forests, lessening the impact of flooding and drought in the area, and protecting biodiversity while demonstrating the touristic potential of the valley’s unique culture and history (Figure 1). Within its 734,000 ha area, five large ecoregions are represented: Tucuman-Bolivian forest, inter-Andean dry forest, Chaco Serrano, Gran Chaco, and the Yungas (Ibisch et al. 2002). Previous vegetation studies have found 36 vegetation series composed of 2,415 individual plant species including 55 that are endangered (Fundación Natura Bolivia 2009). The ANMI includes the municipalities of Cabezas, Gutiérrez, Samaipata, Vallegrande, Postrervalle, Pucará, and Moro Moro; according to the 2001 Bolivian census, the area has a population of approximately 69,000 persons for whom the principal activity is agriculture and ranching (Instituto Nacional de Estadistica 2001).

Yumao, located within the Gutiérrez municipality, is home to approximately 20 families, of which one third are indigenous Guarani. Yumeños are primarily subsistence farmers and fishers. However, Gutiérrez municipality has developed a tourism plan to promote sport fishing, ecotourism, and historical and cultural attractions with the hope of diversifying the local population’s income, while contributing to the conservation of the natural patrimony that the region possesses.

There is some confusion as to the specific ecotype of the study area. According to one of the early biogeographies of the region, Yumao is characterized primarily as Chaco Serrano, with some traces of Gran Chaco (Cabrera and Willink 1980). The National Service for Protected Areas (SERNAP) describes the zone as a high-to-medium altitude mountainous area, and Navarro et al (2007) describe the site as a transitional area between dry forest and Bolivian-Tucumano. Yumao receives less than 1,500 mm of precipitation per year and experiences a dry season that lasts more than five months. In terms of soil types, FAO’s (2007) soil map classifies the area as BK2-C, a mixture of calcic cambisols and chernozems. Yumao covers approximately 6,000 ha but due to jurisdictional conflicts among capitanías (traditional Guarani land administration units) only about 1,600 ha are under conservation. It was with this area in mind that the study was designed.

Methodology

Sampling Methodology

Designed for silvicultural purposes, permanent sample plots are being used with increasing frequency to carry out qualitative and quantitative comparisons between study sites. They also offer the possibility to observe long-term changes in species phenology and forest dynamics. In this manner studies can show how forests respond to intervention; determine change in measured characters like trees’ diameter at breast height, basal area, and/or volume; establish mortality and recruitment; and update forest carbon sequestration amounts.

Following the protocol developed by BOLFOR (the Bolivian Forest Service, BOLFOR 1995, 1998, BOLFOR and PROMABOSQUE 1999), eight permanent sample plots of 50 x 50 m (0.25 ha each for a total of 2 ha) were installed in Yumao to measure trees, bushes, herbaceous plants, soil, and necromass. Rather than taking GPS points for every tree in the plot, we took a GPS point at each plot corner and divided the plot into 10 m by 10 m subplots. In each subplot, the tree locations were recorded as x, y distances from the subplot corner (Figure 2).

Figure 2. Layout of permanent plots. Each 50m x 50m plot was divided into 10m x 10m sub-plots. Point 0,0 is located in the south-west corner of the plot.

Regarding degraded forest, conflicts exist among the protocols developed by BOLFOR; specifically, their 1999 protocol states that the plot location should not be changed if they fall within poor-quality forest. However, in another publication they recommend increasing the distance between plots by a factor of 1 to 1.15 if the plot falls within degraded forests or is too near to the study area’s limit (BOLFOR 1995). The present study installed plots in forests that did not have evidence of burns or agriculture, increasing the distance between sites where necessary.

Measurement and Identification

Each tree with a diameter larger than 10 cm was marked with a numbered aluminum plaque, identified to species, measured (height, DBH) and characterized according to percent coverage by woody vines, health, canopy position, exposure to light, and crown quality. Dead trees were also measured but not marked.

In sub-plots one and twenty-five, located at the extremes of the plot, all the saplings and bushes measuring between 2.5 and 9.9 cm diameter were identified and their height and diameter were measured. We chose to locate the subplots as such because they would be in areas least disturbed from the process of delineating and measuring the plots. Vegetation samples were collected and pressed in the field. Each registered specimen was identified using morphological keys, the herbarium at the Museo Noel Kempff Mercado, and the expertise of Bolivian botanists.

Biomass Calculation

One approach to quantifying carbon biomass stores consists in inferring changes from long-term forest inventory plots. As such, regression models are used to convert forest inventory data such as height, DBH, and wood density as determined by tree species into an estimate of aboveground biomass (AGB). Ideally these allometric equations are developed specifically for the study area in question.

Unfortunately the dry forests of Bolivia, Paraguay, and Argentina lack specific regression equations, and developing them was beyond the scope of this study due to restrictions on harvest and time limitations. Instead, we chose to compare the most widely used generalized equations developed for dry forests:

(Brown 1997)

(Chavé 2005)

Differentiating the two equations is the fact that the Brown equation relies solely on the datum of DBH whereas Chavé’s equation includes the tree’s height and density as well.

Results

Forest Structure

The plots were distributed among flat, hilly and peak locations; some demonstrated past evidence of human intervention such as cattle pasture (Plot 3) and wood harvest (Plot 2). The canopy measured between 10 and 18 meters in height with a predominance of Phyllostylon rhamnoides, Anadenanthera colubrina, Pisonia zapallo, Calycophyllum multiflorum, and Ruprechtia apetala species.

The understory of the plots varied according to topography. For example, on the slopes vegetation was sparse and dominated by garabata (Bromelia serra) and cactus. In total, 52 tree species belonging to 24 botanical families were identified (for a full list see Appendix 1) of which Ulmaceae, Fabaceae, and Malvaceae have the highest importance values, based on their relative frequency, relative density, and relative dominance (Figure 3).

Figure 3. Importance value by family. Importance value is calculated by taking the average of the family’s relative frequency, relative density, and relative dominance. In Yumao’s forest Ulmaceae, Fabaceae, and Malvaceae have the highest importance values.

The incidence of new species became less and less frequent as the inventory was completed. This allowed us to assess the reliability of our inventory and could be used to extrapolate the total number of species that would be present in the entire study area. In terms of diameter, the structure of our sample site showed a reverse J curve, with a higher number of small-diameter trees in almost every plot (Figure 4).

Biomass and Stored Carbon

To calculate biomass according to Chavé et al.’s (2005) equation, which includes wood density, this study used Zanne et al.’s (2009) tables. Some regional species (e.g., negrillo, cuchimara, chichapi, lanza, athyana), which were not included in this source, were available from Argentina’s National Institute for Industrial Technology (INTI 2012). If a species did not have a recorded density, the average of the genera, the family, or the entire list (0.639 g/cm3) was used (Dauber et al. n.d.). There was no density data available for the Cactaceae family so they were removed from the calculations. Standing dead trees were measured using the same methodology but only taking 70% of the equation’s result to account for decomposition and the subsequent loss of biomass.

Biomass is an approximate measure of the carbon contained in trees with DBH greater than 10 cm. Specifically, the carbon is roughly 50% of the aboveground biomass (Brown and Lugo 1992, IPCC 2006). The plots in Yumao (2 ha total) had a biomass—for live trees, dead trees, and shrubs—of 495.42 t following Chavé et al. (2005), and therefore carbon content of 123.86 t/ha) and 384.97 t following Brown (1997) (or 96.24 t/ha of carbon). Biomass was also calculated for herbaceous plants, leaf litter, necromass, and soil but is not included in these figures.

Discussion

Forest Structure

The majority of the trees identified and measured in Yumao were those typically associated with topographic conditions unique to dry forests and the Chaco. Some species however, in particular those of the Cactaceae family, are more associated with Bolivian-Tucuman forest. In general, this combination of species indicates what Navarro et al. (2007) characterize as pre-Andean transitional dry and semi-arid Chaco forest. The incidence of new species became less and less frequent as the inventory was completed, which is what one would expect in a typical species accumulation curve.

Tropical dry forests are generally composed of smaller trees and are less diverse than tropical rain forests. In fact, on a scale of several hectares, tropical dry forests have approximately half the number of species found in wet forests (Murphy and Lugo 1986, Gentry 1995). Although family richness is due to the high variability between dry forests, which makes generalizations across areas difficult, the 52 tree species identified in Yumao fall within the parameters established by several studies across dry forests. For example, Murphy and Lugo (1986) recorded a range of 35 – 90 species and Gillespie et al. (2000) noted between 34 – 81 species in a summary of eight studies over 28 lowland tropical dry forest sites.

In a forest the dynamics of growth and development of their populations is quite variable and a complex relationship exists between species and even between individuals of the same species. One of the indicators that can determine the degree of stability of a forest in time is the distribution of diameter sizes. Generally in wild multiple-age populations, individuals of smaller diameter are much more numerous than those of larger diameter. The structure of our sample site was consistent with this idea (Figure 4).

Figure 4. Diameter distribution of Yumao forest. Our study area shows a higher number of small-diameter trees in almost every plot.

Biomass Estimation and Stored Carbon

Usually sample size is determined by admissible error on the estimate of the parameter of interest, in this case biomass. In terms of determining the biomass of the forest in Yumao, preliminary calculations revealed that more study plots are necessary. Using the formula,

 

where the -value is 2, represents variance, and the desired precision () is +/- 10, resulted in a recommendation for 44 or 90 plots in Yumao. The former figure only refers to the aboveground biomass, whereas the latter relates to all carbon sources including soil. The sample-size calculator developed by Winrock (http://www.winrock.org/Ecosystems/tools.asp) recommends setting up 42 study plots based on our preliminary data and according to the criteria developed for land-use land-change studies. In order to carry out a kriging calculation, a group of geostatistical techniques to extrapolate values from the study area, future studies in the AMNI RG-VC will require between 50 and 60 plots. It would therefore be misleading to compare our preliminary results with other more comprehensive biomass studies.

Recommendations

As part of the study done for Fundación Natura Bolivia, several recommendations were made in a formal report and presentation (Arnstein 2012). Foremost among them were: implementing the use of satellite imagery, reconsidering the use of permanent plots, and studying the specific context of dry forests in biomass measurement.

To facilitate fieldwork and improve the precision of measurements and subsequent carbon calculations it would have been useful to divide the project area into relatively homogenous strata of human intervention and/or elevation. Useful tools to accomplish this goal would be “ground-truthed” satellite images, aerial photographs and vegetation, soils, and topographic maps. I would also recommend that the locations of the plots be determined by map, sizing them to correlate with the size of remote sensing image pixels, such as Landsat. The disadvantage of permanent plots is that their location can be known and can therefore be treated in a different way than the rest of the study site. In a community as small as Yumao it will be difficult to assure that this does not occur.

In regard to future carbon studies, according to Brown (1997), biomass estimations in the majority of forests can be based on trees with a diameter equal to or greater than 10 cm. However, for forests composed of shorter trees such as those found in arid areas like Yumao it is recommended that all trees with diameters above 5 cm be measured. Additionally, approximately three percent of the trees measured were not broad-leafed species (i.e. belonging to the Cactaceae family) for which densities have not yet been calculated.

This research provided much information about forest structure and species composition in the under-studied transitional area between dry and Bolivian-Tucuman forest types. In addition, Fundación Natura Bolivia now has basic data about forest biomass in the area and a more solid methodology which they will use to expand the study to the entire AMNI-RG. The knowledge they continue to gather will be used to not only monitor forest growth and change but to also provide alternative carbon market-based financing mechanisms to support Bolivian smallholders.

Acknowledgements

This study would not have been possible without the support of Huascar Azurduy, Tito Vidaurre Sanchez, Maria Teresa Vargas, and Oswaldo Maillard of Fundación Natura Bolivia; Aquilino Molina Olivera and Ivan Linneo Forronda of Museo de Historia Natural Noel Kempff; CIAT Bolivia; funding from Yale’s Tropical Resources Institute; and the gently mocking counsel of my ad-hoc advisors Gordon Geballe, Florencia Montagnini, Tim Gregoire, and Mark Ashton. This article is dedicated to the community of Yumao which housed me, fed me, and invited me to go fishing.

Bibliography

Arnstein, E. 2012. Cuantificación de carbon almacenado en los bosques de la comunidad de Yumao, provincial cordillera, Bolivia. Santa Cruz.

BOLFOR (Bolivian Forest Service). 1995. Guía práctica y teórica para el diseño de un inventario forestal de reconocimiento. Available online: http://www.cadefor.org/index.php?option=com_remository&func=fileinfo&id=183&Itemid=65

BOLFOR. 1998. Monitoreo de parcelas permanentes de medición en el Bosque Chimanes. Available online: http://pdf.usaid.gov/pdf_docs/PNACG716.pdf

BOLFOR and PROMABOSQUE (Programa de Desarrollo Forestal Industria). 1999. Guía para la instalación y evaluación de parcelas permanentes de muestreo. Available online: http://pdf.usaid.gov/pdf_docs/Pnacg821.pdf

Brown, S. 1997. Estimating biomass and biomass change of tropical forest. Forestry paper. Rome: FAO.

Brown, S. and A.E. Lugo. 1992. Above ground biomass estimates for tropical moist forests of the Brazilian Amazon. Interciencia 17, 8-18.

Cabrera, A.L. and W. Willink. 1980. Biogeografía de América Latina. Serie de Biologia Monografía 3.

Chavé, J., C. Andalo, S. Brown, et al. 2005. Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia 145, 87–99.

Dauber, E., J. Terán and R. Guzman. nd. Estimaciones de biomasa y carbono en bosques naturales de Bolivia. Revista Forestal Iberoamericana 1, 1-10.

FAO (Food and Agriculture Organization of the United Nations). 2007. Digital Soil Map of the World 1:5.000.000 scale. Available online: http://www.fao.org/geonetwork/srv/en/metadata.show?id=14116.

Fundación Natura Bolivia. 2009. (Dirección de Áreas Protegidas del Gobierno Departamental Autónomo de Santa Cruz). Plan de manejo área natural de manejo integrado Rio Grande Valles Cruceños. Available online: http://www.naturabolivia.org/Informacion/Publicaciones/PLAN%20DE%20MANEJO%20ANMI%20RG-VC%20DIAP-NATURA.pdf

Gentry, A.H. 1995. Patterns of diversity and floristic composition in Neotropical montane forests. In: Biodiversity and Conservation of Neotropical Montane Forests, eds. S.P. Churchill, H. Baslev, E. Forero, and J.L. Luteyn, 89 – 102. New York: The New York Botanical Garden.

Gillespie, T.W., A. Grijalva, and C.N. Farris. 2000. Diversity, composition, and structure of tropical dry forest in Central America. Plant Ecology 147, 37-47.

Huayrana, A.M.F. 2012. Esquemas de Acuerdos Reciprocos por Agua (ARA) implementando iniciativas productivas amigables con el medio ambiente, Cuenca Arriba. Available online: http://www.ecosystem-alliance.org/sites/default/files/documents/FUNDACION%20NATURA_BOLIVIA_MH1.pdf

Ibisch, P.L., K. Columba and S. Reichle (Eds.) 2002. Plan de conservación y desarrollo sostenible para el bosque seco Chiquitano, cerrado y pantanal Boliviano. Editorial FAN: Santa Cruz.

INTI – CITEMA (Instituto Nacional de Tecnología Industrial - Centro de Investigaciones Tecnológicas de la Madera). Listado de densidades secas de maderas. Available online:http://www.inti.gov.ar/citema/densidad_cientifico.pdf.

Instituto Nacional de Estadística. 2001. Censo de población y vivenda – 2001: Poblacion por organizaciones comunitarias. Available online: http://www.ine.gob.bo/comunitaria/comunitariaVer.aspx?Depto=07&Prov=07&Seccion=05.

IPCC (Intergovernmental Panel on Climate Change). 2006. Guidelines for national greenhouse gas inventories. Available online: http://www.ipccnggip.iges.or.jp/public/2006gl/.

Murphy, P.G. and A.E. Lugo. 1986. Ecology of tropical dry forest. Annual Review of Ecology and Systematics 17, 67-88.

Navarro, G., W. Ferreira, A. Fuentes, et al. 2007. Mapa de vegetación de Bolivia a escala 1:250 000. Available online: http://conserveonline.org/workspacesbol.veget/bol_vegetSIG/view.html.

United Nations Framework Convention on Climate Change (UNFCC). 2006. Issues relating to reducing emissions from deforestation in developing countries and recommendations on any further process. Available online: http://unfccc.int/resource/docs/2006/sbsta/engmisc05.pdf.

Zanne, A.E., G. Lopez-Gonzalez, D.A. Coomes, et al. 2009. Data from: Towards a worldwide wood economics spectrum. Dryad Digital Repository. doi:10.5061/dryad.234.

Citation

Arnstein, E. 2014. Forest Inventory and Quantification of Stored Carbon in the Bolivian Chaco. Tropical Resources Bulletin 32-33, 108-118.

Appendix 1

Common Name	Scientific Name	Family
ajo ajo	Gallesia integrifolia (Sprengle) Harms	Phytolaccaceae
algarrobo	Caesalpinia paraguariensis (Parodi) Burkar	Fabaceae
bouganvilla, zapallo (plot 2)	Bougainvillea modesta Heimerl.	Nyctaginaceae
cacha	Aspidosperma quebracho-blanco Schltd.	Apocynaceae
café con leche, sombra del toro	Sideroxylon obtusifolium (Humb. Ex. Roem & Shult.) T.D. Penn	Sapotaceae
cala pierna, roble	Amburana cearensis (fr. Allen) A. Co. Sm.	Fabaceae
caracore	Cereus sp.	Cactaceae
caraparí	Cereus comarapanus Cardenas	Cactaceae
cari cari	Acacia loretensis Mac.	Fabaceae
cedro	Cedrela lilloi C. DC.	Meliaceae
chari	Parapiptadenia excelsa (Griseb.) Burkart	Fabaceae
chichapí	Celtis chichape (Gard.) Mig	Cannabaceae
chika chika	Randia cf. boliviana Rusby	Rubiaceae
chiki chiki	Agonandra excelsa (Griseb)	Opiliaceae
chorokete	Ruprechtia triflora Griseb.	Polygonaceae
coca de cabra comomosí	Capparis speciosa Griseb.	Capparidaceae
cordia, unknown 6	Cordia alliodora (Ruiz & Pavon) Cham.	Boraginaceae
cuchi mara	Loxopterygium grisebachii Hieron. & H. Lorentz	Anacardiaceae
cuchi	Astronium urundeuva (Allemao) Engl	Anacardiaceae
curupau	Anadenanthera colubrina (Vell. Conc.) Benth	Fabaceae
cuta	Phyllostylon rhamnoides (Poisson) Taub.	Ulmaceae
itapallo	Urera baccifera (L.) Gaudich. Ex Wedd.	Urticaceae
jichituriqui	Aspidosperma pyrifolium C. Martius	Anacardiaceae
lanza	Saccellium lanceolatum Hump. & Bonpl (Cordia saccellia)	Boraginaceae
limoncillo	Ximenia americana L.	Olacaceae
mani, tipilla (plot 1)	Diplokeleba floribunda N.E. Brown	Sapindaceae
momoki	Caesalpinia pluviosa D.C.	Fabaceae
negrillo	Achatocarpus praecox Griseb	Achatocarpaceae
palo blanco	Calycophyllum multiflorum Griseb.	Rubiaceae
palo cala	Crotalaria cf. anagyroides Kunth	Fabaceae
palo corcho	Aralia soratensis Marchal	Araliaceae
peroto	Pseudobombax marginatum (A St. Hil.. Juss & Cambess.) A. Robyns	Bombacaceae
pica pica	Cnidoscolus tubulosus (Muell. & Arg.) I.M. John- ston	Euphorbiaceae
picana	Casearia gossypiosperma Briquet.	Salicaceae
piñón	Jatropha macrocarpa Griseb.	Euphorbiaceae
pitajaya	Cleistocactus cf. baumannii (Lem.) Lem.	Cactaceae
quina	Myroxylon balsamum (L.) Harns	Fabaceae
quinilla	Pogonopus tubulosus (DC.) Schumann	Rubiaceae
sacharoza	Pereskia sacharosa Griseb.	Cactaceae
sawinto	Myrcianthes pungens (O. Berg.) Legrand	Myrtaceae
soto	Schinopsis haenkeana Engl.	Anacardiaceae
soto (plot 1)	Schinopsis quebracho-colorado (Schltdl.) F. Barkley & T. Meyer	Anacardiaceae
taijibo	Tabebuia serratifolia (Vahl) Nicholson	Bignoniaceae
tala	Celtis iguanaea (Jac.) Sarg.	Cannabaceae
tarara	Myroxylon peruiferum L.f.	Fabaceae
timboi	Enterolobium contortisiliquum (Vell.) Moronge	Fabaceae
tinajera	Aparisthmium cf. cordatum (Juss.) Bail.	Euphorbiaceae
tipa	Tipuana tipu (Benth.) Kuntze	Fabaceae
toborochi	Ceiba insignis (HBK) Griss & Semirsi	Malvaceae
trichilia	Trichilia claussenii C.DC.	Meliaceae
ulala	Cereus hankeanus K. Schum	Cactaceae
ulala (bush)	Cereus cf. kroenleinii N.P. Taylor	Cactaceae
uña de gato	Acacia praecox Griseb	Fabaceae
wauyacan	Machaerium scleroxylon Tul.	Fabaceae
wuayacan blanco	Chloroleucon foliolosum (Benth) GP Lewis	Fabaceae
wayabuta	Ruprechtia apetala (Cf) Wedd.	Polygonaceae
wayabuta (plot 1)	Ruprechtia cf. exploratricis Sandwich	Polygonaceae
zapallo	Pisonia zapallo Griseb.	Nyctaginaceae
unknown 1, unknown 5 (understory)	Athyana weinmannifolia L.	Sapindaceae
unknown 3 (understory)	Allophylus edulis (A. St._Hil & Camb.) Hieron. Ex. Niederl | Sapindaceae| |unknown 4 (understory) | Coursetia cf. brachyrhaphis Harms | Fabaceae| |unknown 5 | Simira macrocrater (K. Schum) Steyerm | Rubiaceae|