Reconnaissance of Salmonid Redd Abundance and Juvenile Salmonid Spatial Structure in the Smith River with Emphasis on Coho Salmon (Oncorhynchus kisutch) MARCH 31 2014 FINAL REPORT TO THE CALIFORNIA DEPARTMENT OF FISH AND WILDLIFE FISHERIES RESTORATION GRANTS PROGRAM GRANTEE AGREEMENT: P1010504 ON BEHALF OF THE SMITH RIVER ALLIANCE AND THE CALIFORNIA DEPARTMENT OF FISH AND WILDLIFE ANADROMOUS FISHERIES RESOURCE AND MONITORING PROGRAM Reconnaissance of salmonid redd abundance and juvenile salmonid spatial structure in the Smith River with emphasis on Coho Salmon (Oncorhynchus kisutch) Final report to the California Department of Fish and Wildlife Fisheries Restoration Grants Program Grantee agreement: P1010504 Prepared by: Justin M. Garwood1 and Monty D. Larson2 Abstract We investigated two essential population viability metrics of salmonids in the Smith River basin (Oregon and California), with ESA listed coho salmon as the focal species. First, we monitored adult salmonid escapement and distribution for two consecutive years (2011‐2013) using live fish, carcass, and redd counts as defined in California’s Coastal Salmonid Monitoring Plan. Second, we developed a new protocol for monitoring the summer spatial structure of juvenile salmonids and adult coastal cutthroat trout during the summers of 2012 and 2013 using multiple‐pass snorkel surveys in an occupancy modeling framework. To implement these studies, we developed two unbiased sample frames tailored specifically to identify stage‐based coho salmon habitats. We compiled empirical species distribution data and physical stream attributes into a geographic information system model that was later verified in the field. We divided each sample frame into survey reaches resulting in 161.8 kilometers of stream habitat (68 reaches, 30 sub‐reaches) for the adult sample frame and 298.1 kilometers (126 reaches, 40 sub‐reaches) for the juvenile spatial structure sample frame. We completed 388 and 398 spawning ground surveys throughout the Smith River basin for the 2011‐2012 and 2012‐2013 seasons, respectively. We made 389 and 129 live adult coho salmon observations in 2011‐2012 and 2012‐2013, respectively. All live coho salmon observations occurred in Mill Creek except one individual was observed in the Rowdy Creek sub‐basin during the 2011‐2012 season. We recovered 82 and 24 coho salmon carcasses in 2011‐2012 and 2012‐2013, respectively. All coho salmon carcasses were observed in Mill Creek except one individual in Morrison Creek in 2012‐2013. We were able to verify 90 and 25 individual coho salmon redds for the 2011‐2012 and 2012‐2013 seasons, respectively. All verified redds were found in the upper Mill Creek sub‐basin. Since our coho salmon observations were almost exclusively clustered in the Mill Creek, we determined that our redd population estimates for the whole sample frame were biased high and unreliable based largely on large between‐reach error estimates. However, Chinook salmon and steelhead estimates were determined for the sample frame since these species were more evenly distributed throughout the basin. We estimated total coho salmon redd abundance in the Mill Creek sub‐basin as 482 (95% CI: 464 ‐ 501) and 227 (95% CI: 217 ‐ 236) redds for 2011‐2012 and 2012‐2013 seasons, respectively. Hatchery origin salmonids were observed spawning throughout sampling frame with the mean hatchery proportion of Chinook salmon carcasses ranging from 7.5% to 22.7% and mean hatchery proportion of live steelhead ranging from 8.5% to 11.1%. Our results highlight the limitations of spawning ground surveys when your target species is rare and narrowly distributed while other species are common and widespread. We used multi‐scaled occupancy models to estimate the probability of salmonid occupancy at the sample reach and at the sample unit (within reach) simultaneously while accounting for species detection probabilities. In 2012 we detected juvenile coho salmon in 17 out of 41 surveyed reaches in five portions of the watershed. Eleven (65%) of the reaches with coho salmon were non‐natal rearing areas. Estimated large‐ scale occupancy of juvenile coho salmon equaled 0.42 (SE=0.08) while estimated small‐scale occupancy equaled 0.68 (SE=0.01) resulting in a proportion of total area occupied (PAO) of 0.29. In 2013 we detected juvenile coho salmon in 23 out of 60 surveyed reaches in four portions of the watershed. Nine (39%) of the reaches with coho salmon were non‐natal rearing areas. Estimated large‐scale occupancy of juvenile coho salmon equaled 0.39 (SE=0.06) while estimated small‐scale occupancy equaled 0.60 (SE=0.02) resulting in a PAO of 0.23. All other salmonid species had much wider spatial distributions with reach‐level occupancy estimates ranging from 0.71 (SE=0.07) to 1.00 depending on species and age class. Based on our surveys, we found the Smith River coho salmon population had two remote inland sub‐populations and a core coastal plain sub‐population. Coho salmon juveniles used a variety of non‐natal rearing habitats highlighting diversity in life‐history and complementary resource needs. 1California Department of Fish and Wildlife, 5341 Ericson Way, Arcata, CA 95521, [email protected] 2Smith River Alliance, PO Box 2129 Crescent City, CA 95531 Table of Contents Introduction ......................................................................................................................................................................... 1  Monitoring Approach ..................................................................................................................................................... 1  Materials and Methods ................................................................................................................................................... 3  Study Area .......................................................................................................................................................................... 3  Smith River Salmonid Populations ........................................................................................................................... 3  Sample Frame Development ....................................................................................................................................... 5  Sample Unit Development ........................................................................................................................................... 7  Spawning Ground Survey Frame .............................................................................................................................. 7  Spatial Structure Survey Frame ................................................................................................................................. 8  Sample Draw Procedure ............................................................................................................................................... 8  Sampling Rate ................................................................................................................................................................... 8  Field Methods .................................................................................................................................................................... 9  Spawning Ground Survey Statistical Methods ................................................................................................... 11  Spatial Structure Statistical Methods .................................................................................................................... 14  Database and Data Storage ........................................................................................................................................ 15  Spawning Ground Survey Results ........................................................................................................................... 15  2011‐2012 Spawning Ground Survey Conditions and Effort ...................................................................... 15  2011‐2012 GRTS Spawning Ground Surveys .................................................................................................... 18  2012‐2013 Spawning Ground Survey Conditions and Effort ...................................................................... 31  2012‐2013 GRTS Spawning Ground Surveys .................................................................................................... 31  Mill Creek Spawner Survey Census ........................................................................................................................ 35  Coastal Cutthroat Trout and Pacific Lamprey ................................................................................................... 38  Spatial Structure Survey Results ............................................................................................................................. 41  2012 Sampling Effort and Coho Salmon Occupancy ....................................................................................... 41  2013 Sampling Effort and Coho Salmon Occupancy ....................................................................................... 41  Occupancy of Other Salmonid Species .................................................................................................................. 49  Discussion ............................................................................................................................................................................ 49  Spawning Ground Surveys ......................................................................................................................................... 49  Spatial Structure Surveys ........................................................................................................................................... 52  Restoration Recommendations .................................................................................................................................... 55  Literature Cited................................................................................................................................................................. 56  i List of Tables Table 1. Estimated stream kilometers available to anadromous salmonids by species and the total stream kilometers included in both the spawning survey and summer spatial structure sample frames in the Smith River basin, Del Norte County, California and Curry County, Oregon ..................................................................................... 7  Table 2. Summary statistics of spawning ground reach survey effort and reach survey availability based on flow conditions for the winter of 2011‐2012, Smith River basin, Del Norte County, CA. Surveys occurred from November 1, 2011 to February 28, 2012 ........................................................................................................................................... 16  Table 3. Summary of live adult and salmonid carcasses observed by species and reach from November 1, 2011 to February 28, 2012, Smith River basin, Del Norte County, CA. Live salmonid totals do not represent individual fish observations since live individuals could be observed over multiple survey periods. All observed salmonid carcasses were uniquely tagged with numbered jaw tags so totals represent individual carcass observations .................................................................................................................................................................................... 19  Table 4. Descriptive statistics for observation date of live fish, observation date of known species redds, observation date of carcasses, and carcass fork lengths for the 2011‐2012 spawning ground survey season in the Smith River basin, Del Norte County, CA. Totals include data from GRTS drawn reaches and the Mill Creek Lifecycle Monitoring Station census. .................................................................................................................................................... 21  Table 5. Proportion of observed hatchery‐origin salmonids summarized by species, observation type, and major sub‐basin, during the winter 2011‐2012 spawning ground surveys conducted throughout the Smith River basin, Del Norte County, CA.......................................................................................................................................................... 26  Table 6. Summary of total observed redds separated by reach and species for the winter of 2011‐2012, Smith River basin, Del Norte County, CA. Surveys occurred from November 1, 2011 to February 28, 2012. Location codes with shaded cells were not GRTS drawn for the annual survey but indicate they were surveyed to complete the annual upper Mill Creek Life Cycle Monitoring Station census .................................................................... 28  Table 7. Confusion matrix, statistics, and number of redds by species for the 2011‐2012 and 2012‐2013 spawning ground survey seasons in the Smith River basin, Del Norte County, CA. Redds were predicted with the kNN algorithm using known species redds and live fish locations as a training dataset. Model performance was assessed using a leave one out cross validation. Data are from GRTS drawn reaches and the additional Mill Creek Life Cycle Monitoring Station census reaches ............................................................................................................ 30    Table 8. Estimated total number of redds by species in the Smith River spawner survey sample frame for the winter of 2011‐2012. Components of estimated variance are broken down to the estimation of the number of redds within the reach and estimation of redds by expanding the sample reaches to the entire frame (sample error). ................................................................................................................................................................................................................. 31  Table 9. Summary statistics of spawning ground reach survey effort and reach survey availability based on flow conditions for the winter of 2012‐2013, Smith River basin, Del Norte County, CA. Surveys occurred from November 6, 2012 to February 27, 2013. Location codes with shaded cells were not GRTS drawn for the annual survey but indicate they were surveyed to complete the annual upper Mill Creek census ......................... 32  Table 10. Summary of live adult and salmonid carcasses observed by species and reach from November 6, 2012 to February 27, 2013, Smith River basin, Del Norte County, CA. Live salmonid totals do not represent individual fish observations since live individuals could be observed over multiple survey periods. All observed salmonid carcasses were uniquely tagged with numbered jaw tags so totals represent individual carcass observations .................................................................................................................................................................................... 33  ii List of Tables continued Table 11. Descriptive statistics for observation date of live fish, observation date of known species redds, observation date of carcasses, and carcass fork lengths for the 2012‐2013 spawning ground survey season in the Smith River basin, Del Norte County, CA. Totals include data from GRTS drawn reaches and the Mill Creek Life Cycle Monitoring Station census. .................................................................................................................................................. 34  Table 12. Proportion of observed hatchery‐origin salmonids summarized by species, observation type, and major sub‐basin, during the winter 2012‐2013 spawning ground surveys conducted throughout the Smith River basin, Del Norte County, CA.......................................................................................................................................................... 36  Table 13. Summary of total observed redds separated by reach and species for the winter of 2012‐2013, Smith River basin, Del Norte County, CA. Surveys occurred from November 6, 2012 to February 27, 2013. Location codes with shaded cells were not GRTS drawn for the annual survey but indicate they were surveyed to complete the annual upper Mill Creek Life Cycle Monitoring Station census.. ............................................................ 37  Table 14. Estimated total number of redds by species in the Smith River sample frame for the winter of 2012‐ 2013. Components of estimated variance are broken down into the estimation of the number of redds within the reach and estimation of redds by expanding the sample reaches to the entire frame (sample error)........... 38  Table 15. Estimated total number of redds by species within the Mill Creek Life Cycle Monitoring Station for the winter of 2011‐2012. Components of estimated variance are broken down to the estimation of the number of redds within the reach. There is no between‐reach variation since all reaches were surveyed. ....... 40  Table 16. Estimated total number of redds by species within the Mill Creek Life Cycle Monitoring Station for the winter of 2012‐2013. Components of estimated variance are broken down into the estimation of the number of redds within the reach. There is no between‐reach variation since all reaches were surveyed. ....... 40  Table 17. Spatial structure survey effort during the summers of 2012 and 2013, Smith River Basin, California and Oregon. ...................................................................................................................................................................................................... 41  Table 18. Summary statistics of coho salmon occupancy and relative abundance based on snorkel surveys occurring in 41 GRTS drawn reaches during the summer of 2012, Smith River Basin, California and Oregon. 42  Table 19. Occupancy estimates, proportion of area occupied, and relative count densities if salmonids for the summer spatial structure survey during 2012 and 2013, Smith River basin, Oregon and California. ................... 43  Table 20. Summary statistics of coho salmon occupancy and relative abundance based on snorkel surveys occurring in 60 GRTS drawn reaches during the summer of 2013, Smith River Basin, California and Oregon. 44  iii List of Figures Figure 1. Map of the Smith River Basin, Del Norte County (California) and Curry County (Oregon). Stream lines indicate potential anadromous salmonid stream habitat based on this studies sample frame development process. .................................................................................................................................................................................... 4  Figure 2. Ordered steps for developing an unbiased population‐level sampling frame for salmonid spawning ground surveys; adapted from Garwood and Ricker (2011). ...................................................................................................... 6  Figure 3. Spawning ground survey effort and timing in the Smith River basin (Del Norte County, CA) as it relates to mean daily river discharge. Panel A represents the 2011‐2012 survey and panel B represents the 2012‐2013 survey. The dashed red line represents the maximum discharge (16,000 cubic feet per second) where spawner surveys could be safely completed in smaller streams without being impaired by decreased water clarity. ................................................................................................................................................................................................... 17  Figure 4. Number live salmonids, identified to species and survey period, observed during spawner surveys occurring over two winters in the Smith River basin, Del Norte County, CA. Panel A represents the 2011‐2012 survey and panel B represents the 2012‐2013 survey. ............................................................................................................... 20  Figure 5. Map showing annual survey reaches, distribution of observed adult Chinook salmon, and verified Chinook salmon redds, Smith River Basin, Del Norte County, CA. Note: redd location symbols are displayed above fish observation symbols and may obscure fish observations in reaches with numerous verified Chinook salmon redds................................................................................................................................................................................. 22  Figure 6. Map showing annual survey reaches, distribution of observed adult coho salmon, and verified coho salmon redds, Smith River Basin, Del Norte County, CA. Note: redd location symbols are displayed above fish observation symbols and may obscure fish observations in reaches containing high densities of verified coho salmon redds. .................................................................................................................................................................................................. 23  Figure 7. Map showing annual survey reaches, distribution of observed adult steelhead, and verified steelhead redds, Smith River Basin, Del Norte County, CA. Note: redd location symbols are displayed above fish observation symbols and may obscure fish observations in reaches containing high densities of verified steelhead redds. ............................................................................................................................................................................................. 24  Figure 8. Number of uniquely tagged salmonid carcasses, identified by species and survey period, during spawner surveys occurring over two winters in the Smith River basin, Del Norte County, CA. Panel A represents the 2011‐2012 survey and panel B represents the 2012‐2013 survey. ....................................................... 25  Figure 9. Map showing annual survey reaches and the distribution of observed adipose fin clipped adult hatchery Steelhead, adipose or left ventral fin clipped adult Chinook Salmon, hatchery Chinook salmon constructing redds, and a maxillary clipped coho salmon from the Klamath River; observed in the Smith River Basin, Del Norte County, CA. ..................................................................................................................................................................... 27  Figure 10. Number of individual salmonid redds observed by survey period during spawner surveys occurring over two winters in the Smith River basin, Del Norte County, CA. Panel A represents the 2011‐2012 survey and panel B represents the 2012‐2013 survey. Line plots represent percentages of redds identified to species by survey period through direct observations of live fish actively building or guarding individual redds. .................................................................................................................................................................................................................. 29  iv List of Figures continued Figure 11. Map showing the distribution of observed adult coho salmon, and verified coho salmon redds in the Mill Creek spawning ground census (Life Cycle Monitoring Station) area during the winters of 2011‐2012 and 2012‐2013, Smith River Basin, Del Norte County, CA .......................................................................................................... 39  Figure 12. Map showing annual spatial structure survey reaches and the spatial distribution of pools containing juvenile coho salmon from 2012 and 2013, Smith River Basin, California and Oregon ........................ 45  Figure 13. Map showing annual spatial structure survey reaches and the spatial distribution of pools containing juvenile Chinook salmon from 2012 and 2013, Smith River Basin, California and Oregon. ................ 46  Figure 14. Map showing annual spatial structure survey reaches and the spatial distribution of pools containing juvenile trout (spp.) from 2012 and 2013, Smith River Basin, California and Oregon. .......................... 47  Figure 15. Map showing annual spatial structure survey reaches and the spatial distribution of pools containing adult cutthroat trout from 2012 and 2013, Smith River Basin, California and Oregon ......................... 48  Figure 16. Stream sections in (A) Baldface Creek (North Fork Smith River), and (B) upper South Fork Smith River, where juvenile coho salmon were detected during both years of the study. Note large accumulations of LWD and small granite alluvium in Baldface Creek example (A). ........................................................................................... 52  List of Appendices Appendix A. Sample Frame Development Resources and Metadata .................................................................................... 60  Appendix B. Number of pools occupied and density of Chinook Salmon, Unidentified juvenile trout (not identified to species), and adult Cutthroat Trout for all GRTS drawn reaches surveyed during spatial structure sampling in the Smith River, June ‐ September 2012. .................................................................................................................. 61  Appendix C. Number of pools occupied and density of Chinook Salmon, Unidentified juvenile trout (not identified to species), and adult Cutthroat Trout for all GRTS drawn reaches surveyed during spatial structure sampling in the Smith River, June – August 2013. .......................................................................................................................... 62  v Acknowledgments Without the generous help of numerous individuals, groups, and agencies, we would know much less about the biological resources of the Smith River basin than we do today. We would like to thank Patty McCleary and Grant Werschkull of the Smith River Alliance for taking on this project with enthusiasm and attention to ensure the project was a success throughout. The vision and collaboration of the Smith River Alliance provided a lasting motivation for all involved resulting in a highly organized effort. Beatrijs deWaard provided excellent grant support throughout the study. We thank all of the exceptional field biologists who assisted us in establishing the California Department of Fish and Wildlife’s Coastal Salmonid Monitoring Program in the Smith River basin with rigorous data collection standards: Gaytha Babcock, Diana Baetscher, John‐Deibner‐Hanson, Tara Dettmar, John Hagan, Jesse Nolan, Marisa Parish, Melissa Reneski and Jolyon Walkley. We also thank AmeriCorps Watershed Stewards Project (WSP) volunteers: Diana Baetscher, Adam Cockrill, Jen Olson, and Tim Sandborn who helped ease busy days and inspire us with their budding enthusiasm for salmonids. We also thank Michael van Hattem and Scott Bauer of CDFW for allowing their WSP members to join our team during critical survey periods. Rachael McCain (National Park Service) provided field assistance at critical survey periods and helped us overcome logistical challenges with sampling on State and National Park lands. The Mill Creek Salmonid Monitoring Program assisted greatly with the spawning ground surveys in 2011‐2012 so we could census the upper Mill Creek watershed. Field staff included Paul Albro, Thomas Dunklin, Zack Larson, Rod McLeod, and Bradford Norman. Many thanks to Frank Kemp and Charles Bartolotta of CDFW for providing last minute office space at Lake Earl Wildlife Area during our first season and to Jeff Bomke of State Parks and Rachael McCain of National Park Service for the opportunity of leasing park office and housing space. We thank Philip Bairrington, Mary Kuehner, Brenda Tuel, and Marisa Parish for administrative assistance throughout the study. Seth Ricker and Ken Lindke provided detailed solutions to estimating adult salmonid redd abundance and many conversations on analytical approaches to our data. We are in debt to Wade Sinnen, Michael Wallace, Sarah Borok, Frank Kemp, Charles Bartolotta, Mary Claire Kier, and Scott Bauer for loaning vehicles during times of need. This work would not have progressed as quickly or smoothly without planning insight provided by Mike McCain, Jim Waldvogal, Zack Larson, Scott Bowman, Tom Weseloh, Dan Burgess, Andy Van Scoyk, Rod McLeod, Ian Reid, Pat Righter, Michelle Gilroy, and James Simino. Last, we would like to thank all the generous landowners who provided critical access to stream sections on their properties. vi Introduction Observed population declines of coho salmon (Oncorhynchus kisutch) in freshwater habitats in California have led to both federal and state listings under the federal (ESA) and California (CESA) Endangered Species Acts (Federal Register 1997, CDFW 2002). These listings have initiated the development of recovery plans which include delisting goals (CDFW 2004, draft NOAA SONCC recovery document) for the Southern Oregon Northern California Coho (SONCC) Evolutionary Significant Unit (ESU). The Smith River basin has been identified as a functionally independent coho salmon population in the central diversity strata for the SONCC ESU by NOAA (McElhany et al. 2000, Williams et al. 2006, Williams et al. 2008) and is recognized as a recovery unit by California Department of Fish and Wildlife (CDFW 2004). The ‘population’ has been defined as the scale used to assess population viability (Williams et al. 2006). For a coho salmon population to meet or exceed a viable threshold, it must show a low risk of extinction over 100 years (McElhany et al. 2000). Therefore, to determine recovery for the SONCC ESU, numerous long‐term population monitoring programs need to be established and maintained across the ESU. NOAA established four viable salmon population (VSP) parameters to determine a population’s risk of extinction. These parameters include: abundance, productivity (population growth rate), spatial structure, and diversity (McElhany et al. 2000). Trend monitoring for these VSP parameters is the measure by which extinction risk and recovery status of an ESU is evaluated. To address data needs for the viability assessment, CDFW and NOAA (National Oceanic and Atmospheric Association) cooperatively developed the Coastal California Salmonid Monitoring Plan (CMP). Boydstun and McDonald (2005) and Adams et al. (2011) describe the strategy, general design, and general methods that are used in CMP monitoring. The current major funding source in California for VSP trend monitoring of ESA and CESA listed salmonids is through the Federal and State supported Fisheries Restoration Grants Program (FRGP) where funding is allocated based on population‐specific monitoring goals and focus species which are defined by the grants program. Coho salmon are currently the only ESA listed salmonid in the Smith River basin and thus are the only focus species identified in the watershed by FRGP. The Smith River has been ranked by the North American Salmon Stronghold Partnership Initiative as among the highest for salmonid conservation value. Furthermore, the Smith River is one of two watersheds in California described as “irreplaceable” with respect to salmonid population resiliency and biodiversity (Wild Salmon Center 2012). However, the status of the coho salmon population is among the least understood in California as assessments conducted previously were limited in scale, and almost exclusively restricted to a single sub‐basin. This uncertainty around the status of Smith River coho salmon prevents managers from critically assessing ESA recovery goals and prioritizing an effective restoration strategy defined by NOAA and CDFG (Beechie et al. 2003, CDFG 2004). The work described in this report represents the first comprehensive effort for monitoring viable salmon population parameters in the Smith River basin as defined by California’s Coastal Salmonid Monitoring Program. Monitoring VSP parameters of anadromous salmonids in basins having habitat resiliency and high salmonid conservation value will also benefit recovery priorities throughout the SONCC ESU by comparing data across a wide range of habitat conditions. Monitoring Approach We developed this coho salmon monitoring effort to assess two of the four viable salmonid population parameters: Abundance and Spatial Structure (McElhany et al. 2000). Each monitoring component requires well planned study designs, sampling protocols, analysis and reporting metrics, and data storage (Adams et al. 2011). Application of various monitoring components also needs to be standardized across multiple salmonid populations in order to assess population metrics at the ESU scale. Notwithstanding, the implementation of the CMP has only occurred in recent years for much of the monitoring area and methods 1 are being refined as lessons from new monitoring programs and data sets are becoming available to program managers. Population Abundance Abundance is perhaps the most important population metric since it can generally be used to assess overall extinction risk without needing to understand all the species‐specific factors influencing the population (McElhany et al. 2000). Spawning ground surveys are the primary monitoring method used for tracking salmonid population abundance trends in the northern monitoring area (Boydstun and McDonald 2005, Adams et al. 2011). Surveys are confined to an annual sample of stream reaches where redds, live fish, and carcasses are counted across multiple survey periods throughout a season (Gallagher et al. 2007). Total redd production is the primary abundance metric and is carried out using flag‐based mark‐recapture of individual redd features in a population model. The total number of redds are estimated for each survey reach and these totals are used to expand the estimate across the entire sample frame (Boydstun and McDonald 2005). Although this monitoring effort was designed for coho salmon, all salmonid species were incorporated into data collection and analysis based on the need to divide individual redds into separate species. Ultimately redds are converted to adult numbers based on adult to redd correction factors produced at local life cycle monitoring stations or from the scientific literature (Gallagher et al. 2010, Adams et al. 2011). Spatial Structure The spatial structure of a population refers both to the spatial distributions of individuals in the population and the processes that generate that distribution (McElhany et al. 2000). Spatial structure is important for assessing viability because understanding extinction risk for population abundance trends occurs at longer timescales than measured changes in the spatial arrangement of the population. Understanding patch use, patch size, patch connectivity, and patch colonization and extinction processes of the population will help managers define source patches while also protecting isolated patches that are much more vulnerable to extinction (Adams et al. 2011). For coho salmon, juvenile life stages are the most widely distributed across the riverscape with habitats being spatially and temporally dynamic (Wigington et al. 2006, Henning et al. 2006, Anderson et al. 2008, Koski 2009, Flitcroft et al. 2013). Two distinctive periods representing a high likelihood of contrasting stream habitat availability include the winter and summer. We suggest both periods are critical to understanding spatial structure dynamics and sampling strategies should be developed for each. For example, estuaries have been shown to be important temporal rearing locations for coho salmon during the winter (Koski 2009, Wallace and Allen 2009). Methods for monitoring juvenile salmonid spatial structure have not been formally developed by the CMP. However, Adams et al. (2011) suggested juvenile salmonid surveys be conducted during the summer on an annual basis in a sampled fraction of reaches throughout a population. We adapted a snorkel survey protocol by Webster et al. (2005) to sample for juvenile coho salmon throughout a randomly selected set of reaches with pools defined as the primary sampling unit. We based our design on an occupancy modeling framework that incorporates both reach‐level and pool‐level occupancy while accounting for imperfect detection rates (Nichols et al. 2008, MacKenzie et al. 2006). By tracking occupancy at both scales, we were able to determine the overall proportion of area occupied during the summer rearing period. Results from each year can be directly compared to assess the relative change in annual spatial structure. Our study is the first attempt at formalizing sampling methods and a statistical framework specifically for measuring juvenile salmonid spatial structure in California so this work should be considered a pilot effort. As such, our methods have not been reviewed or endorsed by the CMP. We hope results from this study will offer critical empirical data to further the development of an accepted state‐wide spatial structure monitoring component. Methods in the occupancy modeling construct are currently rapidly evolving suggesting opportunities to use new tools and methods in the near future. 2