It is clear that there are large gaps in the existing knowledge of the impact of noise on wildlife populations. In invertebrates and lower vertebrates (fish, reptiles, amphibians) there is relatively little study on the effects of road noise with no clear indication of a strong adverse response, at least for the levels of noise likely to be encountered from road traffic. For reptiles and amphibians, effects appear to be localized and likely due to mortality or a barrier to movement. Recent studies on the effect on toads in burrows near roads strongly indicate that further study on this or similar behaviors is warranted. For birds, noise can apparently have a significant effect; however, the results are not universal with some species being adversely affected, many unaffected and still others becoming more common near even interstate highways. Mammals (particularly large species) may avoid noise, however, there is evidence (particularly for smaller species) that additional habitat and corridors for movement are provided by roadways.
The most urgent requirement is to determine why noise - the presumptive cause – has such variable effects and to determine if the effect is attributable to noise alone or if other factors and/or interactions are present. This could be addressed through introduction of appropriate noise levels into naïve areas or through studies of individual responses in controlled laboratory settings to determine where background noise is having an effect (e.g. distance of transmission of calls, ability of birds to locate others, patterns of behavior, reproductive success etc.).
Since direct masking of vocalization is unlikely to be the significant factor in many cases, future studies could also look at other indicators of stress including physiological indices such as an increase in sympathetic nerve activity affecting pupils, heart, digestive system, adrenal medulla, blood vessels and musculature (Borg and Møller c.f. Algers et al.,(3)). In stressed animals, the hypothalamus would signal an increase in ACTH (adrenocorticotropic hormone) and TSH (thyroid stimulating hormone) from the pituitary gland and the resultant changes (e.g. corticosteroid levels, blood glucose levels, electrolyte balance) could be measured either field or laboratory studies to determine the level of stress. A number of additional physiological effects of noise on animals have been summarized including changes in endocrine, digestive, blood, immune and reproductive function (see Algers et al(3); Manci et al.(81) and references therein) and could be looked at as indicators of stress and deviations in any of these (from control or reference populations) could help to explain the results seen. This approach has been suggested as a possible course of action recently.(99)
Two important points to consider in the design of studies are 1) the density of a given species is not necessarily an absolute indicator of the best habitat (i.e. sometimes individuals are relegated in significant numbers to less desirable habitat because of territoriality by dominant individuals)(118), and 2) greater behavioral response (i.e. movement away from highway) does not necessarily indicate species that are at greatest need of protection.(55) Thus, any plans for conservation must consider the quality of the habitat and the sensitivity of the population or community under consideration as well as the degree of the effect on a given species.