ISSN 0038-3872 S sssy nM SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Volume 114 TIN Number 1 114(1) 1-62 (2015) Zi eiooz QQ N010N1HSVM MN '3AV NOUfUllSNOO QNV IS H101 S2 HNWN 30NVH3X3/SN0lllSin03V [ Ad03 'ainiliSNI NV1N0SH1IWS April 2015 Southern California Academy of Sciences Founded 6 November 1891, incorporated 17 May 1907 © Southern California Academy of Sciences, 2015 2014-2015 OFFICERS Julianne Kalman Passarelli, President David Ginsburg, Vice-President Edith Read, Recording Secretary Ann Dalkey, Treasurer Daniel J. Pondella II and Larry G. Allen, Editors - Bulletin Brad R. Blood, Newsletter Shelly Moore, Webmaster ADVISORY COUNCIL Jonathan Baskin, Past President John Roberts, Past President John H. Dorsey, Past President Ralph Appy, Past President Brad R. Blood, Past President 2012-2015 Bengt Allen Shelly Moore Ann Bull Dan Cooper Mark Helvey BOARD OF DIRECTORS 2013-2016 Ann Dalkey Julianne K. Passarelli Edith Read Danny Tang Lisa Collins 2014-2017 David Ginsburg Gordon Hendler Andrea Murray Tom Ford Gloria Takahashi Membership is open to scholars in the fields of natural and social sciences, and to any person interested in the advancement of science. Dues for membership, changes of address, and requests for missing numbers lost in shipment should be addressed to: Southern California Academy of Sciences, the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007-4000. Professional Members $60.00 Student Members 30.00 Memberships in other categories are available on request. Fellows: Elected by the Board of Directors for meritorious services. The Bulletin is published three times each year by the Academy. Submissions of manuscripts for publication and associated guidelines is at SCASBULLETIN.ORG. All other communications should be addressed to the Southern California Academy of Sciences in care of the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007-4000. Date of this issue 15 July 2015 ® This paper meets the requirements of ANSI/N!SO Z39.48-1992 (Permanence of Paper). Bull. Southern California Acad. Sci. 114(1), 2015, pp. 1-11 © Southern California Academy of Sciences, 2015 Possible Stock Structure of Coastal Bottlenose Dolphins off Baja California and California Revealed by Photo-Identification Research R.H. Defran,1’* Marthajane Caldwell,2 Eduardo Morteo,3,4 Aimee R. Lang,5 6 Megan G. Rice,7 and David W. Weller5 1 Cetacean Behavior Laboratory, San Diego State University, 11060 Delphinus Way, San Diego, CA 92126, USA 2 Marine Mammal Behavioral Ecology Studies Inc., 8429 Cresthill Avenue, Savannah, GA 31406, USA 3Instituto de Ciencias Marinas y Pesquerias, Universidad Veracruzana, Calle Hidalgo #617, Col. Rio Jamapa, C.P. 94290, Boca del Rio, Veracruz, MX 4Instituto de Investigaciones Biologicas, Universidad Veracruzana, Av. Dr. Luis Castelazo Ayala SIN, Col. Industrial Animas, C.P. 91190, Xalapa, Veracruz, MX 5Marine Mammal & Turtle Division, Southwest Fisheries Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, 8901 La Jolla Shores Drive, La Jolla, CA 92037, USA 6 Ocean Associates, Inc., 4007 North Abingdon Street, Arlington, VA 22207, USA 1 California State University, San Marcos, 333 S. Twin Oaks Valley Rd., San Marcos, CA 92078, USA Abstract. — Boat-based photo-identification research has been carried out on bottle- nose dolphins in eastern North Pacific coastal waters off northern Baja California, Mexico and southern and central California, USA from 1981 to 2001. Within these waters, bottlenose dolphins routinely travel back and forth between coastal locations while generally staying within a narrow corridor extending only 1-2 km from the shore. Inter-area match rates for 616 dolphins photo-identified between 1981-2000 in four California coastal study areas (CCSAs) of Ensenada, San Diego, Orange County and Santa Barbara averaged 76%. To explore possible southern range limits for these dolphins, photo-identification surveys were carried out in the coastal waters off San Quintin, Baja California, Mexico between April- August 1990 (n= 8 surveys) and July 1999 to June 2000 («= 12 surveys). The 207 individual dolphins identified off San Quintin were compared to the 616 dolphins identified in the CCSAs. The inter-area match rate between San Quintin and the CCSAs was 3.4% (n=l dolphins). This low rate contrasts sharply with the much higher average match rate of 76% observed between the CCSAs. These differences in match rates suggest that both a California coastal stock and coastal Northern Baja California stock may exist, with only a limited degree of mixing between them. The common bottlenose dolphin ( Tur slops truncatus) is the most frequently encountered cetacean in the nearshore waters of California and Baja California, Mexico. Two distinct bottlenose dolphin ecotypes occur in these waters: a coastal form that is typically found within 1-2 km of shore (Carretta et al. 1998; Defran and Weller 1999; * Corresponding author: rh.defran@gmail.com 1 ^thso/v^ JUL 2 1 2015 .pBRARieS 2 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Bearzi 2005) and an offshore form that is distributed in deeper waters, typically greater than a few kilometers from shore (Defran and Weller 1999; Bearzi et al. 2009). Differentiation of these two ecotypes, which are managed as separate stocks by the National Marine Fisheries Service (Carretta et al. 2013), is supported by morphological (Walker 1981; Perrin et al. 2011), photographic (see Shane 1994) and genetic data (Lowther-Thieleking et al. 2014). The California coastal stock is small, estimated to contain about 450-500 individuals (Dudzik et al. 2006; Carretta et al. 2013) that are distributed between Monterey, California and Ensenada, Baja Mexico (Defran et al. 1999; Hwang et al. 2014), with occasional sightings as far north as San Francisco, California1. Photo-identification research has been carried out on the coastal stock off California, and to a lesser extent off Northern Baja California, since the early 1980s. Areas off California and Baja California where photographic data have been collected include: (1) Ensenada, (2) San Diego, (3) Orange County, (4) Santa Monica Bay, (5) Santa Barbara, (6) Monterey Bay and (7) San Francisco Bay (Fig. 1). In general, photo-identification data have shown that California coastal dolphins display little site fidelity to any portion of their distribution (Defran et al. 1999; Hwang et al. 2014). Instead, they routinely travel back-and-forth within their range, on some occasions in excess of 900 km, while at the same time typically staying very near shore (Defran et al. 1999; Hwang et al. 2014). Records from the nineteenth century suggest that coastal bottlenose dolphins may have once occurred in Monterey Bay and San Francisco Bay (Dali 1873; True 1889; Orr 1963). More recent studies, however, considered the northern range boundary to be located off Los Angeles County up until the early 1980s (Norris and Prescott 1961; Dohl et al. 1981; Leatherwood and Reeves 1982). The 1982-83 El Nino Southern Oscillation (ENSO) dramatically impacted the coastal marine ecosystem off California and Baja. It was during this ENSO event that California coastal stock dolphins extended their northern range back to Monterey Bay (Wells et al. 1990). This northern range extension has persisted to the present day (Riggin and Maldini 2010; Maldini et al. 2010; Cotter et al. 2011) and now extends even further north to San Francisco Bay and most recently, Bodega Bay1. The southern boundary of the California coastal stock is less well known but photo- identification data demonstrate that it extends to at least Ensenada (Defran et al. 1999; Hwang et al. 2014). In this research, boat-based photo-identification surveys of coastal bottlenose dolphins were carried out south of Ensenada off San Quintin Bay, Baja California (Figs. 1 & 2). The goal of this research was twofold: (1) to examine the degree of overlap between coastal dolphins photo-identified off San Quintin and those photo- identified in study areas off Ensenada, San Diego, Orange County, and Santa Barbara, and (2) to use photo-identification data to determine if the southern range of the California coastal stock extended as far south as the San Quintin area. Methods The general design used in this study was the same as our earlier studies that compared independently collected bottlenose dolphin photo-identification catalogs from California and Baja California (Defran et al. 1999; Hwang 2014). 1 Szczepaniak, I., W. Keener, M. Webber, J. Stem, D. Maldini, M. Cotter, R.H. Defran, M. Rice, G. Campbell, A. Debich, A. Lang, D. Kelly, A. Kesaris, M. Bearzi, K. Causey, and D. Weller. 2013. Bottlenose dolphins return to San Francisco Bay. Poster presented at the 20th Biennial Conference on the Biology of Marine Mammals, Dunedin, New Zealand December 9-13. BOTTLENOSE DOLPHIN STOCK STRUCTURE, SAN QUINTIN, BAJA CALIFORNIA 3 Fig. 1 . Coastal locations where California coastal stock bottlenose dolphins have been photo-identified. Point Conception and Punta Colonet are included to indicate the northern and southern coastal boundaries of the Southern California Bight. Study areas marked with an asterisk indicate those that were compared to San Quintin sightings (Table 1). Study Area This study was conducted in the coastal waters south of San Quintin Bay, Baja California, during two independent study periods: 1) April, June and August 1990, n= 8 surveys (Caldwell 1992); and 2) July 1999 to June 2000, n= 12 surveys (Morteo et al. 2004). The San Quintin study area was located approximately 376 km south of San Diego and about 200 km south of Ensenada. Within the study area, the survey track extended 4 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 32 km southward from a point 8 km east of Azufre Point (30o23’50” N; 115°54,42” W) south to Rosario Canon (30°09’06” N; 115°48’27” W) (Fig. 2). Most surveys in 1990 began at the Base Camp near El Socorro and extended 18 km south to Rosario Canon. During the 1999-2000 study period, surveys began about 8 km east of Azufre Point and extended 26 km south to Hondo Creek (Fig. 2). Photo-Identification Surveys and Photographic Data Analysis Survey methodology and photo-identification analysis procedures employed during both study periods in San Quintin followed those used previously in the Ensenada, San Diego, Orange County and Santa Barbara study areas, hereafter referred to as California coastal study areas (CCSAs). Detailed descriptions of these procedures are provided BOTTLENOSE DOLPHIN STOCK STRUCTURE, SAN QUINTIN, BAJA CALIFORNIA 5 Table 1. Summary information on survey effort, study period, photographic data, and data sources for all study areasa. Study area Number of surveys (complete, partial) Study period Number of dolphins identified San Quintin 20 (20, 0) 19901, 1 999-20002 207 Ensenada 23 (23, 0) 1985-19863, 1999-20004 129 San Diego 241 (157, 84) 1981-19895, 1996-19996&7 518 Santa Barbara 73 (55, 18) 1987 & 19893, 1998-19997 213 Data sources: 1 Caldwell (1992), 2Morteo et al. (2004), 3Defran et al. (1999), 4 Guzon-Zatarain (2002), 5Defran and Weller (1999), 6Dudzik (1999), 7 Lang (2002). aSome numbers differ from those given in original data sources due to refinement and revision of the dataset over time and the elimination of sightings not meeting the specified photographic quality criteria. elsewhere (Caldwell 1992; Defran and Weller 1999; Defran et al. 1999; Dudzik 1999; Lang 2002; Morteo et al. 2004) but are briefly described here. Photographic surveys involved slow travel in small boats while moving parallel to the coast and outside the surf line; generally within 500-750 m of shore and corresponding to water depths between 4 m to 10 m. Surveys were conducted in sea state and visibility conditions adequate for finding and photographing dolphins. Although past data demonstrated that most coastal bottlenose dolphins are typically found within 500 m of the shore (Hanson and Defran 1993; Defran and Weller 1999; Bearzi 2005; Carretta et al. 2013), two or more observers, nevertheless, visually searched the area from the shore to ~ 2 km offshore to ensure complete coverage of coastal waters. Once a group of dolphins was sighted, initial estimates of group size, as well as information on time, location, environmental conditions and behavior were recorded. Following initial estimates of group size, the survey vessel maneuvered to a distance from the dolphins suitable for photo-identification. Thirty-five millimeter SLR film cameras equipped with telephoto lenses were used to photograph all dolphins (marked and unmarked) within a group. Initial estimates of group size were revised as necessary, and contact with the group was maintained until photographic effort was completed, or dolphins began exhibiting avoidance behavior. Identical procedures were repeated as the vessel resumed travel on the predetermined survey route and as additional dolphin groups were encountered. The best quality photograph of every dolphin was scanned and converted into a high- resolution digital image. Of these, only high quality photographs of dorsal fins with two or more distinctive dorsal fin notches were used for analysis. Distinctive dorsal fins were those that had sufficient notching on the leading or trailing edge such that they could be matched to high quality dorsal fin photographs from other sightings (Urian and Wells 1996; Defran and Weller 1999; Defran et al. 1999; Mazzoil et al. 2004). Only unambiguous matches were accepted as resightings (i.e., a re-identification of a previously identified individual). Dorsal fin images from selected CCSAs (marked with an asterisk in Fig. 1) were analyzed and maintained in the Cetacean Behavior Laboratory at San Diego State University. The combined photo-identification catalog for the CCSAs consisted of 616 individuals identified during two sample periods: (1) 1981 to 1989, and (2) 1996 to 2000. Table 1 provides a summary of survey effort, study period, photographic data and data sources for each of the CCSAs. 6 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 0) o Q 25.4 cm) O. mykiss has been observed (Krug et al. 2012). Benthic macroinvertebrates (BMI) are an important food source for both P. clarkii and O. mykiss (Angradi and Griffith 1990, Nystrom and Graneli 1996). Competition for food resources and disruption of BMI community functionality is a potential concern. The complexity of functional feeding groups (e.g., gatherers, filterers, scrapers, predators) can be a measure of the functional integrity of BMI communities and a reflection of its capacity to cycle nutrients (Wallace and Webster 1996). Disturbance to the benthic community, such as the introduction of non-native fauna, can alter BMI community composition and cause unanticipated changes in freshwater ecosystems (Covich et al. 1999). Changes in BMI abundance, diversity, and feeding group complexity can indicate such community disturbance. In Topanga Creek, drought induced low flows in 201 1-2014 resulted in isolated refugia pools and reduced numbers of O. mykiss redds and young of the year1. However, P. clarkii were able to successfully reproduce and inhabit the shallow riffles and fragmented reaches inaccessible to O. mykiss. In September 2013, the Resource Conservation District of the Santa Monica Mountains (RCDSMM), in conjunction with the Watershed Stewards Program (WSP), launched a citizen science program that 1 ) removed crayfish from several refugia step-pool habitats within a 200 meter reach of Topanga Creek, 2) measured crayfish demographics (sex/length), and 3) monitored water quality (dissolved oxygen, pH, salinity, conductivity, turbidity, water temperature), nutrient levels (nitrate, nitrite, ammonia, phosphate), and BMI community metrics. Materials and Methods Topanga Creek (34° 6T1”N 118° 36’ 18” W, gradient 1 to 6%) is the main drainage of a small coastal watershed (approximately 47 km2) located within the Santa Monica 1 Krug, J., R. Dagit, Stillwater Sciences, and J.C. Garza. 2014. Lifecycle monitoring of Oncorhynchus mykiss in Topanga Creek, California. Final Report Prepared for CA Department of Fish and Wildlife, Contract No. P0950013. January 2014. 14 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Mountains National Recreation Area in southern California. The study reach consisted of 400 continuous meters in Topanga Creek, starting at 3500 m (upstream from the ocean) and ending at 3900 m. The study area included a downstream 200 m crayfish removal reach (RR), and an upper 200 m non-removal reach (NRR; Fig. 1). Both reaches were relatively uniform in geomorphological features, including a similar distribution of pools, step-pools, runs, and riffles, substrate type, and percent canopy cover. No introduced barriers of any sort were incorporated into the study reaches; however, natural low-flow boulder barriers separated the RR from the NRR. A total of ten volunteer crayfish removal events took place between September 2013 and April 2014. Water quality, nutrient, and BMI samples were collected in both 200 m reaches during removal events between November 2013 and April 2014. Crayfish were removed throughout RR with 7.6 cm hot dog pieces attached to hemp strings. The presence of federally listed O. mykiss prevented setting traps of any kind. Crayfish were counted, sexed, and measured (cm) from the tip of the rostrum to the end of the tail in midline. Removed crayfish were donated to a local wildlife rescue or used for educational purposes. Water samples were collected from three pools within each 200 m reach an hour prior to removal. Each site was tested for air temperature (mercury thermometer), salinity (ATC 300011 SPER SCIENTIFIC salt refractometer), pH (Oakton pHTestr 30), conductivity (Oakton ECTestrl 1), dissolved oxygen (DO) and water temperature (YSI 55 DO meter). All probes were calibrated within a week prior to the collection date. Nutrient and turbidity sampling was conducted once a month from November 2013 through April 2014 at 3500 m, 3550 m, and 3600 m in RR and at 3700 m, 3800 m, and 3850 m in NRR. Samples were tested for nitrate-N (ppm), nitrite-N (ppm), ammonia-N (ppm), orthophosphate (ppm) and turbidity (NTU) within eight hours of collection using LaMotte SMART3 colorimeter and LaMotte 2020we turbidity meter. BMI samples were collected according to CA Rapid Bioassessment protocol2 in November 2013, December 2013, February 2014, and April 2014 at three comparable sites in RR and NRR. Each sample was composed of nine kicks into a 1-ft. wide D-frame net (three transects and three kicks per transect). Samples were preserved in 95% ethanol and processed within a month from collection date. BMI were identified to genus, or lowest possible taxonomic level using a 40x magnification dissecting microscope. P. clarkii was recorded but not included as a benthic macroinvertebrate for analysis. For quality assurance, 10 percent of samples were randomly selected and re-identified by a second processor. First and second identifications were compared and scored for accuracy, resulting in an estimated error of 1.6%. Paired t-tests were applied to determine any significant difference between the two reaches in crayfish demographics, water quality, nutrient levels, and biotic integrity metrics of BMI communities. Regression analyses were performed to compare water quality metrics to crayfish removal and to analyze the relationship between catch per unit effort and water temperature. Simpson’s Index of Diversity (Simpson 1949) was calculated for each BMI sample and analyzed by paired t-test to compare biodiversity. Simpson’s was also applied to samples categorized by functional feeding groups (gatherers/filterers, scrapers, predators, or other) to compare feeding group complexity. Southern Coastal California Index of Biotic Integrity (SCC-IBI; Ode et al. 2005) metrics 2 Ode, P.R. 2003. CAMLnet: list of California macroinvertebrate taxa and standard taxonomic effort. Aquatic Bioassessment Laboratory, Rancho Cordova. Retrieved September 10, 2014 from http:// www.safit.org/ste.html. PROCAMBARUS CLARKII IN TOPANGA: REMOVAL AND ECOSYSTEM EFFECTS 15 3700-3900m 3500-3700m NEVADA San Francisco'? Jose Topanga Creek Watershed itlFORNh Los Angeles* Topanga Watershed Crayfish sampling site Topanga Creek Sampling Locations 1 in = 1 miles Source: Imagery - ESRI Topanga Watershed - CalWater Sampling Locations - RCDSMM Projection: NAD 1983 Albers **Distances are linear meters from the ocean Fig. 1. Map of Topanga Creek Watershed and the crayfish study reaches (3500-3700 Removal Reach (RR); 3700-3900 Non-Removal Reach (NRR)). 16 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES H3 O flj u W VI u Tfc 250 200 150 100 50 SEP 13 OCT 13 NOV 13 DEC 13 JAN 13 FEB 13 FEMALE ■ MALE Fig. 2. Total number of male and female crayfish removed each month Oct. 2013 to Feb. 2014. (number of EPT, Coleoptera and predator taxa, and percent tolerant, intolerant, non- insect, and collector-gatherers + collector-filterers) were applied and scored for all BMI samples. Results Ten volunteer removal events between September 2013 and April 2014 (203.25 person- hours) resulted in the removal of 345 P. clarkii; 166 females and 179 males (Fig. 2). The average length of crayfish removed was 7.61(±0.348 SE) cm. There was no significant difference between male and female average length or number removed. The first event (9/21/ 2013) resulted in the most captures with more than four times as many crayfish removed than any proceeding month. The catch per unit effort (CPUE) in the study period November 2013 to February 2014 ranged from 0.1 to 3.0 crayfish per person per hour, and increased significantly with warmer water temperatures (R2= 0.67, F= 12.27, /?<0.05). An increase of approximately 0.26 CPUE was calculated for every 1°C increase in temperature (Fig. 3). The comparison of water quality and nutrient concentrations between the RR and NRR showed no significant differences, except for salinity. Salinity showed a statistical difference between reaches (paired two-tailed, /(3)=-4.65, p<0.02). The NRR had higher salinity throughout the course of the study, although levels in both reaches ranged from 0-2 ppm. The four BMI samples collected from the NRR in November 2013, December 2013, February 2014, and April 2014 contained a total of 645 individuals and 38 taxa. The samples collected from the RR contained a total of 3,642 individuals and 51 taxa. A total of four phyla were represented including Arthropoda, Annelida, Mollusca, and Nematoda. BMI abundance was significantly higher (paired two-tailed r(3)=3.59, p <0.04) in the RR (Fig. 4). In both reaches, there was an increase in BMI abundance from November through April. The NRR had significantly lower richness (paired one- tailed f(3) = 2.74, p< 0.04). However, species diversity as measured by Simpson’s Index of PROCAMBARUS CLARKII IN TOPANGA: REMOVAL AND ECOSYSTEM EFFECTS 17 *2 3.5 3 2.5 2 •a 3 u <0.05) (Fig. 6). Additionally, P. clarkii were collected more often in NRR BMI samples (3.1%, 20 individuals total) than in RR (<1%, 7 ind.). Discussion The invasive Procambarus clarkii has been shown to have severe effects on native aquatic wildlife in southern California streams (Riley et al. 2000, Gamradt and Kats 2002, Rodriguez et al. 2005, Cruz et al. 2006, Feminella and Resh 2006, Correia and Anastacio 2008, Ficetola et al. 2011). In Topanga Creek, benthic macroinvertebrate abundance and species richness were significantly higher in the 200m RR where crayfish were actively managed by hand- removal than in an adjacent NRR. This result is consistent with previous reports that correlate non-native crayfish presence to reduced BMI abundance in freshwater systems (Charlebois and Lamberti 1996, Stewart et al. 1998). In the RR, BMI samples contained between 23 and 51 distinct taxa and in the NRR, richness ranged from 6-38. This finding corroborates previous studies that have found that P. clarkii invasions lead to loss of BMI diversity (Rodriguez et al. 2005, Correia and Anastacio 2008). Functional feeding group diversity was lower in the NRR, and % of tolerant organisms was higher. Increased abundance of BMI in RR indicates higher productivity for a number of taxa. Six distinct taxa had more than 100 individuals in one or more samples from the RR including Viviparidae and Hydrobiidae, Chironomidae, Hyalellidae, Coenagrionidae, Ostracoda, and Physa. Only two taxa had more than 100 individuals in any one NRR sample: Chironomidae and Ostracoda. A major distinction between community was that Viviparidae and Hydrobiidae were most abundant taxa in RR, but relatively rare in NRR (3%). The relative rarity of freshwater snails (scrapers) in the NRR diminished feeding group complexity. Procambarus clarkii predation on Viviparidae in this reach is one possible driver of reduced abundance of the genus, although micro-habitat differences within the 400 m study reach are another potential factor. Higher abundance, species richness, feeding group complexity, and a smaller proportion of tolerant species indicate that the BMI community in RR was in better ecological condition than in NRR. As crayfish are generally the largest species within the BMI community, a comparison of BMI sample proportional dry weight of taxa groups would further our understanding of P. clarkia effects on trophic-level productivity by providing a quantitative measure of biomass. The ecological implications of invasive P. clarkii in Topanga Creek could be severe if they significantly disrupt benthic macroinvertebrate communities. BMI make up the primary consumer trophic level and play an integral part in nutrient decomposition and cycling through riparian systems. Changes at this level could impact higher trophic organisms such as California newts (species of special concern) and southern California steelhead trout (endangered). How the continuation of drought conditions within the region will continue to affect the population and impact of P. clarkii is uncertain; reduced flows and higher temperatures place stress upon aquatic natives, it renders riparian habitat more preferential for crayfish. Water quality and nutrient results between reaches were less notable. Salinity was the only parameter to differ significantly, which may be influenced by a groundwater seep in NRR at 3900 m3. Some studies have suggested P. clarkii may be a source of bioturbation 3GeoPentech. 2006. Hydrogeologic Study Lower Topanga Creek Watershed, Los Angeles County, CA. Prepared for the RCD of the Santa Monica Mountains. Topanga, CA. 20 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES (Mueller 2007, Yamamoto 2010), however, results in this study showed no significant difference in turbidity between the RR and the NRR. The level of effort per crayfish removed increased over the course of the study at a rate that correlated to decreasing water temperatures. While decreased activity is one possible factor, diminished crayfish numbers due to removal efforts is another. Removal events might be most efficient in warmer months; however a more extensive study including more removal areas and a longer time period is needed to determine whether there is a relationship between temperature and catch per unit effort, as well as to more completely characterize the effects of crayfish on water quality and the benthic macroinvertebrate community in Topanga Creek. Acknowledgements We would like to extend a special thanks to the following individuals and organizations for their contributions to this project: K. Vander Veen (Calvary Christian School), C. Najah (Topanga Youth Wildlife Project), Daniel Paz (Verbum Dei intern), RCDSMM Stream Team, Watershed Stewards Program, California Conservation Corps, and AmeriCorps. Funding was provided by LA County District 3, Supervisor Zev Yaroslavsky. This paper also benefitted from review by three anonymous reviewers. Literature Cited Angradi, T.R. and J.S. Griffith. 1990. Diel Feeding Chronology and Diet Selection of Rainbow Trout (Oncorhynchus mykiss) in the Henry’s Fork of the Snake River, Idaho. Canadian Journal of Fisheries and Aquatic Sciences, 47:199-209. Barnes, R. 1974. Invertebrate Zoology. Philadelphia, PA: W.B. Saunders Company. Charlebois, P.M. and G.A. Lamberti. 1996. Invading Crayfish in a Michigan Stream: Direct and Indirect Effects on Periphyton and Macroinvertebrates. Journal of the North American Benthological Society, 1 5(4):55 1-563. Correia, A. and P. Anastacio. 2008. Shifts in aquatic macroinvertebrate biodiversity associated with the presence and size of an alien crayfish. Ecological Research, 23(4):729. Covich, A.P., M.A. Palmer, and T.A. Crowl. 1999. The Role of Benthic Invertebrate Species in Freshwater Ecosystems. BioScience, 49:119-127. Cruz, M., R. Rebelo, and E. Crespo. 2006. Effects of an introduced crayfish, Procambarus clarkii, on the distribution of south-western Iberian amphibians in their breeding habitats. Ecography, 29(3): 329-338. Feminella, J. and V. Resh. 2006. Submersed macrophytes and grazing crayfish: an experimental study of herbivory in a California freshwater marsh. Ecography, 12(1): 1-8. Ficetola, F.G., M.E. Siesa, R. Manenti, L. Bottoni, F. De Bernardi, and E. Padoa-Schioppa. 201 1. Early assessment of the impact of alien species: differential consequences of an invasive crayfish on adult and larval amphibians. Diversity and Distributions, 17(6):1 141-1 151. Gamradt, S. and L. Kats. 2002. Effect of Introduced Crayfish and Mosquitofish on California Newts. Conservation Biology, 10(4):1 155-1 162. Geiger, W., P. Alcorlo, A. Baltanas, and C. Montes. 2005. Impact of an introduced Crustacean on the trophic webs of Mediterranean wetlands. Biological Invasions, 7(1):49— 73. Holmes, S. 1924. The genus Cambarus in California. Science, 60(1 555):358— 359. Kats, L., G. Bucciarelli, T. Vandergon, R. Honeycutt, E. Mattiasen, A. Sanders, S. Riley, J. Kerby, and R. Fisher. 2013. Effects of natural flooding and manual trapping on the facilitation of invasive crayfish-native amphibian coexistence in a semi-arid perennial stream. Journal of Arid Environments, 98:109-112. Krug, J., E. Bell, and R. Dagit. 2012. Growing up fast: diet and growth of a population of Oncorhychus mykiss in Topanga Creek, California. California Fish and Game Bulletin, 98(l):38-46. Momot, W.T. 1995. Redefining the Role of Crayfish in Aquatic Ecosystems. Reviews in Fisheries Science, 3:33-63. — , H. Gowing, and P.D. Jones. 1978. The dynamics of crayfish and their role in ecosystems. American Midland Naturalist, 99:10-35. PROCAMBARUS CLARKII IN TOPANGA: REMOVAL AND ECOSYSTEM EFFECTS 21 Mueller, K.W. 2007. Reproductive habits of non-native red swamp crayfish ( Procambarus clarkii) at Pine Lake, Sammamish, Washington. Northwest Science, 8 1 (3):246— 250. Nystrom, C.B. and W. Graneli. 1996. Patterns in benthic food webs: a role for omnivorous crayfish? Freshwater Biology, 36(3):63 1-646. Ode, P.R., A.C. Rehn, and J.T. May. 2005. A quantitative tool for assessing the integrity of southern California coastal streams. Environmental Management, 35(4):493-504. Riley, S.P.D., G.T. Busteed, L.B. Kats, T.L. Vandergon, L.F.S. Lee, R.G. Dagit, J.L. Kerby, R.N. Fisher, and R.M. Sauvajot. 2005. Effects of urbanization on the distribution and abundance of amphibians and invasive species in southern California. Conservation Biology, 19:1894-1907. Rodriguez, C.F., E. Becares, M. Fernandez- Alaez, and C. Fernandez- Alaez. 2005. Loss of diversity and degradation of wetlands as a result of introducing exotic crayfish. Bioinvasion Science, 7:75-85. Simpson, E.H. 1949. Measurement of diversity. Nature, 163:688. Stewart, T.W., J.G. Miner, and R.L. Lowe. 1998. An experimental analysis of crayfish ( Orconectes rusticus ) effects on a Dreissena- dominated benthic macroinvertebrate community in western Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences, 55:1043-1050. Stenroth, P. and P. Nystrom. 2003. Exotic crayfish in a brown water stream: Effects on juvenile trout, invertebrates and algae. Freshwater Biology, 48(3):466-475. Vodopich, D. and R. Moore. 1999. Biology Laboratory Manual. The McGraw Hill Companies Inc. USA. Wallace, J.B. and J.R. Webster. 1996. The Role of Macroinvertebrates in Stream Ecosystem Function. Annual Review of Entomology, 41:115-139. Yamamoto, Y. 2010. Contribution of bioturbation by the red swamp crayfish Procambarus clarkii to the recruitment of bloom-forming cyanobacteria from sediment. Journal of Limnology, 69(1):102. Bull. Southern California Acad. Sci. 114(1), 2015, pp. 22-32 © Southern California Academy of Sciences, 2015 Soil Organic Carbon and Nitrogen Storage in Two Southern California Salt Marshes: The Role of Pre-Restoration Vegetation Jason K. Keller,* Tyler Anthony, Dustin Clark, Kristin Gabriel, Dewmini Gamalath, Ryan Kabala, Julie King, Ladyssara Medina, and Monica Nguyen Schmid College of Science and Technology, Chapman University, Orange, CA Abstract. — Soil organic carbon and nitrogen storage represent important ecosystem services provided by salt marshes. To test the importance of vegetation on soil properties, we measured organic carbon, total nitrogen, and belowground biomass in two southern California salt marshes. In both marshes, cores were collected from areas which differed in dominant vegetation cover prior to the restoration of tidal influence. There were no differences in organic carbon or total nitrogen density between vegetation classes at either site; however, a relationship between belowground biomass and soil organic carbon suggests that vegetation may influence soil properties. Salt marshes provide a number of important ecosystem services, including habitat for fish and bird species, food web support for adjacent marine environments, nutrient removal from the landscape, and carbon storage in long-lived soil pools (e.g., Zedler and Kercher 2005). Despite their importance, these ecosystems have been lost at alarming rates. Recent estimates suggest that on a global scale, 25% of salt marshes have been lost since the 1800s with ongoing loss rates of an additional 1-2% per year (Mcleod et al. 2011). While comparable estimates of loss rates in southern California are limited, it is likely that salt marsh loss in the region is considerably higher than the global average. Grossinger et al.1 used US Coast Survey T-sheets from the late 1800s to estimate a historical area of 7,711 ha of vegetated intertidal marsh along the South Coast of California (from Point Conception to the Mexico border). Sutula et al.2 estimated that approximately 1681 ha (4,153 acres) of intertidal estuarine wetlands remain in the same region. While a direct comparison between these values should be viewed with caution due to differences in methodologies, the apparent dramatic loss in wetland area highlights the impact of historical anthropogenic activities on Southern California wetlands. More recently, losses of salt marsh habitat in the Pacific region were negligible between 2004-2009 (Dahl and Stedman 2013), suggesting that rates of loss have slowed. Further, ongoing conservation and restoration activities are aimed at maintaining the services provided by the remaining wetlands in the region (Callaway and Zedler 2009). * Corresponding Author: jkeller@chapman.edu Grossinger, R.M., E.D. Stein, K.N. Cayce, R.A. Askevold, S. Dark and A.A. Whipple. 2011. Historical wetlands of the southern California coast: an atlas of US Coast Survey T-sheets, 1851-1889. San Francisco Estuary Institute Contribution #586 and Southern California Coastal Water Research Project Technical Report #589, 55 pp. 2 Sutula, M„ J.N. Collins, A. Wiskind, C. Roberts, C. Solek, S. Pearce, R. Clark, A.E. Fetscher, C. Grosso, K. O’Connor, A. Robinson, C. Clark, K. Rey, S. Mrrissette, A. Eicher, R. Pasquinelli, M. May and K. Ritter. 2008. Status of Perennial Estuarine Wetlands in the State of Califonia. Southern California Coastal Water Research Project, 48 pp. 22 SOIL ORGANIC CARBON AND NITROGEN IN RESTORED SALT MARSHES 23 A great deal of recent attention has focused on capitalizing on ecosystem services provided by salt marshes as a potential means to support ongoing restoration and conservation efforts. In particular, there is a growing interest in quantifying carbon storage in salt marshes (Chmura et al. 2003; Mcleod et al. 2011; Pendleton et al. 2012). Salt marshes, along with other vegetated coastal ecosystems including mangroves and sea grass beds, are particularly effective at storing carbon in their soils because anaerobic conditions generally limit decomposition of primary productivity in these ecosystems (Megonigal et al. 2004; Tobias and Neubauer 2009) while a continuous supply of sulfate limits production of the greenhouse gas methane (Poffenbarger et al. 2011). Further, salt marshes continuously accrete new soils vertically to cope with sea level rise, which allows for new layers of soil carbon to be accumulated through time (Kirwan and Megonigal 2013; Morris et al. 2002). This so-called “blue carbon” could conceptually be traded in emerging carbon markets, although there are a number of ecological, political and economic questions surrounding this possibility (Edwards et al. 2013; Pendleton et al. 2013; Sutton-Grier et al. 2014; Ullman et al. 2013). Concomitant with storing “blue carbon”, salt marsh soils serve as an important sink for nitrogen, and this ecosystem service may also be valuable in the context of restoration and conservation efforts (Lau 2013). We have previously measured soil organic carbon storage in two restored salt marshes in Huntington Beach, California (Keller et al. 2012). This work showed that soil organic carbon was generally higher in a marsh that had been restored for two years than in an adjacent marsh that had been restored for 22 years. This suggests that the assumption that restoration projects share a common starting point and predictably accumulate soil carbon through time needs to be critically evaluated. In particular, we hypothesized that initial site conditions, such as extant vegetation, may be as important as time following restoration when determining soil carbon storage, and perhaps when determining other belowground ecosystem properties. Here, we further explore this possibility by measuring soil carbon and nitrogen storage, as well as belowground biomass, in two additional southern California salt marshes. In the first marsh, which had been restored for three years, we compared belowground properties from areas which differed in vegetation coverage prior to restoration. In the second marsh, which had not yet been restored, we compared areas dominated by dramatically different pre-restoration vegetation communities. Materials and Methods Site Description The Huntington Beach Wetlands used for this project are remnants of a larger marsh that historically existed at the mouth of the Santa Ana River in northern Orange County, California (33° 39’ N, 117° 59’ W). The majority of this marsh area was isolated from tidal exchange by the mid- 1940s due to development and flood control measures, but various wetland restoration efforts, including reconnection to tidal exchange, have been taking place since the 1980s3. To explore the importance of pre-restoration vegetation on belowground carbon and nitrogen dynamics, we collected samples in both the Magnolia and Newland Marshes (Fig. 1). 3 Jones & Stokes Associates, I. 1997. Talbert Marsh restoration project five-year postrestoration monitoring report. Final. December. (JSA 96-300.) Sacramento, CA. Prepared for Huntington Beach Wetlands Conservancy, Huntington Beach, CA. 24 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Efforts to restore the 16.6 ha Magnolia Marsh, including reestablishment of tidal influence as well as the recreation of historical tidal channels, were completed in 2010 (Gordon Smith, Huntington Beach Wetlands Conservancy, personal communication). We utilized Google Earth images from October of 2007 to identify locations which differed in pre-restoration vegetation coverage. Specifically, we collected four soil cores from areas with vegetation cover prior to restoration (“vegetated”) and two soil cores from areas with limited vegetation cover (“un vegetated”; Fig. 1C.). While admittedly qualitative, our designations of vegetation cover are in general agreement with vegetation monitoring efforts at Magnolia Marsh, which show extensive coverage of senescent salt marsh vegetation on the eastern side and limited vegetation on the western side of this site4. Core locations were not selected based on specific vegetation communities, but at the time of collection vegetation was generally similar to other southern California salt marshes, and included: pickleweed ( Salicornia pacified), alkali seaheath ( Frankenia salina), turtleweed ( Batis maritima), and saltgrass ( Distichlis spicata). At the time of our sampling, tidal influence had not yet been restored to the 17.8 ha Newland Marsh, located west of Magnolia Marsh in Huntington Beach. This site currently has two visually distinct vegetation communities; a salt marsh community (“salt marsh”) dominated by plants similar to those found in Magnolia Marsh and a brackish community (“brackish”) dominated by cattail ( Typha sp.) We collected two soil cores from each vegetation community in Newland Marsh (Fig. ID.). Sample Collection and Analysis Soil cores were collected in October-December 2013 following a modification of the protocol described in Keller et al. (2012). Briefly, a 15.3 cm diameter stainless steel tube equipped with a sharpened bottom edge was inserted to an average depth of 41 cm below the soil surface (range 32-48 cm). Care was taken to minimize soil compaction. Upon extraction of the core, soils were sliced into 2 cm depth increments using a serrated knife and returned to the laboratory at Chapman University for processing. Each depth increment was weighed and then passed through a 2 mm sieve within 1 week of collection (when necessary, soils were stored at 4°C until sieving). Material >2mm was subse- quently washed with distilled water over a 1-mm sieve and live roots and rhizomes were collected and dried at 60°C to a constant mass. Four depths from a core collected in the brackish community at Newland Marsh had highly organic soils, which did not pass easily through the 2 mm sieve. Belowground biomass was removed by hand from these depths and the remaining (unsieved) soil was processed as described below. Subsamples of soil that passed through the 2 mm sieve were dried at 60°C to determine percent moisture for each depth increment. Percent moisture values were used to calculate the total dry mass of soil based on the total wet mass collected at each depth. Dried soils were ground to a fine powder using an IKA All Basic Analytical Mill (IKA Works, Inc., Wilmington, NC, USA). Organic carbon and total nitrogen were measured using a Costech elemental analyzer (Costech Analytical Technologies Inc., Valencia, CA, USA). To remove inorganic carbon, soil samples were acidified with 50 pL of 1M HC1 and dried overnight at 37°C twice before carbon and nitrogen analysis (Craft et al. 1991). 4Whitcraft, C., B. Allen and C. Lowe. 2013. Huntington Beach Wetlands Restoration Project Monitoring Program Methodology and Data Summary. Prepared for Huntington Beach Wetlands Conservancy and National Oceanic and Atmospheric Administration - MSRP. SOIL ORGANIC CARBON AND NITROGEN IN RESTORED SALT MARSHES 25 Fig. 1. Location of Huntington Beach Wetlands (A.) and of Magnolia Marsh (outlined in red) and Newland Marsh (outlined in blue) (B.). Soil cores were collected from areas identified as vegetated (n=4) or un vegetated (n=2) prior to restoration in Magnolia Marsh based on imagery from 2007 (image date: 10/22/2007) (C.). Soil cores were collected from areas dominated by brackish (n=2) or salt marsh (n=2) vegetation in Newland Marsh (image date: 4/16/2013) (D.). Image source: Google Earth for all panels. Organic matter content was determined as loss on ignition (LOI) following combustion at 400°C for at least 10 hours. Statistical Analyses Organic carbon and total nitrogen concentrations were multiplied by the total dry mass of soil to calculate the mass of organic carbon and total nitrogen in each depth increment. These values were subsequently summed over the 0-10 and 0^10 cm depth increments and expressed as organic carbon or total nitrogen densities (g cm-3) based on the total volume of these depth ranges (Keller et al. 2012). The 0-10 cm depth increment includes the majority of roots found in these sites while the 0^40 cm depth increment includes the entirety of the soil core. In cases where soils cores did not extend to a depth of 40 cm, the average elemental and mass values from the 3 deepest depth increments were used for all missing depths to 40 cm. In 2 cores from Newland Marsh, this approach was used for the 38-40 cm depth increment. In a single unvegetated core from Magnolia Marsh, averages were used for the 32-40 cm depth increments. A similar approach was used to calculate total belowground biomass (g) in the upper 10 and 40 cm of each soil core. 26 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES % Organic Carbon % Organic Carbon 0 5 10 15 0 10 20 30 40 Fig. 2. Depth profiles of soil organic carbon content (A., B.) and total nitrogen content (C., D.) in cores collected from Magnolia and Newland Marshes. In Magnolia Marsh, cores were collected from areas identified as vegetated (n=4) or un vegetated (n=2) prior to restoration. In Newland Marsh, cores were collected from areas dominated by Brackish (n=2) or Salt Marsh (n=2) vegetation. Independent t-tests were used to compare organic carbon densities, total nitrogen densities and belowground biomass in the 0-10 and 0^10 cm depth increments between vegetated and unvegetated cores in Magnolia Marsh and between brackish and salt marsh cores in Newland Marsh. All data were normally distributed; however, data frequently failed to meet assumptions of equal variance between groups based on the Levene’s Test. In cases with unequal variances, we used the more conservative t-test output that did not make assumptions about equal variance (IBM Corp 2012). Differences were considered significant at /?<0.05 for all t-tests. Regressions were used to explore relationships between LOI, organic carbon and total nitrogen content as well as relationships between soil organic carbon density and belowground biomass. All analyses were completed using Version 21 of the IBM SPSS statistical package (IBM Corp 2012). Results Organic carbon content was highest in surface soils and decreased with depth at both Magnolia Marsh and Newland Marsh (Fig. 2A and B.). Vegetated cores at Magnolia Marsh had higher average organic carbon concentrations than unvegetated cores in the upper 10 cm, but these differences disappeared at deeper depths (Fig. 2A.). Average carbon density to a depth of 10 cm in vegetated cores at Magnolia Marsh was nearly double the carbon density in unvegetated cores; however, there were no significant differences in carbon density between vegetated and unvegetated cores over either the SOIL ORGANIC CARBON AND NITROGEN IN RESTORED SALT MARSHES 27 Table 1. Mean (± 1 SE) soil organic carbon density, total nitrogen density and belowground biomass in soil cores collected from Magnolia and Newland Marshes. All values were summed to a depth of either 10 cm or 40 cm. There were no significant differences between vegetated and unvegetated samples in Magnolia Marsh or brackish and salt marsh samples in Newland Marsh at either depth. Magnolia Marsh Newland Marsh Vegetated (n=4) Unvegetated (n=2) Brackish (n=2) Salt Marsh (n=2) Organic Carbon Density (g cm-3) 0-10 cm 0.023 ± 0.0014 0-40 cm °-013 ± °-0002 0.012 ± 0.0045 0.013 ± 0.0024 0.026 0.014 ± 0.0092 ± 0.0018 0.022 ± 0.0044 0.016 ± 0.0016 Total Nitrogen Density (g cm-3) 0-10 cm 0.0019 ± 0.00013 0^10 cm 0.0012 ± 0.00030 0.0012 ± 0.00003 0.0012 ± 0.00020 0.0018 0.0012 ± 0.00060 ± 0.00005 0.0020 ± 0.00030 0.0014 ± 0.00015 Belowground Biomass (g) 0-10 cm 12.9 + 4.0 0-40 cm 17.2 ± 5.4 4.5 ± 3.4 6.0 ± 2.1 11.8 19.0 ± 11.2 ± 18.1 6.0 ± 3.0 8.3 ± 2.8 0-10 or (M-0 cm depths (Table 1). Cores from the brackish community at Newland Marsh generally had higher average organic carbon content than cores from the salt marsh community, although variability between cores was high, especially in the brackish community (Fig. 2B). There were no significant differences in organic carbon densities between brackish and salt marsh cores in Newland Marsh when calculated over the 0-10 or CMO cm depths (Table 1). Patterns of soil nitrogen through the depth profile mirrored organic carbon concentrations at both Magnolia and Newland Marshes (Fig. 2C and D), reflecting a strong relationship between organic carbon and nitrogen in the soils. There were no significant differences in total nitrogen density between vegetated and unvegetated cores in Magnolia Marsh or between brackish and salt marsh cores in Newland Marsh at either the 0-10 or CMO cm depth increments (Table 1). Similar to organic carbon and total nitrogen, belowground biomass was generally higher in surface soils and decreased with depth (Fig. 3). Average total belowground biomass in both the 0-10 and 0-40 cm depth increments was nearly 3-times higher in the vegetated cores compared to the unvegetated cores in Magnolia Marsh; however, these differences were not statistically significant at either depth increment (Table 1). In Newland Marsh, average total belowground biomass in both the 0-10 and 0^10 cm depth increments was approximately twice as high in brackish cores compared to salt marsh cores, but these differences were not significant at either depth range (Table 1). Total organic carbon density in the 0-10 cm depth increased with increased belowground biomass in the same depth range (/?=0.03; r2=0.48; Fig. 4). There was no relationship between organic carbon density and belowground biomass in the 0^1-0 cm depth increment (/?= 0.63; Fig. 4). Across all sites, organic carbon content increased with increasing concentrations of organic matter (measured as LOI; /><0.001; r2=0.96; Fig. 5A.). Similarly, total nitrogen content was highest in samples with high organic matter content (p<0.001; r2=0.96; Fig. 5B.). Discussion and Conclusions Tidal influence had been restored at Magnolia Marsh for 3 years prior to sampling for this project and had yet to be restored at the nearby Newland Marsh. Despite different restoration histories, the upper 40 cm of soil in both sites stored between 0.01 3-0.0 1 5 g cm-3 28 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Belowground Biomass (g) Belowground Biomass (g) 02468 10 12 14 0 2 4 6 8 Fig. 3. Depth profiles of belowground biomass in Magnolia (A.) and Newland (B.) Marshes. Cores were collected as described in Figure 2. of organic carbon (Table 1). These values are lower than the global average soil organic carbon density of 0.039 ± 0.003 g cm-3 provided by Chmura et al. (2003). Soil organic carbon density measured in the current project was also lower than the values of 0.034 g cm-3 and 0.023 g cm-3 measured in the adjacent Brookhurst Marsh and Talbert Marsh which had been restored for 2 and 22 years, respectively (Keller et al. 2012). Taken together, these results verify our previous assertion that time since restoration does not appear to be the primary control of soil organic carbon content in this salt marsh landscape. This conclusion is in contrast to previous chronosequence studies which have documented increased soil carbon through time following restoration (e.g., Cornell et al. 2007; Craft et al. 2003). However, Streever et al. (2000) suggested that inter-site differences in ecosystem properties may be greater than differences that emerge through time following restoration. We previously hypothesized that site-specific differences in pre-restoration vegetation may play a particularly important role in determining soil carbon density (or other soil conditions) at these sites (Keller et al. 2012). The current project provides limited support for this hypothesis. While there were trends towards higher soil carbon and nitrogen in the vegetated cores in Magnolia Marsh and the brackish cores in Newland Marsh (Fig. 2), these differences were not significant at either site (Table 1). It is worth noting that there was considerable spatial variability in soil properties even within a plant community type within the same marsh (especially in the brackish community in Newland Marsh). The reasons for this variability are unclear, but could include differences in marsh elevation, vegetation community and/or decomposition dynamics which are known to interact to influence carbon content and rates of soil accretion in marsh ecosystems (Kirwan and Megongial 2013). Future work should consider this variability when attempting to account for carbon storage within an entire marsh ecosystem. Across both sites, 48% of the variability in soil organic carbon density in the upper 10 cm was explained by belowground biomass in the same depth interval (Fig. 4), suggesting that vegetation community can perhaps influence soil properties. Root and rhizome dynamics are rarely studied in wetland environments due to logistical constraints (e.g., Iversen et al. 2012), but these belowground processes may be important for understanding soil carbon and nitrogen dynamics. Decreasing belowground biomass with depth has been observed previously (Saunders et al. 2006) and may be driven by both biotic factors SOIL ORGANIC CARBON AND NITROGEN IN RESTORED SALT MARSHES 29 Belowground Biomass (g) Fig. 4. Relationship between soil organic carbon density and belowground biomass in the upper 10 cm (closed symbols) and the upper 40 cm (open symbols) of salt marsh soil cores collected from both Magnolia and Newland Marshes. (i.e., competition between species) and abiotic factors (i.e., flooding and oxygen availability or their interaction). Modeling approaches have explored the links between root productivity and soil carbon content (e.g., Mudd et al. 2009), and Langley et al. (2009) demonstrated that organic matter production in the form of fine roots in response to elevated atmospheric C02 was the primary driver of increased rates of accretion in a brackish marsh. There was a strong relationship between soil organic carbon content and organic matter content (LOI) across all samples analyzed in the current project (Fig. 5). This relationship was similar to those reported by Craft et al. (1991) and Callaway et al. (2012) using salt and brackish marsh soils from North Carolina and San Francisco, California, respectively (Fig. 5), suggesting that this relationship is relatively robust across climate and vegetation types. The quadratic form of this relationship results from an increased fraction of organic carbon in organic matter in soils with higher organic matter contents. For example, organic matter from the 0-2 cm depth increment contained 42 ± 3 (mean ± 1 SE) percent carbon compared to 22 ± 3 percent carbon in organic matter from the 8-10 cm depth increment. These values are all below the 58% of organic matter predicted to be carbon based on the van Bemmelen factor (commonly used to convert organic matter to organic carbon) and are generally below the more recent estimate of 50% carbon suggested by Pribyl (2010). The deviations from these values are particularly pronounced in deeper (older) soils which might suggest that carbon is being lost from organic matter through time, perhaps through microbial respiration or through export of dissolved carbon. 30 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fig. 5. Relationship between organic matter content (measured as loss on ignition, LOI) and organic carbon content (A.) and total nitrogen content (B.) in salt marsh soils collected from both Magnolia and Newland Marshes. Previously published relationships from Craft et al. (1991) and Callaway et al. (2012) are included for comparison. Craft et al. (1991) also reported a relationship between total soil nitrogen content and organic matter content (LOI), suggesting that relatively simple measurements of LOI might provide indirect information on soil carbon and nitrogen. We also observed a strong relationship between soil nitrogen content and soil organic matter content (Fig. 5); however, our soils had a higher percent of soil nitrogen for a given organic SOIL ORGANIC CARBON AND NITROGEN IN RESTORED SALT MARSHES 31 matter content (i.e., lower C:N) than those analyzed by Craft et al. (1991). Thus, while the relationship between organic carbon and organic matter appears to be robust across climates and vegetation types, the relationship between nitrogen and soil organic matter may be much more site-dependent and generalized relationships should be viewed with caution. A lack of a consistent accumulation of soil organic carbon along a chronosequence of southern California salt marshes (from pre-restored to 22 years post-restoration) suggests that site-specific factors may be as important as time since restoration in controlling the “blue carbon” accumulation in these systems. Pre-restoration vegetation, as either the presence or absence of vegetation in Magnolia Marsh or as different vegetation communities in Newland Marsh, also did not play the key role in determining soil organic carbon (or total nitrogen) content in these marshes. However, a strong relationship between belowground biomass and soil organic carbon means that vegetation does likely play some part in determining soil properties. Acknowledgements We thank the School of Earth and Environmental Sciences within the Schmid College of Science and Technology at Chapman University for funding this project as the laboratory component of the Fall 2013 Ecosystems Ecology course. Angelina Delgado, Justin Drzymkowski, Kaitlin Fuller, Nicolas Lapointe, Jacob Lopez, Daniel Moore, Cassandra Oregel, Steven Pham, Jesse Simons, and Ryan Ugale provided valuable field and laboratory assistance. Amber Garcia and Jocelyn Paez from Orange High School were instrumental in completing the carbon and nitrogen analyses of these soils. The Board of the Huntington Beach Wetlands Conservancy under the leadership of Dr. Gordon Smith graciously provided access to these sites. Literature Cited Callaway, J.C., E.L. Borgnis, R.E. Turner, and C.S. Milan. 2012. Carbon sequestration and sediment accretion in San Francisco Bay tidal wetlands. Estuaries and Coasts, 35:1163-1181. — and J.B. Zedler. 2009. Conserving the diverse marshes of the Pacific Coast. Pp. 285-307 in Human Impacts on Salt Marshes: A Global Perspective (B. R. Silliman, E. D. Grosholz and M. D. Bertness, eds.) University of California Press. Chmura, G.L., S.C. Anisfel, D.R. Cahoon, and J.C. Lynch. 2003. Global carbon sequestration in tidal, saline wetland soils. Global Biogeochemical Cycles, 17. doi:10.1029/2002GB001917. Cornell, J.A., C. Craft, and P. Megonigal. 2007. Ecosystem gas exchange across a created salt marsh chronosequence. Wetlands, 27:240-250. Craft, C., P. Megonigal, S. Broome, J. Stevenson, R. Freese, J. Cornell, L. Zheng, and J. Sacco. 2003. The pace of ecosystem development of constructed Spartina alterniflora marshes. Ecological Applications, 13:1417-1432. Craft, C.B., E.D. Seneca, and S.W. Broome. 1991. Loss on ignition and Kjeldahl digestion for estimating organic carbon and total nitrogen in estuarine marsh soils: calibration with dry combustion. Estuaries, 2:175-179. Dahl, T.E. and S.M. Stedman. 2013. Status and trends of wetlands in the coastal watersheds of the Conterminous United States 2004 to 2009. U.S. Department of the Interior, Fish and Wildlife Service and National Oceanic and Atmospheric Administration, National Marine Fisheries Service, 46 pp. Edwards, P.E.T., A.E. Sutton-Grier, and G.E. Coyle. 2013. Investing in nature: restoring coastal habitat blue infrastructure and green job creation. Marine Policy, 38:65-71. IBM Corp. 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp. Iversen, C.M., M.T. Murphy, M.F. Allen, J. Childs, D.M. Eissenstat, E.A. Lilleskov, T.M. Sarjala, V.L. Sloan, and P.F. Sullivan. 2012. Advancing the use of minirhizotrons in wetlands. Plant and Soil, 352:23-39. 32 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Keller, J.K., K.K. Takagi, M.E. Brown, K.N. Stump, C.G. Takahashi, W. Joo, K.L. Au, C.C. Calhoun, R.K. Chundu, K. Hokutan, J.M. Mosolf, and K. Roy. 2012. Soil organic carbon storage in restored salt marshes in Huntington Beach, California. Bulletin of the Southern California Academy of Sciences, 111:153-161. Kirwan, M.L. and J.P. Megonigal. 2013. Tidal wetland stability in the face of human impacts and sea-level rise. Nature, 504:53-60. Langley, J.A., K.L. McKee, D.R. Cahoon, J.A. Cherry, and J.P. Megonigal. 2009. Elevated C02 stimulates marsh elevation gain, counterbalancing sea-level rise. Proceedings of the National Academy of Sciences, USA, 106:6182-6186. Lau, W.W.Y. 2013. Beyond carbon: conceptualizing payments for ecosystem services in blue forests on carbon and other marine and coastal ecosystem services. Ocean & Coastal Management, 83:5-14. Mcleod, E., G.L. Chumra, S. Bouillon, R. Salm, M. Bjork, C.M. Duarte, C.E. Lovelock, W.H. Schlesinger, and B.R. Silliman. 2011. A blueprint for blue carbon: toward an improved understanding of the role of vegetated coastal habitats in sequestering C02. Frontiers in Ecology and the Environment, 9:552-560. Megonigal, J.P., M.E. Hines, and P.T. Visscher. 2004. Anaerobic metabolism: linkages to trace gases and aerobic processes. Pp. 3 1 7-^424 in Biogeochemistry (W. H. Schlesinger, ed, ) Elsevier-Pergamon. Morris, J.T., P.V. Sundareshwar, C.T. Nietch, B. Kjerfve, and D.R. Cahoon. 2002. Responses of coastal wetlands to rising sea level. Ecology, 83:2869-2877. Mudd, S.M., S.M. Howell, and J.T. Morris. 2009. Impact of dynamic feedbacks between sedimentation, sea-level rise, and biomass production on near-surface marsh stratigraphy and carbon accumulation. Estuarine, Coastal and Shelf Science, 82:377-389. Pendleton, L., D.C. Donato, B.C. Murray, S. Crooks, W.A. Jenkins, S. Sifleet, C. Craft, J.W. Fourqurean, J.B. Kauffman, N. Marba, P. Megonigal, E. Pidgeon, D. Herr, D. Gordon, and A. Baldera. 2012. Estimating global "blue carbon" emissions from conservation and degradation of vegetated coastal ecosystems. PLOS ONE, 7:e43542. Pendleton, L.H., A.E. Sutton-Grier, D.R. Gordon, B.C. Murray, B.E. Victor, R.B. Griffis, J.A.V. Lechuga, and C. Giri. 2013. Considering "coastal carbon" in existing U.S. federal statutes and policies. Coastal Management, 41:439^-56. Poffenbarger, H.J., B.A. Needelman, and J.P. Megonigal. 2011. Salinity influence on methane emissions from tidal marshes. Wetlands, 31:831-842. Pribyl, D.W. 2010. A critical review of the conventional SOC to SOM conversion factor. Geoderma, 156: 75-83. Saunders, C.J., J.P. Megonigal, and J.F. Reynolds. 2006. Comparison of belowground biomass in C3- and C4-dominated mixed communities in a Chesapeake Bay brackish marsh. Plant and Soil, 280: 305-322. Streever, W.J. 2000. Spartina alterniflora marshes on dredged material: a critical review of the ongoing debate over success. Wetlands Ecology and Management, 8:295-316. Sutton-Grier, A.E., A.K. Moore, P.C. Wiley, and P.E.T. Edwards. 2014. Incorporating ecosystem services into the implementation of existing U.S. natural resource mangement regulations: operationalizing carbon sequestration and storage. Marine Policy, 43:246-253. Tobias, C. and S.C. Neubauer. 2009. Salt marsh biogeochemistry - an overview. Pp. 445^192 in Coastal Wetalnds: An Integrated Approach (G. Perillo, E. Wolanski, D. Cahoon and M. Brinson, eds.) Elsevier. Ullman, R., V. Vilbao-Bastida, and G. Grimsditch. 2013. Including Blue Carbon in climate market mechanisms. Ocean & Coastal Management, 83:15-18. Zedler, J.B. and S. Kercher. 2005. Wetland resources: Status, trends, ecosystem services, and restorability. Annual Review of Environment and Resources, 30:39-74. Bull. Southern California Acad. Sci. 114(1), 2015, pp. 33—41 © Southern California Academy of Sciences, 2015 Identical Response of Caged Rock Crabs (Genera Metacarcinus and Cancer) to Energized and Unenergized Undersea Power Cables in Southern California, USA Milton S. Love,* 1* Mary M. Nishimoto,1 Scott Clark,1 and Ann Scarborough Bull2 1 Marine Science Institute, University of California, Santa Barbara, CA 93106 2 Bureau of Offshore Energy Management, 770 Paseo Camarillo, Camarillo, CA 93010 Increasingly, energy generation facilities (i.e., wave and wind) are being sited in offshore marine waters. The electricity generated from these facilities is transmitted to shore through cables carrying alternating (AC) or direct (DC) current. If DC is used, it is converted to AC for the North American grid at onshore stations. While these currents produce both electric and magnetic fields, only the magnetic field, here called an electromagnetic field (EMF), is emitted from the cable. Some marine vertebrates and invertebrates can detect EMFs (summarized in Normandeau et al. 201 11). However, while it is clear that organisms can detect EMFs, less well understood is how these animals respond behaviorally to this stimulus, and concerns have been raised regarding how these organisms might interact with energized subsea cables1. Among fishes, a few field or quasi-field studies have produced what appear to be minor or equivocal responses. For instance, in a study of three species of elasmobranchs held in offshore mesocosms and subjected to EMF, there were some statistically significant differences in behavior; however these differences were inconsistent among individuals within a species2. In other studies, migrating European eels (Anguilla anguilla ) in the Baltic Sea slowed, but did not halt, their swimming speed around an energized cable (Westerberg and Lagenfelt, 2008), and the movement of a number of fish species did not appear to be affected by an energized cable off Denmark3. Along the Pacific Coast of the United States, fishers have also raised this issue4; one of the specific issues is how crabs (which form major fisheries along the Pacific Coast) might respond to energized power cables. There have been few studies on the behavioral changes that invertebrates might show in the presence of EMF although a small laboratory study implied that Dungeness crabs ( Metacarcinus magister) were attracted to * Corresponding author: love@lifesci.ucsb.edu 1 Normandeau, Exponent, and T. Tricas, and A. Gill. 2011. Effects of EMFs from undersea power cables on elasmobranchs and other marine species. U.S. Dept. Int., Bur. Ocean Energy, Management, Regulation, and Enforcement, Pacific OCS Region, Camarillo, CA. OCS Study BOEMRE 2011-09. 2 Gill, A.B., Y. Huang, I. Gloyne-Philips, J. Metcalfe, V. Quayle, J. Spencer, and V. Wearmouth. 2009. COWRIE 2.0 Electromagnetic Fields (EMF) Phase 2. EMF-sensitive fish response to EM emissions from sub-sea cables of the type used by the offshore renewable energy industry. COWRIE Ltd. COWRIE- EMF-1-06. 3 DONG Energy and Vattenfall A/S. 2006. Review Report 2005 The Danish offshore wind farm demonstration project: Horns Rev and Nysted offshore wind farms environmental impact assessment and monitoring. The Danish Offshore Wind Farm Demonstration Projects. 4 Pacific Fisheries Management Council (PFMC). 2010. Letter from PFMC to Federal Energy Regulatory Council, dated 19 June 2010. Titled COMMENT Reedsport OPT wave Park Project, FERC No. 12713. Accessed 11 December 2013. http://www.pcouncil.org/wp-content/uploads/Cmt_Reedsport_ OPT_FERC.pdf 33 34 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES a zone of high EMF and that crabs in some zones with elevated EMF levels were somewhat more active than control animals5. Needed are studies that address how organisms respond to an in situ energized submarine power cable. The presence of energized and unenergized AC submarine cables in close proximity to one another off the coast of southern California allowed us to conduct such an experiment on crabs. The experiments took place off Las Flores Canyon (34°28’N, 120 02’W), southern California, USA. Here several energized and unenergized submarine power cables, identical in construction, lie unburied on the seafloor and extend to offshore oil and gas platforms (Fig. 1). We selected two cables for this study; one was energized and the other unenergized. The two cables run parallel to each other, perpendicular to shore, and are approximately 7 m apart. Note that in an ongoing study we have determined that the EMF around the energized cable dissipates to background levels at a distance of about one meter. We used stiff plastic perforated boxes (88 cm x 57 cm x 23 cm) that were secured to the sea floor with sand anchors at a bottom depth of 10 m. Each box was placed so that one end was in contact with one of the two cables. In all, twelve boxes were installed, six adjacent to the energized cable and six adjacent to the unenergized one. The boxes were installed at intervals of 2.5 meters along each cable, half on the east side and half on the west side and these alternated from one side to the other (Fig. 1). To reduce the chances of crabs visually sensing the cable, plastic panels were attached to the end of each box closest to the cable and identical panels were attached to the boxes on the end farthest from the cable. To further reduce the chances that the crabs could sense a difference between the cable end and the noncable end, we also removed the common brown macroalgae Pterygophora californica that occurs on the cables but does not live on the adjacent sea floor. With the boxes in place along the energized and unenergized cables, divers stocked each with one adult crab of either Metacarcinus anthonyi or Cancer productus, for an experimental trial. Each crab, which was randomly selected from a stock of legal-sized crabs provided by a commercial crab fisherman, was dropped through a hinged hatch, which was centered in the middle of the cage. One hour after emplacement, divers recorded the position of the crab within the box by visually dividing it into two halves, the portion closest to the cable being designated “near-half’ and that furthest from the cable “far-half’ (Fig. 1). A second diver then opened the box to record EMF values (in microteslas - pT) with a handheld EMF detector (EMF 1390 from General Tools & Instruments). Readings were taken on the floor of each box at the edge closest to the cable and on the floor of that box furthest from the cable. The boxes were then leaving the crab in the box. Divers returned 24 hours later to observe where the crabs were positioned in the boxes and recorded EMF values. The crabs were then removed from the boxes and new, previously untested, crabs inserted for the next trial. Four sequential, 24-hour trials comprised an experiment. A total of four experiments were conducted in 2013 (10-14 June, 9-13 September, 30 September^- October, and 7-1 1 October). Crabs were selected randomly for each box. Gender was recorded for each crab with exception of the first experiment. The primary question we addressed in this study is whether crabs responded differently to the two types (energized and unenergized) of cables. The observations made 1 hour 5 Wilson, C.S. and D.L. Woodruff. 2011. A preliminary study on the effects of electromagnetic fields on the burial behavior and location of the Dungeness crab, Cancer magister. Pacific Northwest National Laboratory, Prepared for the U.S. Dept. Energy, Contract DE-AC05-76RL01830, PNNL-20729. CRABS AND UNDERSEA POWER CABLES 35 Fig. 1. The location of the energized and unenergized cables used in the experiments and the orientation of six of 12 boxes. The distance between the cables, about 7 m, is not drawn to scale. 36 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES and 24 hours after the crabs were set in the cages were evaluated separately. We used the generalized linear model (GLM) approach to determine if crabs along the energized cable were found nearer or farther from the cable compared to crabs along a non-energized cable. A crab’s position, in the half of the box near or far from the cable, was the response variable. Given the binomial distribution of the response variable, a logistic regression model was used with a logit link function. We used JMP software to fit each GLM to the data by Firth bias-adjusted maximum likelihood estimation of the parameter vectors6. The most complete GLM model analyzed included the effects of experiment (1-4), trial (1^1) nested within experiment, side of cable that the cage was set (west, east), and type of cable (energized and unenergized) as well as the intercept. A likelihood-ratio Chi-square test evaluated the hypothesis that all the model parameters were zero. We also examined a sequence of simpler GLM models to identify the best-fit model that might include as few as one predictor. Akaike’s information criterion (AIC) was used to select between candidate models. To determine if the genders responded differently to the energized and non-energized cables, we first added gender as a predictor in the complete GLM model using data from all but the first experiment when gender was not recorded. We used the same method above to determine the best-fit model. We also parsed the data by gender to determine if either male or female crabs, separately, responded differently to the two types of cables. Again, we used the same GLM approach described above to determine if cable type alone or with the other explanatory factors had a significant effect on a male or female crab’s position in a box. The EMF at the end of the boxes closest to the energized cable ranged from a mean of 46.2 pT to 80.0 pT during the experiments, and the readings on the far end of the boxes never exceeded 0.9 pT (Table 1). Along the unenergized cable, EMF did not exceed 0.2 pT in the near half or far half of the boxes during the experiments. A total of 192 crabs were used in this study; 24 crabs in each of four experiments on each cable (Table 2). The positions of all 192 crabs were observed 1 hour after emplacement. A total of eight crabs were recorded as lost 24 hours after emplacement during the four experiments; three crabs in boxes along the unenergized cable and five crabs along the energized cable. Escapement was not possible and loss of crabs was likely due to predation by octopuses. The crabs responded no differently in the boxes along the unenergized and energized cables. Both 1-hour and 24-hours after the crabs were set in the boxes, there were no apparent differences in the proportion of crabs near the two types of cable regardless of the side of cable where the boxes were set (Fig. 2). For a given observation period, experiment, trial nested within experiment, side of cable that the cage was set, and type of cable had no significant effect on the position of crabs in the boxes as evident from the GLM that was not significantly different from the intercept model (1 hour: n=192, -log likelihood =5.676, X2= 11.351, DF= 17, p=0.838, AIC=295.901. 24 hours: n=184, -log likelihood =7.946, X2= 15.892, DF = 17, p=0.532, AIC=281 .037). None of the GLMs that incorporated fewer explanatory factors could predict with statistical significance the variability in crab responses in the boxes next to the cables one hour or 24 hours after deployment. The proportion of crabs near the two types of cables 24 hours after deployment was highly variable across experiments regardless of side of the cable the box was set (Fig. 2). 6 Schwarz, C.J. 2013. Sampling, regression, experimental design and analysis for environmental scientists, biologists, and resource managers. http://people.stat.sfu.ca/cschwarz/Stat-650/Notes/ PDFbigbook-JMP/. CRABS AND UNDERSEA POWER CABLES 37 Table 1. Level of electromagnetic field (microteslas - pT) in those parts of boxes closest to unenergized and energized cables as read one hour and 24 hours after crabs were inserted. EMF readings at the farthest end of the boxes were <0.1(iT at the unenergized cable and <0.9 fiT at the energized cable. The lower n in experiments 1 and 4 were due to the flooding of the housing containing the EMF meter after the first day of observations, which led to failure of the devices. However, note that the energized cable used in this experiments has been in continuous use for many years and did not fail during the course of these studies. Experiment Cable Type 1 hr 24 hr X sd n X sd n 1 Unenergized 0.0 0.0 6 - - 0 Energized 46.2 11.4 6 - - - 2 Unenergized 0.0 0.0 24 0.1 0.0 24 Energized 57.0 7.4 24 55.5 8.7 24 3 Unenergized 0.0 0.0 24 0.1 0.0 24 Energized 54.2 9.3 24 56.1 0.0 24 4 Unenergized 0.1 0.0 6 0.1 0.1 6 Energized 80.0 19.7 6 51.0 10.1 6 Combining the observations from the four experiments, the proportion of crabs found close to the two types of cable changed little between the observations made one hour and 24 hours after the crabs were set in the boxes (Fig. 3). One hour after emplacement, 53% (51 of 96) of the crabs set along the unenergized cable and 55% (53 of 96) of the crabs along the energized cable were observed in the near-half of the boxes (Fig. 3). The log- likelihood test of the GLM showed no cable-type effect on crab response (n=192, -log likelihood=0.042, X2=0.084, DF=1, p=0.772, AIC=270.876). The AIC for this single- factor model indicates that it is no worse fit of the 1-hour data than the GLM of all explanatory factors. In comparison, 24 hours after emplacement 56% (52 of 93) of the crabs set along the unenergized cable and 51% (46 of 91 of the crabs set along the energized cable were in the near-half of the boxes (Fig. 3). Although a slightly greater proportion of crabs were nearer the unenergized cable than the energized cable, Table 2. Number and gender (F = female, M = males, Unk = unknown) of crabs used in four experiments. Gender of crabs in experiment 1 was not determined. Loss of crabs between one hour and 24 hours was likely due to predation by octopuses. Unenergized Energized F M Unk Total F M Unk Total Grand Total Experiment 1 1 hr 24 24 24 24 48 24 hrs 23 23 24 24 47 Experiment 2 1 hr 17 7 24 17 7 24 48 24 hrs 17 7 24 17 7 24 48 Experiment 3 1 hr 17 7 24 22 2 24 48 24 hrs 15 7 22 19 2 21 43 Experiment 4 1 hr 18 6 24 17 7 24 48 24 hrs 18 6 24 16 6 22 46 38 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 14 Unerwgized cable, 1 hr j§jj£ fwar-lmlf ■ br-hsT ^ F? 11111b 14-, 12- 10- Unenergized cable, 24 hrs ggiw«Mur ■ far-fta* m i g EM mm Eai£ Energized cable, 1 hr 14-i ®5 n©ar4wiir 14-, Energized cable, 24 hrs S8t near -ha r W#f* East SicJeorcafeie Fig. 2. The number of crabs positioned in the near-half and far-half of boxes on the west side and east side of the energized and unenergized cables by experiment, one hour and 24 hours after deployment. cable type in the single factor model had no effect on crab response (n=184, -log likelihood =0.266, X2=0.5318, DF=1, p=0.466, AIC=259.897). As was the case for the 1-hour observations, the proportion of crabs found close to the two types of cable did not differ 24 hours after the crabs were set in the boxes. Some of the crabs were found in the opposite half of the box when reexamined 24 hours later. Along the energized cable, 23.1% (21 individuals) of 91 crabs moved to the half of the box that was closer to the cable from the half farther, and 27.5% (25 individuals) moved to the half of the box farther from the half nearer. Along the non-energized cable, 21.5% (20 of 93 individuals) moved to half of the box that was closer to the cable, and 18.3% (17) moved to the farther half of the box. Movement of crabs within the boxes between the one-hour and 24-hour observations is unknown. The addition of gender to the complete GLM faired no better using data from experiments 2-4 when gender was recorded (1 hour: n=144, -log likelihood =6.632, V= 13.265, DF= 14, p=0.506, AIC=221.950. 24 hours: n=137. -log likelihood =7.136, CRABS AND UNDERSEA POWER CABLES 39 Crab position, genders combined, experiments combined, after one hour and 24 hours near-half Cable type Fig. 3. The number of crabs positioned in the unenergized cables after one and 24 hours. Cable type •-half and far-half of boxes adjacent to energized and X2= 14.272, DF=14, p=0.430, AIC=21 1.300). None of the GLMs that incorporated fewer explanatory factors could predict with statistical significance the variability in crab responses in the boxes next to the cables one hour or 24 hours after deployment. Specifically, cable type had no effect on a crab’s position in the boxes regardless of gender (Fig. 4). One hour after emplacement, 54% of the females next to the unenergized cable (26 of 52 crabs) and 50% of the females next to the energized cable (28 of 56) were found in the near half of boxes (n=108, -log likelihood =0.080, A2 =0.160, DF=1, p=0.689, AIC= 155.643). Twenty-four hours later, a slightly higher proportion of crabs were found next to both types of cables, 58% of the females (29 of 50 crabs) next to the unenergized cable were found in the near-half of boxes, whereas 52% of the females set along the energized cable (27 of 52 crabs) were in the near-half. Again, the females responded no differently to the two cable types (n=102, -log likelihood =0.190, A2=0.380, DF=1, p=0.538, AIC= 146.285). Males also responded no differently to the two cable types. One hour after emplacement, 65% of the males next to the unenergized cable (13 of 20 crabs) and 50% of the males next to the energized cable (8 of 16) were found in the near half of the boxes (Fig. 4). Although it appears that a greater proportion of males were found nearer the unenergized cable than energized cable, cable type in the single factor GLM had no statistically significant effect on male crab response (n=36, -log likelihood =0.410, X2=0.820, DF=1, p=0.365, AIC=54.8330). Twenty-four hours 40 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Crab position, by gender, summed across experiments, after one hour and 24 hours b) 24 hours 30- b) 24 hours 20- Cabf® type Fig. 4. The number of female and male crabs positioned in the near-half and far-half of boxes adjacent to energized and unenergized cables, one hour and 24 hours after eployment. later, 50% of the males next to the unenergized cable (10 of 20 crabs) and 53% of the males next to the energized cable (8 of 15) were found in the near half of the boxes (n=35, -log likelihood =0.019, Y=0.038, DF=1, p=0.846, AIC=55.228). Pacific Coast crab fishers have voiced several concerns regarding crabs and their potential responses to the EMF generated by submarine power cables. These concerns generally relate to whether crabs are either attracted to, or repulsed by, EMF. If either of these occurs, crab migrations might be compromised and, more specifically, crabs might not walk over a cable to reach a baited trap. While this experiment does not address all of CRABS AND UNDERSEA POWER CABLES 41 these concerns, it does imply that these two crab species may not respond either positively or negatively to the levels of EMF generated by this specific energized cable. Literature Cited SAS Institute Inc. 2013. JMP Pro 11.0.0. Westerberg, H. and I. Lagenfelt. 2008. Sub-sea power cables and the migration behaviour of the European eel. Fish. Manage. Ecol., 15:369-375. Bull. Southern California Acad. Sci. 114(1), 2015, pp. 42-53 © Southern California Academy of Sciences, 2015 Recent Decline of Lowland Populations of the Western Gray Squirrel in the Los Angeles Area of Southern California Daniel S. Cooper1 and Alan E. Muchlinski2 1 Cooper Ecological Monitoring, Inc., 255 Satinwood Ave., Oak Park, CA 91377 2Department of Biological Sciences, California State University, Los Angeles, 5151 State University Drive, Los Angeles, CA 90032 Abstract. — We provide an overview of the distribution of lowland and otherwise isolated populations of the western gray squirrel ( Sciurus griseus) in the Los Angeles area of southern California, an area that has experienced a recent and ongoing invasion by the non-native eastern fox squirrel (, Sciurus niger ), an urban-adapted species introduced a century ago. Away from its strongholds in the western Santa Monica Mountains, San Gabriel Mountains, and Santa Ana Mountains, the western gray squirrel is resident locally in both the Santa Susana and the Verdugo Mountains, in Griffith Park, in low hills at the eastern periphery of the San Gabriel Valley and in Claremont, and along the Santa Ana River canyon near Yorba Linda. It also persists east of the Los Angeles area in residential areas of Redlands and Yucaipa, which as of 2014 are still outside the range of the eastern fox squirrel. Here we document several gray squirrel extirpation events within its lowland range, and discuss factors influencing its persistence and its extirpation. The western gray squirrel ( Sciurus griseus) is a large tree squirrel native to forests of the western United States and extreme northwestern Mexico, with the subspecies S. g. anthonyi common and widespread in oak- and pine-dominated areas of the hills and mountains of southern California (Wilson and Reeder 2005). In the Los Angeles area, a region we define as extending from eastern Ventura County east through Claremont and south through the coastal plain into Orange County to the base of the San Joaquin Hills, it also occurs in human-modified habitats, including large city parks and golf courses, where scattered trees, particularly conifers, provide year-round food and shelter. It is one of two tree squirrels in the Los Angeles area, the other being the eastern fox squirrel {Sciurus niger), a non-native introduced in the early 1900s, and now abundant throughout much of the Los Angeles area of southern California (Jameson and Peeters 1988, King et al. 2010). As discussed by Linders and Stinson (2007) western gray squirrels are closely tied to oak and evergreen woodland, and serve two main roles in maintaining native woodlands: they harvest and bury acorns throughout the woodland, and disperse the seeds and fruit of various oak woodland component tree and shrub species, such as California bay {Umbellaria calif ornica). They also forage heavily on truffle-like mycorrhizal fungi found in leaf litter and loose soil, which aid oaks in fixing nitrogen and retaining water through dry months. During foraging, western gray squirrels deposit the spores of these fungi through their droppings, thus spreading them throughout the oak woodland and promoting the health of its trees. Because of this close association with oaks, the presence Corresponding author: dan@cooperecological.com 42 WESTERN GRAY SQUIRREL IN LOS ANGELES AREA 43 of western gray squirrels may serve as an indicator of oak woodland health. By contrast, the eastern fox squirrel is highly generalist in its food sources, requires a much smaller home range (becoming super-abundant in urban settings), and occurs in a much wider array of habitats than S. griseus away from the major mountain ranges in the region (Gatza 2011, Ortiz 2014). The history and origin of western gray squirrel populations on the floor of the Los Angeles Basin are poorly understood. Today, most lowland populations of S. griseus are strongly associated with planted pines and other conifers, which may now be crucial habitat elements for the species. It was presumably naturally present at lower elevations when oak woodland (mostly Quercus agrifolia ) once covered large areas of now- urbanized places like the San Gabriel Valley, a pattern shared by numerous lower montane plant and wildlife species that are able to persist locally at lower elevations in suitable areas of canyons and woodlands (Cooper 2011). Later, as the region developed, populations of S. griseus may have retreated to large urban parks and more wooded residential areas, where it persisted through most of the 1900s, including those at the base of the San Gabriel Mountains foothills from Pasadena east into Claremont (an area referred to as the “mesa” by early naturalists, e.g., Grinnell 1898). It is also possible that they colonized these areas later by moving down from the surrounding foothills, or that both patterns occurred, with isolated lowland populations “winking” out periodically, replenished by animals from surrounding highlands. Whatever the history, in the years between the late 1990s and the mid-2000s, S. griseus became scarce or altogether absent within many of these same neighborhoods. Clear instances of its extirpation and subsequent replacement - directly or indirectly - by the non-native eastern fox squirrel are now well documented (e.g., Muchlinski et al. 2009, Guthrie 2009, King et al. 2010). Sciurus niger became established in the neighborhoods surrounding the eastern Santa Monica Mountains in the western Los Angeles Basin during the decades following its introduction in 1904, it only arrived in the San Gabriel Valley around 1990, the east San Gabriel Valley around 1998, and the Claremont area and Orange County in the early 2000s (Guthrie 2009, King et al. 2010). In recent years, S. niger has also colonized much of urbanized Santa Barbara County (P. Collins, pers. comm.) and portions of San Diego County, the latter also following an early introduction (King et al. 2010). Now virtually ubiquitous throughout the Los Angeles area from the San Fernando Valley east to San Bernardino County and south through Orange County, S. niger appears to still be absent at several urban-edge locations at the margins of the Los Angeles area, including parks and neighborhoods in Redlands and Yucaipa, San Bernardino Co. (Ortiz 2014); canyons in the lower San Gabriel Mountain foothills from the Sunland-Tujunga area east through Claremont (Gatza 2011), and along the Santa Ana River at Gypsum Canyon, near Yorba Linda, Orange Co. (AEM, unpubl. data). Only a handful of local naturalists have noticed this turnover, and few published data exist on the range of S. griseus in the Los Angeles area prior to the arrival of S. niger. Today, only a few populations of S. griseus remain away from the larger mountains [typically below around 457 m (1500’) a.s.l.], with only a handful, at the far eastern periphery being free of S. niger. To ensure the ecological integrity of these remaining populations of S. griseus - and of their habitat patches - particularly in areas where S. niger has not yet invaded (or at least where it is not completely dominant), it is important that remaining populations of S. griseus be identified and recognized by conservation agencies and organizations. Since the late 1990s, we (DSC and AEM) have been making notes on the occurrence of S. griseus in the Los Angeles area, as described 44 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Distribution of western gray squirrel ( Sciurus griseus) JpF] Core WGS populations # Persisting Irregular/Unknown O Extirpated Freeways/major highways /s/ Counties 305 m a.s.l.). South of here, the low range of hills in the eastern San Gabriel Valley referred to as the San Jose Hills apparently serves as an ecological connection between the San Gabriel Mountains and the Puente-Chino Hills, which then connect to the much larger Santa Ana Mountains to the south (see Cooper 2000). Here, the species persists only at Bonelli Park (San Dimas) and in the “Industry Hills” near La Puente, and several extirpations have been very recent (e.g., observed by DSC at Walnut Creek Park in Covina in 201 1 but not sine e;fide AEM). Puente-Chino Hills Western gray squirrels occurred in multiple canyons and open space areas from Diamond Bar and Rowland Heights west into Whittier and La Habra Heights, and south into Brea, and Chino Hills State Park during the late 1990s (DSC, unpubl. data). A population in Turnbull Canyon in the Whittier Hills (far western Puente Hills) was apparently extirpated in the late 1960s following a major fire that burned many mature oaks (J. Schmidt, in litt.), indicating that even by then some loss had occurred. By the late 2000s they had been extirpated west of Harbor Blvd., with replacement by S. niger, WESTERN GRAY SQUIRREL IN LOS ANGELES AREA 49 including along Powder Canyon in Rowland Heights/La Habra Heights, where S. griseus was present in late 2005 (1, R. Erickson, unpubl. data) yet absent by 2007 (DSC, unpubl. data; fide L. Longacre). The latter location is particularly notable, as the canyon is directly contiguous to hundreds of acres of natural habitat, has been protected as part of the Puente Hills Landfill Conservation Authority, and has seen little if any land use change in the past 20 years. A devastating fire in 2008 that burned most of Chino Hills State Park resulted in the immediate loss of most western gray squirrel populations there, with only a very small number of individuals persisting in oak woodland in the remote center of the park, north of San Juan Hill (A. Ing, pers. comm.). Pomona Valley! Claremont While still present at Rancho Santa Ana Botanic Gardens and along the San Gabriel foothills through the northern portion of Claremont (e.g., San Dimas Canyon, Marshall Canyon, fide AEM), western gray squirrel has been recently extirpated from several areas, and replaced by S. niger, within the city of Claremont to the south, including the Claremont Colleges area (Guthrie 2009). There are apparently no historical or recent records of S. griseus from the eastern Pomona Valley nor along the lower Santa Ana River Valley upstream of Prado Dam. Redlands! Yucaipa (San Bernardino County) Western gray squirrels are widespread and conspicuous residents of the San Bernardino Mountains. However, lowland populations away from the lower foothills persist (as of 2014) at University of Redlands, Sylvan Park, Ford Park, and Prospect Park (Ortiz 2014). The species has also been reported in the “Sunset Hills” area of Redlands just south of Interstate 10 and in an apparently small area of Yucaipa (including Third St.), where they are found in mature pines in a residential area (CSULA web survey). These populations do not appear to be in contact with S. niger as of 2014, and are much higher in elevation than other lowland sites discussed. However, because they are persisting away from the main mountain ranges in what is still obviously lowland (non-montane or foothill) habitat, we have included them here. South Orange County In contrast to the report by Pequegnat (1951) that the western gray squirrel was not found in the Santa Ana Mountains, the species is present in several oak-filled canyons in the Santa Ana Mountains (e.g., Trabuco Canyon, CSULA web survey and J. Ortiz, via email; Modjeska Canyon/Tucker Wildlife Sanctuary, CSULA web survey; Whiting Ranch Wilderness Park, R. Hamilton, via email). Additional reports to the CSULA web survey locate western gray squirrels at the suburban-wildlands interface west of Lake Elsinore. Whether they are recent (post- 1950s) arrivals to this range is not known. Away from the Santa Ana Mountains, two small populations are known from Oak Canyon Nature Center in the Anaheim Hills, and along the “Santa Ana River canyon” where the Chino Hills meet the northern Santa Ana Mountains (AEM, unpubl. data; B. Leatherman, via email). We know of no records from the San Joaquin Hills, where S. niger has been present in residential areas since around 2010 (R. Erickson, via email). Like much of San Bernardino (and Riverside) County, S. niger has only recently (late 1990s) penetrated Orange Co., but it is now widespread and common into Irvine (D. Willick, via email). 50 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Other Areas DSC (unpubl. data) observed a small number of what appeared to be western gray squirrels in pines at the golf course at the Palos Verdes Country Club near Malaga Cove on the Palos Verdes Peninsula in the 1990s; in this same area in roughly the same time period, a local naturalist observed what appeared to be a single individual in the same area (R. Melin, via email). A recent visit to this area (October 2014) revealed that it still supported a dense forest of coast live oak, toyon ( Heteromeles arbutifolia ), many mature planted conifers and eucalyptus, and a riparian strip running through the golf course (DSC, pers. obs.). And, whereas the eucalyptus plantation here has apparently been established for more than a century (Gales 1988), due to the extreme isolation of this area from any other known S. griseus populations, its coastal location, and the possibility that this population derived from deliberately introduced individuals (or pertains to the eastern gray squirrel, Sciurus carolinensis), we consider a Palos Verdes population to be "hypothetical” for now until more information is uncovered that would support its inclusion in the current range of the species. Discussion Our investigation into the distribution of the western gray squirrel in the Los Angeles area elucidates its status as essentially a foothill species that is now rare and declining below around 457 m elevation, particularly in areas where it has come into contact with the eastern fox squirrel. Away from its main strongholds in the western Santa Monica Mountains, the San Gabriel Mountains, and the Santa Ana Mountains, small, isolated populations persist only in the Santa Susana Mountains, Griffith Park, the Verdugo Mountains and San Rafael Hills, the San Jose Hills, the Chino Hills, at Rancho Santa Ana Botanic Gardens in Claremont, and in Redlands/ Yucaipa. Based on local naturalists’ observations, several lowland populations appear to have declined in the past five years, including that in Bonelli Park, the San Rafael Hills, Chino Hills State Park, and along the Santa Ana River Canyon near Yorba Linda. Invariably, extirpations have occurred concurrently with colonization by the ubiquitous S. niger. It is probably unlikely that truly extirpated, isolated lowland populations in the area will re-develop on their own. Areas of recent extirpation (or near-extirpation, where S. griseus is no longer resident but may occur irregularly) are typically separated from the nearest presumed source population by more than a kilometer, and generally by dense residential or urban development. Multi-lane freeways now provide formidable barriers between these areas of extirpation and source populations of S. griseus. Remarkably, animals do persist in a handful of lowland areas with very limited habitat, including the Industry Hills in La Puente, which suggests that certain small, isolated subpopulations may act as “refugia”, perhaps from pathogens that periodically sweep through larger and more intact populations. Of course, these same refugia are vulnerable to their own extinction events, and so are almost certainly temporary. Erkabaeva (2013) demonstrated that the length of projected coexistence of the two squirrel species in a given habitat fragment depends upon both the size of the habitat fragment and the structure of the habitat within the fragment, with length of coexistence associated with a higher diversity of food bearing tree species and coniferous trees. Sciurus griseus had a high probability of going extinct within a relatively short period of time (10 to 40 years) in small to medium-sized habitat fragments. The presence of the S. niger in the same habitat brought about extinction in a shorter period of time. WESTERN GRAY SQUIRREL IN LOS ANGELES AREA 51 Competition with other squirrel species has been suggested as a potential cause of S. griseus decline (or a contributor to its current patchy distribution) in the region, but the mechanisms involved in this relationship need further study. Extirpation sites generally support very high densities of S. niger , yet this species simply occurs at higher densities in general. Sciurus niger is highly urban-adapted, and occurs at all the sites where S. griseus has vanished, and we have not confirmed a site where S. griseus has been extirpated and where S. niger is completely absent. Still, King (2004) found few interactions among S. niger, S. griseus, and even California ground squirrel ( Spermophilus beecheyi) in her study area where all three occur in San Dimas, California (eastern Los Angeles Co.), and Ortiz (2014) also observed very few aggressive interactions between S. niger and S. griseus in her local study areas. Regardless of the mechanism, the loss of S. griseus in these areas - and region-wide - may be associated with a profound ecological change and degradation of seemingly healthy oak woodland and other habitat, particularly in wildland areas where replacement has occurred (e.g., the Puente-Chino Hills). Larger wildland areas where S. griseus is persisting in the presence of S. niger are of particular interest because these appear to offer the basic habitat needs of both species, at least for some period of time, and possibly in different areas of the landscape. The discovery of nests of S. griseus well into protected open space such as in the rugged Cedarbend/Whiting Woods area of the Verdugo Mountains (DSC, unpubl. data) and at San Dimas Canyon Park (King 2004) suggests a pattern of edge-avoidance, possibly related to increased competition with the eastern fox squirrel at the urban edge. However, this pattern breaks down at sites like Fern Dell in Griffith Park, where S. griseus occurs a few feet from houses and dense urbanization (DSC, unpubl. data). Here, supplemental feeding or food provisioning may simply be “propping up” the population of S. griseus which has also been aided by the abundance of planted trees providing additional food sources (fruits and nuts). Although we have made a few direct incidental observations of supplemental feeding (e.g., unshelled peanuts dropped at Fern Dell in Griffith Park being carried off by S. griseus), it probably occurs widely. Other vegetative characteristics that allow S. griseus to persist here include some amount of closed-canopy woodland (or woodland-like groves of trees) with an open understory rich in non-woody debris and leaf litter; older, mast-producing trees for food; and at least a few very tall trees for nest placement (Linders and Stinson 2007), characteristics that still apply to many parks in the region. More proximate factors in the decline of S. griseus relevant in our study area include death from injury and disease. Mortality from roadkill has been shown to be a major (if localized) factor in squirrel deaths in studies in Washington state (Linders and Stinson 2007), and S. griseus is frequently detected as roadkill in the Los Angeles area (pers. obs.). Many sites at the urban-wildland interface, including sites with documented S. griseus extirpations have roads along a canyon bottom, making squirrels that live in low densities and that forage on the ground particularly vulnerable. Other important causes of death and/or population decline include necrotic mange (found in many populations of S. griseus but oddly, apparently undocumented in the introduced S. niger in California, per King 2004); habitat quality decline from removal or disruption of the forest canopy due to development, tree-cutting, or fire; soil trampling and compaction (which reduces the biomass of fungi and perhaps other foods); and extreme natural events such as prolonged drought, which work synergistically to wipe out small populations. However, considering how modified the current habitat of many lowland S. griseus populations is (e.g., planted pines on golf courses), habitat transformation would seem to be a relatively minor threat. 52 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Based on the continuing trend of extirpation in the region, we consider all existing lowland populations of S. griseus to be highly imperiled throughout the Los Angeles area. We estimate one of the largest intact populations within the urban core of the region, that at Griffith Park, at well under 50 individuals, and even here it is geographically limited within the park itself, with most of the population in two adjacent canyons (DSC, unpubl. data). Smaller, more isolated populations such as that at Rancho Santa Ana Botanic Gardens and at various patches in the San Jose Hills are now “landlocked” by freeways and urbanization and are probably much more imperiled; populations here and the Chino Hills are now surrounded and infiltrated by S. niger (fide A. Ing), and they may not be able to resist continued invasion by this species. In the case of Redlands/Yucaipa, it is likely only a matter of time before S. niger colonizes and saturates the residential areas and parks where S. griseus currently occurs alone. Should re-introduction of S. griseus to lowland areas be attempted, we recommend this be limited to large, protected areas of natural habitat; however, reintroduction into areas where S. niger has already saturated the surrounding landscape and S. griseus has disappeared, such as at Franklin Canyon Park in Beverly Hills or along the lower Arroyo Seco in Pasadena, seems unlikely to succeed in the long term. Another possibility might be the modification of large closed landfills that have trees with a significant amount of closed canopy and that produce appropriate food items. We refer readers to Gatza (2011) for information on a Habitat Suitability Model that would support S. griseus while not being conducive to S. niger. Landfills within large urban areas often cover hundreds of hectares, and modification of portions of these landfills with corridors between suitable habitat fragments could provide new habitat for “lowland” western gray squirrels. We would not recommend introducing individuals from outside into areas of continued occurrence, such as Griffith Park, which would have the potential to introduce an unknown pathogen into vulnerable, isolated populations. Literature Cited Cooper, D.S. 2000. Breeding birds of a highly-threatened open space: the Puente-Chino Hills, California. Western Birds, 31:213-234. — . 2011. Rare plants of Griffith Park, Los Angeles, California. Fremontia, 38(4): 1 8—24. Erkabaeva, K. 2013. Habitat structure and extinction risk modeling of Sciurus griseus in long-term coexistence habitats of southern California. M.S. thesis, California State Univ., Los Angeles. Gales, D. 1988. Handbook of Wildflowers, Weeds, Wildlife, and Weather of the South Bay and Palos Verdes Peninsula, Third Edition. FoldaRoll Company, Palos Verdes Peninsula, California. Garrett, K. and J. Dunn. 1981. Birds of Southern California: Status and Distribution. Los Angeles Audubon Society, Los Angeles. Gatza, B.P. 2011. The effects of habitat structure on western gray squirrels and invasive eastern fox squirrels. M.S. thesis, California State Univ., Los Angeles. Grinnell, J. 1898. Birds of the Pacific Slope of Los Angeles County. Pasadena Academy of Sciences Publication No. 11. Guthrie, D. 2009. Suburban Squirrels. Chaparral Naturalist, 49(1), September/October 2009. Jameson, E.W. and H.J. Peeters. 1988. California Mammals. Univ. of California Press, Berkeley, CA. King, J.L. 2004. The current distribution of the introduced fox squirrel ( Sciurus niger ) in the greater Los Angeles metropolitan area and its behavioral interaction with the native western gray squirrel (Sciurus griseus). M.S. thesis, California State Univ., Los Angeles. — , M.C. Sue, and A.E. Muchlinski. 2010. Distribution of the eastern fox squirrel ( Sciurus niger) in southern California. The Southwestern Naturalist, 55(1):42M9. Lewis, S.A. 2009. Factors that allow the native western gray squirrel ( Sciurus griseus) and the introduced eastern fox squirrel ( Sciurus niger) to coexist in certain habitats within California. M.S. thesis, California State Univ., Los Angeles. WESTERN GRAY SQUIRREL IN LOS ANGELES AREA 53 Linders, M.J. and D.W. Stinson. 2007. Washington State Recovery Plan for the Western Gray Squirrel. Washington Dept, of Fish and Wildlife, Olympia, 128+viii pp. Muchlinski, A.E., G.R. Stewart, J. L King, and S.A. Lewis. 2009. Documentation of replacement of native western gray squirrels by introduced eastern fox squirrels. Bull. So. Calif. Acad. Sci., 108:160-162. Ortiz, J.L. 2014. Behaviors of the native western gray squirrel ( Sciurus griseus ) and the invasive eastern fox squirrel ( Sciurus niger ) in Los Angeles and surrounding counties. M.S. thesis, California State Univ., Los Angeles. Pequegnat, W.E. 1951. The biota of the Santa Ana Mountains. Journal of Entomology and Zoology. Nos. 3 and 4. Wilson, D.E. and D.M. Reeder, Editors. 2005. Mammal species of the world: a taxonomic and geographic reference. Third Edition. Smithsonian Institution Press, Washington, D.C. Bull. Southern California Acad. Sci. 114(1), 2015, pp. 54-57 © Southern California Academy of Sciences, 2015 A Young-of-the-Year Giant Sea Bass, Stereolepis gigas Buries Itself in Sandy Bottom: A Possible Predator Avoidance Mechanism Michael C. Couffer1 and Stephanie A. Benseman2 lGrey Owl Biological Consulting 2California State University, Northridge The adult giant sea bass, Stereolepis gigas, (GSB) is the largest teleost inhabiting California’s shallow rocky reefs, attaining a length of about 2.3 m (7 ft) and a maximum weight of around 256 kg (563 lbs) (Baldwin and Keiser 2008). They range from Humboldt Bay, California to Oaxaca, Mexico, including the Gulf of California (Miller and Lea 1972). Adults consume a wide variety of prey and occupy rocky bottom habitat ranging from approximately 7^10 m (25-130 ft) of water (Miller and Lea 1972) and can forage over sandy bottom, away from rocky reefs (Baldwin and Keiser 2008). After their peak commercial catch in 1932, at just over 1 14,000 kg, the population quickly crashed and their numbers have remained depressed ever since; this has inhibited detailed research (Pondella and Allen 2008). Young-of-the-year (YOY) GSB pass through various color phases and morphological changes during early development. These transitions help it to appear cryptic, while hiding to avoid predators during a vulnerable stage of life. When less than 2.5 cm (1 in), these YOY appear black with several small white spots around its face (Fig. 1). This black stage has very large black dorsal and pelvic fins, with transparent pectoral, anal, and caudal fins. The black juveniles morph through a “brown” stage, to a bright orange fish (Fig. 2). The black dorsal fin changes to orange, while the enormous pelvic fins remain black. Color expands outward to include half of the pectoral and anal fins, and the entire tail remains clear. The white spots remain from the earlier stages, and small black spots also appear (Pers. obs., and Benseman, unpublished data). These YOY appear to frequent open, sand and mud-bottomed habitat between 2-30 m (7-100 ft) for the first few months after settlement (Benseman, unpublished data). During a focused SCUBA survey for YOY GSBs at Veteran’s Park in Redondo Beach, Los Angeles County, California, Michael Couffer located a roughly 2.5 cm (1 in) long orange juvenile GSB in 5.5 m (16.5 ft) of water, floating upright in the bottom of a shallow sandy depression with its dorsal and pelvic fins closed. The bottom was clean sand without surface detritus. When approached, the GSB raised its dorsal and pelvic fins and left the depression, moving slowly within 30 cm of the bottom. The fish was photographed to record the sighting time in image metadata, and followed from about a meter away to acclimate the fish to human presence so that it could be photographed in profile. After the fish had moved about 9 m, Mr. Couffer approached to half a meter to photograph its spot pattern. The fish startled, and darted toward the bottom at an angle. As the fish reached the bottom, it turned on its side at the last instant and buried into the soft sand by undulating its body like a flatfish. It pushed its head beneath the sand and undulated until the entire fish was buried in under three seconds. I took several photos of Corresponding author: mikecouffer@gmail.com 54 A POSSIBLE PREDATOR AVOIDANCE MECHANISM 55 Fig. 1. An 18mm YOY Giant Sea Bass from Newport Beach, California. Fig. 2. A 75mm YOY Giant Sea Bass from Newport Beach, California. 56 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fig. 3. Scales on the side of the buried YOY Giant Sea Bass show through the sand to the left of the insert. the exact spot where the GSB had disappeared (Fig. 3), and then put a 15 cm net over the spot, working the net’s frame down into the sand. The fish remained buried. I dug my hand deep into the soft sand under the net, lifted a ball of sand containing the fish up into the net. As the sand fell away to the sides of the ball, the GSB burst out of the sand and up into the net. I measured the GSB at 32 mm, and released it. The GSB darted beneath me as I knelt on the sand. I pushed off the bottom, but the fish was gone. The GSB’s actions appeared to be a flight response, a possible last-ditch effort to evade predation in an area where there was no available cover for shelter. Unlike flatfish that may cover themselves with soft sediment to ambush their prey (Gibson and Robb 1991 ), or certain benthic gobiid fishes that have mutualistic relationships with shrimp that dig holes for shelters (Horinouchi 2008; Thacker et. al. 2011), this GSB was behaving as if actively trying to avoid detection by a “predator”. Senoritas, Oxyjulis calif ornica, are also known to bury themselves to avoid predators, but this occurs mostly at night, with the senorita remaining buried for protection (Hobson, E.S. 1968), and not as an immediate escape response. This GSB predator evasion method should prove highly effective, as few predators could dig in the sand for the fish after burial. The bottom was so uniform that if the observer had looked away from the spot where the fish had buried, its location would have been lost (Fig. 3). The open expanse of sand and mud bottoms that these small YOY GSB utilize make ideal nursery areas since there is an abundance of food, such as mysids and other arthropods (Dahl 1952), and relatively few inhabitants, including predators (McLachlan 1990). However, when a juvenile does encounter a predator, it must rely on its cryptic coloration and shape, and other types of active and passive predator avoidance strategies. The burying behavior observed may be a successful last-resort predator avoidance strategy for GSB, and is certainly the first one documented. A POSSIBLE PREDATOR AVOIDANCE MECHANISM 57 Acknowledgements We would like to thank R. H. Defran of San Diego State University, L. G. Allen of California State University, Northridge, and D. J. Pondella II for reviewing this note. Literature Cited Baldwin, D.S. and Kaiser, A. 2008. Giant sea bass, Stereolepis gigas, status of the fisheries report. Cal. Dept. Fish Game. p. 8. Dahl, E. 1952. Some aspects of the ecology and zonation of the fauna on sandy beaches. Oikos, 4(l):l-27. Hobson, E.S. 1965. Diurnal-Nocturnal Activity of Some Inshore Fishes in the Gulf of California. Copeia, 291-302. Horinouchi, M. 2008. Patterns of food and microhabitat resource use by two benthic Gobiid fishes. Environ. Biol. Fish., 82:187-194. Gibson, R.N. and Robb, D.L. 1992. The relationship between body size, sediment grain size and the burying ability of juvenile plaice, Pleuronectes platessa. L. J. Fish Biol., 40(77):1— 778. McLachlan, A. 1990. Dissipative beaches and macrofauna communities on exposed intertidal sands. J. Coast. Res., 6(1):57— 71. Miller, D.J. and Lea, R.N. 1972. Guide to the Coastal Marine Fishes of California. Calif. Dept. Fish. Game, Fish Bull., 157-249. Pondella, D.J. II and Allen, L.G. 2008. The decline and recovery of four predatory fishes from the Southern California Bight. Mar. Biol., 154:307-313. Thacker, C., Thompson, A., and Roje, D. 2011. Phylogeny and evolution of Indo-Pacific shrimp- associated gobies Gobiiformes: Gobiidae. Mol. Phylog. Evol., 59( 1):168— 176. Bull. Southern California Acad. Sci. 114(1), 2015, pp. 58-62 © Southern California Academy of Sciences, 2015 Nelson’s big horn sheep ( Ovis canadensis nelsoni ) trample Agassiz’s desert tortoise ( Gopherus agassizii ) burrow at a California wind energy facility Mickey Agha,1 David Delaney,2 Jeffrey E. Lovich,3 Jessica Briggs,4 Meaghan Austin3 and Steven J. Price1 1 Department of Forestry, University of Kentucky, Lexington, KY 40546, USA 2U.S. Army Construction Engineering Research Laboratory, P.O. Box 9005, Champaign, IL 61826-9005, USA 3 US. Geological Survey, Southwest Biological Science Center, 2255 North Gemini Drive, MS-9394, Flagstaff, Arizona 86001, USA 4 Colorado State University, Fort Collins, CO 80523, USA Research on interactions between Agassiz’s desert tortoises ( Gopherus agassizii ) and ungulates has focused exclusively on the effects of livestock grazing on tortoises and their habitat (Oldemeyer, 1994). For example, during a 1980 study in San Bernardino County, California, 164 desert tortoise burrows were assessed for vulnerability to trampling by domestic sheep ( Ovis aries). Herds of grazing sheep damaged 10% and destroyed 4% of the burrows (Nicholson and Humphreys 1981). In addition, a juvenile desert tortoise was trapped and an adult male was blocked from entering a burrow due to trampling by domestic sheep. Another study found that domestic cattle ( Bos taurus) trampled active desert tortoise burrows and vegetation surrounding burrows (Avery and Neibergs 1997). Trampling also has negative impacts on diversity of vegetation and intershrub soil crusts in the desert southwest (Webb and Stielstra 1979). Trampling of important food plants and overgrazing has the potential to create competition between desert tortoises and domestic livestock (Berry 1978; Coombs 1979; Webb and Stielstra 1979). Native ungulates like Nelson’s big horn sheep ( Ovis canadensis nelsoni ) co-occur with desert tortoises in portions of the desert southwest. Due to habitat and partial dietary overlap of various annual forbs and grasses at certain elevations (Ernst and Lovich 2009; Oehler et al. 2003), there is potential for contact between these species. Although there are data demonstrating damage and destruction of desert tortoise burrows caused by domestic ungulates (Nicholson and Humphreys 1981; Avery and Neibergs 1997), it is previously undocumented if native sheep like Nelson’s big horn sheep are capable of similar impacts to tortoise burrows. On 29 September 2013, we documented desert tortoise burrow collapse caused by Nelson’s big horn sheep trampling at a wind energy facility in Riverside Co., California, USA (33°57'06"N, 116°40,02,'W, WGS84). In the summer of 2013 (1 June 2013 to 14 November) 48 Reconyx and Wildgame motion sensor trail cameras were deployed at the entrances of desert tortoise burrows during an ongoing investigation of the effects of wind energy generation on behavior and activity of this species (Lovich et al. 2014). Cameras were mounted on 1.5 m foot tall steel stakes inserted into the ground approximately 1 m from desert tortoise burrow entrances that were known to be occupied or used recently. Cameras were activated by movement of wildlife via an infrared sensor, Corresponding author: steven.price@uky.edu 58 BIG HORN SHEEP AND TORTOISES 59 Fig. 1 . Active desert tortoise burrow collapse caused by Nelson’s big horn sheep in a series of 4 motion sensor camera images. and programmed to take 1-5 photographs at a trigger speed of 0.2 sec. Each month, an investigator checked each camera and downloaded photos onto a data storage device. Lastly, surface air temperatures were collected every 30 minutes from an onsite Remote Automated Weather Station (WWAC1; accessed via the MesoWest website (http:// mesowest.utah.edu/index.html). Our motion sensor cameras recorded three Nelson’s big horn sheep approach a north facing active desert tortoise burrow (previously occupied by an adult male desert tortoise on 13 June 2013) at 1140 h. From 1241 h to 1303 h, several Nelson’s big horn sheep gathered below the entrance of the burrow, brushing loosened soil around the entrance of the burrow with their hooves, eventually causing the outer walls to collapse (Fig 1 .). Three different Nelson’s big horn sheep then proceeded to lie down and in the process compact the soil, rocks and sticks on top of the newly collapsed entrance from 1304 h to 1429 h. Fig. 2. Nelson’s big horn sheep at the entrance of desert tortoise burrow. Nelson’s big horn sheep may have been eating tortoise feces or soil, or simply investigating the burrow. 60 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 2013-09-02 11:26:01 ;!*. • J u&>J m i 3§llli PCBOO HVPERFIRE PRO Fig. 3. Domestic cattle walking past the entrance of a desert tortoise burrow. Several Nelson’s big horn sheep remained standing at the burrow from 1430 h to 1542 h. During these observations ambient air temperature ranged from 30.56 C to 32.50 C. Over the course of the camera-trapping study, Nelson’s big horn sheep were also recorded walking past or standing at the base of seven different desert tortoise burrows at various other locations throughout the study site. Photographs also revealed what appeared to be Nelson’s big horn sheep grazing near the mouth of the burrow (Fig. 2). Since few plants grow in the mouth of active tortoise burrows, the sheep may have been eating soil or possibly the fresh feces of desert tortoises that are comprised mostly of partially digested grass and forbs. In addition to big horn sheep, domestic cattle were captured by motion sensor cameras walking past the base of four desert tortoise burrows (Fig. 3). The images we recorded are the first documented evidence of Nelson’s big horn sheep trampling a desert tortoise burrow and subsequently collapsing the outer walls of the burrow in the process. Nelson’s big horn sheep employ various strategies of seeking shade and cooler soil for bedding (Cain et al. 2008), and it appears that north-facing slopes (location of collapsed tortoise-burrow) may provide such a site. Alternatively, previous studies of big horn sheep have documented extensive movement and occasionally large descents from mountain ranges to use mineral licks at lower elevations, as they provide sodium which is crucial to physiological functions (Bangs et al. 2005; Holl and Bleich 1987; Watts and Schemnitz 1985). Since most terrestrial plants have low concentrations of sodium (Weeks and Kirkpatrick 1976), Nelson’s big horn sheep may be mining essential minerals brought to the surface by tortoises through excavation of their burrows (Ernst and Lovich 2009; Turner et al. 1984). One study demonstrating soil ingestion, or geophagy, by bighorn sheep ( Ovis canadensis) in Alberta, Canada found their feces contained as much as 30% soil in some samples (Skipworth 1974). Ingestion of desert tortoise burrow soil may be important to Nelson’s big horn sheep as it could be a source of certain minerals (Beyer et al. 1994). Lastly, we hypothesize that relatively high plant productivity at the site (Ennen et al. 2012b; Lovich et al. 2015) attracts ungulates (Oehler et al. 2003), both domestic and native to the study area. Moderate winter precipitation produces an abundance of annual food plants for both desert tortoises and big horn sheep at the study site. Trampling and collapsing active desert tortoise burrows may entomb resident individuals (Loughran et al. 2011; Nichols and Humphries 1981), although in the majority of observed burrow collapses at the site, tortoises were able to excavate BIG HORN SHEEP AND TORTOISES 61 themselves (Loughran et al. 2011). In light of our observation, trampling may have greater impacts to slope dwelling rather than valley dwelling desert tortoises. Furthermore, female desert tortoises nest at the entrance and within burrows (Agha et al. 2013; Ennen et al. 2012a); consequently, trampling may negatively impact tortoise egg clutches or entomb emerging neonates (Berry 1978). Entombment of desert tortoises within burrows can cause physiological stress to the animal (Loughran et al. 2011), thereby leading to potential mortality (Lovich et al. 2011). We are unaware of any cases where bighorn sheep behavior resulted in mortality of desert tortoises and suspect that such interactions between the species are rare in comparison to interactions involving domestic ungulates. Acknowledgements Our research was supported by the California Energy Commission-Public Interest Energy Research Program (Contract NO.: 500-09-020), the California Desert District Office of the Bureau of Land Management, U.S. Army Construction Engineering Research Laboratory, and the Desert Legacy Fund of the California Desert Research Program. Research was conducted under permits from the United States Fish and Wildlife Service, California Department of Fish and Game, and the Bureau of Land Management. Earlier versions of the manuscript benefited greatly from comments offered by Vernon Bleich. Special thanks are given to A. Muth of the Boyd Deep Canyon Desert Research Center of the University of California, Riverside, for providing accommodations during our research. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Literature Cited Agha, M., J.E. Lovich, J.R. Ennen, and E. Wilcox. 2013. Nest-guarding by female Agassiz’s desert tortoise {Gopherus agassizii ) at a wind-energy facility near Palm Springs, California. The Southwestern Naturalist, 58:254-257. Avery, H.W., and A.G. Neibergs. 1997. Effects of cattle grazing on the desert tortoise, Gopherus agassizii : nutritional and behavioral interactions. In Proceedings: Conservation, Restoration, and Manage- ment of Tortoises and Turtles- An International Conference, 13-20. Bangs, P.D., P.R. Krausman, K.E. Kunkel, and Z.D. Parsons. 2005. Habitat use by female desert bighorn sheep in the Fra Cristobal Mountains, New Mexico, USA. European Journal of Wildlife Research, 51:77-83. Berry, K.H. 1978. Livestock grazing and the desert tortoise. In Transactions of the North American Wildlife and Natural Resources Conference (USA), 1978:505-519. Beyer, W.N., E.E. Connor, and S. Gerould. 1994. Estimates of soil ingestion by wildlife. Journal of Wildlife Management, 58:375-382. Cain, J.W., B.D. Jansen, R.R. Wilson, and P.R. Krausman. 2008. Potential thermoregulatory advantages of shade use by desert bighorn sheep. Journal of Arid environments, 72:1518-1525. Coombs, E.M. 1979. Food habits and livestock competition with the desert tortoise on the Beaver Dam Slope, Utah. Proceedings of the Desert Tortoise Council, 1979:132-147. Ennen, J.R., J.E. Lovich, K.P. Meyer, C. Bjurlin, and T.R. Arundel. 2012a. Nesting Ecology of a Population of Gopherus agassizii at a Utility-Scale Wind Energy Facility in Southern California. Copeia, 2012:222-228. Ennen, J.R., K. Meyer, and J.E. Lovich. 2012b. Female Agassiz’s desert tortoise activity at a wind energy facility in southern California: The influence of an El Nino event. Natural Science, 4:30-37. Ernst, C.H., and Lovich, J.E. 2009. Turtles of the United States and Canada. Johns Hopkins University Press. Holl, S.A., and V.C. Bleich. 1987. Mineral lick use by mountain sheep in the San Gabriel Mountains, California. Journal of Wildlife Management, 51:383-385. 62 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Lovich, J.E., D. Delaney, J. Briggs, M. Agha, M. Austin, and J. Reese. 2014. Black bears ( Ursus americanus ) as a novel potential predator of Agassiz’s desert tortoises ( Gopherus agassizii ) at a California wind energy facility. Bulletin of the Southern California Academy of Sciences, 113: 34-41. , J.R. Ennen, K. Meyer, M. Agha, C. Loughran, C. Bjurlin, M. Austin, S. Madrak. 2015. Not putting all their eggs in one basket: bet-hedging despite extraordinary annual reproductive output of desert tortoises. Biological Journal of the Linnean Society, 115.2:399^-10. , J.R. Ennen, S.V. Madrak, and B. Grover. 2011. Turtles, culverts and alternative energy development: an unreported but potentially significant mortality threat to the desert tortoise ( Gopherus agassizii ). Chelonian Conservation and Biology, 10:124-129. Loughran, C.L., J.E. Ennen, and J.E. Lovich. 201 1 . Gopherus agassizii (Desert tortoise). Burrow collapse. Herpetological Review, 42:593. Nicholson, L., and K. Humphreys. 1981. Sheep grazing at the Kramer study plot, San Bernardino County, California. In Proceedings of the 1981 symposium of the Desert Tortoise Council, 163-194. Oehler Sr., M.W., R.T. Bowyer, and Y.C. Bleich. 2003. Home ranges of female mountain sheep, Ovis canadensis nelsoni: effects of precipitation in a desert ecosystem. Mammalia, 67:385-402. Oldemeyer, J.L. 1994. Livestock grazing and the desert tortoise in the Mojave Desert, p. 95-103. In R. B. Bury and D. J. Germano (eds.), Biology of North American tortoises. U.S. Dept. Int. Natl. Biol. Surv. Fish Wildl. Res. 13. Skipworth, J.P. 1974. Ingestion of grit by bighorn sheep. Journal of Wildlife Management, 38:880-883. Turner, F.B., P. A Medica, and C.L. Lyons. 1984. Reproduction and survival of the desert tortoise (Scaptochelys agassizii) in Ivanpah Valley, California. Copeia, 1984:811-820. Watts, T.J., and S.D. Schemnitz. 1985. Mineral lick use and movement in a remnant desert bighorn sheep population. Journal of Wildlife Management, 49:994-996. Webb, R.H., and Stielstra, S.S. 1979. Sheep grazing effects on Mojave Desert vegetation and soils. Environmental Management, 3:517-529. Weeks Jr., H.P., and C.M. Kirkpatrick, 1976. Adaptations of white-tailed deer to naturally occurring sodium deficiencies. Journal of Wildlife Management, 40:610-625. SMITHSONIAN LIBRARIES 3 9088 01817 2296 CONTENTS Possible Stock Structure of Coastal Bottlenose Dolphins off Baja California and California Revealed by Photo-Identification Research. R.H. Defran, Marthajane Caldwell, Eduardo Morteo, Aimee R. Lang, Megan G. Rice, and David W. Weller ... J 1 Removal Efforts and Ecosystem Effects of Invasive Red Swamp Crayfish (Procambarus clarkii) in Topanga Creek, California. Crystal Garcia, Elizabeth Montgomery, Jenna Krug, and Rosi Dagit 12 Soil Organic Carbon and Nitrogen Storage in Two Southern California Salt Marshes: The Role of Pre-Restoration Vegetation. Jason K. Keller, Tyler Anthony, Dustin Clark, Kristin Gabriel, Dewmini Gamalath, Ryan Kabala, Julie King, Ladyssara Medina, and Monica Nguyen 22 Identical Response of Caged Rock Crabs (Genera Metacarcinus and Cancer ) to Energized and Unenergized Undersea Power Cables in Southern California, USA. Milton S. Love, Mary M. Nishimoto, Scott Clark, and Ann Scarborough Bull 33 Recent Decline of Lowland Populations of the Western Gray Squirrel in the Los Angeles Area of Southern California. Daniel S. Cooper and Alan E. Muchlinkski 42 A Young-of-the-Year Giant Sea Bass, Stereolepis gigas Buries Itself in Sandy Bottom: A Possible Predator Avoidance Mechanism. Michael C. Couffer and Stephanie A. Benseman 54 Nelson’s Big Horn Sheep (Ovis Canadensis nelsoni ) Trample Agassiz’s Desert Tortoise ( Gopherus agassizii) Burrow at a California Wind Energy Facility. Mickey Agha, David Delaney, Jeffrey E. Lovich, Jessica Briggs, Meaghan Austin, and Steven J. Price 58 Cover: Young-of-the-Year Giant Sea Bass from Newport Beach, California. Photo by permission of Michael C. Couffer.