Roads are responsible for thousands of roadkill per year worldwide. Different taxonomic groups are vulnerable to this mortality source, but amphibians are especially vulnerable. In fact, very high mortality rates can occur during time concentrated periods when individuals migrate to and from reproduction areas (e.g., ponds). In addition, the activity of this taxonomic group is dependent on weather conditions, also influencing roadkill patterns. In this study we aim to assess the influence of the lunar cycle on the number of six species of amphibians killed on the road each night and total number of amphibian roadkill, while accounting for weather conditions of each night (minimum temperature, precipitation in previous hours, relative humidity, wind). Amphibian roadkill were monitored daily by car (30-40 km/h) for two years (from March 2010 to May 2012) along 37 km of National and Municipal roads in Southern Portugal (Évora district). The data here presented concern the species with higher roadkill rates in the study area (two urodeles and four anurans): Sharp-ribbed Salamander (Pleurodeles waltl), Fire Salamander (Salamandra salamandra), Iberian Painted Frog (Discoglossus galganoi), Western Spadefoot toad (Pelobates cultripes), Spiny Common Toad (Bufo spinosus), and Natterjack Toad (Epidalea calamita). We
performed Generalized Linear Mixed Models for zero inflated and over dispersed data, in order to evaluate the importance of the lunar cycle and weather variables in each species and total mortality. We excluded the period of June to August after data exploration, as those months corresponded to minimum values of amphibian roadkill. We assumed that roadkill events are a result of both breeding and non-breeding movements. Overall, the increase in minimum temperature, relative air humidity and rainfall in the previous hours (8-24 h) to road surveys increases the probability of higher roadkill numbers for all studied species. However, the lunar cycle also influenced roadkill numbers of some species. Darker nights had higher numbers of roadkill of Pleurodeles waltl, while moonlit nights had higher numbers of Salamandra salamandra. There were higher roadkill numbers for Pelobates cultripes in Last Quarter nights when compared with First Quarter. Higher total roadkill numbers were mainly determined by a combination of wet and relatively warm nights: higher maximum and cumulative rainfall, relative air humidity and minimum temperature in previous hours of road surveys. Additional lunar effects on number of roadkill were detected for three species, although these effects were species-specific. Animals that are more active in moonlight may be at an advantage if their visual acuity is better than that of their predators. So, differences between species in the response to moonlight may be due to species’ ecology, physiology or differences perceived in predation risk. Temporary mitigation actions, such as amphibian catches by volunteers, warning signs or temporary road closures, can be more successful if conducted at the right time (i.e., at periods with higher movement rates). Our work provides information to improve the timing of these actions.
Spain
