Artificial lighting reduces the effectiveness of wildlife-crossing structures for insectivorous bats

In an attempt to improve cost-effectiveness, it has become increasingly popular to adapt wildlife crossing structures to enable people to also use them for safe passage across roads. However, the required needs of humans and wildlife may conflict, resulting in a structure that does not actually provide the perceived improvement in cost-effectiveness, but instead a reduction in conservation benefits. For example, lighting within crossing structures for human safety at night may reduce use of the structure by nocturnal wildlife, thus contributing to barrier and mortality effects of roads rather than mitigating them. In this study, we experimentally evaluated the impact of artificial light at night on the rate of use of wildlife crossing structures, specifically underpasses, by ten insectivorous bat species groups in south-eastern Australia. We monitored bat activity before, during and after artificially lighting the underpasses. We found that bats tended to avoided lit underpasses, and only one species consistently showed attraction to the light. Artificial light at night in underpasses hypothetically increases the vulnerability of bats to road-mortality or to the barrier effect of roads. The most likely outcomes of lighting underpasses were 1. an increase in crossing rate above the freeway and a decrease under the underpasses, or 2. a reduction in crossing rate both above freeways and under the underpasses, when structures were lit. Our results corroborate those of studies on terrestrial mammals, and thus we recommend that underpasses intended to facilitate the movement of wildlife across roads should not be lit.

Testing the feasibility of wireless sensor networks and the use of radio signal strength indicator to track the movements of wild animals

Context Wireless sensor networks (WSNs) are revolutionising areas of animal behaviour research and are advantageous based on their ability to be deployed remotely and unobtrusively, for long time periods in inaccessible areas. Aims We aimed to determine the feasibility of using a WSN to track detailed movement paths of small animals, e.g. rats (Rattus spp.) 100–400g, too small for current GPS technology, by calibrating active Radio Frequency Identification (RFID) tags and loggers using Radio Frequency Signal Strength Indicator (RSSI) as a proxy for distance. Active RFIDs are also called Wireless Identification (WID) tags. Methods Calibration tests were conducted using a grid of loggers (n=16) spaced at 45-m intervals in clear line-of-sight conditions. WID tags (n=16) were placed between the loggers at 45-m intervals. Eight ‘walks’ were also conducted through the grid using a single WID tag. This involved attaching the tag to a small bottle of water (to simulate the body of an animal), towed around the grid using a 1-m long tow line attached to a volunteer walker. The volunteer also held a GPS device that logged their track. Models were constructed to test the effects of distance, tag movement and individual differences in loggers and tags on the reliability of movement data. Key results Loggers were most successful at detecting tags at distances <50m. However, there was a significant difference in the detection probabilities of individual loggers and also the transmission performance of individual tags. Static tags were less likely to be detected than the mobile tag; and although RSSI was somewhat related to distance, the reliability of this parameter was highly variable. Implications We recommend caution in the future use of current radio frequency ID tags in wireless sensor networks to track the movement of small animals, and in the use of RSSI as an indicator of individual distance values, as extensive in situ calibration is required. ‘Off the shelf’ devices may vary in performance, rendering data unreliable. We emphasise the importance of calibrating all equipment in animal tracking studies to reduce data uncertainty and error.

Longevity in wild populations of the squirrel glider Petaurus norfolcensis

Petaurus norfolcensis were trapped between December 1996 and April 2005 at 3 sites in south-eastern Australia. In 2005 nine individuals marked in previous years were captured. Age estimates at first capture were based on body mass, tooth wear and pouch condition. Most recaptured individual were propertly assigned to the correct age category (>3 years), but the age of 2 were underestimated. This indicates uncertainty in using physical characteristics, chiefly wear of upper incisors, to estimate age in P. norfolcensis.One individual captured in 2005 was over 7 years 9 months old. This compares to captive P. breviceps and P. australis records of 9 years and 16 years respectively.

The influence of time since fire and distance from fire boundary on the distribution and abundance of arboreal marsupials in Eucalyptus regnans-dominated forest in the Central Highlands of Victoria

The impact of time since fire after two consecutive wildfires 44 years apart (1939 and 1983) within the same area, and the distance from the fire boundary (<100 m or 500-2000 m), were investigated in relation to the distribution and abundance of arboreal marsupials in 1994. Arboreal marsupials were censused by stagwatching and spotlighting in two relatively young age classes of mountain ash (Eucalyptus regnans) dominated forest in the Central Highlands of Victoria. Five species of arboreal marsupial were detected, but only three were detected in sufficient numbers to determine habitat preferences. Petauroides volans (greater glider) was statistically more abundant in 1939 regrowth forests, while Trichosurus caninus (mountain brushtail possum) showed no significant preference for either age class of forest. All but one record of Gymnobelideus leadbeateri (Leadbeater's possum) came from young forest, though the effect of age-class was not statistically significant. Distance from fire boundary explained little or no variation in mammal distribution or abundance. While the actual number of hollow-bearing trees was similar in both age classes of forest, the long-term lifespan of hollow-bearing trees in more recently burnt forest is predicted to be lower than in unburnt or not recently burnt forest. Post-fire salvage logging following the 1983 wildfires appears to have reduced the number of hollow-bearing trees at sites burnt in 1983.