MARINE BIOLOGICAL INVASIONS: THE DISTRIBUTIONAL ECOLOGY AND INTERACTIONS BETWEEN NATIVE AND INTRODUCED ENCRUSTING ORGANISMS by CHAD LeROY HEWITT A DISSERTATION Presented to the Department ofBiology and the Graduate School of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy August 1993 APPROVED: Dr. I. Lorraine Heisler 11 '" III An Abstract of the Dissertation of Chad LeRoy Hewitt in the Department of Biology for the degree of to be taken Doctor of Philosophy August 1993 Title: MARINE BIOLOGICAL INVASIONS: THE DISTRIBUTIONAL ECOLOGY AND INTERACTIONS BETWEEN NATIVE AND INTRODUCED ENCRUSTING ORGANISMS Approved: Dr. I. Lorraine Heisler Whether community membership is of limited or unlimited nature is a longstanding issue in ecology. Assembly studies have provided insight into the contributions of competition, diversity, and history on the development of community structure. In these studies however, the colonizers are drawn from the same species pool in which all members have had an evolutionary history. Thus interacting species have potentially evolved life history strategies in response to one another and have altered the resistance of native assemblages to species insertion. The human-mediated introduction of species provides an opportunity to ask questions pertaining to the resistance or susceptibility of communities to invasion in the absence of co-evolved traits. Whether a co-adapted, potentially co-evolved species pool can resist the invasion of a species with which none of the community members have had evolutionary "experience" has rarely been experimentally examined. The marine encrusting communities of Coos Bay estuary, Oregon, have been and continue to be inocculated by non-indigenous species from a range of donor regions These communities form two distinct clusters dominated either by native species limited to marine sites in the lower bay or introduced species in the brackish waters of the upper bay IV The settlement phenologies of native and introduced species in both the lower and upper bay exhibit significant differences in the duration, timing, and density of settlement. Questions pertaining to the resistance of native communities to invasion by non- indigenous species were examined in the face of catastrophic disturbance and established adult assemblages. In the presence of introduced species, disturbed patches (bare settlement panels) quickly attained a species equilibrium and diversity. In contrast native communities continued to gain species after 17mo. Experimental manipulation of established native and invaded assemblages (14mo) such that native assemblages were placed in direct contact with invaded assemblages in 25%, 50% and 100% treatments allowed adult-adult interactions to be controlled. These manipulated assemblages were then reciprocally transplanted between the two sites. Early mortality of introduced species and lack oflarval input at the native site resulted in reduced invader success. At the invaded site invasion success was correlated with two factors, initial native space occupancy and invaded assemblage treatment density. VITA NAME OF AUTHOR: Chad LeRoy Hewitt PLACE OF BIRTH: Bakersfield, California DATE OF BIRTH: December 17,1960 GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon University ofHawaii, Manoa University of California, Berkeley DEGREES AWARDED: Doctor ofPhilosophy, 1993, University of Oregon Bachelor of Arts, 1984, University of California, Berkeley AREAS OF SPECIAL INTEREST: Marine Community Ecology Community Selection and Evolution Introduced and Colonizing Species PROFESSIONAL EXPERIENCE: Teaching Assistant, Department ofBiology, Oregon Institute ofMarine Biology, University of Oregon, Eugene, 1987 - 1993 Research Assistant, Department of Biology, Oregon Institute ofMarine Biology, University of Oregon, Eugene, 1988, 1991 - 1992 Field Biologist, Marzet Incorporated, 1991-1993, P.I.: Dr. Dan Varoujean Field Zoologist, UC Davis and National Park Service, 1991, P.I.: Dr. Tom Suchanek v VI Field Zoologist, Ecological Consulting Incorporated, 1990, P.I.: Dr. Glenn Ford Field Zoologist, Dames and Moore Environmental Firm, 1989, P.I.: Dr. Tom Suchanek Teaching Assistant, Department ofZoology, University ofHawaii, Honolulu, 1985 - 1987 Teaching Assistant, Research Diving Program, University of California, Berkeley, 1985 AWARDS AND HONORS: Walter Moberly Memorial Award 1989 PUBLICATIONS: Boero, F. and C.L. Hewitt. 1992. A hydrozoan, Zanclella bryozoophila, n.g., n.sp., symbiotic with a bryozoan, with a discussion of the Zancleoidea. Can. J. Zool. 70: 1645-1651. VB ACKNOWLEDGEMENTS Several people have contributed, guided and encouraged me during the course of my academic career. These individuals have molded my love of invertebrates and of the marine environment to such an extent that I feel compelled to acknowledge their roles here. I would like to thank Chuck Fisk for his effervescent humor and love of all wild things, including a small boy with endless questions. Jim Rossas, Jack Grosse, and Ron Loch for friendship and guidance. My parents have always been there to support my academic yearnings. They have supported me in many ways. My father's love ofboth knowledge and books, and his father's love of the sea, have found places in my heart. I would like to thank them for their willingness to allow a child to pursue a dream, even though it took twenty-five years. Wayne Sousa provided the knowledge to help fill in the gaps, and gave freely of his limited time. During my stay at the Bodega Marine Laboratory 1 was befriended by rNando Boero and Cadet Hand who taught me to appreciate little things as well the big ones. I am deeply grateful to both of them for sharing their love of the hydroids. The late Ralph Smith was one of the most amazing professors of invertebrate zoology I shall ever meet, and has inspired me always. I will miss his keen mind and flying erasers. Throughout the process of this doctoral program there were innumerable people to whom I have become indebted. First I would like to thank the many people who have given oftheir time and energy to help with field work. These include (but are not limited to) Doug Warrick, Dan Varoujean, Kraig Slack, and the scores ofgraduate and undergraduate students that filtered through aIMH only to be exploited by my relentless need during the spring tides. Vlll Portions of this study were funded by the Sigma-Xi Grants in Aid-of-Research and the Lerner Grey Fund for Marine Biology. Support was also provided by the employment opportunities provided by Tom Suchanek, Glenn Ford, and Dan Varoujean. Lynda Shapiro provided research assistantships, but more importantly allowed me to use her image analysis system, without which much of this research would not have been done. Jerry Medler and Nathan Tublitz for sitting in on committee meetings and providing me with feedback that helped shape the dissertation. Lorraine Heisler and Peter Frank for their willingness to accept the responsibility for my education. Peter's advice and prodding have been greatly appreciated; I am overjoyed to have been able to interact with him. Lorraine Heisler, my adoptive advisor, has provided me with hours of stimulating discussion and much food for thought on a diverse array of subjects. She has also been one of my best critics, forcing me to think critically about my own work. I am honored to have been hooded by her. Jim Carlton, who introduced me to the various concepts of introductions and the ramifications of introduced species, has been my mentor. I appreciate his ability to guide yet allow room to pursue a subject near to his heart without restriction. The interactions between my committee members have been exemplary, they truly worked together to provide a cohesive, supportive unit. Lastly I wish to thank my wife, Vicki Hewitt for her unflagging support, love, understanding, and perseverance during the last six years. She has stood by me while I have explored the realm of the ocean at the oddest hours, seen me off on various plane trips to give talks, present posters or to work. She has graciously listened while I talked (ad nauseum) of marine invertebrates, and most importantly has been there when I felt like not going on. This degree, in many respects, is as much hers as it is mine. Thank you. While I acknowledge the debt lowe to all of the aforementioned for the successes of this dissertation, any failings that appear herein are solely my responsibility. Chapter IX TABLE OF CONTENTS Page I. THE DISTRIBUTION AND ECOLOGY OF NATIVE AND INTRODUCED ENCRUSTING ORGANISMS IN THE COOS BAY ESTUARY . 1 Introduction 1 Study Area 4 Mechanisms of Biological Invasions 4 Materials and Methods 9 Results 13 Discussion 28 II. THE ASSEMBLY OF ENCRUSTING COMMUNITIES IN THE LOWER COOS BAY ESTUARY . 83 Introduction 83 Study Sites 88 Materials and Methods 89 Results 96 Discussion 118 III. AN EXPERIMENTAL STUDY OF NATIVE COMMUNITY INVASION BY NON-NATIVE SPECIES 169 Introduction 169 Study Sites 176 Materials and Methods 176 Results 182 Discussion 198 APPE1-.TDIX A. B. COOS BAY TRANSECT RAW DATA .. RAW DATA FOR COMMUNITY ASSEMBLY EXPERIMENT . 245 252 c. D. RAW DATA FOR NORTH JETTY TRANSPLANT EXPERIMENT . RAW DATA FOR POINT ADAMS JETTY TRANSPLANT EXPERIMENT . x 261 272 LITERATURE CITED . 283 Xl LIST OF TABLES Table Page 1. Mechanisms ofEncrusting Species Introduction into Coos Bay. Receiving areas are as follows: CR, Coos River; VB, Upper Bay; MB, Middle Bay; LB, Lower Bay; and SS, South Slough 5 2. Sample Sites within Coos Bay. Transect names are CB, Coos Bay and SS, South Slough. Region Codes Follow Table 1 .. 10 3. Physical Measurements for Coos Bay Stations 14 4. Native Species List for Five Regions of Coos Bay. Region Codes Follow Table 1. Asterisks Identify Possible Introductions 16 5. Introduced and Cryptogenic Species List for Five Regions of Coos Bay. Region Codes Follow Table 1. Asterisks Identify Possible Natives 19 6. Botrylloides violaceus Dispersal Following Secondary Introduction. Mechanism Codes Follow Table 1 22 7. Percentage ofTotal, Native and Introduced Species Recruiting Per Month in the Lower and Upper (Isthmus Slough) Coos Bay Estuary 26 8. Non-parametric ANOVA Table Comparing Ranked Native and Introduced Recruitment Rates and Month 29 9. Native Species Presence/Absence in Community Assembly Experiment 1 at the North Jetty Between April 1989 and August 1990 97 10. Native Species Presence/Absence in Community Assembly Experiment 1 at the Point Adams Jetty Between April 1989 and August 1990 98 XII 11. Introduced Species Presence/Absence in Community Assembly Experiment 1 at Both Sites Between April 1989 and August 1990 99 12. Mobile Fauna Species List 101 13. Mean Percent Cover for All Species at the North Jetty During Experiment 2 Between September 1990 and February 1992 103 14. Mean Percent Cover for All Species at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992 104 15. Summary ofMonthly Community Richness Statistics During Experiment 2 Between September 1990 and February 1992 106 16. Stochastic Boundedness of Species Richness for North Jetty and Point Adams Jetty Replicates During Experiments 1 and 2. Panel Replicates are Denoted by Letter, S is the Mean Species Richness, CV is the Coefficient of Variation, and N is the Number of Sample Periods. See Text for Description of Stochastic Boundedness 111 17. Means ofInitial Native Community Statistics for the Transplant Experiment Between September 1990 and February 1992. Standard Deviations are in Parentheses Below the Mean 183 18. Mean Percent Cover for the 25% Transplant Treatment at the North Jetty. Months and Days are from the Beginning of the Experiment (September 1990) 186 19. Mean Percent Cover for the 50% Transplant Treatment at the North Jetty. Months and Days are from the Beginning of the Experiment (September 1990) 187 20. Mean Percent Cover for the 100% Transplant Treatment at the North 'etty. Months and Days are from the Beginning of the Experiment (September 1990) 188 Xl1l 21. Proportion ofReplicates Each Introduced Species Invaded During the Transplant Experiment. Replicate Numbers in Parentheses Underneath the Treatments .................................. 189 22. Mean Percent Cover for the 25% Transplant Treatment at the Point Adams Jetty. Months and Days are from the Beginning of the Experiment (September 1990) ....................... 191 23. Mean Percent Cover for the 50% Transplant Treatment at the Point Adams Jetty. Months and Days are from the Beginning of the Experiment (September 1990) ....................... 192 24. Mean Percent Cover for the 100% Transplant Treatment at the Point Adams Jetty. Months and Days are from the Beginning of the Experiment (September 1990) ....................... 193 25. Mean Percentage of the Dominant Introduced Species' Areas Derived From Recruitment or Immigration. A) 25% Treatment; B) 50% Treatment; and C) 100% Treatment ......................................................... 195 26. Ranked Averages ofIntroduced Species Abundances on Replicate Panels for Two Time Periods: Early Period, December 1990 to February 1991; and Late Period, December 1991 to February 1992. A) North Jetty 25% and 50% Treatments; and B) Point Adams Jetty 25%, 50% and 100% Treatments 197 LIST OF FIGURES XlV Figure Page 1. Inset Map (A) of Oregon with Coos Bay Circled. Map of Coos Bay (B) with Sample Stations and Major Geographical Locations. Sample Station Codes Follow Table 2. Scale Equals 3 km ................................................ 38 2. Average Temperature and Salinity Measurements Along the Coos Bay Transect During July 1989. Solid Line Represents Salinity (ppt) and the Line with Closed Circles Represents Temperature caC) .............................................. 39 3. Physical Measurements from the Ocean Station (Site 2): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature °C); b) Hydroclimagraph of Site 2, January (1) and December (12) are Labelled 40 4. Physical Measurements from the Charleston Boat Basin (Site 15): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature °C); b) Hydroclimagraph of Site 15, January (1) and December (12) are Labelled ................................................. 42 5. Physical Measurements from the South Slough(Site 18): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature °C); b) Hydroclimagraph of Site 18, January (1) and December (12) are Labelled ................................................. 44 6. Physical Measurements from the Haynes Inlet (Site 9): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature °C); b) Hydroclimagraph of Site 9, January (1) and December (12) are Labelled ................................................. 46 xv 7. Physical Measurements from the Isthmus Slough (Site 12): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature DC); b) Hydroclimagraph of Site 12, January (1) and December (12) are Labelled 48 8. Maximum and Minimum Annual Temperature Values (Solid Lines) and Temperature Range (Triangles and Dotted Regression Line) as a Function ofDistance from Ocean (temperature range = 0.51 e distance + 3.66) 50 9. Maximum and Minimum Annual Salinity Values (Solid Lines) and Annual Salinity Range (Triangles and Dotted Regression Line) as a Function ofDistance from Ocean (salinity range = 8.04 e e(distance) +2.48) 51 10. Frequency of Native and Introduced (and Cryptogenic) Encrusting Species by Taxa 52 11. Pie Diagrams ofNative and Introduced (and Cryptogenic) Encrusting Organisms by Taxa 53 12. . Frequency ofIntroduced and Cryptogenic Species By Taxa 13. Introduced Species Abundances in Five Regions of Coos Bay According to Their Affinities with Specific Introduction Mechanisms. Mechanisms: I, Wooden Hulled Vessel Fouling; II, Atlantic Oyster Culture; III, Japanese Oyster Culture; IV, Modern Introduction Mechanisms (See Text) . 14. Native Species Richness Correlated with Distance (km) from the Ocean (pooled r2 = 0.65, n = 20). In(Native Spp Richness) = -0.2(km from Ocean) + 3.65 15. Native Species Richness as a Function of Annual Salinity Range at Each Site (pooled r2 = 0.64, n = 20). In(Native Spp Richness) = -0.2(Salinity Range) + 6.43 16. Coos Bay Encrusting Community ~-Diversity. Sorensen's Between-Site Similarity as a Function of the Difference Between Site Specific Annual Salinity Range 54 55 56 57 58 XVI 17. Introduced Percentage of the Community as a Function of Distance From the Ocean (pooled r2 = 0.95, n = 20). (Introduced Spp %) = 3.32(km from Ocean) + 1.19 59 18. Introduced Percentage of the Community as a Function of Native Species Richness (pooled r2 = 0.70, n = 20). In(Introduced Spp %) = -0.04(Native Spp Richness) + 4.03 .. 60 19. Similarity Dendrogram Based on S0rensen's Similarity Measure Comparing Species Presence/Absence Data for all Sites in Coos Bay. Dendrogram was Constructed Using UPGMA Clustering Technique 61 20. Monthly Recruitment Phenologies ofNative and Introduced Encrusting Organisms in Lower Coos Bay and Isthmus Slough (Upper Bay). Species Recruitment During the Study is Represented by Filled Boxes 62 21. Average Number of Species Settling per Month in the Lower Bay Between August 1988 and December 1990. Solid line represents all Species, Triangle Symbols are Introduced Species and Circles are Native Species 64 22. Average Number of Species Settling per Month in the Upper Bay Between August 1988 and December 1990. Solid line represents all Species, Triangle Symbols are Introduced Species and Circles are Native Species 65 23. Frequency of the Length of the Recruitment Period for Native and Introduced Species 66 24. Kite Diagram of Average Recruitment Rates (Recruits·Panel-1·Mo-l) for Native Species in the Lower Bay. All Species are Represented at the Same Scale, Asterisks Represent Recruitment at Low Densities 67 25. Kite Diagram of Average Recruitment Rates (Recruits·Panel-1·Mo-l) for Introduced Species in the Lower Bay. All Species are Represented at the Same Scale, Asterisks Represent Recruitment at Low Densities 68 XVII 26. Schizoporella unicornis Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years 69 27. Botrylloides violaceus Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 70 28. Botryllus schlosseri Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 71 29. Conopeum tenuissimum Recruitment (Average Number of RecruitsePanel-leMo-I) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 72 30. Bowerbankia gracilis Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years 73 31. Cryptosula pallasiana Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 74 XVIll 32. Alcyonidium polyoum Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 75 33. Distaplia occidentalis Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years 76 34. Balanus glandula Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 77 35. Bugula pacifica Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 78 36. Cheilopora praelonga Recruitment (Average Number of Recruitsepanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years 79 37. Spirorbid spp. Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years ......... 80 XIX 38. Serpulid spp. Recruitment (Average Number of Recruits·Panel-I.Mo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years 81 39. Hippotlwa hyalina Recruitment (Average Number of Recruits·Panel-1.Mo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years 82 40. Map of the Lower Coos Bay Showing the North Jetty (Site 3) and Point Adams Jetty (Site 14). Scale Equals l.OkIn 126 41. Army Corps of Engineers Flow Field Diagram for the Lower Coos Bay During Flood Tide (COE 1979) 127 42. Army Corps of Engineers Flow Field Diagram for the Lower Coos Bay During Ebb Tide (COE 1979) 128 43. Diagram illustrating the 2x2 Panel Array in Plan (A) and Side (B) Views and Concrete Block Design (C). See Text for Further Explanations of Design . 129 44. Species Accumulation at the North Jetty During Experiment 1 Between April 1989 and August 1990. Lines Represent Mean Species Richness (n =4) with Standard Deviation Error Bars ,. .. 130 45. Species Accumulation at the Point Adams Jetty During Experiment 1 Between April 1989 and August 1990. Lines Represent Mean Species Richness (n =4) with Standard Deviation Error Bars 131 46. Species Accumulation at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data 132 I f xx 47. Species Accumulation at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data 133 48. Shannon-Weaver Diversity (H') Change at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ............................................................................ 134 49. Shannon-Weaver Diversity (H') Change at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ........................................................................... 135 50. Evenness (1') Change at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ............................................................. 136 51. Evenness (1') Change at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ............................................................. 137 52. McNaughton's Dominance Index (MD) Change at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ............................................................. 138 53. McNaughton's Dominance Index (MD) Change at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data ............................................................. 139 XXI 54. The Number of Species That Comprise 75% of the Living Cover at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data 140 55. The Number of Species That Comprise 75% of the Living Cover at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data 141 56. The Proportion of Total Space ofIntroduced Species at the North Jetty and Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 142 57. Introduced Species Proportion of Species Diversity (H') Correlated with the Introduced Species Proportion of Community Richness. Triangles Represent Individual Replicate Monthly Community Samples. The Line Represents a Direct Correlation 143 58. The Number of Species that are Settled Upon (Fouled) by Species X as a Function of the Number of Species that can Settle Upon (Be Fouled By) Species X. Species Identified are Botrylloides violaceus, Bv; Botryllus schlosseri, Bs; Schizoporella unicornis, Su; Distaplia occidentalis, Do; Balanus glandula, Bgl; Serpulid spp., Serp; and Cheilopora praelonga, Cpo 144 59. The Association Between Native and Introduced Species Cover on Replicate Panels at Point Adams Jetty as a Function of the Period ofImmersion (see Text) 145 60. Experiment 1 Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regression are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis 146 XXII 61. Experiment 2 Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regression are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis 147 62. The Total Species Pool and Native-Only Species Pool Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regression are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis 148 63. The Total Species Pool and Native-Only Species Pool Immigration and Extinction Rate Regression. Immigration Regressions are Represented by Thick Lines; Extinction Regression are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis 149 64. Bray Curtis Mean Similarity Between Replicate Panels at the North Jetty During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean with Range Bars (n = 4) ISO 65. Bray Curtis Mean Similarity Between Replicate Panels at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean with Range Bars (n = 4) lSI 66. Bray Curtis Mean Similarity Between Replicate Panels Within and Between the North Jetty and Point Adams Jetty Sites During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean of Four Replicate Panels 152 67. Mean Cumulative Number of Species at the North Jetty and Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. Symbols Represent Replicate Panel Data, Lines Represent Means......................... 153 XXIII 68. Realized Species Richness as a Function of Cumulative Species Richness for the North Jetty and Point Adams Jetty During Experiment 2. Symbols Represent Monthly Panel Data; Thick Line Represents the North Jetty Regression and the Thin Line Represents the Point Adams Jetty Regression 154 69. Balanus glandula Settlement Densities During Experiment 2 Between September 1990 and February 1992. Symbols Represent Four Replicate Panels 155 70. Balanus glandula Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 156 71. Cohort Survival ofBalanus glandula Following Peak Recruitment in August and September 1991 During Experiment 2 at the North Jetty. Symbols Represent Data from Four Replicate Panels. Regression is Based on Data from all Four Panels 157 72. Obelia sp. Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 158 73. Hippothoa hyalina Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 159 74. Cheilopora praelonga Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 160 75. Schizoporella unicornis Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 161 76. Overgrowth Survival by Schizoporella unicornis Colonies as a Function of Time of Overgrowth . 162 XXIV 77. Eudistylia vancouverensis Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 163 78. Serpulid polychaetes Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 164 79. Distaplia occidentalis Proportion of Total Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 165 80. Botrylloides violaceus Proportion ofTotal Space During Experiment 2 at the North Jetty and Point Adams Jetty Between September 1990 and February 1992. Lines Represent the Mean with Standard Deviation Bars (n = 4) 166 81. Percent Cover ofThree Common Species at the North Jetty in August 1990 (Experiment 1), August 1991 (Experiment 2) and February 1992 (Experiment 2). Species are as Follows: Balanus glandula, Bgl; Cheilopora praelonga. Cp; and Hippothoa hyalina, Rh. ................................................... 167 82. Percent Cover ofFive Dominant Species at the Point Adams Jetty in August 1990 (Experiment 1), August 1991 (Experiment 2) and February 1992 (Experiment 2). Species are as Follows: Schizoporella unicornis, Su; Botrylloides violaceus, Bv; Cheilopora praelonga, Cp; Hippothoa hyalina, Hh; and Distaplia occidentalis, Do............................................... 168 83. Experimental Design for the Reciprocal Transplant Experiment: 100% Native (4 replicates), 50% Native and 25% Native Treatments (3 Replicates Each) at the North Jetty and Point Adams Jetty. For Explanation ofPanel Design See Text ................................................................................ 205 84. Species Accumulation in the 25% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 206 xxv 85. Species Accumulation in the 50% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 207 86. Species Accumulation in the 100% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 208 87. Shannon-Weaver Diversity (H') Change in 25% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 88. Shannon-Weaver Diversity (H') Change in 50% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 89. Shannon-Weaver Diversity (H') Change in 100% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 90. Evenness (J') Change in 25% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) . 91. Evenness (1') Change in 50% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) . 92. Evenness (J') Change in 100% Treatment at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) . 93. The Number of Species That Comprise 75% of the Living Cover in 25% Treatments at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) . 209 210 211 212 213 214 215 XXVI 94. The Number of Species That Comprise 75% of the Living Cover in 50% Treatments at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) . 216 95. The Number of Species That Comprise 75% of the Living Cover in 100% Treatments at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 217 96. Species Accumulation in the 25% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 218 97. Species Accumulation in the 50% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 219 98. Species Accumulation in the 100% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 220 99. Shannon-Weaver Diversity (H') Change in 25% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) " ... ......... 221 100. Shannon-Weaver Diversity (H') Change in 50% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 222 XXVII 101. Shannon-Weaver Diversity (H') Change in 100% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 223 102. Evenness (J') Change in 25% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 224 103. Evenness (1') Change in 50% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 225 104. Evenness (1') Change in 100% Treatment at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 226 105. The Number of Species That Comprise 75% of the Living Cover in 25% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 227 106. The Number of Species That Comprise 75% of the Living Cover in 50% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 3) 228 107. The Number of Species That Comprise 75% of the Living Cover in 100% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; The Line is the Mean (n = 4) 229 XXVIII 108. The Relative Contribution to Species Diversity in 25% Treatments at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. The Upper Line is the Mean HI Based on All Living Cover; The Lower Line Represents the Mean H' Based on Native Species Only; and the Area Between the Two Lines is the Contribution ofIntroduced Species to the Total H' (n = 3) 230 109. The Relative Contribution to Species Diversity in 50% Treatments at the North Jetty During the Transplant Experiment Between September 1990 and February 1992. The Upper Line is the Mean H' Based on All Living Cover; The Lower Line Represents the Mean H' Based on Native Species Only; and the Area Between the Two Lines is the Contribution ofIntroduced Species to the Total H' (n = 3) 231 110. Introduced Species Percent Contribution to Species Diversity at the North Jetty as a Function of the Introduced Species Percentage of the Community for 25% and 50% Treatments Between September 1990 and February 1992. Symbols Represent Individual Replicate Data; Lines Represent the Regressions (Thick Line for 25%; Dashed Line for 50%) 232 111. The Relative Contribution to Species Diversity in 25% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. The Upper Line is the Mean H' Based on All Living Cover; The Lower Line Represents the Mean H' Based on Native Species Only; and the Area Between the Two Lines is the Contribution ofIntroduced Species to the Total H' (n = 3) 233 112. The Relative Contribution to Species Diversity in 50% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. The Upper Line is the Mean HI Based on All Living Cover; The Lower Line Represents the Mean H' Based on Native Species Only; and the Area Between the Two Lines is the Contribution ofIntroduced Species to the Total H' (n = 3) 234 113. The Relative Contribution to Species Diversity in 100% Treatments at the Point Adams Jetty During the Transplant Experiment Between September 1990 and February 1992. The Upper Line is the Mean H' Based on All Living Cover; The Lower Line Represents the Mean H' Based on Native Species Only; and the Area Between the Two Lines is the Contribution ofIntroduced Species to the Total H' (n = 4) 114. Introduced Species Percent Contribution to Species Diversity at the Point Adams Jetty as a Function of the Introduced Species Percentage of the Community for 25%, 50%, and 100% Treatments Between September 1990 and February 1992. Symbols Represent Replicate Panel Data; Lines Represent the Regressions (Thick Line for 25%; Dashed Line for 50%; and Dotted Line for 100%) . XXIX 235 236 115. Change in Mean Introduced Species Area (cm2) at the Point Adams Jetty for 25% (A), 50% (B), and 100% (C) Treatments During the Transplant Experiment Between September 1990 and February 1992. Symbols Represent Replicate Panel Data; Solid Lines are the Mean Areas (n = 3 for 25% and 50% Treatments; n = 4 for 100% Treatments) 237 116. Change in Mean Introduced Species Area (cm2) for 25% Treatments at the Point Adams Jetty (n = 3): Abundances of Individual Introduced Species Between September 1990 and February 1992. Thick Line Represents Total Introduced Species Cover. Species are as Follows: Botrylloides violaceus, Bv; Schizoporella unicornis, Su; Botlyllus schlosseri, Bs; and Clyptosula pallasiana, Crypto , 238 117. Change in Mean Introduced Species Area (cm2) for 50% Treatments at the Point Adams Jetty (n = 3): Abundances of Individual Introduced Species Between September 1990 and February 1992. Thick Line Represents Total Introduced Species Cover. Species are as Follows: Botlylloides violaceus, Bv; Schizoporella unicornis, Su; Botrylllls schlosseri, Bs; and Ctyptosula pallasiana, Crypto 239 xxx 118. Change in Mean Introduced Species Area (cm2) for 100% Treatments with Low Initial Native Species Cover at the Point Adams Jetty ( n = 2): Abundances ofIndividual Introduced Species Between September 1990 and February 1992. Thick Line Represents Total Introduced Species Cover. Species are as Follows: Botrylloides violaceus, Bv; Schizoporella unicornis, Su; Botryllus schlosseri, Bs; and Cryptosula pallasiana, Crypto ............... 240 119. Change in Mean Introduced Species Area (cm2) for 100% Treatments with High Initial Native Species Cover at the Point Adams Jetty (n = 2): Abundances ofIndividual Introduced Species Between September 1990 and February 1992. Thick Line Represents Total Introduced Species Cover. Species are as Follows: Botrylloides violaceus, Bv; Schizoporella unicornis, Su; Botlyllus schlosseri, Bs; and Cryptosula pallasiana, Crypto ............... 241 120. Change in Mean Percentage of Total Introduced Species Area that is Derived from Recruitment for 25%, 50% and 100% Treatments at the Point Adams Jetty Between September 1990 and February 1992 (n = 3 for 25% and 50% Treatments; n = 4 for 100% Treatments) ......................... 242 121. Change in Mean Percentage of All Botlylloides violaceus Area that is Derived from Recruitment for 25%, 50% and 100% Treatments at the Point Adams Jetty Between September 1990 and February 1992 (n = 3 for 25% and 50% Treatments; n = 4 for 100% Treatments) ......................... 243 122. Change in Mean Percentage ofSchizoporella unicornis Area that is Derived from Recruitment for 25%, 50% and 100% Treatments at the Point Adams Jetty Between September 1990 and February 1992 (n = 3 for 25% and 50% Treatments; n = 4 for 100% Treatments) ...................... 244 1CHAPTER I THE DISTRIBUTION AND ECOLOGY OF NATIVE AND INTRODUCED ENCRUSTING ORGANISMS IN THE COOS BAY ESTUARY Introduction Interest in the effect of introduced species on ecosystem function has risen since Elton's (1958) seminal monograph The Ecology ofInvasions by Plants and Animals, as documented by the number of volumes dealing with this question published during the last decade (Groves and Burdon 1986; Mooney and Drake 1986; Kornberg and Williamson 1987; MacDonald et al. 1987; Joenje et al. 1987; Drake et al. 1989; di Castri et al. 1990; Groves and di Castri 1991; Rosenfield and Mann 1992; Mills et al. 1993; Nalepa and Schloesser 1993). Despite the broad geographic scope of these works (Britain, North America, Hawaii, Australia, South Mrica, mainland Europe) this recent attention has focused in large part on freshwater and terrestrial systems, with little attention given to the nearshore marine environment (Carlton 1989, 1992b). In part this may be due to the lack of a clear consensus as to what constitutes the native·biota in coastal ecosystems. Most biological surveys of coastal marine communities commenced well after many invasions occurred (Carlton 1979a, b, 1989, 1992b; di Castri 1989, 1990). Once biological surveys were conducted, the assumption that the organisms encountered were endemic until proven to be introduced has resulted in large numbers of widely distributed, "cosmopolitan" species whose biogeographic origins are largely unknown (cryptogenic species sensu Carlton 1979b; Chapman and Carlton 1991). Consequently our understanding of the degree to which marine communities in general have been biologically altered by human-mediated invasions is minimal. 2Several general patterns that mayor may not be transferable to the marine environment can be distilled from recent work on invasions in freshwater and terrestrial communities. That certain kinds of communities or ecosystems are more susceptible than others to invasion by non-native species is now well documented (Elton 1958; Carlton 1979a, b, 1989, 1992a, b; Fox and Fox 1986; Crawley 1987; Carlton et al. 1990). For example, islands, with their relatively depauperate faunas, are readily invaded by new species (Elton 1958; Pimm 1987). These island biotas are continually changing around an equilibrium number of species due to species' invasion and extinction events on the local (island) scale (MacArthur and Wilson 1967). Similarly, the effect of disturbance (both local and regional) in "resetting" a community to an earlier successional state has been linked to the success of invasions (Fox and Fox 1986). Species-poor regions such as the post-glacial Northwest Atlantic and the Laurentian Great Lakes (two regionally disturbed areas) have received considerable numbers of invaders (Vermeij 1991; Rosenfield and Mann 1992; Mills et al. 1993; Nalepa and ScWoesser 1993). The converse of this last pattern is that certain communities "resist" the invasion of exotics. Elton (1958) proposed that the ecological resistance of communities to invasion by exotics was related to native species diversity (i.e., richness). The Eltonian idea of ecological resistance is distinct from the dynamic equilibrium ofMacArthur and Wilson (1967). Elton suggested that species rich communities will prevent the establishment of exotic species through an intricate series of synecological interactions including (but not limited to) competition with existing natives, predation by natives, parasitism and disease. Thus in this view recipient communities selectively allow entry into the system but are closed to the majority of invaders (closed systems: Roughgarden 1989). In contrast, MacArthur and Wilson (1963, 1967) proposed that communities were unlimited membership systems (Roughgarden 1989) in which the leading determinant ofinvasion success is not based on recipient community attributes, but on the dispersal ability of the 3invader species. Once a species has arrived in the new community, it is assumed that synecological assortative processes return the system to an "equilibrium" number of species. On the Pacific coast the majority of biological introductions have been in bays and estuaries with relatively few successful open coast introductions (Carlton 1974, 1979a, b, 1989). The large estuaries of the Pacific Northwest are geologically young (10,000 - 15,000 yrs old; Atwater et al. 1977; Carlton 1979a, 1992b; Ricketts et al. 1985; Nichols and Pamatmat 1988) and consequently may have had an impoverished native brackish water fauna. It has been proposed that the "immature" (i.e., in the process of assembly), depauperate communities of these estuaries are more susceptible to the introduction of species from other regions of the world (Carlton 1974, 1979a, b, 1992b; Nichols and Thompson 1985) than the mature, species rich communities of the open coast. Alternatively, this apparent estuarine susceptibility may be an artifact of the availability of dispersal mechanisms (see below) that are more likely to transport estuarine species than than species from open coastal marine communities (Carlton 1979b, 1992b). A faunal survey of the sessile, encrusting organisms (the "fouling" community) of Coos Bay, Oregon USA, was undertaken with two aims: 1) to determine the distribution of the native and introduced species along an estuarine gradient that is similar to those described for other Pacific coast estuaries; and 2) to discern any correlation between native and introduced species distributions within Coos Bay. Thirdly the monthly recruitment patterns in lower and upper bay communities were examined to detect differences in the phenology of recruitment for native and introduced encrusting species. 4Study Site The Coos Bay estuary, Oregon (43 0 19' 30"N, 1240 19' 30"W) comprises two sub- estuaries: South Slough, a drowned syncline of the Empire Formation and the Coos River, an L-shaped bar-built estuary (see Figure 1). In the past these two estuaries had separate entrances but the southward extension of the sand spit (North Spit) shifted the Coos River mouth southward until it merged with South Slough. Combined, the present-day estuary covers approximately 10,000 acres (see Figure 1) and is classified as a well-mixed, drowned river mouth. The Coos River discharges between 90 and 5,500 c.fs. seasonally, averaging 2,200 c.fs. South Slough has few minor tributaries (Winchester Creek) from its watershed of25,000 acres. Coos Bay is a heavily man-modified estuary and has lost approximately 90% of the tidal habitat (e.g., mud and sand flats, salt marshes) to land reclamation, diking and filling (Hoffnagel and Olson 1974). Mechanisms ofBiological Invasions Four temporally distinct mechanisms of species introduction into Coos Bay have operated between 1850 and the present and have been elucidated for other estuaries such as San Francisco Bay (Carlton 1979b, 1987, 1989, 1992b). These mechanisms include I) wooden ship hull foulinglboring; II) Atlantic oyster (Crassostrea virginica) culture; III) Japanese oyster (c. gigas) culture; and IV) modern mechanisms which include ballast water transport, modern mariculture practices, and intra-coastal and intra-estuarine ship traffic. Table 1 summarizes the approximate periods of operation for each of these mechanisms and the areas of Coos Bay that have differentially been affected. Wooden-hulled vessels (mechanism I) operated between Coos Bay and San Francisco Bay, Portland, and Puget Sound between 1853 (initial settlement of Coos Bay by western Table 1. Mechanisms ofEncrusting Species Introduction into Coos Bay. Receiving areas are as follows: CR, Coos River; UB, Upper Bay; MB, Middle Bay; LB, Lower Bay; and SS, South Slough. Range of Transport ID Mechanism Period Inter-provincial Regional Local Donor Region Receiving Area >2000km 2000km> <20km <20km in Coos Bay Wooden Vessel Shipping 1850's - 1950's 0 0 0 Atlantic, Europe SS, LB, MB, UB II Atlantic Oyster Culture 1870's - 1930's 0 0 0 Atlantic, Europe SS,MB,UB III Japanese Oyster Culture 1930's - 1950's 0 0 0 Japan/Asia SS,MB,UB IV Modern Mechanisms a) Ballast Water 1940's - present 0 0 Asia, Australia LB,MB,UB b) Other Mariculture present 0 0 - SS,MB,UB c) Coastwise Shipping 1850's - present 0 0 - ALL Fishing Boats Private Craft Replica Vessels V1 6man) and the 1950's (Kemble 1957; Douthit 1986). This coastwise shipping (between bays and estuaries of the Pacific Coast) was initially restricted to the lower Coos Bay (Empire to the mouth) but in 1856 the waterfront of North Bend (a mill town) was developed for timber shipping to San Francisco Bay. The Collector of Customs for the Coos Bay harbor district reported that between 1867 and 1874 a total of227 coastwise ships had arrived and departed Empire City - an average of28 vessels per year. By 1883 the annual number of ships had risen to 167 (Douthit 1981, 1986; Case 1983). The coastwise traffic increased to accommodate the coal and timber industries as well as passenger transport necessitating harbor improvements, including entrance jetties (North Jetty 1890's; South Jetty 1920's) and dredging up to Isthmus Slough. Over 100,000 tons of coal were extracted between 1896-1897, the bulk of which was shipped to San Francisco (Baldwin 1981; Case 1983). Much of the timber logged and milled in Coos Bay was shipped south in the earlier period by the lumber brigs due to their superior handling in the Coos Bay entrance, but as jetty placements stabilized the entrance they were soon replaced by coastwise lumber schooners (Kemble 1957). Consequently the intra-coastal trade may have distributed much of the San Francisco Bay exotic fauna to Coos Bay via hull fouling, until the late 1930's. The first transoceanic vessel arrived in Coos Bay from Japan during 1921 (Case 1983). In the following years trade with Asian ports increased to a mean of 133,000 tons (s.d. = 29,000) until World War II. During the post-war years (1946 - 1952) foreign shipping in Coos Bay rose to a mean of243,000 tons (s.d. = 80,000) with foreign imports remaining below 500 tons (Case 1983). Thus between the mid 1920's and the 1950's wooden hulled vessels had the potential to transport Asian hull fouling and boring organisms to Coos Bay directly. Atlantic oyster culture (mechanism II) appears to have been limited to a single inoculation event at the entrance of South Slough in 1872 (Douthit 1986; P. Baker pel's. 7comm.), although numerous undocumented movements of this species may have occurred. Consequently while this mechanism may not have directly introduced species into Coos Bay, Atlantic oyster culture is known to have introduced scores of species to primary receiving regions including San Francisco Bay, Humboldt Bay, and Willapa Bay from 1890 to the 1930's (Carlton 1974, 1979b). From these primary receiving areas numerous encrusting species may have been secondarily spread via coastwise shipping traffic (wooden hulled fouling: mechanism I) and later, with Japanese oyster culture (mechanism III). This secondary invasion step would necessarily place additional filters or bottlenecks (Carlton 1979b, 1987) on the Atlantic species transported to Coos Bay and may explain the absence from Coos Bay of species that are present in San Francisco Bay or Willapa Bay. The Japanese oyster, Crassostrea gigas, was first introduced to the larger bays of the Pacific Coast in 1919 (Steele 1964). These introductions of adult oyster communities (the oysters and encrusting and infaunal associates: mechanism III) continued until the 1950's when the practice of rearing spat negated the need to ship adults (Carlton 1979b; Qualman 1983). The sources of C. gigas material moved into Coos Bay have been varied and include Willapa Bay, Hood Canal, Netarts Bay, and to a lesser degree Humboldt Bay (Qualman 1983; P. Baker pers. comm.). The details of seeding and transport of oysters between the Pacific bays are mired in the lost records of the oystermen. The primary areas in Coos Bay used for the production ofJapanese oysters are South Slough and Haynes Inlet. The muddy sand flats ofEastside (Figure 1) have also been used as oyster clearing grounds in the past (Qualman 1983; Douthit 1986). The final group of transport mechanisms (mechanism IV) includes ballast water, recent mariculture plantings, and modern intracoastal and intra-bay small boat traffic. The transport of ballast water and associated plankton has been elucidated in general (Carlton 1985; Williams et al. 1988), and for Coos Bay specifically (Carlton and Geller 1993). 8This water may be discharged as the vessel navigates the channel from the entrance to the primary docking areas in the upper bay (Figure 1, sites 7 and 12) but the bulk of the ballast (mean = 1.09 x 1Q71iterse ship-l, S.d. = 2.7 x 106; Carlton and Geller 1993) is released at the docking facility and thus the upper bay is most likely to receive introductions. The current mariculture plantings have occurred in the same regions as those used for Crassostrea gigas culture (mechanism III) and include the re-inoculation and subsequent re-establishment of the native Olympia oyster, Ostrea conchophila (lurida). The Olympic oyster is native to the Pacific coast but it is herein treated as an introduced species into Coos Bay for the following reasons: 1) 0. conchophila (lurida) went locally extinct from the Coos Bay estuary in prehistoric times due to a temporary change in siltation patterns (Stubbs 1973; P. Baker pers. comm.); 2) in the intervening period 0. conchophila (lurida) did not naturally re-establish a viable population (p. Baker pers. comm.); and 3) it was re-inoculated in 1992 by local mariculturists. Having become re-established in Coos Bay, Ostrea conchophila (lurida) is now a member of an encrusting community newly composed of a large number of Atlantic and Japanese encrusting species that were absent when the oyster became extinct. For the resident community of invaders, O. conchophila (lurida) thus becomes an invader itself. As opposed to the primary (1°) invasion mechanisms described above, the secondary (2°) introduction of species via intracoastal and intra-bay small boat traffic (commercial and private fishing vessels, sailing vessels, replica antique vessels) continues to redistribute species introduced originally to primary receiving areas (e.g., San Francisco Bay and Willapa Bay). These vessels have the propensity to develop lush hull fouling due to lack of upkeep (Crisp 1958; Crisp and Southward 1959, Skerman 1960; Carlton and Scanlon 1985; Carlton and Hodder 1993). The primary mode of transport within the Coos Bay region was historically by water. Until the 1940's a variety of within bay vessels operated as far up the Coos River as Alleghany and down into the South Slough, acting as 9passenger and cargo carriers. These small vessels may have moved organisms introduced by coastwise or foreign shipping (mechanism I) from the larger shipping docks to areas throughout the bay. Carlton and Hodder (1993) studied species transport between Pacific Coast bays by the Golden Hinde II, a replica of Sir Francis Drake's vessel Golden Hinde. They demonstrated that a variety of species were collected and transported between estuaries and bays on the hull of this replica vessel, despite the application of copper based antifouling paints. Species were transported between each adjacent pair of bays studied (Yaquina Bay -7 Coos Bay -7 Humboldt Bay -7 San Francisco Bay) as well as in an additive fashion between all bays. Materials and Methods Coos Bay Transect Physical Measurements Temperature and salinity (conductivity) were characterized at four sites in Coos Bay (sites 2, 9, 12 and 15 ofFigure 1; see also Table 2) during spring low tide series of each month from September 1988 to September 1990. A fifth site in South Slough was sampled by researchers conducting studies in the National Estuarine Research Reserve (site 18 ofFigure 1 and Table 2) over the same time period and the data are presented for comparison. Salinity and temperature were collected with a YSI temperature-salinity meter. Measurements were taken at approximately -2.0' Mean Lower Low-Water (MLLW). An additional bay wide (0 to 32 km) transect was conducted during flood tide in July 1990, in which temperature and salinity measurements were collected at one mile intervals (Figure 2). Table 2. Sample Sites within Coos Bay. Transect names are CB, Coos Bay and SS, South Slough. Region of the bay follows codes in Table 1. Site ID# Transect Region Site Name River Mile Km' Substrate Physical Measurement Site I CB LB Coos Head 1.25 2 Rock 2 CB LB Ocean Station 1.5 2.4 Shell, Rock 0' 3 CB LB North Jetty 2.25 3.6 Rock 4 CB LB Fossil Point Jetty 2.5 4 Rock 5 CB ME Sitka Dock 4 6.4 Panels, Pilings 6 CB ME Coos Bay Dredge 6 9.6 Shell 7 CB ME Weyerhauser Dock 8 12.8 Panels, Piling, Rock 8 CB ME Pony Slough 9 14.4 Floats, Pilings 9 CB UB Haynes Inlet 10 16 Pilings, Rock 0' 10 CB UB Larson Slough 12 19.2 Rock II CB UB City Dock 14 22.4 Floats, Pilings 12 CB UB Isthmus Slough Float IS 24 Floats, Floating Logs 0' 12 CB UB Isthmus Slough Dredge 15 24 Shell, Rock, Bark 12 CB UB Isthmus Slough Panels 15 24 Panels 13 CB CR Coos River 20 32 Floats 14 SS SS Point Adams Jetty 2 3.2 Rock 15 SS SS Charleston Boat Basin 2.75 4.4 Floats, Pilings 0' 16 SS SS Hallmark Fisheries Dock 3 4.8 Floats, Pilings 17 SS SS Port of Coos Bay Dock 4 6.4 Floats, Pilings 18 SS SS South Sloug!L 5 8 . Oyster Shells, Logs 0' o II Biological Communities The distributions of encrusting (i.e. sessile) organisms in the Coos Bay estuary were surveyed along two estuarine transects: Coos Bay (CB) and South Slough (SS). Fourteen. sites were selected (nine in the CB transect and five in SS) based in part on the consistent presence of hard substrate, accessibility and location relative to other sites (Table 2). Long term settling or "fouling" panels of sanded black acrylic plexiglass (200cm2) were placed at three sites, Sitka Dock, Weyerhaeuser Dock, and Isthmus Slough to supplement the availability of retrievable hard substrate. At each subsequent sampling (roughly six month intervals between September 1988 and 1990) a species list was compiled for the more obvious fauna (macro-fauna) during a qualitative site survey. Approximately 400 cm2 of the encrusting community from each site was collected, placed in seawater and transported to the Oregon Institute of Marine Biology (DIME), where samples were examined under dissecting microscopes to aid in species detection and identification of the less obvious species. Species were scored for presence or absence at each site. Site species lists were then compared between sample times and collapsed so that a single list was obtained for each site where species were scored for presence at any sample time. This results in liberal (maximum) distributions of species within the bay without regard to temporal fluctuations. Surveyed substrates included rock (basalt jetty materials and conglomerate material surrounding tide gates), cobble, live and dead shell material, wood and bark, pilings and floating docks (see Table 2). All collections were made between 0.0' and -2.5' MLLW. Three additional dredge sites (-20.0' to -30.0' MLLW) were sampled in November 1988 and October 1989 in the lower (site 2: 2.4 km), middle (site 6: 9.6 km), and upper (site 12: 24 km) portions of the bay. 12 Species presence and absence were compared between sites by calculating a S0rensen's similarity coefficient (1948) and constructing a similarity dendrogram. The S0rensen's similarity coefficient compares the number of species shared between two samples versus the total number of species in those samples. All between-site similarities were calculated and a similarity dendrogram was derived based on a hierarchical clustering technique, the unweighted pair grouping method (UPGMA; Wilkinson 1990). Recruitment Phenologies and Abundances Two sites, one in the upper bay (Isthmus Slough, site 12, Figure 1) and one in the lower bay (Charleston Boat Basin, site 15, Figure 1), were selected to study the recruitment dynamics of native and introduced species. At each site four sanded, black acrylic recruitment panels 50 cm2 (7.2cm X 7.2cm) were exposed horizontally at -1.0' MLLW for 30 days, at which time the panels were collected and replaced by a clean set. The collected panels were placed in seawater, transported to OIMB and the bottom surface examined with a dissecting microscope. All organisms were identified to species or if species identification was not possible an illustration was made for future reference and the settlement panel was placed in running seawater and examined until identification was possible. In this fashion a catalogue of identification illustrations was developed for use with early recruits. After recruitment panels were examined they were cleaned and sanded for reuse. This experiment continued at the two sites from August 1988 to December 1990. Due to differences in the number of days between collections, the species recruitment data were standardized over a 30-day month and averaged over the four replicate panels (recruitsepanel-1emo-1). The data include the phenology or timing of recruitment (number of speciesemo-1), number of months in which recruits were observed for each individual ij , 1 13 species (mo· species-I), and the rates oflarval recruitment on bare substrate (recruits·panel-l·mo-l). Monthly samples were averaged over years (January to July = 2 years, August to December = 3 years) in order to calculate the duration and abundances of recruitment. As the numbers of native and introduced species are greatly different at each site, the values (number ofspecies·mo-l) were converted to percentages within group (native or introduced). These data (percent ofgroup·mo- l) were arcsine transformed and compared within site by t-test. Length of recruitment period for each species (mo· species-I) was compared between groups at each site by t-test. Seasonal variation patterns in native and introduced species recruitment abundances (recruits·panel-l·mo-l) were found to violate assumptions of homogeneity of variances (Fmax test) after log (n+ I) transformation and thus were analyzed by a Kruskal-Wallis non-parametric two-way ANOVA (Sokal and Rohlf 1981; Zar 1984). Results Coos Bay Transect Physical Measurements Temperature and salinity measurements are presented in Figures 3 through 7 as both line graphs and "hydroclimagraphs" (sensu Hedgpeth 1957) of monthly means for each sampled site. Temperature and salinity of the lower bay station (site 2, Figure 3) vary little throughout the year, ranging between 7° and 13°C and 28 and 33%0 (parts per thousand) (see Table 3). In contrast, the Charleston Boat Basin (site 15, Figure 4) experiences seasonal reductions in salinity (17%0 in January). The Charleston Boat Basin sample station is located on the OIMB dock adjacent to a stream outlet feeding an anadromous Table 3. Physical Measurements for Coos Bay Stations. Site ID ~ - Name Salinity (ppt) Temperature (DC) Mean SD Min Max Mean SD Min Max 2 Ocean Station 31.3 1.5 29 33 10.6 1.4 8.4 13 15 Charleston Boat Basin 25.5 5.7 17 32.5 U.5 1.5 9.3 13.5 18 South Slough 25.1 8.1 12 33 13.1 4.0 7 18.5 9 Haynes Inlet 17.1 8.7 2.9 30 14.1 2.7 10 18 12 Isthmus Slough 17.6 8.0 4.5 29 14.9 5.3 4.5 22 .4 15 fish ladder. The December through March reductions in salinity reflect a slight delay from the peak precipitation period for the Pacific Northwest (rain year begins October). South Slough (site 18, Figure 5) experiences a wide range of salinities (12 to 33%0). Similarly, South Slough experiences summer warming of the shallows as demonstrated by the mid-summer elevated temperatures (18°C) returning to oceanic temperatures in winter (7 0 to lOOC). The upper bay sites, Haynes Inlet (site 9, Figure 6) and Isthmus Slough (site 12, Figure 7), experience salinity variations from 3%0 in mid-winter to 29-30%0 in summer. This wide variation is due in large part to the riverine input of the Coos and Millicoma Rivers whose outflows increase during the rain year. Haynes Inlet experiences a temperature range (100 to 18°C) approximately equal to that of South Slough. Isthmus Slough however reaches elevated summer temperatures above 20°C and winter lows of 80%. By September 1990, S. unicornis colonies had bleached (turned white) and were presumably dead, while the colonies ofB. schlosseri had regressed and were not abundant (together occupying <5% on the transplanted dock). The colonies of Botrylloides violaceus however had grown considerably. In addition newly recruited colonies on adjacent docks and floating logs increased the percent space covered by B. violaceus to 60% on the transplanted dock and about 20% in the adjacent encrusting community. 21 Over the course of the ensuing years Botrylloides has overwintered and successfully insinuated itself into the Isthmus Slough introduced-species community. Densities in subsequent years have dropped to 5 - 10% of the total space. I have followed the spread ofBotrylloides down the bay over the years 1991, 1992 and 1993 by its appearance at each of the Coos Bay transect sites. The rate of spread is approximately 2 kmeyr-1 (Table 6) reaching the Pony Slough (site 8) after three years. Measurements of water flow at Isthmus Slough at peak ebb tide convert to approximately 3.6 kmeyr-1, and Japanese studies of the large (> Imm) lecithotrophic tadpole larvae ofBotrylloides violaceus have shown that settlement can occur up to 10 hr after release (Saito et ai. 1981; Boyd et ai. 1990). Berrill (1949) has shown that in congeners settlement occurs between 5 min and 12 hr after release (mean of 1 hr in the laboratory). Thus the rate of spread is well within the capability of a single cohort. As previously discussed, the native Olympia oyster, Ostrea conchophila (lurida), was re-established in Isthmus Slough by mariculture in the summer of 1992. As with Botrylloides, 0. conchophila (lurida) has spread down the bay and is presently found in Pony Slough (site 8). The Isthmus Slough community has also been the receiving area for a second ascidian, Diplosoma mitsakurii. The native D. macdonaldi has recently been synonymized with D. listerianum (G. Lambert pers. comm.), and the common Japanese D. mitsakurii may be indistinguishable from D. listerianum, although Nishikawa (1990) retained the species name. The combined factors of the absence of a Diplosoma from any site in the bay between 1988 - 1991, the sudden appearance of common colonies (> 15% cover) in the summer of 1992, and the presence of ascidian tadpole larvae in ballast water (Carlton and Geller 1993) lead me to view this species as D. mitsakurii, a Japanese ballast water introduction. Table 6. BOtlylloides violaceus Dispersal Following Secondary Introduction. Mechanism codes follow Table 1. Distance(km) Furthest Spread Date Spread Site ID# Site Name Dollor Region Mechanism unknown - 18 South Slough 10 invasion Japan III June, 1990 20km 12 Isthmus Slough 20 invasion South Slough IV June, 1991 1.6km 11 City Dock expansion Isthmus Slough June, 1992 3.2km - Coast Guard* " Isthmus Slough May, 1993 4.8km 8 Pony Slough " Isthmus Slough * - Coast Guard Cutter Citrus Dock at 19.2km N N 23 This may have been a transient invasion since it disappeared from permanent panels in Isthmus Slough, and has not been collected since September 1992 (subsequent collections in November 1992, February and April 1993). A second potential Japanese ballast water invader is the bryozoan Alcyonidium sp. collected once from Isthmus Slough. Carlton and Geller (1993) reported Alcyonidium larvae in Japanese ballast water. They were able to isolate and culture these cyphonautes larvae in the lab and induce settlement (1. Carlton pers. comm.). It is impossible to distinguish a transient (failed) invasion event of Japanese Alcyonidium sp. from the seasonal colonization of the upper bay by native Alcyonidium polyoum? due to the morphological similarity ofAlcyonidium species. In the Charleston Boat Basin (site 15), the bryozoan Watersipora edmondsonii? (identification by W. Banta, pers. comm.) is a new introduction to this community. Between 1988 and the appearance of W. edmondsonii? in 1990, extensive surveys of the Charleston fouling community had not detected its presence, nor was it found in surrounding habitats (point Adams Jetty, site 14). Watersipora, a deep maroon to burgundy red bryozoan, reached moderate abundance in 1990 (5% space) and in the following years was found at nearby sites (sites 14 and 17) until September 1992. Since then the abundance of Watersipora has declined «1% space) and it is now absent from Point Adams Jetty (site 14). The intracoastal traffic between California ports and Coos Bay is the most probable mechanism responsible for this introduction. In spring 1992 a Watersipora sp. that is morphologically very similar to the Coos Bay species appeared in the San Francisco Bay encrusting communities (1. Carlton pers. comm.). The native species richness (a-diversity or point diversity) declines in highly significant exponential decay curves along the South Slough (r2 = 0.54, p. ..... :~ 15 ('0 (J) 10 5 0 20 J F M A M J J A SON 0 Time (Months) -..() o '-' ~ 15 ::J ..... ~ Q) 0.10 E Q) .- 12 u6~=uu,-s....~ 5 30 Salinity (ppt) o ~1--------+--------1'--------.,. 25 35 42 Figure 4. Physical Measurements from the Charleston Boat Basin (Site 15): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature 0C); b) Hydroclimagraph of Site 15, January (1) and December (12) are Labelled. 43 353025 ·_·_·_-_·_-1 I Q) L... ::J ..... ..................................... \~II ~ /Q) lE- G) I- 15 20 Salinity (ppt) 105 ..L..-j!....-..--II---+-I-+-1-".._ +.,-_'+1-"'-- ~'+',-"··_·····-+1-·'-"·--+I-····-~·---t-I=~=j F M A M J J A SON D Time (Months) ...........................- -- _.-._-. 35 30 25 ---- ..... Q. 520 >-~ .s 15 ro if) 10 5 0 J o 5f o L.--.,If-----+--t._-+-_+___- 1--+-----+-1----+-~-t-----+---+----_+--....J A) B) 44 Figure 5. Physical Measurements from the South Slough (Site 18): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature DC); b) Hydroclimagraph of Site 18, January (1) and December (12) are Labelled. 45 A) 35 30 25 -. ..... 0. ~20 ~ :S 15 co Cf) 10 5 0 J ~ :J .... ..., ~ - Q) ..... c.:~ 15 E ro Q) CI) r- 10 5 0 J F M A M J J A S a N 0 Time (Months) B) 25 ..,-----------------------_ ......... o o --~ 15 JlO l' : 1-····--+--+--+---+---+-----+----+-----+---+--+-----+---J-.--+I-J o 5 10 15 20 Salinity (ppt) 25 35 48 Figure 7. Physical Measurements from the Isthmus Slough (Site 12): a) Average Temperature and Salinity Data by Month (Solid Line Represents Salinity and the Line with Closed Circles Represents Temperature °C); b) Hydroclimagraph of Site 12, January (1) and December (12) are Labelled. 25 -r------------------------, B) 49 ~ :::l ~ - :t::: .!: 15 ro (f) 10 5 0 J 20 ......... o o '-" :;:; (\J z 20 58 Figure 16. Coos Bay Encrusting Community p-Diversity. Sorensen's Between-Site Similarity as a Function of the Difference Between Site Specific Annual Salinity Range_ 25 • ____9 ---------. _ • • • beta diversity =0.10 5 10 15 20 Change in Env. Gradient (Sal Range) • --- 20 ,--------------------------, x Q) • "0 C C/) -c 15 o C/) c Q) L- -s if) 10 +--- ......... -" ·C ro E 5 if) c ro Q) ~ O+----+---+----+~____,f___-_+_-__+__-__+_-____+_-_+-_____1 o Figure 17. Introduced Percentage of the Community as a Function of Distance From the Ocean (pooled r2 = 0.95, n = 20). (Introduced Spp %) = 3.32(km from Ocean) + 1.19. 10 15 20 25 Distance From Ocean (km) 59 3530 .. South SloughA COOS Bay 5 100 en Q) '0 80Q) c.. (f) "C 60Q) () ::J "C 0 40L........ C ;:R0 20 0 0 1 I 100 A----------------------------, 60 10080 • South Slough~ Coos Bay 20 40 60 Native Species Richness ....\. . • o -!-----+---+----+----+--~~-""l..t_-"---'AIL--""-"_1"-""---""-="-:::=""=t="=""="=''''~""~-+-----j o Figure 18. Introduced Percentage of the Community as a Function of Native Species Richness (pooled r2 = 0.70, n = 20). In(Introduced Spp %) = -0.04(Native Spp Richness) + 4.03. ..... 5i 25~ Q) a. (/) Q) "0 Q) C- OO "0 Q) U .g 50 e ..... c Figure 19. Similarity Dendrogram Based on S0rensen's Similarity Measure Comparing Species Presence!Absence Data for all Sites in Coos Bay. Site Codes Follow Table 1. Dendrogram was Constructed Using UPGMA Clustering Technique. I 10 Site Name Coos Bay Dredge S Sitka Dock 4 Fossil Point Jetty 14 Point Adams Jetty 3 North Jetty Coos Head 2 Ocean Station 16 Hallmark Fisheries 17 Port of Coos Bay 15 Charleston 18 South Slough 7 Weyerhaeuser 9 Haynes Inlet 8 Pony Slough 12 IS Float 12 IS Fouling 12 IS Dredge 11 City Dock 10 Larson Slough 13 Coos River Percent Similarity 100 90 80 70 60 50 40 30 20 10 0 1 r f-- I- ~ I- ~- ~ 61 62 Figure 20. Monthly Recruitment Phenologies of Native and Introduced Encrusting Organisms in Lower Coos Bay and Isthmus Slough (Upper Bay). Species Recruitment During the Study is Represented by Filled Boxes. Native Soecles Balanus glandula Me1l1d1um senile Obela spp.? ScypHstomae (Aurela sp.?) f;':W'·"" A1cyonid1um polyoum BugUa pacifica Calopora horrida CaUibugula cllala Chellopora proelonga Conopeum reticulum Cr1brt6na annulata Crisis occidentalis Dendrobeania Ichenoides Electra crustulenta Hlppo1hoa hya6na Microporena caHfomica MicroporeHa ciliata Porella cotumbiana Smlttoldea pro6fica Trtcelaria erecta Mytilus trossUus Pododesmus cepio Eudstyla spp. Serpulds Splrorbids Terebemd sp. M Terebe.d sp. S Halclona sp. Leucosolenia sp. Astidia ceratoldes Che!yosoma productum Cnernidocarps finmarkiensis Dislap6a occidentals Pyura haustor Styela gibbsi Styela montereyensis Isthmus Slough (Upper Bay) I J I F I M·r=A I M I J I J I A I S I 0 I N I D I CT-- J8ll!!II3!C- J I I I I I I c::::::::J--T -'] I I I I I I I I I I I I I 0\ w Introduced and Cryptogenic Species Balanus improvisus TlbUarta croce. Obela spp.? AIcyonIdium sp. Conopeum tenuisslmum Sctlzoporefla unicomis Barentsi. benedeni Ostrea Iurida H.6chondrta bowerbanki H.6clona sp. Botry1lus 5chlosseri Botrylloides violaceus Diplosoma mitsakurii Molguta mamattensi5 Bowerbankia graci~s? / 0..l...-i----1----l--1---+--+--+---+--+--t--t----j.--.J 64 oNosM A M J J A Time (month) FJ ~_::::::'3~~~----I.--~-------------------.------- ---------.-------------- . 30 -.---=c-----------------------, C> c: :e 252 u a> 0::: 20 (/) a> "g 15 a. en '0 10 L- a> .0 E 5 ::J Z Figure 21. Average Number of Species Settling per Month in the Lower Bay Between August 1988 and December 1990. Solid Line Represents all Species, Triangles Represent Introduced Species and Circles Represent Native Species. 12 -r---------------------~ Figure 22. Average Number of Species Settling per Month in the Upper Bay Between August 1988 and December 1990. Solid Line Represents all Species, Triangles Represent Introduced Species and Circles Represent Native Species. 65 M A M J J A Time (month) FJ C) c: ~ 10 l-() Q) 0::: 8 CJ) Q) .~ 6 Q. Cf) '0 4 l- Q) .c E 2 ::J Z 6~----------- 66 • Introduced• Native 2 3 4 5 6 7 8 9 10 11 12 Length of Recruitment Period 5 (,I) '~4 en '03 ....(I) .0 §2 Z 1 0 1 Figure 23. Frequency of the Length of the Recruitment Period for Native and Introduced Species. Figure 24. Kite Diagram of Average Recruitment Rates (Recruits·Panel-1·Mo-l) for Native Species in the Lower Bay. All Species are Represented at the Same Scale, Asterisks Represent Recruitment at Low Densities. Alcyonidium Distaplia Balanus Cheilopora Spirorbids Hippothoa Bugula ------<0>-- • • 100 -,-+--+--+---J--II--I--I--I--+--+-+--t---J JFMAMJJASOND Month 67 Figure 25. Kite Diagram of Average Recruitment Rates (Recruits·Panel-I.Mo-I) for Introduced Species in the Lower Bay. All Species are Represented at the Same Scale, Asterisks Represent Recruitment at Low Densities. Cryptosula Bowerbankia Conopeum Botryllus Schizoporella Botrylloides :> :: ----====0=:::=--- ----==~====-- 100 -'-+--+--+------+----;----;I--I--l----+--+--+--+__' JFMAMJJASOND Month 68 69 2-.---------------------------, 1210 11986 7 Month 5432 A.O~r=e~~::::t::::-~_._-+---+_+_*_~ii;=+~~~~ 1 Figure 26. Schizoporella unicornis Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years. ~ m ~ Q) c 1.5 m 0- -"CC ~ 1 m ......... +J C Q) E~ 0.5 2 u Q) 0:: Figure 27. Botrylloides violaceus Recruitment (Average Number of Recruits·Panel-1·Mo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years. ~ 2 -I t1.5t 111 c a.> E:t:: 0.5 ._._._.....__....._-_._._..... ::J L..() a.> 0:: 70 .A. A o Ar--;-A--;-.--+--4Ir;;-=-t- --t-.tc---t--\---+-----A-t---A-~ 1 2 3 4 5 6 7 8 9 10 11 12 Month 71 0.8 -,----------------------------, 1210 11986 7 Month 5432 O-=~__'!'"'_:II~__..~__a'__t_~_+__+__+__A______+_____'~+______,o"__~~~ 1 Figure 28. Botryllus schlosseri Recruitment (Average Number of Recruits·Panel-1·Mo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. ~ co 32(j) c: 0.6 co ~ "C .5 g>OA co --..... c: Q) E :t::: 0.2 2() Q) 0:: 72 12109 .A . :.il. .A 6 7 8 Month 5432 o ~~I;=~~-"JlIl~=:--+---.--+---+--+-.-+-----,k-~~---A~--+~ 1 0.1 Figure 29. Conopeum tenuissimum Recruitment (Average Number of RecruitsePaneI-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. ...-... >. co "'C ~ 0.08 c co ..e- -g 0.06 _- -.. _-.- -._._ . C> >co -::: 0.04 . c· Q) E ..... ·20.02 (,) Q) 0:: 73 2...-------------------------~ 1210 11986 7 Month 5432 O..""I"""96-+---4!1l-------1------.41Io-+~~___*-+-_+__+__6__~~+____,k__t=41Il~~ 1 Figure 30. Bowerbankia gracilis Recruitment (Average Number of Recruitsepanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years. -. >-co 32 Q) c: 1.5 co .e- "C c: ~ 1 co '-'" ..... c: Q) E~ 0.5 ::J I- U Q) 0:: 74 0.4-.----------------------1 1210 119678 Month 5432 o ~~~t----;4t_+____.-+-__._~~--f-+_~~_____.--+-~q.-_._~ 1 Figure 31. Cryptosula paUasiana Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. ~ co :E Q) c: CO c. -"'C .!:: g>0.2 CO -- -c:Q) E - ·2 () Q) 0::: 75 0.4 ,--------------------~ 1210 1196 7 8 Month 5432 A.O~+---lII1~~=-T-_Ir_~~~..___+___t_-+-._+______ll~__.-+-__k___t_____=lIl. 1 Figure 32. Alcyonidium polyoum Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. ~ ro 32 "'ij) c: ro c.. -"Cc: g>0.2 ro "'-'" -c:Q) E -'2 u Q) 0::: 76 5.----------------------------, 1210 11 .... 2 986 7 Month 5432 Figure 33. Distaplia occidentalis Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. Numbers Adjacent to Symbols Represent the Number ofReplicates at that Position. 77 1210 1198 ......................._- -..- - . ... 6 7 Month 5432 O.-.-+-*--+--....-+-......,...........,.--I!E--+--+--+--.---+-~--+-~-+-.-+~ 1 3--.------------------------------, Figure 34. Balanus glandula Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. ~ ro 32 2.5 Q) c:: ro .e- 2 "C c:: g> 1.5 ro -- -~ 1 E -'5ts 0.5 Q) 0:: 78 12 ~------------- ----, 1210 11986 7 Month 543 .............- __..- .....................................................__.... 2 0.tr-t--...~~~~~t=+=$=+=1~~...........,.....4~-l 1 Figure 35. Bugula pacifica Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years. ~ co ~ 10 , .. - ··AIt.c.................................................. .................... j (]) c: co .e- 8 "0 .£: g> 6 co -- -~ 4 E :t::::§ 2 (]) 0:: , • 79 0.08 ,-------~----------------, 1210 11986 7 Month 5432 O,*-~~~~'-r--A-~~-7Itf:4--'-:~Ir--!--*",~~~ 1 Figure 36. Cheilopora praelonga Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Year. >: ro ~ Q) c 0.06 ro .e- "C c ~0.04 ro '-" ..... c Q) E :t:: 0.02 2(.) Q) 0::: 80 3 ....---------------------------, 1210 11986 7 Month 5432 O~_t__:lI~~~___4:_+_~_+_&=~~t___:.6.: 1 Figure 37. Spirorbid spp. Recruitment (Average Number of Recruits·Panel-1·Mo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Year. -~ ~ 2.5 C1> c co .e- 2 "C .5 ~1.5 co -- -a5 1 E - ·2 u 0.5 C1> 0:: 81 1210 11986 7 Month 5432 oL+----A----+-~~~::;t=~~*-H;~-A-~rl:~=t=J. 1 0.04 ~-----------------------------, Figure 38. Serpulid spp. Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means of Four Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average of Monthly Recruitment Rates Between Years. ~ ro ~ Q) c: ro c. -"Cc: g> 0.02 ·-·----··-A·-·-·-·····-······---·-·.··-·---·.·.-----· ro -- -c:Q) E ~ 2() Q) 0::: 82 2-,----------------------~ 1210 11 ------A.----- 96 7 8 Month 5432 -A;----- o~+=4~_4___+_____.ik-t___+_---+--.__+_______+~~+=l~_.--+-_4t___~ 1 Figure 39. Hippothoa hyalina Recruitment (Average Number of RecruitsePanel-leMo-l) by Month Between August 1988 and December 1990. Triangles Represent Means ofFour Replicate Samples During One Sample Period (Approximately 30 Days), Solid Line Represents the Average ofMonthly Recruitment Rates Between Years. >ro :e Q)ffi 1.5 c. -"C .5 ~ 1 ro ....... ..... c: Q) E~ 0.5 2 u Q) 0::: 83 CHAPTER II THE ASSEMBLY OF ENCRUSTING COMMUNITIES IN THE LOWER COOS BAY ESTUARY Introduction The patterns of community development have long interested ecologists (e.g., Clements 1916; MacArthur and Wilson 1963, 1967; Drury and Nisbet 1973; Pimm 1984, 1991). In the face of catastrophic disturbance, such that all species in a patch are driven locally extinct, the trajectory of community assembly will be determined by species colonization from adjacent (or regional) patches. The resulting variation in community structure may be examined from a variety of perspectives, including community richness (diversity), and composition. The conceptual model of island biogeography, developed by MacArthur and Wilson (1967), predicts that through the processes of immigration and extinction a dynamic equilibrium of species number will exist for a given community. Large perturbations may drive all organisms in a patch to local extinction, but the patch will proceed to aC.cumulate species according to patch size, distance from source pool of species (immigration), and as the community develops, the extinction rate of species present in the patch. There is no intrinsic assumption that the developing community is more or less closed to portions of the donor species pool (Roughgarden 1989). Instead the turnover model of island biogeography presents the community as open to all colonists, that is the community has unlimited membership. This non-interactive model has been modified to include the alteration of extinction rates due to competitive exclusion by resident species (Wilson 1969) while maintaining the 84 quality of unlimited membership (interactive model). Additionally, Wilson postulated that two further stages in the shifts of community equilibria might exist in ecological and evolutionary time. With increased time the species combinations will result in stable patterns (assortative equilibria) and as these species combinations persist through evolutionary time they will co-adapt (to environmental conditions) and co-evolve (to one another) resulting in reduced extinctions from competitive interactions. While Wilson proposed that both assortative and evolutionary equilibria would increase the species ,.. equilibrium (S), it has been suggested (Goodman 1975; Osman and Whitlach 1978) that it 1\ is equally likely for the species equilibrium (S) to decrease. Many authors have tested the species equilibrium in a variety of insular habitats (Simberloff 1974). These include oceanic islands (MacArthur and Wilson 1963, 1967), lacustrine islands (Kadmon and Pulliam 1993), defaunated mangrove islands (Simberloff 1969; Simberloff and Wilson 1969, 1970; Wilson and Simberloff 1969), caves (Culver 1970), artificial substrates in aquatic (patrick 1967, 1968, 1975) and marine (Osman 1977, 1978; Schoener 1974 a, b) systems, and laboratory microcosms (Dickerson and Robinson 1985, 1986). The marine encrusting community has proven to be a model system with which to test concepts of insular faunal buildup (Schoener 1974a, b; Osman 1977, 1978). In most studies the results have supported the predictions of the model, although a few exceptions have resulted in further development of the theory (Brown 1971; Brown and Kodric-Brown 1977). Robinson (1981) has discussed the inapplicability of assembly experiments (i.e., defaunated island, artificial substrate, island microcosms) to study the assortative and evolutionary equilibria of Wilson (1969), but has demonstrated the ability of these same studies to assess the robustness of the interactive and non-interactive aspects of the theory. Alternatively, the temporal patterns of species composition have been examined in an effort to predict community change. Given the same set of potential colonists, will all 85 communities converge on "stable" community states or is the sequence of species accumulation non-deterministic, resulting in a general lack of pattern? Three successional models were reviewed by Horn (1974) and Connell and Slatyer (1977). These included the "facilitation," "tolerance," and "inhibition" models of community development. Facilitation is the classical Clementsian view of community succession: a single climax community will be approached only after passing through earlier stages of community development (Clements 1916; Margalef 1968; Odum 1969). In each stage the resident species "prepare" the way for the next rank of colonists (Clements 1916; Horn 1974). In the second model of succession the early colonizers will "tolerate" the appearance of later arrivals (Horn 1974; Connell and Slatyer 1977). Early colonists are ephemeral species which are not resistant to the appearance oflonger lived species but are negatively affected by the presence of these late arrivals. As the early recruits die either due to natural mortality or through competitive interactions, the later arrivals dominate the community; the result is a community which is resistant to the early recruiting species. The third model predicts that all colonists arrive at random with no predictable sequence. These early arrivals however are hypothesized to actively inhibit the recruitment of new colonists by exploitative competition (Connell and Slatyer 1977). Of the three models of succession, the inhibition model has received the greatest level of support in marine communities (e.g., Sutherland 1974, 1977a, b, 1978; Woodin 1976; Connell and Slatyer 1977; Dean and Hurd 1980; Smedes and Hurd 1981). Thus the history of species accumulation is capable of determining subsequent additions or deletions of species in the community. Robinson and Dickerson (1987) and Drake (1990b) have empirically demonstrated that in aquatic laboratory microcosms the order (sequence) of species immigration can drastically alter the ultimate community composition. These alterations can be either deterministic or non-deterministic such that several alternate community states are possible from identical species trajectories 86 (Drake 1990b). Similarly, Sutherland (1974, 1978, 1981) and Sutherland and Karlson (1977) have demonstrated that in the encrusting communities at Beaufort, N.C., the resident species of specific assemblages are higWy resistant to subsequent larval recruitment. In these examples the arrival at one of several ultimate community states is greatly influenced by historical effects of species arrival. The discussion of community assembly and successional change in species composition has concerned systems in which all potential colonists are drawn from the same source pool of regional biota. It has been proposed that these systems have, through evolutionary time, developed coevolutionary constraints such that specific assembly pathways are closed (Ehrlich and Mooney 1983; Rummel and Roughgarden 1983, 1985; Drake 1990a, b). Thus the "ghosts" of past competitive interactions may have altered the assembly states away from unstable trajectories resulting in generally stable communities (Lewontin 1969; Jackson 1981; Sutherland 1981). The phenomenon of introduced species, that is the addition of new species to a regional pool by human-mediated mechanisms, has resulted in drastic alterations (additions and subtractions) in local ifnot regional species pools. The addition of new species to the species pool may be viewed as a permanent perturbation experiment (Sutherland 1981; Pimm 1984, 1987, 1991) in which an alteration in the trajectory of community development and ecosystem function may be observed (Vitousek et al. 1987; Vitousek 1990; Moulton and Pimm 1983,1986; see also Mooney and Drake ]986, Drake et al. 1989). Carlton (I979b, 1987, 1989) has reviewed the mechanisms and patterns of human-mediated species introduction to the marine communities of the Pacific coast of North America. The majority of successful introductions have been in bays and estuaries with relatively few invaders on the open coast. It has been suggested that the geologically young bays and estuaries of the Pacific coast (Atwater et al. 1977) aboriginally had a 87 depauperate native fauna (Carlton 1979b; Ricketts et al. 1985) and were as a consequence readily invaded by introduced species (Elton 1958; Carlton 1979b, 1989, 1992b). In contrast the absence of introduced species on the open coast has been proposed to be a function of the ecological resistance (Elton 1958) of the species rich native marine communities. The communities of the Coos Bay estuary have been and continue to be invaded by non-indigenous species (Carlton 1979b, 1987, 1989; Chapter 1). Within the Coos Bay estuary a changeover occurs from native species dominated, marine communities near the ocean to introduced species dominated, brackish water communities in the upper bay. In the lower bay however communities exist which have comparable native species richness, but only one of these has been successfully invaded by nine introduced species (Chapter 1). Elton (1958) proposed that species rich communities would resist invasions, but that if a successful invasion did occur the novel species would cause a drastic alteration of the community structure, and significantly alter the trajectory of community development (Pimm 1984, 1987, 1991; Vitousek et al. 1987; Vitousek 1990). In this chapter, I examine the development of these native-species rich, native and invaded communities in the lower bay of Coos Bay, Oregon. The following questions are asked: 1 ) Does the pattern of community assembly differ between the native and invaded sites? 2) Are there successional or directional patterns in species accumulation in communities at either site? 88 Study Sites Two study sites, the North Jetty and Point Adams Jetty (Figures 1 and 40), were selected in the lower portion of the Coos Bay estuary (43 0 19' 30"N, 1240 19' 30"W) based on the distributions of native and introduced species described previously (Chapter 1). These sites are separated by less than 2km, and are physically similar, with similar temperature and salinity regimes. Both are basalt rock jetties of varying ages: the North Jetty is approximately 100 yrs old (construction began ~1900) and the Point Adams Jetty is 20 yrs old (constructed in 1974). The North Jetty has a diverse native marine \ fauna of 83 species. Two additional species of unknown biogeographic origin (cryptogenic species) occupy less than 1% of the total space (Carlton 1979,1989; Chapman and Carlton 1991). The Point Adams Jetty has 66 native species (and the two cryptogenic species), a subset of the North Jetty fauna. This site has been invaded by 9 species of sessile invertebrates. The disjunct distribution of non-native species at these sites may be due in large part to dispersal limitation (i.e., a lack of intra- and trans-bay transport mechanisms). Velocity field diagrams of the lower bay (developed by COE 1979) for maximum flood (Figure 41) and maximum ebb (Figure 42) tides illustrate the difficulty a passively dispersing planktonic larva would have in crossing the channel between the North Jetty and Point Adams Jetty. Preliminary support for this was obtained from an exploratory experiment conducted in July 1989 where 10 oranges were dropped 50m from the Point Adams Jetty during a moderate ebb tide. Within 30 min all 10 oranges had passed less than 30m from the OIME boathouse (Figures 1 and 40) shore. Oranges are useful drogues. They are positively buoyant and float just under the water-air interface; they are passively dispersing; and they are highly visible. 89 Materials and Methods In order to assess the under rock encrusting community, settlement panels of black acrylic sanded to approximate a natural surface were placed amidst the low intertidal (-1.5' to -2.0' MLLW) jetty rocks of the two selected study sites. These panels consisted offour 50cm2 (7.2cm X 7.2cm X 0.6cm) subpanels (quadrants) arranged in a 2x2 array such that each quadrant could be individually removed, but as a unit they represent a single 200cm2 settlement surface. It has been shown that surface anomalies (e.g., barnacles, Leggo™ bumps, bolt heads) alter the settlement patterns of many invertebrates (Barkai and Branch 1988; Dean 1981; Walters and Wethey 1991). In order to create modular panels without attachments breaking the planar settlement surface the following design was developed. Each subpanel has a permanently attached 3cm stainless steel machine screw cemented (Poly-PoxyTM underwater cement) into a centered 0.3cm countersunk hole in the back. Four subpanels are then fitted and attached to a back panel (with appropriate holes) using wing nuts for ready field removal. Two 10cm stainless steel flathead bolts are attached to the back panel and face away from the subpanels (Figure 43). The entire array (back panel with four subpaneis) is placed (with the settlement surface down) inside one space ofa concrete building block (15cm X 15 cm X 32 cm) which has been prepared by drilling two holes on either top surface of either side. The bolts are threaded through the drilled holes and attached with stainless steel wing nuts (for easy field removal). Plastic spacers maintain the panel at the mid-point of the space and approximately 1.5cm from the walls. Thus each concrete block has two panel arrays, one on either side (Figure 43). To reduce or prevent the action of mobile benthic fauna (e.g., crabs, seastars, and fish) the concrete block openings were covered by 0.7 cm VEXARTM plastic mesh. One side had permanently attached mesh cemented to the concrete block. 90 On the opposite side of the block the mesh was cemented on the bottom side, the remaining three sides had VELCROTM hook strips sewn to the mesh with the VELCRO™ loop strips cemented to the concrete blocks providing easy access. Experiment 1 In order to assess the effects of mesh on settlement patterns, a series of no mesh (neither side covered), partial mesh (mesh on the bottom half of both sides), and complete mesh (both sides) treatments was conducted over a sixteen month period between April 1989 and August 1990. Treatments (mesh cover) were randomized between and within concrete blocks such that microsite differences in block placement were not confounded with treatment. Each treatment was replicated three times at the two study sites. A single subpanel (one quadrant of each panel array is equal to 50cm2) was randomly selected as the focal panel and the back marked for later identification. This focal panel was the sample unit in each panel array for the duration of the experiment. At each sample period the concrete blocks were examined and panel arrays were carefully removed in the field. The presence or absence of benthic mobile fauna (cancrid crabs, fish and seastars), the degree of mesh obstruction due to algae or encrusting organisms, and the degree of siltation were noted. The mobile fauna and accumulated sediment were removed and the mesh coverings cleaned before the blocks were replaced. Thus any disturbance-mediated larval release would occur during the 24 to 48 hr period in which the settlement substrates were in the lab. These panel arrays were then transported to the Oregon Institute ofMarine Biology (OIMB), and maintained in running seawater. The focal panel was identified for each panel array by subpanel markings, examined with a dissecting microscope and returned to the field within 24 to 48 hr. Species lists were constructed for each panel array (replicate) from the focal panel. 91 Experiment 2 In a second study, conducted from September 1990 to February 1992, all concrete blocks were 100% mesh treatments (mobile fauna exclosures). All panels were replicated four times at the two sites and the frequency of sampling increased to 13 samples over 17 mo (mean = 38.5 days, s.d. = 14.3). As in experiment 1, a single subpanel (50cm2) was randomly selected for each panel array as the focal panel for the duration of the experiment and was marked for future identification. At each sample period during the spring low tides, the concrete blocks (with two panels in place) were examined (as above for the presence of mobile fauna, mesh obstruction, and siltation), collected from the field, brought to the OIMB docks and the panels carefully removed. This procedure lessened the disturbance effect of panel removal from the blocks. These panels were then transported to OIMB, and maintained in running seawater while high resolution videotapes were made of each panel for later analysis. The panel arrays were returned to the field during low tide within 24 to 48 hr. In order to avoid contamination of native site panel arrays with introduced species from the invaded site the following measures were taken: 1) site collections were staggered over four to six days in which the native site panels were collected during the low tide, videotaped, and redeployed during the following low tide (24 to 48 hr later); the following day the invaded site panels were collected at low tide, video-taped, and redeployed during the subsequent low tide (24 to 48 hr later); 2) the running seawater tables used to hold panels were drained and scrubbed between sample periods (28 to 70 days); and 3) the site panels were consistently held in two separate water tables over the course of the study. The timing of the sample regime described above allowed panel arrays and blocks to be repaired if necessary. During the 24 to 48 hr period, 92 subpanels whose holding screws were loose could be re-cemented and cured out of the water, while the organisms on the opposite side of the panel were still under water. While the concrete blocks were on the OIMB dock, repairs to the mesh and VELCRO fastener system were conducted. In addition the blocks were scraped clean during each tide cycle, and then repeatedly subjected to high-pressure freshwater, with the intent that the concrete blocks not contribute significantly to larval supply. High resolution videotapes (sVHS) were made with a copy-stand mounted sVHS Panasonic color CCD camera with a 50mm zoom macro lens. The entire panel array (200cm2) was placed in a container that allowed the movement of the panel along registered guides. At each sample period the identical physical placement of the panel was obtained. Images of each subpanel (50cm2) in a 200cm2 array were videotaped (four shots) and sixty-four overlapping macroshots (approximately 6cm2 each) were made for the entire 200cm2 panel. In this fashion the video-resolution was approximately Imm2 and the accuracy of species identification from the video was> 90% for most taxa. To further aid in subsequent species identification, a continuous audio recording was made on the videotape in which newly settled, obscured, or unusual colonies (or individuals) were identified. Similarly, arborescent bryozoans and hydroids were moved during the videotaping such that all primary space could be both easily viewed and accounted for. Only the area of basal attachment, rather than the canopy area, was recorded for arborescent species. Similarly, the settlement of a species or growth ofvine-like (runner) species on top of another was counted for both species. However, overgrowth by sheet-like species was counted only once for the apparent "winner" of the outcome, but through time the result of the overgrowth interaction was followed and scored. Thus more than 50cm2 (greater than 100%) could be counted on a given plate. "Winner" is here defined in sheet-like species' encounters as the species which overgrows more than 5mm onto the "losing" colony (Buss and Jackson 1979; Quinn 1982; Buss 1986). 93 High resolution video-images were digitized using the JAVA image analysis software (Jandel Scientific, Corte Madera, California) installed on a 33 Mhz 486DX computer. Individual (or colony) areas of each species were digitized. At each time period maps of the focal panel were made with individual or colony identification and location, thus individual growth and mortality could be assessed. Similarly the recruitment and immigration (lateral growth from adjacent subpanels) of individuals (or colonies) could be distinguished between time periods. Video analysis however presents some difficulty. The difficulty in distinguishing three serpulid species, Crucigera zygophora, Pseudochitinopoma occidentalis, and Serpula vermicularis, with the sampling method used, prevented species level identification and thus "serpulids" was the least discernible taxonomic unit. Similarly "spirorbids" may refer to a species group (Blake 1975). The difficulty of counting percent cover of runner or vine-like species (Jackson 1977) makes it more reliable to use estimates in 5% intervals (equal to 2.5cm2). Terebellids are not truly sessile organisms but were included in this study due to their consistent presence on panels to which they had recruited and the high densities attained at specific times. While capable of leaving the tube, terebellids are functionally sessile much of the time. As with serpulids and spirorbids, the group "terebellid" may have several species. In contrast, "introduced species" does not describe a taxonomic unit but describes the biogeographic origins of the species included in the group. As a group, these species represent an addition to the species pool (an increase in the regional y-diversity) which may alter the rates of species immigration into and extinction from local patches (in this case settlement substrates). Thus, it is desirable to examine their effect as a unit on native species immigration and extinction processes. II 94 Community Structure Colonization curves of species richness (S) were calculated from the summary statistics for each panel at each sample period based on the qualifications for species identification stated above. Additionally the following community indices were calculated from the areas (cm2) occupied by sessile organisms for each panel at each sample period. The Shannon-Weaver information index (H'; Shannon and Weaver 1949) was calculated for species contribution to live cover (total space occupied by living organisms), as H' = -L Piln(Pi), where Pi is the proportion of all occupied space occupied by the i-th species (area of the i-th species/sum of occupied area for all species). The Evenness Index (J') was calculated from the Shannon-Weaver diversity (H') as J'=H'/H'max, where H'max = In(S) (Pielou 1966). Two measures of dominance were calculated, McNaughton's index (MD) is defined as the ratio of the space occupied by the two dominant species (greatest and second greatest areas covered) to the total space occupied by all species (McNaughton 1967). A second index used by Osman (1977) is defined as the smallest number of species that, taken together, account for 75% of the occupied space. Community Dynamics Species immigration and extinction curves (number of species newly immigrating or going extinct per month) relative to the number of resident species were generated for panels at the two sites. The development of the native component of the community was examined as well at the invaded site, such that native species immigration and extinction 95 rates were compared against the total number of species (native and introduced). From "these curves estimates of S were generated for comparison with the colonization curves developed earlier. A primary tenet of the theory of island biogeography is that species numbers in any given patch are determined by a balance between the rates of species immigration and extinction (MacArthur and Wilson 1967). As Chesson (1978) has argued, deterministic concepts of stability are inappropriate for the assessment of this dynamic equilibrium. Instead he has proposed the concept of stochastic boundedness. A bounded system is one in which upper and lower limits (bounds) exist such that at some arbitrary probability they are not exceeded. As Keough and Butler (1983) have pointed out, it is not sufficient to know that a system has bounds, but it is necessary to evaluate the width of the bounds. They have proposed a statistical criterion for "narrow" boundedness to evaluate the degree of stochastic fluctuation. There exist upper and lower bounds for any data set such that the probability of exceeding those bounds is 50% cover) during much of the study (Figure 56). The three introduced species Schizoporella unicornis, Botrylloides violaceus, and Botryllus schlosseri are the most common non-indigenous members of the community. Here I treat the cryptogenic species (origin unknown sensu Carlton 1979b) as members of the introduced group; these species include Bowerbankia gracilis, and Cryptosula pallasiana. Introduced species contribute disproportionately to the community statistics of diversity (H') and dominance. The group (3 introduced and 2 cryptogenic species) comprises 20% of the total species seen during experiment 2 (5 of25 species). The actual y-diversity for Point Adams Jetty is 66 native encrusting species, 5 introduced and 3 cryptogenic species (Chapter 1), and thus the introduced group comprises less than 11 % of the total species pool. The information index may be partitioned such that the relative contribution (percent ofH') of one species (or group of species in this case) can be compared with the expected distribution based on the species' (or group's) percentage of total species present (percent of S; Smith et al. 1979). Introduced species' contribution to the information index is significantly greater than expected based on their relative percentage of community richness in any given assemblage (G[1] = 5.44, p<.05; Figure 57). As noted above, the Point Adams Jetty assemblages tend to be dominated by less than three species at anyone time (Figure 55). Through time the dominant species on anyone panel at time t is more likely to maintain dominance until t+ 1; hence the results from a 108 series of samples on the same panel are confounded with time. To avoid problems with temporal pseudo-replication (sensu Hurlbert 1984), the percent of dominant species (as defined by the 75% index) that are introduced was calculated for each time point and then averaged for the entire time series for a single panel. Thus a single datum was generated for each panel. The average was taken as the estimate of introduced species contribution to dominance and was equal to 69.2% (s.d.= 20.9, n = 4). The effect of introduced species on native species percent cover was a result of overgrowth interactions (pers. obs.) and exploitative competition. The dominant introduced species, Schizoporella unicornis and Botrylloides violaceus, were higWy resistant to larval recruitment (epibiosis), yet were able to recruit Onto a variety of species substrates (Figure 58), resulting in the subsequent overgrowth of the substrate species. This is supported by the temporal shift from moderately to strongly negative correlation coefficients between the abundance of introduced and native species (r2 = 0.58, n = 11, p<.OOI; Figure 59). Community Dynamics The immigration and extinction rates were linearly regressed against the number of species present for all panels at the two sites. For this analysis, only the closed (100% mesh cover) treatment in the first experiment (1989-1990) was used. Analysis of covariance (ANCDVA), using the number of species present as the covariate, showed highly significant differences in immigration rates between adjusted site means (p<. 001), and within-site slopes (p<. 001) and indicated that for both the first experiment (April 1989 to August 1990; Figure 60) and the second experiment (September 1990 to February 1992; Figure 61) a single line was not sufficient to explain all points (pooled regression p>.05). ANCDVA for extinction rates however, showed no significant I109 difference in extinction rates between adjusted site means (between site p>.05), or within-site slopes (p>.05) and indicated that a single regression was sufficient to describe the extinction curve for all points (pooled regression p<.OOI). As can be seen in Figure 61, immigration at the North Jetty during the second experiment is not signicantly different from zero (r2 = 4-10-4, n = 52, p>05). Thus one of the criteria of the equilibrium model has been violated. Both Simberloff (1969) and Schoener (] 974a, b) have argued that the estimates of immigration and extinction may be biased due to unobserved immigration and extinction events between sample periods. For most taxa in this system, immigration is primarily from larval recruitment and once metamorphosed, the taxa are permanently attached. Similarly, an extinction event would generally be identifiable by the calcareous skeletal remains of most taxa. The following analysis must therefore be tempered by knowledge that this violation may underestimate the value ofSfor North Jetty communities. I't From the theOIy of island biogeography, a predidion of the dynamic equilibrium (S) can be generated from the intersection of the immigration and extinction curves. This provides an estimate for comparison with the colonization curves presented earlier (Figures 44 and 45 for experiment 1; Figures 46 and 47 for experiment 2). The predicted species equilibrium at the native site (North Jetty) is 15.9 species during experiment 1, but 7.2 species during experiment 2. As noted for the North Jetty, the curves for extinction during experiment 1 and immigration during experiment 2 are not significantly different from zero and consequently the predictions of dynamic equilibrium must he viewed with some degree of skepticism. At the invaded site (Point Adams Jetty) the predicted values of S are different between the two experiments with 10.9 species predicted in experiment 1 (Figure 60) and 7.1 species predicted in experiment 2 (Figure 61). At the invaded site the immigration and extinction rates of the complete species pool were contrasted with those for the native :;pecies component of the pool in order to 110 h observe the change in the expected species equilibrium (S) both with and without introduced species. The immigration regression lines for the total species pool and the native component show slight changes in slope (Figures 62 and 63). The extinction lines however, do not appear to differ. For the first experiment the total species pool predicts an equilibrium of 10.9 species while the species pool for natives only is 10.7. Similarly, in the second experiment the total species pool and the native pool both predict 7.1 species. The boundedness of species richness was evaluated for experiment 1 (the entire duration) and experiment 2 (months 1 to 5 and 6 to 17). For each panel the mean S, coefficient of variation (CV), n (time periods) and calculated co (Keough and Butler 1983) are presented in Table 16. The results of the one-tailed t-test with n-l degrees of freedom are also presented (lIo: CVx :s;15.3% ; HI: CV > 15.3%). During experiment 1 neither the North Jetty and Point Adams Jetty closed treatment communities were found to be stochastically unbounded (Figures 44 and 45; Table 16). With more extensive samples available for experiment 2, the data were analyzed for the two time periods previously described (months 1 to 5, October 1990 to February 1991; and months 6 to 17, April 1991 to February 1992). During the initial time period, none of the panels at either site satisfied the criteria for boundedness (Table 16, unable to calculate co's due to high CV's). The North Jetty assemblages during the second time period were highly variable (HO rejected) with a single exception (co = 0.27). At the invaded site three panels were bounded (co's < 0.21) with one panel exhibiting high variability (co = 0.41). Through time the experiment 2 assemblages at the native site increase in Bray-Curtis similarity to one another, approaching a mean of 60% with a decreasing range ofvalues (Figure 64). In contrast, on average the Point Adams communities are initially more similar to one another (Figure 65) although they are much more variable. By the end of the experiment, they too have approached a mean similarity of 60% and the range has Table 16. Stochastic Boundedness of Species Richness for North Jetty and Point Adams Jetty Replicates During Experiments 1 and 2. Panel Replicates are Denoted by Letter, S is the Mean Species Richness, CV is the Coefficient of Variation, and N is the Number of Sample Periods. See Text for Description of Stochastic Boundedness. Source Panel S CV N 0) Pvalue Experiment 1 North Jetty A 12.0 48.87 6 0.27 ns B 11.5 48.80 6 0.27 ns C 10.0 43.82 6 0.26 ns Point Adams Jetty A 8.4 26.08 5 0.14 ns B 8.6 35.46 5 0.17 ns C 9.8 39.12 5 0.17 ns D 9.2 29.17 5 0.15 ns Experiment 2 North Jetty months 1 to 5 A 0.5 115.50 4 * B 0.0 4 * C 0.3 200.00 4 * D 1.8 71.90 4 * months 6 to 17 A 3.5 36.26 10 0.38 * B 4.8 42.58 10 0.43 * C 5.8 37.95 10 0.40 * D 3.8 24.18 10 0.27 ns Point Adams Jetty months 1 to 5 A 3.0 98.13 4 * B 3.8 102.99 4 * C 3.3 121.46 4 * D 4.5 82.15 4 * months 6 to 17 A 7.6 6.79 10 0.08 ns B 6.7 39.20 10 0.41 * C 5.7 16.64 10 0.19 ns D 7.0 17.82 10 0.20 ns * - p<.05 111 112 decreased. The two sites however have very different assemblages: the mean montWy similarity between sites (all panel comparisons between sites averaged), demonstrates that while within-site community structure is converging (60% similarity), the between-site similarity is less than 10% at any time (Figure 66). As species temporally appear in the community they either remain within the assemblage or go extinct. The cumulative species distributions for the North Jetty and Point Adams Jetty communities exhibit large initial differences (Figure 67). Within the first 55 days the Point Adams Jetty communities have accumulated on average 7 species, as opposed to 1 species at the North Jetty. After five months the slopes of average accumulated species for the two sites are very similar. The correlation between the realized diversity of a sample and the cumulative diversity the sample has experienced (the history of species in the assemblage) is a measure of the openness of community membership. The predicted correlation for assemblages with unlimited membership would be a slope of 1.0, that is all species will enter and remain in the community. The North Jetty site has a pooled slope of 0.63; that is rougWy two-thirds of the species that have entered the community are still present. In contrast the invaded Point Adams site has a pooled slope of0.42, and thus less than half of the species have successfully entered and remained in the community (Figure 68). Community Composition In order to assess the nature of compositional development (i.e., seasonal, successional, or random), the sequence of individual species appearance in experiment 2 is described below for each taxon. 113 Cirripedia Balanus glandula was the only barnacle encountered during the assembly of communities at the North Jetty and Point Adams Jetty. At the North Jetty, Balanus settled on one panel at low densities between initial deployment and February 1991 (141 days). In the following sample period (April 1991) a small settlement had occurred on three panels (mean = 9 individuals· panel-I, s.d. = 11.7, n = 4) constituting <2% cover on any panel. Beginning in June (day 264) and continuing until September, Balanus settlement (Figure 69) increased from an average 123.5 individuals·panel-I (s.d. = 77.8, n = 4) to 266.7 individuals. panel- I (s.d. = 306.9, n = 4). This increased Balanus contribution to the community from less than 1% in April to an average of 42.8% in August and 32.0% in September (Figure 70). Immediately following the peak recruitment events, there was significant mortality, evidenced by numerous basal plates and dead tests. Cohort survival showed a significant exponential decrease for Balanus following the peak recruitments (Figure 71). Although the cohort survival was low (P50 = 47.4 days), Balanus continued to hold an average of30% space until December 1991 (day 438), after which time mortality continues to reduce the percent cover (Figure 70). At the invaded site, Point Adams Jetty, Balanus glandula was conspicuous in its relative rarity. At no time did its contribution to living cover exceed 10.0% on any panel (maximum 8.2%) and the maximum mean was 3.0%. No significant recruitment events were observed at Point Adams Jetty. Cnidaria Three cnidarians settled on the panels during the course of community assembly, a hydrozoan, Obelia spp., scyphistomae presumably ofAurelia sp., and an anthozoan, 114 Metridium senile. Neither scyphistomae nor Metridium were seen on the panels at the North Jetty. Obelia was present at the North Jetty during all but one sample period (August 1991) but never exceeded 5.0% cover (Figure 72). At the invaded site Obelia colonized after 55 days (December 1990) and remained present at some point on all four replicate panels until September 1991 (340 days). Scyphistomae were present on one panel at Point Adams during two sample periods: June and August 1991. They covered less than 1.0% space at all times. Metridium senile can be relatively large and extremely mobile, in comparison to other members ofthis community. The presence of single individuals on three panels at Point Adams had no discernible temporal pattern. Ectoprocta Sixteen species of ectoprocts. (bryozoans) were identified during community assembly. Eleven species were seen at the North Jetty (Table 13) and 12 at Point Adams (Table 14). Ofthe 11 bryozoans in the North Jetty communities seven species, Bugula pacifica, Cribrilina annulata, Crisia occidentalis, Dendrobeania lichenoides, Oncousoecia ovoid.ea, Rhamphostomella costata, and Tricellaria erecta, were minor components, never attaining more than 2.0% space. Bugula and Dendrobeania recruited during the June sample period; the remainder settled late in the assembly sequence during September, October and November. Ofthe remaining species, the earliest recruits were Hippothoa hyalina and Microporella ciliata in February 1991 and April 1991 respectively. Hippothoa was ubiquitous, being found on all plates for the duration of the study. On three plates it controlled more than 20.0% of available space for short periods (57.1, 36.2, and 20.9%; Figure 73). By the June sample period (day 264), Microporella californica had settled on 115 one plate and throughout the study continued to increase in area (maximum of4%). Cheilopora praelonga was found on all plates at an average of 4.0% space occupancy, and reached a single plate maximum of22.7% in September 1991 (Figure 74). At Point Adams five species attained less than 2.0% space during the study; these were Bowerbankia gracilis? (cryptogenic), Cribrilina annulata, Crisia occidentalis, Conopeum tenuissimum (introduced), and Tricellaria erecta. Cribrilina and Conopeum were early recruiters, settling within the first two sample periods, whereas the other three species settled after August 1991. Hippothoa hyalina recruited during the initial 28 day period (as did Cribrilina annulata) reaching a maximum of 5.0% space on one panel (Figure 73). During the December 1990 sample period (day 55) Alcyonidium polyoum?, Cheilopora praelonga, Microporella californica, and Schizoporella unicornis (introduced) recruited. Alcyonidium, Cheilopora, and M califarnica settled on a maximum oftwo plates, but all three attained maximum plate densities of>20.0%. Alcyonidium occupied a maximum 27.1% (average of 6.91 %), and Cheilopora reached 20.4% (maximum average of5.1%; Figure 74). The introduced Japanese bryozoan, Schizoporella unicomis, settled on aU four Point Adams Jetty plates by day 55, and attained no less than 39.0% cover on all plates (plate maxima: 39.6, 82.2, 89.9, and 93.9%; Figure 75). Schizoporella did not go extinct from a plate once it recruited. Individual colonies had the ability to survive long periods of overgrowth (generally by compound ascidians) of up to 180 days (Figure 76). Oyptosula pallasiana, a cryptogenic species, settled on one panel in June 1991 and reached a maximum density of 5.6%. 116 Mollusca The native mussel, Mytilus trossulus, was the only mollusk observed during community assembly and was a minor component. A single individual settled on one plate at the North Jetty during the August sample period (day 307) at 0.6% cover. No mussels were observed on the plates at Point Adams Jetty. Annelida Four identifiable taxa of polychaetous annelids were found at the two sites during this study. Spirorbids were not found on the panels at the native site. Terebellids were patchily distributed between panels at the North Jetty between December 1990 and May 1991. During anyone sample period one panel had terebellids present at low «2.0%) densities. In November 1991 terebellids on one panel reached densities of 9.6% declining to 1.8% in the following month. Serpulids settled on one panel in June and August 1991 but never achieved densities higher than 1.0%. Eudistylia spp. (E. polymorpha and E. vancouveri) were found on one panel between October 1991 and February 1992, at densities up to 4.1% (Figure 77). Spirorbids at the invaded site were found at less than 1% during one sample (June 1991) on a single panel. Similarly terebellids were found on a single panel at densities less than 1% during December 1991 and January 1992. Serpulids were found to settle early (December 1990) on three of the four panels. This group achieved mean densities of2.7% (a panel maximum of7.7%; Figure 78) in February 1991 through growth of individuals settled in December (one individual per panel). Eudistylia spp. settled on three panels during the later half of the study (August and October 1991 and February 1992) but never achieved densities greater than 2.0%. 117 Porifera No sponges were observed at the native site during the assembly of communities. At the invaded site however, two sponges were present. Halichondrea panicea was present on a single panel and achieved densities of4.0% by November 1991. Leucosolenia spp. was observed during separate sample periods on two different panels and reached densities of 1.4% in February 1992. Tunicata Three native tunicates, Cnemidocarpa finmarkiensis, Distaplia occidentalis, and Styela gibbsi, and one introduced tunicate, Botrylloides violaceus, were observed during the assembly process. At the North Jetty a single Cnemidocarpa recruit was seen during the September 1991 sample, but was not found at the subsequent sample date (October 1991). Single individuals ofDistaplia occidentalis recruited at two different times (June 1991 and October 1991) to two panels. One colony reached densities of 3.2% after two sample periods. A single, newly settled botryllid recruit (single zooid) was found during the first sample period at the native site but did not survive to the next sample period. Invaded site tunicates included Distaplia occidentalis, Styela gibbsi, and Botrylloides violaceus. The solitary ascidian, Styela, settled on three panels but never achieved densities greater than 2.0%. Distaplia, a native colonial, settled on three panels during the September sample period and on the fourth panel in the following month and remained in the assemblages until the last sample date (February 1992; Figure 79). This species reached mean densities of 43.4% in November, and had panel maxima of27.0, 31.1, 50.5, 118 and 85.4%. Botrylloides first settled on three panels during the first sample period and on the fourth panel in June 1991and continued to have larval immigration throughout the study. Maximum mean densities of 52.4% were reached in April 1991 with panel maxima of 17.1, 93.8, and 98.6% in April and 39.8% in August (Figure 80). Discussion The non-interactive model of island biogeography (Wilson 1969) appears to apply to the native communities of the North Jetty. The experiment I assemblages had achieved a level of constancy ("stability") in species number after 16 months but had values of S lower than predicted. The species extinction events however were not correlated with the resident species richness (non-significant regression). Similarly the immigration rates in experiment 2 were not correlated with resident species richness resulting in wide variations (CV> 30%) in species richness which violated the tests of stochastic boundedness (Keough and Butler 1983). Schoener (1974a, b) demonstrated that in several sites around the world the encrusting communities slowly accumulated species and that no biogeographic pattern in S could be discerned. In her studies the communities often failed to achieve a stable species number, which she attributed to seasonal variations in larval supply (immigration rates). The invaded communities in experiments 1 and 2 achieve stable species numbers after the initial faunal buildup. These communities are also adequately described by the non- interactive model of island biogeography since the observed extinction rate was not exponentially correlated with resident species richness. The communities that developed at the invaded site however, became increasingly closed to species addition, that is the membership became limited as space became dominated by the introduced species. 119 Sutherland (1974, 1977a, b 1978) and Sutherland and Karlson (1977) found that certain species assemblages were persistent (stable), resisting larval recruitment (immigration) of all other species (Styela-dominated assemblages) or most other species (Schizoporella-dominated assemblages). In these systems, mortality (senescence) was the predominant means of vacating space and altering community structure. The resultant multiplicity of assemblage types were deemed "alternate stable states" (Sutherland 1974, 1981). ~ ~ Is S a good estimator for these communities? At the invaded site the estimated S for the first experiment for closed treatment assemblages was determined to be 10.9 species (Figure 60). The actual species richness attained by the assemblages varied between 12 and 7 species, although during the last three sample periods the closed treatment assemblages averaged at 10 species (Figure 45). For the second experiment an Sof 7.09 species was predicted from the immigration/extinction equilibrium (Figure 61). The mean species number across the four assemblages varied around 7 species (Figure 47). As has been demonstrated statistically, the apparent boundedness of these communities suggests that there is no trend away from the state in which they are presently found. This is not to say that the species composition will not change, but that the species richness (number) is sufficiently described as a dynamic equilibrium. A The differences between S in experiments 1 and 2 for both sites may be due to the seasonal changes in larval availability. Osman (1977, 1978) has shown that the larval availability varies seasonally and directly alters the colonization (immigration) curves for A encrusting species which alter S accordingly. In the Woods Hole, Massachussetts communities, he demonstrated that the season of exposure may alter not only the rate of species accumulation but also the recovery period to a dynamic equilibrium. The study conducted in Coos Bay did not attempt to assess the seasonal component of community assembly, and with two initial time points cannot demonstrate any effect. 120 Osman and Whitlach (1978) have provided a series of predictions for the effects ofan A increase in the regional species richness, R (y-diversity of Whittaker 1960, 1972) on Sand the extinction/immigration ratio elm (where e is the specific extinction rate and m is the basal immigration rate with °species present in the patch). They found that for R to /\ increase S may either increase, stay the same or decrease depending on the value of elm, 1\ but in all instances the value ofelm must change. For S to increase, the value ofelm must decrease either through a substantial increase in immigration or a reduction in extinction. 1\ If S stays the same, elm must change very little either by a decrease in immigration or a !' slight increase in extinction. For S to decrease a substantial increase in the extinction rate is necessary in a sufficient number of patches to alter the regional elm ratio. The regional diversity in the encrusting communities of the lower Coos Bay (within the Point Adams Jetty "region") has been increased by at least 9 species in ecological time. "Examination of the difference in S (derived from the immigration and extinction rates) between the total species pool and the native component of the species pool suggests that no change in equilibrial state (realized S) has occurred. Yet the elm ratio has decreased due to the high immigration rate of the new members of the community and no apparent reduction in extinction rates (Figures 62 and 63). Osman and Whitlach (1978) provide a caveat; there exist many circumstances in which a species addition may result in the replacement of species such that R does not increase. A species addition can be described as a community perturbation away from equilibrium (Ritte and Safriel1977; Sutherland 1981; Pimm 1984). In the face of dominant invaders this perturbation may be sufficient to force the community from one community state (or trajectory) to a separate "basin of attraction" (Lewonton 1969; Sutherland 1974, 1981). These community level alterations have been documented in several invaded systems (Pimm 1984,1987, 1991; Pimm and Hyman 1987). The Great Lakes of North America have been recently invaded by the zebra mussel, Dreissenia 121 polymorpha, which is altering planktonic and benthic community structure (Mills et al. 1993). Similarly the soft substrate communities of San Francisco Bay have recently been invaded and replaced by the asian clam, Potamocorbula amurensis (Nichols and Thompson 1985; Carlton et al. 1990; Nichols et al. 1990). In this system, the communities that develop in the presence of the introduced species have altered patterns of species accumulation on higWy disturbed patches (clean settlement panels). The native communities develop slowly with seasonal recruitment events that stochastically alter community composition (e.g., barnacle settlement and mortality). In contrast the introduced species dominate the communities at the invaded site and shift the community composition. The results for the native communities in this study do not indicate a successional pattern (directional change) in species composition but point to an open system in which species composition is largely controlled by immigration (Figure 68). These communities are similar to Schoener's (1974a, b) findings that throughout the assembly process species continue to accumulate with no apparent equilibrium. Comparisons of community composition at the endpoints of experiments 1 (August 1990) and 2 (February 1992) as well as the August 1991 sample of experiment 2, demonstrate the lack of a "climax" community at this site (Figure 81). The three taxa, Balanus, Cheilopora, and Hippothoa were consistently present but in varying densities and ratios. The lack of significant immigration at the North Jetty until Hippothoa recruits in February 1991, leaves more than 80% of the space available for settlement. Even after the high settlement ofBalanus (August to September 1990) the amount offree space remains at an average of 40%. Thus the competitive interference and exploitative effects of primary space occupancy are not seen during the 17 months of community development at this site. The Point Adams Jetty communities in contrast are quickly colonized by several species, including the introduced species, Schizoporella unicornis and Botrylloides 122 violaceus. Here, the unoccupied space is quickly reduced (less than 5% in three panels by April 1991). Ifone examines the fundamental species (Sutherland 1977a, b, 1978; Sutherland and Karlson 1977), species which occupy at least 10% of living space at any time during the assembly, the sequence of appearance is highly ordered. Botrylloides and Schizoporella recruit in the first two sample periods. Botrylloides reaches a peak abundance in April 1991and the colonies senesce by June. This exposes the Schizoporella colonies that have survived overgrowth and by August Schizoporella reaches peak abundance. Distaplia recruitment and subsequent growth reduces the cover of Schizoporella again by direct overgrowth. Distaplia senesees shortly after it peaks (November - December) and re-exposes the overgrowth resistant Schizoporella colonies. At this point the space is divided between Schizoporella, Distaplia, and new Botrylloides recruits (on two panels). Occasionally this sequence is altered by the early recruitment and space occupation by a species that is resistant to fouling by other species (e.g., Alcyonidium in one panel and early Schizoporella recruitment in a second). This "successional" pattern manifests itself to some extent in all four panels as demonstrated by comparisons of community composition at the endpoints of experiments 1 (August 1990) and 2 (February 1992) as well as the August 1991 sample of experiment 2 (Figure 82). Sutherland (1974, 1977a, b, 1978) demonstrated that two alternate stable states exist in the encrusting communities ofBeaufort, North Carolina. The solitary ascidian, Styela plicata, and the bryozoan, Schizoporella errata develop dense monocultures which resist the settlement of other species and thus have the propensity to hold space and persist for long periods of time, shifting states only with the massive die-off of the entire assemblage. The adults of both species were good competitors, able to gain and hold space. In this system however, the larvae of Schizoporella errata were poor interference competitors, i.e., they were unable to recruit onto occupied substrate. 123 The Japanese fauna from which Schizoporella unicornis and Botrylloides violaceus were introduced is a diverse assemblage of encrusting species with a large number of tunicates and bryozoans (Hirata 1986, 1987, 1991; Nandakumar et al. 1993). Within these communities the competitive abilities ofSchizoporella and Botrylloides relative to the other bryozoans and tunicates are low. Hirata (1987) considers both species to be transitional in that they settled on fouling panels after 4 months of initial immersion and were replaced after 13 months. Schizoporella never achieved densities greater than 3% and Botrylloides reached maximum densities of20% space. Yet in the invaded communities of Coos Bay these two species are competitive dominants for space with the ability to overgrow a majority (90%) of the native species, generally reaching maxima of greater than 90% space coverage. Both introduced species recruit year-round (Powell 1970; Ross and McCain 1976; Chapter 1) at moderate densities with peak recruitment in mid to late summer at levels of 20 to 45 recruits-panel-1-mo-1. The majority of native encrusting species recruit during limited periods of the year (Chapter 1: Figure 20) at moderate densities. Schizoporella and Botrylloides both have the ability to settle on a wide variety of other species as substrates; that is few native species resist epizooism by Schizoporella and Botrylloides. From the opposite perspective, the introduced species are highly resistant to being settled upon by most species (Figure 58). Two native species have comparable resistance to epizooism, Cheilopora praelonga and Distaplia occidentalis. While Distaplia recruits at similar densities, it recruits during a short period in September (Chapter 1). In Coos Bay the competitive hierarchy of the invaded site demonstrates that, with few exceptions, the introduced species are dominant. Botrylloides and the native tunicate Distaplia tie; both outcompete Schizoporella unicornis; but the ability of the bryozoan to resist the lethal effects of overgrowth prevents competitive exclusion and local extinction. Thus Schizoporella, the inferior competitor, is the most persistent of the three. Todd and 124 Turner (1988) documented a variety of bryozoans with the ability to undergo long periods of non-lethal overgrowth. Schizoporella unicornis is a member of the Scottish encrusting community and they documented its ability to withstand overgrowth by a variety of colonial tunicates, including Botryllus schlosseri and Botrylloides leachii, for 50 to 165 days. They speculated that the alterations in the competitive hierarchy by these examples of incomplete interactions may add an additional stochastic component to community diversity. I have documented similar periods of overgrowth survival in Coos Bay for Schizoporella unicornis (Figure 76). The communities at these two sites, a native, uninvaded site (North Jetty) and the invaded Point Adams Jetty, have very different patterns of community development. The North Jetty communities accumulate species throughout the study but with no discernible pattern. These communities have high quantities of bare space at all times, interspersed with periods of high species specific recruitment (e.g., Balanus glandula). Thus these communities appear to be alternately disturbance and recruitment driven systems. In contrast the invaded site exhibits a sequential addition and replacement of species that is identical in all replicate panels. This directional change in species composition cannot be deemed "successional" because it appears to be driven by introduced species. These introduced species have however, drastically altered the pattern and trajectory of community development through the domination of primary space. As invaders the Japanese species have a unique array of life history traits which contribute to their position in the community. The availability of larval recruits year-round will increase the likelihood of available space being occupied by these species. This recruitment effect is increased by the recruitment abilities of Schizoporella and Botrylloides. The amount of space that is perceived to be available by Schizoporella and Botrylloides includes both the bare space on a panel but also the space occupied by species that are susceptible to being settled upon. Thus at any given time the · 125 ilarvae of these introduced species are more likely to recruit than their native counterparts. Once these species have established colonies, they have the competitive overgrowth abilities to expand and outcompete 90% of the native species. These colonies also resist the larval recruitment by native species, and thus maintain space through a combination of interference and exploitative mechanisms. There is now evidence that the solitary tunicate of Sutherland (1974, 1977a, b, 1978) and Sutherland and Karlson (1977), Styela plicata has been introduced to the Atlantic coast ofNorth America (1. Carlton pers. comm.). Thus Sutherland, without realizing it, elucidated a system in which an alternate stable state was established by an introduced species. The community at the Point Adams Jetty in this study has been invaded by several introduced species which together dominate the space and alter the community 'development. In comparison with the native, uninvaded communities at the North Jetty these introduced species have had drastic effects and are responsible for shifting the community into an alternate basin of attraction (Lewontin 1969). While this alternate state is not stable in the sense of being unchanging (Sutherland 1981; Pimm 1984), the community has experienced a compositional shift that was not previously accessible. 126 Figure 40. Map of the Lower Coos Bay Showing the North Jetty (Site 3) and Point Adams Jetty (Site 14). Scale Equals 1.0km. Fossil Point Jetty OIMS South Slough km o 0.5 1.0 1 , Charleston 127 \ . .~ .. . -~; . \ Figure 41. Army Corps of Engineers Flow Field Diagram for the Lower Coos Bay During Flood Tide (COE 1979). Figure 42. Army Corps of Engineers Flow Field Diagram for the Lower Coos Bay During Ebb Tide (COE 1979). 128 • I j li.. ·.- ~ .. .-' ./ I North Spit Charleston --.-.:....... 4 •~::::::~H"""-H~ri4~~~""~ ...-. .-..--.-- --_ ... - 14.lcm •• Concrete Building Block ~ 129 Machine Screw~ Backing Panel ... ~ ~ (...·.·.·.·.·.·.·.·...·;·.·.·.·.·.·....·.·...1.·...·.....·.·.·......;..:.-...·.·.·...;.....1 Subpanels -------- Panel Array with Four Subpanels Plastic SpacerC) B) A) Figure 43. Diagram Illustrating the 2x2 Panel Array in Plan (A) and Side (B) Views and Concrete Block Design (C). See Text for Further Explanation of Design. 130 20 ,--------~--------:-------~ 20 Or6---+----+-----+---+-----je-----+-----+------l o 5 10 15 Immersion Period (mo) --- Closed -0- Partial ..... Open Figure 44. Species Accumulation at the North Jetty During Experiment 1 Between April 1989 and August 1990. Lines Represent Mean Species Richness (n = 4) with Standard Deviation Error Bars. (/) 15 a> ·0 a> 0- en '0 10 ....hhh•• •• L- a> .c E ::J Z 5 131 20 ~-------------------~ 205 10 15 Immersion Period (mo) ..... Closed -e- Partial ....... Open 0&6----t---+---1----+---+---+---t---__i o Figure 45. Species Accumulation at the Point Adams Jetty During Experiment 1 Between April 1989 and August 1990. Lines Represent Mean Species Richness (n = 4) with Standard Deviation Error Bars. en 15(]) '0 (]) 0- en '0 10 '-(]) .c E :J Z 5 132 14 ,------------------------, 20 • •• • • ··············G··· _ . .•.......................- ..•._.•...- .... o 5 10 15 Immersion Period (mo) o ". ~ o 12 Figure 46. Species Accumulation at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data. (/) .(1) 10 U(1) C. CI) 8 \t- O CD 6 .c E ::J 4 Z Figure 47. Species Accumulation at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data. 133 20 I ·······1 • •• ......................................................••••........................................... .. - _ __ Q- _--_ _ . ............. @ .. o 0 @ ....HO.·········· ....................................•... ··0· •• ··• • @-@)--O-+----+--+-----+--I----+--I--+-J 5 10 15 Immersion Period (rna) 14 12 (j) ,<1> 10 () <1> 0- 8en -0 '- 6<1> .0 E :J 4z 2 0 0 134 20 • 0 .. 0 .;.:. o ® 0 • •• ..e.....J~ ..~.J~ ..... 5 10 15 Immersion Period (mo) o +-,-4IIIIIIle-8-+-~~I\-----t---:--_+--+_-__t---+---_j o 2.5 -,-------------------------, Figure 48. Shannon-Weaver Diversity (H') Change at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data. L- a> > ro ~ 1 c o c c 0.5 ro .c C/) ~ 2 h ••••••• • •••• h •• • .~ a> >Cl 1.5 --_._ - . I -- 135 2.5 .,--------------------------, 205 10 15 Immersion Period (mo) Figure 49. Shannon-Weaver Diversity (H') Change at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data. o .;:::. o L- Q) >CO ~ 1 I C o c c 0.5 co .c CI) £ 2 ~ Q) > • •o 1.5······· .. ·...·_·· _..........._._............_................ . I -- 18 ., 16 • 0- .............. _.... ······er············ - @ • •.................................• . .. • •.......- . ......... 0-. o - © 6 8 10 12 14 Immersion Period (rna) 42 o +-......__-~~.}-Gt+-----,f---+--+--+-l-----t--+---+-+--+--+-+--- o Figure 50. Evenness (1') Change at the North Jetty During Experiment 2 Between September 1990 and February 1992. The Line Represents the Mean of All Replicate Samples, Symbols Represent the Individual Panel Data. 136 0.8 0.2 1 -:;-. ., ';;; 0.6 en Q) c c~ 0.4 w 137 18 @ @ 0 @ .. • 16 . ..• e ~ ....CL . @ ............. • • • 6 8 10 12 14 Immersion Period (mo) I 4 @@ I 2 o +---+-------G~....__c._e_....____4~e____J 0... 143 .... .... .... .... ................_ . f .... ................. .... t A ....................._ _.- . ........ .... ............ ta\.L ...4._...... ..~.... ..~_........... .. .... 1 O.J?--t---+-t--+--+----!-+--+---+---i o 0.2 0.4 0.6 0.8 1 Introduced Proportion of Community I _ 0.8 o c: o t00.6 0. e a.. ~ 0.4 <..> ::J "C o~ 0.2 c: Figure 57. Introduced Species Proportion of Species Diversity (H') Correlated with the Introduced Species Proportion of Community Richness. Triangles Represent Individual Replicate Monthly Samples. The Line Represents a Direct Correlation. Figure 58. The Number of Species that are Settled Upon (Fouled) by Species X as a Function of the Number of Species that can Settle Upon (Be Fouled By) Species X. Species Identified are Botlylloides violaceus, Bv; Botlyllus schlosseri, Bs; Schizoporella unicomis, Su; Distaptia occiden/alis, Do; Balanus glandula, Bgl; Serpulid spp., Serp; and Cheilopora praelollga, Cpo 5 20 1m B91 A Su A Non-native ~ Native 5 10 15 Number of Species on Sp X 1m Serp 4. Bs Fa Do 144 A Bv Cp _.__.....•...._.._ _ _ _.....•..........._ _............•....•.•............._...•...............•......................._. mimi III 1m Fa Fa Fa mm P&I o o 20 x c.. en >. ~ 15 a> "5 o LL .~ 10 u a> c.. en '0 L- a> .c E ::J Z 0.2 +H.. """"'"""""""""""""."."'."""'.' """._,-"","""," .. '" " .. """""".".'.. """"'." '""' ·········1 0.4..,---------------------------, 145 1816 ......................;6: . 14 y=-0.057x - 0.11; r sq = 0.58 6 8 10 12 Immersion Period (rna) 42 ..... ... -1 o Figure 59. The Association Between Native and Introduced Species Cover on Replicate Panels at Point Adams Jetty as a Function of the Period of Immersion (see Text). -c Q) '0 0 +---.--------------------------1 IE Q)8 -0.2 c ~ -0.4 «lQ) ~ -0.6 u -0.8 ..... ..._.._HHH .HHH' 146 10 .----------------------, 2015 ..... Point Adams Jetty 10 Species Richness 5 ..... North Jetty o Figure 60. Experiment 1 Immigration and Extinction Rate Regression Lines as a Function ofResident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regressions are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis. Q) 10 8 c:: c:: o :g 6 c:: ~ w 4 '-o c:: o :o;:l 2 ~ 0> ~0+-------------1,-------:~-----t---~ Figure 61. Experiment 2 Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regressions are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Label1ed with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis. 2 147 1084 6 Species Richness ...... Point Adams Jetty ..... North Jetty o 3.5 (1) 3 ..... ro 0:: 2.5c 0 ts 2c ~ w 1.5 L- 0 c 10 :.;::::; ~ 0.5C> 'E E 0 148 10 -,----------------------~ 15 10.7 0.9 5 10 Species Richness -e- Native-only Species -... Total Species Pool o Figure 62. The Total Species Pool and Native-Only Species Pool Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regressions are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis. Q) ro 8 0:: c:: oU 6 c:: ~ w 4 L- o c:: o :p 2 ~ C> "E 0 E 149 108 -.- Total Species Pool 4 6 Species Richness 2 ....- Native-only o Figure 63. The Total Species Pool and Native-Only Species Pool Immigration and Extinction Rate Regression Lines as a Function of Resident Species Richness. Immigration Regressions are Represented by Thick Lines; Extinction Regressions are Represented by Thin Lines. Predicted Values of the Species Equilibrium (S) are Labelled with Vertical Lines from the Intersection of the Immigration and Extinction Regression Lines to the x-axis. 4 ...---------------------------, Q) +oJ &. 3 c a :g ~ 2 w L- a c 1 a +:i ~ C) 'E 0 +-::::~IIlP""'=-==-----------_+_----_"!" E 100 ~---------------------, 150 1 O~~""~H-"""'"----t--+----l--+---+-+---+----t-----+------i 2 4 5 6 9 10 11 13 14 15 16 17 Immersion Period (mo) Figure 64. Bray Curtis Mean Similarity Between Replicate Panels at the North Jetty During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean with Range Bars (n = 4). ~ 60 °C o~ E ~ 40 c: Q) ~Q) 0.. 20 _ 80 () CD I ~ ........... 100 .-----------------------, _ 80 () (II I T"" .......... ~ 60 "C: ~ j~ ~ 40 c: Q) ~ Q) 0... 20 O+----.JL....-+--+----+--f-""--I--+--t---t---t---+-+---t-i 1 2 4 5 6 9 10 11 13 14 15 16 17 Immersion Period (mo) Figure 65. Bray Curtis Mean Similarity Between Replicate Panels at the Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean with Range Bars (n = 4). 151 Figure 66. Bray Curtis Mean Similarity Between Replicate Panels Within and Between the North Jetty and Point Adams Jetty Sites During Experiment 2 Between September 1990 and February 1992. Solid Line Represents the Mean of Four Replicate Panels. -if- North Jetty vs. Point Adams 152 1816146 8 10 12 Immersion Period (mo) 4 .... North Jetty ""l6.- Point Adams Jetty 100 ..-..() 80CD I ~ '-' ~ 60·C .~ .~ en 40 ... C Q) U '- 20Q) G- O 0 2 20 ~-----------------~---, 153 20 AA A A A.A. ..... A.A. • •• -.-.-e---------- • AA 5 10 15 Immersion Period (rno) • North Jetty A. Point Adams Jetty O-=- I--___e_-+---t-------t---t------!-------t----J o Figure 67. Mean Cumulative Number of Species at the North Jetty and Point Adams Jetty During Experiment 2 Between September 1990 and February 1992. Symbols Represent Replicate Panel Data, Lines Represent Means. ~ Q) .c § 15 z en Q) '0 ~10 en Q) >:.;:; ro :; 5 E :J U 10 -,-------------k---,~_____:lIlli.___-__, 154 20 ..... ..........._ _ _ . 5 10 15 Cumulative Species Richness ... Point Adams Jetty • North Jetty O .......e_-+-----t--+---+---+--f---t--- o fI) fI) <1> c: 8 .c.() a: fI) 6 <1> '0 <1> tJi 4 '0 <1> .~ ro 2 <1>' 0:: Figure 68. Realized Species Richness as a Function of Cumulative Species Richness for the North Jetty and Point Adams Jetty During Experiment 2. Symbols Represent Monthly Panel Data; Thick Line Represents the North Jetty Regression and the Thin Line Represents the Point Adams Jetty Regression. 155 600 ,-----------------------, Figure 69. Balanus glandula Settlement Densities During Experiment 2 Between September 1990 and February 1992. Symbols Represent Four Replicate Panels. 205 10 15 Immersion Period (mo) © Y ··_· __···-····,_·u:~-In it \\ " I- I i ~\ ..--.-- - ---- _-".- _..f: ; ',: .-..-.- ------.. "'1\ • I! \!,;.',:"!.f\ ! ! \ i : ..._....i~.---- "n \\ \\ ······..···tt·- - ..------.- - . /t·····.\.L.;...:,-..,.··,·.······· j \.\ i .~ . \ ", O =..,..,.~~TI-t--o---Gl' F<=::-"~-m·······ro--:c:.'-t---- ---I*---Il1I--tiJ--m-..@;f+--{~m---+----Ir, ;,.:.. ;.;::. :-:.:. o 500 100 en ro ~ 400 :~ "'C c::: :: 300 o '-Q) .c E 200 ::J Z