posted on 2024-11-15, 13:14authored byJohn R Gollan, Michael Ashcroft, Daniel Ramp
Investigating responses to climate often rely on macroclimatic models. This is problematic because of the potential to miss or wrongly attribute relationships. Here we compare the explanatory power of macroclimatic models and near-surface topoclimatic models. Body-size measurements of the ant species, Iridomyrmex purpureus, were collected from separate colonies spanning a range of climatic conditions in a large region (∼75,000 km2) of Australia. Regional regression was used to derive two topoclimatic variables, while ANUCLIM was used to derive macroclimatic variables. Relationships were tested using linear mixed-effect models with Akaike information criterion used as an indication of the relative goodness of fit for each model. Significant trends for both topoclimatic variables with body size were detected but only one of the three macroclimatic variables showed a significant trend. Although the significant macroclimatic variable was correlated with one of the topoclimatic variables, the topoclimatic variable had greater explanatory power. Few studies have considered climatic data accuracy or the effects of inaccurately quantified climatic data on ecological theory. This cannot continue to be ignored. As we show in this study, there is potential for important trends to go undetected and interpretation of results to be completely different.
History
Citation
Gollan, J. R., Ashcroft, M. B. & Ramp, D. (2013). Fine-grained climate data alters the interpretation of a trait-based cline. Ecosphere, 4 (12), 154-1 - 154-10.