Abstract

This study evaluated the ability of surgically sterilized, hatchery reared Gulf sturgeon to be used as sentinels for finding remnant populations and appropriate habitat in the wild. Twelve 1995 year class (eight male and four female), and 20 1998 year class (twelve male and eight female) hatchery reared Gulf sturgeon were obtained from Welaka National Fish Hatchery in Welaka, Florida. The fish were transported to the Warm Springs Regional Fisheries Center in Warm Springs, Georgia, and maintained in a 12-foot diameter circular tank. The fish ranged in length from 811 to 1281 cm and from 4.5 to 5.9 kg in weight. The fish were acclimated for several months prior to attempting any surgical procedures. Each fish was anesthetized by induction in a static bath of MS-222 at a concentration of 250 mg/L and then moved to a recirculating anesthesia machine containing an 85 mg/l solution of MS-222. A 10-cm incision was made along the ventral midline of the fish from the pectoral fins caudally. Bipolar cautery endoscopic scissors were used to carefully cut the tissue attachment of the gonad to the body wall from the liver to a point approximately midway down the length of the fish. A large vessel exists at the most cranial attachment of the gonad to the body. Hemostasis of this vessel was achieved by use of medium hemoclips. Second and third vessels of moderate size are located at the opening of the mullerian duct to the celoem and immediately caudal to the swim bladder. Hemostasis of these vessels was achieved with simple bipolar cauterization prior to severing the tissue. A second eight-cm incision was made in a cranial direction, approximately five cm cranial to the vent, along the ventral midline to allow access to the remainder of the gonadal tissue, while leaving a section of the body wall intact that aided in healing. A single moderately sized vessel exists at the posterior attachment of the gonad to the body wall. Hemostasis of this vessel was achieved with bipolar cauterization prior to removing the gonad. This procedure was repeated bilaterally on the male fish. On female fish, this procedure was only completed on one side and was followed by bilateral double ligation of the mullerian ducts with 2-0 PDS. The two incisions were then closed with 0 Dexon in a horizontal mattress pattern and followed by skin sutures using 2 PDS in a cruciate pattern. Betadine ointment was applied to the incision sites and each fish received 0.2 mg/kg Butorphanol intraperitoneally. The fish were transported back to their holding tank for recovery from anesthesia. The incisions were evaluated weekly and sutures removed when healing was complete. Healing time varied from three to eight weeks. All but one of the 1995 year class animals survived the procedure. The male that did not survive was not a good candidate before the surgery and had never been seen to eat since arriving at Warm Springs. The 1998 year class animals did not tolerate the procedure, and only two of the animals survived the procedure. In all cases, the animals appeared to do very well post operatively, but gradually declined over a period of two to three weeks. Necropsy provided no significant findings or cause of death. Histopathology of the removed tissue showed only gonadal tissue present.

Two weeks prior to their scheduled release, each fish was re-anesthetized using the protocol previously described and a four-cm incision was made approximately midway between the ventral midline and the first row of lateral scoots in the caudal third of the celoemic cavity. A radio transmitter was inserted into this incision and the antenna was routed subdermally through a 14g needle inserted through the dorsum of the animal. The incision was closed with 2 PDS in a cruciate pattern, and the fish recovered in their holding tank. One week prior to the release of the hatchery reared, sterilized fish, five wild fish from the Apalachicola River were captured and implanted with radio transmitters in the same manner as described above. These animals served as wild controls to assess behavior of the hatchery reared and sterilized fish.

Three sterilized female Gulf sturgeon, two 1995 year class and one 1998 year class, were released in the Apalachicola River immediately below Jim Woodruf Dam, where the tagged wild fish were located. Four sterilized, 1995 year class, male animals were released above the dam into Lake Seminole. Of these four animals two were released into the main channel of the Flint River and two were released in the main channel of the Chattahoochee River. Four additional animals, one 1995 year class female, two 1995 year class males, and one 1998 year class male, were released into the Flint River approximately 15 miles north of Lake Seminole. The remaining two 1995 year class animals (one male and one female) were retained at Warm Springs to serve as controls for the sterilization procedure.

From the release date until September 1, 2002 we attempted to locate each fish three to four times per week. Once a fish was located, the following variables were measured; fish location (latitude and longitude) was determined with a handheld Global Positioning System; the anchor and depth finder were used to detect bottom substrate type; depth was determined with a boat-mounted Hummingbird depth finder; turbidity was measured with a secchi disk; both surface and bottom water temperatures, dissolved oxygen, conductivity, and pH were measured using a Hydrolab Quanta. The movements of the animals were determined based on individual locations. Average habitat use was calculated for depth, temperature, and dissolved oxygen. Habitat use was determined for individual fish as well as for all fish combined.

The results of this study showed that our sterilized hatchery reared animals occupied the same niches as their wild counterparts in the Apalachicola River, demonstrating their usefulness as sentinels. Six of the eight radio tagged Gulf sturgeon moved downstream through Jim Woodruff Lock and Dam into the Apalachicola River. Of the two remaining fish, one transmitter began giving a very weak signal and was lost while the other never changed position after the first week and is believed to either be a mortality or shed transmitter. The animals in both Lake Seminole and the Apalachicola River were located near substrates consisting of silt or sand. Silt and sand is the dominant substrate type available within Lake Seminole and the Apalachicola River. However, woody debris and rock outcrops do compose part of the substrate of these waterways. On one occasion, a Gulf sturgeon was found near a rock outcrop. However, no sturgeon was ever located in or near woody debris. Water temperature and dissolved oxygen where animals were located was similar to ambient reservoir and river conditions. Gulf sturgeon in this study did not seek cool water refuges or show any type of selection as related to temperature or dissolved oxygen. The average depth occupied by tagged Gulf sturgeon throughout the study period was 20.7 feet. The average depth where the fish were found in the Apalachicola River was 16.9 feet. The average depth occupied by the fish in Lake Seminole was 21.1 feet, excluding the times when fish were located in the lock chamber. Gulf sturgeon in Lake Seminole were found on most occasions in or near the deep water found along the Flint and Chattahoochee River channels. Only the fish whose signal faded before disappearing moved more than a few hundred meters outside of the channel. There were no obvious differences in depth selection between fish. Most differences appear to be related to fish in lentic versus lotic habitats.

While the animals released above the dam into Lake Seminole did not appear to locate good over summering habitat, the fact that they were able to pass through the Jim Woodruff Lock and Dam is a significant finding, which opens the door for further investigation, and proves the usefulness of sterilized sentinel animals.