ESTABLISHING A PROTOCOL FOR SYSTEMATIC SURVEYS OF STRANDED MARINE MAMMALS IN HUMBOLDT COUNTY. Mary Elizabeth Pacewicz. A Thesis Presented to

Transcription

1 ESTABLISHING A PROTOCOL FOR SYSTEMATIC SURVEYS OF STRANDED MARINE MAMMALS IN HUMBOLDT COUNTY By Mary Elizabeth Pacewicz A Thesis Presented to The Faculty of Humboldt State University In Partial Fulfillment of the Requirements for the Degree Master of Science in Biology Committee Membership Dr. Dawn Goley, Committee Chair Dr. Christine Cass, Committee Member Dr. Julia Parrish, Committee Member Dr. John Reiss, Committee Member Dr. Michael Mesler, Graduate Coordinator December 2014

2 ABSTRACT ESTABLISHING A PROTOCOL FOR SYSTEMATIC SURVEYS OF STRANDED MARINE MAMMALS IN HUMBOLDT COUNTY Mary Elizabeth Pacewicz Between October 2011 and September 2013, I designed and implemented an effort-based beach survey protocol for stranded marine mammals and seabirds in Humboldt County. This study was designed to create a baseline data set that is comparable to other programs in the state. I analyzed differences between historical records at Humboldt State University, findings from the systematic surveys, and findings from another program that conducts effort-based surveys for stranded marine vertebrates, the Coastal Observation and Seabird Survey Team (COASST). A mean of 6.0 x10-2 marine mammal carcasses per kilometer (encounter rate) was documented. Seasonal stranding patterns were detected for California sea lion and harbor seals. Three hotspots, or areas of higher encounter rates, were detected at the southern end of the survey range. The average number of marine mammal strandings per year and species diversity per year were compared to those of the historical stranding records from 1975 through There were more marine mammal strandings and more pinniped species per year documented during systematic surveys than in the historical records. On overlapping beaches, the encounter rate for seabird carcasses was lower for HSU s MMSP than for COASST. HSU s MMSP documented a higher encounter rate for ii

3 marine mammal carcasses than COASST. This may be due to different methods, timing of surveys, or different objectives for the programs. In conclusion, protocol for systematic surveys was designed and pilot-tested, a baseline data set documenting stranding patterns in Humboldt County was established, and collaboration with another effort-based survey program was initiated. iii

4 ACKNOWLEDGEMENTS As I began writing my acknowledgements, I was amazed at how many people not only became an important part of my project but also an integral part of my life over the past three years. I am grateful to have had experienced so much support. I would like to thank my advisor, Dawn Goley, and my committee members, Julia Parrish, Christine Cass, and John Reiss for their support and guidance as I developed my project. I also want to thank the Oiled Wild Life Care Network (OWCN) and Humboldt State University s Biological Sciences Department for supporting my research through grants. Bonnie Brown, Casey Brainerd, Kim McClurg, and Lana Ray, my research would not have been possible without you. You went above and beyond the MMERP internship requirements and worked harder than I could have ever asked. Your commitment to my project when I was still exploring Humboldt County s beaches and not sure what I was doing yet was inspiring. Cara Gates, Mary-Lynn Gibbs, Michelle Kobal, Tara McDonough, Ben Marshall, Shalina Noakes, Kristen Orth, Tiffany Plencner, Johnny Roche, Morgan Schleiderer, and Cari Williams, your assistance in MMERP and with my surveys was invaluable to me. To all MMERPers that participated in beach walks, thank you. I am proud to tell you that in 25 months, we never missed a survey. That is an enormous accomplishment iv

5 and there is no way it could have been done without everyone s tireless efforts and commitment. Ashley Donnell, I could not have asked for a better person to continue my project. You helped make the transition run smoothly and beach surveys are in great hands. Also, thank you for letting me use your computer a countless number of times for my GIS analysis and for all of your patience and assistance when we made maps for what seemed like hundreds of hours. Brittany Bierria and HealthSPORT s aquatic staff, I am grateful for all of your support. Every time I left work, I felt happier than when I entered. Having such caring coworkers was invaluable to me. Allison Fuller, all of the effort you put into helping me get acquainted to our lab and your endless support and guidance means more to me than I can say. You are the hardest worker I have ever met, and I could always count on you for advice and encouragement. Wiley Archibald, as you already know, I wouldn t be finishing graduate school without you. From building shelters in your truck to our walks in the park to taking hours to decide what to do for dinner, some of my best memories in Humboldt are because of you. I am so grateful that we started graduate school together. Now I want to thank my two most reliable Centerville walkers, Brett Carrothers and Kathleen Sholty. Brett, I could never thank you enough for your help and support. You not only assisted with surveys the entire time I was conducting research but were an integral part in continuing them once I finished. What s even more important to me, v

6 though, is our friendship. Our coffee trips, marsh walks, and blackberry picking adventures gave me something to look forward to during the most stressful times. Kathleen, you are the reason Humboldt became my home. I can t imagine what these past three and a half years would have been like without you. Making smoothies, painting nails, and exploring Humboldt would not have been the same without my partner in crime. Thank you for being there for me at the times I needed it the most. Chrissy and Eric, thank you for all of the encouragement you have given me since I started graduate school. You both have done amazing things during these past three years, and I am so proud of both of you and all of your accomplishments. Mom and Dad, I owe all of my accomplishments to you. You have and always will be a part of everything I do, and there is no way I can thank you enough. I am incredibly grateful to have had your endless love and support as I pursued my dreams. You have been an important part of this whole journey starting with driving across the country with me as I moved to Arcata. All of the supportive phone calls and words of encouragement you have given me over the past few years mean more to me than I can possibly tell you and are the reason I am receiving my Master s degree. vi

7 TABLE OF CONTENTS PAGE ABSTRACT...ii ACKNOWLEDGEMENTS...iv LIST OF TABLES.ix LIST OF FIGURES x LIST OF APPENDICES.xiv INTRODUCTION History of Marine Mammal Stranding Networks...1 Previous Analyses of Historical Marine Mammal Strandings....4 Challenges to Interpreting Marine Mammal Stranding Data..4 Significance of Documenting Marine Mammal Strandings in Northern California...11 Objectives..13 MATERIALS AND METHODS..14 Study Area. 14 Survey Methods. 14 Marine Mammal Data Analysis Seabird Data Analysis Comparing COASST Survey Protocol to HSU s MMSP Survey Protocol...25 RESULTS..26 vii

8 TABLE OF CONTENTS (CONTINUED) PAGE Marine Mammal Strandings Seabird Strandings Comparing COASST Survey Protocol to HSU s MMSP Survey Protocol...52 DISCUSSION Marine Mammal Strandings..68 Seabird Strandings Comparing COASST Survey Protocol to HSU s MMSP Survey Protocol...87 LITERATURE CITED..95 APPENDIX A APPENDIX B..103 APPENDIX C..105 APPENDIX D..128 viii

9 LIST OF TABLES PAGE Table 1. Names of each beach in the survey area and corresponding beach length..17 Table 2. Number of carcasses of each species documented during the first year of effortbased surveys, October 2011 through September 2012, and the second year of effortbased surveys, October 2012 through September Table 3. Number of carcasses of each cetacean species documented from October 2011 through September 2012 and October 2012 through September Table 4. Number of carcasses of each marine mammal species documented from opportunistic reports..34 Table 5. Summary of COASST and HSU s MMSP survey results ix

10 LIST OF FIGURES PAGE Figure 1. Map of counties in central and northern California where effort-based surveys for marine vertebrates are conducted Figure 2. Map of Mendocino, Humboldt, and Del Norte County. Humboldt State University s Marine Mammal Stranding Program responds to reports of marine mammal carcasses in all three counties Figure 3. Area of Humboldt County coastline that was systematically surveyed once a month from October 2011 through September Figure 4. Start and end point for each beach segment that was systematically surveyed from October 2011 through September Figure 5. Map of survey area showing location of the 107 marine mammal strandings documented from October 2011 through September Figure 6. The mean number of marine mammal, cetacean, and pinniped carcasses documented per kilometer per year between October 2011 and September Figure 7. The mean number of marine mammal, cetacean, and pinniped species documented per year between October 2011 and September Figure 8. The average number of marine mammal carcasses documented per month from October 2011 through September 2013 because of opportunistic reports compared to the average number of marine mammal carcasses documented per month from effort-based surveys...33 Figure 9. Total number of marine mammals, cetaceans, and pinnipeds documented from opportunistic reports compared to the total number of marine mammals, cetaceans, and pinnipeds documented by HSU s MMSP during effort-based surveys between October 2011 and September Figure 10. The number of marine mammal carcasses that were encountered on surveys after they were initially documented., 37 x

11 LIST OF FIGURES (CONTINUED) xi PAGE Figure 11. Mean encounter rate of California sea lion carcasses documented per season compared to the mean encounter rate of harbor seal carcasses documented per season between October 2011 and September Figure 12. Map of Humboldt County showing hotspots and cold spots for marine mammal strandings Figure 13. The number of carcasses of each species of marine mammal documented in the historical data from HSU s Vertebrate Museum (1975 through 2010) compared to the species diversity documented during effort-based surveys (2011 through 2013) Figure 14. Mean number of marine mammals, cetaceans, and pinnipeds documented per year from compared to the average number documented per year from Figure 15. Mean number of marine mammal, cetacean, and pinniped species documented per year from compared to the average number of species documented per year from Figure 16. Total number of California sea lion carcasses documented per season from 1975 through 2010 compared to the total number of California sea lion carcasses documented from October 2011 through September Figure 17. Total number of harbor seal carcasses documented per season from 1975 through 2010 compared to the total number of harbor seal carcasses documented per season from October 2011 through September Figure 18. Total number of gray whale carcasses documented per season compared to the total number of harbor porpoise carcasses documented per season from 1975 through Figure 19. Number of seabird carcasses documented per kilometer per month during effort-based surveys from July 2012 through September Figure 20. Total number of carcasses of each seabird species documented during effortbased surveys from July 2012 through September

12 LIST OF FIGURES (CONTINUED) PAGE Figure 21. Correlation between the encounter rate for all seabird carcasses and the encounter rate for common murre carcasses during effort-based surveys from July 2012 through September Figure 22. (a) Mean encounter rate of sooty shearwater and brown pelican carcasses during effort-based surveys from July 2012 through September (b) Mean encounter rate of grebes during effort-based surveys from July 2012 through September Figure 23. Map showing the beach segment that was a hotspot for seabird strandings from July 2012 through September Figure 24. Mean encounter rate for seabirds on each beach during effort-based surveys from July 2012 through September Figure 25. Map showing two beach segments that were hotspots for common murre strandings from July 2012 through September Figure 26. Mean number encounter rate for common murres on each beach during effortbased surveys from July 2012 through September Figure 27. Map of Humboldt County s coastline showing where HSU s MMSP s surveys overlapped with COASST surveys Figure 28. Map of marine mammal strandings that were documented by HSU s MMSP between October 2012 and September 2013 compared to where COASST surveys were conducted Figure 29. Percentage of seabird carcasses that were identified as common murres, shorebirds, waterfowl, and unidentified seabird carcasses by COASST compared to HSU s MMSP 63 Figure 30. HSU s MMSP encounter rate for seabird carcasses compared to COASST s encounter rate for seabird carcasses Figure 31. Beach segment that was a hotspot for seabird carcasses on COASST surveys between October 2012 and September xii

13 LIST OF FIGURES (CONTINUED) PAGE Figure 32. The average seabird carcass encounter rate for COASST compared to the average seabird carcass encounter rate for HSU s MMSP between October 2012 and September xiii

14 LIST OF APPENDICES PAGE Appendix A. Level A data sheet Appendix B. Latitude and longitude for start and end points for each beach segment Appendix C. Date, length of time, number of volunteers on the survey, and the number of volunteers being trained for each beach survey conducted between October 2011 and September Appendix D. Number of individuals and percentage of seabird carcasses of each species documented by between October 2012 and September 2013 by COASST and HSU s MMSP xiv

15 1 INTRODUCTION History of Marine Mammal Stranding Networks In 1972, the Marine Mammal Protection Act (MMPA) was established by the United States Congress with the goal of protecting marine mammal species and populations to prevent them from declining to a point where they are not functioning elements in the ecosystem. The MMPA was one of the first proactive conservation efforts, with the goal of conserving and protecting all marine mammals and replenishing species or populations that were already depleted. However, conservation efforts were limited because of a lack of information about the life history, population dynamics, and ecology of marine mammals. To provide a more solid basis for conservation decisions, the MMPA mandated regular assessments of marine mammal populations. These assessments have been based both on studies of wild populations and on stranded marine mammals. Studies of live marine mammals at sea are often limited by high cost and logistical complexity; however, studies of stranded marine mammals can be important sources of information and are less costly and logistically complex (Geraci and St. Aubin 1979; Peltier et al. 2012). A stranded marine mammal is defined as any dead marine mammal on a beach or floating nearshore; any live cetacean on a beach or in water so shallow that it is unable to free itself and resume normal activity; or any pinniped which is unable or unwilling to leave the shore because of injury or poor health (Wilkinson 1991).

16 2 Data have been collected on stranded marine mammals for hundreds of years (e.g., Scoresby 1810; Harmer 1918), with the earliest standardized reporting in the United Kingdom in 1913 (Sheldrick 1976). However, it was not until the 1977 Workshop on Marine Mammal Strandings that a coordinated plan to study marine mammal strandings in the United States was established. At this workshop, four regional marine mammal stranding networks were established as a component of Title IV of the MMPA (Hofman 1991). Agencies within the networks collect data from both live and dead stranded animals with the goal of assessing the health of marine mammals and their ecosystems and documenting mortality events. The effectiveness of the networks was assessed during the Second Marine Mammal Stranding Workshop (Reynolds and Odell 1991) in December Between 1977 and 1987, marine mammal stranding networks had established a strong framework of volunteers and coordinators and conducted efficient responses to unusual stranding events. However, there were inconsistencies in data and specimen collection among regions because each region worked independently. To facilitate collaboration among regions, the National Marine Fisheries Service (NMFS) created a national stranding coordinator position to oversee the regional coordinators. As part of the reauthorization of the MMPA in 1992, the National Marine Mammal Health and Stranding Response Program (MMHSRP) was established. The MMHSRP was created under Title IV of the Marine Mammal Protection Act. National and regional marine mammal stranding networks are one component in this program and contribute to the MMHSRP s goal of collecting data on marine mammals and marine

17 3 mammal populations. To standardize data collection in the field, responders are required to collect standardized Level A data which includes species, sex and age class, location, date, level of decomposition, and morphological data (i.e. straight length and weight), and evidence of human interactions (Appendix A). These data are useful for studying life history, determining species diversity, and tracking mortality caused by anthropogenic interactions. Between 2009 and 2011, 12,545 cetacean and 39,104 pinniped strandings were recorded in the United States. Utilizing these stranding data, biologists have described species life history (Westgate and Read 2007), detected diseases (Greig et al. 2005), documented mortality due to human interaction (Read and Murray 2000), detected unusual mortality events (Danil et al. 2010), and assessed spatial and temporal patterns of strandings (McFee et al. 2006). Another component of the MMHSRP is the coordinated response to unusual mortality events. An unusual mortality event (UME) is "a stranding that is unexpected; involves a significant die-off of any marine mammal population; and demands immediate response (Marine Mammal Protection Act of 1972). UMEs can be caused by disease, human activities, or natural toxins and can be indicators of changes the health of marine mammal populations (Harvell et al. 2004; Kim et al. 2005). Information from stranding networks is useful for detecting an UME because baseline stranding patterns can be compared to the suspected UME to determine when a significant change in strandings has occurred (Becker et al. 1994).

18 4 Previous Analyses of Historical Marine Mammal Strandings In addition to data from the MMHSRP, historical marine mammal stranding records have been used to analyze stranding patterns for cetaceans (e.g., McLellan et al. 2002; Norman et al. 2004; Maldini et al. 2005; Danil et al. 2010). For example, the cetacean stranding records in San Diego County ( ) (Danil et al. 2010) and Oregon and Washington ( ) (Norman et al. 2004) were analyzed to determine trends for species diversity, seasonal and spatial stranding patterns, and signs of human interaction over time. In addition, temporal and spatial patterns in historical cetacean stranding records have been analyzed to define bottlenose dolphin (Tursiops truncatus) stocks on the Atlantic coast (McLellan et al. 2002) and to compare species diversity captured in stranding records to live-animal surveys in Hawaii (Maldini 2005). Challenges to Interpreting Marine Mammal Stranding Data Though species diversity in a region can be described and spatial and temporal stranding patterns can be examined by utilizing historical records, analyses are limited because it is not possible to determine whether patterns are driven by factors such as an increase in human presence on the beach (i.e. effort) or if there are true changes in the number of strandings. In addition, by solely relying on reported marine mammal strandings and by not standardizing the effort for marine mammal surveys, data cannot be compared between programs in the stranding network. Without an effort-based approach, marine mammal strandings reported from crowded, easily accessible beaches are often over-represented compared to more isolated

19 5 locations (Norman et al. 2004). For example, 48% of all stranding records for the main Hawaiian Islands between came from O ahu, which has the largest population (Maldini et al. 2005). Changes in tourism may also alter the frequency of stranding reports, with more reports occurring in locations or during seasons where tourism is high (Danil et al. 2010). In addition, as public awareness of marine mammals has increased, so has the number of strandings that are reported (Norman et al. 2004; Leeney et al. 2008). By relying solely on reports from the public, marine mammal stranding networks can describe species diversity in a region, but without controlling for effort, comparisons among regions cannot be made and conclusions that can be drawn about the baseline marine mammal stranding conditions are limited. To address the challenges produced by inconsistent effort, an effort-based approach to systematically survey the coastline for stranded marine mammals has been established along a number of different coastlines in the United States. Effort can be controlled by surveying the same length of beach, conducting the same number of surveys per month, and using the same number of surveyors to search for carcasses on every beach (Steiger et al. 1989). There have been attempts to systematically survey beaches for marine mammal carcasses, (e.g., Steiger et al. 1989), however, these studies have been short term and over a small geographic range. To date, there has not been any coast-wide collaboration with the goal of consistently surveying for marine mammal carcasses over a large geographic range. This lack of collaboration and inconsistent surveying has limited the MMHSRP s ability to detect UMEs or compare stranding patterns over a large geographic range.

20 6 Though effort-based surveys for marine mammal strandings have not been implemented over a large geographic scale, systematic searches for seabird carcasses have been conducted over a long period of time and over large geographic areas in Europe since the 1950s and in the United States since the 1970s in response to oil spills (Camphuysen and Heubeck 2000). A more recent example of a stranded marine vertebrate monitoring program in California is the Coastal Ocean Mammal and Bird Education and Research Surveys (BeachCOMBERS). BeachCOMBERS is a citizen science program that was established by Hannah Nevins in 1997 to systematically monitor the beaches bordering the Monterey Bay National Marine Sanctuary. BeachCOMBERS partners with the Monterey Bay National Marine Sanctuary, California Fish and Wildlife, and Moss Landing Marine Laboratory. Monthly surveys are conducted by citizen-science volunteers from the San Mateo and Santa Cruz County border south to San Luis Obispo (Figure 1). Volunteers are trained to survey the same segment of beach for stranded seabirds and marine mammals during the first week of every month and record information such as species identity, age class, sex, cause of death (when possible), and signs of human interaction. By systematically surveying the beaches along the Monterey Bay National Marine Sanctuary each month, BeachCOMBERS has been successful in building a long term data set that documents the species of seabirds and marine mammals that strand in central California. From these data, trends in encounter rate and UMEs can be detected. Another program that documents both seabird and marine mammal carcasses during effort-based surveys in central California is Beach Watch. Beach Watch was

21 7 established by Jan Roletto in 1993, and volunteers survey segments of the coast along the Gulf of the Farallones National Marine Sanctuary. Beach Watch partners with the Farallones Marine Sanctuary Association. Surveys are conducted by citizen-science volunteers from Sonoma County south to the San Mateo and Santa Cruz County border (Figure 1). In addition to documenting carcasses, Beach Watch volunteers document the presence of live marine vertebrates on the beach. The long term goal for Beach Watch is to establish a baseline data set for living and stranded seabirds and marine mammals. In northern California, staff members from Redwood National and State Parks (RNSP) conduct effort-based surveys for stranded marine mammals and seabirds on 25.8 km of coastline in northern Humboldt County and southern Del Norte County within park boundaries (Figure 1). RNSP staff began effort-based surveys in 1997, and in 2000, began conducting surveys every month. RNSP s goal is to document patterns in marine mammal and seabird mortalities within the park boundaries. A similar program to BeachCOMBERS, Beach Watch, and RNSP is the Coastal Observation and Seabird Survey Team (COASST) which was established by Julia Parrish at the University of Washington in 1998 to systematically collect baseline data on seabird mortality. Marine mammal strandings are reported anecdotally. The program has expanded to include Washington, Oregon, Alaska, and northern California. Differences in effort-based protocols and verification techniques limit the data Beach Watch can share, but due to consistency in protocol and verification techniques, BeachCOMBERS, RNSP, and COASST are able to share data about seabird mortality along the West coast of the United States.

22 Figure 1. Map of California counties where systematic surveys for marine vertebrates are conducted by BeachCOMBERS, Beach Watch, Redwood National and State Parks, COASST, and HSU s MMSP. 8

23 9 Though stretches of central Humboldt County s coastline are systematically surveyed by COASST volunteers, the program s focus is on documenting seabirds. Therefore, volunteers are trained to survey for and identify seabird carcasses but are not trained to document marine mammal carcasses with Level A data sheets. Though patterns of seabird mortality can be compared along the West coast, it is not possible to compare patterns of marine mammal strandings because of the gaps in information about marine mammal strandings in northern California, particularly central Humboldt County. Humboldt State University (HSU) is a member of the West region stranding network and has the letter of authority and responsibility to respond to and report dead marine mammals in Mendocino, Humboldt, and Del Norte County (Figure 2). HSU has held the letter of authority since Historically, members of HSU s MMSP collected data opportunistically when the public reported stranded marine mammals. By solely relying on opportunistic data, HSU s MMSP did not conduct effort-based systematic surveys, and, therefore, did not establish a baseline data set that could be used to compare trends in marine mammal strandings in northern California. In October of 2011, I developed a protocol to systematically monitor the accessible beaches in central Humboldt County with the goal of extending marine mammal stranding coverage to Mendocino and Del Norte County after the protocol was tested. By implementing effort-based surveys and collecting Level A data from every marine mammal stranding, a baseline data set on the identity and spatial and temporal

24 Figure 2. Map of Mendocino, Humboldt, and Del Norte Counties. Humboldt State University s Marine Mammal Stranding Program (HSU MMSP) responds to reports of marine mammal carcasses in all three counties. 10

25 11 distribution of marine mammal strandings along the coast was established. These baseline data can be used to document spatial and temporal stranding patterns, detect UMEs, and coordinate with other effort-based programs along the coast. Another challenge to relying on opportunistic reports and not controlling for effort is that larger species are more easily spotted and potentially more likely to be reported. Unusual species are also more likely to be reported and responded to. Therefore, there has been a general tendency towards more thorough responses to cetacean carcasses compared to pinniped carcasses (Wilkinson 1991). By implementing effort-based surveys, I eliminated the bias toward documenting unusual species and established a protocol that allows stranding patterns for both cetacean and pinniped carcasses to be analyzed. Significance of Documenting Marine Mammal Strandings in Northern California Systematic surveys for both marine mammal and seabird carcasses have been successfully implemented along the central coast of California (e.g., BeachCOMBERS and Beach Watch), and COASST volunteers and RNSP staff survey beaches in northern California. However, COASST volunteers focus on documenting seabird mortality and the program has not emphasized reporting marine mammal carcasses. Though RNSP covers a large geographic range, effort-based surveys are only conducted on a small proportion of beaches in the range. The lack of effort to systematically document marine mammals along a significant portion of northern California coastline has resulted in a gap in knowledge about marine mammal stranding patterns in this region.

26 12 The species assemblage found offshore of northern California is different than other regions in California, making documenting strandings in this area important. Northern California is the southernmost terminus of the Steller sea lion (Eumetopias jubatus) breeding rookeries, has multiple harbor seal (Phoca vitulina) haul-out and rookery sites, is along the Eastern Pacific gray whale (Eschrichtius robustus) population s migration route, and is the southernmost terminus for the gray whale Pacific Coast Feeding Group (PCFG). In addition, historical stranding data reflect a wide range of cetacean diversity. The wide range of diversity found in the waters off of northern California can be attributed to the highly productive California Current System. By implementing systematic surveys for stranded marine mammals in northern California, we can compare species diversity in northern California to other to areas along the coast more effectively. Marine Protected Areas (MPAs) were implemented in northern California in December The implementation of MPAs along the coast of California is due to the Marine Life Protection Act (MLPA) which was passed in The objective of the MLPA is to protect the habitats of marine species, including marine mammals, from human disturbance, with the goal of conserving marine species and diversity. By establishing a baseline data set for marine mammal stranding patterns, shifts in stranding trends can be analyzed in relation to the implementation of local MPAs. Shifts in stranding trends, for example, a change in the number of strandings per year or a change in species diversity, can contribute to adaptive management decisions when the successes and shortcomings of MPAs in northern California are evaluated.

27 13 Objectives My primary goal as HSU s MMSP field coordinator was to establish and test a protocol for effort-based surveys for marine mammal strandings in northern California that was standardized with other programs along the California coast. Implementing this protocol in northern California would enable the Western stranding region to more effectively function as a network. In turn, it would facilitate the MMHSRP s goal of utilizing baseline data to identify unusual stranding events and document trends in marine mammal health. In order to establish a protocol that could be implemented along the coast, I had several objectives. The first objective was to design an effort-controlled protocol to systematically survey the accessible beaches in central Humboldt County for marine mammal strandings that was comparable to other programs along the California coast. The second objective was to establish a baseline data set using this protocol, and to use it to analyze the species diversity and the temporal and spatial trends in marine mammal strandings. The third objective was to compare results from these effort-based surveys to the historical data which were gathered from strandings reported by the public between1975 and Finally, to test the potential to expand the scope of HSU s MMSP, I explored the possibility of partnering with the stranded seabird monitoring program, COASST, which conducts surveys on the same stretch of coastline as HSU s MMSP.

28 14 MATERIALS AND METHODS Study Area Effort-based surveys were conducted on 70.1 kilometers of Humboldt County s kilometer coastline between October 2011 and September 2013 (Figure 3). Every accessible beach in central Humboldt County was surveyed for marine mammal carcasses once a month from October 2011 through September 2013, and seabird carcasses were documented from July 2012 through September Surveys were conducted between 40 34'29.922"N and 41 15'50.491"N (Appendix B). Starting points for surveys were selected by accessibility, and end points for surveys were based on physical barriers (e.g., river mouths) or a beach access where another survey segment would begin (Figure 4). The beach names that correspond with each survey segment can be found in Table 1. Survey Methods Marine mammal and seabird surveys were conducted by two trained surveyors from HSU s MMSP walking in parallel lines along the length of each beach segment. One surveyor walked along the edge of the dunes and the other walked along the wrack line. If dunes were not present, one surveyor walked along the line of vegetation on the beach. If a wrack line was not present, one observer walked in a straight line down the middle of the beach. Surveyors continuously scanned for carcasses.

29 Figure 3. Area in Humboldt County, California that was systematically surveyed once a month from October 2011 through September

30 Figure 4. Starting and ending points for each Humboldt County beach segment that was systematically surveyed from October 2011 through September

31 Table 1. Corresponding number of each beach segment and length as shown in Figure 3. Each beach segment was given a name, typically based on the street name or landmark at the beach access. Number Name of Beach Kilometers 22 Stone Lagoon Dry Lagoon Big Lagoon Agate Beach College Cove Trinidad Beach Indian Beach Baker Beach Luffenholtz Houda Point Moonstone Beach Clam Beach to Moonstone Clam Beach to Mouth of Mad River Mad River Parking Lot to Mouth of Mad River Ma-le'l Dunes to Mad River Parking Lot Manila Community Center to Ma-le'l Dunes Powerline Beach to Manila Community Center Bay Street to Powerline Beach North Jetty to Bay Street South Spit Table Bluff Centerville

32 18 Between October 2011 and September 2013, all beaches were surveyed once a month for marine mammal carcasses (Appendix C). Beginning in July 2012, seabird carcasses were also documented during surveys, using the previously described protocol. Between October 2011 and June 2012, surveyors were trained in the survey methods by conducting surveys with the field coordinator. After June 2012, when seabird carcasses were also documented, surveyors were trained by first observing a beach survey. On their first survey, new surveyors walked as silent observers and did not participate in searching for carcasses. The volunteers in training always walked behind the two surveyors to ensure that they did not give any cues to where a carcass could be found. The purpose of this initial survey was to teach the new volunteer the survey methods and process of carcass identification and documentation. After the initial training survey, volunteers participated in surveys by actively searching for carcasses. However, new surveyors only conducted surveys with a surveyor that was experienced in species identification. On every survey there was always one surveyor that was experienced in species identification. An experienced surveyor was defined as any surveyor that had conducted surveys on a specific beach segment for more than two months. When a new marine mammal carcass was discovered, observers completed a Level A (Appendix A) data sheet, documenting the species, sex, age, size, and location of the carcass. To verify species and carcass condition, photographs were taken of the entire carcass, and if present, the skull, teeth, fore and hind flippers, tags or brands, and other features used to identify species, sex, or age class of the carcass (e.g., sagittal crest or

33 19 baculum). Age class was determined using guidelines from NMFS. Photographs of the straight length and morphological measurements (e.g., flipper length, dorsal fin length, girth) were taken and archived. A measuring tape was included in all photographs as a standard of measurement. Complete morphological measurements were not always possible due to the condition of the carcass (e.g., missing flippers). Straight length was the only morphological measurement required by the Level A data sheet, and regardless of carcass condition, straight length was measured and details about the carcass condition were recorded. When possible, surveyors completed the additional measurements for pinnipeds and cetaceans described in Marine Mammals Ashore: A Field Guide for Strandings (Geraci and Lounsbury 2005). These measurements include girth and length of flippers. Species were identified using criteria from Marine Mammals Ashore: A Field Guide for Strandings (Geraci and Lounsbury 2005). Carcasses were tagged on the skull (if present) with a colored cable tie to ensure they were not recounted on subsequent surveys. If the skull was missing, the carcass was tagged on a hind flipper. If the hind flipper was missing, the carcass was tagged on a fore flipper. Photographs were taken of the cable tie and the location of the cable tie on the carcass was documented on the Level A data sheet. If a marine mammal carcass was found with a cable tie, it was recorded as a re-find but Level A data were not retaken. All Level A data sheets were uploaded to the NMFS marine mammal stranding database and HSU s MMSP s database. In addition, hard copies of the Level A data sheets are stored in HSU s MMSP s office. As the field coordinator, it was my

34 20 responsibility to maintain the data, and future field coordinators will have the same responsibilities. When a seabird carcass was discovered, it was identified using Beached Birds: A COASST Field Guide (Hass and Parrish 2002). Seabirds were identified by foot description and three morphological measurements: tarsus length, wing chord, and bill length. Birds that could not be identified were photographed. A photograph of the foot and tarsus measurement, bill, wing chord, whole bird on it back, and whole bird on its chest were taken and archived for future verification. Seabird carcasses were tagged with a white cable tie to ensure they were not recounted on subsequent surveys. If a seabird carcass was found with a cable tie, it was recorded as a re-find. Data collected from seabird carcasses and photographs of unidentified seabirds were archived. Hard copies of the data sheets are stored in HSU s MMSP s office. Marine Mammal Data Analysis In addition to effort-based surveys, surveyors from HSU s MMSP responded to reports from the public of stranded marine mammals and collected Level A data. Reports from the public are referred to as opportunistic reports. Opportunistic reports that were outside of the geographic range of the systematic surveys were responded to, but were excluded from the current analysis of stranding patterns on central Humboldt County beaches. Opportunistic reports were included in the analyses because all carcasses that surveyors did not collect from the beach were re-sighted during effort-based surveys.

35 21 For the two years of systematic surveys, encounter rate on each individual beach was calculated by dividing the total number of newly discovered marine mammal carcasses on that beach by the length of the beach in kilometers. To determine overall trends in marine mammal strandings on the entire survey area from October 2011 through September 2013, I calculated the mean number of marine mammal, cetacean and pinniped carcasses documented per kilometer per year (encounter rate), as well as the mean number of marine mammal, cetacean, and pinniped species documented per year (average species diversity). The total number of marine mammal carcasses that were documented on surveys following the first encounter was calculated. The longest time that a marine mammal carcass was documented after the initial encounter was also calculated. Seasonal stranding patterns from effort-based surveys (October 2011-September 2013) Seasons are designated as fall (September-November), winter (December- February), spring (March-May), and summer (June-August) to enable comparison with previous studies that examined historical stranding records (e.g., Norman et al. 2004). Seasonal stranding patterns from the two most commonly stranding marine mammals, California sea lions (Zalophus californianus) and harbor seals (Phoca vitulina) were described. Hotspot analysis from effort-based surveys (October 2011-September 2013) A hotspot analysis was used to detect spatial clusters of marine mammal strandings within 1 mile (1.6 km) of each other. The parameter for the hotspot was

36 22 chosen because it was a distance that fell within COASST survey ranges and ensured that all cells fell completely within the survey range. The location of marine mammal carcasses was obtained from the GPS coordinates from Level A data sheets (n=99). Level A data sheets with missing or inaccurate coordinates were not used in the hotspot analysis (n=8). Areas that had a higher number of strandings than expected (hotspots) and areas that had fewer strandings than expected (coldspots) were identified using the Getis-Ord Gi* statistic (ArcGIS 10.1). The Getis-Ord Gi* statistic has been used to locate foraging hotspots for bottlenose dolphins (Tursiops truncatus) (Smith et al. 2013) and sea lion stranding hotspots (Kessina 2013). The null hypothesis for this statistic is that there is complete spatial randomness, or no spatial clustering (Esri 2013). For this study, each stranded marine mammal was given equal weight, and the analysis was based on the comparison of the number of strandings in each 1 mile beach segment and the number of strandings in the surrounding segments. A location where there was a significant spatial clustering of strandings was considered a hotspot, and a location where there was a significant lack of strandings was considered a coldspot. One mile beach segments were chosen after running a Multi-Distance Spatial Clustering Analysis (Ripley s K-function). Ripley s K-function tests different spatial scales to show whether clustering, normality, or dispersion occurs and what spatial scale should be used to detect those patterns (Esri 2013). By using Ripley s K function, it was determined that in HSU s MMSP s survey area, marine mammal stranding hotspots could be detected at small spatial scales, including one mile segments.

37 23 Historical data compared to data from effort-based surveys Historical data from HSU s Vertebrate Museum s database were analyzed to compare with findings from the effort-based surveys and to describe seasonal trends in marine mammal carcass encounters. The first year of records used was Historical marine mammal stranding records that were documented within the range of the systematic surveys and that were documented with a Level A data sheet or specimen collection report from HSU s Vertebrate Museum were used for comparative analysis. The average yearly number of marine mammal, cetacean, and pinniped strandings documented per year from October 2011-September 2013 were compared to the same means from the historical data using a two-sample t-test. The average yearly marine mammal, cetacean, and pinniped species diversity from October 2011-September 2013 were compared to the means from the historical data using a two-sample t -test. Though the method of data collection was different, seasonal stranding patterns from the historical records for the two most common species of stranded pinnipeds (California sea lions and harbor seals) were compared to seasonal stranding patterns from the two years of effort-based surveys. This comparison was done to determine if opportunistic reports reflect the same seasonal patterns as effort-based surveys. Though the two years of effort-based surveys did not have enough data to compare stranding patterns for cetaceans, seasonal stranding patterns for two most common species of stranded cetaceans (Gray whales (Eschrichtius robustus) and harbor porpoises (Phocoena phocoena)), were described using historical records.

38 24 Seabird Data Analysis The number of seabird carcasses documented per kilometer (encounter rate) during effort-based surveys from July 2012 through September 2013 for the entire survey area was calculated. Species diversity and abundance were calculated by counting the total number of species and the number of individuals of each species that were documented from July 2012 through September Seasonal stranding patterns of seabirds from effort-based surveys (July 2012-September 2013) The monthly encounter rate from July 2012 through September 2013 was calculated. Seasonal stranding patterns for the four most commonly encountered seabirds: common murres (Uria aalge), large immature gulls, sooty shearwaters (Puffinus griseus), and brown pelicans (Pelecanus occidentalis) were described. Large immature gull is a category that was created by COASST. Subadult gulls are difficult to identify to species because their plumage is often not the same as an adults for several years. Seasonal patterns for grebes were also described. Hotspot analysis from effort-based surveys (July 2012-September 2013) GPS coordinates for every seabird carcass were not documented, so hotspots for seabird strandings were identified by calculating the encounter rate on each individual beach segment. A beach with an encounter rate that was greater than two standard deviations above the average encounter rate from all beaches was considered a hotspot.

39 25 Comparing COASST Survey Protocol to HSU s MMSP Survey Protocol COASST and HSU s MMSP had some similarities in their protocols. Both protocols required two surveyors to scan for carcasses and did not require surveyors to move debris to dig for carcasses. The goal for both programs was to maintain consistency in the time interval between surveys (e.g., HSU s MMSP surveyors walked each beach once a month). COASST surveyors could walk more frequently but were encouraged to remain consistent in the time interval between surveys. One major difference in protocol was the method for walking the beach. All of HSU s MMSP surveys were conducted by walking in parallel lines down the beach, and surveyors only scanned while walking in one direction. COASST surveyors walked in a zig zag pattern and the width of the beach determined whether surveyors would scan for carcasses while walking in one direction or while walking in both directions. The location of beaches that COASST volunteers surveyed were compared to the location of beaches that HSU s MMSP volunteers surveyed to determine overlap and compare the number and species diversity of marine mammal and seabird carcasses documented per kilometer between programs. For every beach segment COASST volunteers surveyed that fell entirely within HSU s MMSP survey segments, the number of new seabird and marine mammal carcasses that were documented per kilometer (encounter rate) from October 2012 through September 2013 were compared. Hotspots, beaches with greater than two standard deviations above the average encounter rate, were also compared.

40 26 RESULTS Marine Mammal Strandings Results from effort-based surveys Between October 2011 and September 2013, 107 marine mammal strandings (11 cetaceans and 96 pinnipeds) were documented within the survey area (Figure 5). The mean yearly encounter rate of marine mammal carcasses was 6.0 x10-2 (S.E. ± 3.0x10-3 ). Pinniped carcasses were encountered more than cetacean carcasses. The mean yearly encounter rate of cetacean carcasses was 6.5x10-3 (S.E. ± 3.0x10-3 ), and the mean yearly encounter rate of pinniped was 5.7x10-2 (S.E. ± 5.9x10-3 ) (Figure 6). California sea lion carcasses were documented the most (n=29), followed by harbor seal carcasses (n=26) and unidentified otariids (n=25) (Table 2). During the first year of effort-based surveys, 18 California sea lion carcasses were documented and 11 harbor seal carcasses were documented. During the second year of effort-based surveys, 12 California sea lion carcasses were documented and 14 harbor seal carcasses were documented. RNSP s data showed a similar pattern of a higher number of California sea lion strandings compared to harbor seal strandings in 2012 and a higher number of harbor seal strandings compared to California sea lion strandings in In 2012, RNSP documented eight California sea lions and one harbor seal (Bensen 2013). In 2013, RNSP documented three California sea lions and five harbor seals (Bensen 2014).

41 Figure 5. Map of survey area showing location of the 107 marine mammal strandings documented from October 2011 through September

42 Encounter rate Marine mammals Cetaceans Pinnipeds Figure 6. The mean yearly encounter rate of marine mammal, cetacean, and pinniped carcasses from October 2011 through September Error bars represent standard error.

43 Table 2. Number of carcasses of each pinniped species documented from October 2011 through September Species October October September 2012 September 2013 Total Zalophus californianus Phoca vitulina Unidentified otariid Eumetopias jubatus Mirounga angustirostris Unidentified phocid Unidentified pinniped Callorhinus ursinus Total

44 30 The mean yearly encounter rate of California sea lion carcasses was 1.7x10-2 (S.E. ± 4.2x10-3 ), the mean yearly encounter rate of harbor seal carcasses was 1.5x10-2 (S.E. ± 1.2x10-3 ), and the mean yearly encounter rate of unidentified otariid carcasses was 1.5x10-2 (S.E. ± 5.9x10-4 ). The most common cetacean carcass documented was the harbor porpoise, with seven carcasses documented during the two years of systematic surveys (Table 3). The mean yearly encounter rate of harbor porpoise carcasses was 4.2x10-3 (S.E. ± 3.0x10-3 ). The mean marine mammal species diversity per year was 7 (S.E. ±1.0), cetacean species diversity per year was 2.5 (S.E. ±0.5), and pinniped species diversity per year was 4.5 (S.E. ±0.5) (Figure 7). During the two years of effort-based surveys, HSU s MMSP surveyors continued to respond to strandings that were reported by the public. On the same beaches we surveyed systematically, 13 carcasses (12%) were initially documented because of a report from the public (Figure 8). Of these 13 carcasses, six were cetaceans and seven were pinnipeds (Table 4). All of these carcasses were encountered on the systematic survey following the public report except for four that were removed from the beach during the initial response for further analysis. Opportunistic reports accounted for 55% (n=6) of the cetacean carcasses documented and 7% (n=7) of the pinnipeds documented (Figure 9). Harbor porpoises were the only cetacean species originally documented during effort-based. All other cetacean species were initially reported by the public.

45 Table 3. Number of carcasses of each cetacean species documented from October 2011 through September Species October October September 2012 September 2013 Total Phocoena phocoena Eschrichtius robustus Delphinus delphis Mesoplodon peruvianus Total

46 Number of species documented Marine mammals Cetaceans Pinnipeds Figure 7. The mean number marine mammal, cetacean, and pinniped species documented per year from October 2011 through September Error bars represent standard error.

47 Number of carcasses documented Systematic surveys Opportunistic reports 0 Month Figure 8. The average number of marine mammal carcasses documented per month from October 2011 through September 2013 because of opportunistic reports compared to the average number of marine mammal carcasses documented per month documented from effort-based surveys.

48 34 Table 4. The number of carcasses of each marine mammal species documented from opportunistic reports between October 2011 and September Species Opportunistic reports Zalophus californianus 2 Phoca vitulina 2 Unidentified otariid 1 Eumetopias jubatus 1 Mirounga angustirostris 1 Phocoena phocoena 2 Eschrichtius robustus 2 Delphinus delphis 1 Mesoplodon peruvianus 1

49 Number of carcasses documented Opportunistic reports Systematic surveys Total Marine mammals Cetaceans Pinnipeds Figure 9. Total number of marine mammals, cetaceans, and pinnipeds documented by opportunistic reports compared to the total number of marine mammals, cetaceans, and pinnipeds documented by HSU s MMSP during effort-based surveys between October 2011 and September 2013.

50 36 Of the 107 marine mammal carcasses we documented, 102 were tagged and left on the beach. Of these 102 marine mammals, 32 (31.4%) were documented at least one month after they were initially encountered (Figure 10). One Steller sea lion was documented seven months after it was initially encountered. Seasonal marine mammal stranding patterns during effort-based surveys California sea lion carcasses were predominately documented in the fall and winter (65%; n=19). In the fall, an average of 4.8x10-3 (S.E. ± 1.2x10-3 ) and in the winter, an average of 6.5x10-3 (S.E. ± 4.2x10-3 ) California sea lion carcasses were encountered per kilometer (Figure 11). Sex was determined for 83% (n=24) of the California sea lion carcasses, and of those, 88% (n=21) were male. Harbor seal carcasses were predominately encountered in the spring and summer (73%; n=19) (Figure 11). In the spring, 4.8x10-3 (S.E. ± 2.4x10-3 ) and in the summer, 6.5x10-2 (S.E. ± 5.9x10-4 ) harbor seal carcasses were encountered per kilometer. The highest number of pups was documented in April (n=3) and June (n=4), and 91.7% (n=11) were documented from March through September. During the two years of systematic surveys, HSU s MMSP surveyors documented eight Steller sea lion carcasses, seven harbor porpoise carcasses, and two gray whale carcasses. Though these sample sizes were not large enough to analyze seasonal stranding patterns, there seemed to be a trend for harbor porpoise strandings peaking in the summer (57%).

51 Number of carcasses Initial encounter One month Two months Three months Four months Five months Six months Seven months Figure 10. The number of marine mammal carcasses that were encountered on surveys after they were initially documented.

52 Mean encounter rate California sea lion Harbor seal Fall Winter Spring Summer Season Figure 11. Mean encounter rate of California sea lion carcasses per season compared to the mean encounter rate of harbor seal carcasses per season between October 2011 and September 2013.

53 39 Hotspot analysis for systematic surveys (October 2011-September 2013) Three hotspots and two coldspots were detected for marine mammal strandings using the Getis-Ord Gi* statistic (p < 0.05) (Figure 12). One hotspot was located on Centerville Beach (Beach #1) and two hotspots were located on segments of Table Bluff Beach (Beach #2). The two coldspots were located in the northern range of our surveys on Dry Lagoon (Beach #21) and Stone Lagoon (Beach #22). Historical data compared to systematic surveys All marine mammal carcasses documented between 1975 and 2010 were reported to the Humboldt State University s Vertebrate Museum by community members, and those documented between October 2011 and September 2013 included community reports and those from systematic beach surveys. Historical data were not collected using an effort-based protocol, so the total number of marine mammal carcasses documented per year was used for comparison between the two protocols. Between 1975 and 2010, 110 marine mammal strandings were documented within the area of HSU s MMSP s systematic surveys. Of the 110 marine mammal strandings, 55 were cetaceans and 55 were pinnipeds. The average number of marine mammal carcasses documented per year in the historical records was 57 (S.E. ± 0.48). The average number of cetacean carcasses documented per year was 6 (S.E. ± 0.26), and the average number of pinniped carcasses documented per year was 51 (S.E. ± 0.36). The most common marine mammal species documented from 1975 through 2010 was the California sea lion (n=24), followed by gray whale (n=20) (Figure 13).

54 Figure 12. Map of Humboldt County, California with hotspots and cold spots for number of marine mammal strandings identified by using the Getis-Ord Gi* statistic (ArcGIS 10.1). 40

55 Species Species 41 Pygmy beaked whale Baird's beaked whale beaked whale Historical data ( ) Pygmy beaked whale Baird's beaked whale beaked whale Effort-based surveys ( ) Stejneger's beaked whale Stejneger's beaked whale Cuvier's beaked whale Cuvier's beaked whale Short-finned pilot whale Short-finned pilot whale Pacific white-sided dolphin Pacific white-sided dolphin Common dolphin Common dolphin Dall's porpoise Dall's porpoise Pygmy sperm whale Pygmy sperm whale Humpback whale Humpback whale unidentified phocid unidentified phocid unidentified pinniped unidentified pinniped Northern fur seal Northern fur seal Orca Orca Risso's dolphin Risso's dolphin Hubb's beaked whale Hubb's beaked whale Fin whale Fin whale Sperm whale Sperm whale Northern elephant seal Northern elephant seal Steller sea lion Steller sea lion Harbor porpoise Harbor porpoise Gray whale Gray whale unidentified sea lion unidentified sea lion Harbor seal Harbor seal California sea lion California sea lion Number of carcasses Number of carcasses Figure 13. The number of carcasses of each species of marine mammal documented in the historical data from HSU s Vertebrate Museum between 1975 and 2010 compared to the species diversity documented during effort-based surveys between 2011 and 2013.

56 42 More marine mammal strandings were documented per year when systematic surveys were implemented. The average number of marine mammal carcasses documented per year from was significantly lower than the average number of marine mammals documented from (Two-sample t-test, t = , p < 0.01) (Figure 14). The average number of cetacean carcasses documented annually from was significantly lower than the average number of cetacean carcasses documented from (Two sample t -test, t = , p < 0.01) (Figure 14). The average annual number of pinniped carcasses documented from was significantly lower than the average number of pinniped carcasses documented from (Twosample t -test, t = , p < 0.01) (Figure 14). The mean number of marine mammal species per year documented from was significantly lower than the mean number of marine mammal species documented from (Two-sample t-test, t = , p < 0.01) (Figure 15). The mean number of cetacean species documented annually from was not significantly different than the mean number of cetacean species documented from (Two-sample t -test, t = , p > 0.05) (Figure 15). The mean number of pinniped species documented annually from was significantly lower than the mean number of pinniped species diversity documented from (Two-sample t - test, t = , p < 0.01) (Figure 15).

57 Number of carcasses documented per year * * Historical data ( ) Effort-based surveys ( ) * 5 0 Marine mammal carcasses Cetacean carcasses Pinniped carcasses Figure 14. The mean number of marine mammal, cetacean, and pinniped carcasses documented per year in the historical records (1975 through 2010) compared to the mean number of marine mammal, cetacean, and pinniped carcasses documented per year during effort-based surveys (October 2011 through September 2013). Error bars represent standard error. *Indicates a significant difference (p<0.05) between 1975 and 2010 compared to 2011 through 2013.

58 Number of species documented per year * Marine mammals Cetaceans Pinnipeds * Historical data ( ) Effort-based surveys ( ) Figure 15. The mean number of marine mammal, cetacean, and pinniped species documented per year in the historical records (1975 through 2010) compared to the mean number of marine mammal, cetacean, and pinniped species documented per year during effort-based surveys (October 2011 through September 2013). Error bars represent standard error. *Indicates a significant difference (p<0.05) between 1975 and 2010 compared to 2011 through 2013.

59 45 Seasonal stranding patterns from the historical data compared to effort-based surveys Between 1975 and 2010, California sea lion carcasses were predominantly documented in the summer and fall (65%; n=15) (Figure 16). Harbor seal carcasses were predominantly documented in the spring and fall (89%; n=17) (Figure 17). Between 1975 and 2010, 78% (n=15) of the harbor seal carcasses were documented from March through September, and harbor seal pups were only documented from March through May. Historically, gray whale strandings were reported primarily in the spring and summer (82%; n=22) (Figure 18). Harbor porpoises were predominately documented in the summer (64%; n=14) (Figure 18). Of the harbor porpoises documented in the summer, 79% (n=11) were documented in July and August. All gray whale strandings were documented from winter through summer, with no documented strandings in the fall. The highest number of gray whale strandings were documented in April (n=4), May (n=5), and June (n=7).

60 Number of carcasses documented Historical data ( ) Effort-based surveys ( ) 2 0 Fall Winter Spring Summer Season Figure 16. Total number of California sea lion carcasses documented per season in the historical records (1975 through 2010) compared to the total number of California sea lion carcasses documented per season during effort-based surveys (October 2011 through September 2013).

61 Number of carcasses documented Historical data ( ) Effort-based surveys ( ) 2 0 Fall Winter Spring Summer Season Figure 17. Total number of harbor seal carcasses documented per season in the historical records (1975 through 2010) compared to the total number of harbor seal carcasses documented per season during effort-based surveys( October 2011 through September 2013).

62 Number of carcasses documented Gray whale Harbor porpoise Fall Winter Spring Summer Season Figure 18. Total number of gray whale carcasses documented per season in the historical records (1975 through 2010) compared to the total number of harbor porpoise carcasses documented per season in the historical records (1975 through 2010).

63 49 Seabird Strandings The mean monthly encounter rate for seabird carcasses between July 2012 and September 2013, was 1.79 (S.E. ±0.36) (Figure 19). There was a difference between the seabird carcass encounter rate from July through September 2012 compared to the encounter rate from July through September This discrepancy was likely due in part to a brown pelican UME. Common murres were the most frequently documented species (Figure 20), with a monthly encounter rate of 0.84 (S.E. ±0.22) carcasses, followed by brown pelicans with a monthly encounter rate of 0.17 (S.E. ±0.07) carcasses, large immature gulls with a monthly encounter rate of 0.14 (S.E. ±0.04) carcasses, and sooty shearwaters with a monthly encounter rate of 0.10 (0.07) carcasses. HSU s MMSP method for grebe species identification did not differentiate Western grebes (Aechmophorus occidentalis) from Clark s grebes (Aechmophorus clarkia). Therefore, all Western or Clark s grebes were classified as Western grebes. Ninety-two seabirds were classified as Western grebes, but because HSU s MMSP could not verify that the carcasses were Western grebes and not Clark s grebes, all 92 seabirds were reclassified as unidentified large grebes. Seasonal patterns for seabird strandings documented during effort-based surveys The greatest number of seabird carcasses was encountered in the fall. The most common species, common murres, were also encountered most often in the fall. The seasonal trend of spikes in strandings in the summer and fall for common murres followed the overall seasonal trend for seabird strandings (R 2 =0.82) (Figure 21). Large

64 Encounter rate July 2012-June 2013 July 2013-September 2013 Month Figure 19. Seabird encounter rate during effort-based surveys from July 2012 through September 2013.

65 Species 51 Glaucous-winged gull Dunlin Red-necked grebe White-winged scoter Sanderling California gull Black-legged kittiwake Caspian tern Red-throated loon Unidentified shorebird Common loon Unidentified larid Black-footed albatross Red Phalarope Unidentified duck Unidentified procellarid Unidentified raptor Mallard Surf scoter Pigeon guillemont Canada goose Unidentified shearwater Herring gull Double crested cormorant Unidentified alcid Short tailed shearwater Pacific loon Unidentified cormorant Heerman's gull Rhinoceros auklet Northern fulmar Pelagic cormorant Brandt's cormorant Western gull Unidentified seabird Unidentified grebe Unidentified gull Brown pelican Sooty shearwater Large immature gull Common murre Number of carcasses documented Figure 20. Total number of carcasses of each seabird species documented during effortbased surveys from July 2012 through September 2013.

66 Seabird encounter rate R² = Common murre encounter rate Figure 21. Correlation between the encounter rate for all seabird carcasses and the encounter rate for common murre carcasses during effort-based surveys from July 2012 through September 2013.

67 53 immature gull carcasses were most commonly encountered in the summer (Figure 22 a). Sooty shearwater and brown pelican carcasses were most commonly encountered in the summer (Figure 22 a). Grebes were most commonly encountered in the winter (Figure 22 b). Hotspot Analysis The beach segment with the highest encounter rate of seabird carcasses was the Clam Beach to Moonstone segment (Figure 23). Clam Beach to Moonstone had an average encounter rate of 6.31 (S.E. ±1.68) seabird carcasses (Figure 24). The Clam Beach to Moonstone segment was also a hotspot for common murres (Figure 25). Clam Beach to Moonstone had an average encounter rate of 2.41 (S.E. ±0.80) common murre carcasses, and Clam Beach to the mouth of the Mad River had an average encounter rate of 2.32 (S.E. ±1.00) common murre carcasses (Figure 26). Comparing COASST Survey Protocol to HSU s MMSP Survey Protocol Between October 2012 and September 2013, COASST surveyors conducted surveys that partially overlapped with 13 of HSU s MMSP s beach segments (Figure 27). Overall, COASST volunteers conducted more surveys, and HSU s MMSP volunteers surveyed a greater distance (Table 5). On surveys that overlapped with at least part of a COASST survey, HSU s MMSP volunteers documented 35 marine mammal carcasses, and the average encounter rate for marine mammal carcasses 0.05 (S.E. ±0.01). Of these 35 marine mammal carcasses,

68 Mean encounter rate Mean encounter rate Sooty shearwaters Brown pelicans (a) Month (b) Month Figure 22. Mean encounter rate of (a) sooty shearwater and brown pelican carcasses and (b) grebe carcasses each month during effort-based surveys from July 2012 through September 2013.

69 Figure 23. Map showing the beach segment (Clam Beach to Moonstone) that was a hotspot for seabird strandings from July 2012 through September