Understanding amphibian responses to noisy suburban habitats
Dr. Lindsey Swierk and Dr. Jennifer Tennessen, Yale University, New Haven, CT
The 2017 suburban amphibian monitoring season has begun! Our Song Meter SM4 recorders have been deployed to examine the effects of land use change and noise pollution on wood frog choruses. Wood frogs sing in choruses in preparation of their spring breeding season, and it's unknown how noise pollution will affect the abilities of these frog populations to persist in suburbanizing landscapes. We are working to answer this question through a multi-year comparison of suburbanized and forested breeding ponds in Connecticut.
At this point, we have selected and begun to monitor the six ponds that we will track for the next several years. In the third week of February, we placed one Song Meter SM4 at each pond prior to the annual migration of wood frogs. The recorders will collect data throughout each day until wood frogs leave the ponds. These recordings will complement other data on the effects of suburbanization on wood frogs, including behavioral, morphological, and physiological measures that we are collecting this year. While placing the recorders, we had the opportunity to communicate our research to interested residents and their children – all were excited to be living so close to where "real" science was taking place!
The six Song Meter SM4 recorders successfully collected two months' of recordings surrounding the wood frog breeding season at suburban and forested ponds in Connecticut. Wood frog breeding was unusual this year: an unexpected late-season blizzard in March divided the season in half by several weeks. We noted some mortality of adults and fertilized eggs in the breeding ponds following the blizzard, although most choruses managed to recover after the weather warmed. This extreme weather event will allow us to not only examine how wood frog choruses are affected by anthropogenic noise (for example, the large amount of traffic noise that that interfered with some choruses), but also how indirect anthropogenic influences, such as extreme weather events that are predicted to increase in frequency in many global climate change scenarios, will affect chorusing behavior of amphibians. We are currently preparing the sound files for analysis with Wildlife Acoustics' Kaleidoscope program.
The analysis of our Song Meter data is underway! We began our comparison of the effects of suburbanization on wood frog breeding activity by quantifying the number of advertisement calls that were performed in the 5 minutes at the start of every other hour throughout the breeding season. (The record high so far is almost 7000 calls per 5 minutes!) Not only will this allow us to compare breeding season durations, start and end times, and a proxy of the number of animals in the chorus at ponds over a suburbanization gradient, but we will also be able to quantify how different aspects of weather (temperature, wind speed, water temperature, humidity, etc.) affect the call rate at ponds in different environments. Interestingly, the 2016 pilot data from our most and least suburbanized ponds and already show some interesting trends. Frog choruses peak during the nighttime hours in the most suburbanized pond but, in our most forested (least suburbanized) pond, this pattern isn't as apparent. If this trend holds true, it could be evidence that frogs are altering their calling behavior in noisier environments. Stay tuned!
We continue to examine data from the first year of our multi-year study on the effects of suburbanization on wood frog choruses. Our 2017 Song Meter data have already taught us quite a bit about wood frog chorusing behavior across the suburban gradient. Suburban wood frog choruses appear to be more robust to unfavorable weather; colder and windier days appear to be less of an impediment to suburban wood frogs than to those in the forest. Adult male population size, as estimated by call-counting proxy, is not directly related to suburbanization but instead to pond size and habitat. That said, our most suburbanized pond (surrounded by 70% suburban development to 200 m) hosted the smallest population despite its close similarity in size to several of the other ponds in the study. The figure shown here depicts the number of call detections in 5 minutes collected every 2 hours over the breeding season in each of six ponds, from "1" (most forested) to "6" (most suburban). We are currently developing zero-inflated time series statistical models of count data to quantify the effects of multiple environmental parameters on calling rates (e.g., water temperature's effect on calling rates, as shown in the figure depicting a single pond's chorus from our pilot study). Such models will enable us to pinpoint how individual parameters differ in their effects on calling rates between suburban and forested populations. We are also in the process of quantifying characteristics of individual calls within choruses and examining how these relate to each pond's degree of suburbanization.
One of the best aspects of Song Meter data collection is the ability to re-use data in unanticipated ways. The wood frog breeding season in 2017 was interrupted by a late-season blizzard (note the division of calling behavior in the six-panel figure), decimating the breeding wood frog adult population of many ponds in our study area. With our Song Meter data, we were able to document the blizzard's effect on calling behavior, which we plan to compare to other (non-blizzard) years in the future. We hope to be able to explore wood frog population sensitivity and recovery to extreme-weather events, and how suburbanization affects these responses. Contrary to our expectations, we observed that suburban wood frog choruses rebounded more quickly after the blizzard, despite the fact there was no difference in pre-blizzard chorus start dates. We are interested in investigating if suburban development may alleviate the effects of severe weather by causing ponds to warm more quickly, post-blizzard.