|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
SHIFTING CULTIVATION EFFECTS ON BRAZIL NUT (Bertholletia excelsa)
JAMIE NICOLE COTTA
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Jamie Nicole Cotta
To my parents, who made it all possible.
This thesis could not have been completed without the steadfast support of my ever
encouraging advisor, Dr. Karen Kainer, who generously shared her expertise and provided
counsel throughout the research and writing processes. I am equally grateful to my committee
member, Dr. Lucia Wadt, who welcomed me into her home in Brazil three years in a row,
offering not only research advice, but personal and professional guidance that contributed to the
success of my project in Brazil. I also thank the members of my supervisory committee: Dr.
Emilio Bruna and Dr. Kaoru Kitajima, for their comments and suggestions during the writing of
my thesis. I am immensely thankful for Christine Staudhammer's statistical counsel; she has a
true gift for communicating her statistical expertise with forestry professionals. I also thank
Meghan Brennan for sharing her statistical knowledge.
I sincerely thank my international sponsor, EMBRAPA-Acre (Empresa Brasileira de
Pesquisa Agropecudria) in Rio Branco, Brazil, for generously supporting this research. A
number of Brazilian colleagues and assistants provided essential insights and offered their talents
to this research project, including Paulo Wadt, Paulo Rodrigues de Carvalho, Pedro Raimundo R.
de Araujo, Aldeci da Silva Oliveira, Airton, Freire, Sergio, and Rivelino. I would like to
sincerely thank the extractivist families, specifically Valderi, Maria Alzenira, Duda and Bilu
Mendes, and Amarilzo da Rocha Bento, for generously sharing their insights, and providing me
with a true home away from home. Their joyful, affectionate children will also never be
forgotten. I would also like to thank Tim Martin, Christie Klimas, Cara Rockwell, Chris
Baraloto, Marisa Tohver, Roberta Veluci, Alexander Cheesman, Skya Rose Murphy, Shoana
Humphries, Valerio Gomes, and Marco Lentini for their contributions to my final draft.
This research would not have been possible without the support of my Department, the
School of Natural Resources and Environment, and a generous scholarship from the School of
Forest Resources and Conservation. I am also grateful for summer research grants from the
University of Florida Graduate School and the Tropical Conservation and Development
Program. The Institute of Food and Agricultural Sciences, the School of Natural Resources and
Environment, and the UF Graduate Student Council also generously provided me with finds to
return research results to communities.
Most importantly, I thank my family. My parents and sister represent the most admirable
of role models and, through their guidance and living examples, I have learned how to face great
challenges and accomplish personal goals, while never losing focus of others. Because of them I
understand true compassion and generosity of spirit. My grandparents, aunts, uncles, and
cousins have each shared a special part of themselves, which have all contributed to my success
and, without their presence in my life I would not have been able to pursue and complete this
TABLE OF CONTENTS
A CK N O W LED G M EN T S ................................................................. ........... ............. .....
L IST O F T A B L E S ............. ..... ............ ................................................................... . 8
LIST OF FIGURES .................................. .. ..... ..... ................. .9
A B S T R A C T ................................ ............................................................ 10
1 IN T R O D U C T IO N ....................................................................................... .................... 12
2 SHIFTING CULTIVATION EFFECTS ON BRAZIL NUT (Bertholletia excelsa)
REGENERATION ................................. .. ... .... ................... 17
Intro du action ............... ......................... .. ............................... ......................................17
Species Description .................................. ... .. .... ...... ................. 19
Study Area ........................................ 20
M methods ........................ .. ....................................................... 21
Plot Installation and Sampling Scheme........................................................... 21
Regeneration Environments by Forest Type........................................................22
Seedling and Sapling D ensities ............................................... ............................ 24
B excelsa Survival and G row th ........................................................... .....................24
D ata A n a ly sis ............................................................................................................. 2 5
Results .............................. .................. ............. 28
Regeneration Environments by Forest Type ........................................................28
Seedling and Sapling D ensities ............................................... ............................ 29
B excels Survival and G row th ........................................................... .....................29
S e e d lin g s ................................................................2 9
S a p lin g s ................................................................3 0
D iscu ssio n ................... ................... ...................1..........
D en site s E x p lain ed ................................................................................................ 3 1
Seed sources and dispersal ................................. ................................. 31
Seedling survival and growth ................................. .............................33
Sapling survival and growth.....................................36
Greater Ecological and Management Implications ........................................ ...37
3 C O N C L U S IO N ................................................................................................................. 3 9
A T R A N SE C T C R E A T IO N ................................................................................................ 46
B SCHEMATIC OF PLOT INSTALLATION AND SAMPLING SCHEME ..........................47
L IST O F R E F E R E N C E S ............................................................................... ...........................48
B IO G R A PH IC A L SK E T C H ............................................................................... .....................55
LIST OF TABLES
2-1 Mean ( sd) soil properties observed within fallows and mature forest plots in
RESEX Chico Mendes and PAE Chico Mendes. ............. ........................................43
2-2 Mixed model results based on B. excelsa seedling and sapling data subsets. ..................44
2-3 Comparison ofB. excelsa seedling and sapling characteristics in fallow and mature
fo re st ........................................................................................ . 4 5
LIST OF FIGURES
2-1 Location of study sites PAE Chico Mendes and RESEX Chico Mendes in Acre,
B razil ............. .......... .. ......... ............... ................................... 4 1
2-2 Mean ( SE) PPFD at four times of day in four fallows and two mature forest plots.......41
2-3 Bertholletia excelsa seedling and sapling densities (means SE) in fallow and
m atu re forest. ........................................................... ................. 42
2-4 Relationship between B. excelsa seedling height and photon flux density (PPFD)
m e a su re d in 2 0 0 5 ...............................................................................................................4 2
2-5 Relationship between B. excelsa seedling total leaf area and photon flux density
(PPFD ) m measured in 2005 .......................................................... ........ .......43
B-1 M ature forest and fallow plot design ........................................ .......................... 47
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
SHIFTING CULTIVATION EFFECTS ON BRAZIL NUT (Bertholletia excelsa)
Jamie N. Cotta
Chair: Karen Kainer
Major Department: School of Natural Resources and Environment
Brazil nut (Bertholletia excelsa), has emerged as the cornerstone of the extractive economy
in much of the Amazon, but the debate continues as to whether or not current harvest levels have
a detrimental effect on Brazil nut seedling recruitment. Regeneration studies to date have been
conducted solely within mature forest, but my study provides further insight into current Brazil
nut regeneration dynamics, with a unique first look at regeneration in swidden fallows within
two multiple-use areas in Acre, Brazil. Recruitment of individuals < 10 cm dbh was evaluated
within 25 x 25 m and 10 x 10 m subplots in four 9 ha mature forest plots and six fallows (0.5-1.0
ha each), respectively. Individuals < 10 cm dbh were grouped into two size classes: seedlings
(individuals < 1.5 m in height) and saplings (individuals between 1.5 m in height and 10 cm
dbh). General linear mixed model analyses revealed higher densities ofB. excelsa in fallow than
mature forest. Survival and growth of young B. excelsa individuals appear to be enhanced by
higher light levels found in fallows. Greater light availability is positively related to seedling
height and height growth, seedling leaf number and leaf area, and seedling survival, and may
indirectly enhance sapling growth and survival. This enhanced survival and growth likely
contributes to the observed higher densities of seedlings and saplings in fallows. Not only can
anthropogenic disturbance, in the form of shifting cultivation, play a positive role in Brazil nut
regeneration, it could explain current Brazil nut densities and distributions. Finally, in light of
these findings, swidden fallows could potentially be managed for enhanced Brazil nut densities,
which may provide an opportunity for greater income for extractive families while contributing
to the sustainability of Brazil nut extraction in the long term.
Degraded and secondary forests are widespread throughout the tropics, representing 60%
of remaining tropical forests (ITTO 2002). In the Brazilian Amazon alone, over 12% of the
forest mosaic is classified as degraded and secondary forest (ITTO 2002). As these forests
undergo succession they are capable of conferring significant environmental and livelihood
benefits. Secondary forests provide people with food resources, timber and non-timber forest
products, including fuelwood, and quality hunting sites (ITTO 2002). They also regulate water
regimes, protect soils from erosion, store carbon, and serve as biological corridors and refuges
for species (ITTO 2002). In addition, regenerating secondary forests are especially important in
the life histories of a number of tropical plant species, including some that are of particular
socioeconomic and ecological importance, such as Cedrela spp. and Swietenia spp. (Whitmore
1989). Similarly, two other valued species, Astrocaryum murumuru and Dipteryx micrantha,
benefit from seed dispersal to treefall gaps (Cintra and Horna 1997). All four of these species
exhibit gap-opportunistic traits, achieving enhanced growth in higher light conditions found in
canopy openings or young regenerating forest environments.
Brazil nut (Bertholletia excelsa, H.B.K.) is another highly valued tropical tree reported to
be a gap-opportunistic species (Mori and Prance 1990, Myers et al. 2000), benefiting from
canopy openings during early stages of its life history. Brazil nut has emerged as the cornerstone
of the extractive economy in much of the Amazon (Clay 1997), and the seeds, commonly
referred to as nuts, are used for a variety of products, including raw and dried nuts, oils, flour,
medicines, and personal care products (Ortiz 1995), making Brazil nut a versatile and valuable
non-timber forest product (NTFP). To assess recruitment status as related to long-term harvest
sustainability of this valuable species, several regeneration studies have been conducted in recent
years (Viana et al. 1998, Zuidema and Boot 2002, Peres et al. 2003, Serrano 2005), yielding
differing reports on the impact of nut harvest on Brazil nut population structures. However,
studies to date have been conducted solely within mature forest sites. The modern Amazonian
landscape consists of a mosaic of mature forest, regenerating secondary forest, swidden fallow,
and pasture; therefore, in light of the increasing extent of secondary forests in the region, the role
of these sites in critical economic and conservation species recruitment and regeneration
processes should not be underestimated.
In Acre, Brazil, local extractivists report relatively high levels of Brazil nut recruitment in
regenerating swidden fallows, which may be due to specific biotic and abiotic characteristics
associated with these sites, including differences in seed disperser activity (Clay 1997, personal
obs.), light availability (Kainer et al. 1998, Myers et al. 2000, Zuidema 2003), and nutrient
availability (Kainer et al. 1998), as well as increased resistance to pathogens in higher light
environments (Augspurger 1984b). The aim of my study is to assess to what extent swidden
fallows constitute favorable regeneration environments for B. excelsa. Although various studies
have assessed the performance of tropical seedlings in natural gaps (Popma and Bongers 1991,
Cintra and Horna 1997, Myers et al. 2000), typically formed by treefalls, conclusions regarding
recruitment, survival and growth in these canopy openings should not be applied to all cases of
canopy disturbance. For the purpose of this study, I highlight some characteristics unique to
secondary fallow forests, in comparison with treefall gaps, as well as other regenerating forest
types, which may each contribute considerably different influences on B. excelsa regeneration.
Fallows forests differ from treefall gaps in many ways. To begin with, the most common
treefall gap size is < 20 m2 (Brokaw 1982), while swidden plots average 2 hectares (equivalent to
20,000 m2) (Fujisaka et al. 1998). Abiotic conditions, including light and temperature levels,
vary within gaps (Denslow et al. 1990). As a result, one might observe a larger range of light
and temperature levels in larger gap areas, such as swidden plots. Also, herbivore activity may
differ as a result of this difference in gap size (Janzen 1990), as some vertebrates prefer
secondary forest environments (Pefia-Claros and De Boo 2002), but others avoid crossing large
clearings (da Silva et al. 1996).
What may differentiate fallows most from other secondary forests is the influence of land
use history on succession and regeneration (Uhl et al. 1988, Guariguata and Ostertag 2001,
Mesquita et al. 2001). In general, swidden fallows develop after 2 to 4 years of cultivation, and
are left to regenerate for a period of 5 to 15 years (Montagnini and Mendelsohn 1997). In fields
which have been subjected to little burning, Cecropia spp. are often the dominant pioneers to
establish in young fallows, however, where burning has been more frequent, Vismia spp. often
dominate (Borges and Stouffer 1999). The dominance of these early pioneers directly influences
subsequent vegetation succession (Nascimento et al. 2006). For example, in Cecropia-
dominated areas heavy litterfall and increased shading can inhibit seedling establishment
(Mesquita 1995, cited in Nascimento et al. 2006). As succession proceeds, differences in forest
structure, stem density, and species composition can be observed. Fujisaka et al. (1998) found
species richness to be higher in 3 5 year old fallows than in 1 2 year old fallows, and higher
in fallows than in pasture sites. Also, vines were more abundant in mature forest and fallows of
3 5 years than in cultivated sites and pasture (Fujisaka et al. 1998). Furthermore, grass and
weed dominance was reduced in fallows, compared to pasture and agroforestry sites, resulting in
less competitive suppression (Ferguson et al. 2003). More shrub species were found in cropped
and fallow sites than in mature forest and pasture (Fujisaka et al. 1998), and herbaceous plants
were most common in cropped fields and pasture. Finally, as a result of burning and crop
harvesting, nutrient availability may be significantly altered in young fallows (Uhl and Jordan
After a few years of succession, large regenerating clearings may harbor increased
numbers of vertebrates due to greater food availability for certain species than that of mature
forest (Janzen 1990). Composition of seed dispersers may also differ between gap types, as a
result of hunting activities that often coincide with crop cultivation (Janzen 1990).
In large man-made clearings, new recruitment is dependent upon seed dispersal (Ferguson
et al. 2003), if cropping activities are carried out such that adults of desired species are removed
or avoided, and no seeds are present in the seed bank. In this case, nearby patches of mature
forest are required for forest trees to colonize abandoned plots (Uhl et al. 1982). On the other
hand, recruitment may result from both dispersal and adult seed rain in a treefall gap. Although
swidden fallows are often surrounded by other fallows or secondary forests of varying ages
(Ferguson et al. 2003), recruitment processes may not be drastically compromised relative to
those in treefall gaps, since mature forest is often left intact along at least a portion of the
swidden edge (personal obs.). All of the aforementioned differences associated with forest gap
type are relevant to any analysis of recruitment and regeneration processes of species such as B.
excelsa in disturbed areas.
My study offers a new look at the seedling ecology of a valuable NTFP, through a focused
study on the post-germination phase of B. excelsa regeneration within two multiple-use areas in
the southeastern portion of Acre, Brazil. I predicted that densities of B. excelsa individuals
< 10 cm diameter at breast height would be greater in fallow than mature forest for two main
reasons: 1) local extractivists report relatively high agouti seed dispersal in fallows (Clay 1997),
and 2) survival and growth of young B. excelsa individuals is reported to be enhanced in the
presence of higher light availability in forest gaps (Mori and Prance 1990, Myers et al. 2000). In
addition to evaluating seeding and sapling densities, light levels, soil nutrient content, leaf
herbivory level, and number of nearby seed sources were evaluated in both fallow and mature
forest to determine their respective influences on B. excelsa survival and growth. Chapter 2 is
presented as a single, independent chapter ready for submission to a peer-reviewed journal.
Relevant conclusions and management implications are presented in Chapter 3.
SHIFTING CULTIVATION EFFECTS ON BRAZIL NUT (Bertholletia excelsa)
Brazil nut (Bertholletia excelsa, H.B.K.), has emerged as the cornerstone of the extractive
economy in much of the Amazon (Clay 1997), which has given rise to the debate over whether
or not current harvest levels have a detrimental effect on Brazil nut seedling recruitment. Some
scholars report that current harvest levels have minimal to no detrimental effect on seedling
recruitment in selected Brazil nut populations (Viana et al. 1998, Zuidema and Boot 2002), while
others report that regeneration is rare or nonexistent in over-exploited populations (Peres et al.
2003, Serrano 2005). All previous Brazil nut regeneration studies have been conducted solely
within mature forest, however, species regeneration is not limited to this forest type. The
modern Amazonian landscape is a mosaic of mature forest, regenerating secondary forest,
swidden fallow, and pasture; which comprise a continuum of microhabitat suitability for B.
excelsa recruitment and regeneration. An evaluation of Brazil nut regeneration ecology is
needed across the entire mosaic of forest types in order to provide an accurate picture of current
B. excelsa population structures.
Bertholletia excelsa may regenerate more successfully in secondary than mature forest,
based on reports that it is a gap-opportunistic species (Mori and Prance 1990, Myers et al. 2000).
Bertholletia excelsa is one of a number of shade-tolerant species which experience enhanced
seedling survival and growth in canopy openings, characterized by higher light levels (Brokaw
1985, Whitmore 1989, Molofsky and Fisher 1993, Cintra and Horna 1997). According to Myers
et al. (2000), B. excelsa seedlings not only benefit from forest gaps, they require a minimum gap
size of> 95 m2 or > 10.4% global site factor in order to reach sapling size. One large-scale
disturbance yet unexplored for B. excelsa recruitment is that of recent anthropogenic secondary
forests, which, in addition to being widespread throughout Amerindian history (Denevan 1992,
Heckenberger et al. 2003), are becoming increasingly common in the Amazonian landscape
In Acre, Brazil, landholders annually clear 0.5 2 ha patches of older secondary or mature
forest for swidden agriculture. When agricultural sites are abandoned, unique secondary forests
succeed in fallow sites, which may constitute particularly favorable regeneration environments
for B. excelsa. Local extractivists describe fallows as having greater B. excelsa recruitment than
mature forest, associated with a relatively higher incidence of agouti seed dispersal to these
disturbed sites. In addition, seedling survival may also be enhanced in fallow, due to decreased
vulnerability to pathogens in higher light environments (Augspurger 1984b). Subsequent
establishment and growth of B. excelsa may be greater in fallows because of high light levels
(Kainer et al. 1998, Myers et al. 2000, Zuidema 2003) and greater nutrient availability (Kainer et
An evaluation of the potential of fallow sites to provide favorable B. excelsa regeneration
environments will complement previous assessments of tropical seedling performance in other
disturbed sites, specifically natural gaps (Popma and Bongers 1991, Cintra and Horna 1997,
Myers et al. 2000). Although both types of disturbance increase forest light availability, fallows
represent ecologically unique microhabitats which may differentially affect seedling recruitment,
survival and growth. As a result of the shifting cultivation process, which often includes the
application of fire, fallow forests may be characterized by significantly different nutrient stocks
(Uhl and Jordan 1984), plant assemblages (Fujisaka et al. 1998, Ferguson et al. 2003), and
vertebrate activity (Schupp 1988, Cintra and Horna 1997, Pefia-Claros and de Boo 2002).
This study directly tests the hypothesis, based upon aforementioned scientific evidence and
extractivist observations, that B. excelsa seedling and sapling densities are greater in fallows than
mature forest, by assessing the post-germination phase ofB. excelsa recruitment. To understand
some of the mechanisms behind observed recruitment, effects of light, nutrient availability and
proximity of seed sources are also assessed. In addition, one-year survivorship and growth rates
of young individuals, seedling leaf production, and seedling herbivory, which further shape B.
excelsa recruitment, are compared between fallow and mature forest.
Bertholletia excelsa is a monospecific member of the Lecythidaceae family, found in
unflooded (terra fire) forests across the Amazon basin and the Guianas (Mori and Prance
1990). This canopy emergent can reach up to 50 m in height (Mori and Prance 1990), and some
individuals have been recorded at almost 1000 years of age (Vieira et al. 2005). In the two
landholdings surveyed in my study (Filipinas and Cachoeira), densities of Brazil nut trees
> 10 cm diameter at breast height (dbh) were 1.4 individuals ha-1 (Wadt et al. 2005) and 2.5
individuals ha-1 (Serrano 2005), respectively, which are representative of densities reported by
Peres et al. (2003) elsewhere in the study region. Although some authors report that B. excelsa
occurs in groves of 50 100 individuals, with each grove separated from another by distances of
up to 1 km (Mori and Prance 1990), Wadt et al. (2005) reported no existence of groves in my
Mammal communities play a critical role in the regeneration of many large-seeded tropical
tree species (Sork 1987, Asquith et al. 1997, Sanchez-Cordero and Martinez-Gallardo 1998,
Forget et al. 2000, Jansen 2003), via seed dispersal and seed predation. Brazil nut is no
exception, as the regeneration ofB. excelsa is highly dependent upon the presence of agoutis
(Dasyprocta spp.), the primary dispersers of Brazil nut seeds (Huber 1910, Prance and Mori
1978). Agoutis have been observed burying seeds at clearing edges (Peres and Baider 1997) and
in young secondary forest (Forget et al. 2000), and extractivists have often observed agoutis
carrying seeds to fallows (Clay 1997, personal obs.). Agoutis may prefer fallow sites because of
the protection provided against predators by regenerating vegetation (Cintra and Horna 1997).
These reports suggest a possibility of comparable or greater seed dispersal to fallows than mature
forest. Recruitment ofB. excelsa in both mature forest and fallows may also depend upon
density and proximity of reproductive adults, as the proportion of seeds that arrive at a certain
point in the forest should decline with distance from parent tree (Janzen 1970). Agoutis have
been observed dispersing B. excelsa seeds as far as 100 m from seed sources, with the majority
of seeds being dispersed within 25 m (Peres and Baider 1997). Normal seed shadows can be
extended by secondary dispersal, when seeds are unearthed by seed predators and reburied even
further from the adult tree (Mori and Prance 1990, Peres et al. 1997).
Fieldwork was carried out in the southeastern portion of Acre Brazil, between 100 and 110
south of the equator in Acre, Brazil. The region has undulating topography, vegetation classified
as humid, moist tropical forest (Holdridge 1978), and a pronounced three month dry-season from
June to August. Average annual rainfall is from 1600-2000 mm (IMAC 1991). Soils of the
region are classified under the Brazilian classification system as Argissolos (ZEE 2000), which
correspond roughly to U.S. Soil Taxonomy System Ultisols. Recruitment of B. excelsa was
evaluated within RESEX Chico Mendes and the Chico Mendes Agro-Extractive Settlement
Project (PAE Chico Mendes), multiple use areas of 976,570 ha and 24,898 ha, respectively,
separated by 30 kilometers (Fig. 2-1). Extractivists in the two study sites subsist on agricultural
and cattle production, rubber tapping and Brazil nut collection (Gomes 2001), and some timber
extraction in PAE Chico Mendes (personal obs.).
In my study sites, 0.5 1.5 ha of mature or older secondary forest found in close proximity
to households was manually cleared of large trees to initiate the shifting cultivation cycle.
Clearings were created either in areas void of valuable Brazil nut and rubber (Hevea brasiliensis)
trees, or great care was taken to protect individuals of these species. Clearings were burned
once, first cultivated for corn and rice, followed the next year by beans, and then manioc. After a
cultivation period of two to three years, agricultural sites were abandoned, and regenerating
secondary forests formed in situ. Young secondary forests such as these are typically
characterized by lower stand basal area than surrounding mature forest (Saldarriaga et al. 1988).
Although these swidden fallows are often surrounded by other fallows or secondary forests of
varying ages, mature forest bordered at least one length of each fallow border studied. Thus, the
study landscape consisted of a forest matrix with small fallows and pastures embedded within
mature forest expanses.
Plot Installation and Sampling Scheme
Mature forest plots (300 x 300m or 9 ha each) in RESEX Chico Mendes and PAE Chico
Mendes were originally demarcated and delineated in 2002 (Serrano 2005) using transects
opened every 25m and extending for 300m (Appendix A). Transects produced 25 x 25m grids of
9-hectare survey plots (Fig. B-1[A]). From May to July 2005, transects within three of these
original plots in RESEX Chico Mendes were reopened to evaluate all B. excelsa individuals
> 10cm dbh by searching to the right and left of systematically placed parallel lines spaced 25m
apart. To survey 25 % of the area of each mature forest plot for recruitment, four mature forest
subplots (25 x 25 m each) were randomly selected within each hectare. Thus randomization was
restricted to each one-hectare plot, generating a total of 36 subplots (Fig. B-1[A]).
Six fallows formed in RESEX Chico Mendes (0.5 1.0 ha each), and two in PAE Chico
Mendes (1.0 1.5 ha each), were also delineated in 2005 (Fig. B-1[B]), using transects opened
every 10m. Fallow ages ranged from 5 to 12 years, based upon year of abandonment of the
agricultural site and, due to their variable sizes and shapes, were delineated and demarcated with
a Garmin 12XL3 GPS unit. All individuals > 1.5 m in height were located in fallows by
searching to the right and left of parallel lines spaced 10 m apart. Recruitment subplots in
fallows (10 x 10 m) were randomly assigned across the extent of each fallow, and comprised
approximately 25% of each fallow area (Fig. B-1[B]).
Regeneration Environments by Forest Type
To evaluate the influence of seed sources on recruitment, all B. excelsa individuals
> 10 cm dbh were tallied within each 9 ha mature forest and 0.5 1.5 ha fallow, as well as within
a 50m border strip of each plot. Dbh, XY coordinates, and reproductive status were recorded for
To compare understory light availability among 2 mature forest plots and 4 fallows in
2005, photosynthetically active radiation (PAR) was estimated by averaging photosynthetic
photon flux density (PPFD) measurements taken at 10 equidistant sample points running through
the center of each plot, from one plot edge to the opposite edge (Appendix B). Measurements
were taken every 30 m in mature forest plots and measurement spacing varied in fallow sites,
due to variation in fallow size. To capture daytime variations, PPFD was measured in each plot
every 2.5h from 7:30 to 15:00 (ie. four times per day) for three days. PPFD was measured using
10 gallium arsenide phosphide (GaAsP) photocells mounted on a 10-cm disk and connected to a
multimenter (Micronta LCD Digital Multimeter 22-185A; Tandy Corporation, Fort Worth,
Texas). Sensors were calibrated with a standard LI-COR quantum sensor (model LI190SA) and
a Campbell CRO1x data logger.
To determine soil properties within each fallow and mature forest plot, soil samples were
collected in 2005 below the litter layer at two depths (0 20 cm and 20 40 cm). In the fallows,
soil cores were collected at two depths within one randomly selected subplot in each of the four
corners of the fallow, as well as the center-most subplot, to create one composite of five soil
cores for each depth (Appendix B). In the 9 ha mature forest plots, soil cores were collected
within the four hectares located at the covers of each plot, as well within the center-most hectare
(Appendix B). Within each of these five sampled hectares, a composite of five soil cores was
created for each depth, using the same sampling scheme described for soil collection in the
fallows. Two replicates of each composite were dried for 4 days at 650C, and passed through a 2
mm stainless steel sieve. Soil pH was measured at a 1:2.5 soil to water ratio. Extractable P and
K were processed using a dilute double acid extraction (Mehlich-1), with concentrations of P
determined colorimetrically using the molybdate blue method (Murphy and Riley 1962), and K
concentrations were determined using flame emission spectrophotometry. To determine total
potential acidity, H+ and Al3+ were extracted with a buffered solution of calcium acetate at pH 7,
and then titrated with a 0.1 N NaOH. Oxidizable organic carbon was determined on soils passed
through a 1 mm screen, ground in a porcelain mortar, and then digested in a potassium
dichromate acid medium with external heat. All soil analyses were conducted at the Soils
Laboratory of Embrapa-Acre, Brazil (EMBRAPA 1997).
The temperature and moisture environment in fallow and mature forest were evaluated
between July 2005 and June 2006. Mean monthly temperature and percent relative humidity
were determined for two fallow sites and one mature forest site based on measurements taken at
30 minute intervals using HOBO ProSeries data loggers.
Seedling and Sapling Densities
Seedling densities were assessed within all subplots installed in mature forest and fallows.
Sapling densities in mature forest also were assessed within subplots. Due to the ease of locating
saplings in fallow, however, sapling densities were assessed over entire fallow areas, rather than
in subplots only. Individuals < 1.5 m in height were defined as seedlings and individuals
between 1.5 m in height and 10 cm dbh were defined as saplings. This classification scheme was
selected not to differentiate between those individuals still using their seed reserves for growth
and those individuals relying on environmental resources, but rather to look at general classes of
recruitment. Within each subplot, XY location coordinates, height, seedling basal diameter, and
sapling dbh were recorded for all individuals.
B. excelsa Survival and Growth
Survival and annual growth (seedling height and basal diameter and sapling dbh) were
evaluated from June to July 2006. To evaluate the effect of PAR on seedling survival and
growth, PPFD was measured directly above seedlings in two mature forest plots and four
fallows. To capture daytime variations and to measure the full complement of all seedlings in
each selected plot, seedling PPFD was sampled on a rotating basis every 2.5h from 6:45 to 16:45
(ie. five sampling times per day) for three days. To assess the influence of photosynthetic
capacity on survival and growth of all seedlings, leaf area was evaluated for seedlings in three
mature forest plots and four fallows. When leaf number totaled < 5, all observed leaves were
sampled, which was the case for the majority of seedlings. In all other cases, 5 leaves were
sampled, representing the full range of leaf sizes. To avoid destructive sampling, leaves were
traced in the field, and average leaf area was measured in Gainesville, Florida, with a LiCor, Inc.
Model 3000A Leaf Area Meter with Transparent Conveyor Belt Accessory (LiCor, Inc., Lincoln,
Nebraska). Total leaf area was then estimated by multiplying total number of leaves per seedling
by average leaf area.
The influence of invertebrate herbivory on seedling survival and growth was also assessed
by quantifying the percentage of leaves per plant that showed signs of attack in 2005. Vertebrate
leaf herbivory was not assessed because it was not possible to observe or account for the
significant portion of predation which is comprised of leaves that are completely removed by this
type of herbivore.
All statistical analyses were performed using SAS software (Version 9.1, SAS Institute),
excepting Student's t-tests, which were performed using SPSS software (Version 10.0, SPSS
Inc.). Mature forest plots and fallows were treated as statistical blocks for all analyses, and
seedling and sapling densities were determined on a per hectare basis. Average soil values were
included as random effects, with plot nested within forest type, and soil nested within plot, for all
analyses. PPFD values were log transformed prior to all analyses to improve normality and
homoscedasticity of residuals, which were evaluated visually. PPFD levels were compared
between forest types using a repeated measures ANOVA, and soil nutrients were compared
between forest types using a two-way ANOVA. All soil nutrient values were averaged on a plot
basis, and average soil pH, potassium, and percent organic carbon at two depths (0 20 cm and
20 40 cm) were included in explanatory density and growth models. Correlation analysis,
using the PROC CORR procedure, revealed high correlations between the two depths for
phosphorous and potential acidity (H+Al3+), therefore these soil nutrient values were collapsed
into one depth for inclusion in the explanatory models.
Since B. excelsa densities demonstrated a Poisson distribution, they were modeled using a
general linear mixed model and the GLIMMIX procedure. Two-way correlations were
calculated between relevant variable pairs using PROC CORR in order to determine
relationships which merited inclusion in the model. I explored model building by conventional
methods of sequentially dropping non-significant interactions and covariates based on p-values.
In this case, I used a less restrictive significance level (a = 0.10) than the usual arbitrary 0.05
level (Johnson 1999, Burnham and Anderson 2002) to allow all possible significant effects to
remain for the purpose of building the best explanatory model (Bancroft 1968). Because a
primary purpose of data analysis was to evaluate differences between mature forest and fallow,
the indicator variable for forest type was considered fundamental and, therefore, retained in the
density models without regard to its significance level (Bancroft 1968). Densities were first
compared for all plots, with forest type (mature versus fallow) as the only retained fixed effect.
Seedling and sapling densities of mature forest plots previously surveyed in 2004 in PAE
Chico Mendes (Serrano 2005) were included in the comparative density analysis to increase my
sample size. These data were collected using the same sampling scheme described herein, and it
was assumed that no event had occurred that significantly affected these seedling and sapling
densities in this mature forest site (PAE Chico Mendes) between 2004 and 2005. To test this
assumption, 2004 seedling and sapling density data collected mature forest in RESEX Chico
Mendes, using the same sampling scheme, were compared with my 2005 seedling and sapling
density data from this same forest and plots and no statistical differences were detected
(p = 0.57). Finally, no site (PAE Chico Mendes versus RESEX Chico Mendes) interaction was
found when comparing B. excelsa densities; therefore, observations from both sites were
included in the analysis, with site dropped as a fixed effect from the model.
To explain observed densities, a subset of data (2 mature forest plots and 4 fallows) was
analyzed. Tested fixed effects included forest type, number of reproductive adults within 50 m
of each subplot, all plot soil content averages, and plot PPFD averages; insignificant variables
were sequentially removed from the final model. All possible predictor variables were included
in the final model to provide the best explanatory model; however, residuals were not well
Student's t-tests were used to compare initial height, initial basal diameter, height and
basal diameter growth, number of leaves, total leaf area, and leaf herbivory between fallow and
mature forest for all seedlings; all data were normally distributed. Sapling diameter growth,
which was likewise normally distributed, was also compared using a t-test. Logistic regression
was used to analyze differences in seedling survival; tested fixed effects included initial height,
initial diameter, height growth, diameter growth, forest type, all plot soil content averages,
PPFD, number of leaves per seedling, total leaf area, and percent leaf herbivory. Insignificant
variables were sequentially removed from the final model. Initial seedling height and diameter
were highly correlated with total leaf area, therefore, interactions were explored between total
leaf area and initial seedling size.
The GLIMMIX procedure was used to test seedling and sapling growth. Growth values,
some of which were negative, were log transformed (log 10 [1+x]) prior to analyses to improve
normality and homoscedasticity of residuals, which were observed visually. Because data
related to more mechanistic causes of growth (ie. light observed directly above seedlings) were
available, the forest type variable was intentionally excluded from the seedling growth analysis.
However, because these data were unavailable for saplings, forest type was retained for analysis
in the sapling diameter growth model. Tested fixed effects included in initial seedling growth
models were initial height or diameter, forest type, all plot soil content averages, PPFD, number
of leaves per seedling, total leaf area, and percent leaf herbivory, while initial sapling diameter
growth models included initial dbh, forest type, all plot soil content averages, and plot PPFD
averages. Again, all possible predictor variables were included in the final models to provide the
best explanatory models of growth, and insignificant variables were sequentially removed from
the final model; residuals were fairly well distributed. Sapling height growth could not be
compared between forest types, as sapling heights were estimated visually.
Differences in seedling leaf herbivory percentage were analyzed with the GLIMMIX
procedure, accounting for a log-normal distribution of data. Leaf herbivory percentages were
arcsin transformed prior to analysis in order to improve the normality of the response variable.
Tested fixed effects included initial height or diameter, forest type, all plot soil content averages,
PPFD, number of leaves per seedling, total leaf area; insignificant variables were sequentially
removed from the final model. Normality and homoscedasticity of residuals were observed
visually and the residuals were fairly well distributed.
Regeneration Environments by Forest Type
Of all variables measured to characterize regeneration environments (plot light levels,
density of nearby seed sources, temperature and relative humidity, and soil nutrient stores), only
light levels differed significantly between the two forest types. A repeated measures ANOVA
revealed higher average PPFD levels in fallow than mature forest (p = 0.02) (Fig. 2-2). There
were no significant differences between forest types in number of adults within 50 m of
evaluated subplots, nor in observed daily temperature and percent relative humidity monitored in
two fallows and one mature forest site between July 2005 and June 2006. Mean monthly
temperature in the three sites ranged between 23.5 and 24.20C and mean relative humidity
ranged between 90.3 and 92.7%. Although two-way ANOVAs revealed no soil property
differences between forest types (Table 2-1), soil P was significantly different by depth (p =
Seedling and Sapling Densities
Observed mean seedling densities of 12.7 and 5.3 trees ha-1 in fallow and mature forest
(Fig. 2-3), respectively, were different at p = 0.07, according to the general linear mixed model.
Mean sapling densities of 5.2 and 1.3 trees ha-1 in fallow and mature forest (Fig. 2-3),
respectively, were also different, again at p = 0.07. Based upon an analysis of the data subset,
seedling density differences were somewhat explained by the number of adults within 50m of
each subplot (p = 0.06), such that seedling densities increased with number of adults, while
sapling densities were best explained by percent organic carbon at the 0 20 cm depth (p = 0.02)
B. excelsa Survival and Growth
After one year, 26% of surveyed seedlings in fallow and 19% in mature forest (21% of all
seedlings) were either dead or unable to be located, with no survivorship differences between
these two forest types (Table 2-3). Seedling survival was possibly explained by total leaf area (p
= 0.08) and an interaction between percent leaf herbivory and PPFD (p = 0.08) (Table 2-2);
improved survival was associated with greater leaf area and higher leaf herbivory.
Initial seedling height was not significantly different between forest types (Table 2-3),
and, across forest types, seedling height in 2006 was best explained by PPFD measured in 2005
(p = 0.0009) (Fig. 2-4); seedling height increased with PPFD increase. Seedling height growth
was best explained by an interaction between percent leaf herbivory and PPFD (p = 0.02) and
percent leaf herbivory alone (p = 0.02) (Table 2-2). Height growth increased with PPFD, and
increased with herbivory decrease. Seedling growth appeared to increase more with PPFD at
intermediate herbivory levels. A t-test revealed no significant difference in initial seedling basal
diameter between forest types (Table 2-3), however, basal diameter growth was greater in fallow
than mature forest (p = 0.05), according to a t-test. According to the general linear mixed model,
seedling basal diameter growth was best explained by total leaf area (p = 0.05) (Table 2-2), such
that seedlings with more leaf area grew more. Seedling height and diameter growth
demonstrated positive relationships to all tested factors except percent leaf herbivory.
Total leaf area was highly correlated with seedling height, diameter and number of leaves
and, as a result, less significant variables sequentially fell out of the growth models. Total leaf
area was strongly related to PPFD (p = 0.0004) (Fig. 2-5). Observed herbivory level was much
higher in mature forest than fallows (p < 0.0001) (Table 2-3). The mixed model corroborated
that percent seedling leaf herbivory was explained by forest type alone (p = 0.05) (Table 2-2),
but also showed that herbivory was best explained by an interaction between forest type and
number of leaves (p = 0.006) (Table 2-2), such that herbivory decreased with leaf number in
mature forest, but increased with leaf number in fallow.
Estimated mean sapling heights were 5.51 and 2.99 m, in mature forest and fallow,
respectively, which was significantly different (p = 0.02) (Table 2-3). Initial sapling dbh was
also greater in mature forest than fallow (p = 0.05) (Table 2-3). Maximum dbh values were 9.5
cm in mature forest and 7.0 cm in fallow. Nearly 100% sapling survival was observed in both
forest types (only one sapling died in a single fallow site), and therefore no significant difference
in sapling survival between forest types was detected. Sapling dbh growth was much greater in
fallow than mature forest (p = 0.02), and was potentially explained only by forest type (p = 0.07)
Bertholletia excelsa seedling densities were over two times greater in anthropogenically-
disturbed fallows than mature forest, and sapling densities were four times greater in fallow (Fig.
2-3). Many factors influence establishment, survival, and growth of young individuals, including
proximity and density of seed sources (Janzen 1970), seed disperser and predator behavior (Sork
1987, Forget 1994, Molofsky and Fisher 1993, Asquith et al. 1997), light availability
(Augspurger 1984a, Augspurger 1984b, Poorter 1999), and nutrient availability (Ceccon et al.
2003), some of which may be related to microhabitat, or forest type. My data suggest that the
influence of density and proximity of seed sources and light levels, at various stages in the B.
excelsa recruitment process, play a significant role in shaping the observed densities of young B.
excelsa individuals. Differences in densities between forest types, however, could also be a
reflection of additional factors not measured, such as seed dispersal. These factors, which affect
species establishment to differing degrees at various stages in the recruitment process, will be
Seed sources and dispersal
Based on a deeper analysis of a subset of data, which included number of seed sources
within 50m of surveyed subplots, seedling densities were mildly related to the number of nearby
seed sources (p = 0.06). These findings are consistent with scientific evidence that seedling
densities decline with increased distance from parent trees (Janzen 1970). Therefore, higher
recruitment would be expected in fallows that either contain adult B. excelsa trees or are located
adjacent to mature forests with high adult densities, compared to those with no adults nearby.
Although density of nearby adults explained relative densities, seedlings and saplings were often
found in areas with neither a reproductive adult within 50m, nor traces of potential seed sources
that may have suffered mortality prior to my study, confirming seed dispersal distances well
beyond the typical dispersal maximum of 25m (Peres and Baider 1997). These dispersal
distances may be explained by secondary dispersal (Mori and Prance 1990, Peres et al. 1997).
Residuals in these explanatory density models were poorly distributed, with high density
observations being grossly underestimated, suggesting that relevant factors have not been
included in the models. These underestimations could be due to a clumped, and possibly
directed, seed dispersal pattern that has not been accounted for. Bertholletia excelsa seed
dispersal beyond the adult crown is determined nearly entirely by agouti scatterhoarding
behavior (Prance and Mori 1978), and local extractivist reports suggest that agouti dispersal
behavior may be non-random. Extractivists report preferential seed dispersal in fallow
environments (Clay 1997, personal obs.). These accounts are corroborated by observations of
seed burial at clearing edges (Peres and Baider 1997) and in young secondary forest (Forget et
al. 2000), perhaps related to agouti predator avoidance in areas composed of limb and vine
tangles (Cintra and Horna 1997). Agouti scatterhoarding activity may potentially increase the
proportion of seeds in favorable microenvironments (Jansen and Zuidema 2001), which has been
proposed as a more critical influence on regeneration than post-dispersal predation escape or
successful germination (Rey and Alcantara 2000). The importance of disperser site preference
and directed dispersal has been demonstrated for species other than B. excelsa (Uhl et al. 1981,
Schupp et al. 1989, Forget 1994). Although disperser behavior was not tested in this study, it
likely had a significant effect on seed availability for recruitment in fallow and mature forest.
Equally important is the influence of differential seed removal rates (due to retrieval and
predation) between different forest types, which may further affect observed B. excelsa densities.
Some authors report higher seed removal rates of various tropical species in gap environments
(Schupp 1988, Schupp and Frost 1989, Cintra and Horna 1997), while Sanchez-Cordero and
Martinez-Gallardo (1998) report greater tropical seed removal in mature forest than gaps.
Tabarelli and Mantovani (1996), did not find seed removal of B. excelsa to be different between
treefall gaps and undisturbed forest, however, clearings created by anthropogenic disturbances
were not assessed. It is clear that seed disperser and seed predator activities are essential factors
to include in future density evaluations.
Seedling survival and growth
Once seeds are dispersed and young seedlings establish, individuals may demonstrate
differential growth (Augspurger 1984a, Popma and Bongers 1991, Brown and Whitmore 1992,
Poorter 1999) and/or survival (Augspurger 1984b, Cintra and Horna 1997, Brown and Whitmore
1992) in different forest types, which are often differentiated in the literature between understory
and gap environments. In my study, however, forest type explained only sapling diameter
growth (sapling height growth could not be evaluated). Other measured variables, some of
which may be associated with forest type, demonstrated stronger influences on observed survival
and growth of B. excelsa seedlings.
To begin with, total leaf area, which is correlated with leaf number, was a mild indicator of
seedling survival (p = 0.08). Leaves facilitate the acquisition of carbon by means of
photosynthesis; an enhanced carbon gain, due to greater leaf area, may confer relatively higher
fitness (and ultimately survival) to seedlings by various means, including the provision of more
structural material and energy, an increase in nutrient content, and a better tolerance to
environmental stresses (Chabot and Hicks 1982). PPFD, although not highly significant on its
own, may contribute to increased seedling survival, as PPFD was strongly positively correlated
with total leaf area (p = 0.0004), which did explain seedling survival at p = 0.08. Although some
studies have demonstrated greater seedling survival in gaps than understory for some species
(Augspurger 1984b, Cintra and Horna 1997), my results do not directly support these
conclusions, as I found one-year B. excelsa survivorship to be no different between mature forest
and fallow. Kainer et al. (1998) also found no relationship between seedling survival and
comparative light levels in pasture, forest gap, and shifting cultivation sites (Kainer et al. 1998).
The positive effect of leaf herbivory on survival is not so easily explained. Contrary to my
findings, numerous studies demonstrate the negative effect of leaf herbivory on seedling survival
(Howe 1990, Hulme 1996, Vasconcelos and Cherrett 1997). In this study, the influence of
herbivory level on survival may be coincidental. Invertebrate herbivory was not unusually high
in either forest type, and is probably not sufficient to lead to seedling mortality under normal
conditions. In addition, it was not possible to quantify vertebrate herbivory on seedling survival
in this study. Nevertheless, it is likely that a greater percentage of mortality was represented by
completely uprooted seedlings (Ortiz 1995, Zuidema 2003, personal obs.), resulting from seed
predators unearthing and consuming seeds at the base of the plant. Furthermore, only 14 of 94
total seedlings died, therefore concrete differences between factors relating to seedling
survivorship were difficult to detect. Although forest type was not a significant factor in survival
probability, in light of other analyzed effects, it is possible that the forest type variable was
partially masked by other variables such as percent leaf herbivory or PPFD (due to multiple
colinearity), which were both significantly different by forest type (PPFD was higher in fallow
and leaf herbivory was lower in fallow than mature forest). If forest type is partially included in
one of these variables, it may play an undetected role in determining seedling survival
Seedling heights in 2006 were highly positively correlated with PPFD levels measured
directly above seedlings in 2005, which implies a significant advantage to height growth in
higher light environments. In my study, one-year seedling height growth was influenced by an
interaction between PPFD and percent leaf herbivory (p = 0.02), such that height growth
declined with increased leaf herbivory and, while generally increasing with PPFD, was greatest
at intermediate light levels. Previous studies have shown a positive relationship between
seedling height growth and light availability in the tropics (Augspurger 1984a, Denslow et al.
1990, Popma and Bongers 1991), and similar results have been reported for B. excelsa (Poorter
1999, Zuidema 2003), again, with highest growth rates at intermediate light levels.
Uhl and Jordan (1984) found that soil nutrient concentrations did not vary significantly
between agricultural sites and mature forest 5 years after cutting and burning. Similarly, I found
no difference in soil nutrient availability between mature forest and fallows. As a result, soil
nutrients did not explain B. excelsa survival and growth as well as other measured variables,
such as light availability.
Numerous studies have described the negative effects of herbivory on seedling growth
(Marquis 1984, Howe 1990, Hulme 1996, Vasconcelos and Cherrett 1997). In my study,
seedling leaf area did not differ significantly by forest type, but was highly correlated with
seedling height, and the two variables exhibited a significant interaction effect for seedling
height growth. Height growth was positively related to both variables, similar to other studies
which have demonstrated increased growth with greater initial height (Zuidema 2003) and leaf
area (Poorter 1999). Similar to height growth, seedling diameter growth was explained by total
leaf area (p = 0.05), and possibly by an interaction between leaf herbivory and PPFD (p = 0.07),
with a positive relationship to each factor. Zuidema et al. (2003) report a similar relationship
between B. excelsa biomass growth and light availability. Although not directly evaluated,
seedling survival and growth in my study sites may also have been influenced by ant presence.
Leaf cutter ants (Atta spp.) are known to modify soil nutrient concentrations (Lugo et al. 1973,
Haines 1978), which may directly influence the development of young B. excelsa individuals, as
well as influence the competitive environment with other species. In addition, extractivists assert
that leaf cutter ants preferentially predate both B. excelsa seeds and seedlings. The potential
importance of leaf cutter ants in fallow environments is worth particular attention as leaf-cutting
ant populations have been shown to increase after forest clearing (Vasconcelos and Cherrett
1995). Blanton and Ewel's (1985) findings suggest that leaf-cutting species continue to thrive as
succession proceeds, representing the most important herbivore in the successional forests they
studied. I found one of my six surveyed fallows was covered to a great extent in leaf cutter ant
nests (species unknown). The relationship between forest type and ant density, intensity of ant
seed and seedling predation in fallow and mature forest, and the effects of ant populations on soil
composition all merit further consideration, to better understand the role of ant populations in B.
excelsa recruitment and regeneration.
Sapling survival and growth
High sapling survival rates in both mature forest and fallow suggest that, once seedlings
have been well-established, they develop and survive equally well in either forest type. Zuidema
and Boot (2002) also found extremely low mortality rates (1%), over a two-year period, for B.
excelsa individuals > 1 cm dbh. Initial sapling diameter was greater in mature forest, but this can
be explained by the young age of the fallows (5 12 years), as fallow saplings have not had the
opportunity to attain dbh sizes near 10cm within this time frame. No factor included in statistical
analyses, aside from forest type, explained higher diameter growth rates in fallow. However,
saplings in fallow may also be able to invest more in biomass growth than mature forest saplings,
since height growth is greatly enhanced by higher light levels in fallow. Light is cited as the
most frequently limiting resource for seedling growth (Kitajima 1996), a major component of
tropical plant regeneration processes (Chazdon et al. 1996). Increased light availability is likely
one of the most significant advantages conferred upon well-established seedlings in fallow, as
observed PPFD was significantly greater in fallow than mature forest (p = 0.02), and the benefits
of light availability are likely great for individuals up to 10 cm dbh, which continue to compete
to reach the canopy up to, and even beyond, this 10 cm limit.
Secondary forests are, by nature, dynamic and, as a result, are associated with an ever-
changing microenvironment of vegetation, water, and soil nutrients. Therefore, the suitability for
seedling and sapling growth and survival may vary with forest successional stage. However, the
observed higher densities of young individuals in fallow indicate that this forest type, in general,
constitutes a favorable microhabitat for both seedling and sapling development.
Greater Ecological and Management Implications
This research provides a new, and perhaps more accurate, depiction of current Brazil nut
regeneration in tropical forest landscapes, as secondary forest environments have previously
been ignored. The high densities of young Brazil nut individuals in secondary forests created
through shifting cultivation fallowss) clearly demonstrate that Brazil nut regeneration may have
been underestimated in previous studies which were conducted solely within mature forest.
In light of these findings, fallows could potentially be managed for enhanced Brazil nut
densities, which may provide an opportunity for greater income for extractive families while
contributing to the sustainability of Brazil nut extraction in the long term. In the past, some
extractivists have experimented with small-scale Brazil nut enrichment plantings, as a means to
economically enrich their landholdings (Kainer et al. 1998). The benefits of fallow management
may also provide an additional incentive for extractivists to allow fallows to regenerate, rather
than convert them to pasture, which is an increasingly common practice in much of the Amazon.
Shifting cultivation has previously been recognized for its role in shaping Amazonian
ecosystems (Dufour 1990). Pre-Columbian human interventions, in the form of active planting,
were previously considered by Posey (1985) and Balee (1989) as a possible explanation for the
presence of Brazil nut "groves" in parts of the Amazon. Based on the results of this study,
however, it is possible that higher densities are a result of less direct human interventions, such
as shifting cultivation. Not only can anthropogenic disturbance, in the form of shifting
cultivation, play a positive role in Brazil nut regeneration, it could explain current Brazil nut
densities and distributions.
The objective of this study was to compare B. excelsa recruitment and regeneration in
regenerating swidden fallows and mature forest. My results suggest that fallows constitute
favorable microhabitats for both seedling and sapling development. Perhaps most importantly,
survival and growth of young B. excelsa individuals appear to be enhanced by the higher light
levels found in fallow. Greater light availability is positively related to seedling height and
height growth, seedling leaf number and leaf area, and seedling survival, and may indirectly
enhance sapling growth and survival. This enhanced survival and growth likely contributes to
the observed higher densities of seedlings and saplings in fallow sites.
At the conclusion of this study it became clear that seed dispersal and seed predation
should be compared between forest types in future regeneration studies, in order to better assess
the various recruitment advantages, and possibly disadvantages, associated with fallow
environments. In addition, herbivore activity, and its effects on B. excelsa survival and growth,
should be evaluated in greater detail, quantifying the proportion of herbivory exercised by both
vertebrates and invertebrates, in both forest types. Leaf cutter ant behavior in regenerating
fallow fields merits particular consideration. Future studies should also compare B. excelsa
regeneration between different secondary forest types, as well as between fallows of varying
ages, as various successional forest types and successional stages may contribute to differential
survival and growth of young individuals. At the time of thesis submission, soil analyses were
incomplete, therefore conclusions regarding the effects of soil content were limited. Upon
publication of Chapter 3 in a scientific journal, the incorporation of additional soil data should
yield further insights.
Although some significant factors in B. excelsa recruitment processes, such as disperser
activity, were not evaluated in the study, the importance of the results should not be
underestimated. My research focuses on a widespread forest type which has not previously been
included in assessments of B. excelsa population dynamics therefore conclusions regarding
regeneration in fallows may better explain current B. excelsa densities and distributions than
other studies to date. An understanding of the role of fallows in B. excelsa regeneration may also
provide insight into the regeneration requirements of other valuable gap-loving species, such as
those in the Cedrela and Swietenia genera.
L ~ ACRE .
RESEX Chco Mendes r-
-pAE Chico Mendes
Figure 2-1: Location of study sites PAE Chico Mendes and RESEX Chico Mendes in Acre,
Brazil (Figure adapted from Gomes 2001, with permission).
- -*- Fallow
Figure 2-2: Mean ( SE) PPFD at four times of day in four fallows and two mature forest plots.
Data are log transformed.
--- ..- .
-^f ^ ---
* Mature Forest
Figure 2-3: Bertholletia excelsa seedling and sapling densities (means + SE) in fallow and
0 50 100 150 200 250
Figure 2-4: Relationship between B. excelsa seedling height and photon flux density (PPFD)
measured in 2005. Each point represents one observed B. excelsa seedling. Regression line
fitted: y = 0.005x + 0.52 R2 = 0.31.
Figure 2-5: Relationship between B. excelsa seedling total leaf area and photon flux density
(PPFD) measured in 2005. Each point represents one observed B. excelsa seedling. Regression
line fitted: y = 7.93x + 294.97, R2 = 0.23.
Table 2-1: Mean ( sd) soil properties observed within fallows and mature forest plots in
RESEX Chico Mendes and PAE Chico Mendes.
Soil property Fallow Mature Forest
pH 5.5 ( 0.2) 5.1 ( 0.3)
K (cmol*kg-1) 0.4 ( 0.03) 0.3 ( 0.1)
P (mg*kg-1) 2.0 (+ 0.3) 2.7 (+ 0.4)
(HA1+) (cmol*kgb ) 4.0 (+ 0.7) 5.7 (+ 1.0)
Organic Carbon (%) 0.5 (+ 0.1) 0.6 (+ 0.1)
Table 2-2: Mixed model results based on B. excelsa seedling and sapling data subsets.
Model fixed effects
Number of adults within 50m of subplots
Total leaf area
Height growth (cm)
PPFD*Percent leaf herbivory
Potential acidity (0 40cm)
Soil K (0 20cm)
Basal diameter growth (cm)
Total leaf area
Percent leaf herbivory
Number of leaves*forest type
Percent organic carbon
Table 2-3: Comparison ofB. excelsa seedling and sapling characteristics in fallow and mature
forest (all p-values based upon t-tests, except one-year survivorship, which is based
upon logistic regression).
Characteristic N Fallow Mature Forest p-value
Initial seedling height (m) 76 0.67 ( 0.42) 0.59 (+ 0.38) 0.45
Initial seedling basal diameter (cm) 76 0.64 ( 0.34) 0.55 (+ 0.21) 0.15
Seedling height growth (m) 62 0.14 ( 0.28) 0.12 (+0.22) 0.82
Seedling basal diameter growth (cm) 62 0.19 ( 0.25) 0.10 (+0.11) 0.05
Number of leaves per seedling 75 12.8 ( 9.4) 9.7 (7.6) 0.13
Total leaf area (cm2) 68 1003.4 ( 915.4) 710.5 (+ 807.2) 0.20
Leafherbivory (%) 74 16 ( 24) 58 ( 28) < 0.0001
One-year survivorship (%) 66 74 81 0.79
Initial height (m) 31 2.99 (+1.95) 5.51 ( 3.74) 0.02
Initial dbh (cm) 31 2.23 (+2.07) 4.45 ( 3.32) 0.05
Dbh growth (cm) 25 0.47 (+0.39) 0.11 (+0.32) 0.02
Data reported as means (+ sd), excepting percentage data. Reported sapling heights are
To open transects, a compass was placed on a level-cut branch. Next, two sapling poles
where placed in the line of sight of the compass to begin a straight line. Field technicians and
local assistants referred to this line in order to open a straight transect through the forest with a
machete. Reference poles were placed no more than every 15 m apart, with poles placed closer
together in hilly or overgrown areas. Distance was measured with a metric tape to create entire
300 m transects. Each transect was re-measured with metric tape after clearing was completed.
This method facilitated searching and locating B. excelsa individuals over large areas.
SCHEMATIC OF PLOT INSTALLATION AND SAMPLING SCHEME
*25 50 75 10 125 150 175 20 225 250 275 300
N ----- ------
0 25 50 75 100 125 150 175 200 225 250 275 300
0 10 20 30 40 50 60
70 80 90 100
Figure B-1: Mature forest and fallow plot design. A) Nine ha plot of mature forest, with four 25
x 25m subplots randomly selected within each ha, for a total of 36 subplots B) 1.1 ha fallow with
27 10 x 10m subplots. Distance is indicated in meters. U represents recruitment subplot. Light
environments within two mature forest plots and four fallows were characterized by photon flux
density (PPFD) measurements; represents PPFD measurement. Soil samples within each
fallow and mature forest plot were collected below the litter layer at two depths (0-20 cm and 20-
40 cm). Sampling scheme is depicted above for each forest type; @ represents soil sample.
LIST OF REFERENCES
Asquith, N.M., Wright, S.J., Clauss, M.J., 1997. Does mammal community composition control
recruitment in Neotropical forests? Evidence from Panama. Ecology 78(3), 941-946.
Augspurger, C.K., 1984a. Light requirements of neotropical tree seedlings: A comparative study
of growth and survival. J. Ecol. 72(3), 777-795.
Augspurger, C.K., 1984b. Seedling survival of tropical tree species: Interactions of dispersal
distance, light gaps, and pathogens. Ecology 65(6), 1705-1712.
Balee, W., 1989. The culture of Amazonian forests. In: Posey, D.A., Balee, W. (Eds.), Resource
Management in Amazonia: Indigenous and Folk Strategies. Advances in Economic Botany
No. 7, New York Botanical Garden, New York, NY, pp. 1-21.
Bancroft, T.A., 1968. Topics in intermediate statistical methods. Iowa State University Press,
Ames, Iowa, USA.
Blanton, C.M., Ewel, J.J., 1985. Leaf-cutting ant herbivory in successional and agricultural
tropical ecosystems. Ecology 66(3), 861-869.
Borges, S.H., Stouffer, P.C., 1999. Bird communities in two types of anthropogenic successional
vegetation in Central Amazonia. The Condor 101, 529-536.
Brokaw, N.V.L., 1982. The definition of treefall gap and its effect on measures of forest
dynamics. Biotropica 14(2), 158-160.
Brokaw, N. V. L., 1985. Gap-phase regeneration in a tropical forest. Ecology 66(3), 682-687.
Brown, N.D., Whitmore, T.C., 1992. Do dipterocarp seedlings really partition tropical rain forest
gaps? Phil. Trans: Biol. Sci. 335(1275), 369-378.
Burnham, K.P., Anderson, D.R., 2002. Model selection and multi-model inference: a practical
information-theoretic approach. Second edition. Springer-Verlag, New York, New York,
Ceccon, E., Huante, P., Campo, J., 2003. Effects of nitrogen and phosphorus fertilization on the
survival and recruitment of seedlings of dominant tree species in two abandoned tropical
dry forests in Yucatan, Mexico. For. Ecol. Manage. 182(1), 287-402.
Chabot, B.F., Hicks, D.J., 1982. The ecology of leaf life spans. Ann. Rev. Ecol. Syst. 13, 229-
Chazdon, R.L., Pearcy, R.W., Lee, D.W., Fetcher, N., 1996. Photosynthetic responses of tropical
forest plants to contrasting light environments. In: Mulkey, S.S., Chazdon, R.L., Smith,
A.P. (Eds.), Tropical Forest Plant Ecophysiology Chapman and Hall, New York, NY, pp.
Cintra, R., Horna, V., 1997. Seed and seedling survival of the palm Astrocaryum murumuru and
the legume Dipteryx micrantha in gaps in Amazonian forest. J. Trop. Ecol. 13(2), 257-
Clay, J. W., 1997. Brazil nuts: the use of a keystone species for conservation and development.
In: Freese, C.H. (Ed.), Harvesting Wild Species: Implications for Biodiversity
Conservation. The John Hopkins University Press: Baltimore, MD, pp. 246-282.
Da Silva, J.M.C., Uhl, C., Murray, G., 1996. Plant succession, landscape management, and the
ecology of frugivorous birds in abandoned Amazonian pastures. Cons. Biol. 10(2), 491-
Denevan, W.M., 1992. The pristine myth: The landscape of the Americas in 1492. Annals of the
Association of American Geographers 82(3), 369-385.
Denslow, J.S., Schultz, J.C., Vitousek, P.M., Strain, B.R., 1990. Growth response of tropical
shrubs to treefall gap environments. Ecology 71(1), 165-179.
Dufour, D.L., 1990. Use of tropical rainforests by Native Amazonians. BioScience 40(9), 652-
EMBRAPA (Empresa Brasileira de Pesquisa Agropecuaria), 1997. Manual de Metodos de
Analise de Solos. Segunda edicgo. Centro Nacional de Pesquisa de Solos, Rio de Janeiro,
Ferguson, B.G., Vandermeer, J., Morales, H., Griffith, D.M., 2003. Post-agricultural succession
in El Peten, Guatemala. Cons. Biol. 17(3), 818-828.
Forget, P-M., 1994. Recruitment pattern of Voucapoua americana (Caesalpiniaceae), a rodent-
dispersed tree species in French Guiana. Biotropica 26, 408-419.
Forget, P-M., Milleron, T., Feer, F., Henry, O., Dubost, G., 2000. Effects of dispersal pattern and
mammalian herbivores on seedling recruitment for Virola michelii (Myristicaceae) in
French Guiana. Biotropica 32(3), 452-462.
Fujisaka, S., Escobar, G., Veneklaas, E., 1998. Plant community diversity relative to human land
uses in an Amazon forest colony. Biodiversity and Conservation 7, 41-57.
Gomes, C.V.A., 2001. Dynamics of land use in an Amazonian Extractive Reserve: The Case of
the Chico Mendes Extractive Reserve in Acre, Brazil. Thesis, University of Florida,
Guariguata, M.R., Ostertag, R., 2001. Neotropical secondary forest succession: changes in
structural and functional characteristics. For. Ecol. Manage. 148, 185-206.
Haines, B. L., 1978. Element and energy flows through colonies of the leaf-cutting ant, Atta
colombica, in Panama. Biotropica 10(4), 270-277.
Heckenberger, M.J., Kuikuro, A., Kuikuro, U.T., Russell, J.C., Schmidt, M., Fausto, C.
Franchetto, B., 2003. Amazonia 1492: Pristine forest or cultural parkland? Science 301
Holdridge, L.R., 1978. Ecologia basada en zonas de vida. Institute Interamericano de Ciencias
Agricolas, San Jose, Costa Rica.
Howe, H.F., 1990. Survival and growth of juvenile Virola surinamensis in Panama: Effects of
herbivory and canopy closure. J. Trop. Ecol. 6(3), 259-280.
Hulme, P.E., 1996. Herbivores and the performance of grassland plants: A comparison of
arthropod, mollusc and rodent herbivory. J. Ecol. 84(1), 43-51.
Huber, J., 1910. Mattas e madeiras amaz6nicas. Bol. Mus. Paraese Hist. Nat. 6, 91-225.
IMAC Instituto de Meio Ambiente do Acre, 1991. Atlas geografico ambiental do Acre. IMAC,
Rio Branco, Acre, Brasil.
ITTO, 2002. ITTO guidelines for the restoration, management, and rehabilitation of degraded
and secondary tropical forests. ITTO Policy Development Series No. 13. International
Tropical Timber Organization.
Jansen, P.A., 2003. Scatterhoarding and tree regeneration: Ecology of nut dispersal in a
neotropical rainforest. PhD Thesis, Wageningen University, The Netherlands.
Jansen, P.A., Zuidema, P.A., 2001. Logging, seed dispersal by vertebrates, and the natural
regeneration of tropical timber trees. In: Fimbel, R.A., Robinson, J.G., Grajal, A. (Eds).
The cutting edge. Conserving wildlife in logged tropical forests. Columbia University
Press, New York, pp. 35-59.
Janzen, D.H., 1970. Herbivores and the number of tree species in tropical forests. The American
Naturalist. 104 (940), 501-528.
Janzen, D.H., 1990. An abandoned field is not a tree fall gap. Vida Silvestre Neotropical 2(2),
Johnson, D.H., 1999. The insignificance of statistical significance testing. Journal of Wildlife
Management 63(3), 763-772.
Kainer, K.A., Duryea, M.L., de Macedo, N.C., Williams, K., 1998. Brazil nut seedling
establishment and autecology in extractive reserves of Acre, Brazil. Ecol. Appl. 8, 397-
Kitajima, K., 1996. Ecophysiology of tropical tree seedlings. In: Mulkey, S.S., Chazdon, R.L.,
Smith, A.P. (Eds.), Tropical Forest Plant Ecophysiology Chapman and Hall, New York,
NY, pp. 559-592.
Lugo, A.E., Farnworth, E.G., Pool, D., Jerez, P., Kaufman, G., 1973. The impact of the leaf
cutter ant Atta colombica on the energy flow of a tropical wet forest. Ecology 54(6), 1292 -
Marquis, R.J., 1984. Leaf herbivores decrease fitness of a tropical plant. Science 226 (4674),
Mesquita, R.C.G., 1995. The effects of different proportions of canopy opening on the carbon
cycle of a central Amazonian secondary forest. PhD. thesis, University of Georgia, Athens.
Mesquita, R.C.G., Ickes, K., Ganade, G., Williamson, G.B., 2001. Alternative successional
pathways in the Amazon Basin. J. Ecol. 89, 528-537.
Molofsky, J., Fisher, B.L., 1993. Habitat and predation effects on seedling survival and growth in
shade-tolerant tropical trees. Ecology 74(1), 261-265.
Montagnini, F., Mendelsohn, R. 0., 1997. Managing forest fallows: Improving the economics of
swidden agriculture. Ambio 26(2), 118-123.
Mori, S.A., Prance, G.T., 1990. Taxonomy, ecology, and economic botany of the Brazil nut
(Bertholletia excelsa Humb. and Bonpl.: Lecythidaceae). Adv. Econ. Bot. 8, 130-150.
Murphy, L.J., Riley, J.P., 1962. A modified single solution method for determination of
phosphate in natural waters. Anal. Chim. Acta 27, 31-33.
Myers, G.P., Newton, A.C., Melgarejo, 0., 2000. The influence of canopy gap size on natural
regeneration of Brazil nut (Bertholletia excelsa) in Bolivia. For. Ecol. Manage. 127, 119-
Nascimento, H.E.M., Andrade, A.C.S., Camargo, J.L.C., Laurance, W.F., Laurance, S.G.,
Ribeiro, J.E.L., 2006. Effects of the surrounding matrix on tree recruitment in Amazonian
forest fragments. Conservation Biology 20(3), 853-860.
Nations, J.D., Nigh, R.B., 1980. The evolutionary potential of Lacandon Maya sustained-yield
tropical forest agriculture. Journal of Anthropological Research 36, 1-30.
Ortiz, E.G., 1995. Survival in a nutshell. Americas 47, 7-12.
Pefia-Claros, M., De Boo, H., 2002. The effect of forest successional stage on seed removal of
tropical rain forest tree species. J. Trop. Ecol. 18, 261-274.
Peres, C. A., Baider, C., 1997. Seed dispersal, spatial distribution and population structure of
Brazil nut trees (Bertholletia excelsa) in Southeastern Amazonia. J. Trop. Ecol. 13(4), 595-
Peres, C.A., Baider, C., Zuidema, P.A., Wadt, L.H.O., Kainer, K.A., Gomes-Silva, D.A.P.,
Salomdo, R.P., Simies, L.L., Franciosi, E.R.N., Valverde, F.C., Gribel, R., Shepard Jr.,
G.H., Kanashiro, M., Coventry, P., Yu, D.W., Watkinson, A.R., Freckleton, R.P., 2003.
Demographic threats to the sustainability of Brazil nut exploitation. Science 302, 2112-
Peres, C.A., Schiesari, L.C., Dias-Leme, C.L., 1997. Vertebrate predation of Brazil-nuts
(Bertholletia excelsa, Lecythidaceae), an agouti-dispersed Amazonian seed crop: A test of
the escape hypothesis. J. Trop. Ecol. 13(1), 69-79.
Poorter, L., 1999. Growth responses of 15 rain-forest tree species to a light gradient: The relative
importance of morphological and physiological traits. Func. Ecol. 13(3), 396-410.
Popma, J., Bongers, F., 1991. Acclimation of seedlings of three Mexican tropical rain forest tree
species to a change in light availability. J. Trop. Ecology 7(1), 85-97.
Posey, D.A., 1985. Indigenous management of tropical forest ecosystems: The case of the
Kayap6 Indians of the Brazilian Amazon. Agroforestry Systems 3, 139-158.
Prance, G.T., Mori, S.A., 1978. Observations on the fruits and seeds of neotropical
Lecythidaceae. Brittonia 30(1), 21-23.
Rey, P.J., Alcantara, J.M., 2000. Recruitment dynamics of a fleshy-fruited plant (Olea
europaea): Connecting patterns of seed dispersal to seedling establishment. J. Ecol., 88(4),
Saldarriaga, J.G., West, D.C., Tharp, M.L., Uhl, C., 1988. Long-term chronosequence of forest
succession in the upper Rio Negro of Colombia and Venezuela. J. Ecology 76, 938-958.
Sanchez-Cordero, V., Martinez-Gallardo, R., 1998. Postdispersal fruit and seed removal by
forest-dwelling rodents in a lowland rainforest in Mexico. J. Trop. Ecol. 14(2), 139-151.
SAS Institute, Inc., 2004. SAS. Version 9.1. SAS, Cary, North Carolina, USA.
Schupp, E.W., 1988. Seed and early seedling predation in the forest understory and in treefall
gaps. Oikos 51, 71-78.
Schupp, E.W., Frost, E.J., 1989. Differential predation of Welfia georgii seeds in treefall gaps
and the forest understory. Biotropica 21(3), 200-203.
Schupp, E.W., Howe, H.F., Augspurger, C.K., Levey, D.J., 1989. Arrival and survival in tropical
treefall gaps. Ecology 70(3), 562-564.
Serrano, R.O.P., 2005. Regeneracgo e estrutura populacional de Bertholletia excelsa (H.B.K.) em
areas com diferentes hist6ricos de ocupacgo, no Vale do Rio Acre (Brasil). Dissertacgo,
Universidade Federal do Acre, Acre, Brasil.
Sork, V.L., 1987. Effects of predation and light on seedling establishment in Gustavia superba.
Ecology 68(5), 1341-1350.
SPSS Inc., 1999. SPSS Version 10.0. SPSS, Inc., Chicago, IL.
Tabarelli, M., Mantovani W., 1996. Remocgo de sementes de Bertholletia excelsa
(Lecythidaceae) por animals em uma floresta de terra firme na Amaz6nia Central, Brasil.
Revista Brasileira de Biologia 56(4), 755-760.
Uhl, C., Buschbacher, R., Serrdo, E. A. S., 1988. Abandoned pastures in Eastern Amazonia. I.
patterns of plant succession. J. Ecol. 76, 663-681.
Uhl, C., Clark, K., Clark, H., Murphy, P., 1981. Early plant succession after cutting and burning
in the upper Rio Negro region of the Amazon Basin. J. Ecol. 69(2), 631-649.
Uhl, C. Clark, H., Clark, K., Maquirino, P., 1982. Successional patterns associated with slash-
and-burn agriculture in the upper Rio Negro region of the Amazon Basin. Biotropica 14(4),
Uhl, C., Jordan, C., 1984. Succession and nutrient dynamics following forest cutting and burning
in Amazonia. Ecology 65(5), 1476-1490.
Vasconcelos, H.L., Cherrett, J.M., 1995. Changes in leaf-cutting ant populations (Formicidae:
Attini) after the clearing of mature forest in Brazilian Amazonia. Studies on Neotropical
Fauna and Environment 30, 107-113.
Vasconcelos, H.L., Cherrett, J.M., 1997. Leaf-cutting ants and early forest regeneration in
Central Amazonia: Effects of herbivory on tree seedling establishment. J. Trop. Ecol. 13(3)
Viana, V.M., Mello, R.A., Moraes, L.M., Mendes, N.T., 1998. Ecologia e manejo de populaces
de castanha-do-Para em reserves extrativistas Xapuri, Estado do Acre. In: Gascon, C.,
Moutinho, P. (Eds.), Floresta Amaz6nica: Dindmica, Regeneracgo e Manejo. Institute
Nacional de Pesquisa Amaz6nia, Manaus, Brazil, pp. 277-292.
Vieira, S., Trumbore, S., Camargo, P.B., Selhorst, D., Chambers, J.Q., Higuchi, N., Martinelli,
L.A., 2005. Slow growth rates of Amazonian trees: Consequences for carbon cycling.
Proceedings of the National Academy of Sciences 102, 18502-18507.
Wadt, L.H.O., Kainer, K.A., Gomes-Silva, D.A.P., 2005. Population structure and nut yield of a
Bertholletia excelsa stand in Southwestern Amazonia. For. Ecol. Manage. 211, 371-384.
Whitmore, T.C., 1989. Canopy gaps and the two major groups of forest trees. Ecology 70(3),
ZEE Zoneamento Ecol6gico-Econ6mico do Acre, 2000. Recursos Naturais e Meio Ambiente,
vol. I, Secretaria de Estado de Ci6ncia, Tecnologia e Meio Ambiente, Rio Branco, Acre,
Zuidema, P.A., 2003. Ecology and Management of the Brazil nut tree (Bertholletia excelsa).
PROMAB Scientific Series 6, PROMAB, Utrecht, Netherlands.
Zuidema, P.A., Boot, R.G.A., 2002. Demography of the Brazil nut tree (Bertholletia excelsa) in
the Bolivian Amazon: Impact of seed extraction on recruitment and population dynamics.
J. Trop. Ecol. 18, 1-31
Jamie Nicole Cotta was born and raised in Riverside, California, and began her career in
tropical conservation at the University of California, San Diego, where she received a B.S. in
Ecology, Behavior and Evolution. During her studies she conducted behavior observations of
bonobo chimpanzees (Pan paniscus), at the San Diego Zoo, and pygmy lorises (Nycticebus
pygmaeus), at the Center for Reproduction of Endangered Species. She next assisted a Cornell
post-doc with behavioral research on orange-fronted parakeets (Aratinga canicularis) in Santa
Rosa National Park, Costa Rica. Her undergraduate studies concluded with a semester of
Tropical Biology in Monteverde, Costa Rica, where she first became interested in sustainable
tropical forest management. She is most proud of her work as a research assistant for the School
for Field Studies, in Yungaburra, Queensland, Australia, where she managed a tropical plant
nursery and helped to design a restoration planting with a local community tree planting
organization (TREAT). Her tropical research experiences fostered an interest in tropical ecology
and forest restoration in the tropics, leading her to pursue a Masters of Science degree in
Interdisciplinary Ecology at the University of Florida. Upon graduation she plans to apply her
experience and education within an NGO setting, helping businesses and communities adopt
more sustainable practices and lifestyles.