**6. Sturgeon feeding tags**

## **6.1 Background**

Lake sturgeon movements in the field are readily identified using different tagging systems but establishing feeding behaviour is somewhat more complicated because one can not observe feeding directly as lake sturgeon generally do not feed at the surface. However, results reported in this chapter clearly revealed that lake sturgeon spend a significant proportion of time in the water column and were likely feeding on drift concentrated at the inlet and outlet of the lake, and emerging insects in the lake. Consequently a key question was could a sensor be developed to document lake sturgeon feeding? From previous studies on the histology of larval lake sturgeon we knew that there were extensive pressure receptors inside the mouth of lake sturgeon (Dick, unpubl. data). From other observations it was apparent that lake sturgeon utilized the branchial chamber to not only sense and feel the food but also to clean and to expel food with considerable force if the food was found to be unacceptable (Dick, unpubl. data). Furthermore, since lake sturgeons extend their mouth to feed we hypothesized that this may change the pressure inside the branchial chamber. We also knew that lake sturgeon extended the mouth with and without feeding.

Branchial pressure ranges from 50-150 pascals for restrained animals and no studies had attempted to relate branchial pressure to various levels of metabolic activity. We expect pressure to be correlated to oxygen consumption but our initial question was to determine if we could measure differences in the branchial chamber of lake sturgeon. Since lake sturgeon feed by sucking in prey and water this action should result in large pressure pulses interrupting rhythmic ventilation pressure pulses. It should be possible to distinguish mouth movements associated with feeding, coughing etc. The objective was to build a prototype tag to test the feasibility of a pressure tag to monitor branchial chamber pressure and use this as a measure of feeding activity. Previous reports by Webber et al. (2001a) and Webber et al. (2001b) describe the application of pressure tags to measure swimming speeds of fish.

#### **6.2 Methods**

Lake sturgeon used in this study were cultured at the University of Manitoba and subdued with tricaine methanol sulfonate (MS-222). The pressure sensor is a proprietary design with a cannula (PE 160) attached to the positive port, inserted under the tegument and into the parabranchial cavity under the opercular flap such that most of the cannula was not exposed to the environment. The tip of the cannula did not interfere with the movement of the gill filaments. The pressure sensors were powered by a standard bridge voltage (+10v), amplified and sampled at 69Hz. The pressure sensors were calibrated against a column of water of known density at the beginning and end of each experiment. Pressure signals were digitized by a MACLAB data acquisition system (AD Instruments Ltd.) and stored on disk. The resolution of the sensor was 1.85 pascals digital value-1 or 0.0189 cm freshwater at 4oC. The prototype sensor was designed to be attached by wires to the receiver to obtain physiological data. The second sensor was designed to transit the signal directly to a receiver. The experimental setup for the study is shown in Fig. 32

#### **6.3 Results**

The original experiments utilized direct wiring from the sensor and the data are represented by the Analog to Digital conversion (A/D) of the A/D board in the PC **(**Fig. 33).

Lake sturgeon movements in the field are readily identified using different tagging systems but establishing feeding behaviour is somewhat more complicated because one can not observe feeding directly as lake sturgeon generally do not feed at the surface. However, results reported in this chapter clearly revealed that lake sturgeon spend a significant proportion of time in the water column and were likely feeding on drift concentrated at the inlet and outlet of the lake, and emerging insects in the lake. Consequently a key question was could a sensor be developed to document lake sturgeon feeding? From previous studies on the histology of larval lake sturgeon we knew that there were extensive pressure receptors inside the mouth of lake sturgeon (Dick, unpubl. data). From other observations it was apparent that lake sturgeon utilized the branchial chamber to not only sense and feel the food but also to clean and to expel food with considerable force if the food was found to be unacceptable (Dick, unpubl. data). Furthermore, since lake sturgeons extend their mouth to feed we hypothesized that this may change the pressure inside the branchial chamber.

We also knew that lake sturgeon extended the mouth with and without feeding.

Branchial pressure ranges from 50-150 pascals for restrained animals and no studies had attempted to relate branchial pressure to various levels of metabolic activity. We expect pressure to be correlated to oxygen consumption but our initial question was to determine if we could measure differences in the branchial chamber of lake sturgeon. Since lake sturgeon feed by sucking in prey and water this action should result in large pressure pulses interrupting rhythmic ventilation pressure pulses. It should be possible to distinguish mouth movements associated with feeding, coughing etc. The objective was to build a prototype tag to test the feasibility of a pressure tag to monitor branchial chamber pressure and use this as a measure of feeding activity. Previous reports by Webber et al. (2001a) and Webber et al. (2001b) describe the application of pressure tags to measure swimming speeds

Lake sturgeon used in this study were cultured at the University of Manitoba and subdued with tricaine methanol sulfonate (MS-222). The pressure sensor is a proprietary design with a cannula (PE 160) attached to the positive port, inserted under the tegument and into the parabranchial cavity under the opercular flap such that most of the cannula was not exposed to the environment. The tip of the cannula did not interfere with the movement of the gill filaments. The pressure sensors were powered by a standard bridge voltage (+10v), amplified and sampled at 69Hz. The pressure sensors were calibrated against a column of water of known density at the beginning and end of each experiment. Pressure signals were digitized by a MACLAB data acquisition system (AD Instruments Ltd.) and stored on disk. The resolution of the sensor was 1.85 pascals digital value-1 or 0.0189 cm freshwater at 4oC. The prototype sensor was designed to be attached by wires to the receiver to obtain physiological data. The second sensor was designed to transit the signal directly to a

The original experiments utilized direct wiring from the sensor and the data are represented

by the Analog to Digital conversion (A/D) of the A/D board in the PC **(**Fig. 33).

receiver. The experimental setup for the study is shown in Fig. 32

**6. Sturgeon feeding tags** 

**6.1 Background** 

of fish.

**6.2 Methods** 

**6.3 Results** 

Fig. 32. Initial set up to collect data Fig. 33. Sensor on pectoral fin and cannula from sensor. inserted into the branchial chamber with cannula visible.

Fig. 34. Flushing cannula with syringe to remove air bubbles.

Fig. 35. Branchial pressure at 15°C. Fig. 36. Branchial pressure at 22°C. Note occasional negative values.

Movements and Habitat Use by Lake Sturgeon (*Acipenser fulvescens*)

Scatterplot of File (TDFF1355) TDFF1355

OpRate Q10=(R2/R1)10/t2-t1 = (115/70) 10/22-15 = 2.032

Op rate (L) Peak (L) Temp Integ Max-Min

Hour (h)

Scatterplot (TDFF1355) Ratint

=0.89

Rate=21.6 + (4.69(Int)) R2

Pressure Integral (A/D)

8 10 12 14 16 18 20 22

to integration. amplitude

through the cartilage at the base of the pectoral fin (Figs. 33 and 34).

amplitude due to temperature. voltage changes.

14.15 14.25 14.35 14.45 14.55 14.65 14.75

Opercular frequency (beat min-1), , Max-Min (A/D)

(Fig. 44).

RATE +1 STD ERR

Opercular Rate (beats min-1)

in an Unperturbed Environment: A Small Boreal Lake in the Canadian Shield 397

Scatterplot (10281356) N=16K TD1355b

Sampling rate=69.2 Hz Opercular pulse Opercular pulse

(Derivative) Raw/sec


Time

Scatterplot (TDFF1355) RATMAX

=0.90

Rate=19.8 + (0.953(Max-Min)) R2

Max-Min Pressure (A/D)

40 50 60 70 80 90 100 110

Pressure volts (L) Press/sec (R)

Temperature (oC), Integration (A/D)

Fig. 41. Increase in frequency and Fig. 42. High correlation between pressure and

It was decided to build a prototype tag to test the feasibility of a pressure tag to monitor branchial chamber pressure from 12 to 22oC. Frequency (blue circles) increased from 70 to 115 opercular beats-1 (Fig. 42). The calculated Q10 for frequency was 2.03, which describes the general response of most metabolic processes with temperature. The increase in amplitude and frequency was due a metabolic increase in routine metabolic rate. There was a high correlation between pressure in the branchial chamber and voltage changes (Fig. 41). When the TELEPLAY.EXE was used to integrate branchial pressure waveform as an AC neg-pos-neg waveform the integration (red squares) was highly correlated to temperature. Frequency of pulses was highly correlated to both integration (Fig. 43) and amplitude

Opercular Rate (beats min-1)

AVGRATE +1 STD ERR

Fig. 43. Frequency of pulses correlated Fig. 44. Frequency of pulses highly correlated to

The feeding pressure tag (Fig. 45) was tested under laboratory conditions (Fig. 46). A major challenge was determining how to stabilize the cannula and how to attach it to the lake sturgeon. Several methods to attach the tag were attempted, including drilling holes through the scutes and attaching to the dorsal surface of pectoral fin (Fig. 47) and attached to the pectoral fin (Fig. 48). Two methods were tested for placement of the cannula to monitor pressure, 1) attached to the tegument and under the opercle and 2) inserted

Raw data (volts)

> 13.996 13.997 13.998 13.999 14 14.001 14.002 14.003 14.004 14.005 14.006 14.007 14.008 14.009 14.01

Fig. 37. Ability to rapidly alter branchial Fig. 38. Direct observation of changing frequency. frequency due to stress.

Approximately 1 cm of water pressure is equivalent to 40-50 A/D. The method to attach the wiring to the body wall is illustrated in Fig. 33. Figure 33 illustrates the sensor attached to the fish with the opercle lifted to observe the end of the cannula inside the branchial chamber. Figure 34 illustrates the priming of the cannula and the removal of air bubbles. For the majority of the time, data from the opercular cavity had a regular pattern exhibiting consistent amplitude and frequency (Figs. 35 and 36). However, peaks varied in amplitude in both positive and negative directions. Peak amplitude was approximately 5.5 cm (230 AD) at 15oC and increased to 9.5 cm (400 AD at 22oC) and the period ranged from 150 sec at 15oC to 40 sec at 22oC (Fig. 37). The peak amplitude and frequency increased with temperature (Fig. 35 at 15oC and Fig. 36 at 20oC). Figures 36 and 37 illustrate how quickly an individual can alter the opercular frequency in response to activity, metabolism and stress. Figure 38 demonstrates the changing opercular frequency of lake sturgeon as a result of stress. Figures 39 and 40 illustrate that immediately after a large pressure pulse (feeding peak) the regular breathing movements were larger than the preceding ones. Regular pulses increased in frequency and amplitude in response to temperature. Amplitude (green diamond) increased from 2 cm (45 AD) to 2.5 cm (110 AD) (Fig. 42).

Data Amplitude (a/d)

Sturgeon#1 plot (10281456) N=16K TD1456b TDick/DWebber Oct 28/2000 Time (h) Pressure (digital value) -300 -200 -100 0 100 200 300 14.967 14.96727 14.96755 14.96783 14.9681 14.9684 14.96866 14.96894 14.9692 14.9695 14.96977 T. Dick, D. Webber Sturgeon Branchial Pressure Oct. 28/2000 69.2 hz 1 sec Operculum Closed -negative pressure Mouth open - positive pressure

are higher immediately after feeding. change to normal pulse.

Fig. 39. Regular breathing movements Fig. 40. Feeding pulse is followed by rapid

Data Amplitude (a/d)

Approximately 1 cm of water pressure is equivalent to 40-50 A/D. The method to attach the wiring to the body wall is illustrated in Fig. 33. Figure 33 illustrates the sensor attached to the fish with the opercle lifted to observe the end of the cannula inside the branchial chamber. Figure 34 illustrates the priming of the cannula and the removal of air bubbles. For the majority of the time, data from the opercular cavity had a regular pattern exhibiting consistent amplitude and frequency (Figs. 35 and 36). However, peaks varied in amplitude in both positive and negative directions. Peak amplitude was approximately 5.5 cm (230 AD) at 15oC and increased to 9.5 cm (400 AD at 22oC) and the period ranged from 150 sec at 15oC to 40 sec at 22oC (Fig. 37). The peak amplitude and frequency increased with temperature (Fig. 35 at 15oC and Fig. 36 at 20oC). Figures 36 and 37 illustrate how quickly an individual can alter the opercular frequency in response to activity, metabolism and stress. Figure 38 demonstrates the changing opercular frequency of lake sturgeon as a result of stress. Figures 39 and 40 illustrate that immediately after a large pressure pulse (feeding peak) the regular breathing movements were larger than the preceding ones. Regular pulses increased in frequency and amplitude in response to temperature. Amplitude (green diamond) increased from 2 cm (45 AD) to 2.5 cm (110 AD)

Fig. 39. Regular breathing movements Fig. 40. Feeding pulse is followed by rapid

are higher immediately after feeding. change to normal pulse.

Pressure (digital value) -300 -200 -100 0 100 200 300

> 14.967 14.96727 14.96755 14.96783 14.9681 14.9684 14.96866 14.96894 14.9692 14.9695 14.96977

Operculum Closed -negative pressure

Mouth open - positive pressure

Sturgeon#1 plot (10281456) N=16K TD1456b TDick/DWebber Oct 28/2000

Oct. 28/2000 69.2 hz

T. Dick, D. Webber Sturgeon Branchial Pressure

1 sec

Time (h)

Temperature (oC)

Fig. 37. Ability to rapidly alter branchial Fig. 38. Direct observation of changing

frequency. frequency due to stress.

Scatterplot (PEAK) Graph TDPEAKa Oct 28/2000 TDick/DWebber

Time (h)

13 13.4 13.8 14.2 14.6 15

Mouth extention period (sec)

Peak Amplitude (A/D)

Period (L) Temperature (R) Data (R) Peak (L)

(Fig. 42).

Fig. 41. Increase in frequency and Fig. 42. High correlation between pressure and

It was decided to build a prototype tag to test the feasibility of a pressure tag to monitor branchial chamber pressure from 12 to 22oC. Frequency (blue circles) increased from 70 to 115 opercular beats-1 (Fig. 42). The calculated Q10 for frequency was 2.03, which describes the general response of most metabolic processes with temperature. The increase in amplitude and frequency was due a metabolic increase in routine metabolic rate. There was a high correlation between pressure in the branchial chamber and voltage changes (Fig. 41). When the TELEPLAY.EXE was used to integrate branchial pressure waveform as an AC neg-pos-neg waveform the integration (red squares) was highly correlated to temperature. Frequency of pulses was highly correlated to both integration (Fig. 43) and amplitude (Fig. 44).

to integration. amplitude

Fig. 43. Frequency of pulses correlated Fig. 44. Frequency of pulses highly correlated to

The feeding pressure tag (Fig. 45) was tested under laboratory conditions (Fig. 46). A major challenge was determining how to stabilize the cannula and how to attach it to the lake sturgeon. Several methods to attach the tag were attempted, including drilling holes through the scutes and attaching to the dorsal surface of pectoral fin (Fig. 47) and attached to the pectoral fin (Fig. 48). Two methods were tested for placement of the cannula to monitor pressure, 1) attached to the tegument and under the opercle and 2) inserted through the cartilage at the base of the pectoral fin (Figs. 33 and 34).

Movements and Habitat Use by Lake Sturgeon (*Acipenser fulvescens*)

increased with temperature.

tightening strap.

**7. Summary** 

in an Unperturbed Environment: A Small Boreal Lake in the Canadian Shield 399

Positive pulses were always associated with opening of the mouth and negative pressure with closing of the mouth. Amplitude of these rhythmic pulses generally ranged from 50 to 100 pascals for all lake sturgeon. We also observed that all fish periodically made rapid mouth movements that resulted in considerably larger pressure pulses (800 pascals) compared to the rhythmic ventilations pulses described previously. These pulses were caused by the sudden projection of the jaw approximately 3-4 cm outward form the mouth. Pressure amplitude was often an order of magnitude greater when compared to ventilation pulse pressure. This is interpreted as instances of feeding or feeding attempts. Temperature influences all variables as integral and max-min pressure and frequency of ventilation on pulses increased with temperature. As well, amplitude and period of feeding pulses

Although we did not measure MO2 (oxygen consumption) directly the data on integration of branchial pressure and an AC waveform indicates that integration and amplitude can be used to predict MO2 (energy budgets) sturgeon in nature. This information could be

We have developed a specialized feeding tag for lake sturgeon that functions under laboratory conditions. Inserting the cannula through the cartilage above the pectoral fins had a minimal affect on the fish; however, we have yet to find a satisfactory method to hold the tag securely to the fish for more than 12 days. Internal placement of the tag is not an option as the wires would then have to come from the tag through the body wall to the sensor. The prototype tag weighed 46 gm in air and the next stage of development will be to reduce the weight of the tag size considerably (we are already using depth tags that have a much lower weight than the V16s). Even the prototype tag can be attached to large lake sturgeon (over 25 kg) and the preferred attachment site will likely be the pectoral fin. The next tags will have to weigh less than 15 gm in air, be more streamed lined to reduce resistance and mode of attaching to the boney fins rays will need to accommodate self

Lake sturgeon (*Acipenser fulvescens*) in Canada in the early 1900s were reduced to remnant populations over most of their historic range and extirpated from much of the Great Lakes and Lake Winnipeg. Populations continued to decline over the next 100 years due to commercial fishing pressure, hydroelectric and other industrial developments. This led in the early 2000s to the Committee on the Status of Endangered Wildlife in Canada recommending that lake sturgeon be listed as threatened or endangered in various regions of Canada. Most of the current research on lake sturgeon is related to environmental assessment for hydroelectric developments from perturbed areas where populations are low. The purpose of this research was to study a lake sturgeon population in an unperturbed system, the Pigeon River at Round Lake on the west side of Lake Winnipeg, Manitoba, Canada. Round Lake is a small isolated lake with a typical fish community found in the boreal region of Canada. The size of the sturgeon population relative to other fish species in the lake was determined by randomly set standard gang gillnets and all sturgeon caught were tagged with external and PIT tags and returned to the wild. Lake sturgeon comprised about 10% of the total population of fish. The main food item of lake sturgeon was mayflies and a detailed stomach analyses indicates that mayflies are important food for several other fish species. Since we were interested in determining how lake sturgeon, from

combined with temperature and feeding data to predict seasonal growth rates, etc.

Fig. 45. Tag attached to pectoral fin. Fig. 46. Collection of data from tag in tank.

Fig. 47. Tag attached to dorsal scutes. Fig. 48. Tag attached to right pectoral fin and connected to sensor situated on the left pectoral fin.

The feeding sensor pressure tag gave identical results to the data collected from the prototype experimental data. The major problem was attachment of the tag as the longest time for attachment was 12 days. The best location was on the surface of the pectoral fin and surprisingly there was little influence on normal use of the fin by lake sturgeon in a tank. Attaching the tag to the scutes was the least effective as the sharp boney scutes severed both wire and heavy fishing line with ease. Once the tag was not firmly attached the cannula was dislocated and either became clogged with mucous or was outside the branchial chamber and was unable to measure any pressure changes. There was no evidence of infection when the cannula was inserted through the tegument and the cartilage and once the cannula was removed there was no infection. The point at which the cannula was inserted was undetectable within 2 weeks of its removal.
