Changes in tree and caterpillar communities 

in secondary and primary forests along a climate 

gradient in the Southern Yucatán, Mexico 

 

August 2007 

 

MSc. Thesis            Tijl Anton Essens  

Wageningen University, Chairgroup Forest Ecology and 

Forest Management, the Netherlands. 

Supervision 

Prof. Dr. Frans Bongers, Wageningen University, the Netherlands;  

Dr. Carmen Pozo de la Tijera, Ecosur, Unidad Chetumal, Mexico; 

Dr. Ir. Henricus F.M. Vester, Ecosur, Unidad Chetumal  

and University of Amsterdam. 

Supervisors: Dr. Carmen Pozo de la Tijera (director of thesis) 

Associate researcher, Insect Ecology and Systematics, Ecosur, División, El Colegio de 

la Frontera Sur, Unidad Chetumal. Avenida Centenario km 5.5., Chetumal, Quintana 

Roo, CP 77014, Mexico, telephone 01(983) 835 0440 ext 4301 Fax: ext 240.  

Email: cpozo@ecosur‐qroo.mx

 

Prof. Dr. Frans Bongers 

Professor, Tropical Forest Ecology, Wageningen University, Lumen no.100, 

Droevendaalsesteeg 3, 6708 PB Wageningen, the Netherlands, telephone  0031 317 

478029.  

Email: Frans.Bongers@wur.nl

 

Dr. Ir. Henricus F.M. Vester  

Senior researcher, Tree and Forest architecture, Ecosur, División, El Colegio de la 

Frontera Sur, Unidad Chetumal. Avenida Centenario km 5.5., Chetumal, Quintana 

Roo, CP 77900, Mexico, telephone 01(983) 835 0440 and Invited Researcher, 

University of Amsterdam, Institute of Biodiversity and Ecosystem dynamics, 

Kruislaan 318, 1098 SM Amsterdam, the Netherlands. 

Email: hvester@ecosur.mx, hvester@science.uva.nl  

 

 

 

 

 

Author 

Tijl Anton Essens  

E‐mail: tijl.essens@gmail.com

Forest Ecology and Forest Management Chairgroup  

FEM 80436, AV2007_11, Wageningen University, the Netherlands. 

Index

Summary

1. Introduction……………………………………………………..….…...……1

1.1 Background………………………………………………...….………1

1.2 Vegetation of the Calakmul Reserve……………………..……..……..2

1.3 Disturbed habitats and community ecology …………………………..5

2. Hypotheses………………………………………………………..…..…..…..7

3. Materials and Methods…………….…………………………….…………..8

3.1 Vegetation sampling design………………………………………. ….8

3.2 Fauna sampling design………………………………………..….......10

3.3 Identification caterpillars……………………………………...…..….11

3.4 Statistical analyses…………………………………………..…..........12

Community composition……………………………………...……....12

Species diversity and richness……………………………..………....14

Nocturnal versus diurnal…………………………………..………….16

Equitability of caterpillar communities.………………………..…….17

Relationships between caterpillars and vegetation

structure and richness…………..……………………….………….…18

4. Results………………………………………………………………..………19

4.1 Vegetation……………………………….……………………..……..19

4.1.1 Community composition…………………...……….……...…..19

4.1.2 Diversity……………………….....……………….……...…....22

4.2 Caterpillars.…………………………………..…………….…….…...…..28

4.2.1 Community composition………………………………….….... 28

4.2.2 Diversity………………………...……………………….….......30

4.2.3 Nocturnal versus diurnal..…….……………………………...…32

4.2.4..Equitability of the caterpillar communities..…………….…… 33

4.2.5 Relationships between caterpillars and vegetation

structure and richness …………………………………………..……34

5. Discussion………………………………………………………………...….37

5.1 Changes of plant and caterpillar community structure on the gradient

of climate and forest age..…………………............................…………...37

5.2 Plant and caterpillar diversity on the gradient of climate…………….39

5.3 Does caterpillar diversity increase along a gradient of forest age?.....40

5.4 Does woody plant diversity increase along a gradient of forest age?.43

5.5 Relationships between caterpillars and vegetation

structure and richness...…….....……………………………………….... 45

6. General conclusions …………………………..……………..………..……46

Acknowledgements……………………………………………………………..…....47

References

Appendix

 

 

 

 

 

Changes in tree and caterpillar communities in

secondary and primary forests along a climate

gradient in the Southern Yucatán, Mexico

Tijl Anton Essens

Summary

The present study deals with the community composition and species richness of

caterpillars and woody plants in tropical primary and secondary forests along a

precipitation gradient in the Yucatán Peninsula, Mexico. This precipitation gradient

determines forest distribution that ranges from deciduous at the dry end of the

gradient, through an intermediate zone and to the humid semi-evergreen forests. To

investigate possible changes of caterpillar and woody plant communities on a gradient

of stand development, or grade of ‘forest disturbance’, we selected three forest age

classes in the ranges 5-10 years, 10-30 years and old-growth forests. We expected

significantly different community compositions in three climate zones (H1), and

significantly different community compositions in the three forest age-classes (H2). In

addition, the expected response of the two biotic communities is that they become

increasingly diverse from the dry to the humid climate zone (H3) and also from

disturbed to undisturbed habitats (H4). Moreover, the older forests were expected to

be more equitable compared to the young forests. Finally we tried to identify if there

is a relation between caterpillars with respect to the number of woody plant species of

different size classes and average canopy height of forest plots (H5).

Taking into account the species lists and species abundances, the composition of

caterpillar and vegetation communities across climate zones and forest age was tested

for significance with non-parametric ANOSIM test, analogue to the ANOVA. Results

showed that there is an overall significant dissimilarity among caterpillar communities

across climate zones (p=0.01) as expected (H1), and also in woody plant communities

among the three mature forest types of the climate zones (p=0.001). The change in

species composition of plant communities is also found along the gradient of forest

disturbance. Here, caterpillar communities among the three forest age-classes differ

significantly (p<0.01) as expected (H2), as well as plant communities (p=0.001).

Moreover, there is a tendency of an increasing dissimilarity of the caterpillar and tree

communities among the forest age-classes from the humid climate zone towards the

dry climate zone.

Sample-based rarefaction procedures were used to estimate the increase of the

caterpillar and plant species diversity with increasing sampling effort, while

individual-based rarefaction was applied to account for differences in density of

individuals in different habitats. The woody plant communities in the old growth

forests are sufficiently sampled to give a good indication of diversity across the

climate gradient. In contrast to the expected increase of species from the dry to the

wet climate zone (H3), the individual-based richness of woody plant species in the old

growth forests is highest in the medium climate zone. Individual-based richness of

woody plants in old-growth forest was indirectly tested for significance using the

Clench model. Despite the tendency that the forest in the medium climate appears

richer, this difference was not significant but is does contain the highest number of

unique species present of woody plant species. Similarly, the caterpillar communities

tend to be the most diverse in the medium climate zone and maintain the highest

number of unique species.

Furthermore, the results indicate that for woody plant species of ≥10 cm DBH, the

individual-based rarefied diversity and the Simpson index tend to increase with forest

age in all climate zones as expected (H4). The same patterns was found for woody

plant species 3≤ x 10 cm DBH in the medium climate zone, but the dry and the humid

climate zone tend to be more diverse in the forest patches of 10-30 years old. The

caterpillar communities show higher individual-based diversity and Simpson indices

in the secondary forests of 5-10 years old in the dry and the medium climate zone; in

the humid climate zone the secondary forests of 10-30 years appears most diverse.

Even so, significant differences between caterpillar diversity in forest age-classes

were only found in the dry climate zone, where the mature forest is significantly

different from the two secondary forests phases (H4). Similar to the results of woody

plant species, we found most unique caterpillar species in the medium climate zone.

Caterpillar log rank abundance rank plots did not reveal noticeable distinct

equitability for caterpillar communities on the gradient of forest age. The

hypothesized relation between caterpillars and the vegetation species richness and

structure cannot be not confirmed by ordination (CCA) (H5).

The results suggest that caterpillar and plant diversity are directly influenced by

climate and forest age causing a different species biotic compositions and abundances.

However, there is little statistical evidence showing a consistent response of

increasing or decreasing species diversity along the climate and the disturbance

gradient. One of the future challenges is to characterize meaningful groups within

caterpillar and woody plant communities in order to show effects of forest change

throughout various developmental stages upon caterpillar species composition and

diversity.

1. Introduction

1.1 Background

The South of the Mexican peninsula Yucatán holds large seasonal tropical forests.

The main protected area in this zone is the Calakmul Biosphere Reserve, located

towards the border of Guatemala (Figure 1). It extends over 723 185 ha and presents a

precipitation gradient creating pronounced differences in vegetation and

deciduousness, forming an ecocline (Vester et al. 2007). The Calakmul Reserve is part

of the Mesoamerican corridor, which connects several reserves in Belize and

Guatemala. Like many other tropical forests it contains numerous species and large

carbon stocks (Primack et al. 1988).

The Tropical Ecosystem Environment Observations by Satellites (TREES) project has

identified the Southern Yucatán Peninsular Region as a hot spot of tropical

deforestation. In the last three decades high rates of deforestation due to an increasing

population and changes in land use systems throughout the southern Yucatán, have

influenced the ecological functions and characteristics of the above mentioned

ecocline (Turner et al. 2001).

Together with deforestation by human activities, natural disturbances (i.e.; hurricanes,

fires) have substantially altered the forest structure in some areas of the Calakmul

region, leading to a mosaic of forest age-stands, including a variety of plant and

animal composition and diversity.

1

El Colegio de la Frontera Sur (ECOSUR) together with Clark University, Virginia

State University and Rutger University study the vulnerability of the coupled human

environmental system in the Southern Yucatán peninsular Region (SYPR), a NASA

funded program. One of the program’s projects investigates the effects of land use

change on biodiversity (vegetation and fauna). The present study contributes to this

project comparing species richness and composition in secondary and primary forests

on a climate gradient which governs forest types from deciduous to semi-evergreen.

The contrast mature – secondary forest is taken as representative for the increasing

deforestation and the humid - dry gradient is taken as model for a possible long-term

drying trend. The biotic groups used in this study are plants and Lepidoptera,

specifically caterpillars.

1.2 Vegetation of the Calakmul Reserve

The Calakmul Reserve contains a gradient of vegetation that correlates to a north-

south rainfall gradient of 900-1500 mmyr-1. The length and intensity of the dry season

determines largely the vegetation type and deciduousness (Figure 1). Three forest

types can be recognized in the study area (Flores and Espejel 1994, Martínez and

Galindo-Leal 2002, Pérez-Salicrup 2004). Firstly, tropical low semi-deciduous dry

forest (canopy height <15 m) is mainly found in the uplands of the drier northwest

part of the study region (Lawrence et al. 2004). It presents an annual precipitation

ranging from 900-1000 mm yr-1. More than 75% of the trees loose their foliage during

the dry season between April and May. The intensity of leaf loss and its southward

extent varies annually, apparently linked to the amount of precipitation received

2

during the dry season. The most common tree species for this forest are Thouinia

paucidentata, Beaucarnia pliabilis, Guaiacum sanctum, Lonchocarpus yucatanensis,

Bursera simaruba, Haematoxylum campechianum, Ceiba schotti, Pseudobombax

ellipticum and Maytenus schippi.

Figure 1. The Southern Yucatán Peninsular Region, with a precipitation gradient from the dry North to

the more humid Southern border of the Calakmul region. Circles represent the field stations in 1) low

stature semi-deciduous forest, 2) medium stature semi-evergreen and 3) high stature semi-evergreen.

The second forest type in the Calakmul Reserve is medium statured tropical semi-

evergreen forest, with canopy height between 15 and 25 m. It presents a shorter dry

season and less leaf loss (25-50% of the trees are deciduous). The annual precipitation

ranges from 1000-1200 mm yr-1. Tree species commonly found in these forests are

Brosimum alicastrum, Manilkara zapota, and Pouteria reticulata. The third forest

type is high tropical semi-evergreen (canopy height is generally >25 m). The

dominant tree species belong to the family Sapotaceae, the most common species are

Manilkara zapota, M. chicle, Pouteria sapota, P. amygdalina, P. campechiana and P.

reticulata (Galindo-Leal, 2001). This forest type presents a less characteristic dry

12

318000´ N

18015´ N

19000´ N

18045´ N

18030´ N

18015´ N

90015´ W 90000´ W 89045´ W 89030´ W 89015´ W

Dry forests Humid forests Deforested areas Reserve boundaries

3

season and leaf loss is related to particular species that (partially) loose their foliage

during the dry season. The annual precipitation ranges from 1200-1400 mm yr-1. High

semi-evergreen forests are located in the east and south of the study area (especially

near the Guatemala border).

Figure 2. Impression of a vegetation mosaic in

Calakmul with agricultural crops chilli and maize

(in the front), secondary forest patches (right

middle) and old growth forests (background).

Location of this photo: Dos Lagunas Sur,

Campeche, Mexico (T. Essens).

The most important economic activities in this region are agriculture and forestry,

which have lead to a mosaic of mature forests and secondary forest and agricultural

fields (Figure 2). The traditional ”slash and burn” agriculture is a common practice

and characterized by cutting patches of mature or secondary forest followed by

burning the vegetative remains. Afterwards the patches are cultivated with crops such

as maize, beans and squash. After one year of agricultural activities comprising two

harvests, the fields are abandoned. In the period that follows, regeneration of

vegetation establishes successional communities of vegetation.

This agricultural practice leads to a mosaic of young forest stands with predominantly

pioneer species and older stands that hold both pioneers and long living species. Also

forest structure differs in the forest patches (e.g. basal area and the distribution and

4

variability of canopy heights) and related abiotic factors (e.g. light intensity, soil

moisture, humidity). Because caterpillars often consume a limited number of plant

species we can expect the changing forest composition results in different caterpillar

composition. Therefore we hypothesize that anthropogenic impact described above in

different vegetation types results in different caterpillar (Lepidoptera) composition.

1.3 Disturbed habitats and community ecology

Different theories have been developed explaining species richness on gradients of

disturbances. It is generally accepted that disturbances cause a turnover in the species

composition, principally due to differences in times for establishment and degrees of

adaptation to a more complex ecosystem, and a decline of species richness for both

flora (e.g. Eggleton et al. 1998) and invertebrate fauna (e.g. Brown 1994, Roth et al.

1994, Schowalter and Ganio 1999, Holloway 1998, Floren and Linsenmair 2001).

Frequently cited hypotheses that formulate causal relationships between disturbance

and species diversity are the Intermediate Disturbance Hypothesis (Connell 1978, Fox

and Connell 1979, Sheil and Burslem 2003) and the Dynamic Equilibrium Model

(Huston 1979, 1994). Connell (1978) hypothesizes that intermediate levels of

disturbance will maximize diversity, while other studies indicated multiple equilibria

(Hanski et al. 2002), non linearity and threshold effects (Nyström et al. 2000).

Caterpillars were collected, in order to investigate their responses to changing woody

plant communities along a forest age gradient. It has been postulated that Lepidoptera

provide a suitable group to study the effect of habitat change because they are

responsive to changes in their habitat and therefore potentially good ecological

5

indicators (Eberhardt and Thomas 1991, Brown 1997, Thomas et al. 2001, Pozo

2006). Their sensitivity to changes is explained by their strong relation to their habitat

during their life cycle (Ehrlich and Raven 1965, Ehrlich 1984, Murphy and Wilcox

1986). Lepidoptera have a so-called holometabolic life, also called complete

metamorphism, which includes four life stages as an egg, larva, pupa and imago. The

dependence of the larval stages on certain host plants, and the dependence of flowers

for honey when adult (Gilbert 1980, Jennersten 1988, Rausher and Feeny 1990) are

relations typically affected by some types of habitat disturbances such as land use

change (Ehrlich et al. 1972, Murphy et al. 1990). In addition, the egg laying behaviour

of butterflies in their search for host plants seems to be directed towards optimal

conditions for offspring survival (Smiley 1978, Thompson and Pellmyr 1991, Bernays

and Chapman 1994, Santiago Lastra et al. 2006). Moreover, the distribution of many

butterfly species appears to be restricted by climatic conditions (Pollard 1979, Turner

et al. 1987, Dennis and Shreeve 1991) and so we might find strong effects of forest

perturbation in the various forest types.

6

2. Hypotheses

This research aims at contributing to our understanding of the dynamics of vegetation

and Lepidoptera communities along the climatic gradient from deciduous to semi

evergreen forests and the influence of anthropogenic disturbance on vegetation and

Lepidoptera communities. Three categories of forest age-classes were used to describe

the disturbance gradient, considering the youngest vegetation as “high disturbance”

and mature forest as “low disturbance”. The age is calculated from the moment fields

were abandoned by the farmers. The following hypotheses will contribute to respond

the questions how the structure and diversity of caterpillar and vegetation

communities changes along the gradients of forest age and climate and if caterpillar

diversity is related to vegetation diversity and structure. The response of biotic

communities to climate zones and forest age will be discussed in the light of existing

theories on these themes.

H1. Tree and caterpillar species composition are different in the climate zones.

H2. Tree and caterpillar species composition are different in the forest age-classes.

H3. Tree and caterpillar species diversity increase along the dry to wet gradient.

H4. Tree and caterpillar species diversity decrease with decreasing forest-age.

Furthermore, an exploratory analysis was done to investigate the link between

diversity and canopy height and on the other hand the diversity and composition of

caterpillar communities.

H5. There is a relation between caterpillar diversity and composition and tree

diversity vegetation structure.

7

Looking for caterpillars in the canopy

3. Materials and Methods

3.1 Vegetation sampling design

Fieldwork was done in the ejidos Conhuas, Zoh Laguna (and the bordering ejido

Nuevo Becal) and Dos Lagunas Sur that represent dry, medium and wet forests

respectively. Two sets of circular plots of 500 m2 were established. The first set is a

mixture of plots located along the climate and the disturbance gradient. For each

climate zone it comprises 12 plots in forest patches of three age groups: 4 plots in 5≤

x ≥10 year old vegetation; 4 plots in 10< x <30 year old vegetation and 4 plots in old

growth forest (in total 36 plots, Table 1). The second set is more at landscape scale

and each climate zone contains 32 mature forest plots (total 96 plots). The mature

forest plots used in the first set were randomly selected from the second set of three

times 32 plots in each old growth forest type.

For all plots, woody plants ≥10 cm DBH were identified, and diameter at breast

height (DBH) was measured as well as their total height (Figure 3). In the centre of

each plot a circular subplot of 100 m2 is located where all species 3≤ x <10 cm DBH

were identified, and their DBH and total height measured. During the field trips the

local guide and a botanist identified the plants. Samples of the plants were collected

using a botanical press for their posterior identification in the herbarium of Ecosur and

Merida (CIC).

8

Table 1. The time spend on searching caterpillars and the amount of surface area covered. In the first

column, three forest types distinguished in the Calakmul area are shown. Within each of the three forest

types, the 12 circular plots of 500m2 are selected in three forest age-classes. The surface area covered

for searching caterpillars, amount to a total of 2000 m2 / forest age-class. Survey hours are calculated

for five fieldtrips (5 fieldtrips*4 plots *4 people*1 survey hour). The surface area covered for searching

caterpillars, amount to a total of 2000 m2 / forest age-class.

forest types Plots on a gradient of disturbance Surface area (total of plots)

Survey hours

80 80 80

Tropical low semi- deciduous dry forest (dry)

4 plots in mature forest 4 plots old secondary forest (10< x <30 yr) 4 plots young secondary forest (5≤ x ≥10 yr)

2000 m2 2000 m2 2000 m2

Σ 160

80 80 80

Medium tropical semi-evergreen forest (medium)

4 plots in mature forest 4 plots old secondary forest (10< x <30 yr ) 4 plots young secondary forest (5≤ x ≥10 yr )

2000 m2 2000 m2 2000 m2

Σ 160

80 80 80 High tropical semi-

evergreen forest (humid)

4 plots in mature forest 4 plots old secondary forest (10< x <30 yr ) 4 plots young secondary forest (5≤ x ≥10 yr )

2000 m2 2000 m2 2000 m2

Σ 160 Sum 36 plots of 500m2 18000 m2 Σ 480

W

Figure 3. Plot design for

m2. On the right, an imp

higher strata.

N

2 m

500 m2

100 m2

E

S

vegetation and caterpillars. On the left the two circular plots of 500 and 100

ression of a 500 m2 plot scanned for caterpillars at the forest floor and the

9

3.2 Fauna sampling design

Pozo et al. (2003) found that different butterfly families in the Calakmul region occur

in different periods of the year. Therefore, it was decided that sampling of caterpillars

should cover all seasons. The sampling took place in the set of 36 plots (table 1). Two

fieldtrips were done in the wet season (July-November), two fieldtrips in the season

called ´´Nortes´´ (December-January), a period of cold northern winds and finally one

sampling in the dry season (February – June).

Furthermore, Lepidoptera are found in all strata of the forest (Schulze et al. 2001) and

we assumed that caterpillars consume leaves from shrubs as well from full grown

trees, thus, caterpillars and pupae were collected in a horizontal and a vertical

component in the thirty-six plots. Each plot was surveyed by four people for 1 hour.

Two people spent 1 hour in effective searching for caterpillars in the under storey up

to 2 meters (horizontal component) and two other people searched for 1 hour in the

higher region of the forest (vertical component, fig. 3). Professional climbing gear and

a ladder were used to reach a height >2 meters. Pilot investigations revealed that on

average 4 big trees (>20 meters in height) and about 5 smaller trees (<15 meters in

height) were scanned in old growth forest during this time.

During survey time, caterpillars were handpicked, photographed and inserted into 5

ml vials filled with 98% alcohol. Unfortunately, this is presently the only way to keep

the damage to soft-tissued insects to a minimum. The stored caterpillars were used for

visual identification based on morphological characteristics using a microscope.

10

3.3 Identification of caterpillars

In Calakmul 430 species of diurnal Lepidoptera (Rhopalocera) were found (Pozo et al.

2003, Martinez et al 2005). For moths the number of species may be up to 7 times as

large (Heppner 1991). The species diversity of Lepidotera in general requires a high

level of experience for identifying species. Therefore it was necessary that larvae were

collected in the field and identified on a later moment. All individuals are deposited in

the reference collection (INE number QNR.IN.018.0497) at Ecosur, Chetumal.

Identification took place based on morphological characteristics using identification

keys and photos (e.g. Peterson 1962). The photo is an important tool for identification

as colours and shapes tend to change in alcohol. Moreover, the photos were matched

with photos of species that occur in Guanacaste, Costa Rica

(http://janzen.sas.upenn.edu/Wadults/searchcat2.lasso). For this work, all individuals

have been identified at family level, and species were separated in morpho-species.

11

Gonodonta ssp. Noctuidae

3.4 Statistical analyses

Community composition

The assumptions of homogeneity in the variance and normal distribution of the

abundance data of vegetation and caterpillars in each treatment were tested with the

Kolmogorov-Smirnov test using SPSS for Windows (2006). Deviations of normality

and of homogeneity in variance were in all cases significant. Therefore, further

analyses for community composition were done with non-parametric tests. The

ANOSIM test, an analogue to the ANOVA, was employed to test for significance in

the composition of caterpillar and vegetation communities among climate zones and

forest age using PRIMER software (Clarke and Gorley 2006). The ANOSIM results

in dissimilarity values in community composition of compared groups. Importantly,

the dissimilarity is calculated using the overlap of species lists and the species

abundances, which makes this test appropriate for comparative community

composition assessments.

Before ANOSIM was executed, transformations were applied to the abundance data

in order to limit the effect of zeros and the contrast of extreme high and low

abundance values. The abundance data of woody vegetation in the 36 plots were root

transformed. The abundance data of woody vegetation in the sample set of 96 plots in

old growth forest were log transformed. The caterpillar abundance data were 4th root

transformed. Then, Bray-Curtis resemblance matrices for woody vegetation and

caterpillars were generated (Bray and Curtis 1957) and used for differences in

community composition between the climate groups and forest age-classes. The Bray-

12

Curtis resemblance values show plot to plot combinations, and creates e.g. a 36 by 36

matrix (of which half contains values). Furthermore, the Bray-Curtis matrix expresses

resemblance values varying between 0 and 1, where 0 is no resemblance and 1 means

complete resemblance. Every value reflects the resemblance of one plot with another

considering the species list and their abundances. For the ANOSIM test, plots were

grouped based on categories of forest age and climate and the groups were tested

against each other for significance.

As mentioned, ANOSIM is an approximate analogue of the standard univariate 1- and

2-way ANOVA (analysis of variance) tests. I executed two types of tests with

ANOSIM. The first test is two-way nested, a hierarchical design which tests for the

differences between climate groups using age groups per climate as samples. This test

gives three main outcomes. The first outcome gives a global RANOSIM for the overall

difference between age groups across all climate groups, the second is a global

RANOSIM, which expresses the overall difference between climate groups using age

groups as samples. Finally, the program provides a table with pairwise differences

between climate groups. The two-way nested option was selected for the vegetation

and the caterpillar data set of 36 plots according to the hierarchical lay-out of the

sampling design, where climate is the first level and forest age-class is the second.

The second type of test executed with ANOSIM, is the one-way pairwise test which

analyses differences of vegetation and caterpillar assemblages found in the forest age-

classes for each climate zone independently. Also this test gives two types of

outcomes; the first is a global RANOSIM for the overall difference between age groups

within a climate group, the second is a table with post hoc pairwise tests showing

13

specifically the dissimilarities between the communities found in the forest age-

classes. This test was applied to vegetation and caterpillar data for the 36 plots and for

the vegetation data in the mature forest using information from the 96 plots.

The RANOSIM values can vary between -1 and 1. A value of 0 or lower means that there

are no differences or relation between sample caterpillar communities, while 1 means

that the samples of the groups are very distinct. In addition, the Bray-Curtis

resemblance matrices were used to create MDS plots (multi-dimensional scaling). The

plots show caterpillar and vegetation communities of each plot in ordinate space. The

interpretation of the MDS is in accordance with the resemblance matrix: sample sites

that are close together represent more similarity than points that are far from each

other.

Species diversity and richness

Bar graphs were drawn showing total numbers of plant and caterpillar species in each

habitat. Caterpillars and woody plant species in old growth forests were further

investigated for the number of shared and unique species.

In order to see if there are patterns or tendencies of floral and faunal diversity,

individual-based and sample-based accumulation graphs were generated.

Accumulation curves are recommended by Soberon y Llorente (1993) with the aim to

compare species lists in different environments. An advantage of its application is to

use them as predictive tools for biodiversity studies (Soberon y Llorente 1993). For

14

the calculations on expected species number, number of samples and individuals, the

software EstimateS has been used (Colwell 2004). Sample-based and individual-based

rarefaction curves were drawn for plant and caterpillar data. First, sample-based

accumulation curves were drawn to investigate how caterpillar and plant species

richness (MaoTau = expected species) changed over increasing sampling area. The

curves give the opportunity to evaluate the expected species richness for samples

simulating samples drawn at random from the pooled samples for all plots (Colwell

2004). Theoretically the asymptote of the curve is the maximum number of species in

the area. The steepness of the curve indicates whether plot to plot variation is high or

low.

Individual-based rarefaction was developed as a measure of species richness by

rarefaction to overcome sample size and species density effects, allowing the

comparison between communities where, for example, densities of animals and plants

are very different. The accumulation curves plots the number of expected species and

numbers of individuals (Gotelli and Colwell 2001) to investigate the levels of richness

between old-growth forest types and forest age-classes within a climate zone.

Individual-based accumulation curves were generated for the woody plant species in

the sample set of 96 old-growth forest plots. This procedure was repeated for

caterpillars in the different forest age-classes and for woody vegetation in the 36 plots.

The larvae of Lepidoptera have not systematically been studied in Mexico and

therefore not much is known about the study object. Moreover, the study area is

heterogeneous with many rare species (Pozo et al. 2003). In a similar situation,

Soberon y Llorente (1993) advocated that the Clench model should be used to obtain

15

asymptotes (Clench 1979). We used the Clench model on the individual-based

richness along the gradient of forest disturbance. Various comparable points were

interpolated on the Clench curves (Colwell et al. 2004) which, in turn, were used to

apply a t-test (Magurran 2004). In this way individual based rarefied diversity that

accounting for sample size and density effects can be tested indirectly. The Clench

model and t-tests were restricted to the data of caterpillars along the gradient of forest

disturbance and plants in old growth forest types to compare the biodiversity. An

interval of 50 individuals was applied on Clench curves, representing the number of

species along increasing number of individuals, to obtain the number of species that

were used to execute the t-test. For small diameter plants in the 500m2 plots, the level

on which the number of individuals (800 individuals) could be compared was much

higher for the 100m2 plots (500 individuals) for the trees with a larger diameter.

Therefore, to run the t-test for plants, 3≤ x <10 cm DBH (500m2 plots) 10 points were

interpolated, while for the plants ≥10 cm DBH (100m2 plots) 16 points were obtained.

For caterpillars 10 points were used. Though the t-test is our main interest, the curves

of the Clench model also allow to make tentative predictions about the dimensions of

plants and caterpillar communities by extrapolating the curve to obtain the asymptote.

Lande et al. (2000) advocate the use of the Simpson index as an unbiased measure of

diversity for smaller sets of non-parametric samples. The Simpson index is a

dominance measure that weighs towards abundances of the commonest species rather

than providing a measure of the species richness. It calculates the probability of two

individuals randomly drawn belonging to different species (Magurran 1988, 2004).

Accordingly, calculations were performed using the same methods.

16

Nocturnal versus diurnal

The response of Lepidoptera to forest disturbance is further explored by comparing

the relative differences of moth and butterfly populations throughout in the selected

habitats. We calculated the relative proportion of butterfly and moth larvae in relation

to the total number of caterpillars.

Equitability of the caterpillar communities

Rank order abundance plots, sometimes referred to as dominance-diversity curves,

were used as a technique to describe the equitability of caterpillar communities across

the different forest age-classes. The rank order abundance curves simply plots log

number of individuals of each species against the rank in a series of most to less

abundant. The equitability of a caterpillar assemblage can be interpreted from reading

the steepness of the line in the first 5 ranks and the length of the tail. The prototype

disturbed habitats have relatively few ranked species and the first ranked present very

high abundances. This causes a short and steep curve. Undisturbed habitats have some

abundant species, but not as numerous as disturbed habitats. In large samples from

high diverse habitats, there tends to be a long tail of rare species (between one and 5

individuals only).

17

Relationships between caterpillars and vegetation structure and richness

Multivariate analyses were executed with the intention of examining the relationship

between caterpillar communities and the structure and richness of forest plots, with

the software CANOCO (Ter Braak and Van Tongeren 1995). First, a detrended

correspondence analysis (DCA) was executed to examine which ordination technique

was most appropriate. According to the result of the DCA, the Canonical

Correspondence Analysis (CCA) was applied (Lepš and Šmilauer 2003). The

variables incorporated for forest structure are canopy height, and for diversity (1) the

number of woody plant species 3≥ x <10 cm DBH, (2) the number of woody plant

species ≥10 cm DBH, (3) the number of woody plant species ≥3 cm DBH and finally

forest age of the plot. Each of these variables will be evaluated for a possible relation

between vegetation and caterpillar assemblages.

18

4. Results

4.1 Vegetation

4.1.1 Community composition

Differences in woody plant communities in three Old-growth forest types

The data used for these results are based on the 96 plots in old growth forest types

located in the three climate zones. In total 182 species of woody plants ≥10 cm DBH

were identified (1499 individuals). Among plants 3≤ x <10 cm DBH, 148 species

were identified (1762 individuals). Together, the 96 old-growth forest plots contain

232 woody plant species ≥3 cm DBH.

The sampled vegetation in both size classes shows significant (p = 0.001) and strong

overall dissimilarities among the three old-growth forest types (Table 2). The

difference between the plant composition of the medium and the humid forest is

smallest, while the other combinations are more dissimilar.

Diameter classes 3≤ x <10 cm DBH ≥10 cm DBH

Statistics RANOSIM p RANOSIM p Overall dissimilarity 0.723 0.001 0.735 0.001

Dry / Medium 0.766 0.001 0.795 0.001

Dry / Humid 0.871 0.001 0.777 0.001

Medium / Humid 0.541 0.001 0.662 0.001

Table 2. Dissimilarity (RANOSIM) statistics and significance level for plants of two size classes in the three

mature forest types

19

The MDS plots for mature forests illustrate that the samples which belong to the same

forest type are clustered rather well. Only some samples are dissimilar from the

cluster to which they are associated. In the 3≤ x <10 cm DBH size class the cluster is

more separated (Figure 4) though not expressed clearly in the statistics (table 2).

Figure 4. The sample set of 96 plots in mature forests for trees

≥10 cm DBH (left) and plants 3≤ x <10 cm DBH (right) in MDS

based on the Bray-Curtis similarity matrix.

Differences in Plant Communities across Climate and Forest Age

The data used for these results are based on the 36 plots located in the three age-

classes along the climate gradient. In total 130 species of plants ≥10 cm DBH were

identified (818 individuals) and 132 species of plants 3≤ x <10 cm DBH (792

individuals).

Overall dissimilarity of communities among climate groups in this dataset was not

significant, for woody plants ≥10 cm DBH (RANOSIM = 0.284, p = 0.089) nor for woody

plant communities 3≤ x <10 cm DBH (RANOSIM = 0.218, p = 0.175). The largest

difference is, as expected, between the dry and the humid climate group (≥10 cm

DBH RANOSIM = 0.63, p = 0.1 and 3≤ x <10 cm DBH RANOSIM = 0.444, p = 0.1).

20

Differences in plant communities in the gradient of forest disturbance

Overall variation in woody plant communities among the three forest age-classes

across the climate zones shows significant dissimilarities. They were slightly larger in

plant communities 3≤ x <10 cm DBH (RANOSIM = 0.627, p = 0.001) than for plants ≥10

cm DBH (RANOSIM = 0.547, p = 0.001).

Considering each climate zone as an independent unit to test the variation among

forest age-classes, larger dissimilarities of plant communities ≥10 cm DBH were

found in the dry climate zone and the humid climate zone (Table 3). The two most

dissimilar plant communities in the size class 3≤ x <10 cm DBH were found in the

medium climate zone and the dry climate zone respectively.

Table 3. Dissimilarity (RANOSIM) and significance levels for woody plant communities among and

between forest age-classes in each climate zone. Old-growth forest is abbreviated with mf (mature

forest).

Moreover, the results indicate that the woody plant communities of the two secondary

forests are more similar to each other than to the mature forests. There is one

exception; the secondary humid forests in the size class 3≤ x <10 cm DBH (RANOSIM =

0.49, p = 0.029) is more dissimilar than the combination of old secondary and mature

forest (RANOSIM = 0.46, p = 0.057).

Climate zones Dry Medium Humid

Diameter class 3≤ x <10 cm ≥10 cm 3≤ x <10 cm ≥10 cm 3≤ x <10 cm ≥10 cm

Statistics RANOSIM p RANOSIM p RANOSIM P RANOSIM p RANOSIM p RANOSIM p

Overall dissimilarity 0.67 0.001 0.656 0.001 0.752 0.001 0.436 0.001 0.595 0.001 0.55 0.001

5-10 yrs /10-30 yrs 0.458 0.029 0.422 0.029 0.320 0.057 0.25 0.057 0.49 0.029 0.323 0.667

5-10 yrs/mf 0.807 0.029 0.885 0.029 0.99 0.029 0.625 0.029 0.734 0.029 0.667 0.029

10-30 yrs/mf 0.74 0.029 0.542 0.029 1 0.029 0.49 0.029 0.464 0.057 0.771 0.029

21

4.1.2 Diversity

Diversity and distribution of vegetation in old-growth forests over a climate gradient

In the sample set of 96 plots in mature forests, the number of woody plant species in

both size classes was highest in medium stature semi-evergreen forest, located in the

medium climate zone.

0

20

40

60

80

100

120

Dry Medium Wet0

20

40

60

80

100

Dry Medium Humid

Num

ber o

f pla

nt s

peci

es 3

-10

cm D

BH

Num

ber o

f pla

nt s

peci

es >

10 c

m D

BH1

Shared species Medium and Humid

Shared species Dry and Humid

Shared species Dry and Medium

Uniques

Shared species all localities

4147

3330

24

42

83

101

7866

83

64

Figure 5. Total number of plant species ≥3 x <10 cm (left) and ≥10 cm DBH (right) in old-growth

forests of the Calakmul region.

The medium climate zone shows clearly the largest number of shared species >10 cm

DBH. For species 3≥ x <10 cm DBH the humid and the medium climate zone contain

the same number. Besides, the medium climate zone harbours most unique species

(approximately 50% of all plant species 3≥ x <10 cm DBH and 40% of plant species

≥10 cm DBH). Moreover, dry forests contained the smallest number of species

(Figure 5).

The woody plants in 96 plots in old-growth forest on the climate gradient were further

examined with sample-based and individual-based rarefaction curves (Figures 6 and

7). The sample-based curves have the advantage that they give an indication if enough

samples have been taken to characterize a community (Figure 6.1 and 6.2).

22

Number of individuals

0 200 400 600 800 1000 1200 1400

Num

ber o

f pla

nt s

peci

es >

10 c

m D

BH

0

20

40

60

80

100

120

Table 4. Species number based on sample-based (Sb) (figure 6.1 & 6.2) and individual- based (Ib)

rarefaction curves (7.1 & 7.2), and their standard deviation (SD).

Although the curves do not reach the asymptote, they have leveled to a reasonable

level to obtain an acceptable impression of which is the richer old-growth forest. The

Figures 6. Sampled-based rarefaction curves for plants 3≤ x <10 cm DBH in samples of 100 m2 (figure 6.1)

and plants ≥10 cm DBH in samples of 500 m2 (figure 6.2) in the mature forests of the Calakmul region.

Figures 7. Individual-based rarefaction curves (figure 7.1) for

plants 3≤ x <10 cm DBH and plants ≥10 cm DBH

(figure 7.2) in mature forests.

Climate zones Dry Medium Humid Statistics Sb SD Ib SD Sb SD Ib SD Sb SD Ib SD 3≤ x <10 cm 64 2.88 60 2.79 83 4.53 76 4.27 66 5.5 66 5.5 ≥10 cm 78 4.67 77 4.63 101 4.87 88 4.51 83 4.87 83 4.87

Number of samples

0 5 10 15 20 25 30 35

Num

ber

spe

0 cm

DBH

cies

3-1

of p

lant

0

20

40

60

80

100

Number of samples

0 5 10 15 20 25 30 35

Num

ber o

f pla

nt s

peci

es >

10 c

m D

BH

0

20

40

60

80

100

120

Fig. 6.1 Fig. 6.2

Number of individuals

0 100 200 300 400 500 600 700

Num

ber o

f pla

nt s

peci

es 3

-10

cm D

BH

0

20

40

60

80

100

Fig. 7.2 Fig. 7.1

High stature semi-evergreen forestMedium stature semi-evergreen forestLow stature semi-deciduous forest

Num

ber o

f pla

nt s

peci

es >

10 c

m D

BH

23

curves suggest that with increasing sample effort the medium semi-evergreen forest in

the medium climate zone is the richest and the dry one is the least rich, although the

difference between dry and humid forest is not certain given the variation.

In the individual-based rarefaction curves, the number of woody plant species that

occur in high stature semi-evergreen forest represents the number density on which

richness can be compared across the three forest types (Figure 7.1 and 7.2). The

results are similar to the sample-based rarefaction, showing that the medium stature

semi-evergreen forest seems to maintain more species than the two other forest types.

The plotted Clench models illustrate that the plant communities of the old growth

forests seem sufficiently sampled to make a meaningful comparison. The richer forest

is again the medium climate zone (Figure 8). However, the t-test executed with the

Clench values did not show significant differences between plant diversity of the

forest types and we ought to assume that the groups have equal variances.

0

20

40

60

80

100

120

0 800 1600 2400 3200 4000 4800 5600

No. individuals

No.

spe

cies

High stature semi-evergreen forestMedium stature semi-evergreen forestLow stature semi-deciduous forest

Figure 8. Accumulation curves based on the individual-based richness of woody plant species ≥10 cm

DBH, calculated according to the Clench model. The vertical line indicates the maximum number of

individuals to which points had been interpolated, meaning that the sampling effort and the number of

expected species reached until this point; beyond this line are extrapolations using the Clench model.

24

Plant species richness on the gradient of forest disturbance

Plant richness derived from sample- and individual-based rarefaction methods reflect

similar patterns, however, high values for sample based diversity are generally

suppressed compared to individual-based rarefaction (Table 5 and see Appendix 1, 2,

3, 5 for the curves). Individual-based plant richness in the smaller diameter class (3≤ x

<10 cm DBH) in the medium climate zone shows an increase of plant species along a

gradient of forest age. In the two other climate zones woody plant richness of the

young and the mature forests are equal or similar.

Climate zones Dry Medium Humid Statistics Sb SD Ib SD Sb SD Ib SD Sb SD Ib SD 3≤ x <10 cm DBH Forest 5-10 yrs 21 2.52 21 2.52 17 3.6 14 3.00 21 3.74 17 3.12 Forest 10-30 yrs 30 4.66 21 3.72 21 3.11 21 3.11 37 3.58 25 2.70 Mature forest 20 2.49 18 2.3 35 5.32 35 5.32 22 3.39 21 3.74 ≥10 cm DBH Forest 5-10 yrs 9 2.39 9 2.00 nd nd nd nd 12 3.02 12 3.02 Forest 10-30 yrs 36 5.02 12 1.84 15 2.4 15 2.4 19 2.81 5 1.46 Mature forest 31 2.94 16 2.37 49 4.78 29 2.88 38 4.16 10 1.90 ≥3 cm DBH Forest 5-10 yrs 28 3.02 28 3.02 17 3.12 17 3.12 29 4.69 29 4.69 Forest 10-30 yrs 54 5.96 28 3.50 30 3.99 25 3.50 47 3.81 30 2.73 Mature forest 39 3.84 28 3.03 68 5.72 45 4.14 49 4.32 35 3.50

Table 5. Rarefied plant species richness for different size classes in 36 plots in different age-classes

based on sample-based (Sb) and individual- based (Ib) rarefaction curves and their standard deviation

(SD). Nd means no data: EstimateS was not able to calculate richness parameters in the four young

secondary forest plots of the medium climate zone; here, only two tree species ≥10 cm DBH were

found. Note that rarefaction does not account for across climate comparison; these data are exclusively

for across forest age comparison within one climate zone!

For plants with a diameter ≥10 cm DBH in the dry and the medium climate, there is

an increase of individual-based rarefied richness with forest age. In the humid zone,

25

the sample-based richness does shows an increase in species with age, but not when

rarefied with the number of individuals (Table 5).

Individual-based plant richness considering both size classes (plants ≥3 cm DBH)

shows an increase of species with age in both the medium and the humid climate

zone. In the dry climate, plant species numbers are equal in both young and old forest

types (Appendix 4).

The results of the Simpson index shows that in particular the forests in the medium

climate zone show that the chance that two random woody plant species individuals

drawn from the population in mature forests is much higher compared to the

secondary forests (Table 6 and Appendix 6).

Table 6. Simpson estimators in forest age-classes in each climate zone. The calculation could not be

executed for plants ≥10 cm DBH in the wet climate zone, because one plot did not contain plants in this

size class.

Contrastingly, index values for woody plant species 3≤ x 10 cm DBH in the dry as in

the humid climate zone and the index values for plants ≥3 cm DBH in the dry climate

Climate zones Dry Medium Humid 3≤ x <10 cm DBH Forest 5-10 yrs 10.65 4.04 6.33 Forest 10-30 yrs 20.94 6.74 14.00 Mature forest 14.02 14.06 12.95 ≥10 cm DBH Forest 5-10 yrs 4.23 5.64 Nd Forest 10-30 yrs 19.57 5.64 5.17 Mature forest 21.53 18.78 14.23 ≥3 cm DBH Forest 5-10 yrs 14.21 4.64 8.00 Forest 10-30 yrs 27.19 8.37 8.57 Mature forest 23.49 24.42 20.58

26

zone present show higher species diversity in the old secondary forest compared to the

mature forest.

The vegetation communities in the forest mosaic

In order to compare the richness of the whole forest mosaic in each climate, species of

all forest age-classes were brought together (Figure 9). The medium forest samples

share the largest proportions of species found occurring in both dry and humid forests

and harbours in both size classes the lowest number of unique species.

Figure 9. Distribution of plant species 3≤ x <10 cm DBH and ≥10 cm DBH in 36 plots, with total species

number lumped for all forest age-classes.

0

20

40

60

80

100

Dry Medium Humid

Num

ber o

f spe

cies

plan

ts >

10 c

m D

BH

Shared species Medium and HumidShared species Dry and HumidShared species Dry and MediumUniquesShared species for all localities 0

20

40

60

80

Dry Medium Humid

Num

ber o

f spe

cies

plan

ts 3

-10

cm D

BH

27

Automeris spp. Saturniidae

4.2 Caterpillars

4.2.1 Community composition

Variation of Caterpillar Communities across Climate zones

From the 4200 individuals we identified 341 morpho-species, distributed over 5

butterfly families and 23 moth families. There is an overall significant variation

among caterpillar communities in the climate zones (global RANOSIM = 0.391, p = 0.011),

but there are no significant differences of paired climate groups, e.g. the largest

dissimilarity between communities was found in the combination of the dry and the

humid climate group (RANOSIM = 0.667, p = 0.1).

Differences between Caterpillar Communities across Forest age-classes

The caterpillar composition of three forest age-classes differ significantly across all

climate zones (RANOSIM = 0.449, P = 0.001). For each climate zone separately, significant

overall dissimilarities of caterpillar assemblages in the forest age-classes, were found

in the dry climate zone (RANOSIM = 0.495, p = 0.001), in the medium climate zone

(global RANOSIM = 0.266, p = 0.006) and in the humid climate zone (global RANOSIM =

0.128, p = 0.003) (Table 7). It shows that the dissimilarity of caterpillar assemblages

in the dry climate zone is most dissimilar and decreases along the dry to wet gradient.

For each climate zone separately, the most distinct caterpillar communities on the

gradient of forest age can be found between secondary forests (both young and old

secondary) and mature forests.

28

The caterpillar assemblages in ordinate space show that caterpillar community in the

dry climate zone is most dissimilar from the humid forests and the caterpillar

communities of the medium climate zone takes an intermediate position between

those two (Figure 10.1 and 10.2). In addition, the plots located in humid forest are

much more dispersed compared to the much more consistent cloud of samples located

in the dry forests (Figure 10.2).

Table 7. RANOSIM and significance levels for overall and pairwise differences in caterpillar communities.

Climate zones Dry Medium Humid

Statistics RANOSIM P RANOSIM P RANOSIM P

Overall dissimilarity 0.459 0.001 0.266 0.006 0.128 0.003

Forest 5-10 yrs / forest 10-30 yrs 0.51 0.029 0.135 0.171 0.011 0.373

Forest 5-10 yrs / mature forest 0.833 0.029 0.24 0.114 0.294 0.002

Forest 10-30 yrs / mature forest 0.198 0.114 0.469 0.029 0.083 0.082

Fig. 10.2

Fig. 10.1

Figure 10. Caterpillar assemblages of plots in ordinate

space (figure 9.1).On the right plots with the same

samples and equal projection as figure 9.1. The

difference consist of standardized symbols and grey

tones of sampled assemblages for the climate gradient

(figure 9.2) and forest age (figure 9.3).

Fig. 10.3

29

4.2.2 Diversity

Differences in Climate and Age class groups

Dry forestsShared species 10-30 and mfShared species 5-10 and mfShared species 5-10 and 10-30UniquesShared species all age classes

Most caterpillar species were found in the old secondary forests of the medium

climate zone. There is little difference between the absolute number of caterpillar

species among the mature forest types (Figure 11), but large differences age class 10-

30 years.

0

20

40

60

80

5_10 10_30 mf

Dry forests

Num

ber o

f spe

cies

0

20

40

60

80

5_10 10_30 mf

Medium forests

0

20

40

60

80

5_10 10_30 mf

Humid forests Figure 11. Absolute number of species, unique to an age class

within a climate group or shared with another age class in the

same climate group. Number above bars indicates total number

of species. Whiskers present the standard deviation for sampled

caterpillars in each forest age class group. Old-growth forest is

abbreviated with mf (mature forest).

Rarefied species richness of caterpillar communities does not express straightforward

patterns of how diversity changes along the forest-age gradient (Table 8 and

Appendix 5). The individual-based richness of the young secondary forest and the

mature forest are rather similar throughout climate zones. However, the old secondary

forest presents an extraordinary increase in species diversity from the dry to the humid

climate zone.

30

Table 8. Rarefied sample-based (Sb) and individual-based (Ib) species diversity. Note that rarefaction

does not account for across climate comparison; these data are exclusively for across forest age

comparison within one climate zone!

Significant differences between the caterpillar richness in the different forest age-

classes were only found in the dry climate zone; the dry young secondary forest and

old secondary forest (p=0.001) as well as the old secondary forest and the mature

forest (p<0.001).

The Simpson index shows similar patterns as found for individual based rarefaction

(Table 9 and Appendix 6). Also here, the chance that two random caterpillar species

drawn from the young or old secondary forests belong to different species is larger

than in mature forests.

Climate zones Dry Medium Humid Statistics Mean Mean Mean 5-10 yrs 6,11 11,25 8,99 10-30 yrs 2,45 7,17 16,92 mature forest 5,91 8,64 8,6

Table 9. Simpson index values in forest age-classes for each climate zone.

Climate zones Dry Medium Humid Statistics Sb SD Ib SD Sb SD Ib SD Sb SD Ib SD 5-10 yrs 54 3.80 45 3.48 63 3.66 61 3.66 48 4.52 32 3.33 10-30 yrs 25 3.98 17 2.84 73 5.79 48 4.26 51 4.02 51 4.02 mature forest 45 3.66 45 3.66 58 4.29 58 4.29 59 3.96 38 3.17

31

The caterpillar communities in the forest mosaic

The highest total caterpillar species number was found in the medium climate zone,

followed by the humid and dry climate zone respectively (Figure 12). Furthermore,

the medium climate zone shares the largest proportions of species found in both dry

and humid forests. With respect to unique species to a climate group, the humid zone

contains three species more than the medium forests.

020406080

100120140160180

Dry Medium Humid

Precipitation range (sites)

Num

ber o

f spe

cies

Figure 12. Absolute number of species, unique to a locality or shared with another. Number above bars

indicates total number of species.

Shared species Medium and HumidShared species Dry and HumidShared species Dry and MediumUniquesShared species for all localities

154 162

121

4.2.3 Nocturnal versus diurnal

The relative contribution of the occurrence and abundance of caterpillars that belong

to diurnal butterfly species showed that this group is rather small compared to the

nocturnal species throughout the sampled habitats (Table 10). The occurrence of

diurnal species is less than 10% and their abundances vary between 4 and 11%

compared to nocturnal species. Moreover, the abundance and the occurrence of

diurnal species are rather constant throughout the forest ages, which suggest that the

32

proportion of diurnal against nocturnal butterfly larvae does not change substantially

throughout the various habitats.

Climate zones Dry Medium Humid Statistics Occ (%) Ab (%) Occ (%) Ab (%) Occ (%) Ab (%) 5-10 yrs 6,1 8,8 4,7 5,9 7,1 10,7 10-30 yrs 6,0 8,6 4,0 6,1 7,8 10,7 mf 6,3 8,3 7,2 9,0 4,0 4,3

Table 10. Relative proportion of the number of diurnal species (Occ) and the relative abundances of

diurnal species (Ab) in relation to the total number and abundance of all species.

4.2.4 Equitability of the caterpillar communities

More than half of all species were found in low abundances (1 individual) in disturbed

as well as in undisturbed forests (Figure 13). There are no visible patterns in the

steepness of the curves representing on the one hand disturbed and undisturbed forests

on the other. In dry and medium forests Cisthene menea (Arctiidae) was particularly

abundant, and to a lesser extent also in humid forests. The highest ranked species in

the old growth humid forest is occupied by Chlosyne gaudealis (Nymphalidae).

Caterpillars with the higher ranks in all forest age groups belong to Pyralidae,

Elachistidae and Crambidae, of which the adults are small nocturnal species.

33

0

0,5

1

1,5

2

2,5

0 20 40

Range

Log

abun

danc

e ca

terp

illar

s

600

0,5

1

1,5

2

2,5

0 20 40 60 8

Range

Log

abun

danc

e ca

terp

illar

s

0

Dry MediumCis-men Cis-men

Pyr Pyr

0

0,5

1

1,5

2

2,5

0 20 40 60

Range

Log

abun

danc

e ca

terp

illar

s

Figure 13. Rank order log abundance plot for dry, medium and humid climate zone, each with the three

forest age-classes. The arrows exemplify the species Cisthene menea (Arctiidae) and Pyraloids in the

highest ranks.

4.2.5 Relationship between caterpillars and vegetation structure and richness

The diversity of plant species expressed in species number did not explain the

distribution of caterpillar diversity, neither did age or average height of the canopy of

the plots. With the unconstrained ordination DCA it was found that the first axis

explains 5% of the caterpillar species variability, and the first plus the second axis

reached 9.1% (Table 11). This is quite high given the species numbers registered

Humid 5-10 years10-30 yearsmature forest

Cis-men

Pyr

34

(Lepš and Šmilauer 2003), and this percentage accounts for more than 30 species of

the total number of caterpillar species registered.

1 2 3 4 Total inertia Eigenvalues 0.650 0.539 0.458 0.392 13.015 Lengths of the gradient 4.087 3.680 3.630 3.097 Cumulative variance of species data (%)

5.0 9.1 12.7 15.7

Sum of eigenvalues 13.015 Table 11. Results of the DCA analysis

The results of the CCA analysis indicate that 16.3% of the variability is explained by

all the variables (related to structure and richness of vegetation) studied, but just two

of them, number of species of all vegetation ≥3 cm DBH (Veg all) and height of the

canopy (Altura 3) were significant and contributed with 43% of the explained

variability (Figure 14, Table 12).

Table 12. Results of the CCA analysis for the environmental variables. The environmental variables at

plot level are canopy height (Altura 3), the number of woody plant species 3≤ x <10 cm DBH (Veg 3-

10), number of woody plant species ≥10 cm DBH (Veg 10), number of woody plant species ≥3 cm

DBH (Veg all) and forest age (Age).

Envir. variable F p Eigenvalue Accum. Variability explained (%) Altura 3 1.342 0.001 0.494 23.3 Veg all 1.181 0.023 0.452 43.7 Veg 10 1.136 0.105 0.415 63.2 Veg 3-10 1.071 0.251 0.441 82.3 Age 1.01 0.457 0.473 100

35

Figure 14. The species-enviromental

variables-samples triplot of CCA. Species

are represented by triangles, 1-32 circles

indicate plots and the arrows the

environmental variables.

-3 4

-46

AgeVeg 3-10

Veg .10

Veg all

Altura 3

1

2

3

4

5

6

78

9

10

11

12

13

14

15

16 17

18

19

20

21222324

25

2627

28

2930

31

32

33

34

35

36

36

5. Discussion

5.1 Changes in plant and caterpillar community structure on the gradient of

climate and forest age

The results show that the plant community structure is significantly different among

forests with different precipitation ranges. The strongest evidence is provided by the

results of the set of 96 samples which demonstrated overall and pairwise significant

differences of the three mature forest types selected. These findings in agreement with

previous observations that the medium sized semi-evergreen forest and high semi-

evergreen forest fall are floristically distinct (Martínez and Gallindo-Leal 2002, Peréz

Salicrup 2004). The same authors also argue that low semi-deciduous and medium

sized semi-evergreen forest contain similar species, and vary only in terms of

structure and abundance. Vester et al (2007) demonstrate how different structural

characteristics of low and medium sized old growth forest types are different from

high semi-evergreen forest. In disagreement with Martínez and Gallindo-Leal (2002)

and Peréz Salicrup (2004) the analyses in the work presented here, indicate that low

and medium sized old growth forest maintain also very distinct plant communities in

terms of species composition. The results show a high proportion of unique species

for each old growth forest type, moreover there is a small proportion of species that

occur in all old growth forests. The caterpillar species composition is significantly

different across the climate zones but not between pairwise combinations. The

absence of significance may be resolved with more sampling effort similarly to plant

communities.

37

We found overall significant dissimilarity among the woody plant communities of the

forest age-classes. Similarly the results show dissimilar caterpillar communities but

the separation was less compared to woody vegetation. At the same time, our data

show that dissimilarity in caterpillar communities between age-classes was strongest

in dry forests while the lowest in the humid climate zone. Vester et al. (2007) found in

the dry climate zone a much more marked reduction of butterfly diversity in the

gradient of disturbance compared to the medium climate zone. Our results and those

of Vester et al. (2007) findings can be related to the idea that changes in caterpillar

community structure along forest age and climate is characterized by an increase of

dissimilarity in caterpillar communities among forest age-classes from the humid

climate zone towards the dry climate zone. The cause for the stronger divergence of

caterpillar assemblages in dry forests along the gradient of forest age is perhaps

related to pronounced unbalanced abiotic conditions, and slower regeneration and

establishment of original plant species during succession that may affect the

caterpillar distribution. Some of the causes may be found in increased and more

variable temperatures, reduced soil water availability and susceptibility to fires,

commonly observed in secondary forests of the dry tropical zones (Nepstad 1999,

Harvey and Eastman 2005). Obviously, the described disturbances also play a role in

the recovery process of secondary forests in the humid areas, but to a lesser extent

because more frequent and long-lasting precipitation replenishes water losses leading

to the recovery of vegetation and more stable environmental conditions.

The gradual increase in overall dissimilarity between caterpillar communities of forest

age-classes from the humid climate to the dry zone, is not unambiguously supported

by the change in woody plant communities. Similar to the response of caterpillar

38

communities, there is evidence of a contrasting response of plant communities in the

dry (high dissimilarity) and the humid forest (low dissimilarity). However, the

medium climate zone accounts for the lowest dissimilarity for plant communities in

the diameter class ≥10 cm DBH but also the highest dissimilarity for plants

communities 3≤ x <10 cm DBH. Driving attributes in the succession and diversity of

tropical vegetation are variable in the gradients considered, which possibly obscure

the expected uniform response in woody plant communities. To illustrate some of

these aspects that vary along the precipitation gradient are nutrient cycling (Lawrence

2002, Santiago et al. 2005), seed and seedling growth (Ray and Brown 1995, Khurana

and Sing 2001) animal removal and dispersal of seed and seedlings (Uhl 1987), while

plot-to-plot variation could be attributed to timing of abandonment and land use

history (Chinea and Helmer 2003) and related presence of mycoryzza (Allen et al.

2003).

5.2 Plant and caterpillar diversity on the gradient of climate

It is generally suggested that the number of tree species per unit area increases with

rainfall and decreases with seasonality in tropical lowland forests (Gentry 1982, 1988;

Wright 1992, Specht and Specht 1993, Clinebell et al. 1995, Aplet et al. 1998).

Although not significant, we found a tendency which is not accordance with this

hypothesized increase i.e.; the number of woody plant species seems the highest in

medium semi-evergreen forest. Moreover, the number of unique species is also

highest in mature medium semi-evergreen forest.

Although not significant, the disturbed and undisturbed forests in the medium climate

zone maintain probably the highest caterpillar richness, in accordance with Pozo et al.

39

(2003) who found a comparable evidence for butterflies. The high caterpillar richness,

this is mainly due to the high caterpillar diversity found in the younger forest age-

classes. Pozo et al. (2003) also demonstrated that the proportion of shared butterfly

species with other climate zones was high, while unique butterfly species in each

forest type was a quarter of all species (Pozo et al. 2003). In the present study, the

proportion of unique caterpillar species was higher than the proportion of shared

species, especially in medium semi-evergreen forest (>50%). In the other climate

zones and forest age-classes the proportion of unique species was also larger than a

quarter. Although this result may be an artefact of insufficient sampling, it could

indicate that the importance of unique species for creating the differences among

Lepidoptera communities between the three distinguished climate zones may be

higher than generally was assumed. This difference can possibly be related to the

differences in selected species; butterflies can move from one forest patch to another,

whereas caterpillars are less mobile.

5.3 Does caterpillar diversity increase along a gradient of forest age?

The conventional perspective on the effect on anthropogenic disturbance is an overall

loss of biodiversity (e.g. Phillips 1997). The diversity of the caterpillar communities

of the Yucatán peninsula did not present a consistent response to disturbance.

Recently, butterfly diversity in forests patches of secondary forests were reported to

sustain higher diversity compared to mature forests in the Calakmul region (Vester et

al. 2007). This effect was attributed to the ‘source-sink relationship’ referring to the

dispersion of butterflies across forest ages in the entire forest mosaic. As mentioned,

we did not reveal clear patterns of caterpillar species diversity on the gradient of

forest age, and particularly the caterpillar diversity in the old secondary forest is

40

variable. Only the dry climate zone shows significant differences between caterpillar

communities that belong to different forest age-classes.

As we have shown already at the level of caterpillar community structure, it is clear

that habitat modification causes species composition change. Another measure of

caterpillar species composition change can be expressed by shared and unique species

distribution along forest age. Research on butterflies in the Calakmul zone showed

that about 60% of the butterfly species were shared between old-growth forest and

forest <10 years (Vester et al. 2007). In our work, the proportion of shared caterpillar

species is less than 10%, and points out the importance of unique species contributing

to the variation caterpillar diversity among habitats. In general it may be that the

number of unique species in our work is relatively higher because we also dealt with

nocturnal species, which considerably increases the chance of encountering unique

species.

The results suggest that primary forests in Calakmul do not present higher diversity

compared to secondary forests. In the light of the ecological theories explaining

diversity on a gradient of forest disturbance, this type of response could be related to

the intermediate disturbance theory (Connell 1977). Previously, it has been argued

before that decreases as well as increases of butterfly diversity in response to forest

disturbance can occur. For example, various studies showed that low disturbance

levels have a positive effects on the local butterfly diversity (Hill 1999, Lovejoy et al.

1986, Brown 1991, Wood and Gillman 1998) while other found a decrease of

butterfly and moth diversity (Bowman et al. 1990, Thomas 1991, Spitzer et al. 1993,

1997, Kremen 1994, Hamer and Hill 2000, Hill et al. 1995, 2001, Hill and Hamer

41

1998, Brown 1997, Lewis et al. 1998, Willott et al. 2000, Lewis 2000, Fermon et al.

2000, 2001).

Previous studies learn us that species (or guilds) have their own environmental

requirements and frequently lead to more sophisticated predictions instead of

considering Lepidoptera community as a whole. For example, the species composition

of Pyralidae did not change in disturbed and undisturbed sites but this family did

show different abundances (Fiedler and Schulze 2004). Similar, it has been

documented that the diversity of moths remained unchanged whereas butterfly

diversity alters at different levels of fragmentation (Daily and Ehrlich 1996). Also in

Borneo it appeared that Sphingidae do not reflect the human induced disturbance

(Schulze and Fiedler 2003). Despite a different response from nocturnal species

compared to diurnal species in various studies, our work showed that the proportion

of diurnal/nocturnal species is rather constant throughout young and old forests in the

Calakmul area. These issues make clear that it is worthwhile to opt for different

analytical measures for future habitat quality studies such as the use of indicator

families and target species, which could enhance the discriminatory power to identify

the responsiveness of Lepidoptera upon disturbance (Basset et al. 2004).

There may be also another effect that causes the rather high but variable caterpillar

diversity in the secondary forests. This effect is related is related to the abundance and

quality of host-plants in the secondary forests, which are characterized with few but

dominant plant species. To illustrate this relation, I will give two examples with the

Lepidoptera families, Hesperidae and Nymphalidae. A number of abundant species of

from the family of Hesperidae feed frequently on Cryosophila stauracantha, Sabal

japa and Chamedorea oblongata (Palmea), while Nymphalidae have often been found

42

on Croton arboreus (Euphorbiaceae), Lonchocarpus xuul (Leguminosae) and Hampea

trilobata (Malvaceae) (this is a personal observation). These plant species are very

abundant in young and disturbed forests (and dry forests). The diversity measures we

used do not account for high abundances of certain plant species as an explanatory

variable. Here, the use of caterpillar guilds related to plant growth form (Janz and

Nylin 1998, Cleary 2003) may yield clearer indicator groups. It leaves no doubt that

this aspect calls for studies to assess the above mentioned effects.

5.4 Does woody plant diversity increase along a gradient of forest age?

In general, it is likely that the disappearance of old growth forests will lead to a

decrease in organisms, particularly of those species which have restricted habitat

requirements (Thomas 1991). Our results indicate that ongoing conversion of mature

forests to secondary stands in the Calakmul region causes a steady decrease of trees

species ≥10 cm DBH in the medium climate zone. Also in the humid climate zone

there is a decrease of sample based species diversity of plant ≥10 cm DBH with

increasing forest-age.

When diversity is rarefied based on individuals, the interpolation limits the curves

causing serious loss of data, at the expense of old secondary and the primary forest.

More sampling effort is needed in the young secondary forest to express a pattern in

the humid climate zone. The rarefied curves of diversity in the dry climate zone are

still more precarious to interpret as the curves are crossing each other. The moment

the curves for two communities intersect, the ranking of the observed species richness

reverses and poses serious limitations in comparisons for species diversity (Lande et

al. 2000). Moreover, the interpolation, based on the young secondary forest species

43

diversity, leads also here to significant losses of data in the old secondary and primary

forest. With respect to plants 3≤ x <10 cm DBH in the dry forest, the problems

described above, applies here too.

It was intended to overcome problems related to smaller sampling sets using the

Simpson index. In general, the Simpson diversity index for plants and caterpillars

present similar response as the individual based rarefaction statistics. The most

apparent difference is that the Simpson index displays a stronger division between the

plant diversity in the secondary forests compared to the mature forests. The Simpson

index confirms the earlier findings of the individual based rarefaction, and supports

the hypothesized increase of plant species richness with increasing forest age,

particularly for the medium and the humid climate zone.

5.5 Caterpillar community equitability

Low levels of equitability suggest the dominance of disturbance or pollution tolerant

species (Favila and Halfter 1997, Floren and Linsenmair, 2001). We expected these

low levels of equitability in younger age-classes compared to mature forests in

Calakmul. This links up with the hypothesis that species diversity decreases with

decreasing forest-age, as equitability is a measure of how well species abundances are

distributed in the community. The rank order abundance curves did not expose

patterns of distinct curves of caterpillar communities in young forests compared to old

forests and the expected steeper of curves that represent caterpillar communities in

disturbed environments was not distinguished. Apparently, the resources in these

44

disturbed forests are sufficiently abundant and stress is sufficiently low to allow a

large diversity with a distribution of abundances similar to mature forests.

5.5 Relationships between caterpillars and vegetation structure and richness

Examination of the CCA did not show clear caterpillar species assemblages

concentrated around forest plots, nor the variables of plant diversity and forest

structure were evidently related to caterpillar distributions. It would be advisable to

repeat the exercise for each climate zone independently. Moreover, the selected

measures for plant diversity and structure are not suited to provide evidence for a

relation with caterpillar communities. The relation between plant and caterpillar

diversity may be clearer when taking into account the undergrowth, particularly herbs.

There are studies executed in south-east Asia forest plantations with high plant

species richness in the undergrowth turned out to support a significantly richer moth

family i.e.; geometrids (Intachat et al. 1997, 1999) as well as other moth (Chey et al.

1997) and butterfly faunas (Ghazoul, 2002) which is also an important family in our

study area.

45

Automeris banus Saturniidae

6. General conclusions

This work contributes to the present investigation on the effects of land use change on

biodiversity by comparing species richness and composition in secondary and primary

forests in the gradient from the dry to the humid climate zone. This work also adds to

current understanding on tropical insect biology, because knowledge on caterpillar

diversity is poor in the literature, even more if it is intended to explain the ecological

aspects involved. The medium climate zone appears to maintain a high diversity of

caterpillars and plants compared to the dry to and the humid climate, but statistical

analyses could not confirm this pattern. There is not enough evidence to support the

idea of a gradual loss of caterpillar diversity over the gradient of forest disturbance.

Instead, caterpillars communities in secondary forests can be highly diverse but the

variation between communities in forest age-classes is large and still more effort is

needed to identify the response of caterpillar communities. I believe that it is

important at this stage to interpret the tendencies rather than statistical differences

only to reflect gradual differences in the forest mosaic. Finally present and future

research on the adult stages of Lepidoptera, host plant interaction, species guilds and

the effect at different levels of the Lepidoptera community structure may enhance our

understanding.

46

Acknowledgements

This work is a product created at El Colegio de la Frontera Sur, Ecosur, Unidad

Chetumal and Wageningen University. I thank the Southern Yucatán Peninsular

Region (SYPR) project involving Clark University, the University of Virginia, El

Colegio de la Frontera Sur, and Rutger University. Its principal sponsors have been

NASA-LCLUC (Land Cover and Land Use Change) program (NAG5-6046 and

NAG5-11134, NNG06GD98G), the Center for Integrated Studies of the Human

Dimensions of Global Environmental Change, Carnegie Mellon University (NSF SBR

95-21914), and NSF-Biocomplexity (BCS-0410016). I would like to thank Hans

Vester, Frans Bongers and Carmen Pozo for reading and giving comments on my

thesis, as well as Jorge Montero and Hector Abuid Hernandez Arana for statistical

assistance. We owe our thanks to those people that have worked on the large set of

mature forest plots, Richard van Sluis en Maarten Debruyne, and specifically in the

ejido of Conhuas, Rafael Espinoza López, Jaime Haas Tzuc, Wilbert Poot Pool,

Daniel Poot Pool, William Naal Segovia, Hans van der Wal, financial aids for these

colleagues came from SEMARNAT in Plan de manejo de Balam Kin en Balam Ku. I

am also grateful to the people I have been working with both in the field and in the

laboratory, Euridice Leyequien Abarca, Angeles Islas Luna, Miguel Xijun, Emigdio

May Uc, José and Magarito and local field guides that have played an invaluable role

for their knowledge Don Magdaleno of the ejido Conhuas, Don Diego of the ejido of

Dos Lagunas, Don Nico of the ejido Nuevo Becal and Don Eliseo of the ejido Zoh

Laguna.

47

Cited References

Allen, E.B., Allen, M.F., Egerton-Warburton, L., Corkidi, L., Gomez-Pompa, A., 2003. Impacts of

early- and late seral mycorrhizae during restoration in seasonal tropical forest, Mexico.

Ecological Applications 13 (6): 1701-1717.

Aplet, G.H., Hughes, R.F., Vitousek, P.M., 1998. Ecosystem development on Hawaiian lava flows:

biomass and species composition. Journal of vegetation science 9 (1): 17-26.

Basset, Y., Novotny, V., Miller, S. E. and Springate, N. D., 1998. Assessing the impact of forest

disturbance on tropical invertebrates: some comments. Journal of Applied Ecology 35:

461-466.

Bernays, E. and Chapman, R., 1994. Host-Plant Selection by Phytophagous Insects. Chapman and Hall,

London.

Blair, R.B., 1999. Birds and butterflies along an urban gradient: surrogate taxa for assessing

biodiversity. Ecological Applications 9:164–170.

Bowman, D.M.J.S., Woinarski, J.C.Z., Sands, D.P.A., Wells, A., McShane, V.J., 1990. Slash-and-burn

agriculture in the wet coastal lowlands of Papua New Guinea: response of birds,

butterflies and reptiles. Journal of Biogeography 17: 227 239.

Bray, J.R. and Curtis, J.T., 1957. An ordination for the upland forest communities of Southern

Wisconsin. Ecol Monographs 27: 325-349.

Brown, K.S. and Hutchings, R.W., 1997. Disturbance, fragmentation, and the dynamics of diversity in

Amazonian forest butterflies. pp. 91–110 in Laurance, W. F. & Bierregaard, R. (eds).

Tropical forest remnants: ecology, management, and conservation of fragmented

communities. The University of Chicago Press, Chicago.

Brown, K.S. Jr., 1994. The use of insects in the study, inventory, conservation and monitoring of

biological diversity in Neotropical habitats, in relation to traditional land use sytems.

Decline and conservation of Butterflies in Japan, III: 128-149, 1996.

Brown, K.S. Jr., 1997. Diversity, disturbance and sustainable use of Neotropical forests: insects as

indicators for conservation monitoring. Journal of Insect Conservation, 1, 25-42.

Brown, K.S., 1991. Conservation of Neotropical Environments: Insects as Indicators, p. 350-404. In:

Collins N.M. and Thomas, J.A. (Eds). The conservation of insects and their habitats.

London, Academic Press, 450p.

Chao, A., 1984. Non-parametric estimation of the number of classes in a population. Scandinavian

Journal of Statistics 11, 265-270.

Chao, A., 1987. Estimating the population size for capture-recapture data with unequal catchability.

Biometrics 43, 783-791.

Chey, V.K., Holloway, J.D., and Speight, M.R., 1997. Diversity of moths in forest plantations and

natural forests in Sabah. Bull. Entomol. Res. 87: 371–385.

Chinea, D.J. and Helmer, E.H., 2003. Diversity and composition of tropical secondary forests

recovering from large-scale clearing : results from the 1990 inventory in Puerto

Rico Forest ecology and management 180: 227-240.

Clarke, K.R. and Gorley, R.N., 2006. PRIMER v6, User Manual / Tutorial. PRIMER-E, Plymouth.

Cleary, D. F. R., 2004. Assessing the use of butterflies as indicators of logging in Borneo at three

taxonomic levels. J. Econ. Entomol. 97 (2): 429-435.

Cleary, D.F.R. and Mooers, A.Ø., 2004. Butterfly species richness and community composition in

forest affected by ENSO-induced burning and habitat isolation in Borneo. Journal of

Tropical Ecology 20: 359-367.

Cleary, D.F.R., 2003. An examination of scale of assessment, logging and ENSO-induced fires on

butterfly diversity in Borneo. Oecologia 135: 313-321.

Clench , H. K. 1979. How to make regional lists of butterflies: some thoughts. Jour. Lep. Soc. 33: 215-

231

Clinebell, H.R.R., Phillips, O.L., Gentry, A.H., Stark, N. and Zuuring,, H., 1995. Predictions of

neotropical tree and liana species richness from soil and climatic data. Biodiversity and

conservation, 4: 56-90.

Colwell, R.K, 2005. User's Guide & application: http//:viceroy.eed.uconn.edu/estimates. University of

Connecticut, USA, 36 pp.

Colwell, R.K., 2004. EstimateS. Statistical Estimation of Species Richness and Shared Species from

Samples. Version 7.

Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science 199: 1304–1310.

Daily, G.C., Ehrlich, P.R.., 1996. Nocturnality and species survival. Proc Natl Acad Sci 15,

93(21):11709-12.

Dennis, R.L. and Shreeve, T.G., 1991. Climatic change and the British butterfly fauna: opportunities

and constraints. Biological Conservation 55: 1-16.

Eberhardt, L.L. and Thomas, J.M., 1991. Designing environmental field studies. Ecological

Monographs 61(1): 53-73.

Eggleton, P., Bignell, D. E., Sands, W. A., Mawdsley, N. A., Lawton, J. H., Wood, T. G. Dupuy, J. M.

& Chazdon, R. L., 1998. Long-term effects of forest regrowth and selective logging on

the seed bank of tropical forests in NE Costa Rica. Biotropica 30: 223–237.

Ehrlich, P.R. and Raven, P.H., 1965. Butterflies and plants: A study in coevolution. Evolution 18: 586-

608.

Ehrlich, P.R., 1984. The structure and dynamics of butterfly populations. In: Vane-Wright, R.I.,

Ackery, P.R. (Eds.), The Biology of Butterflies. Symposium of the Royal Entomological

Society of London, No. 11. Academic Press, London, UK, pp. 25-40.

Ehrlich, P.R., Breedlove, D.E., Brussard, P.F. and Sharp, M.A., 1972. Weather and the regulation of

subalpine populations. Ecology 53(2): 244-247.

Favila, M.E. and Halfter, G., 1997. The use of indicator groups for measuring biodiversity as related to

community structure and function, Acta Zool. Mex 72: 1-25.

Fermon, H., Schulze, C. H., Waltert, M., Mühlenberg, M., 2001. The butterfly community of the

Noyau Central, Lama Forest (Republic of Benin), with notes on its ecological

composition and geographic distribution. African Entomology 9: 177-185.

Fermon, H., Waltert, M., Larsen, T. B., Dall’Asta, U., Mühlenberg, M., 2000. Effects of forest

management on diversity and abundance of fruit-feeding nymphalid butterflies in

southeastern Côte d’Ivoire. Journal of Insect Conservation 4: 173-189.

Fiedler, K, Schulze, C.H., 2004. Forest Modification Affects Diversity (But Not Dynamics) of Speciose

Tropical Pyraloid Moth Communities. Biotropica 36 (4): 615–627.

Floren, A. and Linsenmair, K.E., 2001. The influence of anthropogenic disturbances on the structure of

arboreal arthropod communities. Plant Ecology 153: 153-167.

Flores, J.S., Espejel, I., l994. Tipos de Vegetación de la Península de Yucatán. Etnoflora Yucatanense.

Fascículo 3, Universidad Autónoma de Yucatán, 135 pp.

Fox, J.F. and Connell, J.H., 1979. Intermediate disturbance hypothesis. Science 204 (4399): 1344-1345.

Fuller, R.M., Groom, G.B., Mugisha, P., Ipulet, P., Pomeroy, D., Katende, A., Baily, R. and Ogutu-

Ohowayo, R., 1998. The integration of field survey and remote sensing for biodiversity

assessment: a case study in the tropical forests and wetlands of Sango Bay, Uganda.

Biological Conservation 86:379–391.

Gentry, A.H.,1982. Patterns of neotropical plant species diversity. Evolutionary Biology, 15: 1-84.

Gentry, A.H.,1988.Changes in plant community diversity and floristic composition on environmental

and geographical gradients. Anal of the Missouri Botanical Graden, 75: 1-34.

Gilbert, L.E., 1980. Food webs organization and the conservation of neotropical diversity, in Soulé

M.E. and Wilcox, B.A. (editors). Conservation biology: an evolutionary-ecological

perspective. Sinauer Associates, Sunderland, Massachusetts, USA.

Gotelli, N.J. and Colwell R.K., 2001. Quantifying biodiversity: procedures and pitfalls in the

measurement and comparison of species richness. Ecology Letters 4: 379-391.

Gretchen, C., Daily, P.R., 1997. Nocturnality and Species Survival

Proceedings of the National Academy of Sciences of the United States of America 93

(21): 11709-11712

Halfter, G.A., 1998. Strategy for measuring landscape biodiversity. Biology International 36: 3-11.

Hamer, K.C. and Hill, J. K., 2000. Scale-Dependent Effects of Habitat Disturbance on Species

Richness in Tropical Forests. Conservation Biology 14 (5), 1435–1440.

Hamer, K.C., Hill, J.K., Lace, L.A. and Langan, A.M., 1997. Ecological and Biogeographical Effects

of Forest Disturbance on Tropical Butterflies of Sumba, Indonesia Journal of

Biogeography 24(1): 67-75.

Hamer, K.C., Hill, J.K., Lace, L.A., Langman, A.M., 1997. Ecological and biogeographic effects of

forest disturbance on tropical butterflies of Sumba, Indonesia. Journal of Biogeography,

24, 67-75.

Hanski, I., Pytry, J., Pakkala, T. and Kuussaari M., 2002. Multiple equilibria in metapopulation

dynamics. Nature 377, 618 - 621.

Harvey, E. and Eastman, R., 2005. The impact of human disturbance on local climate in the Yucatán,

Mexico. Anais XII Simpósio Brasileiro de Sensoriamento Remoto, Goiânia, Brasil, 16-21

abril 2005, INPE, p. 3045-3052.

Heliconius and Passiflora. Science 201:745-747.

Heppner, J.B., 1991. Faunal regions and the diversity of Lepidoptera. Tropical Lepidoptera. Vol. 2, no.

suppl. 1.

Hill, J.K. and Hamer, K.C., 1998. Using species abundance models as indicators of habitat disturbance

in tropical forests. Journal of Applied Ecology 35 (3), 458–460.

Hill, J.K., Hamer, K.C. , Tangah, J., Dawood, M., 2001. Ecology of tropical butterflies in rainforest

gaps. Oecologia 128: 294-302.

Hill, J.K., Kramer, K.C., Lace, L.A., Banham, W.M.T., 1995. Effects of selective logging on tropical

forest butterflies on Buru, Indonesia. J. Appl. Ecol. 32: 754-60.

Holloway, J.D., 1998. The impact of traditional and modern cultivation practices, including forestry, on

Lepidoptera diversity in Malaysia and Indonesia. Dynamics of Tropical Communities (ed.

Newbery, D.M., Prins, H.H.T. & Brown, N.), pp. 567-597. Blackwell Science, Oxford.

Holloway, J.D., Kirk-Spriggs, A.H., Chey, A.H., 1992. The response of some rain forest insect groups

to logging and conversion to plantation. Philosophical Transactions of the Royal Society

of London, Series B, 335: 425-436.

Howard, P.C., Viskanic, P., Davenport, T. R. B., Kigenyi, F. W., Baltzer, M., Dickenson, C. J.,

Lwanga, J. S., Mathews, R. A. and Balmford, A., 1998. Complementarity and the use of

indicator groups for reserve selection in Uganda. Nature 394:472–475.

Huston, M. A. 1979. A general hypothesis of species diversity. Am. Nat. 113: 81–101.

Huston, M. A. 1994. Biological diversity: the coexistence of species on changing landscapes. –

Cambridge Univ. Press.

Intachat, J., Chey, V.K., Holloway, J.D., and Speight M.R. 1999. The impact of forest plantation

development on geometroid moth population and diversity in Malaysia. J. Trop. For. Sci.

11: 329–336.

Intachat, J., Holloway, J.D., and M.R. Speight, 1999. The impact of logging on geometroid moth

populations and their diversity in lowland forests of Peninsular Malaysia. J. Trop. For.

Sci. 11: 61–76.

Janz, N. and Nylin, S., 1998. Butterflies and Plants: A Phylogenetic Study. Evolution, Vol. 52 (2):

486-502

Jennersten, O., 1988. Pollination in Dianthus deltoides (Caryophyllacae): effects of habitat

fragmentation on visitation and seed set. Conservation biology 2: 359-367.

Khurana, E. and Singh, J.S., 2001. Ecology of seed and seedling growth for conservation and

restoration of tropical dry forest : a review. Environmental Conservation (2001), 28: 39-

52

Kitching, R.L., Orr, A.G., Thalib, L., Mitchell, H., Hopkins, M.S. & Graham, A.W., 2000. Moth

assemblages as indicators of environmental quality in remnants of upland Australian rain

forest. Journal of Applied Ecology 37 (2), 284-297.

Kremen, C. 1992. Assessing the indicator properties of species assemblages for natural areas

monitoring. Ecological Applications 2(2): 203-217.

Kremen, C., 1994. Biological inventory using target taxa: a case study of the butterflies of Madagascar.

Ecological Applications 4, 407-422.

Lande, R., DeVries, P.J. and Walla T.R., When species accumulation curves intersect: implications for

ranking diversity using small samples. Oikos 89 (3): 601-605.

Laurance, W.F., Lovejoy, T.E, Vasconcelos, H.L., Bruna, E.M., Didham, R.K, Stouffer, P.C., Gascon,

C., Bierregaard, R.O, Laurance, S.G, Sampaio, E., 2002. Ecosystem Decay of Amazonian

Forest Fragments: a 22-Year Investigation. Conservation Biology 16 (3): 605.

Lawrence, D. and Foster, D., 2002. changes in forest biomass, litter dynamics and soils following

shifting cultivation in southern Mexico: an overview. INCI, 27, 8, p.400-408.

Lawrence, D., Vester, H., Pérez-Salicrup, D., Eastman, J. R., Turner, B. L., II, and Geoghegan, J. 2004.

Integrated Analysis of Ecosystem Interactions with Land-Use Change: The Southern

Yucatán Peninsular Region. In DeFries, R., Asner, G. and Houghton, R. Ecosystem

Interactions with Land Use Change, pp. 277-292. Washington, D. C., American

Geophysical Union.

Lawton, J.H., Bignell, D.E., Bolton, B., Bloemers, G.F.; Eggleton, P., Hammond, P.M., Hodda, M.,

Holt, R.D., Larsen, T.B.; Mawdsley, N.A., Stork, N.E., Srivastava, D.S., Watt, A.D.,

1998. Biodiversity inventories, indicator taxa and effects of habitat modification in

tropical forest. Nature, Volume 391, Issue 6662, pp. 72-76.

Lepš, J. and Šmilauer, P., 2003. Multivariate Analysis of Ecological Data Using CANOCO. Inc

NetLibrary,Cambridge University Press.

Lewis, O. T., 2000. Effects of experimental selective logging on tropical butterflies. Conservation

Biology 15: 389-400.

Lewis, O.T., Wilson, R.J., Harper, M.C., 1998. Endemic Butterflies on Grande Comore: habitat

preferences and conservation priorites. Biological Conservation 85: 113-121.

Lovejoy, T..E., Bieregaard, R.O., Rylands, Jr., A.B., Malcolm, J.R., Quintela, L.H., Harper, Brown,

K.S., Powell, Jr., A.H., Powell, H.O.R., Schubart, H.O.R., Hays, M.B., 1986. Edge and

other effects of isolation on amazon forest management. In Conservation Biology (ed. M.

Soulé), pp. 257-285. Sinauer Associates, Inc, Sunderland, MA.

Magurran, A. E. 1988. Ecological diversity and its measurement. Princeton Univ. Press.

Magurran, A. E. 2004. Measuring biological diversity. Blackwell publishing.

Martinez, A.M., Pozo, C., May Uc, E., 2005. Las mariposas (Rhopalocera: Papilionidae, Pieridae y

Nymphalidae) de la selva salta subperennifolia de la region de Calakmul, Mexico, con

nuevos registros. Folia Entomol. Mex., 44(2): 123-143.

Martínez, E., Galindo-Leal, C., 2002. La vegetación de Calakmul: descripción, composición y

distribución. Boletín de la Sociedad Botánica de México. 71:7–32.

Murphy, D.D. and Wilcox, B.A. 1986. Butterfly diversity in natural habitat fragments: a test of the

validity of vertebrate based management. Pages 287-292 In Vernier, J., Morrison, L. and

Ralph, C.J. (Eds.) Wildlife 2000: modeling habitat relationships of terrestrial vertebrates.

The University of Wisconsin Press, Madison, Wisconsin, USA.

Murphy, D.D., Freas, K.E. and Weiss, S.B. 1990. An environment-metapopulation approach to

population viability for a threatened invertebrate. Conservation biology 4(1): 41-51.

Nepstad, D. C., A. Verissimo, A. Alencar, C. Nobre, E. Lima, P. Lefebvre, P. Schlesinger, C. Potter, P.

Moutinho, E. Mendoza, M. Cochrane, and V. Brooks. 1999. Large-scale impoverishment

of Amazonian forests by logging and fire. Nature 398: 505-508.

Novotny, V., Scott E. Miller, S.M., Leps J., Basset, Y., Bito, D., Janda, M., Hulcr, J., Damas, K.,

Weiblen, G.D., 2004. No tree an island: the plant-caterpillar food web of a secondary rain

forest in New Guinea Ecology Letters 7 (11), 1090–1100.

Nummelin, M., 1998. Log-normal distribution of species abundance is not a universal indicator of

rainforest disturbance. Journal of Applied Ecology, 35, 454 457.

Nystrőm M., Folke, C., Moberg, F., 2000. Coral reef disturbance and resilience in a human-dominated

environment. Trends Ecolological Evolution 15:413–417.

Pérez-Salicrup, D.R., 2004. Forest types and their implications. In: Turner, B. L., II, J. Geoghegan, and

D. R. Foster, editors. Integrated land-change science and tropical deforestation in the

southern Yucatán: final frontiers. Oxford University Press, Oxford, United Kingdom.

Phillips, O.L. 1997 The changing ecology of tropical forests. Biodiversity and Conservation, 6: 291

311.

Pollard, E., 1979. Population ecology and cange in range of the white admiral butterfly Ladoga camilla

L. in England. Ecol. Entomology 4: 61-74.

Posa, M.R.C. and Sodhi, N.S., 2006. Effects of anthropogenic land use on forest birds and butterflies in

Subic Bay, Philippines. Biological Conservation 129 (2): 256-270.

Pozo, C., Martinez, A.L., Uc Tescum, S., Salaz Suarez, N., and Maya Martinez, E., 2003. Butterflies

(Papilionoidea and Hesperioidea) of Calakmul, Campeche, Mexico. The Southwestern

Naturalist 48 (4) 505-525.

Primack, R. B., Bray, D., Galletti, H. A. and Ponciano, I., eds., 1998. Timber, Tourists, and Temples:

Conservation and Development in the Maya Forests of Belize, Guatemala, and Mexico,

Island Press, Washington, D.C.

Rausher, M.D. and Feeny, P., 1980. Herbivory, plant density, and plant reproductive artes: the effect of

Battus philenor on Aristolochia reticulata. Ecology 61: 905-917.

Ray, G. J. and Brown, B. J., 1995. Restoring Caribbean Dry Forests: Evaluation of Tree Propagation

Techniques. Restoration Ecology 3: 86–94.

Roth, D. S., Perfecto, I. and Rathcke, B., 1994. The effects of management systems on ground-

foraging ant diversity in Costa Rica. Ecol. Appl. 4: 423–436.

Santiago Lastra, J., Garcia Barrios, L.E., Rojas, J.C., Perales Rivera, H., 2006. Host selection behaviour

of Leptophobia Aripa (Lepidoptera: Pieridae). Florida Entomologist: 89 (2) 127–134.

Santiago, L.S., Schuur, E.A.G. and Silvera, K., 2005. Nutrient cycling and plant–soil feedbacks along a

precipitation gradient in lowland Panama. Journal of Tropical Ecology 21: 461-47

Schowalter, T. D. and Ganio, L.M., 1999. Invertebrate communities in a tropical rain forest canopy in

Puerto Rico following Hurricane Hugo. Ecol. Entomol. 24: 191–201.

Schulze, C.H. and Fiedler, K., 2003. Hawkmoth diversity in Northern Borneo does not reflect the

influence of anthropogenic habitat disturbance. Ecotropica 9: 99–102.

Schulze, C.H., Linsenmair, K.E. and Fiedler, K., 2001. Understorey versus canopy: patterns of vertical

stratification and diversity among Lepidoptera in a Bornean rain forest. Plant Ecology

153: 133–152.

Smiley, J. 1978. Plant chemistry and the evolution of host specificity: New evidence for

Soberón, J. and Llorente, J., 1993. The use of species accumulation functions for the prediction of

species richness. Conservation Biology 7 (3):480-488

Southwood, T.R.E. and Kennedy, C.E.J., 1983. Trees as islands, Oikos 41, 359-371.

Specht, A. and Specht, R.L. 1993. Species richness and canopy productivity of Australian plant

communities. Biodiversity and Conservation 2: 152 167.

Spitzer, K., Jaroš, J., Havelka, J., Lepš, J. 1997. Effects of small-scale disturbance on butterfly

communities of an Indochine montane rainforest. Biological Conservation 80 (9) 15.

Ter Braak, C. J. F. and Van Tongeren, O. F. R., 1995. Data Analysis in Community and Landscape

Ecology. Cambridge University Press. 321 pp.

Thomas, C.D., 1991. Habitat use and geographic ranges of butterflies from the wet lowlands of Costa

Rica. Biological Conservation 58: 269–281.

Thomas, J.A., Bourn, N. A. D., Clarke, R. T., Stewart, K. E., Simcox, D. J., Pearman, G. S., Curtis, R.,

and Goodger, B., 2001. The quality and isolation of habitat patches both determine where

butterflies persist in fragmented landscapes. Proc. R. Soc. Lond. B (268): 1791-1796.

Thompson, J. and Pellmyr, O., 1991. Evolution of oviposition behavior and host preference in

Lepidoptera. Annu. Rev. Entomol. 36:65–89.

Turner, B.L., Villar, S.C., Foster, D., Geoghegan, J., Keys, E., Klepeis P., Lawrence D., Mendoza,

P.M., Manson, S., Ogneva-Himmelberger, Y., Plotkin, A.B., Salicrup, D.P., Chowdhury,

R.R., Savitsky, B., Schneider, L., Schmook, B., Vance, C., 2001. Deforestation in the

Southern Yucatán Peninsular Region: An Integrative Approach. Forest Ecology and

Management 154 (3): 343-370.

Turner, J.R.G., Gatehouse, C.M. and Corey, C.A., 1987. Does solar energy control organic diversity?

Butterflies, moths and the British climate. Oikos 48: 195-205.

Uhl, C., 1987. Factors Controlling Succession Following Slash-and-Burn Agriculture in Amazonia.

The Journal of Ecology 75 (2): 377-407.

Vester, H.F.M., Lawrence, D., Eastman, J.R., Turner, B.L., Calme, S., Dickson, R., Pozo, C.,

Sangermano, F., 2007. Land Change in the Southern Yucatán and Calakmul Biosphere

Reserve: Implications for Habitat and Biodiversity. Ecological Applications, 17 (4): 989-

1003.

Watt, A.D., 1998. Measuring disturbance in tropical forests: a critique of the use of species-abundance

models and indicator measures in general. Journal of Applied Ecology 35 (3), 467–469.

Watt, A.D., Stork, N.E., Eggelton, P., Srivastava, D., Bolton, B., Larsen, T.B. and Brendell, M.J.D.,

1997. Impact on forest loss and regeneration on insect abundance and diversity. Forests

and Insects (eds. Watt, A.D., Stork, N.E. and Hunter, M.D.), pp. 274-286. Chapman and

Hall, London.

Willott, S. J., Lim, D. C., Compton, S. G., Sutton, S. L. (2000) Effects of Selective Logging on the

Butterflies of a Bornean Rainforest. Conservation Biology 14, 1055-1065.

Wood, B., Gillman, M. P., 1998. The effects of disturbance on forest butterflies using two methods of

sampling in Trinidad. Biodivers. and Conserv. 7: 597-616.

Wright S.J., 1992, Seasonal drought, soil fertility and the species density of tropical forest plant

communities. Trends in Ecology and Evolution, 7: 260-263.

Webpages

Janzen Halwach webpage of Caterpillars Data Collection, Costa Rica:

http://janzen.sas.upenn.edu/Wadults/searchcat2.lasso

Barcode of Life, University of Guelph, Canada:

http://www.barcodinglife.org/views/login.php

Bardaxima perses Notodontidae

Identification keys and photo collections for Lepidoptera

Beutelspacher C.R., 1992. Catalogo de la colección Roberto Muller (Lepidoptera: Heterocera) del

museo de historia natural de la ciudad de México. Cuadernos 15. Instituto de Biología.

Universidad Nacional Autónoma de Mexico, Secretaria de Biología, UNAM, México

D.F., México.

Carter, D., 1992. Butterflies and Moths. Dorling Kindersly Limited, London, UK.

Costa, C., Ide, S., Simonka, C.E., 2006. Insectos inmaduros. Metamorfosis e identificación.

Monografias Tercer Milenio,Vo. 5. SEA, CYTED & RIBES, Zaragoza, Spain.

Novell, Jr. C.V, 2005. A field guide to moths of eastern North America. Virginia Museum of Natural

History. Special publication Number 12. Martinsville, Virginia, U.S.

DeVries, P.J., 1987. The butterflies of Costa Rica and their natural history. Papilionidae, Pieridae,

Nymphalidae. New Jersey, USA.

DeVries, P.J., 1997. The butterflies of Costa Rica and their natural history.Volume II: Riodinidae. New

Jersey, USA.

McCord, D., Floridas fabulous butterflies. World Publications, Tampa, Singapore.

Miller, J.C, Janzen, D.H., and Hallwachs, W., 2006. 100 Caterpillars. Portraits from the tropical forests

of Costa Rica. London, England.

Minno, M.C. and Emmel, T.C., 1993. Butterflies of the Florida Keys. Gainesville, Florida, USA.

Pescador Rubio, A., 1992. Manual de identificación para las mariposas de la familia Sphingidae

(Lepidoptera) de la estación de biología ´Chamela´, Jalisco, México. Cuadernos 22.

Instituto de Biología, Universidad Autónoma de México, Secretaria de Biología, UNAM,

México DF., México.

Peterson, A., 1962. Larvae of insects. An introduction to Neartic species. Part I. Lepidoptera and plant

infesting Hymenoptera. Ann Arbor, Edwards Brothers. 315p.

Pinas F., 2000. Mariposas del Ecuador. Butterflies and Moths of Ecuador. Arctiidae: Arciinae y

Pericopnae. Vol. 20. Quito, Ecuador.

Pinas Rubio, F. and Iván Manzano Pesantez, 1997. Mariposas del Ecuador. Vol. 1. Géneros. Quito,

Ecuador.

Scott, J.A., 1986.The butterflies of North America. A natural histrory and Filed Guide. Stanford

University Press, Stanford, USA.

Tyler, H.A. et al., 1994. Swallowtail Butterflies of the Americas. Scientific Publishers, Inc. Gainsville,

USA.

Vazquez Garcia, L., 1989. Tipos en la colección entomología del instituto de biología. Cuadernos 2.

Instituto de Biología, Universidad Autónoma de México, Secretaria de Biología, UNAM,

México DF., México.

Appendix Appendix 1 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Dry climate zone. Appendix 2 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Medium climate zone. Appendix 3 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Humid climate zone. Appendix 4 Individual-based species richness curves for caterpillars in the forest age-classes in the climate zones. Appendix 5 Rarefied diversity for plants and caterpillars in the forest age classes Appendix 6 Simpson index for plants and caterpillars in the forest age classes Appendix 7 Clench model for caterpillars in the forest age classes in each climate zone

Appendix 1 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Dry climate zone.

0 1 2 3 4 50

10

20

30

40

50

60

70

0 1 2 3 4 50

10

20

30

40

50

Num

ber o

f spe

cies

P

lant

s ≥3

cm

DB

H

0 1 2 3 4 50

10

20

30

40

Num

ber o

f spe

cies

P

lant

s ≥3

- <1

0 cm

DBH

Num

ber o

f spe

cies

P

lant

s ≥1

0 cm

DB

H

Samples Samples 5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

0 20 40 60 80 100 120 140 160 180 2000

10

20

30

40

50

60

70

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

Individuals Individuals0 20 40 60 80 100 120

0

10

20

30

40

50

Num

ber o

f spe

cies

P

lant

s ≥3

cm

DB

H

0 20 40 60 800

10

20

30

40

Num

ber o

f spe

cies

P

lant

s ≥3

- <1

0 cm

DBH

Num

ber o

f spe

cies

P

lant

s ≥1

0 cm

DB

H

Appendix 2 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Medium climate zone.

0 1 2 3 4 50

10

20

30

40

50

60

Num

ber o

f spe

cies

P

lant

s ≥3

cm

DB

H

0 1 2 3 4 50

10

20

30

40

50

60

70

80

0 1 2 3 4 50

10

20

30

40

50

10-30 yrsmf

Samples Samples Samples 5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

Individuals 0

Individuals 20 40 60 80 100 120 140

0

10

20

30

40

50

5-10 yrs 10-30 yrsmf

Individuals

0 20 40 60 80 100 120 140 160 180 2000

10

20

30

40

50

60

10-30 yrsmf

Individuals

0 50 100 150 200 250 3000

10

20

30

40

50

60

70

80

Appendix 3 Sample-based and individual-based species richness curves for plants in the forest age-classes in the Humid climate zone.

0 1 2 3 4 50

10

20

30

40

50

60

0 1 2 3 4 50

10

20

30

40

50

Num

ber o

f spe

cies

P

lant

s ≥3

- <1

0 cm

DBH

Num

ber o

f spe

cies

P

lant

s ≥1

0 cm

DB

H

0 1 2 3 4 50

10

20

30

40

50

Num

ber o

f spe

cies

P

lant

s ≥3

cm

DB

H

Samples Samples 5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

Samples 5-10 yrs 10-30 yrsmf

0 20 40 60 80 100 120 1400

10

20

30

40

50

Num

ber o

f spe

cies

P

lant

s ≥3

- <1

0 cm

DBH

0 20 40 60 80 100 120 140 1600

10

20

30

40

50

0 50 100 150 200 250 3000

10

20

30

40

50

60

Num

ber o

f spe

cies

P

lant

s ≥3

cm

DB

H

Num

ber o

f spe

cies

P

lant

s ≥1

0 cm

DB

H

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

5-10 yrs 10-30 yrsmf

Individuals IndividualsIndividuals

Appendix 4 Individual-based species richness curves for caterpillars in the forest age-classes in the climate zones.

0 100 200 300 400 500 600 700

Individuals

Num

ber o

f cat

erpi

llar s

peci

es

0

80

20

40

60

0 100 200 300 400 500 600 7000

20

40

60

80

0 100 200 300 400 500 600 7000

20

40

60

80

5-10 years10-30 yearsmature forest

Individuals 5-10 years10-30 yearsmature forest

Humid forestsMedium forests

5-10 years10-30 yearsmature forest

Dry forests

Num

ber o

f cat

erpi

llar s

peci

es

Individuals

Individual-based species richness curves for the age classes within each of the three types of climate. The arrows indicate the limiting curve for rarefying richness in the

other habitats. In the graph of Humid forests an example is given how rarefied richness is read from the graph.

Appendix 5 Rarefied diversity for plants and caterpillars in the forest age classes

Rarefied diversity Dry climate zone Medium climate zone Humid climate zone

Sample based

Indiv. based

Sample based

Indiv. based

Sample based

Indiv. based Plant diversity Diameter class ≥3x<10 cm DBH

No

of S

peci

es

0204060

5-10 yrs 10-30 yrs mf

0102030

5-10 yrs 10-30 yrs mf

020406080

5-10 yr s 10-30 yr s mf

0

20

40

60

5-10 yrs 10-30 yrs mf

0

20

40

60

5-10 yrs 10-30 yrs mf

010203040

5-10 y rs 10-30 yrs mf

Diameter class ≥10 cm DBH

No

of S

peci

es

0204060

5-10 yrs 10-30 yrs mf

0102030

5-10 yrs 10-30 yrs mf

020406080

10-30 yr s mf

0

20

40

60

10-30 y rs mf

0

20

40

60

5-10 yrs 10-30 yrs mf

010203040

5-10 y rs 10-30 yrs mf

All woody species ≥3 cm DBH

No

of S

peci

es

0204060

5-10 yrs 10-30 yrs mf

0102030

5-10 yrs 10-30 yrs mf

020406080

5-10 yrs 10-30 yrs mf

0

20

40

60

5-10 yrs 10-30 yrs mf0

20

40

60

5-10 yrs 10-30 yr s mf

010203040

5-10 yrs 10-30 yrs mf

Caterpillar diversity

No

of S

peci

es

0

20

40

60

5-10 yrs 10-30 yrs mf

0

20

40

60

5-10 y rs 10-30 yrs mf

0

20406080

5-10 y rs 10-30 y rs mf0

20406080

5-10 yrs 10-30 yrs mf

0

20

40

60

5-10 yrs 10-30 yrs mf

0

20

40

60

5-10 yrs 10-30 yrs mf

Appendix 6 Simpson index for plants and caterpillars in the forest age classes

Simpson Index Dry climate zone Medium climate zone Humid climate zone Plant diversity Diameter class ≥3x<10 cm DBH

No

of S

peci

es

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

5-10 year s 10-30 year s mf

Diameter class ≥10 cm DBH

No

of S

peci

es

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

10-30 year s matur e f or est

All woody species ≥3 cm DBH

No

of S

peci

es

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

5-10 year s 10-30 year s mf

05

1015202530

5-10 year s 10-30 year s mf

Caterpillar diversity

No

of S

peci

es

0

5

10

15

20

5-10 year s 10-30 year s mf

0

5

10

15

20

5-10 year s 10-30 year s mf

0

5

10

15

20

5-10 year s 10-30 year s mf

Appendix 7 Clench model for caterpillar species richness in the forest age classes in each climate zone

Dos Lagunas

0

20

40

60

80

100

0 300 600 900 1200 1500

No. individuals

No. s

peci

es

5-10 yrs10-30 yrsMf

Humid Naadzkan

0

20

40

60

80

0 300 600 900 1200 1500 1800

No. individuals

No. s

peci

es 5-10 yrs10-30 yrsMf

Dry

Nuevo Becal

0

20

40

60

80

100

0 300 600 900 1200 1500 1800

No. individuals

No.

spe

cies 5-10 yrs

10-30 yrsMf

Medium

Climate zones Dry Medium Humid Statistics 5-10 yrs 46 % 56 % 35 % 10-30 yrs 35 % 36 % 57 % mature forest 60 % 52 % 37 %

The percentage of registered species with respect to the

total number of expected species (asymptote).