High total dissolved gas (TDG) levels and excessive suspended sediment (SS) concentrations pose serious threats to fish survival during flood season. However, little information is available

on the effects of TDG supersaturation with varying SS concentrations on fish. In this study, laboratory experiments were performed to investigate the effects of TDG supersaturation with

varying SS concentrations on five-month-old river sturgeons (Acipenser dabryanus). The test fish were exposed to combinations of SS concentrations (0, 200, 600 and 1,000 mg/L) and TDG levels

(125, 130, 135 and 140%), and their mortality and median lethal time (LT50) were quantified. The fish showed abnormal behaviors (e.g., quick breathing, fast swimming and an agitated escape

(130–140%), there was no significant increase in LT50 with increasing SS. Therefore, river sturgeon showed weak tolerance of TDG-supersaturated water with SS.

In the last ten years, seven high dams (higher than 250 m), such as the Xiaowan Dam, Xiluodu Dam, and Jinping Dam, have been built in China. When these high dams discharge flood water, total

dissolved gas (TDG) supersaturation commonly occurs. A high level (128%) has been reported 180 km downstream of the Three Gorges Dam1. In addition, the discharged flood contains a large

quantity of sediment. The cumulative effects of high TDG levels and suspended sediment (SS) might seriously influence aquatic ecosystems2. Previous studies indicated that TDG supersaturation

threatened the survival of fish dwelling downstream of dams due to gas bubble disease (GBD)3,4. Similarly, mortality due to GBD was found in the Columbia River Basin5,6,7,8. Recently, many

studies on the effect of TDG supersaturation on endemic species, such as rock carp (Procypris rabaudi Tchang), Chinese sucker (Myxocyprinus asiaticus) and David’s schizothoracin

(Schizothorax davidi), in the Yangtze River have been conducted9,10,11,12. In 2014, China Central Television (CCTV) reported that a large number of fish died from TDG supersaturation in the

Jinsha River during flood discharge13. Furthermore, previous studies showed that SS at high concentrations had adverse effects on arctic grayling (Thymallus arcticus), Coho salmon

(Oncorhynchus kisutch) and coral reef damselfish (Acanthochromis polyacanthus)14,15,16. Increasing SS reduced the foraging and growth of coral reef damselfish (Acanthochromis

polyacanthus)16. The survival of common smelt (Retropinna retropinna) and Coho salmon (Oncorhynchus kisutch) was reduced by increasing SS concentrations17,18.

The mean sediment content in the flood season remained high in the Yangtze River from 1956 to 2010 (e.g., 535 mg/L in the Min River among the tributaries of the Yangtze River and 1,660 mg/L

in the Jinsha River in the upper Yangtze River). Moreover, TDG levels are mostly between 120% and 140% during flood discharge19,20. However, the potential harm due to higher SS

concentrations in addition to TDG supersaturation has not been previously considered. The population size of the river sturgeon, a class national protected animal in China, has decreased

greatly, and the species is facing extinction owing to habitat destruction. Therefore, the aim of this study was to investigate the effect of TDG supersaturation with varying SS

concentrations on river sturgeon from the upper Yangtze River. The data resulting from this study can provide a scientific basis for the protection of river sturgeon and other species and

will play an important role in establishing appropriate environmental standards to improve fish habitat.

The experimental proposal was approved by the Ethics Committee for Animal Experiments of the Sichuan Academy of Agricultural Sciences and Xihua University. All experiments were executed in

accordance with the animal management regulations of Sichuan Province in China.

The river sturgeon is a rare species in the upper Yangtze River and inhabits 8–10 m shallow waters. The species usually moves in slow-flowing water and can tolerate the wide water

temperatures (1–32 °C). River sturgeon grows fast and the water temperatures of 18–25 °C increase the growth. However, the growth rate of river sturgeon slows down when the water temperature

is below 13 °C or above 28 °C. The species is bred at the Sichuan Fisheries Research Institute under government authorization. In this experiment, five-month-old juveniles were acquired

from the Sichuan Fisheries Research Institute. Healthy juveniles (mean fork length ± standard deviation (SD), 13.4 ± 0.7 cm; mean weight ± SD, 7.1 ± 1.2 g) with a uniform body size were

selected. Experimental sediments were collected from the Jinsha River. Based on data from China’s river sediment communique21, the annual average sediment diameter was approximately 7 μm in

the Yangtze River in 2000–2010. Sediment with a median diameter of 7 μm was chosen for the experiment, matching the particle size in the Yangtze River.

The experimental device was mainly composed of a water flume, a pressure vessel, a pump, an air compressor and experimental tanks22. The sediment was maintained in suspension by using a

submerged pump placed in the bottom of the water flume. Turbid water from the water flume was sent into the pressure vessel and mixed with air from the compressor to create high

TDG-supersaturated water with SS. The resulting water was mixed with sandy water with 100% TDG to create different levels of TDG-supersaturated water with SS. The water temperature was

maintained at 21–22 °C by the temperature controller. A Point Four Tracker (Point Four Systems Inc., Canada) was used to measure the TDG level and water temperature. The pH and dissolved

oxygen (DO) were monitored by using a digital pH meter (JENCO Model 6010, China) and a dissolved oxygen meter (Oxi 3210 SET 3 Inc., Germany), respectively. Sediment concentration and

turbidity were measured with a turbidimeter (Hach 2100, China).

The test fish were acclimated for five days in a tank (720 L) in the Key Laboratory of Fluid and Power Machinery, Ministry of Education (Xihua University, China). During the acclimation

period, the flow-through water temperature was 21–22 °C, the pH value was 7.1–7.4, and the DO was 7.2–8.0 mg/L. Fish were given Limnodrilus hoffmeisteri as food every morning and afternoon.

The rearing tank was cleaned at the same time every day.

Based on field observational data, the TDG levels are mostly below 140% during flood discharge19,20. In this study, we set the TDG levels at 125, 130, 135 and 140% and investigated the

influence of TDG exposure with SS concentrations of 0, 200, 600 and 1,000 mg/L. Control groups were held at a 0 mg SS concentration with the four TDG levels. Two parallel experiments were

performed for each experimental group. Following acclimation and preparation of the experiment, 60 fish were moved into each experimental tank (length: 1 m; width: 0.6 m; water depth: 0.5 

m). During the trial, the water temperature, pH and DO were kept consistent with those of the acclimation period. The flow velocity of circulating water was approximately 0.5 cm/s in each

experimental tank. In addition, abnormal behaviors and symptoms of GBD of the test fish were continuously observed. The time of death of each fish was also recorded during the trial.

Mortality and median lethal time (LT50; the exposure time at which the mortality of fish reached 50% at a specific TDG level) were determined to investigate the effect of TDG-supersaturated

water with SS on river sturgeon.

The LT50 was calculated by the method of Shetty and Alwar23. The difference in LT50 among TDG levels was analyzed using one-way analysis of variance (ANOVA) followed by a least significant

difference (LSD) test. Dunnett’s T3 test was used when the assumption of homogeneity of variance was not met. Regression analysis was used to investigate the relationship between LT50 and SS

at different TDG levels. Furthermore, a two-way ANOVA followed by a post hoc multiple comparison test (Tukey test) was executed in SPSS to determine the effect of TDG supersaturation with

varying SS concentrations on river sturgeon. Data were expressed as the mean ± SD, and the significance level was set at P  0.05). The lowest LT50 value (1.45 h) was reached at 1,000 mg/L SS

and 140% TDG. As shown in Fig. 3, the LT50 values exhibited a significant negative linear correlation with the SS concentrations at different TDG levels (125% TDG: \(y=-\,0.0017x+4.2222\),

R2 = 0.9136, P = 0.015; 130% TDG: \(y=-\,0.0007x+2.4571\), R2 = 0.8224, P = 0.024; 135% TDG:\(y=-\,0.0007x+2.2288\), R2 = 0.766, P = 0.009; and 140% TDG: \(y=-\,0.0007x+2.0612\), R2 = 

0.7997, P = 0.01, where y is the LT50 value, x is SS concentration, and R2 is the regression coefficient).

Linear correlation between the LT50 and SS concentrations at different TDG levels. Regression analysis was used to investigate the relationship. In these equations, y is the LT50 value, x is

SS concentration, and R2 is the regression coefficient. P