Effects of simulated highway noise on heart rates of larval monarch butterflies, Danaus plexippus: implications for roadside habitat suitability

Abstract

Developed countries around the world are criss-crossed with vast networks of roadways. Conservationists have recently focused attention on roadsides as possible locations for establishing pollinator habitat, with the monarch butterfly (Danaus plexippus) featuring prominently in such discussions. However, roadsides are inherently loud, which could negatively affect developing larvae. We conducted a series of experiments testing if simulated highway noise stresses monarch larvae, which we gauged by non-destructive monitoring of heart rates. In two replicated experiments, larvae exposed for 2 h experienced a significant increase in heart rate (16 and 17% elevation), indicating they perceive traffic noise as a stressor. Meanwhile, experiments exposing larvae for either 7 or 12 days to continuous traffic noise both showed no heart rate elevation at the end of larval development, suggesting chronic noise exposure leads to habituation or desensitization. Habituation to stress as larvae may impair reactions to real-world stressors as adults, which could be problematic for a butterfly that undertakes an annual two-month migration that is fraught with dangers. More generally, these results could have far-reaching implications for the billions of insects worldwide that develop near roadways, and argue that further study is needed before promoting roadside habitat for butterfly conservation.

1. Introduction

Worldwide pollinator declines, stemming from loss or degradation of habitats, are a major conservation issue because of the critical importance of pollinators to agricultural production. To address this, new areas of land need to be converted into suitable habitat for bees, butterflies and other native pollinators. In developed nations, conservationists have increasingly focused their attention on roadsides, or verges, i.e. the strip of land adjacent to roads, as potential habitat for pollinators. This idea makes intuitive sense, because such areas are usually government property (not privately owned), and by establishing pollinator-friendly spaces local governments can reduce mowing and make the property aesthetically pleasing to passers-by. However, research into the suitability of roadside areas for pollinators is only just beginning, with some results indicating they could be of conservation value with the appropriate mix of plant types [1]. However, new studies examining monarch butterflies (Danaus plexippus) have shown that roadside habitats support fewer than the expected numbers of eggs and larvae [2,3], suggesting female monarchs avoid egg-laying near roads, or perhaps are killed by cars while laying [4]. Other research indicates runoff from roads can affect hostplant chemistry, which in turn alters monarch development [5]. It is worth noting that one question that has never been asked is ‘do roads stress the insects living next to them?’

An overlooked feature of roadsides by insect conservationists is the constant noise from passing cars. According to U.S. Dept. of Transportation, ‘Levels of highway traffic noise typically range from 70 to 80 decibels (dB) at a distance of 15 meters (50 feet) from the highway’ [6]. Meanwhile, noise levels in a typical home are in the 40–45 dB range. This constant noise could constitute a stressor to nearby insects, especially if they are exposed to it for long periods, such as developing larvae would be. Moreover, despite not having ‘ears’, there is evidence that some Lepidopteran larvae, like the monarch, can sense and react to loud noise such as passing aircraft [7]. A major question, therefore, is would the constant noise of highway sounds negatively affect those insects living near the road? Here, we conducted a series of experiments to determine whether short- or long-term exposure to simulated road traffic noise induces a state of physiological stress in monarch caterpillars, which we gauged by monitoring their heart rates. Given that cardiac output is elevated whenever animals, including insects, become stressed, we surmised that traffic noise should elevate larval heart rates if they perceive it as stressful.

2. Material and methods

(a). Experimental set-up

We conducted experiments using monarch larvae reared in our laboratory (at the University of Georgia, Athens, GA, USA). All larvae had no prior exposure to highway noise. We used two identical laboratory rooms in the same building, both with windows facing the same direction (allowing identical day length and lighting) and both maintained at the same temperature (24°C). One room was quiet (45–50 dB), while the other room had simulated traffic noise played from two speakers connected to a PC computer (see electronic supplementary material, figure S1.1, File S1 and Audio file A1). In the audio, vehicles can be heard passing the recorder at a rate of 1 vehicle every 2–3 s (equivalent to approx. 12 000 cars per day). The sound level next to the speakers registered 65 dB on average, peaking at 75–80 dB when cars ‘passed’ in the audio. For the experiments outlined below, larvae were placed in either the quiet (control) room or the noise room for the duration specified.

(b). Monitoring heart rate

Two procedures were used to estimate heartbeat frequency of larvae, outlined in detail in electronic supplementary material, File S2. Briefly, in the 2016 experiments, we visually observed (under a stereo microscope) the dorsal vessel pumping by watching movement under the skin (figure 1 and electronic supplementary material, Video V1). In 2017, we employed a custom-built infrared probe that senses the dorsal vessel contractions and relays this information to a computer screen (figure 1). Both approaches provide similar estimates of heart rate (electronic supplementary material, File S2). In either case, care was taken to obtain readings within 1 min, to minimize handling effects on heart rates. Pilot work showed that larval heart rates are slow to elevate after handling (electronic supplementary material, File S2).

Figure 1.

Methods for assessing heart rates of monarch larvae. In year 1 experiments, larvae were visually inspected under a dissecting microscope by one observer, who counted beats of the dorsal vessel (heart) just under the cuticle (video available in the electronic supplementary material). In year 2 experiments, we employed a novel device to electronically monitor heart rates, which involved an infrared emitter fixed to an older-model microscope body, which conveys electronic pulses of the dorsal vessel to a computer screen (further description given in electronic supplementary material, File S2). Comparison of both methods across 10 larvae yielded statistically similar heart rates. (Online version in colour.)

(c). Acute exposure

Two acute exposure experiments were conducted (September 2016 and September 2017). In both cases, larvae were held individually in 1 l plastic containers filled with milkweed. In 2016, 20 larvae from two separate lineages were used. In 2017, we used 40 larvae reared from a mix of three lineages. All larvae were three to four generations descended from wild adults collected on the Florida panhandle the previous autumn and/or during spring migration in northern Georgia. On the day of the experiment, one of us (A.K.D.) estimated heart rates of each larva (5th instar). Then half were placed in the noise room, approximately 20 cm from the speakers, and the other half placed in the control room. After 2 h, we obtained post-exposure heart rate readings. Larval mass was also recorded. Heart rates before and after exposure were statistically compared in each experiment (repeated-measures ANOVA).

(d). Chronic exposure

Two experiments were conducted exposing monarch larvae to traffic noise for long periods during development. In September 2016, we exposed 40 third-instar larvae (from four lineages) to noise for one week (2–8 September), and an additional 40 larvae were maintained in the control room. In September 2017, we used 52 larvae from a mix of lineages and exposed them to noise (n = 24) or control (n = 28) for the entire duration of their larval development, even in the egg stage (10–22 September). For logistic reasons, larvae were reared in groups of 5 in 2016. In 2017, larvae were reared individually. Also in the 2017 experiment, we swapped the rooms every 3 days (i.e. moved the equipment and speakers) to control for possible location effects. During the trials, fresh milkweed was supplied daily. When larvae reached 5th instar, we recorded heartbeat frequency and larval mass. Heart rates of exposed and unexposed monarchs were compared for both chronic exposure experiments (ANOVA). Full datasets for all 2016 and 2017 experiments are available in electronic supplementary material, File S1.

3. Results

(a). Acute exposure

In our two experiments, monarch larvae exposed to highway noise for 2 h experienced a significant increase in heartbeat frequency, while larvae placed in our control room for 2 h showed no increase, so that the interaction of treatment × time was significant in both experiments (p < 0.001, figure 2; electronic supplementary material, tables S1.2 and S1.4, File S1). The magnitude of the increase in the noise treatment was approximately 17% in 2016 and 16% in 2017. Larval mass was not a predictor of heartbeat frequency in either experiment and there was a significant family × time interaction (p = 0.0127) in the 2016 experiment.

Figure 2.

Heart rates of monarch butterfly larvae in response to exposure to acute or chronic simulated highway noise, replicated over two experiments. In acute exposure trials, we monitored heart rates before and after exposure to either highway noise or a control (no noise). In chronic exposure trials, larvae were exposed to either noise or control and heart rates were recorded after 7 days in experiment 1, after 12 days in experiment 2. All larvae were 5th instar at the time of measurement. Acute exposure (2 h) to traffic noise led to a 16–17% increase in heart rate, but there was no elevation after chronic exposure (7 or 12 days). (Online version in colour.)

(b). Chronic exposure

After chronic exposure to road noise for 7 days (2016) or 12 days (2017), heart rates of 5th-instar larvae were not significantly different from controls in either experiment (p = 0.208, electronic supplementary material, table S1.6 and p = 0.882, electronic supplementary material, table S1.8), after controlling for family and larval mass. Larval mass did not affect heart rate in either experiment (p > 0.2 for both).

4. Discussion

A universal response to stressors in the animal kingdom, including humans, is an increase in heart rate [8–10]. We used this universal response to ascertain if monarch butterfly larvae perceive anthropogenic road noise as a stressor. Indeed, results from two replicated experiments indicated that traffic noise induces a state of physiological stress in naive monarch larvae. We are confident in this result, given (i) the same result was found in two experiments, (ii) the precautions we took while handling larvae and (iii) the fact that the control exposure did not raise heart rates (figure 2). Moreover, the magnitude of the elevation is consistent with other work where insects were experimentally stressed [10].

It is important to point out that the acute exposure experiments did not replicate a real-world scenario, because we used naive 5th-instar larvae. Our intent with these trials was only to determine whether larvae perceive road noise as ‘stressful’, and our results confirm they do. In the wild, larvae living near roads would be exposed to noise for the duration of their development, which was the rationale for the chronic exposure experiments. Both of the long-term experiments (7 or 12 days) showed heart rates were not elevated by the end of development (figure 2). There are three possible explanations for this result: (i) that monarch larvae are completely unaffected by long-term road noise, (ii) that they find it stressful but eventually habituate to the noise, or (iii) that insect heart rates simply do not (or cannot) remain elevated during long exposures to a stressor. Further study is needed to determine which explanation is most accurate.

Regarding the second scenario above, if monarchs become desensitized to stress as larvae, it could be cause for concern in the adults. Research with other animals indicates long-term exposure to stressors leads to weakened physiological stress reactions, which would affect reaction times or performance during real-world stressors [11]. This would be especially damaging for a butterfly that undertakes an arduous two-month trek that is fraught with stressors, including, ironically, avoiding cars on freeways [4]. Also ironic is that adult monarchs are not likely to be stressed by the actual freeway noise, because they lack any known auditory-sensing structures (Jayne Yack 2016, personal communication).

Research into the effects of anthropogenic noise on animals has burgeoned in recent years, though noticeably lacking are studies of invertebrates [12]. Our study contributes to this knowledge gap, and moreover our approach, to examine heart rate responses of butterfly caterpillars to a noise stressor, is to our knowledge, the first of its kind within this literature body. Furthermore, if these results on monarchs could be generalized to other Lepidopteran larvae, this project adds a new dimension to the issue of roadside habitat for pollinators: it argues the suitability of roadside habitat should not only be judged on the number of larvae present, but by whether the larvae present are physiologically normal and ‘stress-free’.

Data accessibility

All data for the experiments conducted in this project are available in the electronic supplementary material files associated with this manuscript.

Authors' contributions

A.K.D. contributed to the conception and design of the experiments, data collection, data analysis and wrote the manuscript; H.S. and I.Y. contributed to design of the experiments, data collection and logistics of the experiments and provided feedback on the manuscript; and J.P. helped conceive the experiments, collect data for the experiments and provided feedback on the manuscript. All authors agree to be held accountable for the content therein and approve the final version of the manuscript.

Competing interests

We declare we have no competing interests.

Funding

We received no funding for this study.