Je. Taylor et Bj. Fox, Assessing the disturbance impact on vegetation and lizard communities of fluoride pollution interacting with fire and mining in eastern Australia, AUSTRAL EC, 26(4), 2001, pp. 321-337
The effects of atmospheric fluoride pollution on the lizard fauna of the op
en forest of coastal dunes in New South Wales, Australia were examined. Liz
ards were pitfall-trapped at sites with background fluoride levels (< 0.25
mu mol F.g(-1)), or subject to low (1.85-3.4 mu mol F.g(-1)) or high (8.00-
13.2 mu mol F.g(-1)) levels of fluoride pollution. Sites had been disturbed
by fire or mineral sand-mining 4 or 8 years prior to the study. Fluoride p
ollution resulted in significant changes to canopy cover, understorey veget
ation density and ground cover. Where fluoride levels were low in unmined f
orest, there was significantly higher species richness, total lizard abunda
nce and abundance of the most common species, Lampropholis guichenoti and L
ampropholis delicata (Scincidae), compared with areas of high or background
fluoride levels. Both the present and previous studies show that fluoride
pollution is significantly correlated with increased abundance of the most
common lizard species in sand-mined areas, Ctenotus robustus, Ctenotus taen
iolatus (Scincidae) and Amphibolurus muricatus (Agamidae). A discriminant f
unction model of background-fluoride mined sites was used to predict lizard
abundances based on vegetation density, leaf litter density and soil hardn
ess. The model was verified by using it to predict similarities between bac
kground-fluoride sites and fluoride-affected sites. The sites within each p
redicted group were more similar in lizard species composition than when gr
ouping of sites was done by time since mining or fire. With this analysis,
a close relationship between vegetation variables and the lizard fauna, irr
espective of the type of disturbance or time since disturbance, is demonstr
ated. Discriminant function analysis suggested that Ctenotus would be unlik
ely to use unmined forest sites that had been burned within 12 months previ
ously. Thus it seems unlikely that the original open forest of these coasta
l dunes would have supported populations of either C. robustus or C. taenio
latus. In conclusion, the original source of Ctenotus that colonized sand-m
ined areas is more likely to have been the relatively small areas of heath
vegetation in the area. Therefore, sand-mining and atmospheric fluoride pol
lution result in landscape-scale changes to the relative abundance of lizar
d species, with forest species becoming less common and lizard species from
open areas becoming more common.