EXAMINING FOREST RECOVERY THROUGH SOIL SURVEYS:

A Study at the Firestone Center for Restoration Ecology in Costa Rica

Garwen Chen and Sarah Mahlab, PO’09

Environmental Analysis Senior Thesis

Jonathan Wright and Rick Hazlett Pomona College 2008

1

ABSTRACT Humans have been altering tropical forest landscapes for centuries, most destructively

through widespread deforestation, which among other things, results in severe soil degradation

and reduced biodiversity. The purpose of this study was to compare the physical and chemical

properties of soil, along with the invertebrate communities in the soil in eight habitats with

different historical levels of human disturbances and in different stages of recovery. In addition,

a plot of naturally regenerated forest was examined in more detail to relate our study to the forest

recovery process. The study was done at the Firestone Center for Restoration Ecology in

southwestern Costa Rica from June 1 to July 31, 2008.

2

TABLE OF CONTENTS Abstract ii Introduction 1 Loss of native habitats due to human disturbance 1 Tropical forest recovery 2 Value of secondary forests 3 Limitations and challenges of forest recovery 4 Land restoration and management 5 Nature and function of soil 6 Table 1: Essential plant nutrients and their function 7 Importance of soil fauna 8 Table 2: Soil invertebrate and microbial functional groups for common ecosystem processes

in the soil 10

Land use impacts on soil 11 Soil restoration and conservation 14 This study 14 Materials and Methods 15 Site description 15 Figure 1: Map of Firestone Center for Restoration Ecology property 16 Habitat types 17 Figure 2: Photos of habitats 18 Study plot 1 18 Figure 3: Study Plot 1 19 Figure 4: Location of sampled bamboo and nonbamboo sites within plot 1 19 Adjacent pastures and Hacienda Baru 20 Overview of methods 21 Field Sampling 21 Physical analysis 22 Chemical analysis 22 Biological analysis 23 Data analysis 23 Results 23 Physical properties 23 Table 3: Physical properties of soil according to habitat onsit 24 Figure 5: Percent sand, silt and clay in soil according to habitat. These percentages were used

to determine the soil texture 24

Figure 6: pH, soil bulk density and moisture content according to habitat 25 Table 4: Physical properties of soils in nonbamboo and bamboo areas of naturally

regenerated forests 25

Chemical properties 25 Table 5: Chemical properties of soil according to habitat. Macronutrients and micronutrients

listed are essential for plant growth 26

Table 6: Important trace elements in soils of nonbamboo and bamboo areas of naturally regenerated forests

27

Bamboo Foliar analysis 27 Table 7: Amount of trace elements in Guadua angustifolia and Bambusa vulgaris leaves 28

3

Offsites 28 Table 8: Physical properties of soil in offsite samples 28 Table 9: Amounts of trace elements in the soils of nearby pastures and Hacienda Baru

transect 29

Invertebrate communities 29 Table 10: Number of individuals in each taxonomic group in litter samples, according to

habitat. Bold taxon represent categories 30

Figure 7: Abundance of major orders in litter sample according to habitat 31 Table 11: Abundance of individuals in dry extraction samples according to order, according

to habitat 32

Figure 8: Abundance of major orders in dry extraction sample according to habitat 33 Figure 9: Simpson’s Index of Diversity in litter and dry extraction samples according to

habitat 34

Figure 10: Species Richness in litter and dry extraction samples according to habitat 35 Table 12: Invertebrate species abundance in nonbamboo and bamboo area 36

Figure 11: Common invertebrate abundance in nonbamboo and bamboo areas of naturally regenerated forest

37

Figure 12: Simpson’s Index of Diversity in litter and dry extraction samples in naturally regenerated secondary forests

38

Figure 13: Species richness in litter and dry extraction samples in naturally regenerated secondary forests

39

Discussion 39 Primary forest vs. grazed pasture 40 Naturally regenerated forest vs. hardwood forest vs. recovering pasture 42 Thick bamboo vs. thin bamboo vs. banana plantation 43 Offsite sampling 46 Naturally regenerated forests 47 Table 12: Comparison of macro and micronutrients between plot 1 nonbamboo and bamboo

soils, other soils in the naturally regenerated forests, and primary forests 48

Plot 1 49 Land use and forest recovery 50 Limitations and Recommendations 52 Future studies 52 Conclusions 53 Acknowledgements 53 References 54 Appendices

4

INTRODUCTION

“Ultimately, the future of a natural ecosystem depends not on protection from humans but on its relationship with the people who inhabit it or share the landscape with it”

–William R. Jordan III, a founder of the field of restoration ecology

Loss of native habitats due to human disturbance

Humans have been altering tropical forest landscapes for centuries. Of all of the

interactions between humans and tropical forests, deforestation is the most detrimental. The

impacts of deforestation have been well documented in terms of the dramatic loss of

biodiversity, increase in greenhouse gas concentrations in the atmosphere, changes in the

hydrological cycle and accelerated soil erosion, among other things. Tropical forests are being

cleared at an alarming rate although our understanding of their ecology is still limited. An

estimated 60% of the world’s tropical forests were classified as degraded forests in 2000. This

included secondary forests, degraded primary forests, and degraded forest land (ITTO 2002).

Today, rainforests cover only 6% of the total land surface on the globe while at one point, they

covered more than 14% of the world’s land mass (Rain-tree 1996). Between 2000 and 2005,

tropical deforestation rates increased by 8.5%. During that same time period, over 10.4 million

hectares of tropical rainforest were destroyed worldwide (WRI 2008). Although net forest areas

in North American and European countries are more stable or increasing, forest area continues to

decline in other continents (Figure 1). It has been estimated that the remaining forest will

disappear within 40 years at the current rate of destruction (Rain-tree 1996).

Figure 1: Annual net forest loss by region in 2007 (WRI 2008).

In Costa Rica, the rate of conversion of forest to other uses was 600 square km per year,

amounting to an annual loss of 3.9% of their tropical forest in 1982. Between 1941 and 1983,

5

Costa R en the

oll 1999).

of plant litter inputs, and abundance of predators (Camilo and

Zou 2001). The loss of habitat will have a greater effect on terrestrial ecosystems in the tropics

climate change, increasing carbon dioxide levels, or invasive species

(Sala e h

y

ession

such as wind storms, hurricanes, cyclones,

lightnin t of

ata

ica lost 83% of its tropical forested land (Reiners et al.1994). Cattle grazing has be

main cause of deforestation in Latin American for the past few decades (Amelung & Diehl

1992). Costa Rica was once almost entirely forested, but is now almost half covered by pasture

(WRI 1994). It has experienced excessive erosion and highly reduced water quality (H

As seen in Costa Rica, lowland tropical rainforests are particularly vulnerable to habitat loss due

to its accessibility to the expanding human population and our needs.

While natural disturbances are integral parts of ecosystem processes which disrupt

organism populations, communities, and ecosystem structure, human disturbances remove

significant biomass by changing the physical environment or resource availability at unnatural

and hazardous rates. The conversion of tropical forest to pasture for cattle farming or

agricultural purposes removes resources from the basal trophic levels, which directly influences

the abundance and diversity of soil fauna by altering soil physical properties, microclimate

conditions, quantity and quality

than other factors such as

t al., 2000). Many of the consequences of deforestation are deeply troubling althoug

many people are most concerned about the loss of global biodiversity, since the tropics support

most of the diversity on Earth. It is estimated that 50,000 species are lost each year due to

deforestation (Rain-tree 1996).

Tropical forest recovery

Traditionally, forest recovery is viewed in the context of primary versus secondar

succession. Primary succession follows disturbances where soils are removed or buried, leaving

substrate without organic matter and open to colonization (Walker 1999). Secondary succ

follows disturbances that leave the soil intact,

g, fire or biotic disturbances (Whitmore & Burslem 1998). Depending on the exten

soil degradation and remnant vegetation, anthropogenic disturbances can cause either type of

succession (Chazdon 2003). Remnant vegetation plays a crucial role in forest recovery,

promoting rapid increases in species richness, tree density and aboveground biomass (Guarigu

& Ostertag 2001). Other factors that affect forest recovery include the spatial extent and

frequency of disturbances (Chazdon 2003).

6

Different methods can be used to measure forest recovery. Many studies focus on

structural measures of recovery that directly relate to ecosystem function such as basal area,

aboveground biomass, tree height, or stem density (Chazdon 2003). Changes in canopy

structure, frequency and size of canopy gaps, and light availability during forest recovery can

also be trients,

& Ostertag 2001). Pastures that have been subject to overgrazing, repeated weeding

and burning over long periods (>10 years) or have been bulldozed have very slow recovery or no

g-term use of pastures decreases the number of sprouting roots

and red

hrough

also

occur naturally given that seedling production is sufficient. The rate of forest recovery depend a

the seedlings including seed dispersal, seed predation, seedling

germin spects that

used to measure forest recovery (Denslow & Guzman 2000). Additionally, soil nu

carbon stocks, and nutrient cycling components can be used as indicators of recovery of

ecosystem functions (Hughes et al. 2002). It is practical to use multi-variate approaches to

examine patterns of species composition across stands differing in soils and land-use history.

The situation is complex, however, since recovery processes are influenced by the interactions

between historical land use and the natural forests (Thompson et al. 2002).

There are important biotic and abiotic factors that interact strongly with age since

abandonment and affect succession processes. Low prior-land use intensity, small recovering

areas, areas with fertile soil and have nearby remnant forests increase recovery rates dramatically

(Guariguata

recovery (Uhl et al. 1988). Lon

uces the number of tree seeds in soil, both factors essential for regeneration (Bertoncini

and Rodrigues, 2008). The recovery of aboveground biomass is inversely correlated with the

number of years of land use prior to abandonment, but not correlated with forest age (Hughes et

al. 1999).

Naturally regenerated forests

Natural reforestation refers to the process of allowing a given area or ecosystem that had

once been degraded, depleted, or destroyed to regenerate and return to its natural state t

time. These new re-grown forests are known as secondary forests. Third growth forests can

result from severe disruptions in second growth forests. It is assumed that regeneration will

number of factors related to

ations, shade tolerance, survival and growth of seedlings. Other processes and a

affect forest succession include pollination, soil nutrient availability, soil compaction,

7

competition with pasture grass, seasonal drought, low rates of seed colonization, and abiotic

factors such as light, temperature, and moisture (Uhl 1987, Reiners et al. 1994, Aide et al. 199

Value of secondary forests

Secondary forests are important assets for conservation purposes. Ecological

characteristics of secondary forests allow secondary forests to contribute to the conservation o

biodiversity. Secondary forests have high growth rates, provide conditions suitable for

recolonization of mycorrhizae, and produce seeds that are widely dispersed and remain viable in

soil for several years. Secondary species can often germinate and grow on impoverished soils.

Secondary forests serve as foster ecosystems for valuable late secondary species, reduce pest

populations, restore site productivity and provide conditions that help improve soil and w

quality, conserving nutrients, moisture, and soil organic material, all which contribute to the

conservation

6).

f

ater

of biodiversity. Serving as nutrient sinks, they accumulate nutrients rapidly with

time in vegetation, litter, and soil (Brown and Lugo, 1990). Young trees tend to accumulate

ich tend to reuse nutrients (Bowen and Nambiar

1984).

n and

ere

d

rsity

within the first 10-15 years of recovery was dominated by herbaceous vegetation. However,

more nutrients than old trees of older forests wh

The litter production in younger secondary forests (<20 years) is a higher fraction of the

net primary productivity; thus relatively more organic matter is produced and transferred to the

soil than is stored in above-ground vegetation. As a result, recovery of organic matter in soil of

secondary forests is relatively fast (~50 years). Nutrients are accumulated in secondary

vegetation and returned quickly by litterfall and decomposition for uptake by roots (Brow

Lugo 1990).

Limitations and challenges to forest recovery

Because successful recovery in tropical forests is dependent on a number of factors, th

are limits to recovery in tropical forests. Following abandonment of agriculture, degraded land

may fail to initiate recovery in circumstances with infertile soil, lack of residual vegetation an

sources of seed dispersal. In addition, invasive and exotic species may impede recovery of native

vegetation (Denslow et al. 2001). Aide et al. (1996) found that initial forest recovery in

abandoned pastures was slower than recovery following other human or natural disturbances and

that species invading in pasture recovery were different than the species in a natural recovery.

Their study found a strong correlation between plant diversity and recovery age. Plant dive

8

once shrubs and small trees began to establish themselves, the forest began to recover rap

and more tree growth was facilitated. When natural recovery cannot be initiated, restoration

efforts are needed to recover at least some of the fun

idly

ctions and diversity of the heavily degraded

tropical forests. In extreme cases, intervention is needed to restore species composition and soil

ely linked with recovery of aboveground biomass (Chazdon

2003). everely

he

y,

g

d

oration (Chazdon 2003). By examining land use patterns around the world, it is possible

to make connections between patterns in climate, hydrology, sedimentology, atmospheric

hemistry, sustainable productivity, soil properties and biodiversity (Reiners et al. 1994).

at in soils with similar parent material, the major differences in

their pr

al

fertility. Soil fertility recovery is clos

Restoration of soil fertility may be a prerequisite for forest recovery on sites with s

degraded soils and successional recovery is constrained by soil fertility and texture across

regions as well as across soil types within a region (Chazdon 2003). While wildlife and

vegetation management is central to tropical forest recovery and management, initial success in

tropical reforestation begins with land restoration.

Land restoration and management

It is important to understand the basic biology of an ecosystem in order to design

effective restoration strategies. Successful wildlife and vegetation recovery in tropical forests

depends directly on successful soil recovery. While the act of forest conversion to pasture is still

extremely pervasive, little attention has been devoted to finding what happens to the land, t

soil and the living communities in deforested areas that have been allowed to recover. Toda

even as widespread deforestation continues, some of the previously deforested land is growin

back into secondary forest, some is being used for tillage agriculture and the majority is being

transformed into actively grazed cattle pasture (Reiners et al. 1994). Rates of planting forests an

trees are increasing by 2.8 million ha/year, for purposes of production, as well as conservation

and rest

c

Reiners et al. (1994) found th

operties could be attributed to land use. To determine the best restoration strategies that

are site-specific and promote recovery processes, data on local ecology is necessary. Tropic

land restoration must first begin with data collection of the effects and existing conditions of the

soils.

9

Nature and function of soil

Soil is defined as the naturally occurring, unconsolidated cover of the earth, compose

air, water, mineral and organic components. Soil covers 1.2 x 10

d of 8 square kilometers of the

Earth’s y depends

y

lant growth and also the biogeochemical functions for the ecosystem

einers et al. 1994). Phosphorous availability, aluminum toxicity, the depth of water table, the

mount and arrangement of pores of different sizes, and the availability of base-metal cations

nd micronutrients such as boron and zinc in the soil, affect plant species composition most in

opical rainforest (Sollins 1998). The highest levels of both carbon and nitrogen, important

r in the top ten centimeters of forest soil (Reiners et

Plan nutrients for their biological processes, without which they

surface (Freckman et al. 1997). Because the survival of the ecosystem ultimatel

on the fertility of the soil, the value of soils is determined by its capacity to grow plants. Soils

function as a storehouse for water, air, and nutrients, and at the same time, remain permeable.

Furthermore, soils provide support to ground large trees in the land. The diverse and flexible

properties of soils are due to their physical and chemical properties along with biological

functions of the microbes in soil. Soil has often been referred to as our “most precious non-

renewable resource” (Giller 1996).

Soil is often overlooked when admiring the wonders of the tropical rainforest.

Supporting the most productive ecosystem on the planet, soils in the tropics tend to be highl

infertile, due to the high rates of rainfall that leach important nutrients. Soils in forested tropical

regions are classified as laterites, which fall into the soil order of Pedalfers. These soils are a

reddish brown color and contain abundant metal cations (Schaetzl and Anderson 2007).

Leaching by rain leaves high levels of less soluble minerals, such as Al3+ and Fe3+, in the A

horizon, which can often be toxic for plant roots (Reiners et al. 1994).

Soil properties greatly influence the species composition in the plant communities above

ground and by extension the species composition of the entire forest community. They are

indicators of fertility for p

(R

a

a

tr

nutrients for plant growth and function, occu

al. 1994). ts require 13 essential

cannot survive (Table 1).

10

Table 1: Essential function t

plant nutrients and theirNutrien Function in plant Nitrogen Proteins, amino acids Phosphorous

yst, ion transport alcium Cell wall component

Magne

Nucleic acids, ATP Potassium CatalC

sium Part of chlorophyll Sulfur Amino acids Iron Chlorophyll synthesis Copper Component of enzymes Manganese Activates enzymes Zinc Activates enzymes Boron Cell wall component Mo Involved in N fixation Chlorine Photosynthesis reaction (Motavallie et al. 2008)

In addition, plants gradually alter some properties of the soil in which they grow, so i

difficult to tell whether plants prefer certa

t is

in kinds of soil or create them (Sollins 1998). Plants

can cha f

roots

decreas

ontent,

ers,

of

r the forest. Because soils are thermally buffered from atmospheric temperature changes and

ases with depth (Giller 1996), it provides a constant habitat for

nge soil properties through root penetration, water extraction, anchorage and exudation o

compounds in the rhizosphere. In addition, dead roots and litter provide carbon directly into the

soil ecosystem (Angers and Caron 1998). Also, plant cover physically protects the soil from

rainfall and compaction, thus influencing the physical properties. A greater density of plants

results in a greater abundance of macropores in the soil, resulting in increased soil cohesion and

strength and decreased water content (Angers and Caron 1998). The presence of plant

es the possibility of soil erosion.

Giller (1996) found that physical properties of the soil, such as soil type, water c

relative humidity, nutrient concentration, texture, pore space, pH, soil atmosphere, depth of litter

layers, overlying vegetation, temperature variation and degree of bioturbation by macrofauna,

vary horizontally in a landscape. Factors such as soil texture, structural diversity of litter lay

pore size and soil atmosphere vary vertically as well. The physical and chemically properties

soil mirror the patchy nature of the high biodiversity in tropical forests (Giller 1996).

Instead of considering soils an abiotic factor in ecological communities, soil should be

viewed as living communities in themselves. Soils around the world support a diverse

community of invertebrates that invisibly perform many extremely important ecological services

fo

temperature variation decre

11

inverte to

e

e the

ready known, much of the diversity in soil invertebrates remain undescribed

(Freckm

rties

d

1996). In addition, the roles of soil microorganisms include biochemical decomposition and

onal

. 1991, 1 iota are c partially responsible

rnover, formation and decay of soil organic matte ion, N2O and N2

production, CO2 production lopment and stabilization, transport of

materials (particles, organisms or gas), oxygenation of soils, tra ation of

higher lean al. 1997).

brates. The O horizon is dominated by organic, decomposing material as opposed

mineral matter, which dominates the A layer of the soil and those below. The O layer can b

absent in areas with high decomposition rates, which is often the case in the tropics. Becaus

soils are thermally buffered against extreme atmospheric temperatures and moisture, they can

provide an ameliorated microclimate favorable for invertebrates (Schaetzl and Anderson 2007).

Importance of soil fauna

Diversity in soil invertebrates can mainly be attributed to high amounts of niche

partitioning and specialization in the tropics (Giller 1996). While extremely high levels of

diversity are al

an et al. 1997). Soil organisms play important roles in nutrient cycling, food webs, and

soil properties of tropical ecosystems. The ecological interactions between soil fauna, bacterial,

fungi, plants, and the stratum of soil in which they are found facilitate these processes. The

recycling of materials within the tropical ecosystem through soil food webs is closely related to

ecosystem resilience (Moore et al. 1988, 1993). Furthermore, structural properties and

biogeochemical fertility of soils are directly related to the biodiversity found within them (Silver

et al., 1996).

Although the functional roles of soil organisms in maintaining soil fertility and prope

are poorly understood, the close correlation between forest diversity and soil fertility attests to

the importance of maintaining a healthy and diverse soil biota (Swift and Anderson 1993). The

role of soil macrofauna include breakdown of physical structures such as wood and leaf litter an

inoculation of necro mass (Schaller 1968). Soil organisms are the primary decomposers in

forests ecosystems, decomposing 60 to 90% of terrestrial primary production in the soil (Giller

regulation of nutrient pools that facilitate vegetation establishment and further successi

stages (Lodge et al 994). To summarize, soil b ompletely or

for nutrient tu r, nitrogen fixat

nsport and degrad

and consumption, soil deve

pollutants, food for organisms and provisioning of c water (Freckman et

12

Table 2: Soil invertebrate and m on ecosystem processes in the soil Functional group ive ecosystem Organisms

icrobial functional groups for commRepresentatprocesses

Bioturbators cles and micro-

Oligochaetes, crustaceans, mollusks,

s, termites, t roots

Mix and redistribute organic matter, partiorganisms to depth earthworm

ants, planPrimary producers e Plants, plant roots

Shredders Rip and tear organic material to prepare it for decay by other organisms

Isopods, millipedes

Decom

Create new biomass, stabilizsoils

posers Return carbon and other essential nutrients to soils for primary production

Bacteria, fungi

Nitrogen Fixers Convert N2 into usable N Bacteria CO2 producers, trace gas production

Respiration, transfer of form of carbon, denitirication, N2O production

Bacteria

(Freckman et al. 1997)

Termites, millipedes, earthworms and isopods are among the most important decompose

groups (Brussaard 1997). More specifically, earthworms are vital not only because of their

decomposition capacity but also because of their role in redistributing and burying of plan

and regulation of soil carbon and nutrient levels. Earthworm castes are usually enriched with

nutrients such as C, N or P (Aira et al. 2008). They play an important role in nutrient cyc

and bioturbations, such as the creation of channels, pores and aggregates and mounds that

influence the transport of gases and water in the soil. On average, earthworms produce up to 2

tunnel openings per squ

r

t litter

ling,

20

are meter (3- 5 mm in diameter) and bring 10-500 tons per hectare per

year of

s

y

underground soil to the surface (Pimentel and Kounang 1998). As a result, earthworms

are important keystone species that indicate ecosystem status. Ants contribute as well to

bioturbations in the soil (Brussaard 1997), thus influencing the hydrology, aeration and gaseou

composition of the soil in which they live, all properties which are essential for primar

production and organic and nutrient turnover (Brussaard 1997). In addition, soil fauna influence

the percentage of litter mass loss and nitrogen mineralization rates at any given point in time

(González et al. 2001).

13

Physically, burrowing soil invertebrates aerate the soil by digging tunnels and increasi

rates of gas exchange. All of the important invertebrate groups were found in much higher

densities in the forested habitats than in the pasture and the other habitats with current human

impacts or recent histories of human land disturbance (Schaetzl and Anderson 2007).

Nematodes, unsegmented worms that occur in all sizes and trophic levels, are the most abundant

soil organism. They aid in nutrient cycling, help regulate microbial soil populations and transport

bacteria and fungus from one part of the soil to another.

Invertebrates fill a number of niches in the soil foodweb. They serve as root feeder

grazers, shredders and herbivore

ng

s,

s. Collembola and mites eat bacteria and fungi on and near the

roots o

es.

s

e mass of

e O or

per soil

sequently more susceptible to changes in

the physical structure and properties of the soil (Vos and Kooistra 1994). Soil fauna composition

thus varies with depth, since they have to adapt to different environmental pressures.

oil macroinvertebrates can also serve as bioindicators for soil classification,

(Giller 1996). Nematodes are the most numerous and diverse group of

soil org

f plants. The shredders reside closer to the surface of the soil. They shred plant litter as

they feed. Common shredders are millipedes, isopods, termites, some mites and cockroach

Common predators in the soil include spiders, ground beetles, centipedes, pseudoscorpions, ants

and mites, which prey on nematodes, earthworms and many other groups. All of these organism

mix and aerate the soil as they feed. Furthermore, soil organisms are important prey items for

other species of the tropical ecosystem such as birds, mammals, reptiles, and amphibians

(Schaetzl and Anderson 2007).

Giller et. al. (1996) found that there is a strong inverse relationship between th

accumulated organic material on the soil surface and total soil faunal biomass, due to the

abundance of available nutrients. Organisms in the litter layer or top-most layer (either th

A layer) are exposed to a more variable microclimate while organisms that live in dee

layers live in a more uniform environment but are con

S

disturbances and pollution

anisms and can be used as indicators for soil disturbances by pollution, erosion,

pesticides and water quality (Brussaard 1997). Nevertheless, soil fauna responses to

anthropogenic disturbances are poorly understood.

14

Land use impacts on soil

Previous studies looking at how land use impacts soil properties and invertebrate

communities found that conversion of forest to pasture causes substantial changes in soil acidity

base exchange, porosity, nitrogen mineralization and nitrification (Reiners et al. 1994). In

actively grazed pastures, trampling by cattle as a physical force reduces pore space and increas

soil bulk density. Due to this compaction pressure, pastures display a lower rate of soil mixin

by arthropods and small mammals. Also, actively grazed pastures were found to be lower in

acidity and display lower rates of productivity than primary forest. Soil moisture fluctuated

more in the open pasture due to its high levels of sun exposure and thus could influence the r

of decomposition and nutrient cycling. The nature of the vegetation also influenced t

,

es

g

ates

he rates of

nutrien

r,

. Most evidence shows that disturbances, especially in land

manage nd

and

d

t cycling, since grasses cycle base cations more rapidly and generate fewer and less

persistent organic acids than do forests. Johnson et al. (1997) found that total soil carbon was

progressively less in grassland plots compared to carbon abundance in edge plots and intact

forest plots. The same study found that grassland plots had significantly higher iron, coppe

nickel and lead concentrations than forest plots (Johnson et al. 1997). Most importantly, it was

found that land conversion could result in decreased levels of phosphorous, one of the most

limiting of all the macronutrients required for plant growth (Cleveland et al. 2003).

Cleveland et al. (2003) found that when forest was converted to pasture the total soil

organic matter decreased, the amount of radioactive and chemically active trace gas emissions

increased, there was a change in the hydrologic cycle and losses of important limiting elements

were elevated due to leaching

ment, result in reduced diversity in soil communities (Giller 1996). Giller (1996) fou

that quarrying, tillage, increased stocking rates of domestic animals, manure application

burning tussock heath vegetation all lead to changes in community composition and reduce

alpha diversity, defined as the number of species in the area, and biomass in soil invertebrate

communities. Clearing of forest followed by intense cultivation in the tropics led to a decrease in

biomass of 6%, diversity of 17% and taxon richness by 50% of that of the original primary forest

communities (Giller 1996).

Arthropods are easily affected by land use changes because of potential exposure to

unfavorable environmental conditions, such as desiccation, mechanical destruction, soil

compaction or reduced pore volume and disruption of access to food resources. And since

15

biodiversity of soil communities is so important for ecosystem processes, such as decomposition

and nutrient cycling, forest disturbance can be highly detrimental (Giller 1996). Changes

microbial communities because of land use result in a change in the biogeochemical cycles of th

area (Cleveland et al. 2003).

Each year, 75 billion tons of soil is eroded around the world, resulting in a net loss, since

soil forms at an extremely slow rate (Pimentel and Kounang 1998). Deforestation is one of the

leading causes of soil erosion, resulting in land with decreased productivity and diversity of

plants, invertebrates and microorganisms. Areas that lack vegetation cover and roots th

in

e

at anchor

the soil in place are more susceptible to erosion by wind and rain,. In the U.S., cropland

oil due to erosion, while pasture land is second. Each year,

pasture

.05

(nitrogen,

f contamination. Principle approaches include erosion

preven

ssessments

to evaluate current soil conditions and to provide a baseline in which other studies can build on

evaluate recovery process.

experiences the greatest loss of tops

s in the U.S. lose an estimated 6 tons of soil per hectare, although this rate can increase

with overgrazing (Pimentel and Kounanag 1998). By contrast, a forest only loses 0.004 to 0

tons of soil per hectare per year. Sediment that is washed away in erosive processes contains up

to three times the amount of essential macronutrients for plants as the soil left behind

phosphorous, and calcium), compounding the costs of soil biodiversity and fertility.

Soil restoration and conservation

Soil restoration and conservation involves management strategies for restoring soils or

preventing soils from being physically eroded or becoming chemically altered by overuse,

salinization, acidification, or other types o

tion, acidity control, salinity management, prevention and remediation of soil

contamination, mineralization, choice of vegetative cover, and tracking soil organism

populations and species richness. As soil is the basis for the ecosystem, soil restoration and

conservation is important for forest restoration. To allow the ecosystem to thrive, the structural

degradation must first be evaluated and fixed. Thus, it is important to perform soil a

to track and

Soil degradation was found to be a significant barrier in the recovery of abandoned

pastures to forest (Aide et al. 1996). In order to facilitate recovery, the study suggests that

species such as Miconia prasina be planted to initiate canopy cover and provide a favorable

16

microclimate for other tree species. These plants can also provide physical stabilization fro

material. Many restoration efforts can be made to regrow forests from pasture land.

This study

The purpose of this study was to examine the impacts of land use on the soil and th

communities in a tropical rainforest on the Firestone Reserve for Restoration ecology, located in

southwestern Costa Rica. Comparisons were made between the physical properties, chemical

properties and invertebrate communities in eight different habitats with unique land use histories

The eight habitats were primary forest, a naturally regenerated secondary forest, a planted

hardwood secondary forest, a banana plantation, two different species of bamboo, one of which

was being actively managed and harvested, the other which was let to grow free, a pasture that

has been recovering for four years and recently grazed pasture. More specifically, we exam

the soil properties in a study plot that has been used for vege

m root

e soil

.

ined

tation studies in the naturally

regene ts

ersity

ave

chemical

structu

t,

rated forest in more detail. This was done to provide specific soil data for the study plo

so that future studies can use this baseline data to design studies to measure and evaluate the

recovery process of the naturally regenerated areas of the Firestone Reserve. Moisture content,

soil bulk density, relative particle size, color, pH, soil compaction, depth of the organic layer

were among the physical properties measured. The concentrations of many nutrients and trace

elements of the soil parent material were collected. Finally, invertebrate abundance and div

in the litter and the soils were measured at each site to see how human disturbances h

impacted the living soil fauna.

Our study addresses three main questions: 1) How do the soil physical and

re along with soil invertebrate abundance and diversity vary in the eight different habitats?

Specifically, how do physical and chemical properties of soil and invertebrate soil communities

vary between a) the primary forest and the pasture, b) the naturally regenerated secondary fores

the hardwood forest and the recovering pasture, c) the thick bamboo, thin bamboo and banana

plantation, and finally d) how do physical and chemical factors compare within the offsite

samples. 2) How do the physical and chemical properties, as well as the invertebrate

communities vary between bamboo and nonbamboo habitats in Plot 1? 3) Is Plot 1

representative of the larger naturally regenerated secondary forest on the FCRE?

17

This study also aims at providing baseline data from which future ecological studies can

be built and to which more soil data can be added. Based on previous studies, it was pr

that arthropod diversity and abundance would be lower in habitats with more intense land use.

Grazed pasture, thick bamboo, and recovering pasture have had the highest human impacts of all

the habitats, thus it was expected that these would show the greatest differences from

edicted

the other

less disturbed areas. Conversely, it was expected that the primary forest would show the greatest

vertebrate diversity and abundance. Since all of the habitats lie on soil with the similar

arental material, physical properties were not expected to vary widely. Compaction and soil

bulk density were the factors that were expected to vary the most, since direct use and trampling

ould cause these to be changed. The pH was expected to be lower in the pasture compared to

the forests. In Plot 1, we hypothesized that the chemical properties of the bamboo and

onbamboo habitats would be different because it was initially observed that less vegetation was

rowing around the bamboo areas of the study plot. We also hypothesized that the study plot 1

would be representative of the greater naturally regenerated forest areas of FCRE.

in

p

w

n

g

18

MATERIALS AND METHODS

igure 2: Overview of study

onduct a soil

. The majority of the

s,

was cleared in the 1950s and 1960s for cattle pasture. In 1993, the land began its recovery

F

Site description

We sampled 8 sites in the 60-hectare property known as the Firestone Center for

Restoration Ecology (FCRE) in southwest Costa Rica. The area receives around 4424.2 mm of

precipitation annually (Hacienda Baru 2006). The property was acquired by Pitzer College in

2005 and is located near the town of Dominical on the southwest coast of Costa Rica. Serving as

a site for research and restoration efforts, the Firestone Center is an ideal site to c

survey due to its changing land use over the years and the restoration efforts

property is comprised of previously deforested land and is thus in some stage of

recovery/succession, although there are areas of remaining primary forest.

All of the land, except a small tract of forest along the riparian zones and on steep slope

19

process under efforts of land restoration and sustainable forestry in most parts of the property.

Thus, except for the primary forest, all the land was grazed for the same amount of time unt

recovery began to occur, although recovery was initiated at different points in the different

habitats. Geologically, the land in the higher elevations of the property lies on diorite while

areas downslope are shales that are much richer in nutrient baring clays than diorites (pers

commun. Richard Hazlett 2008). A large

il

most

.

amount of the diorite-derived soils have eroded

ownhill and mixed with the shale soils.

d

Figure 3 : Map of Firestone Center for Restoration Ecology property

20

Habitat Types

Primary Forest: Defined in this study as area that has never been significantly impacted by

humans. Deforestation has never occurred here. It occurs the eastern part of the FCRE property,

along Cocoa stream and North Creek, on extremely steep slopes.

Naturally Regenerated Secondary Forest: Described as secondary forest that has been

orests, we took soil samples from the bamboo and nonbamboo areas

within vegetation Plot 1.

its time as a pasture. This forest has been recovering for 15 years. It

occurs on the south side of the property along the Access Road.

ch

e top of the access road. There is virtually no other

vegetation that grows in the same area. This bamboo was planted about 9 years ago.

hin Bamboo: Defined as the area containing bamboo species Bambusa vulgaris, which grows

in forests and allows a lot of other vegetation to grow beneath it. It also occurs in the northwest

orner of the property at the top of the access road, directly adjacent to the thick bamboo habitat.

oncern has been expressed over the fact that this species spreads rapidly into the surrounding

condary forest. Again, this bamboo was planted about 9 years ago.

anana: Defined as the small area containing mainly banana trees, which were planted 8 years

go. The bananas make up a small area on the western border of the property, in an area known

as the “banana trail”.

ecovering Pasture: Defined as the area that was once a grazed pasture but has been allowed to

naturally recover for four years. It mainly consists of bamboo and tall grasses and occurs toward

e top of the access road.

untouched and allowed to recover for 15 years since the land restoration efforts began on the

FCRE. The natural regenerated forest takes up the majority of the FCRE property. In the

naturally regenerated f

Hardwood Plantation: Defined as an assisted secondary forest that was replanted with native

hardwood tree species after

Thick Bamboo: Defined as the area containing bamboo species Guadua angustifolia, whi

grows in large clumps and is being actively managed and harvested on the property. It occurs in

the northwest corner of the property at th

T

c

C

se

B

a

R

th

21

Grazed Pasture: Defined as the area that contains only short grasses, where cattle have recently

razed and trampled. There is evidence of cattle on the land, by means of hoof prints and

manure. The pasture in the study was directly adjacent to the FCRE borders to the northwest, in

Figure 4: Photos of habitats. From left to right- 1st row: primary forest, naturally regenerated secondary forest, hardwood forest, thick bamboo. 2nd row: thin bamboo, banana plantation, recovering pasture and grazed pasture.

Study plot 1

Study plot 1 is a 30m by 30 m area located in the E6 quadrant of the naturally regenerated

secondary forest region of the FCRE (Figure 5). Study plot 1 is part of a set of vegetation study

plots that have been monitored the past few years. The plots are broken up into 1m x 1m

quadrants for sampling purposes. An inventory of the all woody plants within plot 1 has been

corded annually since 2006. In 2008, plot 1 had 324 plants and 33 different species within its

area. It contains fifteen clusters of bamboo species Bambusa vulgaris, all which are located on

g

between Mudd pond and Basilisk pond.

re

22

the northern half of the plot, characterizing the plot with distinct bamboo and nonbamboo

regions within the area (Figure 6). Within plot 1, the areas around the bamboo clusters are

sparse in other vegetation.

Naturally Regenerated Forests

E 6

Fi Location of sampled bamboo nbamboo sites within plot 1 (green and blue respectively) as well as shown. The size of the dots corresponds to the

he higher the number of sprouts in each cluster.

gure 5:her bam (b

and noot boo clusters (yellow) and all other vegetation lue) are number of bamboo sprouts in each cluster. The larger the dot, t

23

In plot 1 of the naturally regenerated forests, five soil and invertebrate samples were

taken in the bamboo areas at sites with bamboo clusters at x-y coordinates <12,19> <16,18>

<14, 24 e

, 25>

>, <17, 24>, <18, 29> (Figure 5). We chose these sites in particular because these fiv

sites had the highest number of stems in each bamboo cluster. Similarly, five soil and

invertebrate samples were taken in the nonbamboo areas along x-transect 25m: <5, 25> <10

<15, 25> < 20, 25> <25, 25> (Figure 5). We chose to sample along this transect because the

nonbamboo areas within plot 1 was already predominantly on the southern side of the plot and

the transect was uniform in elevation.

: Study Plot 1. A: Nonbamboo area. B. Bamboo area

Adjacent pastures and Hacienda Baru

In addition to the eight on site habitats, we also collected data regardin

habitats. We collected data in two additional grazed pastures, one across the s

FCRE in much more open and intensely grazed pasture, and one closer to the

floodplain, which also had cows grazing at the moment of collection. Finally,

physical and chemical properties from the forests on Hacienda Baru, the adjac

that is managed as a conservation area and tourist educational destination. We

samples within the

A

Figure 6

property of Hacienda Baru. The following were the Hacien

primary forest, chilamate trees, cacoa trees, palms near beach and beach, from

cted and analyzed. The transect in Hacienda Baru started inla

forest a

sample was colle

nd ended at the beach. We chose to sample along this transect in orde

preliminary and general idea of how the soil’s physical and chemical properti

proximity to the ocean.

B

g some off-site

treet from the

ocean on the

we measu

ent plot of land

collected a

red the

total 5

da Baru sites:

each only one soil

nd in the primary

r to get a

es varied in

24

Overview of methods

The study was conducted between June 1 and July 31, 2008 at the Firestone Center for

Restoration Ecology in the Punta Renas province on the southwest coast of Costa Rica. Five

replicate plots, determined by previously tagged trees around the property, were chosen in each

of the eight habitats in order to measure the physical and chemical properties of the soil, as well

as the invertebrate communities in the soil and the leaf litter at each site. Measurements included

moisture content, soil bulk density, relative particle size, color, pH, soil compaction, depth of

organic layer, GPS coordinates, macro and micronutrients, as well as trace

the

metals, and

invertebrates in the litter and the O and A layers. The same variables were examined in Plot 1 in

ted secondary forest to compare the difference between bamboo and

nonbam

er

er

surement. Using a long core sample, color was determined according to the

Munsell Soil Color chart and the depth of the O layer was also measured. Two long cores, two

litter sample were then collected, stored in ziplock bags and transported to

the lab ely 0.1

were

sample was dried in an oven at 60ºC for 24 hours then reweighed in order to determine the

ce dry, the sample was run through a six part sieve set, ranging from 5 mm

to 230 mm and soil caught in each size was weighed and used to calculate the soil bulk density

the naturally genera

boo sites.

Field sampling

In the field, the GPS coordinates were recorded by measuring the distance, the angle and

the slope to the nearest GPS tag on the property. In this way, the exact GPS coordinates of each

sample site were calculated. Additionally, the pH was measured to an accuracy of a tenth of a

degree using a soil pH meter (Em System soil pH meter, Demetra, Tokyo Japan). The soil

compaction was measured by averaging three pocket penetrometer readings. The penetromet

used in this study only went less than a centimeter into the soil, so there is question as to wheth

this is a valid mea

short cores and one

for further testing. The litter sample was scraped from an area that was approximat

m2. In addition, 500 grams of Bambusa vulgaris and Guadua angustifolia foliage each

collected from the bamboo habitats of FCRE for chemical foliar analysis.

Physical analysis

Physical and biological analysis was conducted with the cores that were brought back to

lab. For each site, the fresh mass of the soil in one of the long cores was taken, after which the

moisture content. On

25

of the sample. Soil bulk density is defined as the weight of soil per unit volume. With the other

long core collected in the field, a size of particles test was conducted in order to determine the

percent of sand, silt and clay in the sample and thus classify the soil. Sand is defined as any

particle ranging from 2-0.05 mm, silt is 0.05-0.002 mm and clay is <0.002 mm (Schaetzl and

Anderson 2007).

Chemical analysis

Soil and bamboo foliar samples were sent to Activation Laboratories in Ancaster,

Ontario, Canada and the Laboratory of Soil and Foliage at the Center of Agronomical

University of San Jose, Costa Rica for chemical analysis. Analytical methods

. The

.

,

if possible and otherwise

order and counted. Some specimens were preserved in 70% alcohol. Second, one of the short

ores was set up in a Berlese-Tullgren funnel under a heat lamp and a dry extraction was

erformed for 24 hours to collect all the invertebrates in the A layer. A wet extraction was also

erformed but there were no major differences across habitats, so analysis was brief. We did not

ollect or analyze soil invertebrate fauna for offsite pastures and Hacienda Baru samples.

ata analysis

Most of the data were analyzed qualitatively, since five replicates was not enough to run

any statistical tests. Physical properties and chemical properties were noted and invertebrate

rders that appeared in some habitats and not others were focused on. Species abundance,

ecies richness and Simpson’s Index of Diversity was calculated for the invertebrates in each

abitat and those values were compared across habitats. The species richness measurement

veals the number of species in a given area, while the Simpson’s Index of Diversity relates

Investigations,

used to determine the amounts of micronutrients, macronutrients, and other elements in the soils

include instrumental neutron activation analysis (INAA), inductively coupled plasma optical

emission spectrometry (ICP), and inductively coupled plasma mass spectrometry (ICP/MS)

amount of nitrogen in the samples was only analyzed in the eight different habitats in the FRCE

Biological analysis

Soil invertebrate fauna were sampled with two different methods for each habitat. First

leaf litter was collected on site and sorted in the lab. Invertebrates in the litter were collected

using an aspirator. All the invertebrates were then identified to family

to

c

p

p

c

D

m

o

sp

h

re

26

biodiversity to the relative abundance of different species. The formula used to determine the

impson’s Index of Diversity was 1- D where D = Σ (n/N)2 and n= total number of organisms of

particular species and N= total number of organisms of all species.

S

a

27

RESULTS

Physical properties

ita

The physical properties of th ot v een sites. Although soil texture

lightl ee b ll e cla /clay loam/sa y nge

of the texture spectrum. Soil bulk density was highest in the hardwood forest and lowest in the

recovering pasture. Moisture content was highest in the thick bamboo and lowest in the primary

alue ho n 3 s are re nd Fig

sic i so di t onsi

(%) (%) h of

O layer

bulk moisture

FCRE hab t types

e soil did n ary much betw

varied s y betw n ha itats, a were in th y/sandy clay ndy cla loam ra

forest. V s are s wn i Table and trend shown in Figu 7 a ure 8.

Table 3: Phy

al propertSand

es of Silt (%)

il accorClay

ng to habitaTexture

te Color Dept

(cm)

Soil

density (g/cm3)

Soil

content (mL)

Primary 45.33 .33

39.33 C clay, clay

5 YR 4/6, 7.5 YR 4/6

8.2 0.721 0.1608 15 lay, sandy 5 YR 4/4,

loam, sandy clay loam

Natural Regeneration

50.67 15.33

34.0 Clay, sandy clay, clay loam, sandy clay loam

2.5 YR5/8, 5 YR 4/6, 5 YR 5/8

2.8 0.6743 0.2226

Hardwood 46.67 17.0 36.33 Clay, clay loam, sandy clay loam

5 YR 4/6, 5 YR 5/8, 7.5 YR 4/6, 7.5 YR 6/6

1.8 0.7479 0.2086

Thick bamboo

50.67 14.0 35.33 Clay, clay loam

2.5 YR 4/8. 5 YR 4/4, 5 YR 4/6

0.2 0.6729 0.2820

Thin bamboo 41.33 19.0 39.67 Clay, clay loam, sandy clay loam

2.5 YR 4/6, 5 YR 4/6

0.6 0.7168 0.2456

Banana 33.33 20.33

46.33 Clay, clay loam

5 YR 3/4, 5 YR 4/4, 5 YR 4/6, 5 YR 5/8

1.2 0.6680 0.25260

Recovering pasture

42.67 17.67

39.67 Clay, clay loam, sandy clay loam

2.5 YR 4/6, 5 YR 4/6

0 0.6506 0.2547

4.8 0.7184 0.2532 Pasture 43.33 16.67

40.0 Clay, clay loam, loam

2.5 YR 4/6, 5YR 4/6

28

% Clay

% Silt

% Sand

0.2

0.8

1

1.2

0.4

0.6

0

Prim

ary

Natu

ral R

egen

ation

Hardw

ood

Thick

Bam

boo

Thin B

ambo

o

Bana

na

Reco

verin

g Pa

stur

e

Pastur

e

Figure 7: Percent sand, silt and clay in soil according to habitat. These percentages were used to determine the soil texture.

0

45

pHSoil bulk density (g/cm3)moisture (ML)

35

40

5

10

15

20

30

Prim

ary

Natur

al R

egen

ation

Hard

wood

Thick

Bam

boo

Thin B

ambo

o

Bana

na

Reco

verin

g Pa

stur

e

Pastur

e

25

ty and moisture content according to habitat. Figure 8: pH, soil bulk densi

29

Naturally Regenerated Forests

Nonbamboo soils had 44% sand, 10% silt, and 46% clay, classifying the soils with a clay

xture (Table 4). Bamboo soils consisted of 58.67% sand, 5.6 % silt, and 35.07% clay, giving

colors than bam re uniform

c ). The bulk

s ar, in catin t the sity of the ls was sim boo

and bambo eg as sa d, but neither had an organic layer. The soil moi content

t wo reg ns d t sig antly diff either. Table 4: Ph p ies of nonbamb bamboo areas of n reg d fo

si(% text or

k density

l)

il moisture

L)

BAM 46 claR 4/6 5/8, YR .71 629163

te

the soils a sandy clay loam texture. The soils sampled in the nonbamboo soils had a higher

diversity of boo soils, indicatin

soils (Table 4

g that the bamboo soils were mo in

olor than the nonbamboo density and compaction of both areas were

imil di g tha poro soi ilar. Only the A layers of nonbam

o r ion w mple sture of

he t io id no nific erent

ysical ropert soils in oo and aturally enerate rests

sand(%)

lt ) (%)

clay ure col

bul

(g/m

so

(m

NON BOO 44 10 y 5Y

5, 5YR

2. 4/6 0 60 0.

BAMBOO 7 5 .07 san y loam/sandy clay

5YR 4 R /6 0.72 553024

Chemical properties F E hab y

pH did not vary significantly bet habita wever, all ha are mo

cidic than the primary forest. Potassium, calcium, magnesium, sulfur, lithium and sodium were

Other important nutrients for plant growth, such as nitrogen,

phosph

d

58.6 .6 35dy cla 2. /8, 5Y

4 27 0.

CR itat t pes

ween ts (Table 5). Ho bitats re

a

all highest in the primary forest.

orous, zinc and cobalt did not seem to show any pattern in relation to land use.

Manganese, nickel, copper, iron and chromium all displayed a pattern according to land use, in

which the primary and secondary forest had similar concentrations while the more disturbe

habitats all had different, more extreme concentrations. The primary forest has a higher overall

percentage of macronutrients than the other habitats, especially the pasture.

30

Table 5: Chemical properties of soil according to habitat. Macronutrients and micronutrients listed are essential for lant growth.

Element Unit Primary Natural Regeneration

Hardwood Thick Bamboo

Thin Bamboo

Banana Recovering pasture

Pasture p

pH 6.36 6.04 6.18 5.92 6.08 5.92 5.66 6.14

Macronutrients N % 0.28 28 0.35 0.23 0.26 0.26 0.30 0.31 0.K Ca

%%

1.19 0.54 24 0.17 0.25 0.41 0.14 0.42 0.62 0.08 0.26 0.10 0.11 0.14 0.08 0.16

0.95 0.32 22 0.13 0.34 0.31 0.25 0.07 0.039 0.066 0.168 0.184 0.229 0.12 0.12 0.06 0.032 0.02 0.04 0.01 0.03 0.03 0.04

onutrie 7.68 7.47 8.61 15.3 13.8 14.1 17.3 13.3

0.

Mg % 0.21 0.P % S % Micr nts Fe % Mn 1250 1190 1100 2060 2160 2470 2350 2420

146 113 87.7 154 135 186 161 174 36.3 40.6 63.3 6.6 11.2 4.9 11.3 10 < 1 2 1 < 1 < 1 < 1 < 1 < 1

81.4 67.4 129 610 373 657 806 373 20 25 28 28 24 38 41 30

er elem < 2 < 2 < 2 14 < 2 < 2 < 2 < 2

ppm Zn ppmNi ppmMo ppmCu ppm Co ppm Oth ents Au ppm Al % 8.9 4.89 11.1 12.4 9.88 11.8 12.8 10.6

108 133 141 6 37 5 < 1 < 1 41.5 37.4 19.6 12.7 10.9 17.3 14.7 12.9 0.52 0.05 0.17 0.05 0.08 0.08 0.04 0.07 4 142 67 50 128 128 14 4

rally erat Forests

We found the amounts of 65 different trace elements in the nonbamboo and bamboo

boo

mboo areas (Table

). The pH of the nonbamboo soils were slightly more acidic than the bamboo soils.

Cr ppm Li ppm Na % Zr ppm

Natu Regen ed

areas of the naturally regenerated forests (Appendix 1). The amount of nutrients and fertility of

the soils found in nonbamboo and bamboo areas of the naturally regenerated forests were

relatively similar. The amount of macronutrients and micronutrients in nonbamboo and bam

areas were not significantly different (Table 6). Higher quantities of As, Cr, and Sb were found

in nonbamboo areas while Be, Sr, and Zr were found in higher quantities in ba

6

31

T rtant trace elements in soils of no mboo and bamboo a of naturally regenerated forests nit Nonbamboo Bamboo

able 6: Impo nba reas Element UpH 5.84 6.26 Macronutrients K % 0.14 0.13 Ca % 0.05 0.05

% 0.18 0.22 P % 0.079 0.118 S % 0.03 0.03 Micronutrients Fe % 16.5 16.9

Mg

Mg % 0.18 0.22 Mn ppm 1270 1490 Zn ppm 91.6 117 Ni ppm 20.1 17.1 Mo ppm < 1 < 1 Cu ppm 325 529 Other elements As ppm 11 < 0.5 Be ppm 1 Cr ppm 6

m

ar analysis ana f u s and B v r ed amounts of

K, S O., u Zn ount more of

ambusa vulgaris, otherwise, neither Guadua angustifolia nor Bambusa vulgaris

ican a o f ace el (

components of organic allelotoxins.

: Amount of trace elements in Guadua a folia an sa vulgaris leaves elemen

< 0.5

0.9 < 1

Sb ppmSr pp

0.9 28

< 0.1 64.1

Zr ppm

9 66

Bamboo foli

Foliar

N, P, Ca, Mg,

lysis o Guad a angu tifolia ambusa ulga is show similar

, C. Fe, C , and . Guadua angustifolia had a significant am

Mn than B

showed signif

analyzed minerals, and did not analyze

tly gre ter am unts o any tr ements Table 7). The foliar analysis only

able 7 ngusti d BambuT

t Guadua angustifolia Bambusa vulgaris N (%) 2.12 2.08 P (%) 0.33 0.14 Ca (%) 0.5 0.31 Mg (%) 0.12 0.18 K (%) 1.99 2.26 S (%) 0.25 0.49 Fe (mg/kg) 85 86 Cu (mg/kg) 5.5 5 Zn (mg/kg) 8.5 9.5 Mn (mg/kg) 828 178 B (mg/kg) 1 4

32

Off-sites In the two additional offsite grazed pastures that were sampled, the pasture near the ocean

as more acidic than the pasture across the street from the FCRE. The pasture across the street

ad a higher soil bulk density and lower moisture content than the pasture by the ocean. The

of moistur

c nt and soil bulk dens ed t l b n c d

a ntent dec d. In ditio erce sand he so ncre d w

p . The as si r thr ut th sect le 8

T of so fsite sa les Sil

(%ture Color H bulk

ity 3)

oil m ture nten L)

w

h

Hacienda Baru transect showed varying physical properties, especia n telly i rms e

onte ity. As we sampl closer to the beach, he soi ul dek sity in re sea

nd the moisture co rease ad n, the p nt of in t il i ase ith

roximity to the beach pH w mila ougho e tran (Tab ).

able 8: Physical properties il in of mpSand (%)

t )

Clay (%)

Tex p Soil dens(g/cm

Sco

oist (m

P .33 20. .67 Clay 7.5 Y3/3

.3 58 .305 asture across treet

33 0 46 R 6 0.73 0sPasture near ocean 46.67 20.0 33.33 Clay loam 7.5 4/3 5.9 0.6536 0.2991

acienda Baru rimary forest

40.0 13.33 46.67 Clay 7.5 YR 4/3

6.3 0.6609 0.3123

acienda Baru 33.33 13.33 53.33 Clay 2.5 YR 4/8

6.5 0.6698 0.3194

26.67 33.33 Clay loam 7.5 YR 3/2

6.5 0.6472 0.2781

Hp

Hchilamate trees Hacienda Baru cacoa trees

40.0

Hacienda Baru palms near beach

46.67 26.67 26.67 Loam 7.5 YR 2.5/2

6.8 0.7843 0.3156

Hacienda Baru beach

53.33 13.33 33.33 Sandy clay loam

10 YR 3/2

6.5 0.9009 0.2218

Chemical properties in the offsite properties varied most drastically between iron and

manganese in the two pastures and phosphorous and manganese in the Hacienda Baru transe

Phosphorous increased steadily as the sampling gets closer to the beach. Otherwise, nutrien

samples on the Hacienda Baru transect varied in the different habitats (Table 9).

ct.

ts in

33

Table 9: Amounts of trace elements in the soils of nearby pastures and Hacienda Baru transect

P Fe Mn Zn g/L

pasture across street 2 .64 32 site

K cmol(+)/L

mg/L

Ca cmol(+)/L

Mg cmol(+)/L

mg/L

mg/L

Cu mg/L

m

0.51 24.04 2 14 6 2.2 pasture near the ocean 16.06 93 2.8

Baru primary 9 1 10.23 4.44 105 49 3 1.

Baru chilam7 1 2.50 1.90 114 98 3 1.

cacoa t s 27.75 6.23 21 14 4 4.2 aru palms near

2 3.11 0.96 32 12 1 0.aru beach 28 5 6.07 2.92 94 11 3 1.

ities itat types

Invertebrate communities were very different in every habitat. In the leaf ter, a tota f

iduals dist uted in 5 different orders were identified in the primary forest. 236

3 dif ent orders were found in the naturally regenerated secondary forest while

0 individual 14 di ent orders ere found in the hardwood est. A tal of 104

distribu in 10 different orders occurred in the thick bamboo and 144 individuals in

ent orders o urred the thin ba oo. 136 ividua in 17 d erent o ers were

in the bana planta n, 47 indiv als in 9 ers we counte n the overing

d only 5 i ividuals in 4 orders were found in the gr d past . Som f the mo

on-insect groups, such as the Ricinuclei, Glomerida, Penicilla Diplura, Opilione

rpha, Protura, Polydismidae, Lithobiomo ha, Oli haeta, olape ropmorph

ed in som abitat t not othe The mo ommo nsect ers, s as

optera were ch mo abundant in the primary forest, due to the large colonies of ants

(Table 1 d Fi 9). Nem s were found in every habitat in the wet

extraction funnel.

0.46 ND 4.90 86 5 Haciendaforest 0.2 8 Hacienda ate trees 0.2 7 Hacienda Baru ree 0.30 2 Hacienda Bbeach 0.10 8 Hacienda B 0. 7

Invertebrate communFCRE hab

lit l o

439 indiv rib 2

individuals in 2 fer

only 13 s in ffer w for to

individuals ted

14 differ cc in mb ind ls iff rd

counted na tio idu ord re d i rec

pasture an nd aze ure e o re

obscure, n ta, s,

Geophilomo rp goc Sc nd a

only occur e h s bu rs. re c n i ord uch

Hymen mu re

found there 0 an gure atode

34

T : Number of individuals in each taxonomic group in litter samples, according to habitat. Bold taxon represent categories

Primary Natural Regeneration

Hardwood Thick Bamboo

Thin Bamboo

Banana Recovering pasture

Pastur

able 10

e

TAXON Oligochaeta 2 9 0 0 1 1 1 0 Mollusca Gastropoda 3 9 1 1 1 2 1 0 Onychophora 1 0 0 0 0 0 0 0 Crustacea Isopoda 57 34 21 9 20 33 1 0 Diplopoda Glomerida 5 2 0 0 0 0 0 0 Penicillata 13 5 4 0 0 0 0 0 Polydesmida 2 1 0 0 1 1 0 0Chilopoda Geophilomorpha

1 0 2 0 0 1 0 0 thobiomorpha 1 4 0 1 2 0 0 0

colaphendromorpha 0 0 0 0 0 4 0 0 rchanida ranae 12 13 4 1 2 15 1 0 cari 35 8 6 6 28 8 9 0 piliones 4 5 1 0 0 0 0 0 icinuclei 2 0 0 0 1 0 0 2

doscorpiones 33 20 20 0 0 1 0 0 exapoda plura 10 0 0 0 1 0 0 0

rotura 5 10 6 0 0 2 0 0 ollembola 34 41 13 14 20 26 5 1 secta donata 0 1 0 0 0 0 0 0 lattodae 0 3 0 0 0 1 0 0 optera 2 0 0 0 0 0 0 0

0 europtera 1 2 0 0 0

ptera 0 0 1 0 0 0 0 0

21 9 3 17 4 2 0 3 0 0 2 4 0 0 20 20 39 19 7 6 1

1 0 0 0 0 0 0 1 2 1 0 0 0 0

a 0 49 20 29 29 25 21 1 9 2 1 10 14 6 47 5

LiSAAAORPseuHDiPCInOBIsDermaptera 0 1 0 0 0 0 0 NPsco

00 0 1

0 0

0 0 1

0

Orthoptera 0Hemiptera

ra 6

1

Thysanopte 4Coleoptera 44

Lepidoptera 1Diptera 3

2Hymenopter 1TOTAL 43 63 30 4 4 13

35

-20

0

20

40

60

80

100

120

140

Isop

oda

Glomer

ida

Pseu

dosc

orpion

Aran

ae

Penicil

lata

Diplura

Acar

i

Prot

ura

Hemiptera

Coleop

tera

Diptera

Hymen

optera

Colle

mbo

la

Order

Primary

Natural Regeneration

Hardwood

Thick Bamboo

Thin Bamboo

Banana

Recovering Pasture

Pasture

Figure 9: Abundance of major orders in litter sample according to habitat

The invertebrates in the dry extraction, on the other hand, showed less of a pattern than

the leaf litter. The different habitats resembled each other much more. A total of 566 individuals

distributed in 19 orders were counted in the primary forest, 119 individuals in 13 orders were

found in the naturally regenerated secondary forest and 58 individuals in 13 different orders in

the hardwood forest. 74 individuals in 12 different orders were found in the thick bamboo while

67 individuals in 11 orders were found in the thin bamboo. Finally, 93 individuals in 14 orders

were found in the banana, 129 individuals in 15 orders were found in the recovering pasture and

92 individuals in 12 orders were found in the grazed pasture. Again, like the leaf litter, many

obscure, non-insect groups were found in only a few of the habitats instead of being broadly

distributed.

36

Table 11

Bamboo Bamboo pasture Pasture

: Abundance of individuals in dry extraction samples according to order, according to habitat

Primary Natural Regeneration

Hardwood Thick Thin Banana Recovering

TAXON Oligochaeta 5 2 3 0 0 2 5 3 Mollusca Gastropoda 0 0 1 0 1 0 0 0 Crustacea Isopoda 0 3 4 1 3 2 5 0 Diplopoda Penicillata 5 0 0 0 0 0 0 0 Chilopoda Geophilomorpha 2 0 0 0 0 0 0 0 Lithobiomorpha 0 0 1 0 0 0 2 1 Scolaphendromorpha 1 0 3 0 0 0 0 0 Arachnida Aranae 3 0 3 3 0 4 6 5 Acari 9 0 0 6 2 25 11 2 Pseudoscorpion 2 2 0 0 0 0 0 0 Hexapoda Diplura 2 3 2 2 0 1 3 0 Protura 3 3 0 0 0 2 4 5 Collembola 430 36 5 7 5 13 26 4 Insecta Isoptera 4 1 0 1 0 1 0 0 Neuroptera 3 0 0 0 0 0 0 0 Orthoptera 0 0 0 0 0 0 1 0 Hemiptera 13 10 7 17 39 14 12 11 Thysanoptera 2 5 0 4 1 1 2 2 Coleoptera 14 23 12 16 6 15 25 22 Trichoptera 33 Diptera 34 47 57 51 23 117 168 74 Lepidoptera 1 0 0 0 1 0 0 2 Hymenoptera 28 22 13 2 6 9 1 2 TOTAL 566 166 115 125 90 210 297 166

5 9 4 15 3 4 26

37

-50

450

500

0

50

100

150

200

250

300

350

da n ae ura ar

i ra ra ra a ra

Primary

Natural Regenerati400 on

Hardwood

Thick Bamboo

Thin Bamboo

Banana

Recovering Pasture

Pasture

Isop

o

Pseu

dosc

orpio

Aran

Dipl Ac

Hemipte

Coleop

tera

Dipte

Hymen

opte

Colle

mbo

l

Trich

optera

Isop

te

Order

Figu o n d a and beta d ersity were ca ulated for each habitat in the form ess

a y. For b th, trends were m ch more apparent in the litter

s pared to the dry extraction samples. All of the dry extraction numbers exclude the

order Diptera, since the setup was biased towards flies th ere attracted to the light of the

extraction overnight. Litter species richness was found to be much lower in the pasture, however,

the standard error is extremely high so it is hard to draw conclusions about these trends.

re 10: Abundance f major orders i ry extraction sample ccording to habitat

Alpha iv lc of species richn

nd Simpson’s Index of Diversit o u

amples com

at w

38

VA

1.20

1.00

0.80

0.60

0.40

0.00

0.20

Mea

n

PastureRecovering

PastureBananaThin BambooThick BambooHardwoodNatural

RegenerationPrimary

R00001T xextbo

Primary Hardwood

TB

hick amboo

Thin Bamboo

Banana Pasture Recovering Pasture

Natural Regeneration

□ Litter □ Dry extraction

Error bars: +/- 2 SE

Figure 11: Simpson’s Index of Diversity in litter and dry extraction samples according to habitat

Species richness was much higher in primary forest, followed by the two versions of

secondary forest as compared to the other habitats.

39

30.

25.

20.

15.00

10.

5.

0.

00

00

00

00

00

00

Mea

n87654321

VAR00001

Error bars: +/- 2 SE

Tex…

Figure 12: Species Richness in litter and dry extraction samples according to habitat

Naturally Regenerated Forests

Invertebrates in soils of nonbamboo region

A total of 166 invertebrates were identified to order in nonbamboo areas from the leaf

litter and dry extraction. Fifty-eight of these invertebrates were non-insects and 108

invertebrates were identified as insects. High abundances of Isopoda and Acari were found in

the leaf litter samples with the invertebrates often found in the curled edges of the leaves and

high abundances of Coleoptera and Diptera in the dry extraction samples. Araneae, Isoptera,

Gastropod, Polyxenida, Thysanoptera, and Orthoptera were found only in nonbamboo areas of

the natu ich

include

Primary Natural Regeneration Hardwood

Thick Bamboo

Thin Bamboo

Banana Pasture Recovering Pasture

Natural Regeneration

□

□ Dr

Litter

y extraction

rally regenerates forests. Additionally, there was a high diversity of Collembola wh

d families Isotomidae and Onychiuridae.

40

Invertebrates in soils of bamboo region

dry ext

sects. Like the nonbamboo leaf litter samples, Isopoda and Acari were most common. Within

e order Acari, Orbatidae were most commonly found. Furthermore, there were a few different

ecies within the Orbatidae family (Appendix 6). There was also a significant high abundance

f Hymenoptera in the bamboo leaf litter sample due to the collection of an ant colony.

iones, Diplura and Psocoptera were only found in bamboo areas.

We identified 320 invertebrates to their order in the bamboo areas from the leaf litter and

raction samples. Seventy five invertebrates were non insects and 255 invertebrates were

in

th

sp

o

Diplopoda, Opil

Table 12: Invertebrate species abundance in nonbamboo and bamboo areas

NB leaf litter B leaf litter NB dry extraction B dry extraction

TAXA Oligochaeta Haplotaxida 1 1 4 4 Mollusca Gastropoda 3 0 0 0 Crustacea Isopoda 17 23 1 1 Diplopoda Penicillata 2 1 0 0 Symphyla 1 3 0 0 Arachnida Acari 22 35 3 0 Araneae 2 0 1 0 Opiliones 0 3 0 0 Pseudoscorpions 1 2 0 0 Hexapoda Diplura 0 0 0 2 Collembola 8 5 2 0 Insecta Isoptera 1 0 2 0 Psocoptera 0 5 0 0 Orthoptera 1 0 0 0 HomopHemipt

tera 0 0 7 1 era 7 2 2 2

Thysanoptera 1 0 0 0 Coleoptera 0 9 16 5 Trichoptera 0 0 2 1 Diptera 1 3 42 8 Hymenoptera 12 202 4 2 TOTAL 80 294 86 26

41

Common invertebrates

Common invertebrates found in both the nonbamboo and bamboo regions included

Isopoda, Acari, Haplotaxida, Collembola, Hymenoptera, Hemiptera, Coleoptera, and Diptera

More specifically, Isopoda, Acari, and Diptera were in high abundance. Of the common

invertebrates found in both regions, nonbamboo samples had more orders with higher

abundances than bamboo samples (Figure 13). Nonbamboo samples had higher abundances of

.

Collem ola, Hymenoptera, Hemiptera, Coleoptera and Diptera. Bamboo samples had higher

an nonbamboo samples.

b

abundances of Isopoda and Acari th

40

05

101520253035

Isopo

daAca

ri

Haplot

axida

Collem

bola

Hymen

opter

a

Hemipt

era

Coleop

tera

Diptera

Order

abun

danc

e

4550

nonbamboobamboo

Figure 13: Common invertebrate abundance in nonbamboo and bamboo areas of naturally regenerated forest Simpson’s Index of Diversity and species richness

Simpson’s Index of Diversity and species richness was calculated for the nonbamboo

area and bamboo area in study plot 1 (Figure 14 and 15) to give a measure of diversity. Similar

to previous data analysis methods, the dry extraction data excluded the order Diptera because

light for the overnight dry extraction may have attracted a large number of Dipteras that were

originally not in the soil samples. There was no difference in Simpson’s Index of Diversity

the

etween the nonbamboo and bamboo sites (nonbamboo = 0.783, bamboo = 0.753) or in species

two areas of study plot 1 (nonbamboo = 9, bamboo = 8.4). Nonbamboo

b

richness between the

42

sites had a slightly higher Simpson’s Index of Diversity from the leaf litter samples (nonbam

= 0.757, bamboo = 0.608) and a higher species richness from the dry extraction samples

(nonbamboo = 7.8, bamboo = 3.4).

In comparison with the invertebrate diversity from other sites in the naturally regenera

secondary forests, study plot 1’s Simpson’s Index of Diversity for both the leaf litter

extraction samples are relatively similar to the Simpson’s Index of Diversity from other samp

taken from sites outside of the study plots in the naturally regenerated secondary forests (Figure

14). In contrast, the leaf litter and dry extraction samples from other sites within the naturally

regenerated forest

boo

ted

and dry

les

s had much greater species richness than the samples within study plot 1

(Figure

15).

43

0.9

1

0.8y

0

0.1

0.2

0.3

0.4

0.5

0.6

nonbamboo bamboo other sites inNRSF

sites

mea

n si

mps

on's

inde

x of

div

0.7

ersi

tleaf litter

dry extraction

Figure 14: Simpson’s Index of Diversity in litter and dry extraction samples in the nonbamboo and

ithin study plot 1 and other sites in the naturally regenerated secondary forests (NRSF). bamboo sites w

0

5

10

15

20

25

spe

cies

rich

ness

mea

n

leaf litter

dry extraction

nonbamboo bamboo other sites inNRSF

sites

Figure 15: Species richness in litter and dry extraction samples in naturally regenerated secondary forests (NRSF). Species richness was compared between nonbamboo and bamboo sites in study plot 1 and with other sites in the naturally regenerated secondary forest areas.

44

DISCU

the

f organisms transport organic particles, breakdown certain

hemicals and decay particles resistant to decomposition. Other organisms not directly involved

e important role of regulating microbial populations by grazing and

.

ts

us

In addition to invertebrates, plants play a critical role in the physical and chemical

tically between habitats, especially in the

for

t, yet there was four times as

uch diversity of plants in the forest and there were no species that were common to both

and animals that are responsible for seed dispersal tend to avoid open areas,

ll 1999).

SSION

Invertebrates strongly influence the physical and chemical properties of the soil. Because

the decay of organic materials by invertebrates is done in a series of steps, the entire process

requires a diversity of soil fauna. First, shredders must shred, mix and digest organic matter in

the form of leaf litter and increase the surface area of leaf litter for microbial attack, inoculate

soil with gut bacteria and inoculate the soil with digestive enzymes that continue breakdown in

fecal pellets. Next, a different group o

c

in decomposition play th

predation. Many invertebrates possess highly specialized morphology to perform one of these

steps, thus diversity is essential for decomposition and nutrient cycling (Freckman et al. 1997)

The loss of any of the invertebrate functional groups described in Table 2 can be extremely

detrimental to the functioning of the entire ecosystem. Land use impacts that decrease this

diversity, as shown by the abundance and diversity of invertebrates found in the fores

compared to the pasture at FCRE, can be a problem for the nutrient cycle of the system and th

the plant growth and everything that relies primary productivity. We saw a dramatic decrease in

soil invertebrate abundance and diversity in habitats with greater human disturbances, suggesting

a possible loss of some of these decomposition functions.

properties of the soil, and since vegetation varied drama

pasture, the bamboo habitats and the banana plantation, this could be a possible explanation

the differences in the physical and chemical properties found. Although no vegetation survey

was done in this study, the sequence of plant growth is important to examine when looking at

recovering areas and how time since land use impacts the vegetation in the area. Holl (1999)

found that the succession of a pasture into forest depends most strongly on recently dispersed

forest seeds, as opposed to seed germination rates. In a recovering pasture, it was found that

herbaceous species were much more abundant than in the fores

m

habitats. Many birds

thus stalling seed dispersal of many forest species and slowly the rate of recovery (Ho

45

Primary forest vs. grazed pasture

The primary forest and the grazed pasture lie on opposite ends of the land use spectrum,

thus making their comparison the most revealing one in this study.

Physical properties

It was expected that the physical properties would vary most dramatically between thes

two habitats, especially with respect to soil bulk density, moisture content and the depth

organic layer. It was found that moisture content was much lower in the primary forest than

pasture, most likely due to the high levels of plant roots that created pores in the soil to increa

infiltration and the big difference in soil invertebrates found between the habitats. Because of the

high invertebrate abundance and diversity, there are more pores in the soil to facilitate drainage

Strangely, the other physical properties that were measured did not follow the trend. Soil bulk

e

of the

the

se

.

ensity, which is closely related to porosity and by extension moisture content, did not vary

s. This is surprising when considering the differences in physical

mpac

termine in the pasture, due to the large amounts of cow manure, thus the depth of

e organic layer might actually be less than reported in this study.

d

much between habitat

co tion by trampling cattle between the two habitats, since other studies have found that

heavy grazing often reduces air-filled porosity, water infiltration, increases bulk density, and

changes critical soil chemical properties (Greenwood and McKenzie 2001, Pereira et al., 2003,

Sharrow 2007). It was expected to find a much higher soil bulk density in the pasture than the

primary forest. The depth of the organic layer was about twice as thick in the primary forest as

the pasture due to the differences in the plant communities. The O layer was also extremely

difficult to de

th

Chemical properties

Chemical properties were expected to vary dramatically as well between these two

habitats, since the vegetation is so different. The nature of the vegetation influenced the rates of

nutrient cycling, since grasses cycle base cations more rapidly and generate fewer and less

persistent organic acids than do forests (Reiners et al. 1994). This was found to be true in the

case of potassium, calcium and magnesium, all of which were found in much lower

concentrations in the pasture and the presence of grasses compared to the primary forest. In

addition, the loss of these soluble base cations might also be due to higher amounts of leaching

46

because of increased exposure to weather with less dense vegetation, which would also result in

more acidic soil.

Previous studies found that pH was lower in the pasture compared to the forest

leveland et al. 2003), however, the data in this study found a much less significant difference.

et, the primary forest was the least acidic of all the habitats, suggesting that human disturbance

auses acidification through the loss of base salts. Meanwhile, important macronutrients and

icronutrients for plants, such as nitrogen, phosphorous, sulfur, zinc and colbalt compared rather

losely between habitats. There was a big difference in the concentrations of copper, manganese,

lithium and sodium, some of which influence the enzyme processes within

s, the pasture was nearly void of

fe. While a total of 439 individuals were counted in litter samples from the primary forest, only

in the pasture. All the groups important for litter decomposition were found

ts

different. In the forest, litter was easy to collect

rated secondary forest vs. hardwood forest vs. recovering pasture

The naturally regenerated forest, the planted hardwood forest and the recovering pasture

ries of secondary forest. Both the naturally regenerated forest and the

ing

(C

Y

c

m

c

nickel, chromium,

plants (Table 1).

Biological communities

There were extraordinary differences between the primary forest and the grazed pasture

in terms of the biological communities. For all practical purpose

li

5 in total were found

in high abundance in the primary forest, as expected, while they were all absent in the pasture.

Worms, mites, millipedes and isopods were not found in the pasture. Both the species richness

calculation and the Simpson's Index of Diversity were lowest in the pasture among the habita

analyzed. The lower diversity in the pasture is strongly correlated with land use intensity.

However, the nature of the litter was very

because it was mainly fallen leaves. In the pasture, untangling litter from the grasses was

difficult, so there was less volume to analyze when counting and identifying invertebrates. The

relative absence of litter in the pasture might be part of the reason for lower abundance and

diversity in litter invertebrates.

Naturally regene

are all different catego

hardwood forest have been recovering from disturbance for fifteen years while the recover

pasture is only four years old. Thus, comparing these habitats was potentially revealing. It was

47

not expected to find big differences between the two fifteen-year-old secondary forests how

Disturbance intensity ranges from the naturally regenerated forest to the planted hardwood to

recovering pasture, which is the most recently disturbed of these three habitats.

Physical properties

The organic layer was thickest in the naturally regenerated se

ever.

the

condary forest followed by

rties

ure as expected (Cleveland et al. 2003). There was at

from the recovering pasture to the two forested areas. This is a large

ffect the types of invertebrates that have a high enough acidity tolerance to

survive

n and

rue for

s

t

hardwood then recovering pasture, as expected. There was no clear trend, however, in the data

for soil bulk density and moisture content, although the pattern was the same for both,

demonstrating the relationship between the two properties. In all three cases, higher soil bulk

density correlated with lower moisture content. Interestingly, the recovering pasture had the

lowest soil bulk density and the highest moisture content, even though it was the most recently

impacted. The hardwood had the highest soil bulk density and lowest moisture content of the

three habitats.

Chemical prope

pH was lowest in the recovering past

least a 0.5 difference in pH

enough difference to a

. Also, since a lot of the vegetation in the recovering pasture was tall grasses, they likely

cycled base cations more similarly to the pasture than either of the forest habitats, thus

explaining the lower concentrations of potassium in the recovering pasture. Both nitroge

potassium were higher in both 15 year old secondary forest plots while the opposite was t

phosphorous.

Interestingly, phosphorous levels were lowest in the naturally regenerated secondary

forest, followed by the hardwood secondary forest, even though plant growth and diversity wa

extremely high in this area, due to the rapid uptake of standing biomass, which is typical of areas

undergoing succession. This suggests that phosphorous was not in fact a limiting factor in plan

growth as it often is. There were no major differences in calcium, magnesium or sulfur. Iron,

manganese, zinc, copper and colbalt were all much higher in the recovering pasture, most likely

due to the differences in vegetation.

48

Biolog

3

ies

s

he

hese three habitats are different from the rest, since they are defined more by their

vegetat

the amoun t ma t of he thick

bamboo lacked any understory (Fig ) and is being activ anaged and harvested while the

thin bamboo has been left alone to spread and there is a lot of understory growth. The banana

samples were a special case, and in ting to examine since they woul othe occur on

the FCRE property if they had not in entionally plant

Physical p rties

Sin here w orkers ac y chopping down t ck bamb uad

angustifoli d tram g the soil g the time of samp it was ex ed th se

e physical properties of the soil. In addition, since no vegetation

grew be rent in

he

he

ding

ulk density did not vary much between

ical communities

The patterns in the invertebrates follow the levels of disturbance history. A total of 26

individuals were collected in the leaf litter in the naturally regenerated forest, while 130 were

collected in the hardwood and only 47 were collected in the recovering pasture. This difference

in abundance suggests that invertebrates prefer greater input of organic matter through litter fall

and more complex soil crumb structure and microhabitat diversity. Again, the major

decomposers were most abundant in the habitat that was least disturbed. In litter, the spec

richness clearly showed a trend following the intensity of land use, in which the forested habitat

were much higher than the pasture or the bamboo, although this might again be attributed to t

nature and reduced quantity of litter in the recovering pasture.

Thick bamboo vs. thin bamboo vs. banana plantation

T

ion than their land use histories. The two bamboo species differed immensely in terms of

t of direc human nagemen and also the amount undergrowth present. T

ure 4 ely m

teres d not rwise

been t ed.

rope

ce t ere w tivel he thi oo, G ua

a, an plin durin ling, pect at the

activities would influence th

low the thick bamboo, it was expected that the chemical properties would be diffe

some respect and also there would be lower diversity and abundance in the invertebrate

communities. The organic layer was thickest in the banana, followed by the thin bamboo then t

thick bamboo, following the gradient of leaf fall, since there was more leaf mass falling in t

banana grove as opposed to the bamboo habitats. This means that there is a lot more stan

biomass that contributes nutrients to the soil and provides a healthier litter invertebrate

community as compared to the thich bamboo. Soil b

49

habitat

ver,

nificant. This is likely due to the lack of undergrowth below the

Guadua angustifolia, were not able to cycle the gold. Of all the habitats, phosphorous was found

ighest concentrations in the banana plantation, suggesting that plant growth in this area is

t

t

e

e

ertebrate groups can be explained by the simplification of community

structur c

s, which is surprising because of the compaction by the workers and their equipment.

Moisture content was highest in the thick bamboo, followed by the banana then the thin bamboo,

following a trend of understory cover. The lower moisture content in the thin bamboo habitats

can be contributed to the pores created by the roots of the vegetation, shade level and litter depth.

Chemical properties

pH did not vary enough to be discussed in this section. Other chemicals showed minor

variations between these three habitats. The most drastic difference was found in gold, howe

which was 14 ppm in the thick bamboo and less than 2 in the other two habitats, although this

difference is not biologically sig

in the h

not limited.

Biological communities

As expected, the trends in invertebrate abundance followed the trend of land use

intensity. In addition, since the banana had the thickest organic layer, signifying the greatest

amount of leaf fall, it is no surprise that the major decomposer groups were most abundant in tha

habitat. Many of the groups that occurred in the different forest habitats were completely absen

from all three of these habitats. Overall, the simpson’s index of diversity was nearly as low in th

thick bamboo area as the grazed pasture, suggesting that these two areas, where trampling and

land use was most intense, showed the lowest abundance and diversity of invertebrates. Th

absence of many major inv

e. In addition, it is possible that the thick bamboo is excreting a highly toxic allelopathi

compound that is preventing healthy invertebrate communities from establishing themselves.

50

Offsite sampling

These samples were taken in order to compare them to similar habitats in the FCRE

samples and to get more information on the surrounding soils. However, since the samples were

sent to different labs, we were unable to compare the chemical composition of the FCRE samp

with that of the off-site samples due to different analysis methods and units. Only data on the

physical and chemical properties were collected for these sites.

Physical properties

Both of the offsite pastures were sampled while cows were physically grazing, whereas

cows were never present during the 2 months that the sampling was done at the FCRE. Still

bulk density was relatively low in these scenarios, contradicting what was found in previous

studies. In addition, lower moisture content found in FCRE compared to the other two pastu

indicates that differences in compaction.

les

soil

res

or Hacienda Baru samples, the soil became more sandy as the transect progressed

each, grading from estuarine or marshy silts and clays inland to pure, dominantly

silicate e

ocean could contribute be a factor in

phosphorous, calcium and magnesium deposits that were deposited by sea spray and derived

ae, debris, guano and dead animals.

F

toward the b

beach sand. Moisture content at the beach was extremely low, due to the large pore spac

found in more sandy soils. Beach soils also had a high soil bulk density, due to the size of the

particles and thus the inverse relationship between soil bulk density and moisture content was

illustrated well.

Chemical properties

Chemical properties did not vary much between the off-site pastures, which is not

surprising due to the similar vegetation found in both sites. However, there were big differences

in the levels of manganese and iron, which were both found in higher concentrations in the

pasture near the ocean, suggesting that perhaps the plants in that pasture are healthier, due to

their ability to synthesize cholorphyll more effectively and activate enzymes more readily

(Motavallie et al 2008 ). Also, the proximity to the

from marine inputs, such as alg

Interestingly, in the Hacienda Baru samples, there was a trend of increasing phosphorous

concentrations moving from the primary forest toward the beach. This is puzzling, since

51

phosphorous is often considered one of the most limiting nutrients for plants and there w

lower plant growth and diversity found near the beach as opposed to the forest. In Hacienda B

3, the sample taken in the cocoa trees, there were extremely high levels of zinc, magnesium and

calcium, suggesting that the cocoa trees are not able to cycle these nutrients effectively. The

trend of increasing phosphorous with decreasing distance to the beach can be explained again by

sea spray deposits of marine nutrients.

Naturally Regenerated Secondary Forests

Nonbamboo vs. Bamboo

Soil characteristics

In general, the physical properties of the nonbamboo and bamb

as much

aru

oo regions of the naturally

study plot 1 are

relatively uniform. Neither the nonbamboo nor the bamboo areas had an organic layer,

indicating a high decomposition rate in both areas, which corresponds with high rates of

diversity and soil fauna activity in both regions. Both the nonbamboo and bamboo regions had

similar yellow-reddish subsoil colors in the A subsoil horizon, indicating little water logging

(Kohnke and Franzmeier 1995.) which is consistent with the bulk densities. Yellow-reddish

subsoil colors also usually a sign of good aeration (Kohnke and Franzmeier 1995). The low bulk

densities and yellow-reddish soil colors for study plot 1 and naturally regenerated forest areas

were surprising. Given that the FCRE was used for cattle farming, we had expected higher bulk

densities from compaction of previous land use. However, the low bulk densities are consistent

with other studies on land use and recovery in Costa Rica (Reiners et al 1994, Holl 1999).

Like the physical properties, the similar chemical properties of the nonbamboo and

bamboo regions indicate that the chemical elements in the soils of study plot 1 are relatively

uniform despite the differences in vegetation. Slight differences between the nonbamboo and

de a slightly higher percentage of phosphorus, manganese, and copper in the

bamboo soils (Table 13). Phosphorus, a primary nutrient, is used in large quantities by plants

for high r

regenerated secondary forests were rather similar, indicating that the soils in

bamboo soils inclu

energy bonds (ATP) to grow (Clatterback 2008). Micronutrients manganese and coppe

are important for plant enzymes (Clatterback 2008). We cannot determine whether the

increased amounts of nutrients allowed bamboo to grow in that area. In addition, the chemical

properties reveal the effects of previous cattle grazing as most macronutrients are present in

52

smaller quantities in the nonbamboo and bamboo soils than those in the soils of primary for

Both the nonbamboo and bamboo soils have higher quantities of iron, suggesting that leaching

may have reduced the uptake and recycling of nutrients into the forest biomass.

Table 13 : Comparison of macro and micronutrients between plot 1 nonbamboo and bamboo soils, othersoils in the naturally regenerated forests, and primary forests

Element Unit Primary Natural Regeneration Nonbamboo Bamboo Macronutrients

est.

K % 1.19 0.54 0.14 0.13 Ca % 0.62 0.08 0.05 0.05 Mg % 0.95 0.32 0.18 0.22 P % 0.07 0.039 0.079 0.118 S % 0.06 0.032 0.03 0.03 Micronutrients Fe % 7.68 7.47 16.5 16.9 Mn ppm 1250 1190 1270 1490 Zn ppm 146 113 91.6 117 Ni ppm 36.3 40.6 20.1 17.1 Mo ppm < 1 2 < 1 < 1 Cu ppm 81.4 67.4 325 529

Soil fauna characteristics

Results show that soil invertebrate diversity in the nonbamboo and bamboo were

relatively high. The Simpson’s Index of Diversity for bamboo and nonbamboo leaf litter and

extraction samples were similar, all which ranged around 0.75. The bamboo leaf litter sample

had a lower Simpson’s Index of Diversity (1-D= 0.6) which was most likely due to the ant

colony in one of the bamboo samples since Simpson’s Inde

dry

x of Diversity takes account

ofabun also

le for

atter

n both regions. The large number of Hymenoptera (n=202) in the

bamboo r

ve

dance. The species richness between the different areas and collecting methods were

similar except the bamboo dry extraction samples.

Soil macroorganisms including earthworms, termites, and ants are largely responsib

high decomposition rates in tropical regions, which in turn result in low content of organic m

in tropical regions (Lee and Wood 1971). There was no organic layer in the bamboo and the

nonbamboo soils. Earthworms and ants were present in both regions of the naturally regenerated

forests, indicating activity i

leaf litter due to the collection of an ant colony in the sample. Termites, on the othe

hand, were found only in nonbamboo areas (Table 12). Termites are responsible for intensi

mixture of organic and mineral soil components that is typical for surface horizon (Lee and

53

Wood 1971). The lack of termites in the bamboo areas may suggest that termites are unable to

utilize bamboo organic matter.

resentative of secondary naturally regenerated forests?

d

s.

on’s

nces in the chemical composition. Plot 1 soils

ave smaller quantities of macronutrients such as potassium and magnesium and larger quantities

per (Table 13). Plot 1 soils also had less nickel and

molybd e

lants

tities

nerated

e

boo samples are concentrated

ithin a 30m by 30m area. A past study has shown that low carbon accumulations in soils may

such as low soil surface area and warm and humid climate (Richter et al 1999).

Study Plot 1 has a small soil surface area compared to the large naturally regenerated area, and

Plot 1

Is plot 1 rep

The inventories of the vegetation within the naturally regenerated forests of the reserve

are being compiled to gather more information and data in the reserve. It is important to

determine whether the soils of study plot 1 are representative of the greater naturally regenerate

forest areas for further vegetation studies. Results show that the physical properties and soil

invertebrate biodiversity of plot 1 soils are similar to those of other naturally regenerated area

Plot 1 soils and other naturally regenerated forest areas have comparable percentages of sand,

silt, and clay percentages, textures, pH, and soil colors. A comparison between the Simps

Index of Diversity between the nonbamboo and bamboo areas of plot 1 and other sites in the

naturally regenerated secondary forests do not show great differences in soil invertebrate

biodiversity (Figure 14).

However, when the chemical properties of plot 1 soils and other naturally regenerated

forest soils are examined, there seems to be differe

h

of micronutrients such as iron and cop

enum in their soils. These differences in macro and micronutrients may be due to th

recovery rate of plot 1 in comparison with that of the larger naturally regenerated forests. P

require macronutrients in large quantities while micronutrients are required in smaller quan

(Clatterbuck 2008). If plot 1 is recovering at a slower rate than the greater naturally rege

forest area, since micronutrients occur in smaller quantities, it was be easier for plot 1 to

regenerate its micronutrients, whereas since macronutrients are required in such high quantities,

it would take longer for plot 1 to restore high quantities of the macronutrients. Also, the samples

from the naturally regenerated forests are from many different sites dispersed throughout th

naturally regenerated forest area whereas the nonbamboo and bam

w

be due to factors

54

thus it may be more difficult for that small patch to accumulate minerals. Although the physical

properties and invertebrate biodiversity in plot 1 soils are similar to other areas of naturally

regenerated forests on the FCRE, because the chemical elements are important for plant growth

and forest recovery, further studies are needed before we can conclude that plot 1 soils are

representative of the naturally regenerated secondary forests on FCRE.

Vegetation

In Plot 1, it was originally observed that not much vegetation grew around the bamboo

that grew in this naturally regenerated area. The bamboo in this naturally regenerated area was

Bambusa vulgaris. No Guadua angustifolia was observed. The foliar analysis in this study

analyzed minerals, and did not analyze components of organic allelotoxins. From our results,

cannot determine whether Bambusa vulgaris secretes allelotoxins into the soil that inhibits

vegetation growth. Our results merely shows that there are no significant differences in the

chemical compositions of the soils in nonbamboo and bamboo areas.

only

we

other

Current literature does not

nts

and use and forest recovery

Secondary forests deserve immediate conservation attention. Because secondary forests

an accumulate biomass in a relatively short amount of time, they have high potential to be

arbon sinks and provide for other ecosystem services (Alaverz-Yepiz et al. 2008). By

cilitating forest recovery in disturbed area, we can potentially influence the fate of tropical

inforest. We need to implement restoration and conservation practices to reduce erosion and

hemical changes caused by overuse, salinization, acidification or contamination of the soil.

owever, the nature of natural forest regeneration depends directly on the intensity, duration and

haracter of previous land use (Zamora and Montagnini 2007). Thus, we must first address

tructural degradation, such as compaction and reduced porosity, and then facilitate plant growth.

ince forest recovery is directly linked to the restoration of soil fertility, which is in turn closely

suggest that the species Bambusa vulgaris secretes allelotoxins, but given the lack of vegetation

around the bamboo, this may be an interesting topic to explore for future studies. The lack of

vegetation around the bamboo may also just be due to the nature of bamboo clusters. Since

bamboo grows in clusters, the high number of bamboo sprouts at each site requires high amou

of nutrients and outcompete other plant species.

L

c

c

fa

ra

c

H

c

s

S

55

related to above ground biomass, planting trees and adding organic material to the system will

ss. There are a variety of short-term and long term approaches to restoring accelerate the proce

soils that have severely been affected by human land use (Table 14 ). Table 14: Short-term and long-term approaches to soil problems in ecological restoration.

(Dobson et al 1997).

Assisted natural regeneration (ANR) is a method used for forest restoration that aims at

sing low cost planting methods to restore biodiversity and establish diverse commercial

lantations. ANR accelerates succession by removing or reducing barriers to natural forest

egeneration (Shono et al. 2007). Factors limiting post-dispersal seedling establishment include

ompetition from existing plants, lack of appropriate micro-sites for germination, and seed

redation (Doust et al. 2008). In addition, scarcity of nutrients, soil compaction, lack or excess of

oil humidity, high solar radiation and seed availability can limit tree regeneration (Zamora and

ontagnini 2007). In addition, seed dispersal rates are often the limiting factor in plant growth.

f we start with altering the seed dispersal abilities, we can then address other factors, such as

ompetition. Trees can be planted, in the form of a plantation, in order to attract seed dispersers,

uch as birds and bats, and increase the frequency of seed arrival in the disturbed areas. The

resence of tall trees also provides shade that can suppress the growth of grasses and other

u

p

r

c

p

s

M

I

c

s

p

56

herbaceous vegetation that requires high amounts of solar radiation. Regeneration is often slow

r

improve soil fertility and hasten recovery of species

ompo

ing these types of trees will be

ssential in the land recovery of FCRE, since increased soil N can enhance the capacity of the

ystem to support a more complex biological community.

By facilitating plant growth, we can not only enhance further plant growth but also

aintain a healthy fungi community. This is essential in order to enhance decomposition and

utrient cycling in the ecosystem. Fungi, particularly mycorrhizal fungi, which have more

mited dispersal and colonizing ability than bacteria, can be highly susceptible to land use

hange (Kasel et al. 2008). These microbe communities rely heavily on plant litter as their source

f nutrition, thus by facilitating vegetation growth, we can also enhance the fungi which perform

ital ecosystem processes.

We can make a case for the importance of forest conservation based on the data we

ollected in this study on the biological communities. Simply based on the important roles

vertebrates play in nutrient cycling and organic decomposition, as well as being towards the

ottom of a complicated forest food web, their conservation is essential if we want to maintain

ealthy forest communities and diverse tropical vegetation.

IMITATIONS AND RECOMMENDATIONS

imitations

We encountered some difficulties throughout this study. While we sampled 5 replicate

lots from each habitat, this turned out to be too few replicates to do statistical analysis with our

ata. Thus, we were only able to analyze our data qualitatively, which was difficult due to the

mount of data and the number of variables measured. In addition, while we chose the habitats

in open pastures due to degraded soil, but also because there are no good perches or habitats fo

seed dispersers, thus plants that are wind pollinated, such as grasses and ferns, are able to out-

compete tree seedlings (Zamora and Montagnini 2007). In areas with degraded soils, direct

planting of exotic or native trees to

c sition (Chazdon 2008). ANR lays out a set of steps to follow in order to achieve some of

these things (Shono et al. 2007).

Macedo et al. (2008) found that the use of legume trees was an efficient method for re-

establishing the nutrient cycling processes of a forest ecosystem. Certain legume trees form

efficient N2-fixing symbioses with mycorrhizal fungi. Identify

e

s

m

n

li

c

o

v

c

in

b

h

L

L

p

d

a

57

randomly, sometimes it was difficult to find a suitable place to sample with a consistent flat

find a G

selectio rmining significant differences in chemical soil

not

have lo s

was the first soil survey for the FCRE and we hope the data will provide a baseline in which

Forest

Our results indicate the physical properties within the FCRE property has remained

tructural properties have not been exceedingly destroyed by cattle grazing and provides a

facilitating the accumulation of healthy levels of nutrients in the soils. The restoration of

especially keystone soil fauna species. We recommend continually tracking the changes in soil

restora ific habitat within the FCRE.

Future Studies

data for anyone wanting to continue soil research on FCRE or do other ecological studies that

to look

suggest

portant to examine the role of soil at the FCRE in the carbon cycle (Richter et al 1999).

acronutrient, is only stored in organic matter, which is then released by slow decomposition.

n et

l 1997). Along with the data we collected, it would be important to incorporate the other

slope that was near a marked GPS tag on the FCRE property. In the future, it would be useful to

PS that would work under the forest canopy, thus allowing more freedom in site

n. Another challenge we faced was dete

composition in our samples. Although we examined other soil studies in Costa Rica, we did

cal data (such as from Hacienda Baru or Dominical area) to compare our data with. Thi

future soil composition changes can be tracked.

Recovery Recommendations

consistent despite its past history of cattle grazing. This is encouraging in that some of the

s

gateway to restoring the habitat. The next step in restoring the FCRE property includes

macronutrients and micronutrients will greatly help facilitate the growth of the soil community,

composition and soil fauna and further studies to better evaluate and design effective land

tion strategies for each spec

The next step this study is to enter all of the data into GIS, in order to provide accessible

relates directly to soil properties. Also, it would be important to do carbon analysis in the future

at biomass how carbon is cycled in soils in the tropics. Although previous studies

that mineral soils have low carbon accumulation and storage potential, it would be

im

Additionally, further nitrogen analysis in the soils is essential. Nitrogen, an important

m

As a result, nitrogen accumulation is a limiting factor to tropical forest development (Dobso

a

58

factors in soil formation, such as landform age, geologic parent material, slope and elevation.

Slope and elevation would be particularly interesting to examine in FCRE since the entire

e

hill and

Vegetation surveys are already in progress in the naturally regenerated forests at FCRE,

be inte rveys into other habitats. In addition, an expansion of

bambo ry in that habitat. We

CONC

major c . There was an

soil in areas with larger impacts and less

ecovery time. And most drastically, we saw a change in the soil invertebrate abundance and

s opposed to the more disturbed habitats. Since soil fauna has been known to be a good

the naturally regenerated secondary forest and the hardwood forest have the greatest abundance

property lies on a slope, so levels of erosion and nutrients can be compared from the top of th

the bottom. Transects would be a useful tool to look at this.

however, it will be an important step to correlate the vegetation data with the soil data. It would

resting to expand the vegetation su

our Foliar analysis of the bamboo would be to examine the organic allelotoxins that the thick

o might be excreting that is preventing the growth of understo

recommend doing an analysis of the organic compounds in the soil to account for this.

LUSIONS

In this study, we found evidence that deforestation and other human disturbances cause

hanges in the chemical properties of soil and the diversity of soil in fauna

overall decrease in macro and micronutrients in the

r

diversity, with much healthier communities in the forested habitats, especially the primary forest,

a

bioindicator of soil health and human impacts, we can see that this is indeed true. Because both

59

and diversity of invertebrates, besides the primary forest, it is possible for a disturbed area to

umber of things humans can do to facilitate the recovery of forest.

assistan like to thank Professor Juan

helpful history. Thanks to Dr. Diane Thompson, Chris

our

data. W ng set up the experiment in

Costa Rica and for providing a lot of the equipment we used.

recover significantly in a window as short as 15 years. This is encouraging, since there are a

n

ACKNOWLEDGEMENTS

We would like to thank Dr. Jonathan Wright and Dr. Rick Hazlett for their help in the

field in Costa Rica with sampling and experimental setup and also for their continual support and

ce during data analysis and draft revision. We would also

Araya for his help in Costa Rica.

Thanks to Carol Brandt for helping to organize the research in Costa Rica and provide

background information on the FCRE

Gurney, and Ashley Scott that measured the vegetation on Plot 1 this summer for supporting

e would also like to thank Dr. Donald McFarlane for helpi

60

ide, M., Zimmerman, J., Rosario, M., and Marzano, H. 1996. “Forest recovery in abandoned

ira, M., Sampedro, L., Monroy, F., and Domínguez, J. 2008. “Detritivorous earthworms

ties related to land use history of old-growth and secondary tropical dry forests in northwestern Mexico.” Forest Ecology and Management. 256: 355-366.

REFERENCES

Acattle pasture along an elevational gradient in Northeastern Puerto Rico.” Biotropica. 28(4): 537- 548.

Adirectly modify the structure, thus altering the functioning of a microdecomposer foodweb.” Soil Biology & Biochemistry. 40: 2511-2516.

Alvarez-Yepiz J.C., Martinez-Yrizar, A., Burquez, A, and Lindquist, C. 2008. “Variation in

vegetation structure and soil proper

61

g, T., and Diehl, M. 1992. Deforestation of tropical rain forests. Economic causes animpact on development. Tuebingen: Mohr Publishing.

Amelun d

ck.”

y. 6:

amilo, G.R. and Zou, X. 2001. “Soil Fauna in Managed Forests: Lessons from the Luquillo

Experimental Forest, Puerto Rico.” In R.A.Fimbel, A. Grajal, J.G. Robinson . eds. The Cutting Edge: Conserving Wildlife in Logged Tropical Forests. New York, NY: Columbia University Press pp. 289-303.

hazdon, R.L. 2003. “Tropical forest recovery: legacies of human impact and natural disturbances.” Perspectives in Plant Ecology, Evolution and Systematics. 6: 51-71.

hazdon, R.L. 2008. “Beyond Deforestation: Restoring Forest and Ecosystem Services on Degraded Land.” Science. 320: 1458-1460.

latterbuck, W.K. 2008. “ Nutrient Deficiencies in Trees.” The University of Tennessee. Knoxville, TN.

leveland, C., Townsend, A., Schmidt, s., Constance, B. 2003. “Soil microbial dynamics on biogeochemistry in tropical forests and pastures, southwestern Costa Rica.” Ecological Applications. 13 (2): 314- 326.

enslow, J.S., and Guzman, G.S. 2000. “Variation in stand structure, light and seedling abundance across a tropical moist forest chronosequence, Panama.” Journal of Vegetation Science 11: 201-212.

obson, A.P., Bradshaw, A.D., and Baker, A.J.M. 1997. Hopes for the Future: Restoration Ecology and Conservation Biology. Science 277: 515-522.

Angers, D and Caron, J. 1998. “Plant-induced changes in soil structure: processes and feedba

Biogeochemistry. 42: 55-72.

Bertoncini A.P. and Rodrigues, R. R. 2008. “Forest restoration in an indigenous land considering a forest remnant influence (Avai, Sao Paulo State, Brazil).” Forest Ecology and Management. 225: 513-521.

Bowen, G.D., and Nambiar, E.K.S. 1984. Nutrition of Plantation Forests. New York: Academic Press, pp. 516.

Brown, S. and Lugo, A.E. 1990. “Tropical secondary forests.” Journal of Tropical Ecolog

1-32. Brussaard, Lijbert. 1997. “Biodiversity and Ecosystem Functioning in Soil.” Ambio 26 (8): 563-

570.

C

C

C

C

C

D

D

62

Doust, S., Erskine, P., Lamb, D. 2008. “Restoring rainforest species by direct seeding: Tree stablishment and growth performance on degraded land in the wet tropics of

Australia.” Forest Ecology and Management. 256: 1178- 1188.

reckman, D., Blackburn, H., Brussaard, L., Hutchings, P., Palmer, M, Shelgrove, P. 1997. ents.” Ambio. 26 (8):

556-56

le l. 1 he Di of S m t r Tropical Rain Biodiversity and Conservation 5:135-168.

zá izelle, Seastedt, T. 2001. “Soil Fauna Plan er Decomposition in Tr

ubalpin rests.” logy 955-

enw .L., M zie, B 200 raz fect oil phy properties and the c quences f pasture: eview Aust. J. Exp. A . 41: 12 250

rigua .R. and ertag, R 01. otro seco forest su ess change n structural and functiona t Eco and M emen 148:206.

H iend Res Prec tion. ta Rica esto enteEcological Research, http://costa ont.edu/weather.shtml.

lett, Ri ard. 200

l, K. . “Fact miting tropical rainforest nera n aband pa s: seed rain, seed germination, microclimate and soil.” B . 31 (2): 229-242.

l, K oik. M in. E. Sam I.A. 2000. “Tropical Montane Forest Restoration in Costa Rica: Overcoming Barriers to Dispersal and Establishment.” Restoration Ecology. 8: 339-349.

ITTO Guidelines for the restoration, management and rehabilitation of degraded

and secondary tropical forests. IT Policy velop nt Series . 13.

Hughes R. F., Kauffman, J.B. and Cummings, D.L. 2002. “Dynamics of aboveground and soil c n and n en stoc d cy of ble en aloRondonia, Brazil.” Ecosystems 5 -25

so nd Wedin, D. 1997. “Soil Carbon, Nu ts, and Mycorrhizae During Conversion

of Dry Tropical Forest to Grassland.” Ecological Applications. 7 (1): 171-182.

el, S nnet, L bits, J 8. “L use in nces fungal un mp n across central Victoria, s Soil Biology & Biochemistry. 40: 1724-1732.

seedling e

F

“Linking biodiversity and ecosystem functioning of soils and sedim2.

996. TGil r, Pau versity oil Com unities, he “Poo Man’s forest”.

Gon lez, Gr and t Litt opical and S e Fo Eco 82: 964.

Gre ood, K cKen .M., 1. “G ing ef s on s sical onse or a r .” gric 31-1 .

Gua ta M Ost . 20 “Ne pical ndary cc ion: s i

l characteristics.” Fores logy anag t. 185-

ac a Baru Reserve. 2007. Baru erve

rica.jsd.claremipita Cos : Fir ne C r of

Haz

ch 8. personal communication.

Hol 1999 ors li rege tion i oned stureiotropica

Hol .D., L .E., L H.V., uels,

ITTO. 2002. TO De me No

arbo itrog ks an cling availa nitrog ng a land-se gradient in : 244 9.

John n, N. a trien

Kas ., Be ., Tib . 200 and flue soil comm ity co ositioouth-eastern Australia.”

63

Koh n i 9 oi nce Simplifie rove, IL: elan

ss,

Lee, K.E. and Wood, T.G.. 1971. Termites and Soils. New York: Academic Press.

ge, D catena ., Asb .E., anc M.J. 1. Fine fal relatenutrients inputs resulting from Hurricane Hugo in subtropical wet and lower montane

orests o rto R 23 -34

Lodge, D.J., McDowell W.H., and McSwiney, C.P. 1994. The importance of nutrient pulses in al fore 9: 384-342.

Macedo, M., Resende, A., Garcia, P., Boddey, R., Jantalia, C., Urquiaga, S., Campello, E., o, A. “Cha n so nd ks trient d ics 13 years after

ino roge ing tree ore ology Manangement.

alter, D. and Hunt, W.H. tion of microbiota and biota in w grou ood .” Annual Review of Entomology 3 19-43

.H. 1993. “Influence of productivity on stability o l and m ecosyste s.” Science 26 6-90

t con and function in plants.” College of Agriculture ife ces. Un ity

.<http://www.cartage.org.lb/en/them ienc tanica ce tHo es/EssentialPlant/EssentialPlant.htm>

Pereira, I.M., Andrade, L.A., Sa io, E .B., osa, V., 200 se ry ostructure and flora of Caatinga.” Biotropica 35: 154-165.

ente and K g, N. 1 . “Ec y of rosi ecosys .” Ecosystem 5): 416-426.

-tree 96. “Ra est Inf tion, ts, an eserv .” <www.rain-tree.com/facts.htm>

ers, ., Bou , A.F. ons .J, Keller, M. 1994. “Tropical Rain Forest Conversion to Pasture: Changes in Vegetation and Soil Properties”. Ecological Applications 4 (2): 363-377.

ver of so rbon in e-est hing t. Nature 400: 56-58.

nke, J., aPre

d Franzme Inc.

er, D. P. 1 95. “S l Scie d.” Long G Wav d

Lod .J., S , F.N ury C and S hez, 199 litter l and d

rain f f Pue ico. Biotropica : 336 2.

tropic sts. Trends in Ecology and Evolution

Franc 2008. nges i il C a N stoc and nu ynamrecovery of degraded land using legum us nit n-fix s.” F st Ec and

225: 1516-1524. Moore, J. C., W 1988. “Arthropod regula

websmeso belo nd f 3: 4 9. Moore, J.C. and De Ruiter, P.C. and Hunter, W

r rea odel m 1: 90 8.

Motavalli, P., Marler, T, Cruz, F., and McConnell, J. 2008. “Nutrien centrationivers

s of and L Scien

Guam es/Sc es/Bo lScien s/Plan rmon

mpa .V. S Barb M.R. 3. “U -histo n

Pim l, D., ounan 998 olog soil e on in tems s. 1 (

Rain . 19 infor orma Fac d Pr ation

Rein W.A wman , Pars , W.F

Richter, D.D., Markewitz, D., Trumbore, S.E and Wells C.G. 1999. Rapid accumulation and

turno il ca a r ablis fores

64

Sala, E., Aburto-Oropeza, O., Paredes, G., Barrera, J.C. and Dayton, P.K. 2002. “A general l n s rine reserves.” Scien 1-1993.

aller, 968. Soil Animals. The University of Michigan Press: An rbor

aetzl, nd And n, S. 20 Soils: Genesis and Geomorphology. bridge University Press: New York.

rrow, ., 2007. il com ion b zing stoc ilvopast as ev nced

c es in so ysical erties grof t. Sy : 215-22

no, K adawen , Durs 007 plic of A ed Natur ege tion Restore Degraded Tropical Forestlands.” Restoration Ecology. 15 (4): 620- 626.

er, W , Brown and Lu A.E. 6. “Biodiversity and biogeochemistry in tropical

fo s.” In G ians, R rzo, a . Cu an, ed iodiver and systemProcesses in Tropical Fo pp. 49-67. New York, N.Y.: Springer-Verlag.

ins ip. 199 posi in Tro Lo Racology

ft, M d Ande , J.M. 3. Bi ersity and ecosystem function. In E.D. Schultze and H.A. Mooney, eds., Biodiversity and Ecosystem Function, pp. 15-41. Heidelberg, Germany: Sp er-Verl

mpso Brokaw Zimme n J.K., Waide R.B., Everham E.M., L ge D TayloC.M., Garcia-Montiel D. Flue . 200 and istory, en ronm , and

Ecological Applications 12: 1344-1363.

, C. . Facto ntrolli ccession follo slas -burn lture in Am75: 377-407.

os, E.C. and Kooistra, M.J. 1994. The effect of soil structure differences in a silt loam soil under various farm management systems: soil physical properties and simulated land qualities. Agriculture Ecosystems and Environment 51: 227-238.

alker L.R. 1999. Patterns and processes in primary succession. Ecosystems of Disturbed Ground ( ed. LR Walker), pp. 585-611. New York: Elsevier.

Whitmore T.C. and Burslem D.F.R.P. 1998. Major disturbances in tropical rainforests.

Dynamics of Tropical Communities (eds. DM Newbery, HHT Prins and ND Brown), pp. 549-565. Oxford: Blackwell Science.

WRI (World Resources Institute). 1994. World Resources, 1994-95: A Guide to the Global

Environment. New York, N.Y.: Oxford University Press.

mode for designi g network of ma ce 298: 199 Sch F. 1 n A .

Sch R. a erso 07. Cam

Sha S.H “So pact y gra live k in s ures ide by hang il ph prop .” A ores st. 71 3.

Sho ., C g, E. t, P. 2 . “Ap ation ssist al R nera to

Silv . L. , S., go, 199rest . Or . Di

rests, nd H shm s., B sity Eco

Soll , Phill

Forest: Does8. “Factors Influencing Species Soil Matte

Com tion pical wland in r?” E 79 (1): 23-30.

Swi .J. an rson 199 odiv

ring ag. Tho n, J. N, rma od .J., r

, and t M 2. L use h vi ent tree composition in a tropical forest.

Uhl 1987 rs co ng su wing h-and agricu azonia.

Journal of Ecology V

W

65

WRI (World Resources Institute). 2008. Protecting Forests to Save the Climate: REDD Challenges and Opportunities. Washington DC: World Resources Institute.

Zamora, C., and Montagnini, F. 2007. “Seed Rain and Seed Dispersal Agents in Pure and Mixed Plantations of Native Trees and Abandoned PaCosta Ric s E gy. 3) 3-

A Chemical analysis Appendix 1: Chemical properti of abitats and study plot 1 of FCRE

Chemical UPrimary forest

regenerated forest

Hardwood forest

Thick Bamboo

Thin Bamboo Banana

Recovering pasture Pasture

Plot 1- non-bamboo

Plot 1- bamboo

stures at L 4 .

a Selva Biological Station, a e.” R to nratio colo 1 (5 : 45 61

PPENDICES

es all hNaturally

nit

Au b < 2 < 2 < 2 14 < 2 < 2 < 2 < 2 < 2 < 2 pp

Ag ppm < 0.05 < 0.05 < 0.05 < 0.05 0.06 0.13 0.09 < 0.05 < 0.05 < 0.05

Cu ppm 81.4 67.4 129 610 373 657 806 373 325 529

Cd ppm < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1 < 0.1

66

Mo ppm 2 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1 < 1

Pb p 12.2 9 13.2 6 7.3

Ni p 4 63.3 11.2 11.3 10 20.1 17.1

Zn pp 3 87.7 154 135 186 61 174 91.6 117

S 0 0.02 0.04 0.01 0.03 03 0.04 0.03 0.03

Al 8.9 4 11.1 .8 .8 10.6 8.76 12.3

As pp 8 8 4 4 < 0.5 5 6 11 < 0.5

Ba pp 8 416 336 456 1080 69 689 127 140

Be pp .5 0.6 1 0.8 1.4 9 0.6 < 0.1 0.9

Bi pp < 0.1 < 0.1 0.1 < 0.1 < 0.1 .1 < 0.1 < 0.1 < 0.1

Br pp 13 19 8 22 22 21 27 20

Ca 0.62 0 0.26 1 0.14 08 0.16 0.05 0.05

Co pp 28 24 38 41 30 32 28

Cr pp 8 3 141 6 37 5 < 1 56 < 1

Cs pp 1.15 0 0.51 1.05 1.71 1.24 94 2.45 1.42 1.91

Fe 7 5.3 13.8 14.1 .3 13.3 16.5 16.9

Hf pp 0.1 1.5 0.9 2.8 2.6 2 < 0.1 0.3 1.5

Hf pp 4 5 5 8 9 7 7 11 7 7

Ga pp 1 .9 23.4 29 .6 .6 25.3 20.3 32.9

Ge pp .8 0.1 0.3 0.2 0.7 2 0.2 0.2 0.3

Hg pp 1 1 < 1 < 1 < 1 < 1 < 1 < 1

Chemical Un

Naturallregenera d forest

Hardwood forest

ThBa

in mboo Banana

Recov ng pastur Pasture

Plot 1- non-bamboo

Plot 1- bamboo

Ir p 5 < 5 < 5 < 5 < 5 < 5

K 9 0 4 0.25 0.41 14 0.42 0.14 0.13

Li pp 3 .4 19.6 0.9 17.3 .7 12.9 8.3 10

Mg 5 0 0.21 .13 0.34 31 0.25 0.18 0.22

Mn pp 0 1 2470 50 2420 1270 1490

Na 2 0 0.17 0.05 0.08 0.08 04 0.07 0.04 0.02

Nb pp 0.3 0.7 0.6 1.4 4 0.2 0.1 0.6

P 0. 9 0.066 0.184 0.229 12 0.12 0.079 0.118

pm 4.3 5.9 5.2 9.9 12.5

pm 36.3 0.6 6.6 4.9

m 146 11 1

% 0.06 .02 0.

% .89 12.4 9.88 11 12

m 11 < 0.

m 656 19 2

m 0.2 0 0.

m < 0.1 < < 0

m 15 27 1

% .08 0.1 0.1 0.

m 20 25 28

m 10 13 < 1

m .65 0.

% 7.68 .49 8.61 1 17

m < 3.9 0.

m

m 10.6 6 .6 29 28 29

m 0.1 0 0.

m < < < 1 < 1

it forest Primary

y te ick Th

mboo Baerie

pb < < 5 < 5 < 5 < 5

% 1.1 .54 0.2 0.17 0.

m 41.5 7 12.7 1 14

% 0.9 .32 0.22 0 0.

m 125 190 1100 2060 2160 23

% 0.5 .05 0.

m 0.2 5.9 0.

% 0.07 03 0.168 0.

67

Rb pp 10 40.5 .2 25.6 9.3 11.9 m 28.1 6.5 15.9 16.9 11

Re pp 0. .001 0.001 < 01 <

0.001 < 0.001 0.001

Sb pp .9 0.8 0.7 < 0.1 .1 0.6 0.9 < 0.1

Sc pp 3 37.4 .2 36.5 .1 31.5 56.4 54

Se pp 1 1 1.6 7 < 0.1 < 0.1 1

Sn pp 2 < 1 < 1 1 1 1 < 1

Sr pp .4 24.4 .6 .9 39.2 28 64.1

Ta pp .4 < 0.1 0.1 < 0.1 < 0.1 .1 < 0.1 < 0.1 < 0.1

Te pp 1 < 0.1 .1 < 0.1 .1 < 0.1 < 0.1 < 0.1

Ti 0.22 0 3 2 0.55 32 0.21 0.39 0.26

Th pp 1.2 1.2 3 9 7.7 2.7 2.4

Tl pp 0.18 0 1 0.09 09 0.11 0.07 0.09

U pp 2.1 3.9 .4 2.6 5 2.4 1.1 1.9

V pp 85 3 107 165 20 58 203 281

W pp < 1 < 1 < 1 < 1 < 1 < 1 < 1

Y pp < .1 14.4 26.1 24.8 46.7 .8 25 12 12.9

Zr pp 67 128 128 14 4 9 66

La pp .2 7.2 3.4 57.5 .6 33 8.6 9.3

La pp .8 10 24.1 43.2 71.4 .6 39 15.8 12.5

Ce ppm 18.9 0.9 24.3 62.9 62.1 92.9 47 71.5 34 42.7

Chemical UPrimary forest

Naturally regenerated forest

Hardwood forest

Thick Bamboo

Thin Bamboo Banana

Recovering pasture Pasture

Plot 1- non-bamboo

Plot 1- bamboo

Pr ppm 2.3 0.1 2 6.1 10 17.7 5.7 8.4 2.7 3.4

Nd ppm 9.8 0.4 8.5 25.2 38.6 71.4 24 32.7 11.8 14.2

Nd ppm 19 < 5 < 5 22 31 70 22 32 26 13

Sm ppm 2.4 0.1 1.9 5.3 7.5 14.2 5.3 7 2.8 3.2

Sm ppm 3.2 1.1 2.2 5.7 7.7 13.3 5.3 7.1 4.2 3.4

Eu ppm 0.81 0.05 0.63 1.53 2.01 3.96 1.7 1.88 0.83 0.9

Eu ppm 0.7 0.5 0.9 1.6 2.3 4.4 1.8 2.7 1.5 1

Gd ppm 2.9 0.2 2.5 6.1 7.2 13 6.1 6.4 2.9 3.1

m < 0.001 004 < 0 0.002 < 0.001 0.0

m < 0.1 0 < 0.1 < 0

m 30.7 0.7 42.3 38 53

m 0.3 1.2 1.3 1. 0.

m 1 < 1 < 1

m 64.7 6 19.2 29.9 53 19

m < 0.1 0 < < 0

m < 0. 0.1 < < 0.1 < 0 < 0

% .77 0.2 0.61 0.2 0.

m 0.6 2.4 6.7 1.

m .18 0. 0.09 0.11 0.

m 1.8 2.5 3 2.

m 191 13 293 3

m < 1 < 1 < 1

m 17.2 0 26

m 4 142 50

m 8 0 19.2 3 17

m 10.8 4 21

nit

68

Dy ppm 8 4.7 5.5 2.5 2.5 3.2 0.4 2.2 4.5 5.1 8.

Tb pp 0.5 < 0.4 1 0.9 0.9 0.4 0.5

Tb pp < 5 < 0.5 < 0.5 1.1 2.2 < < 0.5 < 0.5 < 0.5

Ho pp 0.8 0.6 1.2 2.1 1.2 0.6 0.6

Er pp 1 4 1.7 3.2 3.5 5.6 3.2 1.6 1.7

Tm pp 1 0.3 0.5 0.5 0.8 0.5 0.2 0.3

Yb pp 4 1.6 3 3.1 4.8 2.7 1.5 1.7

Yb pp 2.9 1 2.6 4.6 6.6 5.4 3.6 2.7

Lu pp < 0.1 < 1 0.2 0.5 0.2 0.2 0.2

Lu pp 0. 7 0.46 0.66 0.87 1.07 0.88 0.56 0.43

Mass .897 1. 1 1.01 1. 21 1.15 0.897 0.945 1.11

m 0.1 0.9 1.8

m < 0.5 0. 0.5

m 0.1 1.1 1.2

m 2. 0. 3.3

m 0.4 0. 0.5

m 0.7 0. 3

m 2. 4.3 4

m 0. 0.4 0.4 0.4

m 0.48 3 0.75

g 0 0 01 1. 1.07

Appendix 2: Offsite chemical analysis pH cmol(+)/L % mg/L H2O Ca CICE SA P Zn Cu Fe Mn ACIDEZ Mg K 5.5 4 5 10 3 1 10 5 0.5 1 0.2 Pasture eet 6.5 0.23 24.04 27.42 1 2 2.2 6 32 14 across the str 2.64 0.51Pasture 5.5 0.56 16.06 21.98 3 ND 2.8 5 86 93 near ocean 4.90 0.46Haciend 5.6 0.34 0.23 15.30 2 1 1.8 3 105 49 a Baru 1 1 4.44 0.29Haciend 5.2 1.36 2.50 6.03 23 1 1.7 3 114 98 a Baru 2 1.90 0.27Haciend 5.7 0.24 27.75 34.52 1 2 4.2 4 21 14 a Baru 3 6.23 0.30Haciend 5.9 0.38 3.11 4.55 8 2 0.8 1 32 12 a Baru 4 0.96 0.10Haciend 5.8 1.96 6.07 0.28 11.23 17 1.7 3 94 11 a baru 5 2.92 5

69

Appendix 3: Nitrogen and MO chemical

analysis

% N MOPrimary forest 0.28 2.7 Naturally Regenerated Sec

3.1 ondary

Forest 0.31 Hardwood forest 3.5 0.28Thick bamboo 5.5 0.35 Thin bamboo 3.3 0.23 Banana plantation 6 3.8 0.2Recovering pasture 0.26 4.2 Grazed Pasture 0.30 4.3 Pasture across the street 2.5 Pasture near ocean 3.1 Hacienda Baru 1 4.5 Hacienda Baru 2 5.9 Hacienda Baru 3 3.5 Hacienda Baru 4 4.7 Hacienda Baru 5 1.6

vertebrate Diversity

ppendix 4: All taxon found in litter samples at FCRE

ON (Order) Sub-order Family Number found

In A TAXOligochaeta 1 Gastropoda 10

nychophora 1 poda Oniscidea Oniscidae 45 poda Oniscidea Trichoniscidae 115

erida 7 icillata 21

lydesmida 5 eophilomorpha 10

OIsoIsoGlomPenPoG

70

Lithobiomorpha 4 Scolapendromopha 4

nae 37 i 45 i Trom diidae 15 ion Trogl 10

46 c 5 a yg 13

26 ona 102 eon 52

t 1 2

a 3 p 1

Ascalaphidae 3 2

t 1 t Aphid 5 t Aradi t Blissi t hyncha Cicad t hyncha Cicad t cerata Mirid 3 t cerata Nabid 7 t Phop 1 t ptear Piesm 5 t Redu t Rhop t cerata Threo 16 t 2

fera 58 3 Byrrh 2

Coleop Adephaga Carab 1 Polyphaga Chrys melidae 1

haga lerid 4 op Polyphaga u 13

Latrid 6 Mord 1

Ocho 3 Scara

Scydm Silphi 1 Staph 20

p a Acroc

Ara Acar Tetranychidae Acar biOpil es obedae Pseudoscorpion Ricinu lei

JapDiplurProtura

idae

Collembola ArthropleCollembola Symphypl a Odana a ZygopteraBlattodae Isopter Derma tera Neuroptera Psocoptera Orthop era Hemip era idae Hemip era dae 4Hemip era dae 9Hemip era Auchenorr ellidae 11Hemip era Auchenorr idae 3Hemip era Gymno ae Hemip era Gymno ae Hemip era alidae Hemip era Hetero atidae Hemip era viidae 2

1HemipHemip

era era

Gymno

alidae coridae

Hemip era Thysanoptera TubuliCoeloptera Coleoptera ida

tera idaeColeoptera oColeoptera Polyp C ae Cole tera Curc lionidae Coleoptera iidae Coleoptera ellidae Coleoptera Polyphaga daeidae Coleoptera Polyphaga bacidae 3

1Coleoptera aenidae ColeopColeop

tera tera

dae ylinidae

Coleoptera 95 10

Tricho tera Dipter eridae 2

71

Diptera cera BombBrachy yliidae 1 Diptera rhapha Callip 2 Diptera Cecid 3

a Chiro e 1 a rhapha Musc 1 a Polic 1 a Psych 26 a Sumi 4 a rhapha Tachi 2 a Tipul 2 a 7 p 2

ta Braco 1 ta Evan 2

Form 2 yta Sirici 1

1

Cyclor horidae omyiid nomidaDipter Nematocera

Dipter Cyclor idae Dipter hopodidae Dipter Nematocera odidae Dipter Nematocera liidae Dipter Cyclor nidae Dipter Nematocera idae Dipter Lepido era Hymenoptera Apocri nidae Hymenoptera Apocri iidae Hymenoptera icidae 28Hymenoptera Symph dae Hymenoptera

Appendix 5: All ta in dry extraction at FCRE

er Family N mber found

xon found Order Sub-ord uOligoc 3 haeta Gastro 9

d a Oniscidae 1 d a Tric 22 i 5 h 2 b 1 p 1

22 Orb 5 Tetr 9

poda Isopo a OniscideIsopo a Oniscide honiscidae Penic latta Geop ilomorpha Litho iomorpha Scola endromorpha Aranae Acari itidae Acari anychidae

72

Acari Trom bidiidae 38Pseudoscorpion 4

r Japy 12 r 17 m phypleo a 504 m ropleon r Kalo 7

pennia Myr dae 3 p 1 p Aly 1 p Aph 3 p enorrhyncha Cica 77

Hemiptera Form ae 2 Hemiptera Gynmocerat Mir ae 5

Gymnocerat Nabi e 1 mip Pho 1

iptera Heteroptera s 1 p ta Pyrr 2 p rat Thy e 1 n a 18 p

Coleoptera Bostrichidae 1 Coleoptera Byrrhida 3 Coleoptera Carabidae 1 Coleoptera Polyphaga Chrysomelidae 8 Coleoptera Polyphaga Cleridae 4 Coleoptera Curculionidae 55 Coleoptera Polyphaga Histeridae 2 Coleoptera Latridiidae 14 Coleoptera Gymnocerata Pyrrhocoridae 1 Coleoptera Polyphaga Scarabacidae 6 Coleoptera Scydmaenidae 4 Coleoptera Staphilinidae 13 Coleoptera 24 Trichoptera 105 Diptera Acroceridae 7 Diptera Brachycera Bombyliidae 1 Diptera Cyclorrhapha Calliphoridae 2 Diptera Cecidomyiid 19 Diptera Nematocera Chironomidae 17 Diptera Nematocera Culicidae 15 Diptera Cyclorrhapha Dropophilidae 9 Diptera Cyclorrhapha Muscidae 13 Diptera Mycetophilidae 12 Diptera Nematocera Psychodidae 317 Diptera Cyclorrhapha Sarcophagidae 1 Diptera Nematocera Simuliidae 38

Cyclorrhapha Tachinidae 9 Diptera Cyclorrhapha Tephritidae 7

Diplu a gidae Protu a Colle bola Sym nColle bola Arth a 27Isopte a termitidae Neuroptera Plani meleontiOrtho tera Hemi tera didae Hemi tera ididae Hemi tera Auch dellidae

icida id

Hemiptera a daHe tera palidae Hem Pie matidae Hemi tera Gymnocera hocoridaeHemi tera Gymnoce a reocoridaThysa optear TubuliferCeleo tera 1

Diptera

73

74

Diptera Nematocera Tipulidae 8 Diptera 131 Lepidoptera Limacodidae 1 Lepidoptera Pyralidae 1 Lepidoptera 1 Hymenoptera Apocrita Braconidae 1 Hymenoptera Symphyta Diprionidae 2 Hymenoptera Formicidae 70 Hymenoptera Apocrita Megachilidae 1 Hymenoptera Symphyta Siricidae 1 Hymenoptera Apocrita Sphecidae 1 Hymenoptera Symphyta Tenthredinidae 1 Hymenoptera 4

Appendix 6: Identified invertebrates in nonbamboo areas of naturally regenerated forests

LEAF LITTER phylum subphylum class subclass order suborder family abundance Annelida Oligochaeta Haplotaxida 1 Mollusca Gastropod 2 Mollusca Gastropod Pulmonata 1 Arthropoda Crustacea Malacostraca Isopoda Oniscidea Trichoniscidae 17 Arthropoda Myriapoda Diplopoda Penicillata Polyxenida 2 Arthropoda Myriapoda Symphyla 1 Arthropoda Chelicerata Arachnida Acari Prostigmata Bdellidae 1 Arthropoda Chelicerata Arachnida Acari Prostigmata Trombidiidae 2 Arthropoda Chelicerata Arachnida Acari Prostigmata 1 Arthropoda Chelicerata Arachnida Acari Prostigmata Eupodina Dictynidae 1

75

Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 11 Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 4 Arthropoda Chelicerata Arachnida Acari 2 Arthropoda Chelicerata Arachnida Araneae 1 Arthropoda Chelicerata Arachnida Araneae Leptonetidae 1 Arthropoda Chelicerata Arachnida Pseudoscorpions 1 Arthropoda Hexapoda Entognatha Collembola Arthropleona Isotomidae 4 Arthropoda Hexapoda Entognatha Collembola Arthropleona Isotomidae 1 Arthropoda Hexapoda Entognatha Collembola Arthropleona Isotomidae 1 Arthropoda Hexapoda Entognatha Collembola Arthropleona Onychiuridae 2 Arthropoda Hexapoda Insecta Isoptera Hodotermitidae 1 Arthropoda Hexapoda Insecta Orthoptera Ensifera Gryllidae 1 Arthropoda Hexapoda Insecta Hemiptera 1 Arthropoda Hexapoda Insecta Hemiptera 1 Arthropoda Hexapoda Insecta Hemiptera Gymnocerata Aradidae 1 Arthropoda Hexapoda Insecta Hemiptera 3 Arthropoda Hexapoda Insecta Hemiptera Pentatomidae 1 Arthropoda Hexapoda Insecta Thysanoptera 1 Arthropoda Hexapoda Insecta Diptera Psychodidae 1 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 9 DRY EXTRACTION phylum subphylum class subclass order suborder family abundance Annelida Oligochaeta Haplotaxida 4 Arthropoda Crustacea Malacostraca Isopoda Oniscidea Trichoniscidae 1 Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 2 Arthropoda Chelicerata Arachnida Acari Prostigmata Bdellidae 1 Arthropoda Chelicerata Arachnida Araneae Labidognatha Dictynidae 1 Arthropoda Hexapoda Entognatha Collembola Isotomidae 1 Arthropoda Hexapoda Entognatha Collembola Isotomidae 1 Arthropoda Hexapoda Insecta Isoptera termitidae 2 Arthropoda Hexapoda Insecta Homoptera Auchenorrhyncha Cicadellidae 2 Arthropoda Hexapoda Insecta Homoptera Auchenorrhyncha Cicadellidae 3 Arthropoda Hexapoda Insecta Homoptera Auchenorrhyncha Cicadellidae 1 Arthropoda Hexapoda Insecta Homoptera Auchenorrhyncha Cicadellidae 1 Arthropoda Hexapoda Insecta Hemiptera Pentatomidae 1

Arthropoda Hexapoda Insecta Hempitera larvae Pentatomidae 1

Arthropoda Hexapoda Insecta Coleoptera Byrrhidae 1 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Scolytidae 11 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Staphylinidae 1 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Pselaphidae 1 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Scydmaenidae 1 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Leiodidae 1 Arthropoda Hexapoda Insecta Trichoptera 1 Arthropoda Hexapoda Insecta Trichoptera 1 Arthropoda Hexapoda Insecta Diptera 1

76

Arthropoda Hexapoda Insecta Diptera Nematocera Chironmidae 2 Arthropoda Hexapoda Insecta Diptera Nematocera Psychodidae 3 Arthropoda Hexapoda Insecta Diptera Nematocera Ceratopognidae 2 Arthropoda Hexapoda Insecta Diptera Nematocera Mycetophilidae 1 Arthropoda Hexapoda Insecta Diptera Nematocera Mycetophilidae 3 Arthropoda Hexapoda Insecta Diptera Culicidae 22 Arthropoda Hexapoda Insecta Diptera Culicidae 4 Arthropoda Hexapoda Insecta Diptera Culicidae 4 Arthropoda Hexapoda Insecta Hymenoptera Cynipidae 1 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 1 Arthropoda Hexapoda Insecta Hymenoptera Apocrita Cyncipoidea 1 Arthropoda Hexapoda Insecta Hymenoptera Symphata 1

Appendix 7: Identified invertebrates in bamboo areas of naturally regenerated forests

LEAF LITTER

phylum subphylum class subclass order suborder family abundance Annelida Oligochaeta Haplotaxida 1 Arthropoda Crustacea Malacostraca Isopoda Oniscidea Trichoniscidae 23 Arthropoda Myriapoda Diplopoda 1 Arthropoda Myriapoda Symphyla 3 Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 10 Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 5 Arthropoda Chelicerata Arachnida Acari Orbatida Orbatidae 13 Arthropoda Chelicerata Arachnida Acari 1 Arthropoda Chelicerata Arachnida Acari Prostigmata Bdellidae 1 Arthropoda Chelicerata Arachnida Acari Prostigmata 4 Arthropoda Chelicerata Arachnida Acari 1 Arthropoda Chelicerata Arachnida Opilliones 1 Arthropoda Chelicerata Arachnida Opilliones 1 Arthropoda Chelicerata Arachnida Opilliones 1 Arthropoda Chelicerata Arachnida Pseudoscorpiones 1 Arthropoda Chelicerata Arachnida Pseudoscorpiones 1 Arthropoda Hexapoda Entognatha Collembola Onychiuridae 1 Arthropoda Hexapoda Entognatha Collembola Onychiuridae 3 Arthropoda Hexapoda Entognatha Collembola Onychiuridae 1 Arthropoda Hexapoda Insecta Psocoptera Troctomorpho 5 Arthropoda Hexapoda Insecta Hempitera 2 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Scolytidae 3 Arthropoda Hexapoda Insecta Coleoptera Dermestidae 1 Arthropoda Hexapoda Insecta Coleoptera Curculionidae 4 Arthropoda Hexapoda Insecta Coleoptera Archostemata Carabidae 1 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 1 Arthropoda Hexapoda Insecta Diptera Culicidae 1 Arthropoda Hexapoda Insecta Diptera Nematocera Psychodidae 1 Arthropoda Hexapoda Insecta Diptera Nematocera Mycetophilidae 1 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 1 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 200+

77

DRY EXTRACTION

phylum subphylum class subclass order suborder family abundance Annelida Oligochaeta Haplotaxida 4 Arthropoda Crustacea Malacostraca Isopoda Oniscidea Trichoniscidae 1 Arthropoda Hexapoda Entognatha Diplura Japygidae 1 Arthropoda Hexapoda Entognatha Diplura Japygidae 1 Arthropoda Hexapoda Insecta Homoptera Auchenorrhyncha Cicadellidae 1 Arthropoda Hexapoda Insecta Hemiptera Miridae 2 Arthropoda Hexapoda Insecta Coleoptera Polyphaga Scolytidae 4 Arthropoda Hexapoda Insecta Coleoptera Staphylinidae 1 Arthropoda Hexapoda Insecta Trichoptera Hydroptilidae 1 Arthropoda Hexapoda Insecta Diptera Chironomidae 2 Arthropoda Hexapoda Insecta Diptera Simuliidae 1 Arthropoda Hexapoda Insecta Diptera 1 Arthropoda Hexapoda Insecta Diptera Psychodidae 2 Arthropoda Hexapoda Insecta Diptera Culicidae 2 Arthropoda Hexapoda Insecta Hymenoptera Formicidae 2