- Top of page
- Electronic References
The effect of water temperature on growth responses of three common seagrass fish species that co-occur as juveniles in the estuaries in Sydney (34° S) but have differing latitudinal ranges was measured: Pelates sexlineatus (subtropical to warm temperate: 27–35° S), Centropogon australis (primarily subtropical to warm temperate: 24–37° S) and Acanthaluteres spilomelanurus (warm to cool temperate: below 32° S). Replicate individuals of each species were acclimated over a 7 day period in one of three temperature treatments (control: 22° C, low: 18° C and high: 26° C) and their somatic growth was assessed within treatments over 10 days. Growth of all three species was affected by water temperature, with the highest growth of both northern species (P. sexlineatus and C. australis) at 22 and 26° C, whereas growth of the southern ranging species (A. spilomelanurus) was reduced at temperatures higher than 18° C, suggesting that predicted increase in estuarine water temperatures through climate change may change relative performance of seagrass fish assemblages.
Virtually all aspects of ectotherm behaviour and physiology are strongly influenced by exogenous factors, with individual survival occurring within a specific range of environmental variation (Roessig et al., 2004). Water temperature is one such exogenous factor, affecting the way fishes behaviourally thermoregulate their feeding, growth and reproductive rates. Increases in global water temperatures can be expected to have major effects on the distribution of coastal fishes, especially via water temperature effects on demography and range (Perry et al., 2005; Munday et al., 2008). The consequences of fish distributional ranges, however, are likely to be complex (Munday et al., 2008; Feary et al., in press), for example, some species of coral reef fishes with ranges encapsulating the temperate edge of the Great Barrier Reef (GBR) appear to have less scope to adjust to shifts in water temperature of up to 2° C in the next 50 years (Munday et al., 2008).
Understanding how increasing coastal water temperatures will alter temperate reef fish community structure is completely lacking in temperate marine science (Booth et al., 2011). Temperate south-eastern (SE) Australian waters have warmed 0·023° C year−1, c. four times the global ocean warming average (Ridgway, 2007), while regional declines in kelp forest cover (i.e. up to 95% regional losses of the giant kelp Macrocystis pyrifera; Johnson et al., 2011) as well as declines and southward range shifts of habitat-forming kelp species (Wernberg et al., 2011) have been occurring. As SE Australian fish communities form a vital part of both Australian and global marine biodiversity (c. 80% of the fish species are Australian endemics; Last et al., 2010) and Australia's fisheries [yielding c. 33% by value of the total Australian fisheries production, ABARE (2010)], the impact of warming coastal waters on SE Australian temperate fishes will have significant ecological and economic importance.
Despite the high importance of rocky reef habitats for fish community structure in SE Australia (Curley et al., 2002), this region is dominated by shallow estuarine seagrass environments (Edgar & Shaw, 1995). These environments form productive and important habitats for fisheries productivity globally (Orth et al., 2006). Such estuarine habitats are already subject to extreme fluctuations in physical factors, such as salinity, turbidity and water temperature, compared with adjacent coastal waters. Despite this, fish assemblages within estuarine habitats can be exceptionally diverse and abundant, comprising species within a wide range of feeding guilds and life stages (Guidetti, 2000; Nagelkerken et al., 2000). These fish assemblages can also include fish species that encompass a wide range of biogeographic origins, with seagrass beds at given latitudes conceivably comprising fish species at the northern, southern and central parts of their ranges (Pollard, 1984). Although high levels of interspecific divergence in response to temperature rise may be expected within these communities, there is little empirical evidence to support this.
Three seagrass-associated fish species with different biogeographic ranges were sampled between February and March 2010 in two estuaries in the Sydney area [Pittwater (33° 37′ 02·80″ S; 151° 18′ 25·24″ E) and Botany Bay (33° 59′ 17·60″ S; 151° 10′ 51·90″ E)]: the eastern striped trumpeter Pelates sexlineatus (Quoy & Gaimard 1824) (Terapontidae) and the fortescue Centropogon australis (White 1790) (Scorpaenidae), which range from subtropical to temperate regions, and the bridled leatherjacket Acanthaluteres spilomelanurus (Quoy & Gaimard 1824) (Monacanthidae), which is found from warm to cool temperate habitats (Froese & Pauly, 2012). Within each estuary, a 10 m long seine (mesh size 16 mm stretch) was drawn through nearshore seagrass beds, with all collected fishes immediately transferred into a large aerated seawater container for transport to the University of Technology, Sydney, aquarium facilities.
All individuals were housed in 100 l glass aquaria under natural light regimes (33·5° S) at ambient estuary temperature [mean = 22° C, summer range 20–24° C, D.J. Booth, unpubl. data; HOBO© logger (www.onsetcomp.com) data] and fed to satiation twice daily on fish pellets (Tetramin; www.tetra.net) for 2 days. Each species was then divided into groups (20–30 individuals) within each temperature treatment (18, 22 and 26° C) and acclimated for 7 days (i.e. maximum temperature change of 0·5° C day-1). After this period, all individuals were measured [total (LT) and standard (LS) lengths (mm)] and weighed (g wet mass). All individuals were within juvenile size classes and were of similar sizes (mean ± s.e. mass): P. sexlineatus: 1·65 ± 0·22 g; C. australis 1·88 ± 0·10 g and A. spilomelanurus 1·44 ± 0·09 g. All fishes were allocated to separate 10 l individual aquaria maintained at each of the three treatment temperatures. Each aquarium contained a small PVC pipe as shelter. Water was changed (50%) daily and dissolved ammonia was monitored. All individuals were fed twice daily to satiation, and food was removed by siphoning during daily water changes. After 10 days all individuals were removed, then measured and weighed (as before).
Somatic growth of each individual was estimated as the instantaneous growth rate (GINST) based on the mass of fish at the beginning (M1) and the end (M2) of the interval of length Δt (10 days): GINST = [ln (M2 M1−1)] Δt−1.
As oceanic warming shifts fish distributions towards the poles (Walther et al., 2002; Parmesan & Yohe, 2003) it will become increasingly important to understand how species ranges will change and how these changes will affect their physiological ability to grow and survive. In this study, the growth response of species with differing latitudinal distributions was examined, and significant species-specific differences in growth response to differing temperature treatments were found (Table 1).
Table 1. ANOVA results of effects of water temperature and fish species on fish-specific growth rate
|Species × temperature||4||3·254||<0·05*|
|Error||96|| || |
For P. sexlineatus, which was collected towards the most poleward latitude of its distributional range, GINST was lower at the lowest temperature [Fig. 1(a), Tukey's HSD GINST 18° C < 22° C = 26° C], indicating that this species may have the potential to survive and grow at higher water temperatures than those at which it was collected (Gardiner et al., 2010). Centropogon australis had its highest growth at the highest temperature (26° C) [Fig. 1(b); Tukey's HSD GINST 18° C = 22° C < 26° C]. Such patterns in growth rate may show that these species have the scope to acclimate to increasing water temperatures (Gardiner et al., 2010; Donelson & Munday, 2012). Lastly, as may be expected, given Sydney is at the northern end of its range, the GINST of A. spilomelanurus was highest at the lowest temperature [Fig. 1(c), Tukey's HSD GINST 18° C > 22° C = 26° C].
Figure 1. Mean ± s.e. (n = 10–12) instantaneous growth rate (GINST) and temperature treatment for (a) Pelates sexlineatus, (b) Centropogon australis and (c) Acanthaluteres spilomelanurus.
Download figure to PowerPoint
There is increasing evidence to suggest that species' response to warming waters may depend in part on how populations of a species throughout its geographic range are impacted by warming waters; local adaptation to thermal gradients may result in species with latitudinally separated populations responding differently to warming waters (Gardiner et al., 2010). This work has shown, however, that species-specific differences at the same latitude, associated with the biogeographic spread of a fish population, may partly determine a species' ability to acclimate to warming coastal waters (Feary et al., in press). For species with populations that extend into subtropical zones (e.g. P. sexlineatus), warming coastal waters may improve their growth performance (Munday et al., 2008; Lek et al., 2012). For species with populations predominantly within cooler temperate regions, however (i.e. A. spilomelanurus), decreased growth may occur with warming waters. Under predictions that coastal water temperatures will rise between 2 and 4° C by 2080 in the SE Australian region (Lough et al., 2012), this work suggests that there may be substantial changes in the growth performance, and therefore possibly survival, of common estuarine fishes within this region. Given the short-term nature of this study at a single latitude, it is important to note that longer acclimations may influence the shape of the thermal reaction norm, with developmental or transgenerational acclimation possible, and that relationships may differ over latitudes (Donelson & Munday, 2012).
Rapid and substantial changes in the geographic distribution of marine fishes have occurred in the last decades within Australia (Last et al., 2010; Feary et al., in press) and globally (Dulvy et al., 2008; Nye et al., 2009), with such range shifts expected to increase in strength and intensity as global climatic conditions change (Booth et al., 2011; Madin et al., 2012). As metabolism, growth, reproduction and ultimately survival of all organisms are closely linked to temperature (Hurst, 2007), understanding the species-specific response to changing water temperatures will allow a greater understanding of the attributes that underpin successful performance in various environments.
We wish to thank reviewers for comments, and the University of Technology, Sydney, for supply of aquarium facilities. D.A.F. was supported by a Chancellors Postdoctoral Fellowship from the University of Technology, Sydney. This study was approved under UTS Ethics protocol 2011-129A. This article is contribution # 114 from the Sydney Institute of Marine Science.