The effects of substrate and sediment burial on survival of developing pallid sturgeon (Scaphirhynchus albus) and shovelnose sturgeon (S. platorynchus) embryos

The shovelnose sturgeon (Scaphirhynchus platorynchus) and endangered pallid sturgeon (S. albus) deposit demersal and adhesive eggs in swift currents, near or over coarse substrate. Hydrographic surveys have demonstrated the dynamic nature of spawning habitats and that coarse substrates may episodically be buried (partially or completely) by fine sediments. To evaluate embryo survival of both species in various substrate conditions, laboratory trials were conducted with substrates of clean glass, gravel, medium-coarse sand (MCS), and fine sand-silt (FSS). Embryos in MCS and FSS were tested three ways: unburied, partially buried, and fully buried (1–2-mm depth). Embryos were exposed to trial conditions for 10 days from the day of fertilization (5 days beyond expected hatching). For both species, mean hatch of normally developed free embryos was highest in unburied treatments where embryos were incubated on substrates and not covered with sediments and ranged from 81.0 to 87.1% for shovelnose sturgeon and 55.2–80.0% for pallid sturgeon. Mean hatch of normal free embryos was lowest where incubating embryos were fully buried by MCS or FSS and ranged from 2.4 to 11.6% for shovelnose sturgeon and 4.8–15.2% for pallid sturgeon. We observed free embryos with physical abnormalities in all treatments; however, the occurrence was most variable in treatments fully and partially buried by MCS. Hatch of both species was also delayed in treatments where embryos were incubated fully and partially buried by MCS. Our results may be useful to estimate the relative suitability of spawning substrates in relevant river reaches.


Introduction
The shovelnose sturgeon, Scaphirhynchus platorynchus Rafinesque, 1820, and imperiled pallid sturgeon, Scaphirhynchus albus Forbes and Richardson, 1905, are benthic, rheophilic species endemic to the Mississippi River basin of the central USA (Bailey and Cross 1954;Keenlyne 1997;Jordan et al. 2016;Phelps et al. 2016). Distribution of the larger, piscivorous pallid sturgeon is generally limited to the turbid, mainstem Missouri and Mississippi Rivers downstream of the Missouri River confluence, and lower portions of larger connected tributaries, including the Yellowstone, Kansas, and Platte Rivers (U.S. Fish and Wildlife Service 1993, 2007, 2014Jordan et al. 2016). The shovelnose sturgeon is also found in smaller rivers throughout the Mississippi River basin 1 3 Vol:. (1234567890) including the Ohio River, and adjacent tributaries (Bailey and Cross 1954;Carlson et al. 1985;Rapp et al. 2011;Phelps et al. 2016). Populations of both species have declined over the last 100 years; population declines are thought to be due to substantial habitat fragmentation and modification, hydrologic alterations, and overharvest (U.S. Fish and Wildlife Service 1993;Hesse and Carreiro 1997;Keenlyne 1997;Mayden and Kuhajda 1997;Phelps et al. 2016). Whereas the shovelnose sturgeon remains relatively common in parts of the Missouri and Mississippi Rivers (Phelps et al. 2016), the pallid sturgeon was listed as endangered in 1990 by the U.S. Fish and Wildlife Service (U.S. Wildlife Service 1993, 2014). Despite confirmation of spawning (DeLonay et al. 2016a, b, c;Elliott et al. 2020), pallid sturgeon suffer from complete recruitment failure upstream from mainstem dams on the Missouri River (U.S. Fish and Wildlife Service 2007;Braaten et al. 2015) and show only limited reproductive success in the open river downstream of Gavins Point Dam (Steffensen et al. 2019). In contrast, shovelnose sturgeon are successfully recruiting in portions of the Missouri (Moos 1978;Keenlyne 1997;Phelps et al. 2016) and Mississippi Rivers (Tripp et al. 2009). It is hypothesized that alteration of spawning habitat abundance, availability, or suitability may help explain poor reproductive success of pallid sturgeon (DeLonay et al. 2016a;Jacobson et al. 2016b;Jordan et al. 2016).
Spawning of pallid sturgeon occurs in swift currents of large, turbulent rivers, including the Lower Missouri River (downstream from Gavins Point Dam in South Dakota to the confluence with the Mississippi River), Upper Missouri River (downstream of Fort Peck Dam in Montana to the headwaters of Lake Sakakawea in North Dakota), and Lower Yellowstone River (from the confluence of the Bighorn River in Montana to the confluence of the Missouri River in North Dakota (Bramblett and White 2001;DeLonay et al. 2016a, b, c). Spawning of pallid sturgeon in the highly modified Lower Missouri River has been documented at several distinct locations along hundreds of river kilometers from 320 to 1,290 km upstream from the confluence of with the Mississippi River (DeLonay et al. 2016a, b, c). Spawning locations in the Lower Missouri River consist of deep and fastflowing (1-1.58 m/s) areas adjacent to the navigation channel along outside revetted banks (DeLonay et al. 2016a, b;Elliott et al. 2020). Documented spawning events were recorded in a narrow zone between stable coarse substrate of bank revetments and migrating sand dunes (DeLonay et al. 2016a;Elliott et al. 2020). Documented spawning sites in the near-natural Lower Yellowstone River is limited to an 8-to 13-km segment upstream from the Upper Missouri River and are characterized by mean velocity of 1.11 m/s ranging from 0.74 to 1.43 m/s (DeLonay et al. 2016a, b, c) Substrate at these sites is predominantly fields of sand dunes interspersed with patches of gravel ranging from small discrete lenses (less than 5 m in size) to tens of meters in size (DeLonay et al. 2016b). Repeat mapping surveys at pallid sturgeon spawning sites in the Lower Yellowstone River revealed that bed morphology and substrate are highly dynamic and change on intra-and inter-annual timeframes with changes in discharge (DeLonay et al. 2016a, b, c). Shovelnose sturgeon have been documented spawning in the mainstem of the Yellowstone, Missouri, and Mississippi Rivers and in smaller tributaries (Christenson 1975;Bramblett and White 2001;Korschgen et al. 2007;Tripp et al. 2009;Richards et al. 2014;Phelps et al. 2016). While little is known about precise locations, shovelnose sturgeon are thought to spawn in current over coarse substrates in the mainstem of large rivers and upstream in smaller tributaries (Christenson 1975;Keenlyne 1997;Bramblett and White 2001;DeLonay et al. 2009).
Sturgeon are long-lived, slow to reach sexual maturity, and produce a large number of gametes on an annual or longer cycle (Billard and Lecointre 2001;Pikitch et al. 2005). The small egg size (2.5-3mm diam.) of shovelnose sturgeon and pallid sturgeon (Chojnacki et al. 2020), together with hydraulic conditions in large, turbid rivers, have precluded direct observation of egg transport and deposition at spawning sites. Characterization of the specific substrate conditions where embryos incubate and develop is therefore lacking. Information to date indicates that shovelnose sturgeon and pallid sturgeon spawn close to the riverbed over or adjacent to coarse substrates where demersal and adhesive eggs are broadcast into swift and turbulent currents. Eggs are thought to become adhesive within 2-13 min after fertilization (Dettlaff et al. 1993). The adhesive nature of eggs is thought to anchor the developing embryos to coarse substrates or within interstitial spaces during incubation (Dettlaff et al. 1993); however, simulations suggest that prior to becoming adhesive, eggs of shovelnose sturgeon and pallid sturgeon may be transported downstream from the spawning site into different habitat types (Chojnacki et al. 2020). There is concern that demersal and adhesive eggs of sturgeon may be buried by mobile dunes of fine sediment, which is thought to reduce hatching success (Duke et al. 1999;Kock et al. 2006;McAdam 2012;Jacobson et al. 2016a).
At documented pallid sturgeon spawning locations in both the Missouri and Yellowstone Rivers, the river channel has been characterized by mobile dunes of fine sediments moving over coarse substrates (DeLonay et al. 2016b;Elliott et al. 2020). Under such dynamic conditions, fine sediment could intermittently cover patches of coarse substrates (DeLonay et al. 2016b;Elliott et al. 2020). Development, survival, and successful hatch of shovelnose sturgeon and pallid sturgeon embryos adhered to coarse substrates and subsequently buried by mobile dunes of fine sediments has not been documented. Lacking this knowledge, it is difficult to assess suitability of habitat based upon substrate composition, particle size, compaction, or presence of interstitial spaces. Similar to other species, we hypothesized reduced embryo hatch rates when buried by fine sediments. Therefore, our objective was to assess survival and hatch of developing shovelnose sturgeon and pallid sturgeon embryos in a variety of substrate conditions to inform future evaluations of suitability of riverine habitat for embryo development. Experimental conditions were designed to simulate possible fates of embryos that may become partially or fully buried by sediments.

Materials and methods
We evaluated survival of developing embryos in a variety of substrate conditions in two successive laboratory trials during April (shovelnose sturgeon) and May (pallid sturgeon) 2016. Experiments were conducted at the U.S. Geological Survey, Columbia Environmental Research Center (CERC), in Missouri using non-chlorinated well water (generally, 305 mg/L hardness as CaCO 3 , 255-270 mg/L alkalinity as CaCO 3 , 670-690 µS/cm conductivity, 7.8-8.0 pH, dissolved oxygen 8.6 mg/L). Shovelnose sturgeon and pallid sturgeon adults, embryos, and hatched free embryos used in this study were treated according to animal care and use guidelines established by CERC (Animal Welfare Plan, policy number 1401). The study plan was approved by the designated animal care and use committee and followed pallid sturgeon propagation and handling protocols established by the U.S. Fish and Wildlife Service (USFWS; (U.S. Fish and Wildlife Service 2019a, b)). Studies performed were covered under Endangered Species permit TE207526-0.

Broodstock and induction
Shovelnose sturgeon and pallid sturgeon embryos were obtained through hormonal induction of reproductively ready adults in the laboratory. Shovelnose sturgeon broodfish, three females and four males, were obtained from the Missouri River in central Missouri in March 2016. Mean weight of female and male shovelnose sturgeon was 1.4 kg (± 0.10 SD) and 1.4 kg (± 0.16 SD), respectively. Pallid sturgeon broodfish, one female and two males, were selected from captive broodstock maintained at CERC. These were hatchery-origin progeny of wild adults from the Yellowstone and Upper Missouri Rivers. The female pallid sturgeon weighed 3.3 kg, and the two male pallid sturgeon weighed 3.7 and 2.6 kg.
Male sturgeon were induced to spermiate by administration of a single intramuscular injection of Luteinizing Hormone-Releasing Hormone analog (LHRHa; Syndel Laboratories, Vancouver, British Columbia) at a dosage of 50 µg/kg body weight administered approximately 24 h prior to expected milt collection. To collect milt, male sturgeon were positioned on a stretcher in a water bath, ventral side up with head and gills submerged in water. Milt was collected by using a flexible piece of poly-vinyl chloride tubing attached to the end of a 30-ml syringe. The tubing end was inserted into the uro-genital vent of the sturgeon and gentle suction applied (Conte et al. 1988;Chebanov and Galich 2011). Milt from each male was transferred into separate oxygenated plastic bags and refrigerated until needed for fertilization (less than 48 ho after collection). On the day of spawning, milt samples from each male were examined for viability at 400-1000 × magnification using a Nikon Labophot light microscope (Nikon Instruments Inc., Melville, New York).
Female sturgeon were evaluated for reproductive readiness using ultrasound and by calculating the polarity index of a sample of oocytes collected via minimally invasive biopsies Wildhaber et al. 2007;Candrl et al. 2010;Chebanov and Galich 2011). Reproductively ready female sturgeon were induced to ovulate with an intramuscular injection of LHRHa at 50 µg/kg body weight and a resolving dose of white sturgeon, Acipenser transmontanus Richardson, 1836, pituitary extract (Argent Aquaculture LLC, Redmond, Washington) at 2 mg/ kg body weight given 12 h later. At ovulation, shovelnose sturgeon were euthanized with a combination of Tricaine-S (MS-222; Syndel Laboratories, Vancouver, British Columbia) and a blow to the head. Eggs were excised from the body cavity through an incision in the abdominal wall. On April 26, 2016, eggs from 3 shovelnose sturgeon were fertilized with equal aliquots of milt (diluted with water at 1:200) from 4 males. At ovulation, on May 12, 2016, eggs from one pallid sturgeon were expressed through a 2-cm surgical incision using gentle abdominal pressure (Conte et al. 1988). After closing the incision with two simple interrupted sutures, the broodfish was returned to a recovery tank. Eggs were fertilized with equal aliquots of diluted milt (1:200) from two males. After fertilization, Fuller's earth (Sigma-Aldrich, Saint Louis, Missouri) was added to fertilized eggs of both species and gently mixed for 20 min to remove adhesiveness as the eggs become sticky and difficult to manipulate minutes after fertilization (Conte et al. 1988;Chebanov and Galich 2011;Van Eenennaam et al. 2012). Fertilized eggs from each female were maintained separately in screened containers floating in a flow-through water bath maintained at spawning temperature (18.0 and 20.5 °C for shovelnose sturgeon and pallid sturgeon, respectively) until cell division began (approximately 4-5 h post-fertilization).

Experimental design
The estimated optimal temperature for survival of shovelnose sturgeon and pallid sturgeon embryos from published sources is 17-18 °C (Kappenman et al. 2013). For shovelnose sturgeon, peak of hatching reported by Kappenman et al. (2013) was 5.0 days at 20 °C and 8.4 days at 16 °C. For pallid sturgeon, the peak of hatching was 4.3 days at 20 °C and 6.4 days at 16 °C (Kappenman et al. 2013). We performed experimental trials lasting 10 days at 18 °C to ensure that the trials lasted beyond the expected hatching period for both species. We tested each species using four incubation substrates -clean glass (control; CG), gravel (GR), medium-coarse sand (MCS), and fine sand-silt (FSS). Control and GR treatments were not covered with additional substrate. Developing embryos incubated in MCS and FSS treatments were tested under three burial scenarios: no burial (NB), partial burial (PB) or full burial (FB; Fig. 1). The intent of the PB treatments was to provide a uniform covering of substrate without fully burying embryos, whereas the intent of the FB treatments was to bury embryos to a depth of approximately 1-2 mm. The experiment followed a randomized complete block design where each of the 8 treatments (CG, GR, MCS-NB, MCS-PB, MCS-FB, FSS-NB, FSS-PB, and FSS-FB) was replicated in ten blocks and was randomly assigned a location within each block.
Gravel and sand from the Missouri River were obtained from Capital Sand Company in Jefferson City, Missouri. Mean particle size of gravel was 28.1 mm. Sand was sieved to 0.3-1.2 mm to achieve MCS. Gravel and MCS were thoroughly rinsed, dried for 24 h at 100 °C, then cooled prior to placement in the test chambers. Sediment with total organic content less than 1% was obtained from the upper Spring River in Jasper County, Missouri. Sediment was sieved to < 1.0 mm, resulting in FSS.
Standard glass laboratory beakers notched at the 1-l level were used as test chambers. The notches were covered with 280 µm (µm) stainless-steel screen to allow water to flow through the chamber while retaining hatched sturgeon free embryos (Fig. 1). The CG treatment had no substrate added to the clean glass of the treatment chamber. In the GR treatment, a layer of gravel approximately 2 particles deep was placed in test chambers. In MCS and FSS treatments, a base layer of 100 ml of either MCS or FSS, respectively, was placed in the bottom of test chambers. Grids (individual cells 13 × 13 × 13 mm) cut from plastic fluorescent light diffusers, were placed in each test chamber to prevent embryos from clustering and/ or facilitate embryo burial. Two small, glass weights were added on top of the grids to hold them in place. Test chambers were filled with 750 ml of well water and placed in a circulating, constant 18.0 °C water bath (4.85-m length, 0.9-m width, 0.37-m depth, filled to approximately 11.5 cm with well water) at least 24 h prior to stocking. Ambient laboratory light was adjusted to approximately 100 lx, except during examination periods.
A total of 80 test chambers (10 replicates per each of 8 treatments) were each stocked with 21 viable embryos (confirmed microscopically) on the day of fertilization. After the onset of cell division, aliquots of fertilized eggs from each female were transferred into separate 2-L polypropylene pitchers with well water at spawning temperature and allowed to acclimate to experimental conditions (18 °C) at least 1 h prior to stocking chambers. Developing embryos were singly pipetted into test chamber grid cells (Fig. 1). For the shovelnose sturgeon experiment, 7 embryos from each of three females were stocked into each test chamber for a total of 21 viable embryos. All 21 pallid sturgeon embryos were from one female. Embryos in the MCS and FSS burial treatments were additionally covered with 25 ml (PB) or 50 ml (FB) of corresponding substrate (either MCS or FSS). Depth of substrate covering each embryo in burial treatments varied slightly due to uneven distribution of substrate that occurred during settling. After the embryos were stocked into the test chambers and additional substrates were applied as necessary, fresh well water was continuously delivered at a rate of 109.6 ± 6.3 ml/ min during the 10-day trials.
Test chambers were examined at least twice daily, and hatched free embryos were removed, counted, and preserved separately in 10% neutral buffered formalin for later microscopic examination (described below). At the conclusion of the trials, each test chamber was emptied, the contents were sieved through 1.2-mm mesh screen, and remaining specimens (hatched or unhatched embryos) were collected and preserved separately in neutral buffered formalin for later examination. Preserved specimens from each replicate were microscopically examined using a Nikon SMZ1500 stereo microscope (Nikon Instruments Inc., Melville, New York). Hatched free embryos were categorized as either normal or abnormally developed. Abnormalities of free embryos included edemas (excessive fluid in the abdomen) and spinal curvatures. While the majority of abnormalities of free embryos may not have been lethal at the time of observation, affected free embryos were unable to swim normally and would have been unlikely to survive in the river (Ruban et al. 2006;Chebanov and Galich 2011). Unhatched embryos were categorized as either nonviable (deceased) or viable based on the presence of a discernable embryo within the

Analysis
Vials of preserved specimens from two test chambers in the shovelnose sturgeon trial inadvertently desiccated awaiting examination, one MCS-FB replicate and one FSS-PB replicate. These two replicates were omitted from the summary and analysis because specimens could not be properly categorized. Percent normally and abnormally developed hatched free embryos, and unhatched viable embryos relative to the total number of eggs stocked in each replicate were recorded (n = 21 in each replicate). All values are reported as mean ± SD. Data are included in Chojnacki et al. (2022). All analyses were performed in R Studio version 4.0.2 (RStudio Team 2019). Prior to statistical analysis, proportion of normally hatched free-embryo data were tested for normality (Shapiro-Wilk normality test). The data were arcsine square root transformed to achieve a more normal distribution. We tested for a block effect in our initial analysis of variance (ANOVA); however, no block effect was detected, and the model was reduced to include only species and treatment as factors. Twoway ANOVA was performed to analyze the effect of species and treatment on proportion normally hatched free embryos. We graphically assessed the normality of residuals. Tukey's Honestly Significant Differences (HSD) post hoc test was used for pair-wise comparisons. A significance level of α = 0.05 was used to judge the results of all statistical tests.

Results
Mean water temperature was 18.03 ± 0.11 °C in both trials. Mean dissolved oxygen was 8.62 ± 0.19 mg/L for all treatments. Mean water hardness was 304.8 ± 5.64 mg/L as calcium carbonate (CaCO 3 ) and mean alkalinity was 261.8 ± 2.69 mg/L as CaCO 3 . Mean pH was 7.98 ± 0.12 and conductivity was 676.0 ± 5.37 µS/cm. Mean ammonia during the trials was 0.03 ± 0.017 mg/L. Hatching of both shovelnose sturgeon and pallid sturgeon began at 5 days post fertilization. In total, we examined 1025 shovelnose sturgeon and 833 pallid sturgeon hatched free embryos, of which 907 shovelnose sturgeon and 787 pallid sturgeon free embryos were normally developed (normal). Although there was a significant difference of percent normal hatch between species (ANOVA: F (1, 142) = 14.1, p < 0.05), species accounted for just 1.8% of the variation whereas, treatment accounted for 77.1% (Table 1). Overall, the mean percent hatch of normal free embryos was similar between species within each treatment, except for the GR treatment in which normal hatch of shovelnose sturgeon was significantly greater (Tukey's HSD: p < 0.05; Table 2 and Fig. 2). In the control treatments, mean percent hatch of normal shovelnose sturgeon was 81.0 ± 15.1%slightly, but not significantly, greater than that of pallid sturgeon at 71.4 ± 17.1% (p = 0.1; Table 2). In general, mean percent hatch of normal free embryos was greatest in unburied treatments (CG, GR, MCS-NB, and FSS-NB; 55.2-87.1%) and lowest in fully buried treatments (MCS-FB and FSS-FB; 2.4-15.2%).
Mean percent hatch of normal shovelnose sturgeon was greatest in unburied treatments (CG, GR, MCS-NB, and FSS-NB) and ranged from 81.0-87.1%; (Tables 2 and 3 and Fig. 2). The lowest mean hatch of normal shovelnose sturgeon occurred in three of the four buried treatments (MCS-FB, FSS-PB, FSS-FB) and ranged from 2.4 to 18.0%. The percent of normally developed shovelnose sturgeon that hatched when incubated in MCS-PB treatments did not differ from unburied treatments (CG, GR, MCS-NB, FSS-NB) but was greater than those from all other buried treatments (MCS-FB, FSS-FB, FSS-PB; Table 3). The percent of normally developed shovelnose sturgeon free embryos that hatched when incubated partially covered with fine sand-silt (FSS-PB) did not differ from full burial by medium-coarse sand treatment (MCS-FB). Percent of normally developed free embryos that hatched in FSS-PB was higher than fully buried with fine sand-silt (FSS-FB), and was lower than in unburied treatments (CG, GR, MCS-NB, FSS-NB) and partially buried medium-coarse sand treatments (MCS-PB). Standard deviation of mean hatch of normal shovelnose sturgeon was highest in the MCS-PB and GR treatments, 24.7% and 17.9%, respectively (Table 2). Mean percent hatch of normal pallid sturgeon was greatest in unburied treatments (CG, GR, MCS-NB, and FSS-NB), ranging from 55.2 to 80.0% (Tables 2  and 3; Fig. 2). The lowest hatch of normal pallid sturgeon was observed in three of four buried treatments (MCS-FB, FSS-PB, FSS-FB), ranging from 4.8 to 25.2%. The percent of normal embryos that hatched when incubated partially covered with mediumcoarse sand (MCS-PB) did not significantly differ from CG, GR, and FSS-PB treatments. The percent of normal embryos hatched in MCS-PB treatments was greater than when incubated in fully buried treatments (MCS-FB and FSS-FB), and less than those in MCS-NB and FSS-NB (Table 3). The percent of normal pallid sturgeon that hatched when incubated partially covered with fine sand-silt (FSS-PB) did not differ from MCS-FB and MCS-PB but was lower than unburied treatments (CG, GR, MCS-NB, FSS-NB).
As with shovelnose sturgeon, standard deviation of mean hatch of normal pallid sturgeon was highest in the MCS-PB and GR treatments at 20.7% and 19.2%, respectively (Table 2).
In total, 118 shovelnose sturgeon and 46 pallid sturgeon free embryos were abnormally developed. Abnormalities of free embryos included edemas and spinal curvatures. The greatest number of abnormalities were observed in MCS-FB treatments; 41.3 ± 31.0% of shovelnose sturgeon and 9.0 ± 17.6% of pallid sturgeon ( Table 2). Abnormalities of hatched free embryos were also high in MCS-PB treatments; 11.4 ± 11.3% of shovelnose sturgeon and 4.8 ± 6.3% of pallid sturgeon. Notably, some free embryos from partial and full burial treatments had severe spinal curvatures where the tail curved forward around the transverse body plane. Few abnormal free embryos (less than 3%) were recovered in other treatments. partially buried in fine sand-silt; FSS-FB, incubated fully buried in fine sand-silt (1-2-mm depth). Different letters denote significantly different hatch rates of normally developed free (P, 0.05). The box represents the interquartile range, the black line within the box represents the median, whiskers represent the values within 1.5 times the interquartile range greater than the 75th percentile and less than the 25th percentile, and black circles represent outliers We microscopically examined unhatched embryos recovered from all treatments and determined that only 34 shovelnose sturgeon and 23 pallid sturgeon embryos appeared viable and capable of hatching. Viable unhatched embryos of both species were recovered in MCS-FB treatments and in MCS-PB treatments and ranged from 6.7 to 10.6% and from 3.8 to 6.7% of stocked embryos, respectively, for shovelnose sturgeon and pallid sturgeon (Table 2). A single viable unhatched pallid sturgeon embryo was recovered in the FSS-PB treatment.

Discussion
Shovelnose sturgeon and pallid sturgeon generally exhibited similar hatch rates of normally developed free embryos in comparable treatment conditions, except the gravel treatment where hatch of pallid sturgeon was lower and more variable. Hatch of normal free embryos of both species was highest in unburied treatments in which developing embryos were placed on top of substrates and not covered with additional sediment. Hatch of normal free embryos of each species in unburied MCS and FSS treatments was similar to control and gravel treatments, suggesting that contact with these sediments did not result in decreased hatch. In general, mean hatch of normal free embryos in the partial burial treatments were intermediate between unburied and fully buried treatments. Hatch of normal free embryos of both species was lowest in both fully buried treatments.
Results of this study clearly indicate that embryos of both species are sensitive to substrate burial, even under conditions of continuous replacement of overlying water and low organic matter. Complete burial by as little as 1-2 mm of either sediment type reduced mean hatch of normal free embryos to approximately 15% or less. Similarly, hatch rates of grass carp, Ctenopharyngodon idella Valenciennes in Cuvier and Valenciennes, 1844, were found to be significantly lower in a laboratory study where eggs were buried to a depth of 2 mm by fine sand, particle size 0.053-0.6 mm (George et al. 2015). Other laboratory trials assessed the duration of burial in 5 mm of sediment (particle size < 1.0 mm) for 4, 7, 9, 11, or 14 days on embryo survival or hatch of white sturgeon eggs and found a negative correlation with duration of burial (Kock et al. 2006). While survival rates of white sturgeon embryos exceeded 80% in control treatments (i.e., without sediment cover), survival rates for buried embryos was 50% at 4 days 30% at 7 days, and 15-20% for those covered for 9, 11, or 14 days after burial (Kock et al. 2006).
In our study, partial and full burial of developing embryos in sediments also resulted in non-lethal  effects including delayed hatch and physical abnormalities of hatched free embryos. We monitored test chambers for 5 days beyond the expected onset of hatch at 18 °C, and at the conclusion of the 10-day trials, recovered viable, unhatched embryos in both partial and full burial treatments of MCS. It is unknown whether these embryos would have hatched given additional time. Our results are consistent with laboratory trials that have shown that burial of white sturgeon (Kock et al. 2006), and grass carp (George et al. 2015) embryos resulted in similar sub-lethal effects (delayed hatch and developmental abnormalities) when covered by a layer of sediment. Free embryos with physical and developmental abnormalities, such as those we observed, may survive for a short time in a laboratory setting; however, these individuals would likely be susceptible to predation or other sources of mortality in the wild.
Reduced exchange of respiratory gases (oxygen and carbon dioxide) due to decreased water renewal rates within sediments immediately surrounding developing embryos (Greig et al. 2005) likely contributed to the reduced hatch of normal free embryos, delayed hatch, and physical abnormalities when embryos were fully or partially buried in fine sediment (MCS and FSS). Fine sediments are known to reduce the intrasubstrate porosity and permeability of gravel substrates, thereby reducing the intrasubstrate water flow and renewal rates (Greig et al. 2005;Kemp et al. 2011). Fine sediments physically coat the surface and block micropores of the chorion and can reduce respiratory gas exchange across the chorion membrane (Greig et al. 2005). Hypoxia has been shown to adversely affect embryonic development resulting in abnormalities and delayed hatching (Silver et al. 1963;Dettlaff et al. 1993;Shang and Wu 2004). Hatching of sturgeon embryos is thought to result from enzymes secreted by the hatching gland combined with active movements by the embryo within the chorion, both of which may be reduced by hypoxic incubation conditions, resulting in delayed hatch (Dettlaff et al. 1993). Delayed or abnormally developing embryos in buried treatments in our study may have experienced chronically low oxygen conditions and an inappropriate micro-environment for successful hatching.
Variation of hatch within fully and partially buried treatments in part may have been due to the challenge of consistently burying the embryos with fine sediment to the same depth, despite careful application effort. Although eggs of pallid sturgeon and shovelnose sturgeon are demersal, they were difficult to bury. Embryos were easily disturbed and tended to become slightly buoyant as sediments were added to test chambers. Similarly, it was noted that the semi-buoyant eggs of the Rio Grande silvery minnow, Hybognathus amarus Girard, 1856, became positively buoyant and were easily resuspended when concentrations of suspended sediments exceeded specific gravity of the egg (Medley and Shirey 2013). It is thought, however, that under natural conditions, sturgeon embryos would be adhered to coarse substrates (Dettlaff et al. 1993;Parsley and Kofoot 2013) and therefore unlikely to become buoyant. Theoretically, the resulting minor inconsistencies in burial depth in our study should be similar to those expected in a natural setting.
Similar to other sturgeons, shovelnose sturgeon and pallid sturgeon are thought to spawn relatively close to the riverbed, where demersal eggs adhere to or within interstitial spaces of coarse substrate (Bruch and Binkowski 2002;Parsley et al. 1993). In the modified contemporary Lower Missouri River, pallid sturgeon spawning events have been documented in deep, high-velocity areas of outside bends, between bank revetments and migrating sand dunes (DeLonay et al. 2016b;Elliott et al. 2020). Documented pallid sturgeon spawning sites in the near-natural Lower Yellowstone River are predominantly characterized by fields of mobile sand dunes interspersed with patches of gravel (DeLonay et al. 2016c). Given the energetic and turbulent conditions at documented pallid sturgeon spawning sites, simulations indicate that eggs may be transported several hundred meters downstream prior to becoming adhesive (Chojnacki et al. 2020), to locations where habitat could vary substantially from that at the spawning site. Although specific locations and conditions of shovelnose sturgeon spawning sites have not been documented, they spawn in the mainstem of large rivers and upstream in smaller tributaries (Christenson 1975;Bramblett and White 2001;DeLonay et al. 2009). Our study indicates that hatch of pallid sturgeon and shovelnose sturgeon normal free embryos was reduced when buried to a depth of 1-2 mm in either fine sediment. These results suggest that the transport of fertilized embryos into depositional zones where burial in substrate may occur is unlikely to support survival and hatch of normal free embryos and that these species require clean substrates for spawning and incubation to support successful reproduction and recruitment. Field studies of many sturgeon species have documented spawning in flow conditions that are likely to minimize accumulation of fine sediments over gravel, cobble, and other coarse substrates with interstitial spaces that are thought to protect embryos from predation and scour by mobile sediments (Parsley et al. 1993;Paragamian and Kruse 2001;Perrin et al. 2003;Du et al. 2011;Baril et al. 2017). Whereas results from this study suggest that embryos may experience mortality, physical abnormalities, or delayed hatch if continually covered by fine sediment, the fate of embryos is unclear if sediments cover and uncover developing embryos in repeated episodes as might be expected to occur in large rivers with mobile dunes of fine sediment.
Microhabitat and abiotic conditions during the 3-8 days of embryo incubation are important factors for successful hatch. Documenting early life mortality in situ is inherently difficult in large, turbulent, and turbid rivers, which preclude direct observation. Laboratory studies improve understanding by replicating and scaling down complex field conditions. Future research in experimental flumes could help further define factors contributing to distribution and survival of Scaphirhynchus embryos in rivers where these species spawn. Although inherently difficult to study in large rivers, enhanced understanding of interactions of newly deposited eggs with complex substrates and near-bed hydraulics, and the subsequent role of adhesion, is needed to improve understanding of conditions necessary for successful incubation and hatch. Results from the present study may be useful to estimate the relative suitability of substrates selected by spawning sturgeon in relevant river reaches.
Acknowledgements The authors would like to thank Eric Brunson and Chris Ivey for assistance with setting up the experiment. James Candrl and Rachel Claunch were instrumental in the induction and spawning of shovelnose sturgeon and pallid sturgeon. We thank David Combs, Sabrina Davenport, Chad Vishy, Marlene Dodson, Mitchel Blehm, Kelsey Perry, and John Calfee for laboratory assistance and sampling. We also thank Marlene Dodson and Killian Kelly for assistance with specimen examination and Parker Golliglee for assistance with statistical analysis. The authors would also like to thank Diana Papoulias, Michael Mac, and three anonymous reviewers for helpful comments on previous versions of this manuscript. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Funding Funding for this research was provided by the U.S. Army Corps of Engineers, Missouri River Recovery Program-Integrated Science Program and the U.S. Geological Survey Ecosystems Mission Area. Data availability Data generated by USGS during this study are openly available as a USGS data release in Sciencebase.gov at https:// doi. org/ 10. 5066/ P919L DSY (Chojnacki et al. 2022).

Declarations
Ethics statement Pallid sturgeon and shovelnose sturgeon adults, embryos, and hatched free embryos used in this study were treated according to animal care and use guidelines established by CERC (Animal Welfare Plan, policy number 1401).

Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.