Associated Data for "Fluctuating asymmetry as an indicator of estuarine fish health in Queensland, Australia"

Fluctuating asymmetry (FA), the minute, random deviations from perfect symmetry in morphological characters, is a potentially useful tool for assessing fish health in estuarine environments. FA has been widely used internationally as an indicator of fish health due to developmental instability in response to various environmental pressures, both natural and anthropogenic. In Australia, however, literature on the subject is sparse and the utility of the method is not well understood. As a rapidly developing, multi-use port, Gladstone Harbour is subject to a variety of stressors from industrial, commercial and community uses. The occurrence of FA in Gladstone Harbour (defined as a Moderately Disturbed estuarine system (MD)) was investigated in this study by assessing fish captured over five seasonal sampling events in 2019-2022. Sampling was undertaken in Gladstone Harbour, as well as at some relatively unimpacted sites in Central Queensland, classified as being of High Ecological Value (HEV) in the relevant Queensland Government Water Quality Objectives. Following the assumption that increased developmental instability, and therefore poorer health, is represented by increased FA in a population, it was hypothesized that fish from the moderately impacted Gladstone Harbour would exhibit higher FA than those from the less-impacted HEV sites. Five suitable estuarine fish species were targeted (Ellochelon vaigiensis, Mugil cephalus, Neoarius graeffei, Netuma thalassina, and Pomadasys kaakan), and a range of bilaterally symmetrical morphometric traits measured. 

There is some debate as to the efficacy of FA as an indicator of fish population health, and methodological problems are evident in the large body of literature. Measurement error (ME) is an important factor in FA studies that is often unacknowledged or poorly accounted for. This study compared two methods of taking FA measurements - (1) taking measurements by hand using digital calipers; and (2) taking measurements from digital photographs using image analysis software – to examine the impact of method on ME. It was hypothesised that image analysis software would reduce the margins for error, and that the data collected using this method would exhibit much lower levels of ME. 

There were no significant differences in FA between sites for all traits for all species. When analysing the data from measurements taken by hand, we found that asymmetry exceeded ME significantly in two or more traits for all species. Taking measurements using image analysis software greatly increased the chances of discovering asymmetry that exceeded ME. Almost all ME indices were decreased using image analysis software, and the results of FA analyses changed, sometimes considerably, when using image analysis software to take and record measurements. However, there were still no significant differences in FA between Gladstone Harbour and the reference sites. While the two different methods of measurement yielded datasets that were strongly correlated, correlation coefficients varied among traits within each species group. 

Although there are no baseline levels of FA to compare the results of this study to, it could be inferred that the FA occurring in the fish in Central Queensland estuaries, both Moderately Disturbed and HEV, is the result of natural variation, or that the differences between the condition of the study sites are not sufficient to impact fish development. While FA is an interesting potential epigenetic measure of stress, it does not appear to respond to low-level differences in estuary condition and may be better suited to monitoring the effects of specific toxicants or of highly polluted waters. This dataset contains replicate photographs of eyes, pectoral fins, pelvic fins, and otoliths of the abovementioned species of fish, as well as Microsoft Excel raw data files, cleaned data files, and manipulated SPSS and PRIMER-e data files.