1 INTRODUCTION

Brine shrimp (Artemia spp.) have a worldwide distribution. They have adapted to hypersaline waterbodies and occur in permanent, semi-permanent, and seasonal waterbodies that dry out in a predictable or unpredictable manner (Lenz and Browne, 1991). Their wide distribution has been helped by birds, including waterbirds, grebes, flamingos, and ducks, that transport Artemia cysts (hard-walled eggs) from one waterbody to another mostly through gut passage (Reynolds et al., 2015; Frisch et al., 2021). One potential disperser of cysts is the eared grebe (Podiceps nigricollis). This bird is a common species found in saline lakes in North America, where its populations vary between 0.5 and 5 million annually (Conover and Bell, 2020). Population of eared grebes in Europe and North Africa are between 110 000 to 170 000 (Waterbirds Populations Portal, http://wpe. wetlands.org/), and these birds occur in France, Balearic Islands, Spain, Romania, Turkey, and Africa (Varo et al., 2011; Amat et al., 2014).

The species Artemia franciscana inhabits Great Salt Lake (GSL) (Fig. 1), Utah, where densities of adult brine shrimp (hereafter referred to as Artemia) have been recorded as high as 8 inds./L of lake water (Wurtsbaugh and Gliwicz, 2001). In GSL, the Artemia population is dynamic and cyclical in nature based on salinity, temperature, nutrients, and phytoplankton species (Barrett and Belovsky, 2020). In the spring, nauplii, which are the first larval stage of Artemia, hatch from over-wintering cysts when water temperatures exceed 3 ℃; this usually occurs in late February or March. These nauplii grow into the first generation of adults. If conditions are favorable, adults will reproduce ovoviviparously, producing free-swimming young. Each year, two or three generations of Artemia are produced in GSL. When food becomes scarce or temperatures drop in the fall, adult Artemia switch to oviparity and produce cysts that can survive adverse conditions. Adult and juvenile Artemia die when GSL water temperatures drop below 4 ℃, while cysts are able to survive through winter (Stephens and Birdsey, 2002).

Artemia cysts have been commercially harvested from GSL and shipped worldwide to be fed to larval fish, shrimp, and other crustaceans at aquaculture farms (Belovsky et al., 2011; Frisch et al., 2021). The cysts are buoyant in GSL, and wind and water currents concentrate them into large streaks, where they can be harvested efficiently. The state's GSL Ecosystem Program monitors the Artemia population and stops the commercial harvest when cyst density falls below 21 inds./L of GSL water. The average cyst density in the fall before the start of the commercial harvest averaged 111 inds./L (A. KIJOWSKI, Utah Division of Wildlife Resources, Great Salt Lake Ecosystem Program, personal communication).

The value of GSL Artemia is not just from the commercial harvest of their cysts but also because they provide food for the millions waterbirds that the lake supports. Over 200 avian species spend time on GSL (Gwynn, 2002). During the fall, aerial surveys found that eared grebes Podiceps nigricollis population on GSL ranged annually between 0.5 and 5 million, which is over half the eared grebes (hereafter referred to as grebes) in North American (Conover and Bell, 2020). The diet of grebes while on GSL consists primarily of adult Artemia, but brine fly (Ephydra spp.) adults and larva are also consumed. Artemia cysts were consumed by 40% of grebes collected from GSL. After cold water temperatures caused a die‐off of adult Artemia, cysts became more prevalent in the diet of grebes (Roberts and Conover, 2013).

Utah's GSL Ecosystem Program monitors the population of Artemia cysts by counting cysts and determining their hatchability and viability (K. STONE, Utah Division of Wildlife Resources, Great Salt Lake Ecosystem Program, personal communication) to make sure there will be enough live cysts to restart the GSL Artemia population in the spring. Hatchability is the percentage of Artemia cysts that hatch into nauplii. Viability is the percentage of the Artemia cysts that contain an embryo capable of hatching when the right environmental conditions occur. Previous studies have shown that passing through the digestive tract of waterfowl aids seed germination of some wetland plant species (Agami and Waisel, 1986; Mueller and van der Valk, 2002; Green et al., 2016; Farmer et al., 2017). The chemicals of a duck's digestive system, as well as the grinding action of the gizzard, help seeds break dormancy (Kettenring, 2016; Marty and Kettenring, 2017). We hypothesize that a grebe's digestive system would be harmful for Artemia cysts and reduce their viability. The hard wall of an Artemia cyst is there to protect the embryo, and if that wall were to crack, we hypothesize it would reduce the probability of the embryo surviving the digestive system but would be beneficial to grebes that are trying to digest them. This research will also answer the question of what, if anything, grebes gain by consuming these cysts; because if their digestive system is unable to crack the cyst's hard wall, the bird will be unable to digest the embryo.

2 MATERIAL AND METHOD

Great Salt Lake is located in the Great Basin and is the largest terminal lake in North America. Its size varies considerably among years depending on the amount of precipitation that falls within its watershed. The highest recorded elevation of GSL of 1 284 m above sea level was recorded during 1986 when GSL's surface area was 5 950 km² (Stephens and Birdsey, 2002). At an elevation of 1 278 m, GSL has a surface area of 3 086 km².

Great Salt Lake has made newspapers across the world because it is continually setting new historic lows. GSL is shallow, and a small change in the lake's surface elevation results in large change in its surface area. For example, a 0.3-m drop in elevation results in a 45-km² reduction in surface area (Conover and Bell, 2020). As water volume of GSL shrinks, its primary production decreases by about the same proportion, which in turn, decreases brine shrimp and brine fly populations, and GSL's ability to sustain avian populations. Furthermore, as its water volume declines, its salinity rises. This is a threat to aquatic birds because if GSL's salinity gets too high, then populations of brine shrimp and brine flies will collapse (Conover and Bell, 2020).

Grebes were collected for this study throughout GSL during the spring migration (4/21/2020) and fall migration (10/28/2020) for another project, but their digestive systems were used for this study. Grebes were shot with a shotgun from a Utah Division of Wildlife Resources boat and were immediately placed in an individual plastic bag and placed on ice. A cyst sample was collected using a tow net from the upper 1 m of the water column at the same time and place where a grebe was collected. Each grebe was numbered, and GPS coordinates were recorded where every bird and cyst sample were collected. Each grebe and the corresponding cyst sample were immediately returned to Utah State University and placed in a freezer.

All birds were analyzed using the protocol developed by Great Salt Lake Ecosystem Program (2016). Each bird was thawed, sexed, aged, weighed, and numerous body measurements were taken. All food was collected from the esophagus to determine what the bird had consumed immediately before it was shot. The contents of the stomach and intestines were collected but kept separate. Contents of the stomach and intestines were a soupy material of digested food. The pH of this material in the stomach and intestine was measured because the average waterfowl gizzard may be acidic with a pH between 2.0 and 3.0, and this low pH may aid to food digestion (Carbonell et al., 2021). Our procedure yielded three cyst samples from each bird/ location; a water sample, stomach sample, and intestine sample. The pH of these samples was measured using an Aprea 1209 pH meter (Columbus, Ohio). The pH of water sample was compared to stomach and intestine samples using a KruskalWallis one-way analysis of variance corrected for ties. This test was used because the data were not normally distributed (Siegel, 1956). The contents of each sample were immediately washed through a series of sieves. The larger sieves filtered out debris but allowed the cysts to pass through them, while the 150-µm sieve caught all cysts, which were remarkedly similar in diameter. For each cyst sample, three subsamples were tested for hatchability and viability. Approximately 300–500 cysts from each subsample were placed in a small petri dish with liners and gridded Metricel^® membranes. Enough water was added to cover the membranes and all cysts. Hatchability was tested by using a microscope to count the number of cysts on the gridded membrane. Cysts were spherical and easy to identify; they were hydrated at 21–22 ℃ by covering them with water for > 24 h. During this period, the tops on the petri dishes were covered to reduce water evaporation. A microscope was then employed to count the number of unhatched cysts and those that had hatched into nauplii. Nauplii were easy to identify because they were yellow while cysts were a gray color. Nauplii also have a snow-cone shape when they first hatch. The mean proportions of cysts that hatched from the three subsamples were used as a measure of hatchability for the sample.

Viability was tested for immediately after completing the hatchability tests. Enough bleach was added to each petri dishes to cover the cysts. After 30 min, the bleach dissolved the cyst's shell so that cysts with embryos appeared as yellow balls while cysts lacking an embryo inside disappeared. These were counted to determine the number of cysts that contained embryos. The mean proportions of cysts that hatched into nauplii or that contained an embryo were used as a measure of hatchability. Hatchability and viability of cysts collected from males were compared to females, and those from adults were compared to juveniles using a KruskalWallis one-way analysis of variance corrected for ties.

3 RESULT

We collected cysts from 29 grebes including 12 males, 17 females or 17 adults and 12 juveniles. The pH of lake water was 7.6+0.8 (mean±SD), stomach contents was 6.4±0.5, and intestine contents 6.0±0.4; these differences were significant (χ²=19.48, df=2, P < 0.000 1). The difference in pH was also significant when the stomach contents was compared to just the intestine contests (χ²=3.99, df=1, P=0.04).

Viability was significantly different (χ²=45.94, df=2, P < 0.000 1) for cysts collected from the water column (29%), stomach (0.7%), and intestines (5%), but not when cysts from the stomach were compared to those from the intestine (χ²=1.34, df=1, P=0.25). Hatchability also was higher for cysts collected from the water column (19%) than from the stomach (0.3%) or intestines (3%). These differences in viability were significantly different (χ²=49.96, df=2, P < 0.000 1), but not when cysts from the stomach and intestines were compared to each other (χ²=1.67, df=1, P=0.20). Hatchability declined 87% and viability declined 83% after passing through the digestive system of grebes.

Cysts collected from the stomachs of male and female grebes did not differ in hatchability or viability, but cysts taken from the intestines of males had higher hatchability and viability than those taken from females (Table 1). Adult and juvenile grebes were similar in cyst hatchability and viability regardless of where the cysts were collected from stomachs and intestines (Table 1).

4 DISCUSSION

A hard chitin shell surrounds the cysts of Artemia; this shell consists of a hypochlorite-soluble chorion and a fibrous, hypochloride-resistant cuticle that can keep the desiccated embryo enclosed within it alive for years (Tajik et al., 2008). The shell protects the desiccated embryo inside it from acids as low as 1.5 pH, temperatures as low as -196 ℃, anabolic condition for at least 32 d, and from some digestive enzymes, including chitinase, pipase, pepsin, and trypsin (Horne, 1966; Hempel-Zawitkowska, 1971). Cysts are able to survive for years under severe dehydration; this ability allows Artemia cysts to repopulate waterbodies that dried up during a drought (Gajardo and Beardmore, 2012).

The main diet of grebes consists of adult Artemia that are soft and easily digestible (Conover and Vest, 2009; Roberts, 2013), but cysts are also found in their digestive system (Caudell and Conover, 2006). These cysts may be inadvertently ingested by grebes, but they are found in such large quantities that their consumption appears to be deliberate during late fall. Roberts and Conover (2013) examined 398 grebes collected on GSL and discovered that Artemia cysts composed 16% of the aggregate biomass in the grebes' esophagi. Vest and Conover (2011) collected over 600 common goldeneyes (Bucephala clangula), 200 northern shovelers (Spatula clypeata), and 100 green-winged teal (Anas crecca) from GSL from October to March; Artemia cysts composed 4%, 52%, and 80%, respectively, of the aggregate biomass in their esophagi. Such concentrations of cysts raise the question of why grebes and ducks consume so many Artemia cysts; their presence in the digestive system may be a byproduct of consuming oviparious females. Alternatively, birds may be ingesting cysts deliberately for their nutritional benefit; after all, a cyst is nutritious; a gram of cysts contains 23 kJ of energy (Caudell and Conover, 2006). However, the cysts' hard shell must be broken before a bird can obtain any nutritional benefit from consuming cysts. Ducks have a muscular gizzard that contains grit and are better able to break the hard objects, such as seeds, than waterbirds, which do not have a muscular gizzard (Proctor et al., 1967). But, can the digestive system of grebes that has evolved to digest adult Artemia quickly also digest cysts?

The ability of grebes' digestive system to break the hard shell of a cyst was examined in this study by comparing the proportion of cysts in GSL water that contained embryos to those in the stomach and intestines. Our results showed that 29% of cysts in the water were viable, compared to 0.7% in the stomach and 5% in the intestines. Hatchability of cysts declined 87% and viability declined 83% after passing through the digestive system of grebes. However, these values may be conservation because the percentages of cysts that were viable in the intestines are probably higher than they would be after they completed the passage through the entire digestive system.

Many of the cysts inside the digestive system of a grebe probably were ingested when the female carrying them was ingested. If some of these cysts were too immature to hatch, this could explain why the hatchability and viability of cysts inside a grebe is less than for cysts inside the water column. Additionally, grebes can digest most cysts they consume and that digestion occurs in the stomach. It is not clear how that happens. The stomach of grebes is not muscular, but perhaps it is acidic and is able to dissolve the hard shell. We tested this and found that the pH of the stomach and intestines are 6.4 and 6.0, respectively. Although both of these were significantly more acidic that GSL water, which has a pH of 7.6, they are not acidic enough to dissolve the shell of cysts. A grebe's stomach contains a dense mass of feathers, which grebes pluck from their own bodies, and swallow. These feathers slow the movement of food items through the stomach and allow only small food items or liquids to pass through to the intestines. Grebes use their tongue to squeeze their food against the top of their throat before swallowing the food so that they ingest little hypersaline water with their food. Adult Artemia excrete salt from their bodies so their salinity is low. A grebe has a body temperature of 40 ℃ (Ellis and Jehl, 2003). We hypothesize that the low salinity and warm temperatures within the grebe's digestive system may cause cysts to hatch while inside the digestive system, and the nauplii are then digested and their nutrients absorbed. Undigested cysts pass intact unharmed through the grebe's digestive system.

When Artemia cysts were fed to killdeer (Charadrius vociferus) and mallards (Anas platychynchos), the first cysts were excreted 5– 10 min after ingestion with the peak occurring 90 min after ingestion (Malone, 1965). Some cysts that were in the caeca were excreted after 20–25 h; no cysts were found in birds after 24 h (Proctor, 1964; Proctor et al., 1967). Grebes migrate in one night from GSL to their wintering ground in the Gulf of California. Presumably, during spring migration, grebes travel in one night from GSL to their breeding grounds (Frank and Conover, 2017). Grebes on GSL consume adult Artemia and cysts both during the fall on their way to wintering grounds in the Gulf of California and Pacific Ocean and during the spring on their way to breeding grounds in Canada and the northern U.S. Our finding that viable and hatchable cysts occur in the digestive systems of grebes indicate that grebes can vector live cysts from GSL throughout the Pacific Flyway. This can explain why Artemia collected in the avian flyways extending across both North and South America were more genetically similar than those across different flyways (Muñoz et al., 2013).

Although grebes digest most cysts that they consume, a proportion pass through the digestive system and would have been able to hatch and reach maturity if defecated into a suitable waterbody. Some viable cysts can hitch a ride to a new waterbody by being buried in a grebe's plumage, but most are carried inside the bird's digestive system (Sánchez et al., 2012). Grebes could transport Artemia by ingesting them on GSL and defecating them into other waterbodies; this transport could provide a mechanism allowing Artemia to repopulate ephemeral waterbodies (Frisch et al., 2021).

Artemia franciscana collected from GSL have been exported across the world to provide food for newly hatched fish and crustacea at aquaculture farms (Marden et al., 2020). Outside of North America, Artemia franciscana is an exotic species. Unfortunately, populations of this exotic Artemia have been established in western Europe and North Africa (Mura et al., 2006; Horváth et al., 2018). This has had drastic consequences for the native Artemia species of Artermia salina and A. parthenogenetica (Browne and Halanych, 1989). A. franciscana has displaced them due to its higher reproductive rate and more efficient filter-feeder system. Its ability to be dispersed by birds helps explain this rapid expansion in the alien range (Amat et al., 2005, 2007; Sánchez et al., 2016).

5 CONCLUSION

Artemia cyst viability (percentage of cysts that contain an embryo) and hatchability (percent of cysts that hatch) was measured from cysts that had passed through the digestive tract of eared grebes and cysts obtained directly from GSL at the same site where each grebe was collected. Hatchability and viability were significantly lower for cysts from the stomach or intestines of grebes than those collected from the water column. These results indicate that grebes nutritionally benefit from eating cysts and that cysts may be an important food source for grebes in late fall after the adult population of Artemia dies off due to the water becoming too cold. Also, enough cysts survive their passage through the digestive system that grebes can vector hatchable cysts to other waterbodies.

6 DATA AVAILABILITY STATEMENT

The data analyzed by this study are available from the corresponding author.

7 ACKNOWLEDGMENT

We thank all of folks with the Utah GSL Ecosystem Program, especially John LUFT, John NEILL, Kyle STONE, and Ashley KIJOWSKI for helping us in the field and lab and providing great advice; they were key to the success of this project. We thank Eve JEPPSEN for her help processing cysts. This research was approved by Institutional Animal Care and Use Committee of Utah State University (10087) and permitted by the state of Utah (1BAND10069, 2COLL10039), the U.S. Bird Banding Lab (21175), and the U. S. Fish and Wildlife Service (MB693916-0).