Transcript Does fertilization influence herbivory during tropical forest restoration?

Does fertilization influence herbivory during tropical forest restoration?
Emma Rosenfield (‘16), Arianna Porter (‘15), Julia Rogers (‘16), Omari Matthew (‘14), Susan Gagliardi, Lauren
Bizzari, Cathy Collins
Results
3.5
120
3
100
2.5
2
Wet Season (No Fertilizer)
Wet Season (Fertilizer)
Dry Season (No Fertilizer)
Dry Season (Fertilizer)
1.5
1
Percent Herbivory
Globally, about 13 million hectares of forests were converted to other uses or
lost through natural causes each year between 2000 and 2010. 1 This includes
tropical forests that provide the plant with numerous resources and services.
Therefore, it is critical that we actively try to restore these landscapes. Our
restoration study is focused on the Firestone Center for Restoration Ecology in
Costa Rica. The land on this property was originally a healthy tropical
rainforest. However, after clearing for grazing, a non-native bamboo (Guadua
angustifolia), was planted and now dominates the landscape. In order to study
the conditions under which healthy rainforests can be restored, we designed an
experiment in this disturbed landscape. In our study, we asked whether
herbivore pressure responds to fertilizer added to tree seedlings in a tropical
rainforest environment.
Nurtrient Content (percent)
Results
Introduction
80
60
Intensity
Extent
40
20
0
0.5
Am
Cen
Guap
K
Mang
Species
MC
RR
V
0
N
P
Nutrient
K
Figure 2. Comparison between nutrient content of plant tissue in fertilized and non-fertilized plots in
the wet and dry seasons. Species sampled was Maya Colorado. Nutrients analyzed were Nitrogen (N),
Phosphorus (P), and Potassium (K). Errors bars show standard error.
Fertilization increased nitrogen and phosphorous content in the tree leaves in
the wet season (pN=0.0012, pP=0.0281), but not in the dry season (pN=0.0813,
pP=0.484) (Figure 2). Potassium showed no significant differences in the wet
(p=0.5798) or dry (p=0.3883) season.
• Differences in nutrient content of leaf tissue across seasons may be due to the
fact that fertilizer was added closer to sampling in the wet season. In the dry
season, sampling was carried out three months post fertilization. Therefore, the
seedlings did not have the fertilizer nutrients still in their leaves because they
had already processed them. Additionally, during the dry season trees may shunt
nutrients to their roots to preserve tree health.
60
1. In the context of a tropical forest restoration experiment,
a. Does fertilization influence herbivory?
b. Does fertilization alter leaf tissue nutrients?
With the intension of studying rainforest restoration, we set up the experiment
by clearing 16 plots within a monoculture of non-native bamboo and planted
16 native trees per plot. On June 3rd, 2013, 150g of fertilizer was added to 8
plots each in a ratio of 10:30:10 (N:P:K). Fertilizer was added again in August
and October 2013. We chose this ratio based on evidence that suggests that
lowland sites are limited by Phosphorous and that Potassium is important.
Each plot contained seven native trees species present in different relative
abundances.
To measure herbivory, we estimated the extent and intensity of herbivory at
the tree, branch, and leaf level in June 2013 and January 2014. Extent is the
number of leaves that are affected by herbivory, while intensity is the amount
of biomass missing from the tree (Figure 1).
To generate leaf nutrient composition, we collected leaf tissue samples and set
them to a laboratory for nutrient chemical analysis.
Buffer area around plots
50
Percent Herbivory (Intensity)
Methods
This figure shows that percent intensity and extent of herbivory vary
among species (pi<0.0001, pe<0.001). Furthermore, percent extent of
herbivory was higher than percent intensity in all species.
Conclusions
Results
Guiding Question
Figure 4. Comparison between percent herbivory, extent and intensity, in all planted tree species.
Species codes stand for names of native trees: Amarillon (Am), Cenazaro (Cen), Guapinol
(Guap), Kapok (K), Manglio (Mang), Maya Colorado (MC), Ron Ron (RR), and Vaco (V). Error
bars show standard error.
40
30
No Fertilizer
Fertilizer
• Fertilizer increased herbivory intensity across all species in the wet season, but
not in the dry season (Figure 3). This difference with season may be due to the
gap between fertilization and sampling in the dry season. As the wet season
data show, insects may concentrate their herbivory on tissues with higher
nutrient content.
20
• Fertilization does not influence the extent of herbivory. This may be due to the
fact that insects are locating leaves of higher quality and eating more of them,
thus increasing the intensity of herbivory, but not the extent.
10
0
Wet Season 2013
Dry Season 2014
Season
Figure 3. Comparison of herbivory intensity between fertilized and non-fertilized tree species
in the wet and dry season. Species sampled was Maya Colorado. Error bars show standard
error.
There was a significant difference in percent intensity between
fertilized plots and those not fertilized in the wet season (p=0.009).
However, this difference was not seen in the dry season (p=0.0706)
(Figure 3).
• In the wet season, herbivores preferred certain tree species (Figure 4).
Difference in herbivory across species may be due to differences in nutrient
uptake. For example, we noticed higher leaf cutter ant damage on Maya
Colorado seedlings.
• These results indicate that fertilizer influences species interactions in the early
stages of rainforest restoration.
12 x 12 m experimental plots (cleared)
Seedlings planted within bamboo
Extent
Extent
Extent
Extent
Intensity
Intensity
Intensity
Intensity
Figure 1. Explanation of extent and intensity measurements
1
http://www.fao.org/news/story/en/item/40893/
Acknowledgements
We would like to thank Dr. Cathy Collins and the rest of the lab team for their
support and guidance, as well as the Firestone Center for Restoration Ecology.