Performing Department
(N/A)
Non Technical Summary
Small-scale scientific studies where birds were prevented from eating insects off plants (typically by placing netting over the plant or cages) showcase the importance of birds in mitigating insect damage on plants. From those studies, we know that insects can explode in numbers if left uneaten by predators such as birds. From small-scale studies, we can only hypothesize what happens if birds continue to decline or disappear completely since large-scale bird exclusion studies are expensive and unethical. Insects multiplying in the absence of birds is alarming because we are facing global declines in birds and invasive insects cause $70 billion dollars in global agricultural crop damages every year. These declines could have detrimental effects to our food production and increase pesticide use, which could have added side effects to other wildlife. To mitigate the effects of global declines of birds, we need to understand what happens when birds disappear at a large-scale. A devastating event that occurred on the US territory of Guam could be an opportunity to study what happens to agricultural crops if birds disappear completely. After WWII, the brown tree snake was introduced on the island via military cargo and eradicated the native birds. Today, the forests are silent and empty, but nearby islands have maintained their native bird population, which allows for nearby comparisons. In this study, my goal is to understand how the loss of birds have affected herbivory and predation on caterpillars in agricultural crops.To understand how the loss of birds have affected insect damage on crops, I will collect leaves from agricultural crops on Guam (no birds) and nearby islands (birds) and compare how much insect damage are on the leaves. To understand predation that occurs on insects on crops in the absence of birds, I will place clay caterpillars on the same plants I collected leaves from and identify the type of predators that could attack real caterpillars. In 2022, I compared insect damage on leaves of 3 species of native plants found on Guam and two nearby islands (birds present) and found that the story was not as simple as expected. Lack of birds does not necessarily lead to high amounts of plant damage by insects like the small-scale studies suggested. From the preliminary study, I found that Guam (no birds) and Rota (birds present) had similarly high levels of plant damage by insects, while Saipan (birds present) have the lowest level of plant damage by insects. This unexpected finding may be due to the high amount of lizard attacks on Saipan (birds present) compared to Guam (no birds) and Rota (birds present) based on a separate study looking at attack rates on clay caterpillars. I found that there were significantly more lizard attacks in Saipan compared to the nearby islands. This suggests that lizards may also play a significant role in controlling pests. My results regarding the attacks on clay caterpillars were true for whether the model was placed on native and introduced plants, suggesting that similar patterns of caterpillar predation may emerge on crop plants. While I observed consistent patterns of attack on clay caterpillars and amount of insect damage on leaves among islands, we do not know if these results hold for agricultural systems. This is an important knowledge gap. This study is also important locally because most of the food on the island of Guam is imported, which makes food very expensive. The high cost of imported goods drives the 24% higher cost of living on Guam compared to the continental US. High damage by insects will also likely reduce the thin profit margins of local farms in Guam, increasing the dependence on imported food. Lastly, as the worldwide bird declines continue, this work will help us understand how our food production can be affected.
Animal Health Component
0%
Research Effort Categories
Basic
95%
Applied
5%
Developmental
0%
Goals / Objectives
The goal of this project is to understand how the loss of birds have affected herbivory and predation on herbivorous larvae in agricultural crops.Objectives:Quantify herbivory on crops in Guam (lacking birds and with reduced vertebrate predation) and the Commonwealth of the Northern Marianna islands (retaining birds and other vertebrate predators).Determine the differences in predation on herbivorous larvae on crops found in Guam (lacking birds and with reduced vertebrate predation) and the Commonwealth of the Northern Marianna islands (retaining birds and other vertebrate predators).
Project Methods
Location:This study will be conducted on three islands in the Mariana Islands chain: Guam (541 km2), Saipan (115 km2) and Rota (85 km2). The islands experience nearly identical temperature and rainfall patterns and are within 193 km of each other.Objective 1.Quantify herbivory on crop plants on Guam (no birds and reduced abundances of other predators) and nearby islands with birds and other predators.Crops:I will work with universities, farmers, and agricultural government agencies. I have already established partnerships with diverse stakeholders from conducting preliminary studies and previous projects. Three crops will be chosen, but selection will depend on what crops are the most accessible at the time of the study. I will sample on a minimum of 20 fields of each crop at least 1 km away from each other on each island. There are many small-scale agricultural operations in the region that typically produce coconut, breadfruit, tomatoes, cucumbers, eggplants, and melons.Herbivory:Using a stratified random design appropriate for the cropping system, within a crop field, 60 plants will be chosen, and from those, 20 leaves will be sampled randomly from each plant among leaves of intermediate age. For the leaves chosen, a photo of the leaf will be processed with ImageJ (Abrà moff, M.D., Magalhães, P.J. and Ram, 2004).Objective 1 Evaluation:Data analysis and visualization will be conducted using the R statistical program language with RStudio (R Core Team, 2018; RStudio Team, 2016). A generalized linear mixed-effects model will be developed to determine how herbivory on the leaves may be affected by island, plant type, and its interactions.Objective 2.Determine differences in predation on herbivorous larvae on crop plants between Guam and nearby islands with birds.Plants:The same crops used for the herbivory study will be used to deploy the clay caterpillars.Clay Models:Clay caterpillarswill be made with green polymer, malleable, non-toxic and nonair-drying clay. The clay caterpillars will be made using a metal clay extruder to create uniform figures. The caterpillars will be approximately 8 mm in diameter and 5 cm long.Clay Caterpillar PlacementUsing a stratified random design appropriate for the cropping system, within a crop field, 60 plants will be chosen. Models will be placed on the center of the leaves. Predation happens at various heights thus, the caterpillars will be stratified along the plant (levels: low, medium, high). The low-level caterpillar will be placed a few inches off the ground to prevent accidentally stepping on it. Models will be left for 3 days. A photo of each side of the caterpillars will be taken after placement to compare marks observed 3 days later. The marks will be assessed using attack examples on clay caterpillars from Low et al., (2014).Objective 2 Evaluation:Data analysis and visualization will be conducted using the R statistical program language with Rstudio (Rstudio Team, 2016; R Core Team, 2018). A generalized linear mixed effects model will be developed to determine how the rates of attack on the caterpillar may be affected by the island, placement level, clay color, plant type and their interactions.Literature Cited:Abramoff, M.D., Magalh.es, P.J. and Ram, S. J. (2004). Image processing with ImageJ. Biophotonics International, 11(7), 36-42.Low, P. A., Sam, K., McArthur, C., Posa, M. R. C., & Hochuli, D. F. (2014). Determining predator identity from attack marks left in model caterpillars: guidelines for best practice. Entomologia Experimentalis et Applicata, 152(2), 120-126.R Core Team. (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.RStudio Team. (2016). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA.