Habitat quality outweighed measured foliar particulate indicators in grazer patch selection in an arid iron-ore landscape

Abstract

Open-pit mining generates airborne particulate matter that can accumulate on vegetation and potentially influence herbivore foraging behaviour. However, the relative importance of foliar particulate deposition compared with habitat quality in driving habitat selection remains poorly understood in arid ecosystems. We investigated whether semi-quantitative foliar particulate indicators derived from scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM–EDS) were associated with habitat patch selection by cattle ( Bos taurus ), sheep ( Ovis aries ), goats ( Capra hircus ), and springbok ( Antidorcas marsupialis ) in an arid shrub–grass savanna adjacent to an operational iron-ore mine in South Africa. Leaf-surface particulate signatures were quantified as elemental weight percentages (wt%) of aluminium (Al), magnesium (Mg), and silicon (Si), and compared between preferred and avoided habitat patches using stratified Wilcoxon (van Elteren) tests with animal identity treated as the stratification variable. Habitat-quality variables were subsequently evaluated alongside the measured foliar particulate indicators. Elemental signatures showed substantial overlap between preferred and avoided habitat patches. Although three element-by-growth-form comparisons were statistically significant, preferred patches consistently exhibited equal or higher elemental values than avoided patches, providing no evidence that the measured foliar particulate indicators were associated with habitat avoidance. In contrast, vegetation height and greenness showed stronger and more consistent relationships with habitat selection across species. These findings indicate that habitat quality outweighed the measured foliar particulate indicators in explaining grazer patch selection within this arid mining landscape. Because the SEM–EDS-derived elemental signatures cannot distinguish mine-derived particulates from natural background dust, these results should not be interpreted as evidence that mining-derived dust has no ecological effects. Rather, they demonstrate that the specific foliar particulate indicators measured in this study were poor predictors of observed habitat use, highlighting the need for future studies incorporating source-resolved dust measurements and quantitative particulate loading.