Fotoecological aspects of Bicellonycha lividipennis Motschulsky, 1854 (Coleoptera: Lampyridae) fireflies: adaptation to sites under ALAN and anthropic influenceAbstract
Fireflies are disappearing around the globe due to habitat reduction, use of pesticides, and especially artificial nightlighting (ALAN), which directly interfere with their sexual communication through bioluminescence. For this reason, they constitute promising nocturnal bioindicators. Here we report ecological aspects of Bicellonycha lividipennis (Motschulsky, 1854), a common marshy field firefly in Brazil, and its occurrence in sites under ALAN and anthropic influences during the past 30 years. This firefly is twilight active and was abundant in flat swampy areas, often impacted by tramped pastures and also in sites under high levels of ALAN, especially High Pressure Sodium (HPS) lamps. The relatively high natural irradiance during twilight at the onset of activity of this firefly (dowelling ≤6 Lux; upwelling~ 0.5–2.0 Lux), the lower overlap of the bioluminescence spectra (570 nm) in the yellow-green window (550–565 nm) with the natural twilight illumination (<550 nm, > 600 nm), and the sunset simulating effect of HPS (>568 nm) may explain the adaptation of this firefly in sites under influence of HPS. The results indicate that B. lividipennis is an opportunistic species which may have recently spread to opened areas, including those under ALAN. HPS could be less impacting for yellow-green emitting fireflies active in open meadows during the short twilight of tropical latitudes.
Keywords
Fireflies; artificial nightlighting; photopollution; Bicellonycha
Resumo
Vagalumes estão desaparecendo em todo o mundo devido à redução de habitat, uso de pesticidas e, especialmente, à iluminação artificial noturna (IAN), que interfere diretamente na comunicação sexual desses insetos por meio da bioluminescência. Por essa razão, eles constituem bioindicadores noturnos promissores. Neste estudo, relatamos aspectos ecológicos de Bicellonycha lividipennis (Motschulsky, 1854), um vaga-lume comum em áreas alagadas no Brasil, e sua ocorrência em locais sob influência de IAN e de ações antrópicas durante os últimos 30 anos. Esse vaga-lume é abundante em áreas pantanosas planas, frequentemente impactadas por pastagens pisoteadas e também em locais com altos níveis de IAN, especialmente sob lâmpadas de vapor de sódio de alta pressão (VSAP). A relativamente alta irradiância natural durante o crepúsculo, no início da atividade dessa espécie (irradiância descendente ≤6 Lux; irradiância ascendente ~0,5–2,0 Lux), a menor sobreposição do espectro de bioluminescência (570 nm) na janela verde-amarelada (550–565 nm) com a iluminação natural do crepúsculo (<550 nm, >600 nm), e o efeito similar ao pôr do sol proporcionado pelas lâmpadas VSAP (>568 nm), podem explicar a adaptação desse vaga-lume em locais sob influência de VSAP. Os resultados indicam que B. lividipennis é uma espécie oportunista, que pode ter se espalhado recentemente para áreas abertas, inclusive aquelas sob influência de IAN. A luz VSAP pode ser menos impactante para vaga-lumes que emitem luz verde-amarelada e que são ativos em campos abertos durante o curto crepúsculo das latitudes tropicais.
Palavras-chave
Vagalumes; luz noturna artificial; fotopoluição; Bicellonycha
Introduction
Fireflies and other adult bioluminescent beetles use their bioluminescence primarily for purposes of sexual attraction, whereas immature stages use it mainly for defense and eventually for illumination or prey attraction. In flashing fireflies, practically all mating communication is carried out by light signals (LLOYD, 1983; LEWIS & CRATSLEY, 2008). Environmental factors, especially the ambient illumination determine the timing and activity of fireflies (DREISIG, 1974). Therefore, it is likely that factors that affect the visibility of the bioluminescent signals, especially higher levels and the spectral range of ALAN (Artificial Light at Night), can dramatically affect the mating behavior of fireflies (LLOYD, 2006).
The perception of bioluminescence in fireflies is dictated by the visual sensitivity, under influence of visual pigments and screening pigments in the eyes, as well as the signal to environmental noise ratio. The perception of the bioluminescence signal of fireflies is affected by the contrast ratio between the bioluminescence spectrum and the photic environment of the habitat (SELIGER et al, 1982 a,b). According to this hypothesis, the twilight active fireflies preferentially display bioluminescent signals in the yellow region to contrast better against the green light reflected by the grass, whereas the dark active fireflies display preferentially green bioluminescence, since there is no longer a competing effect of the reflectance of the grass, and furthermore the visualization of the green bioluminescence is maximized (SELIGER et al, 1982a,b; LALL 1981). Twilight fireflies display screening pigments in the eyes which attenuate the visual sensitivity in the green region, narrowing the visual sensitivity spectrum and shifting the sensitivity to the yellow (LALL et al., 1988; CRONIN et al. 2000). Studies with North American Photinus scintillans firefly showed that the onset of activity of twilight active fireflies is mediated by both short- and long wavelengths (LALL, 1993). Studies with Brazilian fireflies, on the other hand, did not showed a clear trend, since some night active species emitted yellow light, probably due to the much shorter duration of twilight at such higher latitudes which reduces the selective pressure for different BL colors (VIVIANI & BECHARA, 1995). However, a recent study comparing 3 Brazilian fireflies, two twilight active emitting yellow and a twinight emitting green bioluminescence showed that, at least these species, follow the trend of BL color adaption found in Neartic species (LALL et al., 2022).
ALAN has recently been considered the second most important threat to fireflies around the world (LEWIS et al. 2020), but detailed studies are still scarce. The average natural environmental light varies from 0.001 Lux in moonless nights up to >0.3 Lux during full moon, however, in urban environments it may vary from 0.2 Lux for the downwelling light during cloudy nights up to 15 Lux near public street lights (RICH & LONGCORE, 2006; GASTON et al., 2014; CARAVEO, 2024). In Brazil, observations during the last 30 years indicate the disappearance of fireflies around urban centers, however, detailed studies on the effect of ALAN are still very scarce (VIVIANI et al., 2010; SANTOS, 2013). We showed that Photinus sp1 fireflies, found in the campus of Sorocaba of Federal Univ. São Carlos, are strongly affected by the ALAN: areas under illumination near buildings and sport areas lack fireflies of this species, whereas similar areas without illumination display higher presence of these fireflies (HAGEN et al., 2015). More recently, VAZ et al., (2021) made a multiparametric analysis of environmental stressors on the ghost firefly Amydetes fastigiata in Rio de Janeiro state. Using satellite analysis, the authors predicted that ALAN caused by urban sprawl could be a major increasing concern for this firefly. Preliminary studies with fireflies at Maryland Piedmont showed that introduction of light source near the habitat reduce their flashing frequencies of fireflies (COSTIN AND BOULTON, 2022).
Very recently, the effect of spectral composition of ALAN also begun to be investigated. OWENS et al. (2018) showed that light with wavelengths in the blue-green region below 533 nm affects the frequency and brightness of flashes in the firefly Aquatica ficta, whereas wavelengths above 597 nm had no effect. Studies with Photinus obscurellus using LED sources of different intensities showed that, whereas dim blue or red light sources are less disruptive than amber light, both brighter blue and red light are more disruptive. All colors of artificial light reduced the activity of Photinus carolinus, showing that red light is less disruptive to firefly courtship than amber light (OWENS at el., 2022). Furthermore, the pattern of the flash is also changed by the presence of these sources, under blue and red light the flashes are longer, brighter and more frequent, whereas females flashing activity is diminished under white light (OWENS AND LEWIS, 2021).
Despite the above studies, in situ studies on the adaptation of fireflies in sites under influence of specific artificial light sources commonly used in public illumination, including irradiance measurements as well as their spectral composition, are still scant. In Brazil, during the past two decades high pressure sodium lamps (HPS) were commonly used for road illumination in many cities, though they are now being gradually substituted by white light led sources.
Here we report ecological and photoecological aspects of Bicellonycha lividipennis (Fig. 1), a very common firefly in marshy fields in São Paulo state (VIVIANI, 2001), and observations made during the last 40 years, showing the occurrence of this firefly in environments under anthropic influences, especially its persistence in artificially illuminated sites.
Material and Methods
1. Investigated sites
We compared the occurrence of Bicellonycha fireflies in different places under direct or indirect anthropic influence, along water courses inside Sorocaba and Campinas municipalities along the past decades. We also compared the occurrence of B. lividipennis fireflies in artificially illuminated sites and non-illuminated sites in Campinas, Sorocaba and Votorantim municipalities. Fig. 2 shows the physiognomy of the habitats, and Figures S1 and S2 shows the hydrographic map of Sorocaba and Campinas municipalities with the investigated sites. From 1987 through 2024, we made almost yearly visits and observations at Faz Santana (Campinas municipality). From 2012–2015 we selected and investigated in more detail two main sites in Sorocaba/Votorantim municipalities, which were very closely located and displayed similar habitat physiognomies, but distinct environmental illumination patterns: (I) Jd. Clarice in Votorantim municipality, and (II) the campus of UFSCAR (23°35’11.2”S 47°30’58.6”O). From 2017 to 2025 we investigated another swampy area at the Campolim site and Sorocaba river margins inside Sorocaba city. Below is the description of the investigated sites:
Habitats where B. lividipennis fireflies are found: (A) campus of UFSCAR in Sorocaba; (B) Campolim in Sorocaba; (C) Jd Clarice in Sorocaba; (D) another view of Jd. Clarice. This species is found in flat swampy areas, often besides cattail belts.
2. Sorocaba/Votorantim municipality
Jd. America (23°31’00.6”S 47°28’29.9”O) which consisted of a flood retention artificial lake inside the city; Campolim park (23°31’44.1”S 47°28’08.7”O) which consisted of a recreation park with a small creek and two lakes, where the margins of the first had shallow grasses where the fireflies were found (23°30’20.4”S 47°27’04.8”O); Campolim (Sorocaba). This site was a former farm consisting of a swampy field with grasses and cattails around a small creek, with pastures and crops in one side and a secondary growth on the other side. This site was investigated since 2017, and actually is undergoing fast urbanization, with increase of surrounding tall buildings and recently cut by an avenue, with increasing influence of high pressure sodium lamp (HPS). Jardim Clarice (Votorantim; 23°33’47.4”S 47°27’43.1”O) this area was a swampy depression where there is a small stream, mostly dominated by pastures with patches of cattails (Typha latipholia), which was under influence of HPS, in a region with growing urbanization from 2006–2015; campus of UFSCar (Sorocaba, 23°35’11.2”S 47°30’58.6”O). This site is located about 9 km away from Jd Clarice and is also located in a swampy depression, where there was a lake that was gradually overtaken by cattails and was not under direct influence of artificial nighlighting (ALAN). In one side of this lake there is a small remnant of secondary forest, and at the other side, where the study was carried out, there is a slope with cattle pasture (Fig. 2). Sorocaba´s river (Sorocaba), where we spotted 3 sites at the swampy margins along the river inside Sorocaba city; Terras de São Francisco (Salto de Pirapora 23°38’50.8”S 47°35’24.9”O): The condominium has a lake with surroung swamp grasses and cattails. National Forest Ipanema (Flona) (Iperó, 23º25’47.57”S, 47º35’54.26”O) The collections and observations were carried out in the Ipanema farm, close to the dam in an area with remnants of Atlantic forest.
3. Campinas municipality
Santana farm, Sousas – (22°53’38.3”S 46°58’16.6”O), this site has been under investigation during almost 40 years (Viviani, 2001). It consists of a marshy horse pasture bordering Atibaia’s river, constituted by short grasses.
Other sites in Campinas municipality. We also included in this investigation other previously investigated smaller sites located at Jd. das Palmeiras and Jd. Paineiras, which were farm regions that underwent gradual urbanization from 1983–2000, and Regina Coutinho Nogueira Elementary state school and surroundings (Vila Nogueira), a place where B. lividipennis fireflies were first found by one of the authors (VRV), back in 1981–1982, and surroundings.
Limeira-SP. This area consisted of a eutrophyzed bog between in the middle of sugarcane plantations, and the Washington Luís highway on another side.
Itirapina-SP. This site consisted of a swampy pasture inside a farm near the SP310 highway.
4. Insects collection and identification
Fireflies of Bicellonycha lividipennis (Fig. 1) were observed and collected just after sunset in marshy areas during the period from November through March, mainly during moonless evenings. They were identified by comparison with the collection of Bioluminescent Coleoptera under curatorship of V. Viviani at UFSCAR, which were previously identified by comparison with paratypes of Museu de Zoologia of São Paulo University. In the field, they could be easily identified by their typical flash pattern, which consists of single yellowgreen flashes. Adult males were collected with entomological net, whereas females were usually collected perched on the grass. Larvae were collected at night by hand in the marshy ground in the same areas, upon location of their glow.
5. Observation and activity time of fireflies
One of the main goals of this work was to compare the occurrence of flashing fireflies in swampy areas with similar habitat characteristics, under natural (NNL) and artificial nightlighting (ALAN). The onset of activity was considered when the first flashing firefly was observed, and the end when no more flashing fireflies were observed for a period of 10 min. Crepuscular units (CREP) were also calculated according to NIELSEN (1963) to have an additional photoecological parameter, besides environmental irradiances reported below. More than 140 hours of observations have been conducted in different periods and sites, alternating locations each day, to reduce the effect of climatic variations. Attempts to estimate the number of individuals and applying statistics were done, but without much success, because it was not trivial to estimate the flashing populations of this species. Thus, because this species usually occurs in high populational densities during the reproductive season, we considered just 2 categories: present or absent. We consider an established population, when ≥10 individuals were seen in flashing activity during a limited time of 30 s in the visual field (a stretch of 30 m) during several evenings in the reproductive season (November-March) for at least 3 consecutive years. Repeated signals from the same individual were not eligible and individuals with different flashing signals, indicating distinct species, were not considered in such an analysis. We sampled areas with similar swampy habitats under ALAN and without ALAN in different locations in the municipality of Sorocaba. Furthermore, we also considered the occurrence of larvae on the ground, as a proof of adaptation of this firefly to the investigated habitats.
6. Photography
Images were obtained with a Sony Cybershot or Canon Ei5 cameras with a 100 mm Macro lens, and long exposures times.
7. Measurement of environmental irradiances
The environmental vector irradiances were measured in Lux using a field photometer provided with a cosine SKL-310 sensor (SKYE INSTRUMENTS LDT, 2007), which works in an active range between 0.010 and 200 lux. The photometer was allowed to stabilize during 5 min before use. Due to noise, values below 0.010 Lux were not considered. The measurements of natural irradiances were made following JOHNSEN (2012), during the twilight, placing the cosine detector about 1.5 m above the ground (nearly the average height of flying flashing B. lividipennis fireflies), and directing it at 90°, to measure the straight downwelling light from the sky, at 135° to measure the upwelling irradiances from the grass, at 45° to measure the zenith skylight irradiance after the sunset, and directly toward the artificial light source. The measurements were taken in three distinct points several evenings in the same places, mostly during moonless evenings and nights and averages and variations were reported. During the first years of this study (1996–2016) we measured just the downwelling skylight. Then, from 2017 we recorded both upwelling and downwelling measurements. Downwelling light have usually been criticized as a main parameter of environmental light (JOHNSEN, 2012) whereas upwelling light coming from the grass is a better parameter to measure environmental light, especially during the onset of activity of fireflies. However, because during the first decades we just made downwelling measurements, which expectedly have some proportionality with upwelling irradiances, we decided to include all dowelling light measurements made during the past 30 years, besides the upwelling irradiances made during the past 7 years. For comparison, the measurements were also taken during cloudy and clear sky evenings (JECHOW et al., 2017).
8. Measurement of environmental light spectra
Environmental (NNL) and artificial nightlighting (ALAN) illuminance and reflectance spectra were measured using a portable field spectrometer Ocean Optics QE65 Pro (OCEANOPTICS, 2010) with a refrigerated CCD camera. The spectra were scanned from 200–1000 nm and integrated for 100 ms, and the average of 3 scans of were reported. The spectra were obtained by placing the fiber optic detector about 1.5 m above the ground and facing it to 4 distinct angles as described above: at 135° to measure the upwelling irradiances from the grass, at 90° to measure the downwelling light from the sky, at 45°, in order to measure a natural upcoming irradiance from the horizon after the sunset, and directly toward the artificial light source. Several spectra were taken in a single representative point of the observation site along the first 30 min after sunset.
9. Bioluminescence spectra
Bioluminescence spectra were first measured in a Hitachi F4500 spectrofluorometer with the excitation lamp shut off, in the module “luminescence”. The spectra were scanned from 450–700 nm at the scanning speed of 2400 nm/min. We also used a LumiSpetra CCD based spectroluminometer (ATTO, Japan) which has a more constant spectral photoresponse and instantaneously scan the spectra, avoiding artifacts cause by kinetics. Unfortunately, it was not possible to measure the spectra of live firefly flashes, therefore we proceeded with our former experimental procedure to induce constant glows. Live fireflies were anesthetized with chloroform during 1 min and then injected with ca 50 µl epinephrine under the lantern cuticle to stimulate a continuous light emission, according to VIVIANI AND BECHARA (1995). The spectra were taken at 22°C in the first 2 minutes after adrenaline injection, to avoid time-dependent spectral changes that may occur due to physiological effects, such as intracellular pH changes that affect the spectrum of these fireflies. The spectra were scanned 3 times for each male specimen and 3 individuals were used. The estimated error associated with the bioluminescence spectra peak was ±2.5 nm. Slight differences of peak positions were observed from previously published data (VIVIANI & BECHARA, 1995). The measured peak at 564 nm using the Hitachi spectrofluorometer is in agreement with previous results (VIVIANI AND BECHARA, 1995), which were also obtained with a spectrofluorometer. On the other hand, the spectra obtained with the CCD based spectroluminometer (570 nm) yielded a more red-shifted peak which shall be considered more reliable due to the instantaneous scanning and also the more steady spectral photoresponse across the spectrum.
Normalization of the environmental light (NNL), high pressure sodium lamp (HPS) artificial nighlighting (ALAN) and bioluminescence spectra was made to analyze the overlapping of bioluminescence spectra with environmental illuminance. The spectra were normalized and then superimposed. The integration of the overlapping areas was considered. To calculate the overlap values, the Origin 8 software was used.
Results
1. Habitat and biogeographical aspects
This firefly (Fig. 1) has been commonly found in flattened swampy fields around lakes and shallow water courses rich in short grasses such as Brachiaria spp. often besides cattails (Typha latipholia) belts (Fig. 2) throughout Brazil (Fig. 3), in the following states: São Paulo, National Emas Park in Goiás, Mato Grosso do Sul, Tocantins, Rondônia. It is noteworthy that this species was always found in impacted flattened swampy areas under direct or indirect anthropic influence, especially swampy pastures surrounding cattail belts. In most cases, such environments are not natural despite being found in rural areas, and may have arosen from deforestation around small water courses during the past 2 centuries, suggesting the possibility that this firefly may have spread throughout Brazil quite recently.
Seasonality. Fig. 4 shows the seasonality of these fireflies during the past 30 years. In the eighties and nineties, flashing individuals were usually observed from November through March in Campinas municipality (VIVIANI, 2001). Usually the peak of the season, with the highest number of flashing fireflies, were observed during the months of January and February, whereas in March, a few remaining active flashing fireflies could be observed. However, in the past two decades, from 2005 to 2025, the period of activity of this species appeared to have been somewhat extended from October through April (Fig. 4), and the peak of activity was usually observed in February and March. This seasonal shift could be related to the climate changes, in which the average temperatures, especially in the mild climate months of fall (April-June) and winter (June-September) gradually increased, and the dry season extended to October. Furthermore, during the year of 2017, when there was a severe drought with the highest temperatures historically reported in Sorocaba municipality (~38–40°C at 18:00), active adults were observed beginning in September at the Campolim park, a fact that contrasts with all our historical recording.
Flashing pattern and activity ofB. lividipennis. The adult males usually fly quite low, close to the grass, at the height of 1 m above the ground, emitting fast rows of single flashes (Fig. 5). Females stay perched on bushes answer to the male´s flashes, with a slightly more continuous signal. Sometimes, several males can be found perched in a bush flashing near a female. Below 19° C, the fireflies stay mostly perched on the grass, eventually flashing.
Occurrence of flashing B. lividipennis fireflies at Campolim in a site under influence of ALAN: (A) onset of activity; (B) after some minutes; (C) after 20 min; (D) under effect of HPS.
The onset of flashing activity for B. lividipennis fireflies was observed over 4 decades in several locations. These fireflies consistently begin their flashing activity about 10–20 minutes after sunset (~0.903 ± 0.23 CREP units), when there is still a considerable amount of natural light (downwelling: 2–6 Lux; upwelling: 0.5–1.3 Lux; Table 1). The period of flashing activity of this firefly, as measured when the first and after the last flying flashing male individuals were observed, occurs between ~10 and ~60 min after sunset, and the peak of activity is usually observed between 20 and 30 min after sunset (Fig. 6). Usually, after 50 min the activity decreases considerably, but eventual flashing fireflies can be seen until 80 min. Considering that the duration of the civil twilight at the latitude of São Paulo region (~22°) lasts ~25 min (the astronomical sunset is ~80 min) (NIELSEN, 1963), during the reproductive season of this firefly from October through March, the activity of this firefly spans the twilight and beginning of night period.
Ocurrence of Bicellonycha lividipennis fireflies in different sites under anthropic influence São Paulo state.
2. Bioluminescence color
The bioluminescence color is in the yellow-green region, with a measured peak of 564 nm when using spectrofluorometer, and 570 nm when using a CCD-based spectroluminometer (Fig. 7).
3. Occurrence of fireflies in sites under anthropic influences
The B. lividipennis fireflies were observed in several swampy places since 1981. These places included partially degraded areas under direct or indirect anthropic influence, such as sites near houses in the periphery of the cities, being especially common near or within swampy pastures (Table 1). Of a total of 21 investigated sites, 9 were located inside urban areas, 6 in the periphery of urban areas and 6 in rural areas. These fireflies were first observed and collected in small marshy field (~1000 m2), often tramped by walking children, inside the elementary school Regina Coutinho Nogueira in Campinas municipality (Table 1), where they were very abundant in February during the years of 1981 and 1982. At that time, the author did not measure light intensities, but reports that there was some influence of the artificial nightlighting (ALAN) coming from the school lamps in the open corridors and classrooms. These fireflies were also observed in a much larger marshy field near the school in the neighborhood in Parque Taquaral, and in many other localities with marshy fields including Jd. Palmeiras besides a street lights and houses, Jd. Paineiras, horse pasture at Santana farm in Campinas municipality periphery, in a small bog surrounded by sugarcane plantations in Limeira municipality periphery, in Rio Claro city, and several places in Sorocaba downtown and Votorantim (Fig. 5), including along Sorocaba´s river and some places along its tributaries.
4. Natural illumination during flashing activity
The onset of flashing activity of these fireflies begins when there is still a considerable amount of natural light (downwelling: 2.6 ± 0.93 Lux, Range: 1.3–6 Lux; upwelling 135°: 1.24 ± 0.716; Range: 0.3–2.3 Lux; Fig. 5; Table 2), approaching up to 6 lux at the onset of flashing activity. The light coming from the horizon averaged 3.5 ± 1.15 Lux (Range: 1.55–6.6 Lux). During the peak of activity, the downwelling irradiance is between 0.300–0.500 Lux. At the end of activity, after 80 min, the dowelling irradiance decayed to <0.030 Lux in naturally illuminated environments. Comparatively, the downwelling light measured during periods of moonlighting, at night, varies from 0.030 to >0.300 Lux (full moon), in agreement with RICH AND LONGCORE (2006). Therefore, the influence of moonlight during twilight and activity period of this firefly is almost negligible.
5. Occurrence in sites under artificial nightlighting
Of a total of 17 investigated sites in distinct municipalities in São Paulo state, 8 where under different degrees of direct artificial nightlighting (ALAN). Among them, 6 were under direct road high pressure sodium lamp (HPS) illumination (~5–15 m distance from the firefly site) and 3 under the skyglow of the city, especially during cloudy evenings.
Among the investigated areas in Sorocaba region, only the marshy fields at UFSCAR Campus, the Ipanema Farm and Salto de Pirapora were out of reach of direct ALAN (Table 1). The other locations were under the direct or indirect (skyglow during cloudy evenings) influence of street ALAN (Table 2 and Table S1).
Within the 6 areas under the direct influence of HPS ALAN, the downwelling illuminance at the onset of activity averaged 6.15 ± 1.55 Lux, and ranged from 4.5 to 8.6 Lux (Table 2). The upwelling irradiance averaged 2.65 ± 0.95 Lux, and ranged from 1.6–3.8 Lux. At the end of activity, the downwelling irradiance average was 0.657 ± 0.34 Lux, ranging from 0.080 to 1.2 Lux, whereas the upwelling light reflected by the grass averaged 0.272 Lux, ranging from 0.05 to 0.50 Lux.
Near the end of activity, during cloudy nights in such places under indirect influence of ALAN reflected form the city sky, the downwelling sky light ranged from 0.040 to 0.200 Lux. The direct illuminance coming from HPS sources varied from 0.090–0.190 Lux for the mild illuminated sites, up to 6.8 Lux for the most illuminated sites at Jd. Clarice and Parque Campolim (Fig. 8).
Comparison between the bioluminescence spectrum of Bicellonychia lividipennis, environmental and light sources spectra: (A) Upwelling 135°; (B) Downwelling; (C) Upwelling 45°; (D) superimposed Bioluminescence and HPS spectra.
Period of flashing activity in artificially illuminated areas. We found evidences that the flashing activity of fireflies is somewhat delayed in highly illuminated sites. Whereas the peak of flashing activity in environments without artificial night lighting occurred on average 30 min after sunset (downwelling < 0.9 lux) and extended up to 60 min after sunset (< 0.03 lux), in the areas under direct influence of more intense HPS (4 to 6 Lux) the period of activity was postponed about 10 min in the beginning, and even later at the end of activity: (onset) 20 to 30 minutes after the sunset (Fig 6), (peak) 40–45 min after sunset and (end) 180 min after sunset. However, more studies are necessary to measure the extension of time under such circumstances.
6. Spectral composition of the environmental and artificial night lighting in firefly sites
To analyze the influence of the spectral composition of the artificial (ALAN) and natural (NNL) environmental night lighting during the flashing activity period of this firefly, we recorded the environmental irradiance spectra at the time of the onset up and peak of the flashing activity of this firefly (Fig. 8). The lower irradiances at the end of activity of this firefly hampered the possibility to measure environmental spectra, due to the sensitivity of the equipment.
The natural downwelling sky light spectrum measured at the onset of activity of this firefly has a main broad band in the blue-green and another in the amber-red region (~678 nm; Fig. 8B), with a valley in the yellow-green region (550~570 nm), just in the region where the bioluminescence spectrum of B. lividipennis peaks (Fig. 8B). The upwelling light from the grass, on the other hand, is predominantly in the green region but, again, showing a main valley in the yellowgreen region (Fig. 9C). About 30 min after sunset, during the peak of activity of B. lividipennis, the green and especially the red components decreased, while the blue one becomes predominant (Fig. 8B).
Regarding the HPS, whereas this source has sharp and defined lines peaking in the yellow (568 nm and 582 nm) and orange/amber (595 and >612 nm) regions, the irradiance reflected by the grass in sites under influence of HPS (~10–20 m) is somewhat diluted (Fig. 8D) contributing for the higher proportion of light in the amber-red region.
Discussion
Overall, the results show that B. lividipennis firefly is an opportunistic firefly which easily adapts to open swampy fields under indirect anthropic influence, such as pastures and partially illuminated sites near or inside urban areas. In the past centuries, most of Southeastern Brazil was covered by forests like Atlantic and gallery forests, whereas original open marshy areas were more restricted near larger rivers. During the past decades, perhaps during the last two centuries, upon gradual deforestation, the swampy fields dominated by short grasses proliferated along smaller tributaries and this species may have had the chance to further spread especially along such smaller water courses close to pastures. The occurrence near or inside swampy areas affected by cattle, indicate that, somehow, this firefly finds a favorable environment. Perhaps the larvae may find abundance of preys that proliferate near cattle dung, such as nematodes and anellids.
Observation sites and environmental irradiances during the periods of activity of flashing fireflies observed from 2008 to 2020 in Sorocaba region.
This firefly is twilight active. However, in Southeastern Brazil latitudes the twilight duration is quite short (~24 min), and the activity of this firefly (60 min) spans both the twilight and beginning of night. Therefore, this species is likely to be more prone to be affected by artificial nightlighting (ALAN), mainly from the peak to the end of its natural activity. The bioluminescence spectrum of this species, peaks in the yellow-green region (564–570 nm; Fig. 7), clearly displaying a reasonable contrast with either the downwelling and upwelling environmental illuminance spectra at twilight (Fig. 8D), with lower overlap with the photic environment spectra. Similar contrast between BL and relative spectral irradiance at twilight were already reported for the North-American Photinus pyralis firefly, whose BL spectrum peak (563 nm) is close to that of B. lividipennis (SELIGER et al., 1982). This is in good agreement with the bioluminescence contrast and visual sensitivity optimization hypothesis, which predicts that yellow emitting fireflies are better adapted to the twilight to overcome the photic effect of the grass (LALL et al., 1980; LALL et al., 1980b). However, this species is not just a typical twilight active species like North-American species, but span the very short twilight period and the beginning of night.
It is possible that this firefly, similarly to the North-American twilight fireflies, also has photostable screening pigments in the eyes, which filter light reflected by grass in the green region during twilight, maximizing the detection of bioluminescent signals in the yellow region (LALL, 1981). Further studies about the spectral sensitivity of vision of this firefly, will help to understand the effect of ALAN in this firefly.
This firefly was found to be clearly adapted to environments under influence of ALAN, especially high pressure sodium lamps (HPS) which has predominance in the amber region of the spectrum. This is noteworthy, because recent studies using LEDs with North-American twilight fireflies showed that amber light, which is similar to HPS, is more disruptive than dim blue light for firefly flashing activity (OWENS et al., 2021, 2022).
However, when considering the environmental downwelling and upwelling irradiance spectra at twilight, especially the downwelling light coming from the horizon (sunset), which is predominantly in the amber-red region (CARAVEO 2024; JOHNSEN, 2012), there is a considerable overlap with the HPS spectral lines in the yelloworange region, suggesting that HPS may, at least partially, simulate the natural background irradiance during early twilight. On the other hand, despite the existence of considerable overlap between B. lividipennis bioluminescence spectrum and some HPS spectral lines (~50%), especially with in the yellow (568 and 582 nm), and also in the orange regions (595 nm), there are steep valleys between these spectral lines with lower overlap with the bioluminescence spectrum (Fig. 8). Furthermore, below the 568 nm spectral line, there is almost no overlapping with the bioluminescence spectrum of this species. Therefore, the area of highest intensity of the BL spectrum (550–565 nm) lies just in the valley between the environmental irradiance spectrum band in the green region (500–550 nm) reflected by the grass at twilight, and the yellow spectral line of HPS (>568 nm).
Thus, the occurrence of adapted populations of B. lividipennis in sites under HPS can be rationalized as follows: (1) this species begins its flashing activity when there is still considerable higher natural downwelling environmental lighting (≤8 Lux), and therefore also upwelling light reflected by the grass (~1.0–2.0 Lux), in most cases overcoming the local ALAN; (2) because HPS has considerable overlap with the twilight environmental light spectra, especially in the amber region after sunset, it may partially simulate the natural photic environment at the onset of the activity of this species, postponing the end of its activity; (3) the spectral composition of HPS lighting, despite displaying considerable overlap with the bioluminescence spectrum of this species above 568 nm in the yellow/amber region, displays a deep valley below 568 nm, with almost no overlap with the bioluminescence spectrum, leaving a narrow window in the yellowgreen region (550–565 nm) where the bioluminescence signal still contrast with either the environmental upwelling light and HPS, potentially maximizing the visualization of bioluminescence.
Therefore, these results show, for the first time, that HPS may have lower impact on twilight active fireflies during the short twilight of tropical latitudes, provided a similar bioluminescence spectrum in the yellow-green region (564–570 nm), a hypothesis that remain to be tested with other species and confirmed after measuring the visual sensitivity spectrum of this species and experimental tests with different LED sources. However, as already shown with experimental studies with other fireflies (OWENS et al., 2018, 2022), this type of ALAN may have negative effects on other species, especially twinight and night active fireflies that display different bioluminescence spectra and later flashing activity.
Concluding Remarks
The firefly Bicellonycha lividipennis Motschusky 1854, is an opportunistic twilight firefly adapted to open flattened marshy areas throughout Brazil, being often observed in areas under direct or indirect anthropic influence, including pastures partially tramped by cattle and sites under ALAN near or inside urban areas. The occurrence of populations of this firefly in sites under influence of high pressure sodium lamps (HPS) artificial nightlighting (ALAN), indicate that this species tolerates such kind of ALAN under irradiance levels that simulate the twilight at their onset of activity (2–6 Lux). The HPS may also partially simulate the natural illuminance after sunset at the onset of activity of this firefly during twilight. Therefore, the adaptation of this firefly to sites under HPS can be explained by: (1) the early onset of activity of this firefly when there is still high natural environmental irradiances, overcoming those of the HPS; (2) the partial overlap of the twilight environmental irradiance and the HPS spectra, especially in the yellow-amber region, simulating the natural illumination during the activity at twilight, and (3) the lower overlap between the BL spectrum in the yellow-green region between 550–565 nm, with either the natural (<550 and > 600 nm) and the HPS (> 568 nm) irradiance spectra, providing a contrasting window for the BL signal under such photic circumstances. The results indicate for the first time that HPS could be less disruptive for tropical meadows and twilight active fireflies which emit in the yellow-green region.
Supplementary Material
The following online materials are available for this article:
Figure S1
Figure S2
Table S1
Acknowledgements
We acknowledge Prof. Dr. Petrere Jr. by guidance on sampling methods and Prof. Dr. Mauricio Cetra for help in analysis of B. lividipennis. We appreciate the considerations of Profa. Dra. Simone P. Rosa and Prof. Dr. Pedro José Ferreira Filho. We are grateful to FAPESP (2010/05426-8; 2023/16675-9) and CNPq (405060/2021-1) for funding: FAPESP Master Scholarship (2011/16653-8) and CAPES Master scholarship.
Data Availability
Raw data were deposited in Dataverse: https://doi.org/10.48331/scielodata.C6VSGE.
References
- CARAVEO, P. 2024. Troppa Luce fa male: I pericoloi dell´illuminazione artificiale. Edizioni Deadalo.
- COSTIN, K.J. & BOULTON, A.M. 2016. A field experiment on the effect of introduced light pollution on fireflies (Coleoptera: Lampyridae) in the Piedmont Region of Maryland. The Coleopterists Bulletin, 70: 84–86.
- CRONIN T.W., JARVILHTO M., WECKSTROM M. & LALL A.B. 2000. Tuning of photoreceptor spectral sensitivity in fireflies. Journal of Comparative Physiology A, 186:1–12.
- DREISIG H. 1974. Environmental control of the daily onset of luminescent activity in glowworms and fireflies (Coleoptera: Lampyridae). Oecologia, 18: 85–99.
- GASTON, K.J., GASTON, S., BENNIE, J. & HOPKINS, J. 2014. Reducing the impacts of artificial light. British Wildlife, 25(5): 332–339.
- HAGEN O., SANTOS R.M., SCHLINDWEIN M.N. & VIVIANI V.R. 2015. Artificial Night Lighting Reduces Firefly (Coleoptera: Lampyridae) Occurrence in Sorocaba, Brazil. Advances in Entomology, 3: 24–32.
-
JECHOW A., KOLLÁTH Z., RIBAS S.J., SPOELSTRA H., HÖLKER F. & KYBA C.C.M. 2017. Imaging and mapping the impact of clouds on skyglow with all-sky photometry. Scientific Reports, 7(1). doi:10.1038/s41598-017-06998-z
» https://doi.org/10.1038/s41598-017-06998-z - JOHNSEN S. 2012. The Optics of Life: a biologist guide to light in nature. Princeton Univ. Press New Jersey.
- LALL A.B., CHAPMAM R.M., TROWTH C.O. & HOLLOWAY J.A. 1980a. Spectral mechanism of the compound eye in the firefly Photinus pyralis (Coleoptera: Lampyridae). Journal of comparative physiology, 135: 21–27.
-
LALL A.B., SELIGER H.H., BIGGLEY W.H. & LLOYD J.E. 1980b. Ecology of Colors of Firefly Bioluminescence. Science, 210(4469): 560–562. doi:10.1126/science.210.4469.560.
» https://doi.org/10.1126/science.210.4469.560 -
LALL A.B. 1981. Electroretinogram and the spectral sensitivity of the compound eyes in the firefly Photuris versicolor (Coleoptera-Lampyridae): A correspondence between green sensitivity and species bioluminescence emission. Journal of Insect Physiology, 27(7): 461–468. doi:10.1016/0022-1910(81)90097-4.
» https://doi.org/10.1016/0022-1910(81)90097-4 -
LALL A.B., LORD E.T. & TROUTH C.O. 1982. Vision in the firefly Photuris lucicrescens (Coleoptera: Lampyridae): Spectral sensitivity and selective adaptation in the compound eye. Journal of Comparative Physiology, 147: 195–200. doi:10.1007/BF00609844.
» https://doi.org/10.1007/BF00609844 - LALL A.B. 1993. Action spectra for the initiation of bioluminescent flashing activity in males of Twilight firefly Photinus scintillans (Coleoptera: Lampyridae). Journal of Insect Physiology, 39: 123–127.
-
LALL A.B., VIVIANI V.R. & VENTURA D.F. 2023. Spectral tuning of bioluminescence and visual sensitivity in males of Brazilian firefly species inhabiting dim light environments (Coleoptera: Elateroidea: Lampyridae). Journal of Experimental Zoology A Ecological Integrative Physiology, 339(1): 37–45. doi: 10.1002/jez.2647.
» https://doi.org/10.1002/jez.2647 -
LEWIS S.M. & CRATSLEY C.K. 2008. Flash Signal Evolution, Mate Choice, and Predation in Fireflies. Annual Review of Entomology, 53(1): 293–321. doi:10.1146/annurev.ento.53.103106.093346.
» https://doi.org/10.1146/annurev.ento.53.103106.093346 -
LEWIS S.M., WONG C.H., C.M., OWENS A.C.S., FALLON C., JEPSEN S., THANCHAROEN A., WU C., COCK R., NOVÁK M., LÓPEZ-PALAFOX T., KHOO V. & REEDJ.M., 2020. A Global Perspective on Firefly Extinction Threats, BioScience, 70 (2): 157–167. doi.org/10.1093/biosci/biz157.
» https://doi.org/10.1093/biosci/biz157 - LLOYD J.E. 1983. Bioluminescence and communication in insects. Annual Reviews Entomology 38: 131–160.
- LLOYD, J.E. 2006. Stray light, fireflies, and fireflyers. Ecological consequences of artificial night lighting, 345–364.
- NIELSEN E. 1963. Illumination at twilight. Oikos 14: 9–21.
-
OWENS A.C.S., MEYER-ROCHOW V.B. & YANG E.C. 2018. Short- and mid-wavelength artificial light influences the flash signals of Aquatica ficta fireflies (Coleoptera: Lampyridae). PLOS ONE, 13(2): e0191576. doi:10.1371/journal.pone.0191576.
» https://doi.org/10.1371/journal.pone.0191576 - OWENS A.C.S. & LEWIS S.M. 2021. Narrow spectrum artificial light silences female fireflies (Coleoptera: Lampyridae). Insect Conservation & Diversity 14(2): 199–210.
- OWENS A.C.S., DRESSLER C.T. & LEWIS S.M. 2022. Cost and benefits of “insect friendly” artificial lights are taxon specific. Oecologia 199: 487–497.
- RICH C. & LONGORE T. 2006. Ecological Consequences of Artificial Night Lighting. Washington DC: Island Press.
- SANTOS R.M. 2013. Vaga-lumes bioindicadores de poluição luminosa e riqueza de espécies em áreas naturais da serra de Paranapiacaba e em áreas urbanas. 2013. 109 f. Dissertação (Mestrado em Diversidade Biológica e Conservação) – Programa de Pós-Graduação em Diversidade Biológica e Conservação, Universidade Federal de São Carlos – Campus Sorocaba.
- SELIGER H.H., LALL A.B., LLOYD J.E. & BIGGLEY W.H. 1982a. The colors of firefly Bioluminescence I: Optimization model. Photochemistry and Photobiology, 36: 673–681.
- SELIGER H.H., LALL A.B., LLOYD J.E., & BIGGLEY W.H. 1982b. The colors of firefly Bioluminescence II: Experimental evidence for the Optimization model. Photochemistry and Photobiology, 36: 681–688.
-
VAZ S., MANES S., GAMA?MAIA D., & SILVEIRA L. 2021. Light pollution is the fastest growing potential threat to firefly conservation in the Atlantic Forest hotspot. Insect Conservation and Diversity, 14(2): 211–224. doi: 10.1111/icad.12481.
» https://doi.org/10.1111/icad.12481 - VIVIANI V.R. & BECHARA E.J.H. 1995. Bioluminescence of Brazilian fireflies (Coleoptera: Lampyridae): spectral distribution and pH effect on luciferase-elicited colors. Comparison with elaterid and phengodid luciferases. Photochemistry and Photobiology, 62: 490–495.
-
VIVIANI V.R. 2001. Fireflies (Coleoptera: Lampyridae) from Southeastern Brazil: Habitats, Life History, and Bioluminescence. Annals of the Entomological Society of America, 94(1): 129–145. doi:10.1603/0013-8746(2001)094[0129:fclfsb]2.0.co;2
» https://doi.org/10.1603/0013-8746(2001)094[0129:fclfsb]2.0.co;2 -
VIVIANI V.R., ROCHA M.Y. & HAGEN O. 2010. Fauna de besouros bioluminescentes (Coleoptera: Elateroidea: Lampyridae; Phengodidae, Elateridae) nos municípios de Campinas, Sorocaba-Votorantim e Rio Claro-Limeira (SP, Brasil): biodiversidade e influência da urbanização. Biota Neotropica, 10(2): 103–116. doi:10.1590/s1676-060320 10000200013.
» https://doi.org/10.1590/s1676-06032010000200013
