Content uploaded by Ana Maria Benavides
Author content
All content in this area was uploaded by Ana Maria Benavides on May 12, 2014
Content may be subject to copyright.
Spatial Distribution and Population Structure of
Zamia obliqua in Pacific Coast, Chocó, Colombia
Ana Maria Benavides Duque
Corporación Para Investigaciones Biológicas, Medellín Colombia
(versión enero 2008)
Keywords: Zamia, population structure spatial distribution,
Introduction
Zamia obliqua (Cycadales: Zamiaceae) is a beautiful understory long-lived plant with a
large leaves set and rich gold new leaves, represented one of the oldest seed plants.
Zamia obliqua extends from southern Panama to the southern Chocó of Colombia,
deforestation for logging and expanding agriculture in the Chocó region could have a
dramatic effect on Zamia populations.
Chocó region extends from eastern Panama along the Pacific coast of Colombia and
Ecuador (figure). It is considered one of the most species-rich lowland areas in the
word (WWF 2007). In the region, 20% of forest has been destroyed during the past
years and the deforestation rate per year is near to 600 km2, the resource at river and
coastal zones have been over-exploited for decades.
Plant populations in tropical forests are seriously affected by increasing deforestation
and constant extractive activities. In South America, cycad populations are sensitive to
habitat degradation and prone to extinction because the small effective population
sizes, the isolation among the dioecity of the plants (male or female plants), and the
dependence of Curculionidae beetles in the pollination process (Donaldson 2003).
Reproductive failure contributes to declination of cycad population (Donaldson 2003).
Particularly, Zamia obliqua is highly endangered in Panama as a result of habitat
destruction (Stevenson 2004), in Colombia little is known about population status and
life history.
In this study, the population structure and distribution of Zamia obliqua in an area of the
pacific coast at Colombian Chocó was examined. The study goals were to: (1) analyze
the distribution of the life cycle-stages according to environmental, distance and
topographic factors, and (2) evaluate the conservation status of the studied populations
and life history. There are an increasingly interests in documenting and understanding
the spatial structure and population structure of plants in tropical forest, it is hoped that
this preliminary results obtained provide valuable information for conservation and
management guidelines that will allow efficient long term conservation actions.
Methods
Study species Zamia obliqua
Zamia obliqua is one of the most beautiful New World cycads (Stevenson 2004).
Zamia obliqua is a forest understory species that is widespread in primary and
secondary rain forest from near sea level to 500 m in southern Panama to the southern
Chocó of Colombia (Donaldson, 1997). This species has arborescent habit and
individuals up to 4,5 m tall can be found, bearing up to 90 long recurved leaves.
Populations consist mainly of seedlings and juveniles; reproductive adults are
uncommon (Stevenson 2004 and this study). Under field conditions, plant does
regeneration and exhibit commonly vegetative branching (pers. obs.). Like all cycads,
Z. obliqua is dioecious with male and female cones on separate plants. Between 3-6
male cones were found in three plants and 1-3 female cones in three plants, curiously
all female plant were close. The female cones produced from 135 to 300 seeds
(mean=241, n=8). During this study seed cones do occurred in larger (1.5 m)
individual, but contrary to previous reports three male individuals and two female
individuals were shorter than 1 m. Individuals with cones were only observed in
gardens and backyards.
Study Area
The study was conducted in the Tribugá Gulf in the Chocó region of Colombia (figure
1). The Chocó region is one of the top biodiversity hotspots due to its levels of
endemism, diversity, and biological complexity. Forests develop form the cost line up to
the hills that can be 100 m high. The hills are small with smooth slopes and deep soils.
Mangrove areas, flooded forests and non flooded forest can be found in level ground
almost at sea level, and lowland Rain Forest is found in the slopes and top of the hills.
The study area is dominated by mountain forest, mangroves and “firme” forest.
Elevation within the study area ranged from 0 to 100 m above sea level. Average
annual rainfall is between 5500 and 9632 mm (El Amargal Meteorological station,
IDEAM 1998-2003). The rainy season extends from August to October; dry months are
from January to April. The area covers old growth forest with different levels of
perturbation as consequence of extractive activities and patches of secondary
vegetation can be observed. Three sites were visited during the field-work: “El
Amargal” biological research station, Guachalito private reserve and Coquí
surroundings (Afrocolombian communal land) (figure 1). The maximum distance
between two sites was 11.5 km. Previous studies have shown that the best preserved
forests are those located at “El Amargal” Biological Station, and the flora has been
described in detail by Galeano et. al (1999?).
Field work
The sampling took place between September and October of 2007. One of the sites
was at “El Amargal” biological research station, the selected area had a gradual slope
having a high density of Zamia obliqua to establish a permanent rectangular 0.5 ha
plot. Plot bisected Zamia population including the maximum number of individuals. Two
contiguous quadrants of 50 X 50 m were established, each quadrant was subdivided
into four sub-quadrants of 25 X 25 m. All Zamia individual was censured, permanent
marked and referenced. Information of each individual found as number of leaves,
caudex size, longitude of three new and two mature leaves and number of leaflets of
these five leaves. Distance between individuals and percentage of leaflets missed by
herbivory of the five selected leaves were recorded. Forest structure, canopy
openness, conservation state was added.
In Guachalito and Coquí, field exploration included walks through paths. When an
individual was sighted a circular plot of 25 m radius was established around it.
Information of each individual found such as caudex size, number of leaves, longitude
of longest leaf and number of leaflets, distance, and orientation between individuals
was recorded. Percentage of leaflets missed by herbivory of longest leaf and report of
the color of the new leaves was also included. Information about forest structure,
canopy openness, and conservation state of the forest was added. If a Zamia individual
was found outside the area another plot was placed contiguous to the previous one. Al
the plots were geo-reference and its topographic position noted
Zamia individuals were assigned to different size class based on caudex size (seedling,
presence of sarcotesta or/and one leaf, sapling 1-25 cm, juvenile 26- 99 cm, and > 1 m
adult). Previous literature report Z. obliqua reproductive state to plant higher than 1 m
(Stevenson 2001), nevertheless we have observed in backyard and garden shorter
cone Zamia plants (Benavides A.M. & Zuluaga J.S., pers. obs). Despite this
observation, we accord with the previous report because we do not have evidences for
individuals into the forest). Zamia group or patch is defined in this study as a group of
individual/s where each individual is far no more than 100 m.
During field exploration no reproductive individuals were sighted within the forest,
nevertheless in gardens and backyards at Coquí and Guachalito beach, were founded
individuals with female or male cones. Insect on male cone were collected. Additionally
Lycaenidae caterpillars present in Zamia leaves were monitoring from pupa to adult
state and butterflies preserved to entomological collection. Entomological and plants
collections were deposited at CIB and Herbaria Universidad de Antioquia respectively.
Numerical analysis
Pearson correlation was used test the degree of relation between caudex size and
morphological traits. To test for differences in forest and properties between forest age
categories or sites, we used analysis of variance (ANOVA).
The response of Zamia obliqua to environmental variables was explored using indirect
(Principal Component Analysis, PCA) and direct (Canonical Correspondence Analysis,
CCA) multivariate gradient analysis. In these, Zamia obliqua values were present in the
life-cycle stage abundance per group. In CCA, the independent variables were forest
(ordinal, flatland or mountain), slope scale (ordinal 1=0o , 2=1o -15o, 3=16 o-35o, 4=>36o)
population size (nominal, number zamias), slope (nominal), forest stage (ordinal,
secondary forest or primary forest), forest stage (nominal, ranked from 50 (secondary
forest) to 100 (primary forest). For the numerical analyses were performed using MVSP
(version 3.13) and CANOCO (version 4.0; ter Braak & Smilauer 1998). Multiple
distances calculations among Zamia individuals were carried out, after compute
geographic distances from latitude-longitude data, using the R package (Casgrain &
Legendre 2001).
Results
General Pattern
Zamia obliqua individuals reached in this study a maximum height of 4.5 m and beared
from 0 to 90 long recurved leaves (mean=7 n=836). Number of leaves, leaflest and
longitude of longest leaf increase according with the caudex height (Pearson
correlation 0.73, 0.65 and 0.59 respectively p<0.001). Five percentage of individuals
present herbivory in its longest leaf. Butterflies were determined as Eumaeus godortii
(Boisduval, 1870) Lycanidae.
Population distribution and structure
Zamia obliqua individuals tend to be aggregated; the 1010 individuals recorded in 107
plots (radius 25 m) were grouped in 26 patches or groups. The three larger groups
compass 68% of the individuals found and half of the groups have less than 10
individuals. Isolated individuals were uncommon (4 individuals) and adult plants were
more frequent in larger groups.
Population structure based on caudex size gave a reverse J-shaped curved for Zamia
obliqua (figure 2). Seedling and juveniles with 768 individuals (92%) dominate the
population profile adults (>1m caudex size) were uncommon.
Life cycle stages distribution related to environmental factors and distance
Zamia pattern show a shorter gradient length along first two PCA axes, indicating that
groups share life cycle-stage. Nevertheless Zamia patches tend to aggregated
according to the predominance of some of the life cycle-stages (figure 3A).
This pattern is more evident in CCA diagram, where each caudex size category is
closer to its contiguous category. Indicating, as the factors affect in different way each
caudex size category or each life cycle-stage. Analyzed by CCA, seedling was related
to group size and primary forest, as well as to low slope. On the other hand, others life
cycle stages could stand a more pronounced slope, less size group and secondary
forests (Figure 3B).
The abundance of Zamia varies greatly across the landforms. Nevertheless, larger
groups (177, 205 and 212 individual) were located in mature forest, no-flooded low-
lying areas and elevated areas with moderate slope. No plants were located in
mangroves and few individual were located close to a flooded area and elevated area
with extreme slopes.
Distance among individual in the same caudex size category differ from seedling to
juvenile and adult plants. (Table 1). Distance from seedling to juvenile increment to 11
m to 77-78 m and decrease to 25 m from juvenile to adult size class.
Acknowledgments
This study was supported by The Cycad Society Grant. Invaluable help, logistic support
and field work assistance was provided by people at the “El Amargal”, Guachalito and
community of Coquí. Dr Gerardo Lamas identified Lycanidae species (Director Museo
de Historia Natural Universidad Nacional Mayor de San Marcos).
References
Casgrain,P. and Legendre,P. (2001). The R Package for Multivariate and Spatial
Analysis. Département de sciences biologiques, Université de Montréal.
Figure 1. Map of Colombia showing the locations of the study sites at the
Tribuga Gulf, Pacific Coast Chocó Colombia.
Guachalito
El
Amargal
Coquí
5 km
Figure 2. Population structure showing Zamia obliqua individuals per caudex
size.
Figure 3A
Figure 3B
Table 1. Mean distance among Zamia obliqua individuals per each caudex size
category.
Post Hoc Test
Tukey B
a,b
7720 12.0341
3502 52.2444
362 58.0768
5219 76.7662
8776 77.4821
catecaudex
o cm
51-99 cm
> 100 cm
1-25 cm
26-50 cm
N123
Subset for alpha = .05
Means for groups in homogeneous subsets are displayed.
Uses Harmonic Mean Sample Size = 1435.515.
a.
The group sizes are unequal. The harmonic mean of the
group sizes is used. Type I error levels are not
guaranteed.
b.
Appendix
Table present the Zamia obliqua individuals in each caudex size category per group
Group 0 cm 1-25 cm 26-55 cm 56-99 cm > 99 cm
1 117 98 25 46 19
2 7 14 7 15 2
3 0 0 0 1 1
4 0 0 0 1 1
5 0 2 0 0 2
6 2 14 21 8 0
7 0 1 0 0 0
8 0 5 4 3 0
9 0 2 1 1 0
10 0 1 3 1 2
11 41 69 43 38 13
12 0 4 2 2 0
13 0 4 1 1 1
14 0 8 5 4 1
15 0 0 0 1 0
16 2 7 5 4 6
17 0 1 0 0 0
18 0 0 0 1 1
19 0 1 1 0 0
20 0 1 0 0 0
21 14 13 6 3 1
22 13 9 1 2 2
23 54 52 48 11 12
24 0 1 0 1 0
25 0 0 0 1 0
26 2 0 36 19 8