SMOLT MONITORING IN THE LAGUNITAS CREEK WATERSHED 2015

SMOLT MONITORING IN THE LAGUNITAS CREEK WATERSHED 2015 To: California Department of Fish and Wildlife Fisheries Restoration Grant Program Grant # P1130417 Lagunitas Creek Coho Salmon Life Cycle Monitoring Program Prepared By: Eric Ettlinger, Aquatic Ecologist Patrick Doughty, Watershed Stewards Project/AmeriCorps Member Vincent Rogers, Watershed Stewards Project/AmeriCorps Member Gregory Andrew, Fishery Program Manager Marin Municipal Water District, 220 Nellen Avenue, Corte Madera, CA 94925 (415) 945-1193 In association with the National Park Service and the Salmon Protection and Watershed Network (SPAWN) August 2015

Acknowledgements This project was generously funded by the California Department of Fish and Wildlife, Fisheries Restoration Grant Program, in partnership with the National Oceanic and Atmospheric Administration's Pacific Coast Salmon Restoration Fund. The Marin Municipal Water District (MMWD) would like to thank the Gallagher family for granting us access onto their property to conduct this monitoring. Cover image: Inside the Lagunitas Creek rotary screw trap.

Smolt Monitoring in the Lagunitas Creek Watershed 2015 Table of Contents EXECUTIVE SUMMARY... III INTRODUCTION... 2 METHODS... 3 Lagunitas Creek Monitoring... 3 Olema Creek Monitoring... 4 San Geronimo Creek Monitoring... 5 Data Analysis... 5 RESULTS... 5 Lagunitas Creek Monitoring... 5 Olema Creek Monitoring... 8 San Geronimo Creek Monitoring... 9 DISCUSSION... 9 Sampling conditions and effects on estimated smolt abundance and emigration timing... 9 Age and size composition of salmonids... 10 Smolt abundance trends and implications for winter survival... 11 REFERENCES... 12 List of Tables Table 1. Marking schedule at Lagunitas and San Geronimo Creek smolt traps... 4 Table 2. Estimated smolt emigration from Lagunitas Creek, 2006-2015... 6 Table 3. Salmonids captured in the Lagunitas Creek rotary screw trap by length and week... 7 Table 4. PIT-tagged coho detections... 8 List of Figures Figure 1. Trap locations... 13 Figure 2. Lagunitas Creek smolt emigration and stream flow.... 14 Figure 3. Weekly trap efficiency and Lagunitas Creek flow... 15 Figure 4. Daily coho smolt captures at the Lagunitas Creek smolt trap, 2012-2015... 16 Figure 5. Coho smolt captures and water temperature, 2015... 17 Figure 6. Daily steelhead smolt captures at the Lagunitas Creek smolt trap, 2012-2015... 18 Figure 7. Coho emigration estimates and smolt lengths at the Lagunitas smolt trap... 19 Figure 8. Lagunitas Creek smolt estimates... 20 Figure 9. Juvenile coho abundance and subsequent smolt abundance... 21 i

Smolt Monitoring in the Lagunitas Creek Watershed 2015 EXECUTIVE SUMMARY Downstream migrating salmonid smolts were sampled using a rotary screw trap (RST) in lower Lagunitas Creek, near Point Reyes Station (Figure 1). This represents the tenth consecutive year of smolt monitoring at that location. From mid-march through early June the trap was monitored cooperatively by staff and volunteers from the Marin Municipal Water District (MMWD) and the Watershed Stewards Program/AmeriCorps (WSP). A passive integrated transponder (PIT) tag antenna was also in operation at that location to detect PIT-tagged coho smolts. In addition, smolt traps were operated on Olema Creek by National Park Service (NPS) staff and on San Geronimo Creek by staff and volunteers of the Salmon Protection and Watershed Network (SPAWN). A total of 7,373 coho smolts were captured at the RST and 10,643 (± 596) coho smolts were estimated to have emigrated past the trap. The previous fall 18,333 coho fry were estimated to reside upstream of the RST, yielding a winter survival rate of 58%. Coho smolt emigration from Olema Creek was estimated at 2,343 (± 476). On San Geronimo Creek, the coho emigration was estimated to be 62 (± 24) smolts. The total emigration from the watershed was estimated at 12,986 (± 763) coho smolts. A total of 814 steelhead smolts were captured at the RST and an estimated 2,699 (± 594) steelhead smolts emigrated past the trap. The steelhead emigration was the largest seen in at least three years. An estimated 1,855 (± 402) steelhead smolts emigrated from San Geronimo Creek. The Olema Creek smolt trap captured 98 steelhead smolts, but an emigration estimate was not calculated. A record number of Chinook salmon smolts were observed at the RST in 2015. A total of 2,005 Chinook smolts were captured and 3,376 (± 382) were estimated to have migrated past the trap. 1

Smolt Monitoring in the Lagunitas Creek Watershed 2015 INTRODUCTION Lagunitas Creek is a regionally important coastal stream for coho salmon (Oncorhychus kisutch) and steelhead trout (O. mykiss), although run sizes for both species are significantly reduced from historical numbers. Recent coho escapement estimates have averaged approximately 400 individuals, and available data suggest that steelhead runs are similar in size. Chinook salmon (O. tshawytscha) also spawn periodically in Lagunitas Creek and smolts have been observed in four of the last ten years. MMWD has conducted annual smolt surveys on Lagunitas Creek since 2006, as well as in 1983, 1984 and 1985. Summer and fall electrofishing surveys for juvenile (also called fry, 0+ or young-ofthe-year) coho salmon and steelhead trout were conducted in Lagunitas Creek in 1970, 1980, 1982 through 1988, 1990, and annually since 1993. Since 2012 coho fry captured during these surveys have been implanted with passive integrated transponder (PIT) tags. This represents one of the longest data records for juvenile salmonids in coastal streams of California. Surveys have been conducted cooperatively between MMWD, the California Department of Fish and Wildlife (CDFW), the National Park Service (NPS), the Marin Resource Conservation District, the Watershed Stewards Project, and the Salmon Protection and Watershed Network (SPAWN). Systematic coho adult spawner surveys began during the 1982-83 and 1983-84 spawning seasons, and have been conducted annually since 1995-96. Since the early 1980s, stream flows in Lagunitas Creek have been monitored daily by United States Geological Survey gages located in Samuel P. Taylor State Park and near Point Reyes Station. A separate gage is maintained by MMWD on San Geronimo Creek. Water temperature monitoring has been performed by MMWD since the early 1990s. Lagunitas Creek streambed conditions are monitored annually and salmonid habitat is quantified approximately every five years. NPS has monitored salmonid smolt emigration from Olema Creek since 2004, and smolt monitoring was conducted on a tributary to Olema Creek between 1998 and 2004. Smolt monitoring on San Geronimo Creek has been conducted annually since 2006. Smolts leaving San Geronimo Creek are sampled again at the Lagunitas Creek RST, so the Lagunitas Creek smolt estimates include San Geronimo Creek smolts. Since 2012, monitoring in Lagunitas Creek has been conducted as part of CDFW s and National Marine Fisheries Service s (NMFS ) Coastal Monitoring Program (CMP). Lagunitas Creek was established as a CMP life-cycle monitoring station, for the diversity strata north of the Golden Gate. The intent of this project is to continue monitoring for coho salmon and steelhead in the Lagunitas Creek watershed, consistent with methods described in the CMP. This project is being conducted in collaboration with NPS and SPAWN, which are conducting similar monitoring surveys in Olema and San Geronimo Creeks, respectively. Smolt monitoring in the Lagunitas Creek watershed is intended to answer the following questions: What are the trends in coho salmon and steelhead smolt abundance? What are salmonid overwinter survival rates and what factors influence those rates? Do survival rates differ between subwatersheds? 2

Smolt Monitoring in the Lagunitas Creek Watershed 2015 What are coho marine survival rates and how do these rates compare to other populations in the region? METHODS Lagunitas Creek Monitoring A rotary screw trap (RST) with a five-foot diameter drum was installed on March 18, 2015 in lower Lagunitas Creek, approximately 2.1 miles above the Highway 1 Bridge in Point Reyes Station. The trap was situated in a pool directly downstream of a small bedrock cascade, and was in the same location as has been used since 2006. The bedrock cascade concentrates enough flow to operate the RST in the otherwise low gradient reach of the creek. The trap was operated without interruption for a total monitoring period of 83 days. Trap function was visually inspected each day to ensure proper operation. The rotation speed of the trap drum was recorded daily. The trap was occasionally moved either toward or away from the cascade to maintain drum speeds of between three and eight revolutions per minute (RPM). The drum speed did not exceed this target but fell slightly below it on two days. Debris was removed from the trap drum and live box daily and was never abundant enough to prevent the drum from spinning. A ½ -mesh screen in the live box allowed the smallest fry to escape from larger fish. Small gaps in the live box were created behind this screen to allow fry to escape, although many chose not to. Plywood baffles were installed in front of the drum during the second week of monitoring to increase the drum rotation speed and improve trap efficiency. Each day, captured fish were removed from the trap and identified by species. Salmonid smolts and parr were checked for marks such as fin clips, visually inspected for signs of smoltification, measured, weighed, allowed to recover, and then released downstream of the point of capture. Coho smolts were scanned for passive integrated transponder (PIT) tags, implanted the previous fall. Scales were collected from a small subset of coho to determine age. Steelhead displaying characteristics intermediate between smolts and fry/parr (e.g. some loss of scales, some silver color, fading parr marks, etc.) were classified as transitional. These transitional steelhead could not be assumed to be emigrating and were not included in the smolt estimate. Coho were classified as fry, transitionals, or smolts based on the degree of smolt characteristic development. Young-of-the-year coho displaying smolt characteristics (e.g., a dark trailing edge on the caudal fin) were classified as smolts. All Chinook were assumed to be emigrating and classified as smolts. Downstream migrating fry of all species that were less than 70 mm long were tallied into fivemillimeter length bins and not weighed. Adult steelhead that appeared unspawned were released upstream of the bedrock cascade. Spawned steelhead (kelts) were immediately released off the trap. The proportion of migrating fish captured each week (trap efficiency) was determined by recapturing previously marked fish. Up to ten smolts per species per day were given a fin clip unique to the week and location of capture and were released approximately 500 m upstream. Some of these fish were subsequently recovered at the trap a second time and served as the basis for calculating trap efficiencies. Smolts smaller than 70 mm were not fin clipped. Marking of fish was coordinated 3

Smolt Monitoring in the Lagunitas Creek Watershed 2015 with a simultaneous trapping effort conducted by SPAWN on San Geronimo Creek, a tributary to Lagunitas Creek upstream of the MMWD trap location. Each monitoring effort used distinct fin clips to avoid duplicate marks from the two trapping locations (Table 1). Table 1. Marking schedule at Lagunitas and San Geronimo Creek smolt traps, 2014 Week Date Lower Lagunitas (MMWD) Mark Applied San Geronimo (SPAWN) Mark Applied 1 March 16 to March 22 lower caudal clip anal clip (LCAC) upper caudal clip anal clip (UCAC) 2 March 23 to March 29 dorsal & lower caudal clip (DLC) dorsal & upper caudal clip (DUC) 3 March 30 to April 5 lower caudal clip (LC) upper caudal clip (UC) 4 April 6 to April 12 lower caudal clip anal clip (LCAC) upper caudal clip anal clip (UCAC) 5 April 13 to April 19 dorsal & lower caudal clip (DLC) dorsal & upper caudal clip (DUC) 6 April 20 to April 26 lower caudal clip (LC) upper caudal clip (UC) 7 April 27 to May 3 lower caudal clip anal clip (LCAC) upper caudal clip anal clip (UCAC) 8 May 4 to May 10 dorsal & lower caudal clip (DLC) dorsal & upper caudal clip (DUC) 9 May 11 to May 17 lower caudal clip (LC) upper caudal clip (UC) 10 May 18 to May 24 lower caudal clip anal clip (LCAC) upper caudal clip anal clip (UCAC) 11 May 25 to May 31 dorsal & lower caudal clip (DLC) dorsal & upper caudal clip (DUC) 12 June 1 to June 7 No clip No clip In addition to monitoring smolts with the RST, MMWD operated a PIT tag antenna upstream of the RST through the winter and spring of 2014-15. PIT tag monitoring was intended to estimate overwinter survival rates and the factors influencing survival. In 2014 coho between 55 and 65 mm were implanted with 8 mm full duplex tags (FDX) and fish 65mm were implanted with 12 mm half duplex tags (HDX). Fish with HDX tags could be detected at the antenna as well as at the screw trap where all captured fish were scanned with a handheld PIT tag detector. The antennas cannot detect FDX tags, therefore detections of FDX tags could only occur at the screw trap. Olema Creek Monitoring A fyke/pipe trap was installed by NPS staff in mid-march and was operated for 80 days. The trap design was based on traps used by CDFW on the Noyo River (Gallagher 2000). The trap was checked daily, and no more than 30 coho smolts (or up to 50% of the catch that day) were anesthetized with carbon dioxide and marked with PIT tags. Marked smolts were released immediately after recovering from anesthetization at a predetermined site approximately 100 m or at least three habitat units above the trap site. After being measured, all recaptured smolts and unmarked smolts were released immediately in low velocity areas below the trap. Studies using similar methods of marking and tagging have demonstrated little marking mortality (Greis and Letcher 2002) and a study using the same trapping methodology on five northwestern California streams revealed that trap mortality was less than one percent for smolts and less than three percent for fry (Manning 2001). Salmonids were identified to species and life stage (fry, parr, smolt, or adult) prior to being measured. Fry are less than one year old and can be identified by the presence of distinct parr marks and small body size. Age 1+ steelhead and coho were separated into the following morphological 4

Smolt Monitoring in the Lagunitas Creek Watershed 2015 categories: parr (smaller size, parr marks present) or smolt (faint or absent parr marks, silver body, deciduous scales, black fin margins). A random sub-sample of up to ten coho smolts and steelhead parr and smolts were measured to the nearest millimeter (fork length), and weighed to the nearest 0.01 g using an electronic scale. All fish that were anesthetized and marked were also measured and weighed. Any adult steelhead encountered in the trap were released downstream immediately without being measured. Random sub-samples of ten coho fry and ten steelhead fry were measured daily and individuals greater than 40 mm were weighed to the nearest 0.01 g using an electronic scale. Ten individuals of each non-salmonid species were also randomly selected, measured, and weighed. Sub-samples were obtained by taking blind scoops out of the holding bucket with a small aquarium dip net. After processing, each fish is placed in an aerated recovery bucket, keeping larger sculpin in separate buckets to avoid predation on smaller fish. Fish in the recovery bucket were monitored to ensure sedated fish recover fully before being released. In addition to smolt trapping, a PIT tag antenna array was installed upstream of the Olema Creek smolt trap (Figure 1). For more detailed descriptions of smolt trapping methods, please refer to SOP (standard operating procedure) 3, SOP 4, SOP 6, SOP 9, and SOP 11 of the San Francisco Bay Area Network Salmonid Monitoring Protocol version 4.0 (Reichmuth et al. 2010). San Geronimo Creek Monitoring SPAWN staff installed a fyke/pipe trap near the mouth of San Geronimo Creek in mid-march and operated the trap through late May. The trap was checked daily following protocols similar to those used by MMWD and NPS. A subset of coho and steelhead smolts were marked (Table 1) and released approximately 100 m upstream of the trap to estimate trap efficiency. Data Analysis The efficiency of the Lagunitas Creek rotary screw trap and populations of coho and steelhead smolts were estimated using Darroch Analysis with Rank Reduction (DARR) 2.0.2 software (Bjorkstedt 2005, 2010) from mark-recapture data. The DARR 2.0.2 software was developed to allow populations of downstream migrants to be estimated using mark-recapture data, particularly in small watersheds. This program applies a set of algorithms to stratified mark-recapture data to produce an abundance estimate while defining the variability in capture probability and the distribution of recaptured individuals within the strata. RESULTS Lagunitas Creek Rotary Screw Trap The Lagunitas Creek RST captured 7,373 coho smolts, 2,005 Chinook smolts, and 814 steelhead smolts in 2015. Between mid-march and late-april, 12 adult steelhead were captured in the RST and were immediately released, unharmed. Non-salmonid fish species included the following native and non-native species, in order of abundance: Tomales roach (Hesperoleucus symmetricus symmetricus), three-spined stickleback (Gasterosteus aculeatus), sculpin spp. (Cottidae), bluegill (Lepomis macrochirus), black crappie (Pomoxis nigromaculatus), Sacramento sucker (Catostomous occidentalis), Pacific lamprey (Lampetra tridentata), golden shiner (Notemigonus crysoleucas), 5

Smolt Monitoring in the Lagunitas Creek Watershed 2015 white crappie (Pomoxis annularis), channel catfish (Ictalurus punctatus), green sunfish (Lepomis cyanellus), and the first known detection of a spotted bass (Micropterus punctulatus). An estimated 10,643 coho smolts emigrated from Lagunitas Creek during the monitoring period (Table 2). Young-of-the-year coho comprised 9% of the catch, and 13 age 2+ coho smolt were identified. The remaining coho catch was comprised of 1+ fish (1-2 years old) (Table 3). The highest estimated passage occurred during the week of May 4, with 2,098 coho smolts passing through and around the RST (Figure 2). The highest catch for a single day occurred on May 10 when 391 coho smolts were captured. The weekly trap efficiency for coho smolts varied from 25% to 77% (mean 63%) (Figure 3). Coho smolts averaged 104 mm fork length (FL) and weighed an average of 12 g. Table 2. Estimated smolt emigration from Lagunitas Creek, 2006-2015 Year Survey start date Survey end date Coho Steelhead Chinook Observed Estimated Observed Estimated Observed Estimated 2006 21 March 9 June 1,342 2007 15 March 30 May 611 2008 18 March 5 June 2,532 2009 10 March 5 June 3,150 2010 17 March 27 May 631 2011 1 April 20 May 1,684 2012 26 March 31 May 4,339 2013 19 March 7 June 4,942 2014 11 March 4 June 8,415 2015 19 March 9 June 7,373 5,946 (±1,570) 2,776 (±692) 6,101 (±780) 5,711 (±461) 2,129 (±480) 3,300 (±470) 8,315 (±1,372) 7,479 (±504) 15,055 (±1,974) 10,643 (±596) 308 475 449 646 651 829 251 684 448 814 6,949 (±6,133) 3,632 (±2,066) 1,134 (±259) 2,041 (±537) 3,867 (±1,419) 3,753 (±941) 1,991 (±1,252) 1,876 (±380) 1,720 (±478) 2,699 (±594) 237 504 775 2,445 0 0 0 0 0 0 0 0 0 0 0 0 1,229 2,005 2,011 (±241) 3,376 (±382) During the monitoring period, an estimated 2,699 steelhead smolts emigrated from Lagunitas Creek (Table 3). The peak of emigration occurred during the week of April 6, with an estimated 1,208 steelhead smolts (Figure 2). The highest catch for a single day occurred on April 6, when 106 steelhead smolts were captured. For the season the steelhead catch was comprised of 58% fry, 17% age 1+ and 26% age 2+ and older steelhead (Table 2). The weekly trap efficiency for steelhead smolts ranged from 20% to 71% (mean 35%) (Figure 3). Steelhead smolts averaged 169 mm FL and weighed 51 g on average. 6

Table 3. Salmonids captured in the Lagunitas Creek rotary screw trap by length and week, 2015. Smolt Monitoring in the Lagunitas Creek Watershed 2015 Coho Steelhead Week: 1 2 3 4 5 6 7 8 9 10 11 12 13 1 2 3 4 5 6 7 8 9 10 11 12 13 3/16 3/23 3/30 4/6 4/13 4/20 4/27 5/4 5/11 5/18 5/25 6/1 6/8 3/16 3/23 3/30 4/6 4/13 4/20 4/27 5/4 5/11 5/18 5/25 6/1 6/8 Dates 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 3/22 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 6/7 6/14 Length (mm) Age 0+ Age 0+ 20-24 0 0 0 0 0 0 0 0 0 0 0 0 0 7 2 1 1 6 61 23 1 0 1 0 0 0 103 25-29 0 0 1 0 0 0 0 0 0 0 0 0 0 127 293 63 13 25 292 170 8 5 4 0 0 0 ### 30-34 0 0 1 1 0 0 0 0 0 0 0 0 0 2 10 24 8 10 32 4 3 4 3 0 2 2 104 35-39 0 0 0 1 0 0 0 0 0 0 0 0 0 0 5 1 4 3 6 6 3 6 14 3 1 4 56 40-44 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 3 3 5 4 4 11 14 10 4 2 62 45-49 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1 4 4 4 4 24 41 16 2 3 103 50-54 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 2 0 3 0 5 8 23 12 9 5 67 55-59 0 0 6 3 0 2 0 0 0 0 0 0 0 Age 01+ 0 0 0 0 2 3 5 18 22 13 9 4 76 60-64 0 0 7 6 0 0 0 0 2 0 0 0 0 0 0 0 2 0 3 5 1 15 22 9 4 2 63 65-69 0 0 5 1 0 0 0 1 1 2 2 3 0 0 0 0 2 0 2 3 4 2 13 13 3 2 44 70-74 Age 0 1+ 0 1 1 0 0 0 1 1 3 7 4 1 0 0 0 4 0 0 0 1 4 5 3 2 1 20 75-79 0 1 0 0 1 1 0 1 3 17 16 20 9 1 0 0 6 2 6 3 3 2 2 0 2 0 27 80-84 0 0 0 3 1 0 1 3 3 9 24 37 8 0 0 0 8 1 1 6 2 2 0 0 0 2 22 85-89 0 4 0 8 4 10 5 7 14 20 18 43 14 1 0 0 15 4 7 5 1 5 1 0 1 0 40 90-94 1 7 4 18 18 35 21 40 51 51 40 28 4 0 0 4 19 9 11 11 3 5 3 0 0 0 65 95-99 1 8 6 29 54 51 52 67 90 54 33 22 2 0 1 4 16 5 5 6 5 7 4 0 0 0 53 100-104 4 5 6 66 71 87 102 98 91 62 18 5 0 0 1 2 13 1 16 4 3 5 3 0 2 0 50 105-109 2 7 21 43 82 73 84 83 50 49 9 5 1 0 1 1 10 7 12 8 4 9 3 1 0 0 56 110-114 3 8 16 42 68 48 61 46 46 29 13 0 2 1 5 2 12 8 3 7 2 10 2 2 0 0 54 115-119 2 1 16 34 32 21 19 19 30 20 21 3 0 0 1 1 6 4 5 8 6 7 2 2 1 0 43 120-124 1 3 7 24 11 6 6 5 15 22 39 10 3 1 1 0 2 4 2 8 5 8 3 0 0 0 34 125-129 0 5 4 7 8 0 1 1 2 6 21 9 3 0 1 1 4 2 4 3 3 5 2 0 0 1 26 130-134 0 0 1 4 2 1 0 1 0 0 3 6 0 0 2 0 0 3 2 0 0 1 4 0 0 0 12 135-139 0 2 0 0 0 0 0 0 0 0 2 0 0 0 4 2 3 6 3 2 2 2 0 0 0 0 24 140-144 0 1 Age 0 2+ 0 0 0 0 0Numbers 0 0in bold 0 indicate 0 0 0 3 2 10 3 10 9 2 2 0 0Age 2+ 0 0 41 145-149 1 0 0 0 0 1 0 0that scales 0 0were 0collected 0 0 1 5 5 16 5 7 7 2 2 1 0and 0 0 51 150-154 1 0 3 0 0 0 0 0 0 for 0 age 0analysis. 0 0 0 10 10 15 6 10 4 2 1 1 0older 0 0 59 155-159 0 1 1 1 0 0 0 0 0 0 0 0 0 1 8 5 19 12 15 7 2 3 3 0 0 0 75 160-164 0 0 1 0 0 0 0 0 0 0 0 0 0 1 14 11 26 20 13 6 0 1 0 0 0 0 92 165-169 0 0 0 0 0 0 0 0 0 0 0 0 0 1 15 9 13 11 6 5 0 1 0 0 0 0 61 170-174 0 0 0 0 0 0 0 0 0 0 0 0 0 4 17 16 18 9 10 3 0 0 0 0 0 0 77 175-179 0 0 0 0 0 0 0 0 0 0 0 0 0 7 10 14 7 7 1 3 1 0 1 0 0 0 51 180-184 0 0 0 0 0 0 0 0 0 0 0 0 0 3 9 16 9 3 3 1 1 1 0 0 0 0 46 185-189 0 0 0 0 0 0 0 0 0 0 0 0 0 2 12 9 9 4 2 1 0 0 0 1 0 0 40 190-194 0 0 0 0 0 0 0 0 0 0 0 0 0 1 10 8 6 2 1 0 0 0 0 0 0 0 28 195-199 0 0 0 0 0 0 0 0 0 0 0 0 0 3 6 5 4 1 0 1 0 0 0 0 0 0 20 200+ 6 19 14 16 1 2 1 0 0 0 0 0 0 59 Totals Age 0 0 0 21 14 1 3 0 6 10 32 49 107 36 9% 136 310 91 32 51 410 222 39 99 164 79 38 27 58% Age 1+ 14 49 81 278 352 332 352 367 389 313 217 88 11 91% 4 10 14 119 47 72 69 37 64 27 5 4 1 16% Age 2+ 2 6 6 2 2 1 0 0 0 0 0 0 0 1% 30 145 127 171 93 85 50 12 13 6 1 0 0 25% Chinook Week: 1 2 3 4 5 6 7 8 9 10 11 12 13 Length (mm) 30-34 0 0 0 0 0 1 0 0 0 0 0 0 0 1 35-39 1 0 0 2 0 0 0 0 0 0 0 0 0 3 40-44 0 2 1 2 0 1 0 0 0 0 0 0 0 6 45-49 2 4 3 22 3 0 0 0 0 0 0 0 0 34 50-54 0 5 5 77 10 3 5 1 1 1 1 0 0 109 55-59 0 2 9 77 10 23 17 10 8 5 4 1 0 166 60-64 0 2 7 20 11 23 34 31 43 37 13 7 2 230 65-69 0 1 7 7 6 25 34 46 108 79 44 29 4 390 70-74 0 0 2 2 0 7 13 45 103 123 87 44 6 432 75-79 0 0 0 0 0 2 5 30 96 109 82 53 9 386 80-84 0 0 0 0 0 0 3 13 25 33 35 34 6 149 85-89 0 0 0 0 0 0 2 0 5 4 5 3 2 21 90-94 0 0 0 0 0 0 0 0 2 0 1 0 0 3 95-99 0 0 0 0 0 0 0 0 0 0 0 0 0 0 100-104 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Totals 3 16 34 209 40 85 113 176 391 391 272 171 29 7

Smolt Monitoring in the Lagunitas Creek Watershed 2015 A record 3,376 Chinook smolts emigrated from Lagunitas Creek, with a peak emigration of 842 during the week of May 18. The weekly trap efficiency for Chinook ranged from 47% to 73% (mean 60%) (Figure 3). Chinook smolts were 74 mm in length on average. Chinook that were weighed (those at least 70 mm long), averaged 76 mm in length and weighed 4.8 g. Lagunitas Creek PIT Tag Antenna During the summer and early fall of 2014 441 coho fry were implanted with PIT Tags. The vast majority (89%) of these were in Lagunitas Creek because very few coho fry were found in San Geronimo Creek and Devil s Gulch (Table 4). The PIT tag antenna detected 62 of the 326 HDXtagged coho (19%). The RST captured 60 HDX-tagged coho (18%) and 23 FDX-tagged coho (20%). As noted above, FDX tags cannot be detected by the antenna. The detection rates were very similar for HDX and FDX tags (22% and 20%, respectively). Detection rates were also similar for fish from Lagunitas Creek and Devil s Gulch (Table 4). Five coho smolts were documented moving between Lagunitas Creek and Olema Creek. Two smolts that were originally tagged in Olema Creek in 2014 and two others tagged in 2015 were captured in the RST and/or detected at the Lagunitas PIT tag antenna. A fifth coho was tagged in Lagunitas Creek in the fall of 2014 and then detected at the Olema Creek PIT tag antenna on April 18, 2015. Table 4. PIT tagged coho detections Tag Type HDX FDX Tagging Location Fish Tagged in 2014 2015 Antenna Detections Antenna Detection Rate 2015 RST Detections Total Detections Total Detection Rate Lagunitas Creek 311 58 19% 57 70 23% San Geronimo Cr. 2 0 0% 0 0 0% Devil's Gulch 13 3 23% 3 3 23% All 326 62 19% 60 73 22% Lagunitas Creek 80 NA NA 17 17 21% San Geronimo Cr. 0 NA NA 0 0 NA Devil's Gulch 35 NA NA 6 6 17% All 115 NA NA 23 23 20% All All 441 62 14% 83 96 22% Olema Creek Monitoring The Olema Creek smolt trap captured 658 coho smolts and an estimated 2.343 (±476) coho emigrated from the creek. This was roughly average for the period of smolt monitoring, which began in 2004. The peak of the migration occurred during the week of April 20. Coho smolts averaged 111 mm in length. Five coho smolts PIT-tagged in 2014 were detected at the Olema Creek PIT tag antenna. The Olema Creek smolt trap captured 98 steelhead smolts but trap efficiency for steelhead wasn t calculated and total emigration wasn t estimated. Steelhead smolts averaged 150 mm in length. 8

Smolt Monitoring in the Lagunitas Creek Watershed 2015 San Geronimo Creek Monitoring The San Geronimo Creek smolt trap captured 18 coho and 187 steelhead smolts. An estimated 62 (± 24) coho smolts and 1,855 (± 402) steelhead smolts emigrated from San Geronimo Creek. Coho from San Geronimo Creek made up 0.5% of the total watershed emigration and San Geronimo Creek steelhead contributed 69% of the total emigration. The peak of the coho migration occurred during the week of April 20 and the steelhead migration peaked during the week of April 6. DISCUSSION Sampling conditions and effects on estimated smolt abundance and emigration timing The smolt trap was installed on March 18, which is a fairly typical start date, and catches of both coho and steelhead were fairly low during the first few days. Only 93 coho were estimated to have emigrated during the first week (0.9% of the total emigration). Only one PIT-tagged coho was detected by the antenna prior to trap installation, which provides further evidence that few coho emigrated earlier. Steelhead smolt numbers increased after the first three days, and 186 (6.9% of the total emigration) emigrated that week. The trapping period appears to have covered the vast majority of the steelhead emigration. Little rain fell during the monitoring period and flows remained below 30 cfs. Maximum trap rotation speeds never exceeded the target of eight revolutions per minute (RPM) and minimum speeds fell slightly below the target of three RPM on only two days. The timing of the coho emigration was somewhat unusual this year in that there were two peaks, separated by two weeks (Figure 4). Increases in daily coho catches tended to follow increases in water temperature (Figure 5), as has been observed in previous years. Specifically, increases in water temperature of more than 0.7 F per day were generally followed by increases in coho smolts within two days. The sharp spike in the coho catch on April 26 may also have been associated with a rain event that briefly increased stream flows. Coho marked in San Geronimo Creek took between five and 49 days to reach the Lagunitas trap (median = 12 days). There was no apparent relationship between migration speed and date or stream flow. The steelhead emigration this year was defined by an abrupt spike on April 8 (Figure 6), the day after a brief storm. This peak was somewhat later than emigration peaks in recent years but within the typical peak migration period for steelhead. Steelhead did not appear to be influenced by increases in water temperature, but are more responsive to changes in stream flow. The efficiency of the RST at capturing coho was the highest yet observed, at 69%. Relatively low stream flows allowed us to install plywood baffles during the second week of trap operation, which directed smolts into the cone and increased the cone rotation speed. In future years we may install the baffles at the start of trapping, if flows allow. For coho, the efficiency of the RST increased for the first four weeks, and then remained high (Figure 3). For Chinook, the trap efficiency couldn t be determined for the first five weeks because the smolts were too small for fin clipping. Once smolts had grown large enough to mark (>70 mm), our efficiency estimates were 9

Smolt Monitoring in the Lagunitas Creek Watershed 2015 generally high, but may have been influenced by our difficulty seeing some clips on these small fish. Clips on the small anal fin were the most difficult to see, and our apparent recapture rate for these fish was less than 50%. Dorsal and caudal fin clips were easier to see, and 62% of Chinook with both of these marks were observed. The observed recapture rate for Chinook with only caudal clips, which may have included fish with other marks that weren t seen, was 74%. The effect on the emigration estimate of missing secondary fin clips was minimal, since the peak of outmigration was spread over three weeks and included all of the fin clip types. Missing marked fish entirely, which could inflate our estimate, was likely rare since caudal fin clips are fairly obvious. A larger source of bias was our practice of clipping only larger Chinook smolts, which are stronger and presumably better able to avoid the trap. This likely underestimated the trap efficiency and overestimated the size of the Chinook emigration. The trap efficiency for steelhead was above average compared with previous years, but much lower than for coho and Chinook. This may be due to larger fishes better ability to avoid the trap or to a behavioral mechanism such as swimming deeper. The DARR algorithm calculated a trap efficiency of 71% during the second week of monitoring (Figure 3), although only 45% of the steelhead marked that week were recaptured. The reason for this discrepancy is unknown, but since the steelhead catch that week was low, the anomalous trap efficiency had little effect on the emigration estimate. Age and size composition of salmonids Young-of-the-year (YOY) coho grew to the size of small age-1+ coho by early May, and at that point scale analysis was essential for defining the size break between age classes (Table 2). Differentiating non-migratory coho fry from smolts was based on appearance more than size, and some fish classified as YOY based on length were also classified as smolts based on appearance. Coho smolts generally do not begin smoltification until they reach a size of 75-100 mm (Sandercock 1991). Of the 212 coho in that size range that were classified as likely YOY, 102 (48%) appeared to be smolting. These YOY smolts, while interesting, made up only 1.4% of the total smolt catch. Coho smolts at the Lagunitas Creek RST were unusually small in 2015, which fits with the relatively large population (Figure 7). PIT-tagged coho grew an average of 33 mm between tagging in 2014 and recapture in 2015, which was similar to the slow growth seen in 2013-14, despite a smaller population. The reason for this appears related to the near-absence of coho in the tributaries, where winter growth tends to be faster. Coho in San Geronimo Creek and Devil s Gulch in 2012-13 and 2013-14 grew an extra 3-8 mm prior to emigration compared to coho in Lagunitas Creek. Approximately 98% of steelhead smolts appeared to be age 2+ or older. The occurrence of age 3+ and older age classes is difficult to determine due to the lack of obvious length modes greater than 175 mm. We captured one 226 mm steelhead that lacked smolt characteristics and was likely a resident rainbow trout. Steelhead showing at least some signs of smoltification included 70 transitional and 812 fully smolted fish. These few transitional steelhead, which may or may not have been emigrating, did not appear to be a significant source of error in the Lagunitas Creek steelhead smolt estimate. 10

Smolt Monitoring in the Lagunitas Creek Watershed 2015 Smolt abundance trends and implications for winter survival The 2015 coho emigration was the second-largest migration on record for Lagunitas Creek and the watershed as a whole (Figure 8). The total emigration from the watershed was estimated at 12,986 (± 763) coho smolts. This was the fourth straight year of Lagunitas Creek smolt estimates exceeding the hypothetical carrying capacity of 7,000 coho, proposed by Stillwater Sciences in 2008 (Figure 9). Last year we hypothesized that the record coho emigration was related to dry conditions that prevented coho from migrating downstream, where habitat is limited. The 2015 smolt cohort, however, originated almost entirely in Lagunitas Creek, and could migrate downstream at any time. The most likely explanation for increases in winter survival since 2008 is an improvement in winter habitat conditions. Some of that improvement may be the result of MMWD s installation of eight large wood structures in upper Lagunitas Creek, although we have no direct observations of juvenile coho using these structures during the winter. Improvements in habitat conditions may also be related to continued recovery following the flood of 2005-06. Alternative explanations for improved coho winter survival in the last four years could include flow conditions, water temperature or quality, or predator populations, but no correlation was found with these factors. Coho smolt abundance (as opposed to the survival rate) is negatively correlated with the number of days of stream flows greater than 500 cfs. However, whether coho benefit from dry winters is still an open question. While flows during the last four years were below average, they were similar to flows in 2008-09, a year with fewer than average coho smolts. Coho abundance was also found to be negatively correlated with steelhead smolt abundance, although a causal mechanism is not clear. One possibility is that high numbers of one species may encourage the other to emigrate prior to the smolt monitoring period. Given the size difference between species, it s possible that abundant steelhead may encourage coho to migrate early. Steelhead smolt estimates have varied far less than coho estimates since 2007, so the factors influencing that limited variability are harder to identify. Steelhead smolts are positively correlated with YOY steelhead abundance 1.5-years earlier (primarily the same cohort) and negatively correlated with coho fry from the previous summer. As stated earlier, the causal mechanism behind the coho and steelhead correlation is not clear. Age 1+ steelhead may prey on newly-emerged coho fry, or abundant coho fry may encourage steelhead smolts to emigrate earlier. The only year when an early surge of steelhead smolts was documented was 2009, and at the time juvenile coho numbers were low. This argues against coho influencing steelhead emigration, so a more likely interaction is one of steelhead preying on coho when they are very small. The factors that influence steelhead survival between YOY and smolts are still being investigated. Looking ahead to next year, the coho smolt emigration will be determined by juvenile abundance this fall and potentially by habitat-altering floods resulting from the forecasted El Niño. The steelhead smolt emigration may be larger, based on large numbers of YOY steelhead in Lagunitas Creek in 2014, although as stated previously, the factors influencing their survival are unknown. The Chinook smolt emigration will depend on the number of adults spawning this fall. No Chinook smolts were documented in 2012 or 2013, so a Chinook run would depend on strays. In short, little data is available to predict the size of the smolt emigration for any of the salmonids in 2016. 11

Smolt Monitoring in the Lagunitas Creek Watershed 2015 REFERENCES Bjorkstedt, E. P. 2005. DARR 2.0: updated software for estimating abundance from stratified markrecapture data. NOAA Technical Memorandum NMFS-SWFSC-368. Bjorkstedt, E. P. 2010. DARR 2.0.2: DARR for R. http://swfsc.noaa.gov/textblock.aspx?division=fed&id=3346 Gallagher, S. P. 2000. Results of the 2000 Steelhead (Oncorhynchus mykiss) Fyke Trapping and Stream Resident Population Estimations and Predictions for the Noyo River, California with Comparison to Some Historic Information. California State Department of Fish and Game, Steelhead Research and Monitoring Program, Fort Bragg, CA. Report FB-03-Draft, September 2000. 75pp. Greis, G., and B. H. Letcher. 2002. Tag retention and survival of age-0 Atlantic salmon following surgical implanatation with passive integrated transponder tags. North American Journal of Fisheries Management 22: 219-222. Hearn, A., P. Brown, M. Felger and T. Steiner. 2014. Coho and Steelhead Population Monitoring in San Geronimo Creek and its Tributaries, 2013 and 2014. Manning, D.J. 2001. Carrying capacity and limiting habitat analysis for coho salmon (Oncorhynchus kisutch) in streams of northwestern California. Humboldt State University: Masters Thesis, Arcata, California. Moyle, P.B. 2002. Inland Fishes of California, revised and expanded. University of California Press. Berkeley, CA. Reichmuth, M., B.J. Ketcham, D. Fong, S. Carlisle, E. Brown, and M. Koenen. 2010. Salmonid Monitoring Protocol for the San Francisco Bay Area Network: narrative and appendices version 4.0. Natural Resource Report NPS/SFAN/NRR 2010/202. National Park Service, Fort Collins, Colorado. Sandercock, F. K. 1991. Life History of Coho Salmon (Oncorhynchus kisutch). Pages 397-445 in Groot, C. and L. Margolis (eds). Pacific Salmon Life Histories. University of British Columbia Press, Vancouver, British Columbia. Shapovalov, L., and A. C. Taft. 1954. The life histories of the steelhead rainbow trout (Salmo gairdneri gairdneri) and silver salmon (Oncorhynchus kisutch) with special reference to Waddell Creek, California, and recommendations regarding their management. California Department of Fish and Game Fish Bulletin 98. Stillwater Sciences. 2008. Lagunitas limiting factors analysis; limiting factors for coho salmon and steelhead. Final Report. Prepared by Stillwater Sciences, Berkeley, California for Marin Resource Conservation District, Point Reyes Station, California. 12

Tomales Bay Highway 1 Pt Reyes-Petaluma Rd Lagunitas Creek Watershed Lagunitas Creek rotary screw trap and PIT tag antenna Nicasio Reservoir Pacific Ocean Watershed Boundary San Francisco Bay Nicasio C r Platf orm Bridge Rd Nicasio Valley Rd Olema Creek smolt trap Olema Creek PIT tag antenna Sir Francis Drake Blvd Olema Creek Golden Gate National Recreation Area Lagun ita s Devil's Gulch Samuel P. Taylor State Park Creek San Geronimo Creek smolt trap San Geronimo Creek Larsen Cr Highway 1 Marin County Open Space Woodacre Cr. Kent Lake Marin Municipal Water District Legend Marin Municipal Water District Highway 1 Marin County Open Space State Lands Federal Lands 0 0.5 1 2 3 4 5 6 Kilometers Figure 1. Smolt monitoring locations in the Lagunitas Creek watershed. 13

2200 Coho Estimate 55 2000 Coho Observations Steelhead Estimate 50 Number of Smolts 1800 1600 1400 1200 1000 800 600 400 Steelhead Observations Chinook Estimate Chinook Observations Flow 45 40 35 30 25 20 15 10 Mean Daily Stream Flow at the Point Reyes USGS Gage (cfs) 200 5 0 0 Figure 2. Lagunitas Creek smolt emigration and stream flow. 14

100% Trap Efficiency for Coho 50 90% Trap Efficiency for Chinook Trap Efficiency for Steelhead 45 Trap Efficiency 80% 70% 60% 50% 40% 30% 20% Flow 40 35 30 25 20 15 10 Mean Daily Stream Flow at the Point Reyes USGS Gage (cfs) 10% 5 0% 0 Figure 3. Weekly trap efficiency and Lagunitas Creek stream flow. 15

500 450 400 Smoothed Mean 2006-2015 2015 2014 2013 2012 Coho Smolts Captured (New and Recaptures) 350 300 250 200 150 100 50 0 Date Figure 4. Daily coho smolt captures at the Lagunitas Creek smolt trap, 2012-2015 16

400 350 Coho smolts Water temperature Temperature increase > 0.7 deg 59.0 58.0 300 57.0 Coho smolt catch 250 200 150 56.0 55.0 54.0 Water Temperature ( F) 100 53.0 50 52.0 0 51.0 Date Figure 5. Coho smolt captures and water temperature, 2015. 17

120 110 100 Smoothed mean (2006-2015) 2015 2014 2013 2012 Steelhead Smolts Captured (New and Recaptures) 90 80 70 60 50 40 30 20 10 0 Figure 6. Daily steelhead smolt captures at the Lagunitas Creek smolt trap, 2012-2015. Date 18

140 135 130 125 Mean smolt length (mm) 120 115 110 105 100 2015 R² = 0.7769 2014 95 90-2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 Smolt estimate Figure 7. Coho emigration estimates and smolt lengths at the Lagunitas Creek smolt trap. 19

35,000 30,000 Coho population recovery target Olema Creek Coho Lagunitas Creek Coho Lagunitas Creek Steelhead Lagunitas Creek Chinook 25,000 Smolt Estimate 20,000 15,000 10,000 5,000-2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Figure 8. Lagunitas Creek smolt emigration estimates. Note: The coho recovery target assumes an ocean survival rate of at least 8%, resulting in 2,600 adult returns. 20

18,000 16,000 14,000 Coho smolts 2012-2015 Coho smolts 2006-2011 Log. (Coho smolts 2006-2011) 2013-14 12,000 2014-15 Coho Smolts 10,000 8,000 2011-12 2012-13 6,000 4,000 2,000 - - 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 Juvenile Coho Figure 9. Juvenile coho abundance and subsequent smolt abundance. Olema Creek data not included. 21