Stable isotope analysis (SIA) of modern and historical samples to investigate the potential influence of urbanization on carbon and nitrogen isotope rations in murid rodents

The study used stable isotope analysis (SIA) of modern and historical samples to: 1) investigate the potential influence of natural history traits and environmental factors that also included metrics of urbanization on carbon (<i>δ</i>¹³C) and nitrogen (<i>δ</i>¹⁵N) isotope ratios in murid rodents from Gauteng Province, South Africa; and 2) evaluate their potential trophic ecological shifts due to urbanization over a historical time scale. SIA was conducted on biologically inert fur samples of modern and historical samples (i.e., museum-preserved samples collected over a time scale) of three invasive (<i>Rattus rattus</i>, <i>R. norvegicus</i> and <i>R. tanezumi</i>) and three indigenous (<i>Aethomys ineptus</i>, <i>Mastomys coucha</i> and <i>Otomys angoniensis</i>) murid rodents. Generalized linear models (GLM), analysis of variance (ANOVA) and Tukey’s <i>post-hoc</i> analyses showed that sex, age, and sampling year had minimal impact on <i>δ</i>¹³C and <i>δ</i>¹⁵N isotope ratios, but not sampling locality and season. These results were supported by the analysis of percentage trophic overlaps based on sample size-corrected standard ellipse areas (SEAc). Collectively, these results suggest that in the last century, short-term seasonal fluctuations accounted for a larger proportion of isotopic variation than long-term environmental changes. The results also showed that qualitative categorical land use data as a proxy for urbanization did not adequately account for changes in rodent isotope ratios. Consequently, an attempt was made to identify quantifiable potential sources of short-term variation in <i>δ</i>¹³C and <i>δ</i>¹⁵N stable isotope ratios in these rodents. The identified quantitative variables included monthly median temperature and monthly total rainfall as short-term metrics of environmental change and Normalized Difference Vegetation Index (NDVI) as a metric for vegetation cover and impervious surface area (ISA), both as proxies for urbanization over a historical time scale. The analyses found that both metrics of short-term environmental factors and proxies of urbanization had no impact on <i>δ</i>¹³C ratios in any of the three invasive and three indigenous murid rodent species investigated, while <i>δ</i>¹⁵N ratios of the insectivorous indigenous <i>A. ineptus </i>and <i>M. coucha</i> were found to be influenced by monthly median temperature and impervious surface area. Overall, the largest %SSQs were due to <i>error </i>(i.e., residual), suggesting that apart from the metrics analysed and the interaction between them, there may be other unexplained components that are also responsible for the variation. These results suggest that the murid rodent populations investigated may be shielded from the negative impacts of short-term environmental factors and urbanization through the availability of excess food resources in one of the most urbanized landscapes in South Africa. These results also highlight the potential impact of urban landscapes on the trophic ecology of invasive and indigenous murid rodents and their negative impacts on the conservation of other threatened species within urban settings.