Introduction

Temporary waterbodies are extremely common and contain a diverse (Williams, 2006) and often unique fauna (Calhoun et al., 2017). Ostracod microcrustaceans are a major component of this fauna (Williams, 2006). Most ostracod species produce desiccation-resistant eggs, which are used to survive periods of drought and other adverse conditions (Horne, 1993; Strachan et al., 2015). The eggs accumulate in the sediment, forming an egg bank from which active populations are re-established when more favourable conditions return (Brock et al., 2003; Strachan et al., 2015). The desiccation-resistant eggs can be transported via wind, water, or animal vectors (Sandberg & Plusquellec, 1974; De Deckker, 1977; Brendonck & Riddoch, 1999; Finston, 2002; Green et al., 2023; Rosa et al., 2023) and are likely also the main dispersive phase of the ostracod life cycle (Bilton et al., 2001; Figuerola et al., 2005). These eggs are therefore crucial to the persistence of ostracod populations in temporary waterbodies (Horne & Martens, 1998), as well as the persistence of metapopulations across landscapes (Rahman et al., 2023). The durability of the desiccation-resistant eggs of ostracods is not well studied (Fryer, 1996; Rossi et al., 2012; Vandekerkhove et al., 2013) but those of at least some species can remain viable for years and possibly decades (Fryer, 1996; Schön et al., 2012).

The desiccation-resistant eggs or cysts of aquatic invertebrates can be an important source of material for biological studies (Brendonck & De Meester, 2003; Rosa et al., 2021a). In the laboratory, the eggs or cysts can be rehydrated, and individuals hatched and raised for a range of purposes (Vandekerkhove et al., 2005a; Bisquert‐Ribes et al., 2022). For example, the renowned crustacean researcher G. O. Sars described several new species of ostracod, copepod, and cladoceran after hatching and rearing individuals in his laboratory in Norway from dried sediments collected in Australia, New Zealand, and South Africa (Sars, 1885, 1894, 1895, 1896a, 1896b, 1896c, 1897, 1898). Other studies have utilized dried sediments to investigate hatching cues and patterns of emergence (Bailey et al., 2004; Vandekerkhove et al., 2005b; Campagna, 2007; Haghparast et al., 2012), determine the species present in temporary waterbodies (Williams, 1991; Vandekerkhove et al., 2005a), compare community estimates based on egg sampling versus collecting active individuals (Rosa et al., 2021a; Bisquert‐Ribes et al., 2022), and evaluate population histories (Lenormand et al., 2018). Collecting and rehydrating desiccation-resistant eggs or cysts has the advantage of removing the need to collect samples when water is present in the habitat (Martens et al., 1992; Havel et al., 2000; Rosa et al., 2021a). This approach has been widely used for studying invertebrates in freshwater environments (e.g. rotifers, ostracods, cladocerans, branchiopods; Brendonck & De Meester, 2003; Rosa et al., 2021a) but has only been partially explored in salt lakes (Campagna, 2007; Moscatello & Belmonte, 2009).

More than 80% of lakes and wetlands in Australia are saline (Timms, 2005). Most of these have salinities > 3 g/L (Geddes et al., 1981; Timms, 2005) and are shallow and temporary. They predominantly occur in hydrologically closed ancient palaeodrainage basins with hydroperiods dominated by local precipitation (De Deckker, 1983b). Globally, salt lakes are among the most threatened ecosystems (Finlayson, 2016; Saccò et al., 2021). In Australia, salt lakes are under threat from a range of activities, including global climate change and mining (Williams, 2002; Timms, 2005; Jellison et al., 2008). Climate change-induced alterations, such as shorter and less predictable hydroperiods, increased salinities and other changes in water chemistry, and higher UV radiation, can profoundly impact community composition and the distributions of species, potentially leading to local extinctions (Williams, 2002; Timms, 2005; Jellison et al., 2008; Atkinson et al., 2021). Mining activities, including drilling, excavation, and discharge of acidic or saline wastewater, can disturb lake beds and hydrological regimes and water quality, threatening the natural processes of the lake ecosystem (Timms, 2005; Datson, 2009; Mernagh et al., 2016; Marsh et al., 2021). Mernagh et al. (2016) identified salt lakes from six regions in Australia as most likely to contain economically significant deposits of minerals, such as potash. These regions are mainly in remote arid areas (Mernagh et al., 2016) where the lakes may go for years without holding water (Timms, 2006; Timms et al., 2006). Assessing and managing these and other threats to Australian salt-lake ecosystems should be informed by evidence, including a sound knowledge of the biodiversity in these ecosystems and of the tolerances of resident organisms to physical and chemical variables across their entire life cycle (Lawrie et al., 2021).

The ostracod subfamily Mytilocypridinae (Cyprididae) is endemic to Australia and comprises approximately 29 species in a range of genera. (The exact number of genera is uncertain; Rahman, 2024). These species are an important component of the fauna of wetlands, especially salt lakes (Geddes et al., 1981; De Deckker, 1983b; Lawrie et al., 2021; Rahman et al., 2023). Sampling for mytilocypridines (and most other salt-lake invertebrates) has largely been conducted on an opportunistic basis, is biased towards accessible areas (Cale et al., 2004; Quinlan et al., 2016), and has focused on active individuals, mainly adults and late juvenile instars (hereafter called active sampling; Lawrie et al., 2021). Salinity and other physical and chemical factors are likely to be important determinants of the distribution and abundance of mytilocypridine species (Radke et al., 2003; Rahman et al., 2023). Given that adults and late juvenile instars may be more tolerant of physical and chemical conditions than other stages of the life cycle (Geddes, 1976), it is possible that the available information currently overestimates species’ tolerances. Laboratory hatching experiments based on desiccation-resistant eggs collected from salt lakes could help to address some of the above-mentioned knowledge gaps. Several mytilocypridine species have been hatched in the laboratory from eggs in dry sediment samples (e.g. see De Deckker & Geddes, 1980; Campbell, 1995; Timms, 1998; Campagna, 2007; Strachan et al., 2014, 2016; Matzke-Karasz et al., 2016; Rahman et al., 2023). However, we are not aware of any attempts to investigate the viability of mytilocypridine eggs through time. Additionally, attempts to assess the environmental tolerances of the early life cycle stages of mytilocypridines are limited to a study by Martens (1985) that used eggs collected from submerged leaves of Ruppia to assess effects of temperature and salinity on post-embryonic growth in Mytilocypris henricae (Chapman, 1966).

The desiccation-resistant eggs of some cyprinid species [e.g. Bradleystrandesia reticulata (Zaddach, 1844)] may require a period of dormancy before they hatch (Horne & Martens, 1998), although those of other species do not (McLay, 1978). It is currently unknown whether the desiccation-resistant eggs of mytilocypridines require a period of dormancy, but the fact that some species (e.g. Australocypris robusta De Deckker, 1974) inhabit permanent volcanic lakes in western Victoria implies that this might not necessarily be the case (De Deckker, 1983a). The production of desiccation-resistant eggs may be environmentally triggered (Horne & Martens, 1998; Otero et al., 1998), with hatching also influenced by environmental factors such as water presence, oxygen levels, temperature, light exposure, and osmotic pressure (Rossi et al., 2012; Paes et al., 2016). Salinity has been recognized as a critical determinant for hatching (Jiménez-Melero et al., 2023), especially in species inhabiting saline ecosystems (De Deckker, 1983a).

Studies have shown that the desiccation-resistant eggs of a range of invertebrates are unevenly distributed in temporary freshwater bodies (Martens et al., 1992; Thiéry, 1997; Bright & Bergey, 2015; Rosa et al., 2020). The eggs often accumulate in deeper areas, usually in the centre of the water body, where water persists for longer (Martens et al., 1992; Bright & Bergey, 2015), although eggs that are a part of floating debris may concentrate at the margins of the waterbody (Martens et al., 1992). Few equivalent studies have been conducted on salt lake ecosystems, although Campagna (2007) showed that the eggs of Parartemia Sayce (1903) brine shrimps in two large salt lakes (Lake Yindarlgooda and Lake Miranda) in Western Australia were patchily distributed. Understanding the spatial distribution of the desiccation-resistant eggs of an invertebrate within a waterbody is a prerequisite for efficient sampling of these eggs, especially in salt lakes which can be large and heterogeneous (De Deckker, 1983b).

This study used mytilocypridine eggs collected from sediments in salt lakes in Western Australia as source of samples to analyse aspects of the ecology of this ostracod group. The study has three main aims. The first was to assess the distribution of mytilocypridine eggs within a dry salt lake. The second was to use a laboratory trial (rehydration trial 1) to investigate the viability of desiccated eggs from selected mytilocypridine species after more than 2 years in storage. The third was to use a laboratory trial (rehydration trial 2) to assess the range of salinities over which selected species could hatch from eggs and survive to adulthood. The study also compared the ostracod species obtained from rehydration trials to those obtained from active sampling of the same water body, as well as the field and laboratory salinity ranges for these species.

Methods

Egg counts: spatial distribution of mytilocypridine eggs

This aspect of the study focused on identifying the spatial distribution of Mytilocypridinae eggs within a dry salt lake.

Study site

Sediment samples were collected in May 2020 from a small (0.13 ha), shallow, alkaline, and ephemeral salt lake (WH1), located near Wongan Hills in the Western Australian Wheatbelt (Table 1, Fig. 1). We selected this lake because of its small size, and it was known to contain three mytilocypridine species (Mytilocypris ambiguosa De Deckker, 1978, Australocypris n. sp. 1, and Caboncypris kondininensis Halse & McRae, 2004) (Rahman et al., 2023). The lake dries out in summer and fills seasonally, which makes it easier to track its drying and filling cycles; it had been dry for several months when the sediment was collected. Sediment was collected from 18 different sampling points in northern, centre, and southern zones of the lake (Fig. 2). At each sampling point, we used a hand scoop to collect a surface scraping of sediment, measuring approximately 30 cm in width, 30 cm in length, and 3 cm in depth.

Table 1 Details of sediment collection sites for rehydration trials 1 and 2, including site location and area, mytilocypridine species collected at each site via active sampling, dates of active sampling, and salinity and pH of water body at the time of active sampling
Full size table
Fig. 1
figure 1

Map displaying the locations of sites from which sediment was collected for rehydration trial 1 (circle symbols) and trial 2 (triangle symbols). The map was generated using the QGIS software (https://www.qgis.org), and the base map for the IBRA (Interim Biogeographic Regionalisation for Australia) region boundaries (Avon Wheatbelt, Mallee) was collected from the Australian Government Department of Climate Change, Energy, the Environment and Water website (https://fed.dcceew.gov.au)

Full size image
Fig. 2
figure 2

Details of sediment sampling used to investigate the distribution of mytilocypridine eggs within a small salt lake (WH1). Sediment was collected from a total of 18 sampling points, with 6 points located in each of the north, centre, and south zones. Each sampling point covered approximately 30 cm in width × 30 cm in length × 3 cm in depth

Full size image

Sediment processing and egg counting

After collection, the sediment samples were placed in labelled paper bags and transported to the laboratory for further processing. To ensure that the samples were completely dry and free of moisture, which is reported as necessary for stimulating the hatching of desiccation-resistant crustacean eggs (Campagna, 2007), they were oven-dried at 30 °C. Once dried, the sediment from a single sampling point was mixed thoroughly to ensure an even distribution of eggs and then sieved using a 1 mm sieve to remove debris.

Mytilocypridine eggs were identified using information available in the literature (De Deckker, 1974, 1976, 1978; Martens, 1983; Matzke-Karasz et al., 2016). These eggs were spherical and approximately 100 µm in diameter, which is generally larger than the eggs of small ostracods. They exhibited a deep depression on one face and had an orange to red coloration (Fig. 3; De Deckker, 1974, 1978; Martens, 1983). However, we were unable to distinguish between the eggs of different mytilocypridines and so could not assess species-specific egg distributions.

The number of mytilocypridine eggs in 6 g (6 × 1 g portions) of sediment was estimated from each of the 18 sampling points. The egg numbers were counted under a stereomicroscope at 40× magnification.

Fig. 3
figure 3

Desiccation-resistant eggs of Mytilocypridinae in sediment samples from the LQ site. Scale bar = 100 µm

Full size image

Rehydration trial 1: viability of desiccation-resistant eggs

This trial used the above collected sediment samples from WH1 to investigate the viability of the desiccation-resistant eggs of the mytilocypridine species.

The sediment samples were rehydrated in October 2020 (after 6 months of the storage) and kept inundated for three months. This duration was necessary to raise ostracods to adulthood (Ikeya & Kato, 2000; Diaz & Lopretto, 2017; Rosa et al., 2021b) to enable morphological identification based on fully developed male genitalia (Halse & McRae, 2004).

For each sampling point, 20 g of sediment was placed in each of two transparent plastic containers. The same amount of sediment was added to each culture, but the number of mytilocypridine eggs in that sediment was not counted. Subsequently, 400 ml of freshwater (0 g/L reverse osmosis water, RO) was added to one container (freshwater treatment) and 400 ml of saline water (11.5 g/L) was added to other (saline treatment). The two different salinity treatments were used to account for potential variation in the hatching success of different species at different salinities. The saline water was created by mixing RO water with salt collected from saline lakes in Western Australia (provided by the WA Salt Group). Throughout the article, conductivity measurements (mS/cm) were converted to salinity (g/L) using the equation developed by Williams (1986).

The trial was conducted in a temperature-controlled room at 20 ± 1 °C, with a 12-h light/dark regime (Paes et al., 2016). Hatching and water quality parameters such as pH, dissolve oxygen, salinity, and temperature were monitored weekly. Dissolved oxygen, salinity, and pH levels were measured using YSI DO200A, YSI EC300A, and YSI PH100A meters, respectively. The ostracods were fed with Tetra Fin flakes (Sevilla et al., 2013). Once the ostracods reached adulthood, they were collected from the cultures using a 5-ml plastic transfer pipette under a light bulb. Due to the large size (> 3 mm) of mytilocypridines, they were easily sorted through visual observation. Lids were placed on the containers to limit evaporation, but small holes were drilled into the lids to allow airflow and prevent the cultures from becoming anoxic.

To evaluate the length of time desiccation-resistant eggs might remain viable, rehydration trial 1 was rerun in September 2022, i.e. 27 months after the sediment was collected, using the material and methods described above with sediment samples that had been stored in paper bags at room temperature since collection. Values for salinity, pH, and the numbers of each species recovered from both the 6-month and 27-month storage runs are provided in Supplementary Table S1.

Rehydration trial 2: effect of salinity on hatching and development

In this rehydration trial, mytilocypridines were raised from desiccation-resistant eggs under various salinity conditions to identify the salinity range at which selected species can hatch and survive to adulthood.

Study site and sediment collection

Sediment samples were collected in April 2022 from six temporary, alkaline salt lakes (LP, KON, BR1, BR2, 3S7, LQ) located in Western Australia (Fig. 1). Three of these lakes (LP, BR2, LQ) contained water at the time the sediment samples were collected in which case the sediments were collected from the dry margins of the lake. Samples of surface sediment, measuring approximately 30 cm in width, 30 cm in length, and 3 cm in depth (30 W × 30 L × 3 D cm), were collected using a hand scoop from four to five different sampling points in each lake. The sampling points focused on areas where there were high concentrations of dead ostracod shells to ensure the presence of eggs (Fig. 4).

Fig. 4
figure 4

Illustration of the key steps of sediment rehydration procedure used in Rehydration trial 2: A Sediment collection, B Sediment processing, C Identification and enumeration of egg, D Hatching setup, E Monitoring and feeding, and F Harvesting and identification

Full size image

Sediment processing and egg collection

After collection, the sediment samples were oven-dried and sieved using the same process employed in the first rehydration trial. A specific quantity of mytilocypridine eggs was used for the cultures in this trial (details provided below). To extract the eggs, the sieved sediments were passed through stacked 250 µm and 63 µm Endecott® brass sieves, retaining material < 250 µm and > 63 µm (Fig. 4). The eggs were then separated from the sediment under a stereo microscope using a ‘Drosophila’ sorting brush.

Set up and monitoring of hatching and development

For each lake, mytilocypridine hatching and development was assessed in six salinity treatment levels (0, 20, 40, 60, 80, and 100 g/L), with each salinity level represented by 5 replicates + 1 control. Thus, a total of 36 cultures/transparent plastic containers (1000 mL capacity) were prepared for each site. Approximately 20 g of autoclaved lake sediment and 400 mL of water at the designated salinity was added to each container. The sediment was sterilized at 121 °C using dry autoclaving for 30 min in PROHS horizontal steam sterilizer PJ Lab/Lab + to prevent hatching of non-target invertebrates (which had occurred in the first trial). Within each salinity treatment, 50 intact eggs were added to each of the five (non-control) containers. No eggs were added to the sixth/control container to check whether any mytilocypridines could have hatched from the autoclaved sediment (as opposed to the added eggs).

In this trial, commercially available red sea salt was mixed with RO water to achieve the desired salinity levels. Although this salt is primarily designed to mimic the marine environment, it is suitable for these trials because the ionic ratios in Australian salt lakes are generally comparable to those found in marine water (Bayly & Williams, 1966; Geddes et al., 1981).

The trial for the LP site commenced, i.e. sediment was hydrated, in June 2022 to confirm that the culture conditions were suitable for hatching and rearing mytilocypridines. The trials for the other five sites (KON, BR1, BR2, 3S7, LQ) commenced in August 2022. The cultures were monitored weekly for the hatching of mytilocypridines, and any adults that were observed were collected from the cultures and preserved in 100% ethanol. The trial was stopped after three months.

The numbers of adult mytilocypridines harvested from different treatments and replicates were plotted against the salinity treatments in R statistical software (version 4.2.3; https://www.r-project.org) using RStudio platform (Racine, 2012). The mean salinity measurements for each replicate over the course of the trial are provided in the supplementary material (Table S2).

Active sampling

Samples of active (field collection of live ostracods) mytilocypridines (mainly adults) were collected from the same salt lakes from which the sediment samples were obtained. These samples were collected between 2018 and 2022 (before the sediment sampling) using a small handheld dip net with a mesh size of 250 µm (Table 1). Live ostracods were transported to the laboratory, where they were relaxed in a 10% ethanol solution before being preserved in 100% ethanol. Subsequently, the described species were identified using morphological features based on the taxonomic keys in Halse & McRae (2004), supplemented by information on undescribed species listed by (Rahman, 2024). Data on water conductivity (≡ salinity) (as well as on dissolved oxygen, pH, and water temperature) were also collected using a YSI 556 MPS handheld multiparameter probe. The species collected via active sampling of a waterbody were compared to those that were recovered from sediment samples from that waterbody in the rehydration trials. Field salinities (salinity of water when active individuals were collected) were compared to the salinities in which individual of that species were able to hatch and develop to adulthood in the rehydration trials. The field records for salinity include the above data as well as all published records of field salinity for the species involved as compiled by Rahman et al. (2023).

Statistical analysis

Egg counts: spatial distribution of eggs

Sediment samples collected from 18 sampling points in WH1 were grouped into three zones: North, Centre, and South, with each zone containing six points. A generalized linear model (GLM; Nelder & Wedderburn, 2018) was used to investigate the relationship between egg count and zone. The dataset exhibited overdispersion, as the variance was much higher than the mean (mean = 63.39, variance = 6668.96). Thus, a negative binomial GLM was fitted to the data using the formula: Count ~ Zone. In this model, ‘Count’ represents the egg count, and ‘Zone’ refers to the North, Centre, and South zones. The model fitting was conducted in R (version 4.4.0; https://www.r-project.org) using the ‘glm.nb’ function from the ‘MASS’ package (Venables & Ripley, 2002). Differences in egg count among zones were visualized using boxplots in R.

Rehydration trial 1: viability of eggs

A Generalized Linear Mixed Model (GLMM; Clayton, 1996) was employed to investigate the relationship between the hatching of ostracods, salinity in the experimental treatments, and egg viability after storage, while accounting for potential random effects associated with distinct sediment conditions (Formula: Count ~ Condition * Storage + (1 | Box ID)). The response variable, ‘Count’, denotes the number of ostracods harvested and was the dependent variable in the model. The fixed effects include Condition (Fresh and Saline) and Storage (6 months and 27 months). To address potential correlation or variability in intercepts across different sediment groups, the model incorporates a random effect denoted as (1 | Box ID). This random effect allows for any individual differences between groups not explained by the fixed effects. The GLMM model was separately applied to the datasets of Australocypris n. sp. 1 and Mytilocypris ambiguosa. Due to negligible hatching, Caboncypris kondininensis was excluded from the analyses. A negative binomial distribution was used to model both dataset because the variance was higher than the mean for both A. n. sp. 1 (mean = 2.83, variance = 21.83) and especially M. ambiguosa (mean = 8.28, variance = 216.06), indicating overdispersion. All analyses were performed using the R statistical software (version 4.4.0; https://www.r-project.org), utilizing the ‘glmerMod’ function from the ‘lme4’ package for model fitting (Bates et al., 2015).

Rehydration trial 2: effects of salinity

A generalized linear model (GLM; Nelder & Wedderburn, 2018) was used to investigate the relationship between hatching and development and salinity in Australocypris insularis (Chapman, 1966) at site 3S7 and in Mytilocypris mytiloides (Brady, 1886) at site LP. These sites were selected for analysis because, unlike the other four sites, only a single species was recovered from the sediments at these sites and so all eggs added to a culture can be assumed to belong to that species. The data were analysed separately for each site/species because the two trials were not run at the same time (see above). Both datasets showed evidence of overdispersion as the variance was much higher than the mean (A. insularis mean = 6.87, variance = 41.64; M. mytiloides mean = 11.33, variance = 169.13) and there were substantial number of zero counts especially for M. mytiloides. Thus, a zero-inflated negative binomial GLM was fitted to the data using the formula: Count ~ Category. In this model, ‘Count’ represents the harvested ostracod count, and ‘Category’ encompasses various salinity treatments (0 PPT, 20 PPT, 40 PPT, 60 PPT, 80 PPT, and 100 PPT). The model fitting was performed in R (version 4.4.0; https://www.r-project.org) using the ‘glm.nb’ function from the ‘MASS’ package (Venables & Ripley, 2002).

Results

Egg counts: spatial distribution of eggs

The total egg count from sediment from the North zone of WH1 was noticeably higher than that for the Centre and South zones (Fig. 5). In the GLM analysis, the coefficient for the South zone was − 3.160 with a p value of < 0.001, demonstrating a highly significant decrease in the log count of eggs in this zone compared to the North zone (see Supplementary Table S3). The coefficient for the Centre zone was − 0.858 with a p value of 0.162, indicating that the difference in the log count of the eggs between the Centre and North zones was not statistically significant.

Fig. 5
figure 5

Box-and-whisker plots showing the number of mytilocypridine eggs per gram of sediment collected from North, Centre, and South zones in the WH1 site. The number of eggs per gram was estimated from counting eggs in 6 × 1 g subsamples per sampling point. Locations of sampling points are shown in Fig. 2. The box represents the interquartile range (IQR) with the median depicted by a thick horizontal line, while whiskers indicate data range

Full size image

Rehydration trial 1: viability of eggs

A total of 374 adult mytilocypridines were recovered from rehydrated sediment after 6 months of storage, with 74.87% of individuals from the freshwater treatment and 25.13% from the saline water treatment. Most of the harvested individuals were Mytilocypris ambiguosa (68.4%) followed by Australocypris n. sp. 1 (31%). Only two individuals of Caboncypris kondininensis were observed. The numbers from sediments collected from North zone of WH1 were typically higher than those collected from Centre and South zones, especially in the freshwater treatment (Fig. 6). This broadly coincides with the above-noted data suggesting that egg abundance was higher at North zone. (Note that in this trial the same amount of sediment was added to each culture but the number of mytilocypridine eggs in that sediment was not counted.)

Fig. 6
figure 6

Box-and-whisker plot showing numbers of mytilocypridine harvested, under fresh and saline conditions from sediment collected from North, Centre, and South zones at the WH1 site after 6 and 27 months of sediment storage in rehydration trial 1. The box represents the interquartile range (IQR) with the median depicted by a thick horizontal line, while whiskers indicate the data range, and outliers are represented as circles

Full size image

A total of 429 adult mytilocypridines were obtained from rehydrated sediment after 27 months of storage in paper bags at room temperature compared to only 374 after 6 months of storage. Although the exact values differed (perhaps because the number/composition of eggs in the samples was not identical), the same trends were apparent in both datasets. Thus, for the 27-month dataset: (i) a higher percentage of adults was obtained from freshwater (86.01%) compared to saline conditions (13.99%; Fig. 7) and (ii) representatives of the same three species hatched, with most harvested individuals belonging to Mytilocypris ambiguosa (79.3%) and most of the rest belonging to Australocypris n. sp. 1 (20.5%) (see Fig. 7). Only one individual of Caboncypris kondininensis was harvested in this run.

Fig. 7
figure 7

Total numbers of three mytilocypridine species (square = Australocypris n. sp. 1, diamond = Caboncypris kondininensis, and triangle = Mytilocypris ambiguosa) harvested under fresh and saline conditions from sediment collected from the WH1 site after 6 and 27 months of sediment storage in rehydration trial 1. The circle and bar represent, respectively, the mean and standard deviation for each year-condition combination across different sampling points

Full size image

For Australocypris n. sp. 1, the effects of salinity or storage were not individually significant; however, their interaction was negative (see results of GLMM in Supplementary Table S4). The negative estimate for the interaction term suggests a reduced number of adults harvested in saline conditions after 27 months of storage (Supplementary Table S4). In contrast, for Mytilocypris ambiguosa the effects of salinity, storage, and their interaction were not significant (see results of GLMM in Supplementary Table S5).

Rehydration trial 2: effects of salinity

Mytilocypridines hatched and reached adulthood from the sediments from all six waterbodies included in this trial (Fig. 8), but none were hatched in the control group and no non-target species were apparent in the trial (presumably because any eggs present were destroyed when the sediment was autoclaved). Adults of both Australocypris insularis and Mytilocypris mytiloides were raised from four sites (KON, BR1, BR2, LQ), but only the former species was recovered from the rehydrated sediments from the 3S7 site, and only the latter was recovered for those from the LP site (Fig. 8).

Fig. 8
figure 8

Numbers of Australocypris insularis (red) and Mytilocypris mytiloides (blue) obtained from sediment cultures for five replicates (TR1–TR5), six saline treatments (0, 20, 40, 60, 80, 100 g/L) and six sites (3S7, KON, BR1, BR2, LQ, LP) in rehydration trial 2. No hatching was observed in the controls, which were run for each treatment and site (data not shown)

Full size image

Adults of A. insularis were recovered across a very broad salinity range, for example, from every treatment (0–100 g/L) for one site (KON) and between 20 and 80 g/L for the other four sites that included this species (Fig. 8). In contrast, adult M. mytiloides were recovered from between 0 and 40 g/L for four out of five sites that included this species and from the 0 and 20 g/L treatments for the remaining site (Fig. 8). Related to this, A. insularis adults were obtained from treatments above 40 g/L in all five sites where they hatched, whereas no M. mytiloides were found above 40 g/L for any site (Fig. 8). The median salinity from which A. insularis were obtained from the rehydration trial ranged from 47.06 g/L (KON) to 67.50 g/L (BR2), whereas that for M. mytiloides ranged from 21.98 g/L (BR1) to 28.56 g/L (BR2).

Further analyses were conducted for the two sites (3S7 and LP) for which only a single species was recovered from the sediments. For A. insularis from site 3S7, the GLM results indicate a significant increase in the log count of individuals in the salinity categories of 20 PPT and 60 PPT compared to 0 PPT. However, no significant differences in log count were observed for the other salinity categories (100 PPT, 40 PPT, 80 PPT) when compared to 0 PPT (see Supplementary Table S6). Conversely, for Mytilocypris mytiloides from site LP, the GLM results reveal a significant decrease in the log count of M. mytiloides in 40 PPT salinity compared to 0 PPT. No significant differences in log count were found for the 20 PPT category. Since no M. mytiloides were recovered from the sediments at 60 PPT, 80 PPT, and 100 PPT, reliable log count estimates could not be obtained for these categories (see Supplementary Table S7).

Active vs egg sampling

The rehydration trials (egg sampling) and field data (active sampling) yielded the same species or group of species from three waterbodies (WH1, KON, and 3S7) (Table 2). However, M. mytiloides was recovered from the egg sampling for three sites (BR1, BR2, and LQ) but was not detected in the active samples (Table 2). Conversely, A. insularis was present in active samples from LP but was not observed in the rehydration trial (Table 2).

Table 2 Mytilocypridine species collected through active sampling compared to those that were recovered from eggs in dry sediments from the same site in rehydration trials 1 and 2
Full size table

For both M. mytiloides and A. insularis, individuals were detected over a broader salinity range in the field (active sampling) compared to the rehydration trials (egg samples). The difference was greatest for M. mytiloides, with field salinities ranging from 1.3 to 172.9 g/L compared to 1.3–55.3 g/L in the rehydration trials (Fig. 9). The corresponding values for A. insularis were 2–200.2 g/L for field sites and 2.6–100.9 g/L for the rehydration trials (Fig. 9).

Fig. 9
figure 9

A comparison of salinity range observed for Australocypris insularis and Mytilocypris mytiloides in field records (from Rahman et al., 2023) and in rehydration trial 2. The red brackets indicate the range of salinity tested in the rehydration trial. The graph displays the actual salinities measured in the rehydration trials and not the salinity of the water added to the sediment (see Supplementary Table S2)

Full size image

Discussion

Distribution of eggs in waterbodies

The egg count data indicate that mytilocypridine eggs are unevenly distributed in the sediments of waterbodies, notwithstanding that this finding is based on samples from a single small salt lake (WH1). Thus, sediment collections in mytilocypridine studies should generally target areas within a waterbody that are likely to contain a relatively high concentration of eggs. Although sieving can be used to help recover eggs from relatively large amounts of sediment (Campagna (2007), targeted sediment sampling is essential, especially for large lakes. We do not know why the mytilocypridine eggs were unevenly distributed in WH1. For temporary freshwater habitats, invertebrate eggs tend to accumulate in fissures and cracks, surface depressions, and deeper areas because this is where invertebrates tend to aggregate as the waterbody dries (e.g. see Strachan et al., 2014; Wang et al., 2014; Bright & Bergey, 2015). Other factors such as wind patterns, lake morphometry, flooding effects, and the adaptive strategies of invertebrates may also play a role (Thiéry, 1997; Mura, 2005; Rosa et al., 2020). Some ostracods deposit their eggs on plants or detritus (Rosa et al., 2023), which will also foster an aggregated egg distribution. The situation for small, shallow salt lakes may be similar to what happens in temporary freshwater bodies. In large salt lakes, where salinity can become excessively high and lethal to invertebrates during the drying process (Finlayson, 2016), the remains of mytilocypridines and their eggs may accumulate in shallower areas towards the periphery corresponding to the shore line when salinity was around the maximum that the species can tolerate. In extreme conditions, the mytilocypridines may also close their valves tightly and float (Williams et al., 2015), in which case wind action will play an important role in distributing these individuals within the lake (Zhai et al., 2015).

Viability of desiccation-resistant eggs

The results of the first rehydration trial indicate that the eggs of three mytilocypridines (Australocypris n. sp. 1, M. ambiguosa, C. kondininensis) can hatch after 27 months of storage in paper bags at room temperature. Furthermore, the eggs used in this trial were collected from a salt lake that had been dry for several months prior to the sediment being collected. The desiccation-resistant eggs of some other invertebrate species have been shown to remain viable for extended or even extreme (100 + years) time periods after being collected and held in the laboratory (Brendonck & De Meester, 2003). The longest verified duration for an ostracod that we are aware of is 12 years for an unspecified species of Cypridopsis (see Fryer, 1996), although there are unverified reports of ostracod eggs remaining viable for decades (Schön et al., 2012). There was no evidence of a loss of viability in the eggs of either A. n. sp. 1 (at least for the freshwater treatment) or M. ambiguosa between 6 and 27 months of storage. This finding is significant because, from an ecological perspective, average viability is more important than the maximum viability. Our interpretation assumes that egg abundance, which was controlled by adding the same amount of sediment to each replicate, was similar in both runs. Other studies have similarly quantified hatching of invertebrates using controlled sediment aliquots (e.g. Brazil et al., 2022; Trujillo et al., 2023).

Some studies have shown that the hatchability of the eggs of some invertebrate species may decline with time. For example, although the eggs of two species of Anostraca remained viable for 24 months under room temperature storage, their hatchability reduced gradually over time (Thaimuangphol & Sanoamuang, 2017). However, if stored in − 18 °C under dry-dark-anoxic conditions, hatchability was unaffected. Our storage method, which following Campagna (2007) involved drying the sediments to ensure the eggs are desiccated, appears to be suitable for long-term storage of mytilocypridine eggs.

Although it appears that mytilocypridine eggs can remain viable for at least 27 months in dry storage in a laboratory setting, assessing the survival duration of these eggs in the natural environment is challenging. Evidence suggests that the desiccation-resistant eggs of certain copepods and cladocerans can survive for decades when buried under sediments, although hatchability decreases over time (Hairston Jr et al., 1995). Further research is needed to explore the viability of mytilocypridines eggs in situ and the implications of this for population dynamics and persistent in ephemeral saline lakes, especially as the long-term viability of eggs will be a major factor in determining their resilience to prolonged droughts owing to climate change (Vargas et al., 2019).

Active vs sediment/egg sampling and species surveys

For three out of seven salt lakes, the same mytilocypridine species were obtained via active sampling and in the rehydration trials. This included the WH1 site, which had the most (three) species present, and for which sediment samples were collected from across the entire lake. In three of the remaining lakes, an additional species (always M. mytiloides) was obtained from the egg sampling. This likely reflects the timing of the active sampling. Although M. mytiloides is known to occur in salinities ranging from 1.3 to 172.9 g/L, it has a median salinity record of only ~ 19 g/L (Rahman et al., 2023). The salinity of the field sites at the time active sampling took place was > 45 g/L (Table 1), which was possibly suboptimal for this species, especially for egg hatching and development (see below). In general, desiccation-resistant eggs can accumulate in the sediments of water bodies over long periods of time (Hairston Jr et al., 1995; Schön et al., 2012). This means that they are likely to provide a broad temporal view of the species present in a waterbody (Brendonck & De Meester, 2003; Hairston Jr et al., 1995). Consequently, it is common for studies to recover species from sediment samples that were not detected by active sampling, particularly when that active sampling is limited or opportunistic, as was the case in this study (see Martens et al., 1992; Havel et al., 2000; Moscatello & Belmonte, 2004; Vandekerkhove et al., 2005a; Rosa et al., 2021a). It is also possible that a permanent change in the conditions in a lake could result in a situation where sediment samples will contain viable eggs of a species that are no longer capable of hatching in situ (Brendonck & De Meester, 2003; Santangelo et al., 2014).

An additional species (A. insularis) was captured in the active but not egg/sediment samples from one lake (LP). This is probably because the sediment samples from this lake were collected from dry sediment close to the high-water mark as the lake contained a lot of water at the time when the sediment was collected. Australocypris insularis may not be active in the lake when it is full due to the relatively low salinity of the water at this time. Consequently, active individuals of this species, and thus eggs, are not likely to occur around the high-water mark. In any case, this finding highlights the need to ensure that the sediment collections provide good coverage of a lakebed. Quantitative sampling along a transect from the edge of the lake to the deepest area may provide information about changes in the relative abundance of different mytilocypridines as the lake dries.

Physical and chemical variables

Previous information about the salinity of tolerance of mytilocypridines is from active sampling, mainly of adults and late juvenile instars, in field records (Rahman et al., 2023). Although the interpretation of the data from rehydration trial 2 is complicated by the fact that we do not know the relative proportions of eggs of the different species that were added to cultures at sites where the species co-occur, they nevertheless provide the first indication of salinity tolerances in mytilocypridine species across their entire life cycle. Additionally, we were able to quantify the relationship between hatching and development and salinity for one population of each of M. mytiloides and A. insularis (for the two sites where only one species was recovered from the sediment).

The results show that individuals of A. insularis can hatch and grow to adulthood over a very broad range of salinities, including very high salinity (up to 100 g/L from one site), although hatching/development rates, were reduced at the extremes of the range. The results also suggest that the optimal salinity for hatching and development in M. mytiloides is lower than that for A. insularis. For M. mytiloides from LP, hatching and development started to decline at only 40 PPT. This would explain why the mean and median of the salinity distribution of M. mytiloides (Mean = 25.73 g/L, Median = 19.44 g/L) in the field is noticeably less than that of A. insularis (Mean = 55.97 g/L, Median = 48.36 g/L) even though both species occur in the field across a very broad salinity range (Rahman et al., 2023). Thus, although these two species have broad salinity tolerances, they also show evidence of a degree of specialization, which will reduce overlap in their temporal and/or spatial field distributions.

The documented ranges of salinity for A. insularis and especially M. mytiloides were higher than the salinity ranges over which individuals of that species were recovered from sediment samples in hatching experiments. This is most likely because the range over which the individuals of a species can hatch and develop may be lower than the salinities that adults can tolerate (Geddes, 1976), although we cannot rule out the possibility that suboptimal culture conditions might have influenced this outcome. Lawrie et al. (2024) found the reverse for six species of Coxiella gastropods from Australian salt lakes, i.e. salinity tolerances in laboratory experiments exceeded those in the field but these results were based on adults rather across the entire life cycle.

Conservation implications

We have provided information about factors that influence the resilience of mytilocypridine populations to habitat changes, such as increasing water salinity and decreasing hydroperiods, that are occurring as the climate in southern Australia becomes more arid (Atkinson et al., 2021). Although the results show that at least some mytilocypridine species have broad salinity tolerances, they also suggest that some species are more tolerant of salinity than others and that the salinities at which individuals can hatch and develop are lower than those based on field records for adults and other late instars. The results also show that desiccation-resistant eggs of some species can remain viable for periods of at least 27 months. These eggs will potentially buffer populations against a loss of genetic diversity and local extinctions during dry and other unfavourable conditions of at least this duration (e.g. see Hairston Jr et al., 1995). Finally, by revealing the presence of species that were not captured via active sampling, our results highlight the value of using sediment rehydration to capture components of the biodiversity of a salt lake that might not be available for active sampling (e.g. see Vandekerkhove et al., 2005a; Rosa et al., 2021a).

Limitations and future directions

Bet-hedging is a survival technique observed in some invertebrates species from inland aquatic environments (Simovich & Hathaway, 1997). It involves some eggs requiring more than one hydration event to hatch (Hairston Jr et al., 1985; Simovich & Hathaway, 1997; Rogers, 2015). This strategy enhances a population’s resilience to unpredictable and fluctuating environmental conditions (Rossi et al., 2004). Since our rehydration trials only used one rehydration event, these trials may not have captured the hatching potential of all the mytilocypridine eggs present in the sediments.

Our laboratory trials were conducted in a controlled temperature environment set at a constant temperature of 20 ± 1 °C with a 12-h light/dark regime. However, environmental conditions in Australian salt lakes are much more variable than this (Rahman et al., 2023), and if there is variability in hatching stimuli, some eggs may not have hatched under the selected conditions.

The salinity of the water used in the cultures was achieved using a commercial source of lake salt (Rehydration trial 1) and of sea salt (Rehydration trial 2). Using salts harvested from the lake where the mytilocypridines originated might provide a more optimal environment for hatching eggs and raising individuals, especially considering that some mytilocypridine species prefer water with different ionic compositions (Radke et al., 2003).

The egg banks in a temporary waterbody span multiple generations (Hairston Jr et al., 1995). Consequently, when conducting rehydration trials, it is important to consider the possibility that the egg banks in a lake may encompass species no longer capable of completing their life cycle in that lake if conditions have changed (Moscatello & Belmonte, 2009).

Future research should focus on the following:

  • Multiple rehydration treatments: Use multiple rehydration events to assess whether hatching of mytilocypridine eggs is synchronous or staggered (see Pinceel et al., 2021).

  • Long-term viability of eggs: Investigate viability of mytilocypridine eggs, both in the laboratory and in situ, over longer time periods.

  • Temporal perspective: Explore time shifts in the mytilocypridine fauna of sites using desiccation-resistant eggs from different cores/lake depths (see Hairston Jr et al., 1995).

  • Expanded dataset: Use similar experiments to document the tolerance of a broader range of mytilocypridine species to physical and chemical variables over their entire life cycle.

  • Egg identification methods: Test whether methods, such as electron microscopy, can be used identify eggs of different species a priori (e.g. see Timms, 2015; Meyer-Milne et al., 2022), thereby allowing control over the number of eggs of each species used in experiments.

  • Species discovery: Use rehydration trials for species discovery and monitoring in lakes, especially those in remote arid areas, which rarely hold water and are typically poorly studied (Halse & McRae, 2004).