.:..:. :1. , . .LV .. H :J:... . , . V i. ... . 5..., .l.3f.....,....!:l : .1: ._ . 2.. :. 1:13.. 3.. n . . ..,.. . . . .51.."- I a.? gal, A”‘{ 1.; ma?" int? 451.4 .1 l 3‘» m“ 7” Gide 1321‘ “a: silly This is to certify that the thesis entitled Migratory Behavior and Angling of Steelhead in the Pere Marquette River, Michigan presented by Aaron Jon Snell has been accepted towards fulfillment of the requirements for A II Major professor Date December 144 2001 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MIGRATORY BEHAVIOR AND AN GLING OF STEELHEAD IN THE PERE MARQUETTE RIVER, MICHIGAN. By Aaron Jon Snell A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 2001 A B S T ILA C "1‘ MIGRATORY BEHAVIOR AND ANGLING OF STEELHEAD IN THE PERE MARQUETTE RIVER, MICHIGAN. By Aaron J. Snell Radio—telemetry studies indicate that the pumped-source fish passage built around a pulsed-DC electrical sea lamprey barrier on the Pere Marquette River, MI was effective at passing at least 67% of tagged adult steelhead. Steelhead were captured using trap nets and hook-and-line fishing in 2000 and 2001, and their movements were monitored as they approached and ascended the fish ladder. Of the fish tagged at the mouth of the river, 6 "/0 were successful at traveling upstream to the barrier. These fish moved upstream faster in 2001 than in 2000, taking a mean time of 76.8 hours to reach the barrier. Once arriving at the barrier, fish were present for a mean time of 184.4 hours before ascending the ladder. An angler survey was conducted on the Pere Marquette River, MI in order to characterize the spring recreational fishery. The study area consisted of 17 sampling sites located from Baldwin downstream to Ludington on the mainstem of the river. Anglers fishing the Pere Marquette from public access sites during the spring months of 2000 and 2001 most often targeted steelhead, and effort totaled an average of 110,000 hours per year. This is a dramatic increase in effort since the time of the last angler survey, nearly twenty years ago. Anglers caught an average of 19.000 steelhead and 24,000 suckers per spring. Most anglers fishing for suckers were from local counties, while those pursuing steelhead typically traveled greater distances. “We must be the change we wish to see in the world” -Mahatma Gandhi iii ACKNOWLEDGEMENTS First and foremost, I must thank the Michigan Department of Natural Resources- Fisheries Division and the Great Lakes Fishery Commission for the financial support given to this project, and to the Michigan Agricultural Experiment Station and to Ben Gabbard for providing field housing. The Huron-Manistee National Forest provided special use permits for my equipment, and I thank Diane Walker for her assistance. I would also like to make a special note of thanks to Ellie Koon, U. S. Fish and Wildlife Service. Her hard work and patience kept the electrical barrier and, therefore, this project running. I would like to thank Doug Workman for his time, “training”, and most importantly, his fi'iendship. Dan Weisenreder, Jesse Gabbard, Jason Jacks, Mark Mutton, Nick Popoff, and Bill Wellenkamp provided the endless hours of fieldwork necessary to complete this project. In addition to my field crews, many graduate students made the long journey to work (capture steelhead by hook and line) on weekends. Particularly, Mike Wilberg and Brad Thompson, whose help in the field and in the classroom are greatly appreciated. Thanks must be given to Ron Soberalski for his time and effort. His willingness to help was unmatched, and his perspective of local history was most interesting. Also, I thank the “old guys” that stopped at my net site daily to heckle me. I admire their concern for the river and its fish. Thanks to Dick Schwikert and Chuck Smith, of the Pere Marquette Watershed Council, for their time and resources. iv A great deal of thanks must also be given to my advisor, Dr. Tom Coon, and committee members Dr. Dan Hayes and Dr. Rich Merritt. Their guidance and time were invaluable. Last and most importantly, to my wife Karen, whose patience and understanding allowed me to earn this degree. “I love you with all that I am. . TABLE OF CONTENTS LIST OF TABLES ................................................................................. vii LIST OF FIGURES ................................................................................ viii INTRODUCTION .................................................................................... 1 CHAPTER 1. A DESCRIPTION OF THE MIGRATORY BEHAVIOR OF STEELHEAD (Oncorhynchus mykiss) IN THE PERE MARQUETTE RIVER, MICHIGAN ........................................................................................... 7 INTRODUCTION ........................................................................... 9 STUDY SITE ............................................................................... 1 1 METHODS .................................................................................. 15 RESULTS ................................................................................... 21 DISCUSSION .............................................................................. 35 CHAPTER 2. AN AN GLER SURVEY ON THE PERE MARQUETTE RIVER, MICHIGAN .......................................................................................... 40 INTRODUCTION ......................................................................... 42 STUDY SITE ............................................................................... 44 METHODS ................................................................................. 49 RESULTS ................................................................................... 54 DISCUSSION .............................................................................. 67 LITERATURE CITED ............................................................................. 70 APPENDIX A: Vehicle count sheet .............................................................. 74 APPENDIX B: Angler interview form ............................................................ 76 vi LIST OF TABLES Table 1. Sampled public access sites on the mainstem of the Pere Marquette River from Baldwin downstream to Ludington (U SF S 1996) ...................................... 47 2. Total number of vehicles counted at each Pere Marquette River access site during the 2000 and 2001 sampling seasons .................................................... 55 3. Catch per effort (fish/hour) estimates for anglers fishing on the Pere Marquette River, MI, Spring 2000 and 2001 (:2 standard errors).................................62 4. Percentage of anglers fishing the Pere Marquette River in 2000 and 2001, based on vehicle counts, by state or province of residence .................................. 63 5. Percentage of Michigan anglers interviewed on the Pere Marquette River in 2000 and 2001 by county of residence .................................................. 65 vii Figure 1. 2. 10. ll. 12. 13. 14. 15. LIST OF FIGURES Location ofthe Pere Marquette River in Michigan... 12 Location ofthe study site on the Pere Marquette River, MI... .....14 Antenna arrangement at the electrical barrier site in Custer, MI... ......17 Daily catch of steelhead in fyke nets in Ludington, and daily arrival of tagged steelhead and water temperature at the Custer station, Spring, 2000. .. ...22 Daily catch of steelhead in fyke nets in Ludington, and daily arrival of tagged steelhead and water temperature at the Custer station,.Spring, 2001... ...23 Water temperature at the Custer station, Spring 2000 and 2001 ..................... 24 Discharge at the U. S Geological Survey gauge station in Scottville, MI, Spring 2000 and2001.. . 25 Number of tagged (x1 00) and untagged steelhead ascending the fish ladder in Custer, MI, Spring 2001.. .. 29 Number of tagged steelhead passing through the fish ladder and water temperature at the Custer station, Spring 2001... .30 Number of tagged steelhead arriving at Custer and time of arrival, Spring, Number of tagged steelhead passing the barrier at Custer before (barrier off) and during (barrier on) weir operation, Spring 200133 Location ofthe Pere Marquette River in Michigan... ...45 Location of the seventeen sampling sites on the Pere Marquette River, MI used forthe angler survey...... .....46 Estimated total effort at each study site on the Pere Marquette River, MI, Spring 2000 and 2001...... 56 Estimated targeted effort at each sampling site on the Pere Marquette River, MI, viii 16. 17. Estimated targeted effort at each sampling site on the Pere Marquette River, MI, Spring 2001 ................................................................................. 60 Estimated total number of fish caught on the Pere Marquette River, MI, during the Spring 2000 and 2001 sampling seasons ................................................ 61 ix INTRODUCTION The Pere Marquette River is one of the last remaining flee-flowing coastal streams in Michigan. It has been designated a Michigan “Natural River” and a National “Wild and Scenic River”, and is highly regarded by anglers for its resident and anadromous fish populations. Much of the angling pressure occurs in the spring when many native and introduced species from the Great Lakes move into the river for spawning (Kruger 1985). The most sought after species in the spring fishery is the steelhead (Oncorhynchus mykiss), while the longnose sucker (Catostomus catastomus), white sucker (Catostomus commersoni), and several species of redhorse (Moxostoma spp.) also support important fisheries. The sea lamprey (Petromyzon marinus) also enters the river to spawn in the spring. The invasion of the sea lamprey into the Great Lakes has resulted in one of the largest and most intensive efforts to control a vertebrate predator ever attempted (Smith and Tibbles 1980). For the past 50 years, a number of different methods have been used in an attempt to control sea lamprey. One such method, a migration barrier, is intended to prevent passage of adult sea lamprey to upstream spawning habitat. However, migration barriers may limit reproduction by other potamodromous fish species, banning fisheries dependent on spawning migrations. Sea lamprey were first discovered in the Great Lakes in the late 1930’s, and by the mid-1940’s they had contributed to a significant decrease in lake trout (Salvelinus namaycush) populations (Smith and Tibbles 1980). The decimated fishery prompted the development of a sea lamprey control program, and in 1950, mechanical weirs and traps were installed in tributaries to prevent sea lamprey from spawning. Electrical barriers soon followed; however, control measures did not become effective until the development of the chemical TF M (3 -trifluoromethyl-4nitrophenol). This toxicant is applied to streams to destroy ammocetes (larval lampreys). In 1959, TF M was endorsed by the Great Lakes Fishery Commission as the accepted method of control (Dahl and McDonald 1980). Presently, the sea lamprey control program depends largely on TFM as a control measure. However, the high cost of TFM treatment and public sentiment against the use of a chemical control measure has prompted a shift toward alternative control methods. Electric field barriers were first used in 1952. They could be built and operated at less cost and with greater efficiency than mechanical traps, and by 1960, 132 electric baniers had been installed on Great Lakes tributaries (Smith and Tibbles 1980). The barriers used 110-volt, 60 cycle alternating current (AC) to energize an electrode array fixed in the stream. No provisions were made in the barrier to allow other fish species to pass, so traps were installed at the downstream end of the electric field to capture migrating fish, which could then be transported upstream. The traps, however, caused interruptions in the electric field, allowing sea lamprey to pass through. There was also excessive mortality by electrocution of desirable species migrating upstream. Modifications to the barriers included the addition of pulsated direct current (DC) leads to divert fish into traps. This change greatly reduced, but did not eliminate, mortality of the desirable fish (Hunn and Youngs 1980). A truly complete and efficient electrical barrier was not realized during this period of sea lamprey control. It is believed that under ideal circumstances the electrical banier could stop the migration of sea lamprey; however, most barriers were not operated long enough to be accurately evaluated (McLain et a1. 1965). With the adoption of TFM as a control measure, the barriers were only used as a means of assessing the number of spawning lamprey, and by 1966, they were removed from all Lake Michigan tributaries (Smith and Tibbles 1980). Although TFM has been used successfully since it was developed, the rising cost of the chemical has led to a renewed interest in the use of electric barriers. Rivers such as the Pere Marquette, which have large amounts of habitat that must be treated with TFM, are at the top of the list for alternative control methods. In October 1999, an electric lamprey barrier with a fish passage structure was completed on the Pere Marquette River at Custer, MI. Previous attempts at operating an electric weir at this site were unsuccessful. The first weir was operated in 1964 and 1965 and was effective at blocking the lamprey migration, but was hannful to other fish. A new weir was installed and operated in the spring of 1989. Again, the weir was effective at blocking sea lamprey, no mortality of steelhead adults or smolts was observed, and northern pike and suckers were able to pass downstream through the barrier. However, the barrier did not allow any species of fish to pass upstream (Rozich 1989, unpublished). The current barrier began operation on March 21, 2000. Several modifications were made in order to allow upstream passage of steelhead. The fish passage consists of a channel with a series of jumps navigable to steelhead but impassable to non—jumping fish, including sea lamprey. River water is pumped through the channel at a velocity high enough to create an attraction flow, allowing steelhead to swim through the passage and around the electric grid. In addition to the passage, the banier has also been changed fi'om AC to a pulsed DC gradient electric grid, which is generally accepted to be much less damaging to fish (Reynolds 1996). The new system uses controllable electronics and updated software to provide a consistent and uniform electrical field oriented such that a fish moving parallel to the stream flow will always encounter the maximum voltage gradient. Management considerations Steelhead were first introduced to the Pere Marquette River in 1885, and were naturalized shortly thereafter. Steelhead in Lake Michigan tributaries typically spend two years in the river after hatching, and then migrate to the lake in early spring, where they spend another 1 to 3 years before retunring to the river to spawn. The spawning period normally lasts from 4 to 7.5 weeks and spawning takes place in areas dominated by large gravel and cobble substrate (Biette et a1. 1981). Steelhead have been most successful in rivers that provide unobstructed passage fi'om the lake to spawning habitat, and that have water temperatures that are conducive to juvenile survival (e. g., <20°C). The operation of the electric barrier to the Pere Marquette could obstruct passage, thereby limiting the reproductive capabilities of adult steelhead. Little is known about the behavioral response of adult steelhead to a migration barrier, or the energetic costs associated with the delay that may occur while searching for a passage. Palmisano and Burger (1988) used an electric barrier to guide Chinook salmon (0. tshawytscha) adults into a side channel where a weir and trap were installed to capture the fish. The salmon were able to locate the alternate channel but the time that they spent searching for the passage was not recorded. Bjornn et a1. (1998) used radio telemetry to study the migration of Chinook salmon past dams and through reservoirs in the Lower Columbia and Snake River system. They found that 90-94% of the salmon were able to pass over the dams. The median times for salmon to pass the dams ranged from 0.25-0.86 days for the dams without fish traps, and up to 5.4 days for dams that operated fish traps. With an increase in the time it took for a salmon to pass through the lower Snake River, came a subsequent decrease in the likelihood of the salmon completing its spawning migration. Andrew and Geen (1960) discussed how delays below fish passages could adversely affect, or even totally prevent spawning success of sockeye salmon (0. nerka) in the Fraser River. Many of the habitat characteristics of the Pere Marquette River that allow for successfirl reproduction of steelhead are also favorable to the sea lamprey. Ironically, habitat enhancement programs such as streambank stabilization and sand removal, which have been intended to increase salrnonid production, also benefit the sea lamprey (Smith and Tibbles 1980). While sea lamprey begin their upstream. migration as early as April, they do not begin nest construction until the water warms to about 15° C (Manion and Hanson 1980). For successful spawning, sea lamprey require a steady unidirectional flow of water, sand and gravel substrates, and suitable water temperatures (6. g. 10°C to 261°C) (Applegate 1950). Blocking sea lamprey from this spawning habitat can effectively control their population (McLain et al. 1965). In addition to any biological concerns attributed to the banier, it is also important to realize that the operation of the barrier could have an effect on the upstream fishery. The history of trout and sahnon in the Pere Marquette has been well documented (Kruger 1985), but no recent work has been done to quantify the fishery. Species other than steelhead are also important. Suckers, for example, are exploited as they make their spring spawning migration, and although little is known about their importance in the recreational fishery, provisions for their passage were not incorporated in the initial design of the barrier and fishway. During the months of March and April, anglers can be found along the banks of the Pere Marquette attempting to catch the migrating suckers. Upstream of the barrier, in Walhalla, M1, the annual sucker festival attracts large crowds to celebrate the sucker fishery. A better understanding of the spring fishery on the Pere Marquette is critical to the operation of the barrier, because to the anglers, the success of the banier will most likely be measured in how it affects their angling experience. This two-year study was designed to determine the effects of an electrical sea lamprey barrier on the migratory behavior of adult steelhead into the Pere Marquette River, and to characterize the spring recreational fishery. The objectives were to: 1) describe the timing of migration initiation by adult steelhead and the distribution of dates of anival of radio-tagged fish at the Custer weir site, 2) determine the speed of upstream movement by adult steelhead from the old US 31 bridge at Ludington to Custer, through the weir, and from Custer to Bowman Bridge, 3) to determine the percent of steelhead passing the electrical barrier at Custer, and 4) to characterize the spring fishery in terms of total effort, targeted effort, and catch per effort. CHAPTER 1: A DESCRIPTION OF THE MIGRATORY BEHAVIOR OF STEELHEAD (Oncorhynchus mykiss) IN THE PERE MARQUETTE RIVER, MICHIGAN. ABSTRACT Radio-telemetry studies indicate that the pumped-source fish passage built around a pulsed-DC electrical sea lamprey barrier on the Pere Marquette River, MI was effective at passing at least 67% of tagged adult steelhead. I captured steelhead using trap nets and hook-and-line fishing in 2000 and 2001. I tagged a total of 113 fish by either surgical or gastric implantation, and monitored their movements as they approached and ascended the fish ladder. I captured most (87) at the mouth of the river, approximately 20 km downstream of the banier, and captured additional fish (26) directly below the barrier. Of the fish tagged at the mouth of the river, 68% were successful at traveling upstream to the barrier. These fish moved upstream faster in 2001 than in 2000, taking a mean time of 76.8 hours to reach the banier. Once arriving at the banier, fish were present for a mean time of 184.4 hours before ascending the ladder. Fish that ascended the ladder were twice as active in seeking passage as those that did not ascend the ladder. Successful fish moved from one antenna to another a mean of 7.1 times per hour. All fish that entered the ladder eventually ascended, and navigated the channel quickly (mean 0.58 hours). Most passage through the ladder was during daylight hours for tagged and untagged fish. INTRODUCTION The Pere Marquette River is one of the last free flowing coastal streams in Michigan. It is a state “Natural River” and a national “Wild and Scenic River”, and is highly regarded by anglers for its resident and anadromous fish populations. The river supports a naturally-reproducing run of steelhead (Oncorhynchus mykiss), and like many Great Lakes tributaries, has been colonized by sea lamprey (Petromyzon marinas). As part of the effort to control sea lamprey, an electrical barrier was installed on the Pere Marquette in 1999 to prevent their passage to upstream spawning habitat. A pumped- source fish passage built around the barrier is designed to facilitate the upstream migration of steelhead. Electrical barriers have been used successfirlly for completely blocking spawning migrations of sea lamprey (Swink 1999), as well as guiding adult Chinook sahnon (0. tshawytscha) into a channel where they could be trapped (Palmisano and Burger 1988). However, little is known about the behavior of steelhead as they encounter an electrical barrier, or the efficiency of this unique design of fish passage. More common fish passage scenarios, such as hydroelectric dams with pool and fall fish ladders, have been studied extensively in recent years, and passage rates in excess of 90% are common (Bjorn et a1. 2000, Gowans et a1. 1999). This two-year study was designed to determine the effects of an electrical sea lamprey barrier on the migratory behavior of steelhead in the Pere Marquette River, MI. In order to achieve this, I met the following objectives: 1) describe the timing of migration, 2) determine the proportion of fish that succeed in bypassing the weir, 3) measure the rate of ascent from the mouth of the Pere Marquette River to the electrical sea lamprey barrier, 4) measure the amount of time required by fish to bypass the electrical barrier, and 5) to measure the time required by fish to reach spawning habitat after passing the barrier. 10 STUDY SITE The Pere Marquette River is located along the eastem-shoreline of Lake Michigan in west central Michigan (Figure 1). The watershed encompasses an area of 1956 km2 and contains 612 lineal kilometers of streams. From the headwaters near Chase to Pere Marquette Lake in Ludington, the Pere Marquette River courses a distance of about 154 km (MDNR/IFR 1998). The historic mean discharge in the spring (February - April) is 914 cfs (USGS 2001). In 2000, the mean discharge in spring was 671 cfs, and in 2001 it was 1040 cfs. The electrical sea lamprey barrier is located at the Custer Road Bridge, about 2 km south of Custer, MI, and approximately 20 km upstream of the river mouth. There is no documented spawning habitat available to sea lamprey downstream of the banier (Ellie Koon, USFWS, personal communication). A pulsed-DC electrical banier (Smith- Root, Inc., Vancouver, Washington) was first used at the site in 1989. It was successful at blocking sea lamprey but did not allow the upstream passage of non-target species (Tom Rozich, MDNR, personal communication). The current barrier was completed in 1999, and has been modified with state-of-the-art pulsators and electronics, as well as a pumped-source pool-and—weir fish ladder. The fish ladder is constructed with walls of corrugated steel sheet piling, and is 1.2 m wide and about 19.5 m in length. Removable weirs separate 3 pools. Because the electrical barrier does not create an impoundment, water must be pumped into the fish ladder. One pump delivers water to a false weir at the top of the channel at 12 cfs, creating a 20-30 cm head over the removable weirs, and a larger pump is used to create an attracting flow (25 cfs) closer to the entrance of the 11 Figure 1. Location of the Pere Marquette River in Michigan. 12 ladder. Upon entering the passage, fish must navigate the series of four weirs before being returned to the river upstream of the barrier via a 30.5 cm PVC tube. In 2001, a net was placed over the exit end of the tube to allow observers to count and examine passing fish. The barrier was in operation from 21 March-30 June 2000 and 5 April-30 June 2001. The normal operating procedure for the electrical banier is to run continuously for the duration of the sea lamprey spawning migration. However, this protocol was not followed during part of the 2000 field season. A damaged pump prevented the circulation of water through the fish ladder from 31 March to 8 April, so the electrical field was tumed off between the hours of 10:00 and 16:00 to allow fish passage. Several radio-tagged fish passed over the electrical banier during this time, thereby compromising the design of the study. Additionally, at the conclusion of the 2000 field season, stray voltage from the banier was discovered near the entrance to the fish ladder. It is believed that this condition may have existed throughout the period of operation and was corrected for the 2001 field season. The area of study (Figure 2) extended from the mouth of the Pere Marquette River upstream to Bowman Bridge (48 km). Much of the river within this area has a stream gradient of less than 1 meter per kilometer and is characterized by slow runs and deep pools. Generally, streams best suited for trout reproduction have a gradient ranging from 1.9-13 meters per kilometer (Hay-Chmielewski et a1. 1985). In the Pere Marquette, much of this higher quality spawning habitat occurs near and upstream of Bowman Bridge. Therefore, it is likely that fish that were successful in passing through the fish ladder would continue their migration upstream past Bowman Bridge. 13 Cusmr //////Ludmgton Bowman Bfldge Figure 2. Location of the study site on the Pere Marquette River. MI. METHODS I captured steelhead in trap nets near the mouth of the Pere Marquette River, and by hook and line fishing near the mouth, as well as directly below the barrier. I fished the nets continuously for approximately 25 days each year, and checked them daily for the presence of steelhead. In order to minimize mortality, I only selected those fish measuring 50 cm or more in length, migrating upstream, and appearing in good physical condition for the study. I implanted digitally encoded radio transmitters (Lotek Engineering, Inc., Model MCFT-7A; 16 x 83 mm, 12.8g in water) into 113 steelhead (56 in 2000 and 57 in 2001). I anesthetized the fish in a 15-gallon tank of tricaine methane sulfonate (MS-222), outfitted them with a transmitter and an external T-type tag (Floy Tag Co.), and transferred them to a recovery tub prior to releasing the fish at the point of capture. I released all fish caught in the trap nets approximately 200 m upstream of the nets in order to minimize recapture. I released fish caught by angling at the location of their capture. In 2000, I surgically implanted radio transmitters using methods similar to Hart and Summerfelt (1975). The entire procedure, from induction of anesthesia to release took approximately 15 minutes to complete. In response to concerns that the surgical procedure might alter behavior, I used an alternative implant method, gastric insertion, in 2001. In 2001, I surgically implanted transmitters into the peritoneal cavity of 24 fish, and used gastric insertion of transmitters for 33 fish. I executed gastric implants by inserting the transmitter through the esophagus and into the stomach of an anesthetized 15 fish. The antenna extended out the fish’s mouth, and was bent to trail along the fish’s side (Ted Bjomn, University of Idaho, personal communication). I monitored stream temperature and stream discharge for the duration of both sampling seasons. I obtained stream discharge data from the U. S. Geological Survey for the gaging station at Scottville, and monitored stream temperature at the barrier site in Custer and at the net site in Ludington by means of Onset Hobo electronic thennographs. Monitoring movement I monitored movements of the radio-tagged steelhead from the time of tag implantation until the end of the upstream spawning run. I used Lotek SRX-4OO receivers at two fixed (base) stations, and one portable station to monitor fish movement. I used a portable receiver equipped with a boat-mounted Yagi antenna (Lindsay Inc., Lindsay, Ontario) to locate radio-tagged fish between Custer and Pere Marquette Lake, and to ensure that the radio transmitters were functioning prior to implantation. The primary base station was located at the barrier site in Custer. I equipped the site with a recording receiver, and a Digital Spectrum Processor (DSP-SOO) with a fast antenna switching option. The DSP-500 is a digital coprocessor that provides frequency discrimination using real-time analysis. This allowed for simultaneous scanning of multiple antennas and frequencies. The receiver logged all data into memory, which I downloaded daily. I used seven antennas at this station in order to detect the movement of fish near the barrier (Figure 3). I mounted one aerial Yagi antenna approximately 70 m downstream of the barrier (A1), and another aerial Yagi 30 m upstream (A7) of the barrier. I used these two antennas to detect the direction of travel, along with the times of anival and departure in the reception area. I placed three submerged antennas (A2, A3, 16 >8 Flow A7 6 o Al Figure 3. Antenna arrangement at the electrical barrier site in Custer, MI. 17 and A4) across the face of the banier, another at the mouth of the fish passage (A5), and one within the ladder (A6). The submerged antennas allowed me to monitor fish as they encountered and attempted to negotiate the barrier and the fish passage structure. I constructed submerged antennas using RG-58 coaxial cable, and lefi approximately 20 cm of center wire exposed at the end. I calibrated the SRX/DSP unit to determine the reception area of the seven antennas. I positioned the aerial antennas at a 45° angle to the riverbank (directed away from the barrier) to achieve greater range, and to avoid possible overlap with underwater antennas. Areas of overlap in the underwater reception cells ensured that a fish could not swim between antennas without being detected (Figure 3). The second base station was located about 2 km upstream of Bowman’s Bridge. I equipped this station with a recording receiver (SRX-400) housed in an environmental chamber. I used an LA51 solar panel with a 12-volt deep-cycle battery and photovoltaic charge controller (Kyocera, Arvada, CO) to supply power to this remote site. I mounted two aerial Yagi antennas approximately 80 m apart to detect direction of movement, and the date and time of arrival and departure of tagged fish. I obtained movement data of untagged fish from the United States Fish and Wildlife Service, which was in charge of operating the banier. This data included the number of steelhead ascending the ladder on each date of barrier operation, as well as time of passage. Data analysis I divided the river into three sections for data analysis. Section A was the stretch of river from Pere Marquette Lake upstream to the detection limit of A1 at the electrical barrier (20 km). I did not include the fish tagged in the vicinity of the barrier in the 18 analysis of section A. I determined the timing of migration and arrival at Custer by using the dates and times of capture and first detection at the barrier, respectively. The detection zone of the telemetry equipment located at the barrier served as Section B. I calculated the time difference between first detection at the barrier and last detection on the upstream antenna to determine time of passage. Conversely, I calculated downstream time of passage as the fish moved downstream through the array. I included the fish tagged at the barrier in downstream time of passage as well as percent passage, but excluded them from upstream time of passage. Section C extended from the upstream detection limit of A7 to Bowman Bridge, which covers a river distance of approximately 28 km. I included all radio-tagged fish that ascended the fish ladder and continued their migration upstream to Bowman Bridge in the analysis of this section. I considered a fish to have passed the barrier when it moved upstream of antenna seven, was captured upstream, recorded on the Bowman Bridge receiver or, in the case of radio failure, when the banier attendant observed the fish in the capture net on the upstream tube. I determined the activity levels of tagged fish in the vicinity of the barrier by dividing the number of times a fish switched from one antenna to another, by the total time spent at the barrier before ascending or before swimming downstream and not retruning. I did not include fish that were tagged at the barrieriin this analysis of activity. I only carried out a detailed analysis of movement downstream and through the barrier for the 2001 field season. Barrier operations, broken antenna moorings, and a non- functioning antenna (A2) in 2000 made further analysis impractical. l9 For all statistical analysis, I first tested for equality of variance. If variance was equal, I used a pooled t-test, and if variances were unequal, I used Satterthwaite’s approximation. I used Fisher’s Exact Test to compare differences in percent passage between gastrically and surgically implanted fish, males and females, and barrier passage between fish tagged in Ludington and those tagged at the barrier. I performed all statistical tests using or=0.05. 20 RESULTS Timing of migration I captured steelhead in trap nets from 3 March to 28 March, 2000, with many of the fish being caught during the week of 18 March to 24 March (Figure 4). I captured all fish when water temperature was 3°C or warmer, and captured 65% of the fish when water temperature was at least 7°C. Radio-tagged steelhead arrived at the Custer reception area (Section B) from 1 March to 15 April, and most of the arrivals occurred over a one-week period from 22 March to 30 March. In the 2001 field season, I captured steelhead in trap nets from 14 March to 16 April (nets were not fished from 20 March to 1 April), and all fish were caught when water temperature exceeded 2.5° C (Figure 5). I captured seventy-two percent of the fish when the water temperature was 7° C or warmer. The daily catch was generally 1-3 fish per night with the exception of 20 fish. being captured on 8 April. It was during this and the next two days that the majority of the tagged fish arrived at the Custer banier site. Water temperature (Figure 6) and discharge (Figure 7) were very different between the 2000 and 2001 study periods. In 2000, water temperature fluctuated greatly, but warmed much earlier than in 2001. The temperature was consistently above 4°C after 4 March 2000, whereas the river did not reach these temperatures until 30 March in 2001. The river did, however, reach a higher water temperature during the capture portion of the 2001 field season (11.2°C) than it did in 2000 (95°C). Discharge steadily decreased 21 , 00/9 r/vo i 00/v I/VO ff 00/z1/vo j: 00/01/170 1 I 00/80/70 I 00/90/170 : 00/170/170 I 00/z0/vo ’; OO/IE/EO f, 00/62/20 , own/£0 com/£0 —i OO/EZ/EO " 00/IZ/SO 00/6 l/SO OO/L [/80 00/91/50 1: OO/E [/20 ——J: 00/11/2‘0 =—_— 00/60/E0 OO/LO/EO 00/90/20 _ 00/2‘0/50 . ==f 00/10/so will I 12 O 00 \O V N O .—¢ '(3) armaradurar pue Aepmsg 30 raqumN Date Figure 4. Daily catch of steelhead in fyke nets in Ludington, and daily arrival of tagged steelhead, and water temperature at the Custer station, Spring 2000. - capture -— temp . L [:1arrival 1 l l l 1 25 O In 0 V3 N v—t '— '(Q) armeradurar pue Asp/1131.; 50 raqurnN 23 [0/ 9 [/170 [0/7 [/170 [O/Z [/170 [0/0 [/170 [0/80/70 [0/90/70 [ONO/170 ' IO/ZO/VO [0/ [SI/£0 lO/6Z/EO ' IO/LZ/SO 7 IO/SZ/EO IO/EZ/EO [0/ [23/90 [0/ 6 [/EO [O/L [/80 “ 10/9 I/EO [0/5 [/‘50 " 10/ I [/90 - [0/60/‘50 [O/LO/EO [0/ SO/EO [O/SO/EO Date Figure 5. Daily catch of steelhead in fyke nets in Ludington, and daily arrival of tagged steelhead, and water temperature at the Custer station, Spring 2001. ‘ - *r‘ \ \ '~ ~. .. "i“ ~~~~~~~ 4f”- .‘. \' .4»- I - _ I .— ' —lr‘ \~ . AL. s. .” .4... r'“ .— +- ‘s, 4— ‘Q o- ‘ ” Hr— aaaaa AP 6.: «L ' :3 +— o __ C ...... ._ O O - ' "' 2 8 8 T I . ’’’’’’ 4»- l _ r ' ‘- - “I I f .11.- ._'_L l ,J .. l ' ’ a, «P— \, AL ‘ -\ .11.- ,J .I ‘~ - ~ 4- ~ ~~ ‘il' ‘ ~ \ '4‘ L. AL- \ § \ 4r— ,1 A- ' I / -1~ \ i U ‘ . \. a» I. "it“ x . an. I I ~11— C', 7” ~ ‘ .11. I .. ‘ s . .... \ ~‘ ' .4.— ~ . ~ . ‘ o ‘ “l— t l L I I .L r r c l l N O 00 \O V o v-I '— (3) anumadurar 1919M 24 [0/91/170 [0/171/170 [O/ZI/VO [0/01/170 [0/80/170 [0/90/[70 [0/[70/170 [0/Z0/t70 [OHS/£0 [0/62/50 [O/LZ/EO [O/SZ/EO [0/€Z/€0 lO/[Z/EO [0/61/50 [O/LI/€O [O/SI/EO [OISI/EO [0/[ [/50 [0/60/80 [O/LO/fIO [O/SO/EO [O/EO/EO Date Figure 6. Water temperature at the Custer station, Spring 2000 and 2001 l - -2000, r - IO/9l/t70 . [O/VI/VO _ IO/ZI/tO - IO/OI/tIO , roam/to _, [0/90/170 [ONO/170 t. ,_ [OIZO/VO , lO/IE/SO , IO/6Z/SO _ IO/LZ/EO ,, IO/SZ/EO - tom/so _ lO/IZ/EO IO/6l/£O ,, tour/so ,_ 10/91/20 .. IO/Cl/EO _ IO/II/EO , 10/60/EO , [O/LO/EO g lO/SO/EO [0/€0/€0 2000 — 1500 4 1000 4 500 ~ (5.10) 9318mm 25 Date Figure 7. Discharge at the US Geological Survey gauge station in Scottville, MI, Spring 2000 and 2001. throughout the 2000 study period, but increased during the 2001 season. In 2000, discharge was highest on 1 March (1176 cfs) and rapidly fell until 8 March. It then remained between 567 cfs and 767 cfs through the rest of the field season. Conversely, discharge in 2001 was generally increasing, and reached its maximum (2170 cfs) at the end of the capture period. The minimum discharge was 767 cfs. Percent passage In spring 2000, I implanted radio transmitters into 44 steelhead at Ludington. Twenty-seven (61%) were detected at the Custer receiver. Five of these 27 fish passed through the section before the banier began operation (21 March). I tagged twelve more steelhead directly below the barrier, three of which moved downstream and remained stationary. Thirty-one tagged fish were actively seeking passage around the barrier at some time during the study period. Seventeen of these steelhead were able to pass upstream of the banier during the period when the grid was de-energized. Only 1 (3%) fish ascended the ladder when the barrier was fully operational. Of the fish tagged in Ludington that never reached Custer, two were recaptured in other rivers. One was caught by an angler in the Sable River, a Lake Michigan tributary whose mouth is about 9.5 km north of the Pere Marquette River, and the other was captured at the Little Manistee River egg collection weir (MDNR), approximately 34 km north. I found nine fish after tagging that were never detected at the banier, and failed to locate six other fish after tag implantation. In 2001, I radio-tagged 43 steelhead at Ludington. Thirty-two (74% total, 69% surgical, 78% gastric) of these fish were detected at Custer. Of the 32 tagged fish that anived at Custer, nine passed before the barrier began operation (5 April). I tagged an 26 additional 14 steelhead, one of which remained stationary after release (and presumably died or expelled its transmitter), just below the banier. Thirty-six fish were assumed healthy and actively seeking passage around the barrier. At least twenty-four (67%) of these fish were able to ascend the ladder. The successfirl passage of 19 of these fish was recorded on the Custer receiver, and five were accounted for by other means. Two of the 5 successful fish were detected at Custer and at Bowman Bridge, but the Custer receiver did not detect their passage through the fishway. Another of the five fish was captured by an angler upstream of Custer, and its transmitter was nonresponsive. Barrier personnel handled and recorded floy tag numbers from one steelhead with an unresponsive transmitter, and one that had regurgitated its transmitter. The twelve fish that did not ascend the ladder were either located downstream (7 fish), or were not relocated (5 fish). Of the eleven fish tagged in Ludington that never reached the barrier at Custer, three were never located. One was recaptured by an angler in Cooper Creek, a small tributary to Lake Michigan, the mouth of which is about 28 km north of the Pere Marquette, and another was caught by a charter boat in August, about 31 km offshore from Frankfort, MI. Six of the tags remained stationary, and did not reach Custer through the end of the study period. Males and females were equally successful at ascending the fish ladder (males=69% passage, females=65% passage). There was no significant difference (t=1.83, df=31, P=0.08) between the size of fish that ascended the fish ladder (3.4 kg, SE=0.18) and those that did not (3.9 kg, SE=0.35). Additionally, there was no significant difference between percent passage of gastrically tagged fish (74% passage) and surgically tagged fish (54% passage) (Fishers Exact Test, P=0. 14). 27 Of the fish that I tagged immediately downstream of the weir, 85% succeeded in passing through the fishway and 57% of the fish that were tagged at Ludington were successful. These proportions were not significantly different (Fisher’s Exact Test, p=0.086), however, the sample size is relatively small. Rate of ascent to Custer The mean (median in parentheses) time from tag implantation in Ludington to first detection at the Custer receiver station in 2000 was 151.2 hours (103.2 hours) (n=27,SE= 26.4 h, range 24—456 h). Steelhead that I radio-tagged in Ludington in 2001 were detected at the Custer station in a mean time of 76.8 hours (38.4 hours)(n=29, SE=21.6 h, range=16.8- 604.8 h), which was significantly faster than in 2000 (t=2.31, df=54, P=0.02). There was no significant difference in time of passage between surgically and gastrically implanted fish in 2001 (t=0.62, df=20, P=0.54). R_ate of bypassing weir In 2001, tagged and untagged fish exhibited similar patterns in their timing of migration through the fish ladder (Figure 8), suggesting normal behavior of radio-tagged fish. The first radio-tagged fish entered and ascended the ladder on 10 April when the water temperature was 95°C (Figure 9). Most fish, both tagged and untagged, moved upstream through the ladder when water temperature was above 9°C, and remained stable or was on the increase. Eighteen tagged fish ascended the ladder upon first entry, while three ascended on their second entry into the passage. Detailed movements were not available for three fish due to tag loss or failure. The first approach to the electrical barrier was equally distributed across the river width. Six fish first encountered the weir along the north side of the river (A2), six fish 28 1000 “.--..- ___ _5_ o ...-t a .2 m '33. 5 °0 0 no 8 O0 5 a I E] _ , _ i i. I ! ‘ I I o o o o o o o o o oo 0 <1- N [1593019qu [0/0I/90 [0/80/90 [0/90/90 [O/VO/SO [O/ZO/SO [O/OE/VO [0/8Z/l70 [0/9Z/t70 [ 0/ 173/170 [OIZZ/VO [0/0Z/170 [0/8I/170 [0/91/170 [O/VIIVO [0/Zl/170 [0/01/170 [0/80/170 [ 0/90/170 Date Figure 8. Number of tagged (x100) and untagged fish ascending the fish ladder in Custer, MI, Spring 2001. [0/90/ 90 [0/ 70/ SO [O/ZO/ 90 [0/09/70 [0/ 8Z/ 70 » 10/9Z/70 i 10/72/70 IO/ZZ/70 7 10/02/170 [0/ 8 [/ 70 - 10/91/170 [0/7 1/70 ' 1m {/70 00 fl - - - i— , v 7 ‘ _ 1 ‘ \O V” N O OO O V N 0 v-‘ —t v—t y—n “(3) arnraradtuar 1912M pure qsg 30 .raqrunN 30 [0/0 [/70 Date Figure 9. Number of tagged steelhead passing through the fish ladder and water temperature at the Custer station, Spnng 2001 encountered it in the middle of the river (A3), and nine fish encountered it along the south side of the river (four at A5, five at A5). Activity of tagged fish in the vicinity of the barrier ranged from 03-215 antenna switches per hour (sph). Fish that passed the banier appeared to be more active (7.1 sph) than those fish that did not pass (3.5 sph), although the difference was not statistically significant (t=-1.21, df=16, P=0.24). Gastrically tagged fish (6.5 sph) and surgically tagged fish (2.8 sph) were equally as active in their search for passage (t=l .73, df=15, P=0. 1 0). Tagged fish arrived at the Custer station at all times of the day, with 62% aniving during daylight hours and 38% arriving in the dark (Figure 10). Approximately 58% of the 24-hour period at this time of year was daylight. Most (88%) tagged fish that passed through the banier section before the banier was operating (5 April) moved in the dark (Figure 11). Most (86%) tagged fish that ascended the ladder during weir operation moved during daylight hours. The time it took for the tagged steelhead to pass through the weir site (Section B) in 2000 depended on the operation of the banier. The fish that anived before the barrier was turned on (21 March) passed through the section relatively quickly, with a mean time of 1.0 hour (n=4, SE=0.35 h, range=0.43 to 2 h). Once the barrier was operating, only one steelhead successfirlly navigated its way through the fish passage and around the barrier, and it did not pass until 511.2 hours after its arrival at the barrier. Many of the steelhead passed through the section from 31 March to 8 April, when the electrical field was turned off between 10:00 and 16:00 to allow fish passage. The mean time of passage during this period was 105 hours (136 hours)(n=l 1, SE=20.2 h, range=1 to 171 h). 31 DAYLIGHT 4 T "m N .— O 119‘} 30 1:;qu 32 0030 003€Z 00 3ZZ OOZIZ 0030Z 00361 00581 001.1 00191 00391 00371 00381 00321 OOZII 00301 0036 0038 003/. 0039 0019 0037 0038 0033 0031 Time Figure 10. Number of tagged steelhead arriving at Custer and time of anival, Spring, 2001. 00 30 .Som wctam doufiomo to? Go 5:38 macaw can £0 SEE: 8&3 $350 8 SEE 05 wEmmma 32:0on comma“ («o 58:: Z A _ Emmi DEE z z z Z I I I. I. I. I. I I I. I. 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O 0 0 O O 0 0 0 O 0 . EIEEB D , :0 Stamp I HEOE>1 80 hours in their efforts to locate the entrance to the fish ladder or to enter it. Most upstream movement through the ladder can be expected during the latter part of 38 March to the end of April, during daylight hours, and when water temperatures reach 9°C. It is also apparent that minor flaws in the operation of the electrical barrier and fish ladder can have a significant effect on the movement of steelhead. It is important to monitor for and to eliminate stray voltage from the barrier that may block access to the fish ladder. 39 CHAPTER 2: AN ANGLER SURVEY ON THE PERE MARQUETTE RIVER, MICHIGAN 40 ABSTRACT I conducted an angler survey on the Pere Marquette River, MI in order to characterize the spring recreational fishery. The Pere Marquette is a free flowing coastal stream that experiences anadromous runs of native and introduced species. I designed this study to document all anglers, including those fishing for steelhead and suckers. My study area consisted of 17 sampling sites located from Baldwin downstream to Ludington on the mainstem of the river. The section is open to year round fishing and includes a seven mile long stretch designated as a catch-and-release “flies only” area. I found that anglers fishing the Pere Marquette from public access sites during the spring months of 2000 and 2001 most often targeted steelhead, and effort totaled an average of 110,000 hours per spring. This is a dramatic increase in effort since the time of the last angler survey, nearly twenty years ago (22,228 hours per spring). Anglers caught an average of 18,074 steelhead, 7,504 brown trout, and 24,029 suckers per spring. Most anglers fishing for suckers were from local counties, while those pursuing steelhead typically traveled greater distances. Anglers who targeted steelhead were most numerous at upstream sites, particularly between Green Cottage and Gleason’s Landing, and anglers who targeted suckers were most numerous at the three downstream sites. 41 INTRODUCTION The Pere Marquette River is one of the last remaining free flowing coastal streams in Michigan. It is a state “Natural River” and a national “Wild and Scenic River” and is highly regarded by anglers for its resident and anadromous fish populations. The Pere Marquette is believed to be the first river in North America to be stocked with brown trout (MacCrimmon and Marshall 1968), and the historical significance of trout and sahnon to the fishery has been well documented (Kruger 1985). The sahnonid fishery is nationally renowned and provides a significant contribution to local economies; however, there are at least 40 other species of fish that inhabit the river, some of which also contribute to the recreational fishery (MR 1998). Heavy angling pressure occurs in the spring when large numbers of native and introduced species make spawning migrations into the river from Lake Michigan. It is widely known that many of the anglers are targeting steelhead, but it is apparent that suckers are also an important species (Aaron Snell, personal observation). The goal of this study is to describe the spring recreational fishery on the Pere Marquette River. Nelson et al. (1998) conducted a recreational use survey on the mainstem, and Nelson and Smith (1998) give a detailed description of use patterns and current management issues, but did not explore the fishery in detail. Kruger (1985) conducted a year long angler survey on the upper mainstem of the Pere Marquette, which lies in an area designated as a Michigan “Blue Ribbon” trout stream, but no other creel survey work has been done since then. I designed the present study to estimate total 42 effort, catch per effort, and total catch on the entire maintstem of the Pere Marquette River from 23 F ebruary-29 April, 2000 and 2001. 43 STUDY SITE The Pere Marquette River is located along the eastern shoreline of Lake Michigan in west central Michigan (Figure 12). The watershed encompasses an area of 1956 km2 and contains 612 lineal km of streams. From the headwaters near Chase to Pere Marquette Lake in Ludington, the Pere Marquette River courses a distance of about 154 km (MDNR/IFR 1998). A large percentage of the riparian lands along the river are privately owned, limiting public access to commercial or govermnent maintained sites (Table 1). This study focused on seventeen sampling sites located on the mainstem of the Pere Marquette from Baldwin downstream to Ludington (Figure 13). Like many free-flowing rivers, the habitat in the Pere Marquette River changes as it winds its way from its headwaters to Lake Michigan. In general, river gradient is steep in the upper watershed and gradually decreases as the river nears its mouth. The upper mainstem, near Baldwin, is characterized by relatively shallow water and numerous riffle areas that attract large numbers of spawning fish. The river is easily waded, yet provides sufficient access to canoes or boats. Downstream of Gleason’s Landing, much of the river has a stream gradient of less than 1 meter per kilometer and is characterized by slow runs and deep pools (MDNR/IFR 1998). Mid-stream sites (Bowman Bridge to Maple Leaf) offer limited access to wading anglers and several areas of suitable spawning habitat. Downstream sites (Walhalla Bridge to Twin Bridges) are primarily accessed by boat or from shore. Currently, Michigan sport fishing regulations allow year-round fishing on the 44 Figure 12. Location of the Pere Marquette River in Michigan. 45 Lud In g ton CUS N r 17 16 :14 13 12 Bowman Bridge Bald w in See Table 1. For site names Figure 13. Location of the seventeen sampling sites on the Pere Marquette River, M1 for the angler survey. 46 Table 1. Sampled public access sites on the mainstem of the Pere Marquette River from Baldwin downstream to Ludington. Site # County Agency Name Type 1 Lake MDNR M-3 7 Watercraft/Wading 2 Lake Unknown 72nd St Wading 3 Lake USFS Green Cottage Watercraft/Wading 4 Lake USF S Clay Banks Wading 5 Lake USF S J orgenson’s Wading 6 Lake US F S Gleason’s Watercraft/Wading 7 Lake USF S Bowman Bridge Watercraft/Wading 8 Lake USF S Rainbow Rapids Watercraft/Wading 9 Lake MDNR Sulak Watercraft/Wading 1 0 Lake USFS Upper Branch Watercraft/W adin g l 1 Lake Unknown Ackerson’s Wading 1 2 Lake USFS Maple Leaf Wading 13 Lake MDNR Walhalla Bridge Watercraft/Wading 14 Mason USF S Indian Bridge Watercraft/Wading 15 Mason MDNR Custer Bridge Watercraft 1 6 Mason MDNR Scottville Bridge Watercraft l 7 Mason1 MDNR US. 31 Watercraft 1 7 Mason1 Unknown Sutton’ 3 Landing Watercraft USFS-United States Forest Service; MDNR-Michigan Department of Natural Resources. 1 included in the “Twin Bridges” site. 47 mainstem of the Pere Marquette River from Lake Michigan upstream to M-37 in Baldwin. Most tributaries are closed to fishing from 1 October through the last Saturday in April to protect spawning fish. Additional regulations include a “flies-only” catch- and-release section between M-37 and Gleason’s Landing, and a restriction that limits the use of motorized watercraft to the portion of river downstream of Indian Bridge. 48 METHODS I patterned this roving survey after those used by the Michigan Department of Natural Resources (MDNR) and it consisted of two separate sampling components: interviews of angling trips and counts of anglers (Lockwood 1999). The survey followed a stratified design and the data I collected represents angling characteristics for each of the sampling sites during specific calendar and daily periods. The study focused on the entire mainstem, during a 10-week period in the spring of 2000 and 2001, and was designed to quantify all public-access angling. I selected the seventeen sampling sites based on patterns of previous use (Nelson et al. 1998) and personal observation. I selected sampling dates and times by dividing the 10-week sampling period into blocks of two weeks. Within each two-week block, I randomly selected three weekdays (Monday-Friday) and three weekend days. I then assigned a sampling time to each of the three weekdays and three weekend days. A morning (700-1200), aftemoon (1200-1700), and evening (1700-2200) time block were all represented once during the week and once during the weekend. During each of the sampling periods, I used “count-while-interviewing” progressive vehicle counts to estimate effort at each of the 17 sampling sites (Pollock et al. 1994). I randomized the starting point and direction of travel (N euhold and Lu 1957), and used checkpoints along the route to stay on schedule, thereby eliminating potential bias (Wade et al. 1991). In addition to the number of vehicles present at each site, I recorded the date, day of the week, count start time, count end time, starting site, and direction of travel on the data sheet (Appendix A). I tallied vehicles into four categories: 49 Michigan vehicles, Michigan vehicles with trailer, out—of-state vehicles, and out-of-state vehicles with trailer. I also recorded the state of residence for all out-of-state travelers. Angler use was restricted during a portion of the 2001 sampling season due to inclement weather. Residual snow and ice storms closed Sulak until 3/11, Ackerson’s until 3/22, and J orgenson’s and Maple Leaf until 3/25. Additionally, these sites were only accessible by 4-wheel drive on some sampling dates. Twin Bridges was closed for r approximately half of the 2000 season due to road construction. I conducted angler interviews during and after the car counts at each of the 17 sampling sites. All anglers interviewed fished for a minimum of 0.5h (Pollock et al. p 1997). I interviewed most anglers while they were actively fishing; however, anglers i were sometimes met at the completion of their trip. Therefore, I recorded both complete and incomplete-trip interviews on forms patterned after those used by the MDNR (Appendix B). For the purpose of this study, I considered steelhead and rainbow trout to be the same species, and recorded all species of sucker into one general category. In an effort to avoid angler party size bias, I recorded information by individual angler rather than by angling party (Lockwood 1997). Summary parameters I used the following formulas to calculate the summary parameters for estimates of effort, catch per effort, and total catch. Formulas marked with an asterisk (*) and all variance estimates are from Lockwood et al. (1999). I developed the other formulas for use in this study. Total Effort: Car hours*: 50 [3 = F c where, B = estimated total number of car-hours from progressive counts F = fishable hours in a stratum c = total number of cars counted Angler hours*: E=fl5 where, E = estimated angler hours a = number of anglers in a party Boat vs. shore hours: M: “<1"! where, M = proportion of angler hours fished from a boat T = total number of trailers counted per section. V = total number of vehicles counted per section. Angler trips“: 2 i=1 f=-—- 5p= k 1‘: k E T where, a = estimated angler trips t = length of fishing trip 51 k = total number of anglers interviewed Targeted effort': D = estimated effort targeted at steelhead k”, = number of anglers interviewed at site 3 targeting steelhead kc, = number of anglers interviewed at site 3 targeting suckers J = estimated effort targeted at suckers 1Only 2000 data was used for estimation at the Custer site, due to low number of 2001 interviews. The low number of interviews was caused by difficulty in accessing fishermen. Catch per effort“: Incomplete trips: kp . 2 (0%...) kn rp = _- where, c = total catch of a particular species by angler i h = total angler hours fished by angler 1' Complete trips: rp - 1.1—], 52 lncomplete(F) + C 0mplete(F ) 2 Catch per effort (R) = Total catch: C = RE where, R = appropriate catch rate estimator 53 RESULTS I counted a total of 2,044 vehicles in 2000 and 1,948 in 2001 (Table 2). Green Cottage and Gleason’s Landing were the most used sites, with 37% of the total number of vehicles counted at these two sites alone. Custer was the third busiest site in both years of the study. The estimated number of total angler hours (i 2 standard errors) was consistent during both years of the study. In 2000, anglers spent 110,435 (9,844) hours on the river, while in 2001 effort was estimated to be 109,626 ( 10,694) hours. Much of the effort (5 8% in 2000, 57% in 2001) was concentrated in the “flies only” section which offers high accessibility and a large number of spawning fish (Figure 14). Fishing effort was evenly distributed over the weekday (57,036 angler hours) and weekend (53,399 angler hours) strata during the 2000 season; however, more effort was applied during the weekdays in 2001 (65,314 angler hours on weekdays, 44,313 angler hours on weekends). The division of angler hours between boat and shore anglers was similar between years, as well as regions of the river. In 2000, 19% of the anglers fishing in the “flies only” section fished from a boat and in 2001, 22% fished from a boat. Downstream of Gleason’s Landing, 20% of the anglers fished from a boat in 2000, and 25% fished from a boat in 2001. The estimated number of angler trips (3: 2 standard errors) in 2000 was 18,410 (3,006) for steelhead and 4,114 (1,158) for suckers. In 2001, the number of angler trips was similar at 18,015 (4,171) steelhead trips, and 4,116 (1,170) trips targeting suckers. A total of 81.7% of the fishing trips targeted steelhead in 2000, as compared to 81.4% in 54 Table 2. Total number of vehicles counted at each Pere Marquette River access site during the 2000 and 2001 sampling seasons. Number of vehicles Site 2000 2001 M-37 90 107 72"d Street 144 105 Green Cottage 428 393 Clay Banks 1 16 93 Jorgenson’s 68 67 Gleason’s Landing 328 335 Bowman Bridge 51 47 Rainbow Rapids 63 78 Sulak 133 94 Upper Branch 98 96 Ackerson’s 28 26 Maple Leaf 93 91 Walhalla Bridge 54 63 Indian Bridge 54 58 Custer 1 74 142 Scottville 97 108 Twin Bridges 25 45 Total 2044 1948 55 5939118 HIM; OHIAJJOOS l rarsng .2000“,E 1.12001 f “I I I I I . I II 939118 Imam 9313.08 2119mm 399'] =91de 81108193319! [1903-18 Jaddn 3191118 sprdeu Momma}! 931mg umnmog 30000 4 ., . 531059919 sauna £213 suasuaflror 9391103 uaorg _ IS puz L I. LUV <2 .2 g. g. g. .1. O o O O o O o O O o W O V) o m N N F‘ .— srnoq ralfiuv Site Figure 14. Estimated total effort at each study site on the Pere Marquette River, MI, Spring 2000 and 2001. 2001. Trip lengths were about 2 hours longer for steelhead anglers than for sucker anglers. This meant that 87.5% (96,652 hours in 2000, 95,839 hours in 2001) of the actual total effort was spent fishing for steelhead. In both years of the study, anglers interviewed in the “flies—only” section, and several midstream sites targeted only steelhead (Figures 15 and 16). At downstream sites, such as Indian Bridge, Custer, and Scottville, a greater proportion of the effort targeted suckers. Catch rates varied greatly between species (Table 3). Sucker anglers caught only ' suckers, and had the highest catch per effort in both years of the study. The lowest catch per effort was for brown trout, which were caught regularly by anglers targeting steelhead. Catch rates were similar between years for all species of fish, although there appears to be a reduction in 2001 for steelhead and suckers. This possible reduction can most likely be attributed to the large estimated variances. Based on these estimates of catch per effort, the total catch (i 2 standard errors) for 2000 was 19,427 (i10,025) steelhead, 6,766 (i3,817) brown trout, and 33,166 (i225,860) suckers (Figure 17). In 2001, the total catch was estimated to be 16,676 (i12,152) steelhead, 8,242 (i10,325) brown trout, and 14,891 (i40,886) suckers (Figure 17). The majority of anglers fishing on the Pere Marquette River were residents of Michigan, although license plates from 29 other states or provinces were observed (Table 4). Interview data suggest that anglers targeting steelhead were likely to travel greater distances to the Pere Marquette than those anglers targeting suckers. Interviews of anglers fishing for steelhead represented thirty-six Michigan counties. In 2000, about 36% of steelhead anglers were from Mason or Kent counties (Table 5). In 2001, about 34% of anglers interviewed were from Kent or Oakland counties. Sucker anglers, 57 meanwhile, represented only 13 counties within Michigan. Most (76% in 2000, 70% in 2001) anglers targeting suckers were from Lake or Mason county. 58 93pm 11919111 ' 9313118 91121118,“ F‘ IIIIII||I 3°91‘fldmrv L— W suosraypv q 1101!ng Jaddn _ W8 I" SPIdBtI AAOClurtm 931mg tram/“cg gnome“) {— _ 911133 £213 #- — 511951198101: p — 93"”03 “9919 I. — “51% r. _ LEW 30000 , .. I l I T r ‘ O O O O 8 o O O O o O O O o o V) O '13 o m N N .... ...-1 smoq ”130V 59 Site Figure 15. Estimated targeted effort at each sampling site on the Pere Marquette River, MI, Spring 2000. 893998 “WI 911141-1008 .191an E ... 5 13 9313118 auteur 3 '5 1;; a I D. 9313118 91121112111 l >— — 3““ ”“9” q suosrarpv fl “m8 ”“0 _ ““3 [‘4 _ SPIdetl ”‘09“th P ‘ afipug unrumog — s'“°°°°'9 h — ”“98 “‘3 p— _ 51195119810 1‘ P— - °°°°°3 ”“9 — ‘5 9““ .— 25000 ‘ “W I I T I o o o o o O O O O o o o o O In 0 In N v—d '— smoq ralfiuv 60 Site Figure 16. Estimated targeted effort at each sampling site on the Pere Marquette River, MI, Spring 2001. _oom can ooom o5 wESc :2 53¢ @3332 20m 2: co Ewan”. 2mm .3 .5983: :38 3683mm .2 Emmi E833 .3033. 9:383 88on So: .365 32:86 58 n. w 88 I . coca cocoa ”N n w OOOma m. 0 1. w. 088 M Du n 0.0 mu ooomm ooooM 000mm 61 Table 3. Catch per effort (fish/hour) estimates for anglers fishing on the Pere Marquette River, MI, Spring 2000 and 2001 (:2 standard errors). 2000 2001 Target steelhead suckers brown trout steelhead suckers brown trout steelhead 0.201 0.028 0.070 0.174 -- 0.086 (0.089) (0.036) (0.034) (0.1 12) -- (0.094) suckers -- 2.210 -- -- 1.080 -- -- (2.036) -- -- (0.360) -- 62 Table 4. Percentage of anglers fishing the Pere Marquette River in 2000 and 2001, based on vehicle counts, by state or province of residence. Residence 2000 2001 Michigan 93.1% 91.9% Alabama -- 0.05% Arkansas 0.04% -- California -- 0.05% Colorado 0.15% 0.46% Georgia 0.04% 0.10% Idaho 0.15% 0.21% Illinois 2.25% 2.67% Indiana 1.66% 1.33% Iowa 0.15% 0.31% Kentucky 0. 10% -- Maryland -- 0.05% Minnesota 0.04% 0.05% Mississippi -- 0.05 % Missouri 0.04% -- Montana 0.29% 0.21% New Hampshire -- 0.10% New York -- 0.05% North Carolina 0.04% -- Ohio 1.61% 1.39% 63 Table 4 continued (page 2). Residence 2000 2001 Ontario 0.04% 0.10% Pennsylvania -- 0.21% Tennessee -- 0. 10% Texas -- 0.05% Vermont -- 0.05% Virginia -- 0.05% West Virginia 0.15% 0.05% Wisconsin 0.04% 0.36% Wyoming 0.04% -- Sample Size 2044 1948 Table 5. Percentage of Michigan anglers interviewed while fishing for steelhead and suckers on the Pere Marquette River during 2000 and 2001 by county of residence. 2000 2001 Residence Steelhead Suckers Steelhead Suckers Allegan -- 1.72% -- -- Barry 0.80% -- -- -- Bay -- -- 3.75% -- Branch -- -- 1.25% -- Calhoun -- -- 1.25% -- Clare 2.40% -- 1.25% -- Clinton 0.80% -- 1.25% -- Crawford 0.80% -- -- -- Eaton 0.80% -- -- -- Emmet 0.80% -- -- -- Genesee 2.40% -- 1.25% -- Grand Traverse 1.60% -- -- -— Gratiot 1 .60% -- -- -- Ingham 4.80% -- -- -- Isabella 0.80% -- 2.50% -- Jackson -- -- 1 .25% -- Kent 15.20% -- 16.25% -- Lake 5.60% 18.97% 3.75% 24.24% Lapeer -- -- 1.25% 3.03% Livingston -- -- 1.25% -- 65 Table 5 continued (page 2). 2000 2001 Residence Steelhead Suckers Steelhead Suckers Macomb 1.60% -- 1.25% 3 .03% Manistee 0.80% 3.44% 1.25% 3.03% Mason 20.80% 56.90% 8.75% 45.45% Mecosta 7.20% 1.72% 6.25% -- Midland 2.40% -- 3.75% -- Monroe 0.80% -- 3.75% -- Montcahn 0.80% -- -- -- Muskegon 4.80% 12.07% 6.25% 9.09% Newaygo 2.40% -- -- 3.03% Oakland 5 .60% -- 17.50% ~- Oceana 1.60% 3.45% 1.25% 3.03% Osceola -- -- -- 3.03% Ottawa 4.80% -- 1.25% 3 .03% Saginaw 1.60% -- 3.75% -- Shiawasee -- -- 1 .25% -- St. Joseph -- -- 1.25% -- Washtenaw 0.80% -- -- -- Wayne 5.60% 1.72% 6.25% -- Sample size 125 58 80 33 66 DISCUSSION The results of this study show that steelhead are, in fact, the principal species targeted in the spring fishery on the Pere Marquette River. Steelhead anglers can be found along the entire length of the river, but the highest density of anglers is in the “flies only” section. Sucker fishing, while only accounting for about 15% of the total effort, attracts mostly local anglers who spend nearly 14,000 hours on the lower portion of the river. Also apparent from the results of this study and from Nelson et al. (1998) is that the Pere Marquette River has experienced dramatic changes in angler use over the past two decades. In 1982-83, anglers spent an estimated annual total of 74,294 hours fishing the section of river between M-37 and Rainbow Rapids, with approximately 30% of the effort taking place from February to April (Kruger 1985). During 2000 and 2001, this same section of river experienced an average of 69,584 angler hours from late February to late April alone. Kruger (1985) found that anglers fishing the “flies only” section made a total of 2,281 trips during the months of April and May. In my ten-week sampling period, anglers made an average of 12,012 trips to the “flies only” section. Additionally, Kruger (1985) reported that, between March and May, only about 1% of the effort in the “flies only” section was applied by boat anglers. In the two years of the present study, boat anglers applied an average of about 20% of the effort in the “flies only” water, which is probably an underestimate based on the method of calculation used. While it is difficult to make direct comparisons to Kruger (1985) due to methodological differences, it is clear that the number of anglers fishing the Pere Marquette has grown 67 substantially in the past twenty years. Increased use is likely the result of many factors, including management for anadromous species and an increasing human population in the watershed. Interestingly, the users of the Pere Marquette River cite overcrowding as the most negative aspect of their trip (Nelson and Smith 1998). Even with the large increase in total effort over this time period, the catch per effort for steelhead has increased (0.05 fish/hour during spring months in 1983 (Kruger 1985)). Both mandatory and voluntary catch-and-release fishing allow the same fish to be caught multiple times, and undoubtedly contribute to high catch rates and associated total catch. Catch rates of sucker fishermen proved to be difficult to accurately estimate due to large variances. These large variances can be attributed primarily to the low numbers of interviews collected. Additionally, the variance associated with sucker fishing is higher because the catch of suckers per angler (zero to 41 fish) was much more variable than the catch of steelhead per angler (zero to 6 fish). From the results of this study, it is clear that the amount of angler effort has increased substantially on the Pere Marquette River. Most of this pressure comes fiom fishermen that travel from areas outside of the watershed to pursue steelhead in the upper reaches of the river. Sucker fishermen account for about 15% of the total effort, and are typically local anglers that fish the lower portions of the river. 68 LITERATURE CITED 69 LITERATURE CITED Applegate, V.C. 1950. Natural history of the sea lamprey (Petromyzon marinas) in Michigan. U. S. Fish and Wildlife Service Special Scientific Report. 35. Andrew, El, and G.H. Geen. 1960. Sockeye and pink sahnon production in relation to proposed dams in the Fraser River system. Bull. Int. Pacif. Salm. Fish. Commn 11 in Banks, J .W. 1969. A review of the literature on the upstream migration of adult Sahnonids. Journal of Fish Biology. 1:85-136. Banks, J .W. 1969. A review of the literature on the upstream migration of adult Salmonids. Journal of Fish Biology. 1:85-136. Bergstedt, RA, and JG. Seelye. 1992. Current and proposed alternative methods for the control of sea lampreys. U. S. Fish and Wildlife Service. Biette, R.M., D.P. Dodge, R.L. Hassinger, and TM. Stauffer. 1981. Life history and timing of migrations and spawning behavior of rainbow trout (Salmo gairdneri) populations of the Great Lakes. Canadian Journal of Fisheries and Aquatic Sciences. 38:1759-1771. Bjomn, T.C. Personal communication: discussion of radio telemetry techniques used for sahnonid species. University of Idaho. Bjomn, T.C., K.R. Tolotti, J .P. Hunt, P.J. Keniry, R.R. Ringe, and CA. Peery. 1998. Migration of adult Chinook salmon and steelhead past dams and through reservoirs in the lower Snake River and into tributaries. Part 1. Idaho Cooperative Fish and Wildlife Research Unit. University of Idaho. Chapman, W. McL. 1941. Observations on the migration of salmonoid fishes in the Upper Columbia River. Copeia 240-242. Dahl, RH. and RB. McDonald. 1980. Effects of control of the sea lamprey (Petromyzon marinas) on migratory and resident fish populations. Canadian Journal of Fisheries and Aquatic Science. 37:1886-1894. Eiler, J .H. 1990. Radio transmitters to study salmon in glacial rivers. American Fisheries Society Symposium 7:364-369. Gowans, A.R.D., J .D. Armstrong, and I. G. Priede. 1999. Movements of Atlantic sahnon in relation to a hydroelectric dam and fish ladder. Journal of Fish Biology 54:713-726. 70 Hart, LG, and RC. Summerfelt. 1975. Surgical procedures for implanting ultrasonic transmitters into flathead catfish (Pylodictis olivaris). Transactions of the American Fisheries Society 104256-59. Hay-Chmielewski, E.M., P.W. Seelbach, GE. Whelan, and DB. Jester Jr. 1995. Huron River Assessment. Michigan Department of Natural Resources, Fisheries Division, Special Report No. 16. Hunn, J .B., and W.D. Youngs. 1980. Role of physical baniers in the control of sea lamprey (Petromyzon marinas). Canadian Journal of Fisheries and Aquatic Science. 37:2118-2122. Koon, E. Personal communication: Availability of sea lamprey spawning habitat downstream of the Custer sea lamprey barrier. United States Fish and Wildlife Service. Kruger, K. 1985. Pere Marquette River Angler Survey and Brown Trout Evaluation. Masters Thesis. Michigan State University. East Lansing, MI. Laine, A. 1995. Fish swimming behavior in Finnish fishways. Movements of Atlantic salmon in relation to a hydroelectric dam and fish ladder. Journal of Fish Biology 54:713-726. ‘ Lockwood, R.N., D.M. Benjamin, and J .R. Bence. 1999. Estimating angling effort and catch from Michigan roving and access site angler survey data. Michigan Department of Natural Resources. Fisheries Research Report No. 2044. Lockwood, RN. 1997. Evaluation of catch rate estimators from Michigan access point surveys. North American Journal of Fisheries Management 17:611-620. MacCrimmon, HR, and TL. Marshall. 1968. World distribution of brown trout, Salmo trutta. Journal of the Fisheries Research Board of Canada. 25:2527-2598. Manion, P.J., and L.H. Hanson. 1980. Spawning behavior and fecundity of lampreys from the upper three Great Lakes. Canadian Journal of Fisheries and Aquatic Science. 37:1635-1640. McLain, A., B.R. Smith, and H.H. Moor. 1965. Experimental control of sea lampreys with electricity on the south shore of Lake Superior, 1953-60. Great Lakes Fish. Comm. Technical Report. 10: 48 p. MDNR/IFR (Michigan Department of Natural Resources Institute for Fisheries Research). 1998. Streambed gradient data for the mainstem Pere Marquette River. 71 MR (Mainstream Resources). 1998. Comprehensive surface resource assessment of the Big South Branch Pere Marquette River: Lake, Mason, Newaygo and Oceana counties, Michigan. The Pere Marquette Watershed Council and the Great Lakes Fishery Trust. 37 pp. Nelson, C.M., A. Smith. 1998. Visitor and riparian issue scoping for the Pere Marquette Scenic Corridor. Department of Park, Recreation, and Tourism Resources. Michigan State University, East Lansing, MI. Nelson, C.M., P. Johnson, D]. Stynes. 1998. Estimated fall 1996 and spring and summer 1997 recreation use of the Pere Marquette Wild and Scenic River Corridor from selected public access sites. Department of Park, Recreation, and Tourism Resources. Michigan State University, East Lansing, MI. Neuhold, J .M., and K.H. Lu. 1957. Creel census methods. Utah State Department of Fish and Game Publication 8, Salt Lake City. Palmisano, AN. and CV. Burger. 1988. Use of a portable electric banier to estimate Chinook salmon escapement in a turbid Alaska river. North American Journal of Fisheries Management. 82475-480. Pollock, K.H., J .M. Hoenig, C.M. Jones, D.S. Robson, and OJ. Greene. 1997. Catch rate estimation for roving and access point surveys. North American Journal of Fisheries Management 17:1 1-19. Pollock, K.H., C.M. Jones, and TL. Brown. 1994. Angler survey methods and their applications in fisheries management. American Fisheries Society Special Publication 25. Raymond, H.L. 1979. Effects of dams and impoundments on migrations of juvenile Chinook salmon and steelhead from the Snake River 1966 to 1975. Transactions of the American Fisheries Society. 108:505-529. Reynolds, J .B. 1996. Electrofishing. Page 221 in B. R. Murphy and D. W. Willis, editors. Fisheries Techniques, 2n edition. American Fisheries Society, Bethesda, Maryland. Rozich, T.J. Personal communication: Evaluation of the electrical banier operations in 1989. Michigan Department of Natural Resources. Rozich, T.J. 1989. Evaluation of the Pere Marquette River electrical lamprey barrier. Unpublished report, Michigan Department of Natural Resources — Lansing, Michigan. Sapak, J. Personal communication: diel passage of steelhead at the Little Manistee River weir. Michigan Department of Natural Resources. 72 Smith, BR, and J .J . Tibbles. 1980. Sea lamprey (Petromyzon marinus) in Lakes Huron, Michigan, and Superior: history of invasion and control, 1936-78. Canadian Journal of Fish and Aquatic Science. 37:1780—1801. Snell, A]. 1997. Personal observation: Sucker fishing on the Pere Marquette River. Swink, W.D. 1999. Effectiveness of an electrical barrier in blocking a Sea lamprey spawning migration on the Jordan River, Michigan. North American Journal of Fisheries Management. 19:397-405. USGS (United States Geological Survey). 2001. Discharge data for the Pere Marquette River, USGS gauging station #04122500 at Scottville, Michigan 1939-2001. Wade, D.L., C.M. Jones, D.S. Robson, and K.H. Pollock. 1991. Computer simulation techniques to access bias in the roving-creel-survey estimator. American Fisheries Society Symposium 12:40-46. Webb, J. 1990. The behaviour of adult Atlantic salmon ascending the Rivers Tay and Tummel to Pitlochry Dam. Scottish Fisheries Research Report 48. Workman, R.D., T.G. Coon, and DB. Hayes. 2001. A description of the migratory behavior of steelhead (Oncorhynchus mykiss) and longnose suckers (Catostomus catastomus) in the Pere Marquette River, Michigan. Unpublished report, Department of Fisheries and Wildlife, Michigan State University- East Lansing, Michigan. Workman, R.D., D.B. Hayes, and T.G. Coon. In press. A model of steelhead movement in two Lake Michigan tributaries in relation to water temperature. Transactions of the American Fisheries Society. 73 APPENDIX A 74 Date: Start time: Day of week: Michigan 3P Start site (‘4'): End time: Out—of-state cars trailers DOICS cars trailers 110168 M-37 Ix) 7:2"d s: (a) Green Cottage ‘ \— “WI—T _____ _ Clay Banks Ur Jorgenson‘s Gleason's Landing Bowman Bridge Rainbow Rapids Sulak . Upper Branch . Lower Branch . Ackerson's . Maple Leaf . Walhalla Bridge . Indian Bridge 16. Custer l7. Scottville 18. Twin Bridges NOTES: 7S APPENDIX B 76 COMPLETED TRIP ANGLER PARTY INTERVIEW FORM Party inteniew lD3_m_ Date: ___t____ __ Site: Angler lD3__ Day ofweek:___ Interview time:_ a p Anglers in party: Length oftrip: Mode: boat shore Target species: State of residence:_ County of residence: Catch Suckers __ _ Steelhead _ __ Longnose __ _ Brown trout __ __ White _ __ Northern pike _ _ Redhorse spp. __ _ __ __ Rock bass _ __ _ __ Party interview lD#____ Date: _/_/_ Site: Angler ID#__ Day of week:___ Interview time:_ a p Anglers in party: Length of trip: i Mode: boat shore Target species: State of residence:_ County of residence: Catch §p_€_ci_€§ BEE! wed M m: [gm Suckers __ __ Steelhead __ _ Longnose __ _ Brown trout _ _ White __ __ Northern pike __ _ Redhorse spp. _ __ __ _ Rock bass 77 MICHIGAN \‘ TTTTTTTTTTTTT t A n3; ' ll I ll l l ll 1 l l 1293 02318 2557