Search

Variable seasonal stream temperatures are a critical factor in maintaining aquatic invertebrate community patterns. We investigated whether the degree of predictability in a stream's water temperature profile provides insights into the structure and functional predictability of macroinvertebrate communities. Quarterly macroinvertebrate sampling over 11 months at five paired sites along the longitudinal axes of the Keurbooms River, Western Cape, and the Kowie/Bloukrans River, Eastern Cape, was undertaken in conjunction with collection of hourly water temperatures. Macroinvertebrate taxonomic turnover across seasons was higher for sites having lower water temperature predictability values than for sites with higher predictability, while temporal partitioning was greater at sites with greater temperature variability. Macroinvertebrate taxa responded in a predictable manner to changes in their environment. Specialist taxa preferred sites with greater temperature variability, and generalist taxa those with less temperature variability. Water temperature is an important abiotic driver of stream ecosystems, and should be expressed in terms of predictability metrics, as mean values do not adequately measure the components of water temperature that are important for biotic communities. Water temperature predictability provided an indication of the structure, functional predictability and seasonal turnover in macroinvertebrate communities. These findings have implications for streams experiencing temperature modifications due to the onset of climate change. [ABSTRACT FROM AUTHOR]