Total incombustible (mineral) content of Cherax quadricarinatus differs between feral populations in Central-Eastern Australia

Cherax quadricarinatus has been widely translocated within Australia, and a number of self-sustaining feral populations have established, and persisted, in central-eastern Australia for over 20 years: however, the biology and ecology of feral populations remain poorly understood. Using the loss-by-ignition method, this study investigated differences in the total content of incombustible material (as a proxy for total mineral content), between feral C. quadricarinatus populations in southeast Queensland and northeastern New South Wales. A total of 102 C. quadricarinatus were ignited, and percent total incombustible material was not proportional to the body size, or gender of the crayfish. Incombustible content was however, significantly different between some locations of capture (i.e., waterbodies). The site where incombustible content in crayfish was atypical, Lake Ainsworth, is a naturally acidic coastal lake, and we suggest that acidity and low concentration of calcium in that waterbody are likely responsible for the difference in mineral content detected in that population. Mechanism(s) driving the difference detected in the Lake Ainsworth population are unknown, but we suggest the acidic environment could directly impact maintenance of internal calcium reserves in the crayfish (intermoult), during recalcification of the cuticle (postmoult), or both. Limited calcium availability in the lake may also be a direct, or indirect, contributing factor. The ability of C. quadricarinatus to occupy acidic habitats while managing biomineralization challenges possibly could enable additional range-expansion of the species, and potential impacts on both endangered ecological communities and other biota occupying the acidic coastal habitats of Eastern Australia.


INTRODUCTION
Cherax quadricarinatus (Von Martens) is one of 27 Australian species in the genus Cherax (additional species of Cherax are endemic to New Guinea) Lukhaup, Eprilurahman & Von-Rintelen, 2017). The native range of C. quadricarinatus covers the and aquaculture husbandry and production, the wild biology and ecology of the species remains poorly understood .
The species' ability to tolerate the at-times challenging environmental conditions within its native range evidently provides this species with the capacity, and considerable opportunity, to occupy the extensive and rather less environmentally challenging waterbodies elsewhere in Australia. Sufficient evidence is available to conclude that populations of this species are able to survive, and in some cases prosper, in atypical habitats and environmental conditions well outside its native range in Australia and elsewhere. Populations occupy municipal water reservoirs in areas of temperate climate, the subtropics, and tropics in both Hemispheres (e.g., Ahyong & Yeo, 2007), acidic coastal lakes in central-eastern Australia (Leland, Coughran & Furse, 2012), and waterbodies in the semiarid zone of Australia (J.M. Furse, 2014, unpublished data). However, it is not known if occupying these atypical habitats has any biological or physical impact(s) on C. quadricarinatus.
Specimens captured during a previous (2014) and unrelated distributional study of feral C. quadricarinatus populations in southeast Qld and northern NSW (J.M. Furse, et al., 2014, unpublished data) presented an opportunity to investigate any differences between populations in distinctly atypical habitats.
The objective of this study was to determine if the total content of incombustible material in C. quadricarinatus differed between feral populations occupying different waterbodies within the study area.

Collection and storage of crayfish
Crayfish were collected during a study over the Austral Summer-Autumn of 2014 (January-April). A total of 30 locations (i.e., waterbodies) were surveyed in southeast Qld and northeastern NSW, and crayfish were collected from five different waterbodies. The 30 waterbodies were selected for survey on the basis that self-sustaining feral populations of the species were known to have established there, or the species may have established there via translocation, migration through waterways, or overland movement. Crayfish were collected using a standardized trapping protocol (employing baited box traps, also called "bait" or "shrimp" traps, dimensions 250 Â 250 Â 370 mm, 75 mm entrance apertures, ∼2 mm mesh) that was applied at all locations. Trapping was primarily conducted in the morning (0630-1000) and afternoon (1300-1800), with a few late-evening and overnight sets deployed. Single measurements of water temperature, and pH were recorded at time-of-capture, at all locations (HI 98129, pH/Conductivity/TDS Tester; Hanna Instruments Ò , Woonsocket, RI, USA). Crayfish were collected under NSW Department of Primary Industries Fisheries Section 37 Permit P12/0026-2.0 and Qld Fisheries General Fisheries Permit 169932.
Long-term data were obtained from water-supply authorities, or literature, for the following parameters: water temperature, pH, total hardness (as CaCO 3 , NHMRC-NRMMC, 2011), total alkalinity, and calcium and magnesium concentrations. These data covered capture dates of all crayfish in this study (except Lake Ainsworth), and in some cases covered the likely lifespan (i.e., 4-5 years) of those crayfish.
Following capture, crayfish were measured (Occipital-Carapace length) (OCL, Morgan, 1997), and gender (male, female, or indeterminate) was assessed by examination of external reproductive structures. Crayfish were placed in plastic Ziplock Ò bags, stored under refrigeration in the field, before being transported to the laboratory for euthanasia by freezing and subsequent frozen storage.

Laboratory procedure
The percent total incombustible content of crayfish was determined using the Ash Content method of Allen et al. (1974).
Crayfish were thawed at room temperature (∼21 C), and whole crayfish were dried to constant weight at 100 C, preweighed to the nearest 0.0001 g (AND HR-300 analytical balance) and subsequently ground to a fine powder in a domestic coffee grinder (Sunbeam MultiGrinder II). Wherever possible the entire powdered crayfish were ignited (in preweighed crucibles), or in the case of the larger specimens: subsamples (average weight of subsample ∼6.5 g). All powdered crayfish or subsamples were ignited for 1 h at 500 C in an electric muffle furnace, cooled in a desiccator for 30 min before being weighed again. Following final weighing, all ignited samples were examined for any evidence of incomplete combustion (as per Allen et al., 1974). The ash residue of the ignited samples included all incombustible residues, that is, minerals, including metals, and other elements present in the crayfish, plus any uncombusted organic material bound within inorganic matrices, and hereafter is collectively considered as a proxy measure of the mineral content of the samples. The percent dry weight total incombustible material (hereafter incombustible content) of each crayfish was determined by calculation.

Data analysis
Raw data were plotted (scatter and boxplots) and examined for obvious trends. Effects of gender and location were assessed simultaneously by factorial ANOVA, and final analysis used a main effects ANOVA with gender, OCL and location as factors. A post hoc Tukey's LSD provided p-values for differences between locations. All analysis were performed using STATISTICA for Windows (Version 7.1) (StatSoft, Tulsa, OK, USA), critical value for a was 0.05 throughout. Long-term water temperature and water chemistry data were tabulated to show any differences between locations.

RESULTS
Crayfish were collected at all times of the day, from waterbodies within the subtropical climate zone and a temperate-classified area of central-eastern Australia (climate zones based on a modified Koppen system, Bureau of Meteorology, Commonwealth of Australia (BOM), 2017b), with four of the collection locations being large municipal water reservoirs on basaltic strata, and one (Lake Ainsworth) a perched coastal-dune lake underlain by siliceous sand deposits (Timms, 1982). Well-established biotic communities were evident at all locations, and included: algae, macrophytes, invertebrates, and vertebrates (fishes, reptiles, birds): typically, crustacean zooplankter's were highly abundant.
A total of 102 crayfish were collected from the five locations (Table 1), processed and ignited: 58 females, 43 males and one case where gender could not be assigned. The case where gender could not be assigned was excluded from statistical analysis including gender as a factor. All crayfish were captured in fully-calcified (i.e., intermoult) condition, both clean and very dirty exoskeletons were noted, indicating recent moult activity, or otherwise. Water temperature and pH at time of capture differed between locations of capture (Table 1). On average, long-term water temperatures were similar between locations, but some differences in water chemistries were evident between locations (principally pH, total hardness, calcium, and magnesium (Table 2). Water chemistries at Lake Ainsworth and Emigrant Creek Dam were comparable, but Emigrant Creek Dam had, on average, the lowest values in this study for hardness (16.70 mg L -1 ) and magnesium (2.30 mg L -1 ). On average, Lake Ainsworth had the lowest values for pH (5.91) and Notes: Data from Emigrant Creek Dam and Lake Somerset encompass the likely lifespan of the crayfish in this study. No data available for Lake Dyer. Unless otherwise stated, data provided by water-supply authorities. Lake Ainsworth A , data from Timms (1982). Lake Ainsworth B , data from Kadluczka, Howells & Van Senden (1996). calcium (2.21 mg L -1 ), and hardness (22.76 mg L -1 ) was substantially lower than both Lakes Moogerah (86.60 mg L -1 ) and Somerset (50.31 mg L -1 ). A histogram of the raw incombustible content indicated a slight positive skew. A log(x) transformation was applied and rendered the distribution of the data approximately normal. Percent incombustible content of crayfish plotted vs OCL (Fig. 1) showed a weak, nonsignificant positive correlation (R 2 = 0.027, p = 0.10), indicating that correction for OCL was not required for subsequent analysis. There was no significant difference in incombustible content between gender (p = 0.39). The gender Ã location interaction term was also not significant (p = 0.97), but location of capture was significant (p = 8.0 Â 10 -8 ), all by two-way factorial ANOVA.
A boxplot of percent incombustible content by location of capture indicated which locations differed (Fig. 2). The wider 95% confidence intervals (C.I.) for the Lake Moogerah data can be attributed to the natural variation (13% difference) of incombustible content in the smaller (N = 2) sample size at that location, however, these data did not unduly influence, or leverage, any subsequent statistical analysis. A LSD post hoc test indicated Lake Ainsworth as significantly different to Emigrant Creek Dam and Lake Somerset (p = 0.0031 and p = 5.0 Â 10 -7 , respectively). Incombustible content did not differ significantly between any other locations.

DISCUSSION
This study advances our understanding of wild populations of C. quadricarinatus within Australia, and information uncovered in this study is relevant to both the ongoing translocations (and spread) of this species within Australia, but also in other regions elsewhere in the World. It is well established that C. quadricarinatus is a "tough" species of freshwater crayfish (see review in Furse et al., 2015), tolerant of environmental extremes. However, the present study has uncovered a new dimension of the capacity of C. quadricarinatus to cope with challenging environmental conditions. In this case crayfish were collected from established, self-sustaining populations that have persisted well outside this species' native-range for more than 6 years (e.g., Lake Ainsworth) and over 20 years in the case of Lake Somerset. Similarly, all waterbodies supporting these populations were distinctly non-native habitat-types located in non-native climatic zones (e.g., temperate to subtropical climates) (Modified Koppen classification, Bureau of Meteorology, Commonwealth of Australia (BOM), 2017b).
Water chemistry data indicated the waterbodies "grouped" in two categories: locations in NSW with lower pH and hardness, and locations in Qld with neutral pH and considerably higher hardness. Similarities in water chemistries between Lake Figure 2 Boxplot of percent DWT incombustible content of Cherax quadricarinatus vs location of capture. Specimens from Lake Ainsworth had significantly lower total incombustible content compared to Emigrant Creek Dam and Lake Somerset. Refer Table 1  Ainsworth and Emigrant Creek Dam may, in part, be attributable to that region's somewhat uniform basaltic geology, and both locations being situated in the same catchment, ∼7 km apart. While it has been previously reported that C. quadricarinatus can occupy acidic environments (Leland, Coughran & Furse, 2012, Lake Ainsworth: a naturally acidic coastal window lake, Timms, 1982), any long-term physical/physiological impact(s) of this species occupying such habitat are not well documented. This study establishes that there is a demonstrated physical impact on C. quadricarinatus occupying acidic habitat, in this case a significantly lower total incombustible-mineral content compared to crayfish collected from other nonacidic habitats. In this case, we presume the incombustible content was primarily composed of calcareous minerals derived from the crayfish exoskeletons and tissues, and any other residues (e.g., other metals and incombustible nonmetals) were at negligible levels (sensu Jussila, Henttonen & Huner, 1995).
Despite acidic conditions, and in particular having to maintain a calcified exoskeleton (i.e., internal stores of calcium) in adverse conditions, this study supports a conclusion that populations of C. quadricarinatus can persist, and evidently prosper, medium to long-term in acidic habitats, including where calcium is not abundant. An alternative interpretation is that this study has uncovered evidence indicating the C. quadricarinatus at Lake Ainsworth are not actually coping with the pH and low mineral availability, and the population is under-mineralized. This alternative explanation seems less likely given that C. quadricarinatus has prospered and remained so abundant in Lake Ainsworth for nearly a decade.
The exact mechanism driving the lower incombustible content in the Lake Ainsworth population (or conversely the higher incombustible content in other waterbodies) in this study is unclear, but given the acidic condition of Lake Ainsworth, we speculate it is possible that acidity is a factor directly, or indirectly, contributing to the differences reported here.
Biomineralization is a reasonably well-understood phenomenon in the Crustacea and freshwater crayfish, so some speculation that effect(s) of an acidic environment may occur in C. quadricarinatus is appropriate. However, we hypothesize that other aspects of the water chemistry of Lake Ainsworth may also be impacting mineralization in the C. quadricarinatus population.
The mineral that is principally responsible for providing rigidity and strength in the structure of the exoskeleton of freshwater crayfish is calcium carbonate (CaCO 3 ) (Huner, Kowalczuk & Avault, 1978;Reynolds, 2002). Calcium is therefore one of the key elements for growth in freshwater crayfish (Reynolds, 2002): especially so postmoult (Greenaway, 1985;Huner, Kowalczuk & Avault, 1978). Postmoult, part of the mineral requirement for recalcifying the new cuticle is met by retaining calcium from the previous exoskeleton (via storage in the gastroliths located in the foregut, formed premoult) with the remainder derived from the water column, food, or other ingested material such as the exuvium (Adegboye, Hagadorn & Hirsch, 1975;Greenaway, 1985).
Calcareous minerals are a key structural component of crustacean exoskeletons (Bentov et al., 2012), and the acidity of Lake Ainsworth (the most acidic in the study) has the potential to directly (and indirectly) influence calcareous mineral content in the population of C. quadricarinatus. The acidity of the Lake may directly influence, via increased solubility, internal calcareous mineral content premoult (i.e., during decalcification and gastrolith formation), but also postmoult, during both recalcification from internal stores, that is, solubilization reabsorption of gastrolith calcium) and uptake from external sources of calcium. Availability of bicarbonate (HCO 3 -) is reduced in acidic conditions and if insufficient bicarbonate is available from internal sources, this may also directly contribute to reduced internal calcareous mineral content.
It is therefore possible the pH of Lake Ainsworth is, at times, simply outside of the biologically-tolerable range where C. quadricarinatus is able to calcify, or maintain calcification.
Water chemistry data indicates Lake Ainsworth is not a calcium-rich environment, and limited calcium availability may directly influence the mineral content of C. quadricarinatus in the lake. Lower calcium availability may also interact with acidity of the lake, possibly indirectly creating a limiting condition (Reynolds, 2002), or becoming a limiting factor. One possible scenario would be the acidity of Lake Ainsworth (i.e., reduced carbonate availability), interacting with the low calcium levels thus limiting availability of both the components of calcium carbonate. While water chemistry provides potential explanations for the difference in total mineral content, assessed via incombustible content, at Lake Ainsworth, it is possible other unknown (and unmeasured) biotic and/or abiotic factors may be responsible.
It is not possible to determine why incombustible content in the crayfish from Emigrant Creek Dam was significantly higher than Lake Ainsworth despite lower hardness at that location. We speculate that the higher pH and calcium values, and lower magnesium value, at Emigrant Creek Dam may potentially negate any effect of the lower hardness.
Increasing atmospheric CO 2 may increase acidity in the acidic habitats of Eastern Australia, potentially impact rare and endangered native crayfish (Cherax and Tenuibranchiurus spp.) (Beaune et al., 2018), but also feral populations of C. quadricarinatus in those habitats. Ongoing surveys and monitoring of these feral C. quadricarinatus populations may track any changes in the distribution of C. quadricarinatus but may also permit the mechanism(s) driving the differences in incombustible-mineral content reported here to be investigated and documented, in particular any effects of Lake Ainsworth water chemistry on the mineralization and calcareous content of C. quadricarinatus.

Potential management implications
We speculate that the ability of C. quadricarinatus populations to establish and persist, and maintain internal calcium levels, in calcium-poor and acidic conditions, could have important management implications. For instance, the substantial resident population of C. quadricarinatus in the calcium-poor Lake Ainsworth, needs to somehow obtain the calcium (sensu Adegboye, Hagadorn & Hirsch, 1975) necessary for maintaining adequate internal levels. Other biota occupying the lake, especially calcified biota (i.e., other native crayfish), are likely be an accessible and useful source of calcium for C. quadricarinatus occupying the Lake. Leland, Coughran & Furse (2012) reported the smaller and ubiquitous C. cuspidatus appeared to be absent from the lake, but not from the adjacent, acidic, Melaleuca quinquenervia swamps and associated roadside drains. Monitoring of the adjacent habitats by the authors confirms that situation remains unchanged and while C. cuspidatus is present, C. quadricarinatus has not been detected in these nearby acidic habitats. It is unclear if C. quadricarinatus is involved in the apparent absence of C. cuspidatus from the lake, but it is reasonable to suggest the smaller native crayfish (and other calcified biota) have been a relatively available and useful source of calcareous minerals for the larger C. quadricarinatus, via predation and consumption.
Eastern Australia's extensive acidic coastal Melaleuca swamps extend, discontinuously, from the north of Qld to Sydney in the south, and include a number of coastal window lakes similar to Lake Ainsworth. These acidic coastal swamps and lakes are occupied by other rare and/or critically endangered species (including other species of crayfish: e.g., Tenuibranchiurus spp. Coughran, Dawkins & Furse, 2010;Dawkins et al., 2010Dawkins et al., , 2017, and include endangered ecological communities (e.g., freshwater wetlands on coastal floodplains in NSW). It is possible the ability of C. quadricarinatus to occupy and survive in these acidic habitats could potentially lead to direct, or indirect, negative impacts on the various rare and endangered native floral and faunal species in these habitats.
Similarly, while not continuous, the coastal swamps of eastern Australia could potentially act as an "invasion" route that C. quadricarinatus could exploit, possibly providing an otherwise unavailable pathway ∼600 km south, or >1,500 km to the north and west (adjacent the species native range).

Limitations of study
Loss-by-ignition residues may include nontargets (e.g., metals and other incombustible elements) and possible small traces of uncombusted organic material, and this presumably applies to the incombustible material data obtained in this study. Analysis of mineral content, or other constituents of the combustion residues, was not possible in this study, however, it is reasonable to assume residues were principally mineral, and any other incombustible nonminerals were minor components (Bryan, 1967;Jussila, Henttonen & Huner, 1995). Therefore, our standard method, applied to all samples, has resulted in a consistent measure of actual minerals, plus all nontargets, in all samples. Although numbers of crayfish collected from some locations were small (N < 20), most 95% C.I. were tight and the effects of location of capture on incombustible content were strong. As previously outlined, the broader 95% C.I. at Lakes Moogerah and Dyer are attributable to natural variation in the incombustible content in the crayfish captured there (13% and 1.6%, respectively): results need to be interpreted accordingly until the mineral content in these populations has been investigated and clarified.
Recent water chemistry data, for the parameters of interest in this study, do not exist for Lake Ainsworth; the historical water chemistry data available to us did however, permit satisfactory characterization of the lakes water chemistry, in particular: establishing low CaCO 3 and calcium concentrations, and persistent acidic conditions. The 1987 Lake Ainsworth maximum, average and median values for CaCO 3 and calcium indicated high-value induced skew, but as values for CaCO 3 and calcium in the 1996 data were unremarkable, we drew on the more recent 1996 data.

CONCLUSIONS
The incombustible content of feral C. quadricarinatus populations differed significantly between the locations in this study, and the low total mineral content (using incombustible material as a proxy for mineral content) in crayfish from Lake Ainsworth is likely attributable to the lake's pH and limited calcium availability. This study supports other findings that C. quadricarinatus is tolerant of environments that are acidic, but also calciumpoor, and additionally demonstrates that despite these conditions the species can evidently prosper, albeit with some physical impact. This opens the possibility of the species having the potential to occupy many or all of the extensive coastal swamps of central-eastern Australia, with possible implications for the resident endemic flora and fauna.

ADDITIONAL INFORMATION AND DECLARATIONS Funding
The study was supported by The Griffith School of Environment (Gold Coast campus), the Griffith Centre for Coastal Management and the Environmental Futures Research Institute (Griffith University). There was no additional external funding received for this study. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.