Independent Research

Since we have returned from our field trip to the Atlantic Coast, we have been extremely busy preparing for and conducting our independent research projects. It has been a long month taking data and writing papers, but it has been so amazing to get the opportunity to conduct research in one of the most beautiful places in the world.

I had two goals for my project: (1) to determine whether hummingbird feeders have an effect on the pollination of a native plant, Stachytarpheta frantzii, and (2) how pollination is affected across a disturbance gradient. 

Some background information: 

Pollinators are super important in the life-cycle of plants. Hummingbirds are one of the most widely used pollinators, especially in the tropics. They have adapted to their role as pollinators with their ability to hover over flowers, which requires a high amount of energy. This need for energy correlates with high flower visitation in order to sustain their energy needs. 

As the ecotourism industry continues to expand in the tropics, hummingbird feeders have become increasingly popular. Tourists love to see the brightly colored birds, and these feeders provide the ideal way of providing this service to tourists. Hummingbird feeders provide a constant and reliable food source for hummingbirds, but this can have a serious impact on pollination of surrounding flowers. 

Stachytarpheta frantzii is a plant native to Costa Rica. It is very abundant and is literally found everywhere in Monteverde - from gardens, next to high-traffic roads, and near secondary growth forest. This plant also flowers throughout the year, which makes it an ideal food source for insects and hummingbirds in the dry season. This plant was used as my study organism, primarily because of its abundance in Monteverde and that it attracts large numbers of hummingbirds. 

Stachytarpheta frantzii being visited by a Scintillant Hummingbird at one of my sites.

Methodology

I established four sites throughout the Monteverde / Santa Elena region. Each site contained two patches of my study plant. One was the experimental patch (with a hummingbird feeder) and the second was the control patch (without a feeder). These two patches were a minimum of 50 meters apart and had a similar number of flowers. The four sites were chosen along a disturbance gradient, with two sites considered "least disturbed" and the other two considered "more disturbed." Disturbance was quantified by proximity to a nearby forest and the amount of surrounding vegetation. 

Here is a map of my four study sites and where they were located around Monteverde and Santa Elena: 

My homestay house, if anyone was interested, was very close to Finca Santa Maria, in the Cañitas region of Monteverde.

Bajo del Tigre is part of the Children's Eternal Rainforest and was therefore considered the "least" disturbed of the four sites. Cabinas Capuliín is a hotel that is within very close proximity to a forest, and was therefore considered a lesser-disturbed site. Both Finca Santa Maria and Cerro Plano were considered the most disturbed sites.

An example, from Finca Santa Maria, of how my sites were set up with the feeders.

The hummingbird feeders contained a 4:1 water to sugar solution, which closely resembles the nectar hummingbirds frequently find in flowers. Feeders were refilled every 3-4 days at each site. 

Data was collected every morning between 7:00 and 10:30 am. I visited two sites per day, and observed each patch at each site for 30 minutes. Sites were rotated every day, and patches within each site were also rotated to ensure that the time of day did not play a role. Flower visitation was used as a means of quantifying pollination. Every time I was at a patch, I took note of the following: (1) hummingbird species (2) number of flowers visited per bird (3) number of feeder visits, but only for experimental patch, and (4) the total time spent foraging.

Results of my Study

So what I found through all of my data collection was that hummingbird feeders actually played little role in affecting pollination of Stachytarpheta frantzii (actually they played no role since I had zero feeder visits throughout my sites). When I just compared control vs experimental patches, disregarding the feeder altogether, there was a difference at Bajo del Tigre in the number of flower visits between patches. The experimental patch had a much higher mean number of flowers visited than did the control patch for this site. When the number of flowers visited was combined for both patches at each site, I found significant differences between the four sites. There was a trend in there being greater differences between lesser disturbed sites, such as Bajo del Tigre, and Finca Santa Maria, one of the more-disturbed sites. This observation helped show that plants in areas more impacted by humans are more at risk for a decrease in hummingbird visits than plants in areas less impacted by humans.