Dataset DOI: 10.5061/dryad.8kprr4z0g

Description of the data and file structure

We conducted an experiment to assess the rates of injury and mortality attributable to different turbine runners. Experimental procedures were identical for RHT and conventional turbine trials. Control groups were utilized to account for handling, release, and recapture effects. The control groups underwent identical handling to treatment groups, but were released downstream of the operating turbine. The RHT trial consisted of 36 passage events (N = 172 fish): 12 control passages (n = 60) and 24 treatment passages (n=112). The conventional trial also consisted of 36 passage events (N = 180 fish): 13 control passages (n= 65) and 23 treatment passages (n=115). Each passage event consisted of a group of five fish, except for one RHT treatment passage with only two fish. All groups underwent pre-exposure evaluations, release and recapture, post-exposure evaluations, and a 48 h post-passage holding period.

Due to the possibility of severe injury associated with turbine passage, all fish (treatments and control) were anesthetized prior to being placed in the injector chamber. Each group of fish was anesthetized in a 350 mg/L MS-222 solution buffered with 420 mg/L NaHCO₃ and 6.0 g/L NaCl (InstantOcean™). Fish remained in the anesthetic until loss of equilibrium and unresponsiveness to light physical stimulus indicated that they had reached stage IV surgical anesthesia.

Once anesthetized, fish were placed into the water-filled injector chamber, and chamber pressure was increased via air compressor slowly (over a period of 15 to 30 s) to approximately 103 kPa (10.5 m of head) above atmospheric pressure. The injector valve was then opened quickly (<1 s), allowing the pressurized air to push the water and fish into the pipeline, and closed after all fish were observed entering the pipeline (Watson et al. 2022). After passing through the pipeline (including the turbine, for treatment groups) the fish were extracted from the outlet pipe flow via the conical screen and recovered from the recapture tub with a dip net. The duration between removal from anesthetic to recapture at the outlet was kept below one minute to ensure that the fish would remain anesthetized during turbine passage.

After recapture at the outlet, the fish were returned to a buffered MS-222 solution (350 mg/L MS-222) to re-anesthetize. Immediately upon losing equilibrium, each fish was transferred to a lower concentration of buffered MS-222 (100 mg/L) to maintain their anesthetized state while they were inspected for injury. Fish that presented with mortal injuries were immediately euthanized with an overdose of buffered MS-222 solution (500 mg/L).

Once anesthetized or euthanized, each fish was imaged, weighed, measured, and evaluated for injuries. A visual inspection was conducted to identify and note injuries across all external body parts. Injuries were not rated for severity. No internal injury assessments were conducted post passage. Postexposure injury evaluations were performed by a single evaluator experienced in white sturgeon injury identification. Injuries were classified as amputations, hemorrhages, scale/scute loss, fin tears and gouges, or dermal lesions.

After injury assessment, all live fish remaining from each group were placed into a 19 L bucket placed inside one of two partially filled 1136 L circular holding tanks.Any mortalities occurring within the 48 h hold period were recorded and removed. Since fish were not tagged, 48 h mortalities were identified within their group and matched with their post-passage injury assessment based on fork length and other distinguishing features. After the 48 h holding period, all fish were euthanized with an overdose of buffered MS-222 (500 mg/L).

We provide three data files. The first is on the survival of the passed fish "Dryad_Survival_Data_2025-07-01.csv". The second reports the injury data "*Dryad_Injury_Datasheet_7.2.2025.csv" and the third is the output of a small simulation (described below) "FINAL_Simulated_Passage_Data_2025-05-28.csv".

Files and variables

File: Dryad_Survival_Data_2025-07-01.csv

Description: Data on the survival of juvenile white sturgeon undergoing passage trials passed either a restoration hydropower turbine (RHT), a thin-bladed conventional turbine (CONV), or none (CONTROL). Fish were passed at multiple velocities (VELOCITY_RPM). Mortality was assessed both immediately following passage and after a 48-hour recovery period.

Variables

• FISH_NUM: Unique numeric identifier for each individual fish, each fish only used in one passage event
• DATE: Date of passage event (yyyy-mm-dd)
• MASS: Mass (grams) of the fish
• FL: Fork Length (mm) of the fish
• TL: Total Length (mm) of the fish
• TRIAL_ID: Unique identifier for a single passage cohort, fish were passed in groups of 5.
• TREATMENT: Whether a passage event was a "CONTROL" passage with no exposure to a turbine runner, a "RHT" passage where a fish passed the restoration hydropower turbine, or a "CONV" passage where the fish passed a traditional thin-bladed turbine.
• TREATMENT_2: Similar to TREATMENT but reports what sort of Turbine runner was installed in the systems. This is important for control passages where even though fish didn't pass a runner their was either an RHT ("CONTROL_RHT") or CONV ("CONTROL_CONV") runner. Treatments wherein fish did pass a runner are labeled as either "RHT_RHT" or "CONV_CONV".
• VELOCITY_RPM: Rotational velocity of the turbine runner (rotations per minute)
• IMPINGED: A binary value (0 or 1) indicative of whether an individual fish became impinged within the turbine passage system (1 = impinged, 0 = not impinged)
• SEVERED: A binary value (0 or 1) indicative of whether a fish was severed by a turbine blade (1 = severed, 0 = not severed)
• IMMEDIATE_SURV: A binary value of whether a fish is considered to have survived the passage event (1 = survived, 0 = mortality)
• IMMEDIATE_MORT: A binary value of whether a fish is considerd to have died during the passage evnet (1 = mortality, 0 = survived)
• DELAYED_MORT: A binary value of whether a fish survived immedieatly post-passage but died during the 48 hour recovery window (1 = mortality, 0 = survived)
• TOTAL_SURV: A binary value of whether a fish survived 48 hours (1 = survived, 0 = mortality)
• TOTAL_MORT: A binary value of whether a fish died at anytime within the trial (1 = mortality, 0 = survived)

File: Dryad_Injury_Datasheet_7.2.2025.csv

Description: This data set is used to assess the occurrence of injuries to passed fish attributable to the passage event type (TREATMENT). A visual inspection was conducted to identify and note injuries across all external body parts. Injuries were not rated for severity. No internal injury assessments were conducted post passage. Postexposure injury evaluations were performed by a single evaluator experienced in white sturgeon injury identification. Injuries were classified as amputations, hemorrhages, scale/scute loss, fin tears and gouges, or dermal lesions.

Recorded injuries were scored with each occurrence of an injury increasing the score by one unit. For example, a fish with a torn pelvic fin, a torn pectoral fin, and a scrape on its rostrum would receive an injury score of 3. Injuries were not scored differently based on severity due to inherent subjectivity; all injuries, including mortal injuries like severing, received an injury score of 1. Injuries recorded pre-passage were attempted to be matched with injuries recorded post-passage within the 5-fish group. For example, if one fish in a group was recorded with a split pectoral fin prior to passage, and one fish was recorded with the same injury after passage, then it was assumed those were the same fish, and the pre-passage injury count was subtracted from the post-passage count. If a pre-passage injury was not recorded on any of the fish following passage, then the pre-passage injury was disregarded.

Variables

• FISH_NUM: Unique numeric identifier for each individual fish, each fish only used in one passage event
• DATE: Date of passage event (yyyy-mm-dd)
• MASS: Mass (grams) of the fish
• FL: Fork Length (mm) of the fish
• TL: Total Length (mm) of the fish
• TRIAL_ID: Unique identifier for a single passage cohort, fish were passed in groups of 5.
• TREATMENT: Whether a passage event was a "CONTROL" passage with no exposure to a turbine runner, a "RHT" passage where a fish passed the restoration hydropower turbine, or a "CONV" passage where the fish passed a traditional thin-bladed turbine.
• TREATMENT_2: Similar to TREATMENT but reports what sort of Turbine runner was installed in the systems. This is important for control passages where even though fish didn't pass a runner their was either an RHT ("CONTROL_RHT") or CONV ("CONTROL_CONV") runner. Treatments wherein fish did pass a runner are labeled as either "RHT_RHT" or "CONV_CONV".
• VELOCITY_RPM: Rotational velocity of the turbine runner (rotations per minute)
• IMPINGED: A binary value (0 or 1) indicative of whether an individual fish became impinged within the turbine passage system (1 = impinged, 0 = not impinged)
• SEVERED: A binary value (0 or 1) indicative of whether a fish was severed by a turbine blade (1 = severed, 0 = not severed)
• INJURY_SCORE: Total number of injuries, calculated as the sum of injuries detected on body parts minus any pre-trial injuries.
• BODY: Total number of injuries detected on the body
• HEAD: Total number of injuries detected on the head
• EYES: Total number of injuries detected on the eyes
• ROSTRUM: Total number of injuries detected on the rostrum
• MOUTH: Total number of injuries detected on the mouth
• NARES: Total number of injuries detected on the nares
• GILLS: Total number of injuries detected on the gills
• PECTORAL: Total number of injuries detected on the pectoral fin
• DORSAL: Total number of injuries detected on the dorsal fin
• ANAL: Total number of injuries detected on the anal fin
• CAUDAL: Total number of injuries detected on the caudal fin
• PELVIC: Total number of injuries detected on the pelvic fin
• BARBELS: Total number of injuries detected on the barbels
• MORT_SEVERED: A binary value, did this fish suffer a mortal injury via severing by a turbine blade (1 = mortal injury, 0 = no mortal injury)
• MORT_NOTOCHORD: A binary value, did this fish suffer a mortal injury via damage to the notochord (1 = mortal injury, 0 = no mortal injury)
• MORT_FACE: A binary value, did this fish suffer a mortal injury via damage to the face or head (1 = mortal injury, 0 = no mortal injury)
• MORT_OTHER: A binary value, did this fish suffer a mortal injury via an otherwise uncategorized injury (1 = mortal injury, 0 = no mortal injury)
• MORT_TOTAL: A binary value, did this fish suffer a mortal injury (1 = mortal injury, 0 = no mortal injury)

File: FINAL_Simulated_Passage_Data_2025-05-28.csv

To estimate the effect of repeated turbine passages on survival we used our lowest LOO-scoring 48h survival model to simulate passing through one of four different passage conditions: control passage with no turbine exposure, RHT-type runner passage, conventional runner passage, and an optimized conventional runner passage. This ‘optimized conventional turbine’ condition was added to represent the survival of repeated passage under theoretical conditions optimized for fish survival. For each passage event, each simulated fish was assigned a randomly generated runner peripheral speed between 15 m/s and 27.6 m/s. For the ‘optimized conventional turbine’ condition, the randomized runner peripheral speed was limited to between 19.0 m/s and 22 m/s, velocities which yield mean survival probability greater than 0.80 based upon the best fitting model. We used a simulated fish’s assigned velocity and passage treatment to predict its likelihood of survival. Then, if a randomly generated number was smaller than a given fish's predicted survival likelihood, the simulated fish was marked as surviving passage; otherwise, it was deemed to have died during passage. To assess the impact of repeated passage, the simulation was rerun on the dataset of all simulated fish that survived passage to represent a second passage event. This process was iterated to simulate a total of five sequential passage events. Each iteration of the simulation modeled 100 fish per passage condition. We conducted 10,000 interactions for a total of 4,000,000 simulated fish (1,000,000 per passage treatment). The output data reports the passage treatment, the number of fish entering that passage round, the mean survival of fish during that passage event, the number of prior passage events that cohort had experienced (maximum of 5), and the random seed number in R used to set intial compute conditions.

Variables

• TREATMENT: What type of turbine the simulated fish passed, one of 4 options, CONTROL - no turbine, RHT - restoration hydropower turbine, CONV - a conventional thin-bladed turbine, and CONV_HIGH_SURV - a conventional turbine rotating at velocities optimized for survival
• COUNT: The number of fish at the start of a given round
• MEAN_SURV_P: The mean survival rate for fish passing through the given treatment during a given round.
• ROUND: How many prior passages this cohort had been simulated through
• SEED_NUM: A seed number for setting the original compute conditions for the model estimates.