Download document - Dutch Caribbean | Biodiversity database

WAGENINGEN UNIVERSITY
Diet preference of roaming goats (Capra hircus) on columnar cacti in
Bonairian scrublands.
Thesis paper
Author: Nikkie van Grinsven*
*MSc. Student Forrest and Nature Conservation, Wageningen University, The Netherlands
Supervisors: Dr. Pim van Hooft, Dr. Milena Holmgren and Dr. Dolfi Debrot
Chairgroup: Resource Ecology Group
Duration: February-August
Research paper for Thesis speciality Ecology. MSc. programme Forest and Nature Conservation at
Wageningen University of Live Sciences. By Nikkie van Grinsven contact: [email protected]
Course code: REG-80436
Diet preference of roaming goats (Capra hircus) on columnar cacti
in Bonairian scrublands.
Author: Nikkie van Grinsven
Registration number: 89060729120
Supervisor STINAPA and project supervisor IMARES: Dr. Dolfi Debrot
Supervisors from Wageningen University: Dr. Pim van Hooft and Dr. Milena Holmgren
Date: 28-08-2015
Place of writing: Wageningen, The Netherlands
Research location: Washington Slagbaai National park, Bonaire
MSc. Programme Forest and Nature Conservation
I
Acknowledgements
I would like to thank the rangers and management staff at Washington Slagbaai National Park,
especially Henri for all the advice and help with logistics. Special thanks to Paulo Bertuol and Elise
Galitzki for the extra motion camera’s and for helping me with finding the right experimental locations.
I am forever grateful for my right hand man, Kevin Geurts; without him I could not have built even one
test pen. I would like to thank Pim, Milena and Dolfi for all the advice and laughs through Skype and
in the field, and Bart van der Hee for helping me with my report. Last, I would like to thank all people
who have made me feel very welcome in Bonaire.
II
Abstract
The introduction of goats to Bonaire in the 17th century, caused direct and indirect changes in both
terrestrial and marine ecosystems and has led to serious alterations in the intricate interplay of island
ecology. Direct effects such as the preference of goats for young and palatable plants has likely altered
the plant communities from a dry forest towards the current dominance of thorny bushes and cacti,
and overgrazing has led to cascading effects through soil erosion. The subsequent reduction of several
crucial ecological and economical functions such as local climate control and recreation, are starting
to deteriorate the ecosystem services that the inhabitants of Bonaire rely on, particularly with respect
to tourism; a major source of income for the island. After centuries of goat herbivory, columnar cacti
are some of the remaining species still present today on Bonaire. However, monitoring has provided
evidence that cacti endure severe damage by herbivory. Studies conducted in different geographical
areas have shown that goats commonly browse on cacti. Goat populations on Bonaire have remained
high in recent decades, often causing irreversible damage to columnar cacti.
In this study, I investigated the diet preference of goats with respect to three columnar cacti;
Subpilocereus repandus, Stenocereus griseus, and Pilosocereus lanuginosus, in relation to associational
resistance, spine characteristics and foraging by the green iguana. I assessed the dietary preference
through food choice experiments using choice options, such as with or without spines on cacti, with
or without Opuntia to interfere with accessibility, and evaluation of additive damage effect from the
native green iguana. The study consisted of a combination of field and enclosure experiments. Three
different field experiments were simultaneously conducted within Washington Slagbaai national park
during which diet preference, association resistance and additional browse damage were assessed.
Goats were free to enter and leave the experimental locations during these field experiments. In
enclosure experiments 18 goats were monitored in captivity during which the first 2 field experiments
were replicated and herbivory reduction due to cacti spines was added as third enclosure experiment.
Beforehand, I assessed the spine characteristics of the three columnar cacti, and I observed that a
higher density of spines was associated with a decline in spine thickness. This may indicate a trade-off
between shade (density of spines) and defensive traits (thickness). After that, Opuntia and spines were
tested as factors, and their presence was shown to have no reducing effect on goat herbivory. Field
experiments with indigenous Iguana showed no observations of cactus biomass consumption.
Therefore it is safe to assumed that the Iguana rarely if at all, browse on cacti, with the exception of
their fruit and flowers.
During all field and enclosure experiments goats showed a diet preference for S. repandus over other
cacti species, and this preference was not influenced by the removal of spines or the presence of
Opuntia. It is not yet clear why goats prefer this cacti over other species, but we speculate that taste
and its related nutritional value of this cacti species is higher. This study has shown that herbivory by
Bonairian goats on cacti is worrisome. A 60% rate of browsing damage was observed within 48 hours
of cacti placed in the field, and when restricted to compulsory enclosed conditions, 100% of available
cacti were damaged within 24 hours. The findings in this report provide quantitative evidence of cacti
consumption by Bonairian goats, which show severe goat herbivory on S. repandus and a cascading
effect when this cactus is gone. When the S. repandus has been eaten goats will move to the S. griseus
and P. lanuginosus. The large population of goats inhabiting Bonaire will lead to irreversible damage
if not reduced. Similar reports have shown the detrimental effects goat populations can have on
similar semi-arid ecosystems. I recommend a policy that combines the reduction or eradication of feral
goats within the park, together with intensive education among human communities to raise social
awareness and change goat husbandry practises on the island at large.
III
Table of Content
Acknowledgements................................................................................................................................. II
Abstract .................................................................................................................................................. III
List of Tables and Figures ........................................................................................................................ V
Introduction ............................................................................................................................................ 1
Main objective .................................................................................................................................... 4
Research questions ............................................................................................................................. 4
Methods .................................................................................................................................................. 5
Study area ........................................................................................................................................... 5
Research methods .............................................................................................................................. 6
Field experiments................................................................................................................................ 6
Experiment 1; The diet preference of goats for three columnar cacti. .............................................. 6
Experiment 2; Assessing associational resistance between cacti and Opuntia spp. .......................... 7
Experiment 3: Additional browse damage on columnar cacti by green iguana. ............................... 8
Enclosure experiments........................................................................................................................ 8
Experiment 3: Assessing spine protection against herbivory by goats............................................... 9
Results ................................................................................................................................................... 10
Factors of influence on cacti biomass consumption by goats; Opuntia and Spines. (field,- and
enclosure experiment 2 and enclosure experiment 3) ...................................................................... 11
Additive biomass consumption due to wild goats. (field experiment 3) ......................................... 12
The diet preference of goats for columnar cacti. (field and enclosure experiment 1) ..................... 12
Factors of influence on the diet preference of goats on cacti; Opuntia and Spines. (field,- and
enclosure experiment 2 and enclosure experiment 3) ...................................................................... 14
Discussion .............................................................................................................................................. 16
Synthesis & Conclusions ....................................................................................................................... 18
Implications and recommendations ................................................................................................. 18
References ............................................................................................................................................ 19
IV
List of Tables and Figures
Figure 1.1
Figure 1.2
Table 2.1
Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 3.1
Figure 3.2
Figure 3.3
Table 3.1
Table 3.2
Figure 3.4
Figure 3.5
Figure 3.6
Figure 3.7
Table 3.3
Pilosocereus lanuginosus (a), Subpilocereus repandus (b) and Stenocereus griseus (c).
Pictures by: Bihrman.com
Opuntia curassavica (a), Opuntia wentiana (b) and Opuntia elatior (c). Pictures by:
Bihrman.com
Schematic lay-out of goat/experiment division during enclosure experiments
Experimental locations in Washington Slagbaai national park
Subpilocereus repandus cover by Opuntia wentiana and Opuntia elatior. Picture by
Nikkie van Grinsven
Three test plots for each experimental location; (A) control pen, (B) test pen with access
iguanas, (C) test location with free access iguanas and goats
Cacti pieces with the treatment Spines present (a) and Spines absent (b). Pictures by
Nikkie van Grinsven
The number of spines per cm2 per cacti species. Line in box: median, box: 25th and 75th
percentiles, whiskers (Tuckey style). Boxplots not sharing the same letter are
significantly different. Kruskal-Wallis, N = 5, x2 = 92.365, df =2, p <0.001
Cacti characteristics per cacti species. (A) Thickness of spines per cacti species. (B) Spine
length per cacti species. Line in box: median, box: 25th and 75th percentiles, whiskers
(Tuckey style). Boxplots not sharing the same letter are significantly different. Boxplots
not sharing the same letter are significantly different. (A: Kruskal-Wallis, N = 45, p
<0.001; B: Kruskal-Wallis, N = 45, p <0.001) Boxplots not showing Q2 have no data
points between the lowest data point and the median
Biomass consumption per goat/day during enclosure experiments when cacti were
treated with the absence of spines (A), and the absence of Opuntia (B). Field
experiment when cacti were treated with the absence of Opuntia. Line in box: median,
box: 25th and 75th percentiles, whiskers (Tuckey style). Boxplots not sharing the same
letter are significantly different. Boxplots not showing Q1 and/or Q2 consisted for >25%
or >50% of zero values
Mean biomass consumption per day by goats and green iguanas, data expressed mean
+/- SEM
Kruskal-Wallis test between cacti species for biomass consumption during field and
enclosure experiments.
Biomass consumption per cacti species/day when exposed to goats in Washington
Slagbaai National Park. Line in box: median, box: 25th and 75th percentiles, whiskers
(Tukey style). Boxplots not sharing the same letter are significantly different. Boxplots
not showing Q1, Q2, Q3 and/or Q4 consisted for >25%, 50%, 75% or 95% of zero values.
Percentage of observations in which a cacti specie was preferred by browsing goats,
based on first biomass consumption during observations. Results are based on camera
images during the diet preference- and Opuntia trials in both field- and enclosure
experiments.
Biomass consumption per goat/day on columnar cacti during enclosure experiments,
cacti were treated by removal of the spines. Line in box: median, box: 25th and 75th
percentiles, whiskers (Tuckey style). Boxplots not sharing the same letter are
significantly different. Boxplots not showing Q1 and/or Q2 consisted for >25%, or 50%
of zero values.
Biomass consumption per goat/day on columnar cacti during enclosure experiments,
cacti were treated by the addition of Opuntia in the direct surrounding of the cacti. Line
in box: median, box: 25th and 75th percentiles, whiskers (Tuckey style). Boxplots not
sharing the same letter are significantly different. Boxplots not showing Q1 and/or Q2
consisted for >25%, or 50% of zero values.
Results of Kruskal-Wallis test between cacti species during field and enclosure
experiments. Cacti were treated by placing Opuntia in the direct surrounding of cacti
and the absence of Opuntia (control).
8
9
11
12
13
14
15
17
18
18
19
19
20
20
21
21
22
V
Introduction
Overgrazing by introduced herbivores can cause dramatic changes in the structure and functioning of
ecosystems, often threatening native species and reducing ecosystem services (Desender et al., 1999;
Peo et al. 2011; Campbell & Donlan, in press 2015). The loss of vegetation can have detrimental effects
on biodiversity and ecosystem resilience. Isolated ecosystems, such as colonized islands, are most
vulnerable to invasive herbivores due to a lack of natural predators (Clout, 2002; Cromarty et al. 2002;
Mulder, 2013).
Goats were introduced to Bonaire in the 17th century, during the colonization of the Caribbean islands
and now occupy large parts of the island (Cortes, 2012; Simal, 2005), causing direct and indirect
changes in both terrestrial and marine ecosystems (Baraza & Fernandez-Osores, 2013). The
preference of goats for young and palatable plants has likely altered the plant communities from a dry
forest towards the current dominance of thorny bushes and cacti (Skarpe et al, 2012; Cortes, 2012;
Campbell & Donlan, in press 2015).
Overgrazing by goats also has indirect, cascading effects through soil erosion. Soil erosion not only
reduces nutrient availability but also diminish water infiltration creating harsh conditions for plant
survival and recolonization (Peco, 2011). It also leads to a runoff of nutrients and silt into the ocean.
Runoff nutrients from land accumulate on coral reefs, stimulating algae overgrowth, causing
smothering by silt precipitation, and pollution (Walsh, 2011; Smith et al, 2014). The subsequent
reduction of several crucial ecological and economical functions such as local climate control and
recreation, are progressively deteriorating the ecosystem services that the inhabitants of Bonaire rely
on (Cortes, 2012; Smith et al, 2012), particularly with respect to tourism as the key sustainable source
of income for the island (TEEB, Caribbean Netherlands, 2012).
Despite the high goat density and signs of overgrazing, no systematic study has been conducted to
assess the amount of biomass consumed by goats on the Caribbean scrublands, which Bonaire is part
of. Columnar cacti are common native species on Bonaire and show severe signs of damage caused by
goats. There is general concern for the future of columnar cacti by the inhabitants of Bonaire. In
addition to the ecological value, cacti represent a large part of the cultural heritage (TEEB, 2012).
Columnar cacti are used today as fencing material, for food and as livestock fodder (Cortes, 2012).
Research in other arid ecosystems, such as the Galapagos, South America and India, has demonstrated
that if left unchecked, goats browse on cacti populations and cause loss of cacti (Solanki, 1994; Parkes
et al, 2002; Peco et al, 2011; Malo et al. 2011; Smith et al, 2014; Campbell & Donlan, in press 2015).
The dietary preference of Bonairian goats for certain cacti species above others is in part determined
genetically, in part due to a learning curve, and in part by environmental circumstances including the
relative availability of various plants (Solanki, 1994). The presence of heavy grazing has limited or even
stopped regeneration of many if not most (native) plant species for over three centuries, pressuring
the highly adaptable goats to forage on grazing resistant species such as columnar cacti (Solanki,
2000). Columnar cacti have cladodes, which are swollen water-storing stem segments that are
characterized by high moisture and low dry matter (Batista, 2003; Vieira, 2007). High temperature and
low precipitation force goats to graze on plants that are able to hold a lot of water such as columnar
cacti (Solanki, 1994, 2000). Adult columnar cacti are often ringed by browsing goats due to the removal
of the bark at the base of the cacti, after which cacti perish due to the lack of nutrients. Possible new
growth is often limited due to herbivory and the resulting lack of surviving seedlings. The presence of
unpalatable species such as Aristida and Opuntia in Bonairian scrublands, causes competition for light,
water and nutrition sources that cacti due to herbivory cannot withstand (Debrot and Freitas , 1993).
1
Columnar cacti are important for both animals and plants. They play an important role in the diets of
bats and birds during the dry season (Petit & Pors, 1996; Nassar et al, 1997) and often act as nurseplants to the establishment of cacti seedlings (Peeters et al, 2008). Goat herbivory affects the
condition of the columnar cacti populations as previously reported in studies done by Petit &
Freedman, and Muylaert. Their studies show that goat herbivory on columnar cacti artificially shifts
the start of the flowering season of these cacti to later in the year (Petit & Freedman, 1997;Muylaert
et al.2014). Consequently, affecting the bat populations who survive exclusively on these cacti for
resources and reproduction, which affects the resilience of cacti populations by reducing the number
of seedlings (Petit & Freedman, 1997; Petit, 2001, 2011; Muylaert et al., 2014).
Bonaire has three columnar cacti species, namely the Stenocereus griseus (Yatu), Subpilocereus
repandus (Kadushi) and Pilosocereus lanuginosus (Kadushi di Pushi). These three cacti species differ in
their spine morphologies which may influence the goats’ diet preference. Both P. lanuginosus and S.
repandus have thin flexible spines that mostly cover the topmost parts of the cacti, leaving a low
density of spines near the bottom of the cacti stem. However, the density of spines is higher on P.
lanuginosus than on S. repandus, which may result in preference for S. repandus. The S. griseus has
long, thick and hard spines evenly distributed over the surface of the cacti, suggesting that goats will
have the least preference for the S. griseus.
a
b
c
Figure 1.1 Pilosocereus lanuginosus (a), Subpilocereus repandus (b) and Stenocereus griseus (c). Pictures by:
Bihrman.com
The function of cactus spines is protection as well as a reduction in transpiration (Anderson, 2001).
The protective function of spines has been confirmed in a previous cafeteria experiment where the
feeding behaviour of captive goats was assessed by offering twigs with- and without spines (Skarpe et
al, 2012). The study of Skarpe suggests that the presence of spines reduces direct damage by goats
(Skarpe et al, 2012). The susceptibility of cacti to herbivory can be decreased by certain morphological
characteristics of neighbouring plants, such as spines, toxins, odours or shade, causing the goats to
reject or fail to locate its normal fodder (Atsatt & O’ Dowd, 1976; Jezorek, 2011). This form of
interaction is referred to as associational resistance (Barbosa et al. 2009). The presence of unpalatable
plants, like the cacti species Opuntia, in the direct surroundings of the cacti could offer protection
against grazing. Opuntia is often found near columnar cacti and known for its barbed spines, therefore
I hypothesise that associational resistance could occur between columnar cacti and Opuntia.
2
There are three species of Opuntia (Opuntia curassavica, O. elatior and O. wentiana) on Bonaire (fig
1.2). The leaf-like body of the Opuntia spp. is covered with barbed spines called glochids and larger
spines (González-Espinosa & Quintana-Ascencio, 1986). When in contact with mammals these larger
spines bury themselves into the host, causing a deposition of Opuntia fragments throughout the
ecosystem. The protection of focal plants by neighbouring plants with spines such as the Opuntia is a
clear example of associational resistance (Barbosa et al. 2009).
a
b
c
Figure 1.2 Opuntia curassavica (a), Opuntia wentiana (b) and Opuntia elatior (c). Pictures by: Bihrman.com
Vertebrate nectarivores, frugivores and omnivores are supported by the large number of flowers and
seeds that are produced by columnar cacti during the dry season (Petit & Pors, 1996). Cacti response
to damaging by herbivores leads to a reduced number of reproductive parameters (Peco et al. 2011).
Goats often remove spines and this could potentially facilitate browsing by other herbivores. The most
common native herbivore of Bonaire is the green iguana (Iguana iguana). Iguanas feed mostly on new
shoots, leaves, flowers and soft fruits. (Boer de, 2001; Buurt, 2006; Petit, 2011). Literature states that
even though there is no evidence of iguana eating cactus leafs in the wild, in captivity spineless cacti
are often presented and eaten (Marken Lichtenbelt van, 1993; Green iguana society, 2009)
In this study, I investigate goat predation on columnar cacti by conducting a combination of field and
enclosure experiments. First, the foraging behaviour of goat populations on Bonaire was analysed, to
ensure that this particular population actually consumes columnar cacti. Next, the characteristics of
the spines from the columnar cactus species were documented and the possible associational
resistance between Opuntia and cacti was studied. Last, the endemic iguana was included in the
experiments to assess whether this species forages on columnar cacti as well, and could cause
additional damaging effects.
3
Main objective
The diet preference of goats with respect to columnar cacti in relation to associational resistance,
spine characteristics and foraging by the green iguana were investigated.
Objective:
1. Assess a possible diet preference of goats for three columnar cacti.
2. Determine if the presence of spines reduced goat herbivory on cacti and/or influence their
diet preference.
3. Examine if a possible associational resistance between Opuntia and columnarcacti has effect
on goat herbivory on cacti and/or their diet preference.
4. Investigate a possible additive damage effect by the green iguana and goats on columnar cacti.
Research questions
Main question:
What is the diet preference of goats on three columnar cacti in Bonairian scrublands and which
factors mediate this?
Sub questions:
1. Do goats have a dietary preference for certain columnar cactus species?
2. Does the presence of spines reduce goat herbivory on columnar cacti?
b. Does the presence of spines influence the diet preference of goats?
3. Is goat herbivory on columnar cacti reduced by an associational resistance due to the
presence of Opuntia?
b. Does the associational resistance between Opuntia and columnar cacti influence the
diet preference of goats.
4. Does the presence of green iguanas and goats in Bonairian scrublands have an additive
damage effect on columnar cacti?
Hypothesis and predictions
The main hypothesis is that herbivory by goats on columnar cacti is reduced by cactus spines and the
presence of Opuntia in the direct surrounding of these cacti (up to 10 x 15 cm). Without spines goats
no longer have a preference for specific columnar cactus species. A second hypothesis is that goats
will cause an increase in iguana herbivory on columnar cacti, by removing cacti spines and thus
providing access to cacti stems.
Predictions:
1. Due to the morphology of spines, goats will prefer the Subpilocereus repandus (Kadushii),
followed by the Pilosocereus lanuginosus (Kadushii di Pushi) and as last the Stenocereus
griseus (Yatu).
2. The higher the density and thickness of spines on cacti, the lower goat herbivory on
columnar cacti.
b. Goats have no diet preference among spineless columnar cacti.
3. Opuntia in the direct surrounding of cacti reduces goat herbivory.
b. The associational resistance between Opuntia and cacti will not influence goat diet
preference.
4. The presence of both green iguanas and goats in Bonairian scrublands will increase iguana
herbivory on columnar cacti.
4
Methods
Study area
Bonaire, with an area size of approximately 288 Km2, is the third largest of six islands known as the
Dutch Caribbean. It is located 80 Km from the South-American coast of Venezuela, has a semi-arid
climate with a mean annual temperature of 28°C, and a mean annual rainfall of 463 mm (Buurt, 2006).
Bonaire has a semi-arid climate and knows a dry season (March – June) and a rainy season (peak at
October and November). The study was conducted during the start of the dry season, February-May
2015, in the north of Bonaire at Washington Slagbaai National Park, situated at 12° 16' 2 N and 68° 22'
58 W (Google maps, 2015). The area has a maximum elevation of 241 m above sea level and its
perimeter is approximately 9 km with an additional 14 km coastal zone (Simal, 2005). The vegetation
of the park is dominated by xerophyllic scrub of variable compositions dependent on soil composition
and trade winds (Buurt, 2006). Trade winds are characterized by their perseverance and high salt
contents. Soil composition tends to be poorer in the plains and limestone terraces and richer in areas
of volcanic origin (Simal, 2005).
Figure 2.1 Experimental locations in Washington Slagbaai national park.
5
Research methods
The study consisted of a combination of field and enclosure experiments. The field experiments were
conducted on various locations between February and April within Washington Slagbaai national park.
Based on local expert judgement, twelve locations were opportunistically selected across Washington
Slagbaai National Park in areas with similar ground-, shrub- , and tree cover which were located in the
proximity of existing roads (Local expert: MSc. Paulo Bertuol). These locations were representative for
Washington Slagbaai National Park (Fig. 2). The enclosure experiments were conducted in March on
experimental location 11, whereby the goats were monitored in captivity (fig. 2.1) . Eight test pens,
two of them control pens, were built in which a total of 18 goats were monitored for 48 hours, whereas
six goats where monitored for 96 hours for the duration of all three experiments (Table 2.1).
Table 2.1 Schematic lay-out of goat/experiment division during enclosure experiments
Day 1
Day 2
Day 3
Day 4
Experiment 1
Goat 1,2,3
Goat 4,5,6
Goat 7,8,9
Goat 10,11,12
Experiment 2
Goat 4,5,6
Day 5
Goat 10,11,12
Goat 7,8,9
Goat 1,2,3
Day 6
Goat 7,8,9
Goat 10,11,12
Goat 10,11,12
Day 7
Goat 13,14,15
Goat 16,17,18
Goat 7,8,9
Day 8
Goat 16,17,18
Goat 13,14,15
Experiment 3
Control 2
Field experiments
Three different field experiments were simultaneously conducted, and lasted from February until
April. During these field experiments goats were free to enter and leave the experimental locations.
The experiments were conducted on marked test plots within the 12 experimental locations (Fig 2.1).
During field experiment 1, I studied goat diet preference for columnar cacti. During field experiment
2, I studied if there is a possible associational resistance between Opuntia and cacti that could affect
goat herbivory. Last, during field experiment 3 the additional browse damage on columnar cacti by
green iguana was assessed.
Experiment 1; The diet preference of goats for three columnar cacti.
I compared the goat diet preference for three columnar cacti using a completely randomized onefactor design, with three factorial levels. The factor was the species designation and the levels of the
factor the species, P. lanuginosus, S. repandus and S. griseus. The response variable of this experiment
was the amount of cacti biomass consumed by goats within 48 hours. The biomass consumption was
calculated by weighing the cacti before and after each experimental study. Each experimental study
lasted for 48 hours. The number of goats entering the test plot as a co-variable was included, this was
monitored by monitoring cameras (brand: Relux; settings, Medium sensitivity, 10 pictures when
triggered, one picture every 12 hours). One camera was placed at the corner of each test location.
Prior to the start of the trial, a total of three cacti pieces, one of P. lanuginosus, S. repandus and S.
griseus, were randomly placed within each test plot for 48 hours. The pieces of cacti were
approximately 50 cm long and estimated by local expertise to be between 2 and 5 years old; this
estimation was based on the assumption that size and cactus age are positively correlated (Martinez
del Rio et al. 1995). The cacti pieces were collected outside Washington Slagbaai National Park. To
ascertain the accuracy of biomass consumption measurements, a control plot was placed within each
experimental location. The test,- and control plot within each experimental location was 3 x 3 meters
and had an iron pole on each corner. The plots were built approximately 5 meters from each other in
similar conditions. The control plot corrected for water evaporation, by measuring the average weight
loss per experimental study at each experimental location and subtracting this from the biomass loss
measured from the cacti pieces exposed in the corresponding test plots. The control plot was enclosed
by the addition of chicken fencing (1.20 meter high) on each side and on top, preventing goats and
green iguanas from entering. The test plot was an open area marked by iron poles where goats could
freely enter.
6
Experiment 2; Assessing associational resistance between cacti and Opuntia spp.
To assess if a possible associational resistance between Opuntia and cacti has effect on goat herbivory
of cacti, one fully cross-factored, two-factored design, was conducted. The first factor was the species
designation and the levels of the factor were the species, P. lanuginosus, S. repandus and S. griseus.
The second factor was the occurrence of Opuntia in the direct surrounding of the cacti and the levels
of this factor were Opuntia present and Opuntia absent. The cacti with the treatment of Opuntia, were
completely surrounded by me with Opuntia plants, creating an overall thickness of Opuntia cover
between 10 and 15 cm (Fig. 2.2). The Opuntia plants used during this study were Opuntia wentiana
and the Opuntia elatior, these plants were chosen because of their abundance; and the Opuntia
curassavica was excluded because this species has a maximum height of 15 cm and has a protected
status in Bonaire. The response variable was the consumption of biomass and the co-variable the
number of goats entering the test plots. The methods of experiment 2 are the same as previously
described at experiment 1. The experiment was conducted at the same 12 locations as experiment 1
(Fig. 2.1).
Figure 2.2 Subpilocereus repandus cover by Opuntia wentiana and Opuntia elatior. Picture by Nikkie van
Grinsven
7
Experiment 3: Additional browse damage on columnar cacti by green iguana.
In the third experiment, experiment 1 was replicated with the green Iguana as additional herbivore
in order to assess additional browse damage on columnar cacti. The replicate was similar in
experimental design and was simultaneously assessed (February-March). However, there was a
modification in the number of test plots on each experimental location, with the addition of one extra
test plot. The additional plots for the third experiment were constructed by placing four iron poles on
the corners of a 3 by 3 meter square and attaching chicken wire (height 1,20) to the iron poles 20 cm
above ground (Fig.2.3). By doing so, the test location excluded goat herbivores, but enabled access
for green iguanas. The response variable was the consumption of biomass and the co-variable was the
number of green iguanas entering the test location. The consumption of biomass was calculated as
previously described in experiment 1. The number of iguanas entering the experimental locations was
monitored by monitoring cameras (brand: Relux; settings, High sensitivity, 10 pictures when triggered,
one picture every 12 hours). The third experiment was conducted at the same 12 experimental
locations as previously described. The treatments were similar as described in experiment 1.
A
B
C
Figure 2.3 Three test plots for each experimental location; (A) control pen, (B) test pen with access iguanas, (C)
test location with free access iguanas and goats.
Enclosure experiments
Enclosure experiments exclude field factors such as multiple food sources, social interaction, and goat
density. Therefore I included enclosure experiments in my research to assess the per capita effect of
Bonairian goats on columnar cacti. Three different enclosure experiments were conducted in March
to assess goat herbivory on cacti. Enclosure experiment 1 and enclosure experiment 2 were replicates
of field experiment 1 (diet preference) and 2 (associational resistance). Enclosure experiment 3 was
added later on and assessed goat herbivory on spineless columnar cacti.
Animals
In April 2015 eighteen goats (+/- 25 kg of body weight) were used during enclosure experiments to
collect data on diet preference of individual goats and the influence of associational resistance
hereupon. These goats were previously captured within the borders of Washington Slagbaai national
park as part of the local goat removal project, and placed in an enclosed facility.
During the enclosure experiment three times six goats were separated in individually placed test pens
(3 x 3 metres) for a maximum duration of 48 hours per study. One group of six goats was used for two
studies, and by doing so a sample size of 12 was created for each experiment (Table 2.1). All goats had
ad libitum access to 1.5 m2 shade, 10 litre of water and 5 kilo of cacti. For each individual goat the
biomass consumption was assessed by replicating field experiment 1 (diet preference) and 2
(associational resistance). In each enclosure a motion camera (brand: Relux, one picture every 5
minutes) was added to monitor foraging behaviour. The experiment was conducted on location 11
(fig. 2.1) for the duration of three weeks, after 48 hours goats were placed back in the main enclosure
to be removed from the park.
8
Experiment 3: Assessing spine protection against herbivory by goats.
To assess the effect of spines on herbivory by goats on three different columnar cacti, one fully crossfactored, two-factored design, with three levels for the factor was conducted in March during an
enclosure study. The first factor was the columnar cacti and the levels of the factor were the species,
P. lanuginosus, S. repandus and S. griseus. The second factor, was the occurrence of spines on cacti
and the levels of this factor were Spines present and Spines absent. The cacti that had received the
treatment of spines absent, were completely stripped of their spines which resulted in bark damage.
Therefore, cacti that received the treatment of spines, were partially stripped of bark to ensure an
equal amount and type of bark damage (fig. 2.4). The response variable was the consumption of
biomass every 24 hours. The biomass consumption was calculated by weighing the cacti before and
after each experimental study. Each experimental study lasted for 24 hours.
a
b
Figure 2.4 Cacti pieces with the treatment Spines present (a) and Spines absent (b). Pictures by Nikkie van
Grinsven
To ascertain the accuracy of biomass consumption measurements, a control pen was added for each
enclosure experiment in which no goat was placed, control cacti pieces were weighed before and after
each study thus correcting for possible water evaporation of cacti. Prior to the start of the trial, a total
of three cacti pieces, one of P. lanuginosus, S. repandus and S. griseus, were randomly placed within
each test pen for 24 hours. The pieces of cacti were approximately 50 cm long and between 2 and 5
years old. The test and control pens were 3 x 3 meters and had an iron pole on each corner.
Statistics
All data were tested for normality (Shapiro-Wilk) and equality of variance (Levene) to ensure that
parametric test assumptions were not violated. Most data showed no normality or equality in
variances, for consistency all tests were done non-parametric. Kruskal-Wallis tests were used to
determine if the biomass consumption by goats of cacti species were significantly different from one
another and if the treatments were significantly different from the means. The Alpha level was 0.05.
Results were expressed in boxplots or tables. Outliers were defied based on the three sigma rule by
QI Macros 2015 (Word office 2015). Results in tables will be expressed as mean +/- standard error of
the mean (SEM). Statistical analysis was performed using SPSS (IBM Statistics, v.21.0)
9
Results
Spine characteristics of three columnar cacti.
Number of spines per cm2
Spine density differed among cacti species (Fig 3.1, Kruskal-Wallis, N = 5, x2 = 92.365, df =2, p <0.001.),
as did spine thickness (Fig. 3.2A, Kruskal-Wallis, x2 = 92.365, df =2, p <0.001), and spine length (Fig.
3.2B, Kruskal-Wallis, N = 45, p <0.001). P. lanuginosus has a higher density of spines per square cm2
(median: 15.4 spines/cm2) than S. repandus and the S. griseus (5.1 spines/cm2, 1.5 spines/cm2). The S.
repandus has significant more spines than the S. griseus (Tukey, Mean difference=3.9140, p=0.015).
It is clear from Figure 3.1 that P. lanuginosus showed the largest variation in the measured number of
spines per cm2 when compared to that S. griseus and S. repandus.
A
18
16
14
12
10
8
6
4
2
0
p<0.001
C
B
Pilosocereus
lanuginosus
Stenocereus griseus
Subpilocereus
repandus
Figure 3.1 The number of spines per cm2 per cacti species. Line in box: median, box: 25th and 75th percentiles,
whiskers (Tuckey style). Boxplots not sharing the same letter are significantly different. Kruskal-Wallis, N = 5,
x2 = 92.365, df =2, p <0.001.
From Figure 3.2 A and B, it is clear that the S. griseus has the highest median for thickness of spine
(Fig.3.2A, 0.11 cm), but the lowest density of spines per square centimetre. The S. repandus whose
spines are approximately two times smaller than those of the S. griseus (Fig.3.2A, 0.05 cm) (KruskalWallis, x2 = 63.288, df =1, p <0.001), has the highest median of spine length, but not significantly so
when compared to S. repandus (Tukey, mean difference= 11.479, p=0.251). When I compare Figure
3.1 with Figure 3.2A, I can assume that a high density of spines is associated with a lowspine thickness.
7.4
A; p<0.001
B; p<0.001
B
0.25
0.2
0.15
0.1
B
6.4
C
A
0.05
spine length (cm)
thickness of spines (cm)
0.3
B
5.4
4.4
A
3.4
2.4
1.4
0
0.4
P. lanuginosus
S. griseus
S. repandus
P. lanuginosus
S. griseus
S. repandus
Figure 3.2 Cacti characteristics per cacti species. (A) Thickness of spines per cacti species. (B) Spine length per
cacti species. Line in box: median, box: 25th and 75th percentiles, whiskers (Tuckey style). Boxplots not sharing
the same letter are significantly different. Boxplots not sharing the same letter are significantly different. (A:
Kruskal-Wallis, N = 45, p <0.001; B: Kruskal-Wallis, N = 45, p <0.001) Boxplots not showing Q2 have no data
points between the lowest data point and the median.
10
Factors of influence on cacti biomass consumption by goats; Opuntia and Spines.
(field,- and enclosure experiment 2 and enclosure experiment 3)
During enclosure experiments, the amount of cacti biomass consumption per goat per day was not
significantly affected by spine removal, or the addition of Opuntia (Fig. 3.3, enclosure experiments).
During the field experiments, a pairwise test shows that the addition of Opuntia seems to have an
almost significant effect (Fig. 3.3, Friedman’s test; p=0.058). When I compare the medians of the cacti
biomass consumption during the Opuntia experiments, results show that during enclosure
experiments approximately two times more cacti biomass is consumed than during field experiments
(Enclosure exp. 508.5 grams Opuntia absent; Field exp.: 284.8 grams Opuntia absent).
Enclosure experiments
Field experiment
Biomass consumption (grams)
3500
3000
A; p = 0.650
B; p = 0.749
C; Friedman's test:
p =0.058
2500
2000
1500
1000
500
0
Opuntia absent
Opuntia
Spines absent
Spines
Opuntia absent
Opuntia
Figure 3.3 Biomass consumption per goat/day during enclosure experiments when cacti were treated with the
absence of Opuntia (A; Mann-Whitney test, Z=-.454, P = 0.650), and the absence of Spines (B; Mann-Whitney
test, Z=-.319, P = 0.749). Field experiment when cacti were treated with the absence of Opuntia (C; Friedman’s
test, χ2= 3.6, P=0.058). Line in box: median, box: 25th and 75th percentiles, whiskers (Tuckey style). Boxplots
not sharing the same letter are significantly different. Boxplots not showing Q1 and/or Q2 consisted for >25%
or >50% of zero values.
11
Additive biomass consumption due to wild goats.
(field experiment 3)
No biomass consumption was found in one of the 12 reptile pens located on the 12 experimental
locations (Table 3.1). Iguana and blue whiptail lizards were frequently detected inside the reptile plots,
with the use of motion cameras. Nevertheless no biomass was lost and no bite marks where found.
Table 3.1 Mean biomass consumption per day by goats and green iguanas, data expressed mean +/- SEM
Treatment
Control
Opuntia treatment
non-native herbivore
Wild goat
201.9 ± 68.88
native herbivore
Reptiles incl. Iguana
No observations
N
12
68.54 ± 34.11
No observations
12
The diet preference of goats for columnar cacti.
(field and enclosure experiment 1)
During the field experiment, 26.4% of the cumulative cacti biomass used in test plots were browsed
by goats. The proportion of biomass consumption differing significantly among cacti species (KruskalWallis, x2 = 9.845, df = 2, p = 0.007). Post hoc tests showed that the S. repandus had a significantly
higher proportion of biomass consumption than the S. griseus and the P. Lanuginosus, indicating a
preference (Table 3.2). During the enclosure experiment, 78.4% of the cumulative cacti biomass used
in test plots showed browse damage, an increase of 52% compared to the field experiment.
Nevertheless, data shows similar significant results in the diet preference of goats (Table 3.2). During
both field and enclosure experiments, S. repandus was the preferred species, and S. repandus was
preferred secondary (fig, 3.4). However, in no experiment a significant difference in preference was
found between P. lanuginosus and S. griseus (field exp.: Kruskal-Wallis, X2=0.871, P=0.351; Encl. exp.:
Kruskal-Wallis, X2=0.807, P=0.369)
Table 3.2 Kruskal-Wallis test between cacti species for biomass consumption during field and enclosure
experiments.
(I)Cacti species
(II)Cactus species
Field experiment
X
Pilosocereus lanuginosus
Stenocereus griseus
Subpilocereus repandus
Stenocereus griseus
Subpilocereus repandus
Pilosocereus lanuginosus
2
0.871
5.132
7.407
Enclosure experiment
P-value
X2
P-value
0.351
0.023*
0.006*
0.807
7.671
10.600
0.369
0.006*
0.001*
When I compare the median of cacti biomass consumed by goats, results show that the median of the
S. repandus (Figure 3.4; field exp.: 192.5 gram, enc. exp.; 634 gram) was at least four times larger than
that of the P. lanuginosus or S. griseus (field exp.: 0 gram; 0 gram, encl. exp.: 21 gram;134 gram).
Equally, the number of sampled cacti pieces that showed browse damage was notably higher with
pieces of the S. repandus (field exp.: 45.8%; encl. exp.: 100%) (Wilson's Score interval: N = 24, 95 CI:
30-70%; N = 18, 95 CI: 80-100%) when compared with those of the P. lanuginosus or S. griseus (field
exp.: 8.3%; 25%, encl. exp.: 50%; 70%). The most common order in which cacti were approached was
in both experiments similar, the P. lanuginosus and S. griseus were approached after the complete
consumption of the S. repandus (Fig. 3.5). The S. repandus was approached in 89% (Wilson's Score
interval: N = 28, 95 CI: 80-100%) of the cases first, the S. griseus second 71% (Wilson's Score interval:
N = 28, 95 CI: 35-85%) of the cases second and the P. lanuginosus was in 96% (Wilson's Score interval:
N = 28, 95 CI: 80-100%) of the cases approached third or not approached at al.
12
A: Field experiment
B: Enclosure experiment
Biomass consumption (grams)
1600
B; p=0.002
A; p=0.007
1400
B
1200
1000
B
800
600
A
A
A
400
A
200
0
P. lanuginosus
S. griseus
S. repandus
P. lanuginosus
S. griseus
S. repandus
Figure 3.4 Biomass consumption per cactus species/day when exposed to goats in Washington Slagbaai
National Park. Line in box: median, box: 25th and 75th percentiles, whiskers (Tuckey style). Boxplots not sharing
the same letter are significantly different. Boxplots not showing Q1, Q2, Q3 and/or Q4 consisted for >25%,
50%, 75% or 95% of zero values.
4%
7%
Pilosocereus
lanuginosus
Stenocereus
griseus
Subpilocereus
repandus
89%
Figure 3.5 Percentage of observations in which a cactus species was preferred by browsing goats, based on
first biomass consumption during observations. Results are based on camera images during the diet
preference- and Opuntia trials in both field- and enclosure experiments.
13
Factors of influence on the diet preference of goats on cacti; Opuntia and Spines.
(field,- and enclosure experiment 2 and enclosure experiment 3)
The diet preference of the goats was not significantly affected by the presence of spines (Fig. 3.6; Ttest, P. lanuginosus: p=0.543; S. griseus: p=0.997; S. repandus: p=0.952) or Opuntia (Fig 3.7, table 3.3)
when compared to the control group .
Spines absent
Spines
1800
AB
Biomass consumption (grams)
1600
B
1400
B
1200
AB
1000
800
A
600
400
A
200
0
P. lanuginosus S. griseus
S. repandus P. lanuginosus S. griseus
S. repandus
Figure 3.6 Biomass consumption per goat/day on columnar cacti during enclosure experiments, cacti were
treated by removal of the spines. Line in box: median, box: 25th and 75th percentiles, whiskers (Tuckey style).
Boxplots not sharing the same letter are significantly different. Boxplots not showing Q1 and/or Q2 consisted
for >25%, or 50% of zero values.
Opuntia
Opuntia absence
1800
Biomass consumption (gram)
C
C
1600
1400
1200
AB
1000
800
B
600
A
400
200
A
0
P. lanuginosus
S. griseus
S. repandus P. lanuginosus
S. griseus
S. repandus
Figure 3.7 Biomass consumption per goat/day on columnar cacti during enclosure experiments, cacti were
treated by the addition of Opuntia in the direct surrounding of the cacti. Line in box: median, box: 25th and 75th
percentiles, whiskers (Tuckey style). Boxplots not sharing the same letter are significantly different. Boxplots
not showing Q1 and/or Q2 consisted for >25%, or 50% of zero values.
14
Table 3.3 Results of Kruskal-Wallis test between cacti species during field and enclosure experiments. Cacti
were treated by placing Opuntia in the direct surrounding of cacti and the absence of Opuntia (control).
(I)Cactus species
(II)Cactus species
Field experiment
X
Pilosocereus lanuginosus
Stenocereus griseus
Subpilocereus repandus
Pilosocereus lanuginosus
Stenocereus griseus
Subpilocereus repandus
2
Control
2
P-value
X
0.482
0.230
0.006*
0.871
5.132
7.407
Stenocereus griseus
Subpilocereus repandus
Pilosocereus lanuginosus
0.495
1.440
7.407
Stenocereus griseus
Subpilocereus repandus
Pilosocereus lanuginosus
Enclosure
experiment
5.737
0.017*
4.483
0.034*
13.522
0.000*
P-value
0.351
0.023*
0.006*
Control
0.807
7.671
10.600
0.369
0.006*
0.001*
15
Discussion
To understand the negative effects of goat herbivory on Bonaire on columnar cacti, an assessment of
goat diet preference with respect to columnar cacti in relation to associational resistance, spine
characteristics, and foraging by the green iguana was performed.
Cactus spines turned out to be different in size, thickness and density among the three columnar
cactus species, indicating different strategies for shade and herbivory protection. Although many
cactus species produce structural defences in the form of spines to deter feeding by herbivores (Kohl
et al., 2015), most spines of columnar cacti have evolved to provide shade. Furthermore, studies on
the defensive traits of plant spines have produced equivocal results (Kohl et al., 2015; Martorell, 2006;
Skarpe, 2012). I observed that a higher density of spines was associated with a decline in spine
thickness which may indicate a trade-off between shade (density of spines) and defensive traits
(thickness).
During the enclosure experiments, the presence of spines or Opuntia showed no direct effect on the
amount of biomass consumed by goats. Although this finding was unexpected, literature suggests that
goat forage choices are made not primarily on basis of the physical traits of plants, but also on the
basis of nutritional properties by simple sensory cues (Ginane et al., 2005; Alonso-Díaz et al. 2007;
Jansen et al. 2007; Ramírez-Orduña et al. 2008). If I assume as suggested by literature, that the
amount of biomass consumption by goats is not influenced by defensive traits, but based on the
nutritional value of fodder. Than that would explain why goats during the environmental experiments
would consume the same amount of cactus biomass regardless of spines or reduced accessibility due
to the presence of Opuntia. Cacti have become common ruminant feeds in several regions of the world
due to their high biomass yield during dry seasons, suggesting some nutritional value (Batista et al.
2003; Vieira et al. 2008). Nevertheless research has demonstrated that cacti are low on dry matter,
sodium and phosphor, and high in potassium (Vieira et al. 2008). High cell wall content and lignin
contribute to the relatively low food quality of these columnar cacti and lead to a low nutrient intake
by goats, limiting the productivity of goats (Forbes 2007; Ramírez-Orduña et al. 2008). Therefore the
nutritional value of cacti are considered low and goats will search additional food, separately of cacti
to fulfil all their dietary needs (Cerrillo et al. 2006).
Another explanation would be, that during enclosure experiments goats depended solely on cacti for
their daily nutrition. They were thus forced to browse on columnar cacti regardless of spines or
Opuntia, which probably explains why similar amounts of cacti biomass were consumed during the
enclosure experiments, regardless of spines or Opuntia. Therefore I would suggest that goats who eat
approximately 1004 grams of dry fodder a day (Raats, 1998), will eat the same amount of cacti if no
other fodder is available regardless of defensive traits.
During the field experiments, there is evidence, albeit weak, that the presence of Opuntia reduced the
amount of biomass consumed by goats. The negative effect of Opuntia on goat foraging has been
confirmed by field research, during which the relation between goat density and Opuntia density was
assessed (Geurts, in press). The study of Geurts shows, that when the number of Opuntia reaches a
certain density the number of goat will decline. This is consistent with my experimental results that
suggest that the associational resistance between cacti and Opuntia will reduce the amount of cacti
biomass consumed by goats. Therefore I assume that associational resistance between cacti and
Opuntia will reduce goat herbivory and a higher sample size during field experiments will lead to
stronger evidence.
Interestingly, during all field and enclosure experiments goats showed a diet preference for S.
repandus over other cactus species, and this preference was not influenced by the removal of spines
or the presence of Opuntia. It is not clear why goats prefer the S. repandus over other cacti species.
Nevertheless, my experimental results are consistent with the natural patterns of goat damage found
in Washington Slagbaai national park. Field research done by Ende van de, demonstrate a strong
preference for the S. repandus (Ende van de, in press). Columnar cacti are often dominant in very arid,
16
warm desert regions in which they have important ecological roles supporting numerous pollinators,
seed dispersers and frugivorous animals (Williams, 2014). Large amounts of water and other resources
are stored by columnar cacti in able to thrive in such stressful desert environments (Williams, 2014).
Additional research by me, based on water evaporation, found no significant differences in the water
quantity of the cacti. Therefore I assume that the water quantity of the cacti is not responsible for the
observed diet preference. It is possible that the nutritional value of S. repandus is higher in comparison
to S. griseus and P. lanuginosus. Interestingly, this seems to be supported by the human consumption
of the S. repandus in local cuisine. Studies of flavour preferences and aversions suggest that flavour
perception may be linked to the nutritional value of food (Taylor et al, 2004; Goff and Klee, 2009).
Currently, no studies have been done on the nutritional and related flavour characteristics of the cacti
species in this study.
Field experiments with indigenous Iguana showed no observations of cacti biomass consumption.
Therefore it is safe to assume that the Iguana do rarely, possibly even not at all, browse on cacti, with
the exception of their fruit and flowers (Marken Lichtenbelt van, 1992). Which in hindsight is not
surprising considering that the convoluted digestive tract of the green iguana cannot effectively digest
cacti due to their high cell wall and lignin contents (Marken Lichtenbelt van, 1992, 1993; RamírezOrduña et al., 2008).
The order in which cacti are consumed suggest an cascading effect. In almost 89% of the cases S.
repandus was completely consumed after which goats approached P. lanuginosus and S. griseus. The
S. griseus was approached second during all experiments, but not significantly so, with the exception
of the enclosure experiment where Opuntia was present. During enclosure experiments
approximately two times more cacti biomass was consumed than during field experiments, this is most
likely due to the restricted foraging possibilities during the experimental experiments. These lack of
external factors led to sufficient statistical power to detect the diet preference of the S. griseus second
to the S. repandus. Additional power calculation suggest that external factors, such as other fodder
options during the field experiment, led to insufficient power to detect a significant preference of the
S. griseus second to the S. repandus. Therefore the lack of significant evidence during field
experiments should be contributed to methodological restriction. This explanation is consistent with
the pattern revealed by the camera images, who show that in 71% (Wilson's Score interval: N = 28, 95
CI: 35-85%) of the cases the S. griseus was consumed after S. repandus, rendering the P. lanuginosus
third choice.
The experiments presented here show that Bonairian goats pose a serious threat to columnar cacti in
current conditions, especially the S. griseus seems to suffer the most damage. Within 48 hours 60 %
of all S. repandus pieces showed browse damage and when diet options were restricted to cacti, this
increased to 100% within 24 hours. My experimental results are consistent with the damage patterns
found in the Bonairian scrublands by the research of Ende van de (Ende van de, 2015). The presence
of S. repandus seems to have protected S. griseus and the P. lanuginosus from most of the herbivory
for now. However, these results indicate that when S. repandus disappears from the island, S. griseus
and P. lanuginosus will most like be subject to increased herbivory by goats.
17
Synthesis & Conclusions
The aim of this study was to assess the diet preference of goats for columnar cactus species in relation
to associational resistance and spine characteristics. This chapter synthesises the results obtained
from this research and previous literature. The main findings of this research can be summarized as
follows.
1. Bonairian goats prefer S. repandus first, S. griseus second and P. lanuginosus last.
a. During all experiments the consumed biomass of S. repandus was significantly higher
than that of S. griseus and P. Lanuginosus.
b. Even though the S. griseus was not significantly preferred above the P. lanuginosus,
camera observations and additional power calculations suggest that this is due to
methodological restriction. Therefore I would speculate that the S. griseus is preferred
after S. repandus, rendering the P. lanuginosus third choice.
2. Spines on columnar cacti showed to have no effect on herbivory by Bonairian goats.
a. During this study presence of spines did not have a significant effect on biomass
consumption indicating that spines do not play a major role in goat diet preference.
3. There is evidence, albeit weak, for an associational resistance between Opuntia and cacti.
a. In the field experiment goat herbivory was near-significantly affected by the presence
of Opuntia. This is in agreement with the field studies of Geurts and Ende van de (Ende
van de, 2015; Geurts, 2015). Their studies demonstrate, that when the number of
Opuntia reaches a certain density, the number of goats decline which leads to a
reduction of goat herbivore damages found on cacti.
4. The endemic green iguana only rarely, if at all, consumes cacti cladodes.
a. During this study no observation of cactus consumption was recorded by green
Iguana.
Implications and recommendations
My findings quantify the diet preference of Bonairian goats for three columnar cacti. As my findings
demonstrate the threat of goat herbivory on columnar cacti is substantial. Therefore goat herbivory
on columnar cacti can be considered detrimental to the conservation of these cacti. Similar results
have been reported for the threat of goat herbivory on cacti in many arid and semi-arid subtropical
areas of the world (Ricardi & Shimada, 1991; Hamann 1993; León de la Luz 2005; Meléndez-Ackerman
et al. 2008; Jimenéz-Sierra et al. 2007; Baraza & Fernández-Ososres, 2013). This pattern is consistent
with resent field study done by Coolen, Ende van de and Geurts, that demonstrate that goat herbivory
has a negative effect on cacti and limits new growth (Coolen, 2015; Geurts, 2015; Ende van de, 2015).
The primary aim for Washington Slagbaai national park should be to minimize goat herbivory within
the park as much as possible, preferably eliminating it all together. Following experience on
neighbouring island Curacao, policies should combine efforts to reduce or eradicate all feral goats
within the park, together with intensive education among human communities to raise social
awareness and change goat husbandry practises on the island (Costa Gomez & Debrot, 2007).
18
References
Alonso-Díaz, M.A., Torres-Acosta, J.F.J, Sandoval-Castro, C.A., Hoste, H., Aguilar-Caballero, A.J. &
Capetillo-Leal, C.M. (2008). Is goat’s preference of forage trees affected by their tannin or
fibre content when offered in cafeteria experiments? Animal Feed Science and Technology,
141, 36-48
Atkinson, I.A.E. (1996). Introduction of wild life as a cause of species extinctions. Wildlife Biology, 2
135-141
Atsatt, R. & O’ Dowd, D.J. (1976). Plant defence guilds: many plants are functionally interdependent
with respect to their herbivores. Science, volume 193, 24-29
Baraza, E., & Fernandez-Osores, S. (2013). The role of domestic goats in the conservation of four
endangered species of cactus: between dispersers and predators. Applied vegetation Science,
Volume 16, 561-570.
Barbosa, P., Hines, J., Kaplan, I., Martinson, H., Szczepaniec, A. & Scendrei, Z. (2009). Associational
Resistance and Associational Susceptibility: Having Right or Wrong Neighbours. Annual Review
of Ecology and Systematics, 40, 1-20
Batista, A.M., Mustafa, A.F., McAllister, T., Wang, Y., Soita, H. & McKinnon, J.J. (2003) Effects of
variety on chemical composition, in situ nutrient disappearance and in vitro gas production od
spineless cacti. Journal of the Science of Food and Agriculture, 83, 440-445
Boer de, B. (2001). Nos Bestianen, Onze Dieren, Our Animals. Giethoorn Ten Brink: Stichting
dierenbescherming Curaçao.
Buurt, G. (2006). Conservation of amphibians and reptiles in Aruba, Curacao. Applied Herpetology,
volume 3, 307-321.
Campbell, K., & Donlan, C. (in press). Feral goat eradication on islands. Conservation Biology.
Cerrillo, M.A., López, O.O., Nevárez, C.G., Ramírez, R.G. & Juárez, R.A.S. (2006). Nutrient content,
intake and in vitro gas production of diets by Spanish goats browsing a thorn shrubland in
North Mexico. Small Ruminant Research, 66, 76-84
Clout, M. (2002). Biodiversity loss caused by invasive alian vertebrates. Z. Jagdwiss, 51-58.
Clout, M., & Veitch, C. (2002). Turning the tide: The eradication of invasive species. Turning tide of
biological invasion: the potential for eradicating invasive species (pp. 1-3). Auckland, NewZealand: IUCN- The world conservation Union.
Coolen, Q. (2015) The impact of goat herbivory on the vegetation of Bonaire. Thesis report,
Wageningen University
Costa Gomez da, M. & Debrot, A.O. (2007). Caribbean Research and Management of Biodiversity
Annual report 2007. ©Carmabi foundation
Cortes, D.G.Z. (2012). An Ecological-Economic model of the tropical dry forest on the Island of
Bonaire. Institute of Enviromental Studies, Vrije Universitet Amsterdam
Cromarty, P., Broome, K., Cox, A., Empson, R., Hutchinson, W., & Mcfadden, I. (2002). Turning the
Tide: The eradication of invasive species. Eradication planning for invasive alien animal species
on islands - the approach developed by the New Zealand Department of Conservation (pp. 8591). Auckland, New Zealand: IUCN.
19
Debrot, A.O. & Freitas de, J.A. (1993). A Comparison of ungrazed and Livestock-Grazed Rock
Vegetations in Curacao. Biotropica, volume 25, 270-280
Debrot, A.O, de Graaf, M., Henkens, R., Meesters, H., & Slijkerman, D. (2011). A status report of
nature policy development and implementation in the Dutch Caribbean over the last 10 years
and recommendations toward the nature policy-plan 2012-2017. The Hague, The Netherlands:
Imaris Wageningen UR.
Desender, K., Baert, L., Maelfait, J., & Verdyck, P. (1999). Conservation on Volcan Alcedo
(Galapagos): Terrestrial invertebrates and the impact of introduced feral goats. Biological
Conservation, Volume 87, 303-310.
Ende van de, B., (2015) How do the distribution and abundance of cacti species relate to microsite
types and goat presence? Thesis report, Wageningen University
Forbes, J.M. (2007). A personal view of how ruminant animals control their intake and choice of
food: minimal total discomfort. Nutrition Research Reviews. 20, 132-146
Geurts, K., (2015) Abundance and vegetation impact of feral goat on Bonaire. Thesis report,
Wageningen University
Ginane, C., Duncan, A.J., Young, S.A., Elston, D.A. & Gordon, I.J. (2005). Herbivore diet selection in
response to simulated variation in nutrient rewards and plant secondary compounds. Animal
Behaviour, 69, 541-550
Green iguana society (2009). Food information chart. Retrieved on 8 February 2009 :
http://www.greenigsociety.org/
Goff, S.A., & Klee, H.J. (2006). Plant Volatile Compounds: Sensory Cues for Health and Nutritional
Value? Science, 311, 815-819
González-Espinosa, M., & Quintana-Ascencio, P. (1986). Seed predation and dispersal in a dominant
desert plant: Opuntia, ants, birds, and mammals. Frugivores and seed dispersal, 273-284.
Google maps. (2015, January). findlatitudeandlongitude. Retrieved from Washington Slagbaai
National Park: http://www.findlatitudeandlongitude.com/
Gurevich, J., & Padilla, D. (2004). Are invasive species the major cause of extinctions? Trends in
Ecology and Evolution, 470-474.
Hamann, O. (1993). On vegetation recovery, goats and giant tortoises on Pinta Island, Galapagos,
Ecuador. Biodiversity and Conservation, 2, 138-151
Jansen, D.A.W.A.M., Langevelde, F., De Boer, F. & Kirkman, K.P. (2007). Optimisation or satiation,
testing diet selection rules in goats. Small Ruminant research, 73, 160-168
Jezorek, H., Stiling, P. & Carpenter, J. (2011). Ant predation on an invasive herbivore: can an
extrafloral nectar-producing plant provide associational resistance to Opuntia individuals?
Biological Invasions, 13, 2261-2273
Jimenéz-Sierra, S.L., Mandujano, M.C. & Eguiarte, L.E. (2007) Are populations of the candy barrel
cactus (Echinocactus platyacanthus) in the desert of Tehuacán, Mexico, at risk? Population
projection matrix and life table response analysis. Biological conservation, 135, 278-292
Kohl, K.D., Miller, A.W. & Dearing, M.D. (2015). Evolutionary irony: evidence that ‘defensive’ plant
spines act as a proximate cue to attract a mammalian herbivore. OIKOS, volume 124, 835-841
20
Lacle, A., wolfs, E., Beukering van, P., & Brander, L. (2012). Recreational and cultural value of
Bonaire's nature to it inhabitants. Amsterdam: IVM institute for Environmental studies and
University of Amsterdam.
León de la Luz, J.L. (2005) Evaluation of the conservation status of Morangaya pensilis (Cactaceae), a
little known endemic monotypic genus of southern Baja California, Mexico. Oryx, 39, 219-222
Malo, J., Acebes, P., Giannoni, S., & Traba, J. (2011). Feral livestock threatens landscapes dominated
by columnar cacti. Acta Oecologica , Volume 37, 249-255.
Marken Lichtenbelt van, W.D. (1992). Digestion in an Ectothermic herbivore, the green iguana
(iguana iguana): Effect of food composition and Body temperature. Physiological Zoology,
Volume 65, 649-673
Marken Lichtenbelt van, W.D. (1993). Optimal foraging of a herbivorous lizard, the green iguana in a
seasonal environment. Oecologia, 95, 246-256
Martinez del Rio, C., Hourdequin, M., Silva, A., & Medel, R., (1995) The influence of cactus size and
previous infection on bird deposition on mistletoe seeds. Australian Journal of Ecology,
volume 20, 571-576.
Martorell, C., Vega, E., & Ezcurra, E. (2006) Morphological consequences of the trade-off between
growth and reproduction in columnar cactus (Lophocereus schottii). Plant Ecology, 183, 125131
Meléndez-Ackerman, E.j., Cortéz, C., Sustache, J., Aragón, S., Morales-Vargas, M., García-Bermúdez,
M. & Fernández, D.S. (2008) Diet of feral goats in Mona Island reserve, Puerto Rico. Caribbean
Journal of science, 44, 199-205
Mulder, S. (2013, september 9). Herbebossing op Klein Bonaire en Klein Curaçao succesvol. Retrieved
from Natuurbericht.nl : http://www.natuurbericht.nl/?id=11342
Muylaert, R.L., Da Silva, D.M., Aurelio Ribeiro Mello, M., & Aurelio Ribeiro Mello, M. (2014) Interindividual variations in fruit preferences of the yellow-shouldered bat Sturnira Lilium
(Chiroptera: Phyllostomidae) in a cafeteria experiment. Mammalia, volume 78, 93-101
Nassar, J., Ramirez, N., & Linares, O. (1997). Comparative Pollination Biology of Venezuelan
Columnar Cacti and the Role of Nectar-Feeding Bats in Their Sexual Reproduction. Botanical
Society of America, Volume 48, 918-927.
Peco, B., Borghi, C., Malo, J., & Acebes, P. A. (2011). Effects of bark damage caused by feral
herbivores on columnar cactus Echinopsis (=Trichocereus) terscheckii reproductive output.
journal of Arid Environments, volume 75, 981-985.
Parkes, J., Macdonald, N., & Leaman, G. (2002). Turning the tide: The eradication of invasive species.
An attempt to eradicate feral goats from Lord Howe Island (pp. 233-239). Auckland, New
Zealand: IUCN.
Peeters, E., Martorell, C., & Ezcurra, E. (2008). Nurse rocks are more important than nurse plants in
determining the distribution and establishment of globose cacti (Mammillaria) in the
Tehuacan vally, Mexico. Journal of Arid Environments, Volume 72, 593-601.
Petit, S., & Pors, L. (1996). Survey of Columnar Cacti and Carrying Capacity for Nectar-Feeding Bats
on Curacao . Society for Conservation Biology, June., 769-775.
21
Petit, S., & Freedman, C.E (1997). Nectar production of two sympatric species of Columnar Cacti.
Biothropica, 29, 175-183
Petit, S. (2001). The reproductive phenology of three sympatric species on columnar cacti on
Curacao. Journal of Arid Environments, 49, 521-531
Petit, S. (2011). Effect of mixed-species pollen load on fruits, seeds, and seedlings of two sympatric
columnar cactus species. Ecological restoration, 26, 461-469
Raats, J. (1998). Feeding behaviour of free range goats. South Africa: Department of Livestock and
Pasture Science, University of Fort Hare.
Ramírez-Orduña, R., Ramírez, R.G., Romero-Vadillo, E., González-Rodríguez, H., Armenta-Quintana,
J.A. & Avalos-Castro, R. (2008). Diet and nutrition of range goats on a sarcocaulescent
shrubland from Baja California Sur, Mexico. Small Ruminant Research, 76, 166-176
Ricardi, C. & Shimada, A. (1991) A note on diet selection by goats on a semi-arid temperate
rangeland throughout the year. Applied Animal Behaviour Science, 33, 239-247
Simal, F. (2005). Management plan Washington Slagbaai national park Bonaire, Netherland Antilles.
Bonaire: STINAPA- Stichting national parken .
Skarpe, C., Bergstom, R., Danell, K., Eriksson, H., & Kunz, C. (2012). Of Goats and spines - a feeding
experiment. African Journal of Range & Forage science, volume 29, 37-41.
Smith, S., van der Burg, W., Debrot, A., van Buurt, G., & de Freitas, J. (2014). Key elements towards a
joint invasive alien species strategy for the Dutch Caribbean. The Netherlands: Imares
Wageningen UR.
Solanki, G.S. (1994). Feeding habits and grazing behaviour of goats in a semi-arid region in India.
Small Ruminant Research, 14, 39-43
Solanki, G.S. (2000) Grazing behaviour and foraging strategy of goats in semi-arid region in India.
Tropical ecology, 414, 155-159
Taylor, A., Hort, J., & Roberts, D. (2004) Flavour perception. Blackwell, Oxford, United Kingdom, 1-39
TEEB, Caribbean Netherlands. (2012). Tourism value of ecosystems in Bonaire. the Netherlands:
Ministry of Economic Affairs, Agricultural and Innovation of the Netherlands.
TEEB, Caribbean Netherlands. (2012). What's Bonaire's nature worth? : the economics of ecosystems
and biodiversity on Bonaire. Netherlands: Ministry of Economic Affairs, Agricultural and
Innovation of the Netherlands.
Vieira, E.L., Batista, A.M.V., Mustafa, A.F., Araújo, R.F.S., Soares, P.C., Ortolane, E.L. & Mori, C.K.
(2008). Effects of feeding high levels of cactus (Opuntia fiscus-indica Mill) cladodes on urinary
output and electrolyte excretion in goats. Livestock science, 144, 354-357
Walsh, S. (2011). Ecosystem-Scale Effects of Nutrients and Fishing on Coral Reefs. Journal of Marine
Biology, 1-13.
Williams, D.G.; Hultine, K.R. & Dettman, D.L (2014). Functional trade-offs in succulent stems predict
responses to climate change in columnar cacti. Journal of Experimental Botany, 65, 3405-3413
22