Black abalone (Haliotis cracherodii) population structure shifts through deep time: Management implications for southern California's northern Channel Islands

Abstract For over 10,000 years, black abalone (Haliotis cracherodii) were an important resource in southern California, first for coastal Native Americans, then beginning in the nineteenth century, as one of the state's first commercial shellfisheries. By 1993, after years of heavy fishing, rising sea surface temperatures (SST), and the spread of withering syndrome (WS), black abalone populations declined dramatically, resulting in the closure of the Alta California fishery. After nearly 25 years of management and recovery efforts, black abalone are showing signs of ecological rebound along some Channel Island shorelines. These include the presence of juvenile abalone and increasing densities, largely from data collected by Channel Islands National Park (CINP) monitoring efforts that began in 1985. In an effort to apply deeper historical perspectives to modern fisheries management and restoration, we analyzed black abalone size data from San Miguel Island at prehistoric and historical archeological sites spanning the last 10,000 years and compared these populations to those described by CINP biologists between 1985 and 2013. We found a statistically significant relationship between SST and black abalone size distributions during the ancient record, along with dramatic shifts in population size structure toward larger individuals between the nineteenth century and modern periods. A pattern of larger mean black abalone sizes was identified during warm SSTs, when compared against intervals of cooler SSTs. Synthesis and applications. Our study provides a deep historical perspective of abalone population size distributions, patterns within these distributions through time, and parallels to modern abalone populations. Our results may help managers determine whether the current (and future) size and age structure of intertidal black abalone populations around the northern Channel Islands are “natural” and healthy, measured against the 10,000 year history of black abalone fishing in southern California.

2. In an effort to apply deeper historical perspectives to modern fisheries management and restoration, we analyzed black abalone size data from San Miguel Island at prehistoric and historical archeological sites spanning the last 10,000 years and compared these populations to those described by CINP biologists between 1985 and 2013.
3. We found a statistically significant relationship between SST and black abalone size distributions during the ancient record, along with dramatic shifts in population size structure toward larger individuals between the nineteenth century and modern periods. A pattern of larger mean black abalone sizes was identified during warm SSTs, when compared against intervals of cooler SSTs. 4. Synthesis and applications. Our study provides a deep historical perspective of abalone population size distributions, patterns within these distributions through time, and parallels to modern abalone populations. Our results may help managers determine whether the current (and future) size and age structure of intertidal black abalone populations around the northern Channel Islands are "natural" and healthy, measured against the 10,000 year history of black abalone fishing in southern California.

| INTRODUC TI ON
The long-term impacts, both positive and negative, of humans on local environments have become an important research avenue for a variety of disciplines over the last decade (e.g., Balée, 2006;Erlandson & Rick, 2010;Rick & Lockwood, 2012). Seminal research by fisheries biologists has shown that nearshore marine ecosystems are highly susceptible to anthropogenic impacts and effective modern management requires deep historical perspectives of human-marine ecodynamics (e.g., Pauly, 1995;Jackson et al., 2001).
Archeologists and other historical scientists have taken up this call as part of a marine historical ecology research agenda, simply defined as the integration of marine ecology and history (Rick & Lockwood, 2012). One especially fruitful avenue of marine historical ecology studies has focused on the effects of humans on nearshore shellfish communities, incorporating archeological, paleoecological, and historical data into modern management practices and restoration strategies (e.g., Braje, Rick, Erlandson, Rogers-Bennett, & Catton, 2016;Finney, Gregory-Eaves, Douglas, & Smol, 2002;Lotze et al., 2006;Rick et al., 2016;Rogers-Bennett, Haaker, Huff, & Dayton, 2002).
Along the Pacific Coast of North America, black abalone (Haliotis cracherodii) is an interesting species for historical ecology studies.
California black abalone are critically endangered, and their management might benefit from deep historical perspectives. Since 1993, recreational and commercial black abalone fishing has been suspended, and despite careful management and restoration attempts, black abalone were upgraded by the National Marine Fisheries Service from a species of concern to endangered in 2009 under the Endangered Species Act (California Fish & Game Commission, 2005;Neuman, 2009). On San Nicolas and Santa Cruz islands, there have been some signs of recovery (e.g., recruitment and population expansion; Butler et al., 2009), and recruitment has been observed for black abalone at Anacapa, Santa Rosa, and San Miguel islands in recent years (Raimondi, Jurgens, & Tinker, 2015; S. G. Whitaker, personal obs.), but we lack a clear picture of what a "healthy" abalone population looked like prior to population collapse. Modern data only reflect black abalone sizes and densities following decades of intensive commercial harvest, widespread disease, and the local extirpation of southern sea otters (Enhydra lutris), one of the primary predators of black abalone. Some ecologists, however, have begun to incorporate historical and archeological datasets in the development of deep historical baselines to evaluate modern abalone populations (e.g., Rogers-Bennett et al., 2002).
Here, we extend historical ecological studies of Channel Islands black abalone populations by compiling the size structure of black abalone collected from the rocky intertidal zone around San Miguel Island over the last 10,000 years. We use the maximum shell length of black abalone shells from archeological sites spanning the Early (11,700-7,500 cal BP), Middle (7,500-3,500 cal BP), and Late (3,500 cal BP-AD 1782) Holocene, and the historic period (ca. AD  to examine how population size structure changed through deep time and evaluate whether these changes can be attributed to anthropogenic and/or environmental causes. We then compare archeological and historical data to modern size data collected by Channel Islands National Park (CINP) biologists to assess whether the modern (and future) size and age structures of intertidal black abalone populations are similar compared against 10,000 years of continually and intensively harvested abalone populations. In doing so, our data may aid resource managers and restoration biologists in evaluating the health of current and future black abalone populations in southern California.

| Environmental setting and cultural background
San Miguel Island is the western-most and second smallest (37 km 2 ) of the Northern Channel Islands (NCI). Approximately half of San Miguel's 43 linear km of coastline consists of rocky shores that support abalone and other shellfish species, although the length and extent of rocky shore habitats have changed with fluctuations in sea level and dune building (Erlandson, Rick, & Peterson, 2005;Graham, Dayton, & Erlandson, 2003).
Long-term changes in sea surface temperature (SST), patterns of upwelling, marine productivity, and kelp forest cover (Graham et al., 2003;Kennett, 2005) are natural changes that contribute to variation in shellfish productivity and growth rates. Studies of varved sediments from the Santa Barbara Basin provide a high-resolution record of Holocene SST changes and a picture of long-term climatic variability, with resolution at 25-year intervals for the last 3,000 years and 50-year intervals for the rest of the Holocene (Kennett, 2005). During the Holocene, four cold-water intervals (spanning 9,600-8,200 cal BP; 6,300-5,900 cal BP; 3,800-2,900 cal BP; and 1,500-500 cal BP) and four warm-water intervals (spanning 11,000-9,600 cal BP; 8,200-6,300 cal BP, 5,900-3,800 cal BP; and 2,900-1,500 cal BP) occurred, and since ~500 years ago, SSTs have undergone general warming. Short-term changes, such as El Niño-Southern Oscillation (ENSO) events, are difficult to gauge, but are assumed to have occurred in the past, although probably at lower frequencies than today (Kennett, 2005).
The NCI and the coastal mainland were continuously occupied by the Chumash and their ancestors for at least 13,000 years. From the terminal Pleistocene through the Holocene, Native Channel Islanders were maritime hunter-gatherers, who relied on marine foods, particularly nearshore shellfish, as their primary protein source (Braje, Rick, & Erlandson, 2012;Erlandson, Rick, & Braje, 2009). The resulting disposal of shellfish remains in large, coastal K E Y W O R D S applied archeology, historical ecology, shifting baselines shell middens provide researchers with an excellent record of past shellfish harvesting activities and the structure and nature of shellfish populations through time.
While the archeological record does reflect human choice, it can provide a window into the ancient structure of local intertidal ecosystems. Prehistoric hunter-gatherers were driven by subsistence goals in their harvest pursuits and archeologists operationalize what resources they would have pursued using optimal foraging theory (OFT) and human behavioral ecology (HBE). OFT is a useful tool for considering the productivity of different prey choices through time (see Braje, Kennett, Erlandson, & Culleton, 2007). Studies using OFT and ethnohistoric records suggest that black abalone were a highly ranked resource throughout the Holocene based on their large size, meat yield, ease of capture, abundance, and aggregation. When available in the local intertidal system, hunter-gatherers would have targeted black abalone before nearly any other shellfish species.
Although island shell middens often contain a range of black abalone shell sizes, Channel Islanders likely focused on larger individuals because they provide higher meat yields and are often less difficult to access than smaller abalone that often utilize crevice habitat. Thus, we assume that the archeological record of black abalone size will be biased toward adult abalone with juveniles underrepresented and the exclusion of post-larvae abalone. This should be similar for modern monitoring data as post-larvae, and juveniles are as difficult to access today as they would have been in the past.
Terminal Pleistocene (ca. >11,700 cal BP) and Early Holocene archeological sites are relatively uncommon when compared to Middle and Late Holocene sites. Many terminal Pleistocene/Early Holocene sites probably have been lost due to rising sea levels and other taphonomic processes, and most of the early sites on San Miguel consist of low density shell middens and lithic scatters (Rick, Erlandson, Vellanoweth, & Braje, 2005). Zooarchaeological analyses indicate that Early Holocene subsistence economies focused on rocky intertidal shellfish species such as California mussel and black abalone . By the Middle Holocene, the number of recorded NCI sites increases and evidence for large populations settled in permanent shoreline villages is apparent (Kennett, 2005). During the Late Holocene, there were significant alterations to subsistence, settlement patterns, and technology and a marked increase in populations (Erlandson et al., 2009). Many Late Holocene sites are large, permanently occupied coastal villages and high population densities led to the increased dependence on finfish (Kennett, 2005;Rick, Vellanoweth, Erlandson, & Kennett, 2002). Although the relative contribution of shellfish to the protein diets of the Late Holocene Chumash decreased, harvesting of shellfish intensified and the mean size of several key dietary species declined significantly (Braje et al., 2012).
After the last of the Island Chumash were removed to Spanish settlements on the mainland by AD 1822, Chinese fishermen established commercial fishing camps in Alta and Baja California in the 1850s focused primarily on the harvesting of intertidal black abalone (Braje, 2016). Chinese fishermen traveled from mainland bases to the Channel Islands to gather, dry, and process abalone for export, facilitated by spectacular abalone abundance resulting from local sea otter extirpation and the removal of Chumash harvesting pressure.
Between the mid-1850s and the late AD 1880s, Chinese fishermen harvested hundreds of tons of abalone from coastal waters each year (Braje, 2016). The fishery peaked in 1879 when an estimated 1,860 tons of abalone were harvested from southern Californian waters (Rogers-Bennett et al., 2002). By the late nineteenth century, the Chinese fishery was in decline due to growing anti-Chinese sentiment and passage of targeted, racist laws and, perhaps, declining intertidal abalone populations (Braje, 2016). Japanese immigrants filled the void and shifted the focus to subtidal abalone species by ~1898, taking as many as 40-50 dozen red (Haliotis rufescens), green (H. fulgens), and pink (H. corrugata) abalone per day. Black abalone were considered undesirable compared to other species and fished primarily as bait for most of the 20th century (Altstatt et al., 1996;Parker, Kaaker, & Togstad, 1992), until declines in subtidal species stocks led to the redevelopment of a commercial export fishery in 1968. The black abalone fishery experienced a rapid rise and peaked in 1973, followed by a severe harvest decline due to heavy fishing pressure, coastal development, and pollution (Altstatt et al., 1996).

| Abalone ecology
Black abalone are a smooth-shelled species of abalone that live in the intertidal and shallow subtidal zones of rocky reefs, rarely venturing into deeper waters (Tissot, 1992). Black abalone, like other abalone species, are broadcast spawners and commonly found aggregating (Ault, 1985). Growth rates for Haliotis spp. are rapid during early years, but slow as age and size increase .
Black abalone, in particular, are considered juveniles until they reach 45-50 mm in length and, on average, grow to around 115 mm, with some reaching a maximum of 215 mm (Ault, 1985).
Environmental conditions seriously affect the health of abalone populations, directly and indirectly. Inhabiting an environment in which land and sea overlap, black abalone are susceptible to environmental changes affecting both the ocean and atmosphere (Harley & Rogers-Bennett, 2004). California abalone species are linked to kelp forest ecosystems as giant kelp (Macrocystis pyrifera) is one of their primary food sources and, when coupled with red algae, is thought to produce the fastest growth rates (Leighton & Boolootian, 1963).
Warmer water temperatures are typically correlated with fewer nutrients in the water column (Edwards & Estes, 2006), resulting in lower qualities and densities of kelp and, thus, a reduced food source (Vilchis et al., 2005). Consequently, warm SSTs are linked to reduced abalone growth rates and reproductive success, and an increase in the expression of WS (Harley & Rogers-Bennett, 2004).
Several factors limit California abalone populations. Their slow growth rate and preferred habitat in the intertidal and shallow subtidal make them susceptible to terrestrial and marine predators, including ocher sea stars, octopi, cabezon, crabs, spiny lobsters, and particularly humans and sea otters (Harley & Rogers-Bennett, 2004).
The historical extirpation of otters from Channel Islands waters allowed abalone populations to explode, reaching densities of up to 100 individuals per square meter.
A catastrophic decline of black abalone in southern California occurred in the late 1980s through the mid-1990s as WS, a Rickettsialeslike disease, systematically eradicated entire populations (Altstatt et al., 1996;Raimondi et al., 2002). Commercial fishermen first recognized signs of the disease, empty shells, and deteriorated black abalone along the south shore of Santa Cruz Island in 1985. A year later, diseased abalone were observed and documented by CINP biologists on Anacapa Island. Subsequently, the disease was recognized throughout the Channel Islands and the southern California mainland (Altstatt et al., 1996;Harley & Rogers-Bennett, 2004;Raimondi et al., 2002). Some studies indicate that the effects of WS are intensified by stressful environmental conditions such as warm SST (Davis, Richards, Haaker, & Parker, 1992;Tissot, 1988Tissot, , 1990) and reduced food supply (Tegner & Dayton, 1987), although Lafferty and Kuris (1993) found no association between die-offs and warm SST or changes in kelp density. The disease does not target any particular age or size class, and in areas where mass die-offs have occurred, very few individuals remain (Raimondi et al., 2002).
Declines in abalone populations are assumed to have further led to a lack of success in juvenile recruitment likely due to low densities of reproductive adults which, in turn, results in reduced larval dispersal and the loss of suitable habitat, caused by the colonization of encrusting species (Raimondi et al., 2002). Because current recruitment patterns do not appear to be sufficient to replenish mainland populations of black abalone in southern California and WS continues to impact the existing population, several management and restoration strategies have been proposed (Moore, Finley, Friedman, & Robbins, 2002;Raimondi et al., 2002;Rogers-Bennett et al., 2002;Vilchis et al., 2005). Restoration activities have included the establishment of protected areas, the relocation of existing populations to create high density areas of black abalone that may result in increased reproduction and recruitment, and the outplanting of larvae, juveniles, or adults reared in hatcheries (Moore et al., 2002).

| MATERIAL S AND ME THODS
We analyzed San Miguel Island black abalone shell measurements from archeological sites spanning the last 10,000 years (Erlandson, Rick, Braje, Steinberg, & Vellanoweth, 2008), and modern black abalone shell measurements collected by CINP biologists using fixed plots and a timed-search protocol at monitoring locations between 1985 and 2013 (Table 1; Figure 1). We compiled 1,986 archeological black abalone shell measurements from 26 ancient or historical archeological sites spanning 10,000 years. Additional measurements were available, but excluded based on their small sample size (n < 15).
We included only measurements from whole or nearly whole shells and for which total length could be accurately estimated. Shells were measured in the field and in the lab, from a combination of excavated assemblages and shells eroding from archeological sites. For TA B L E 1 Sample size and black abalone length measurements for temporal periods and sea surface temperature intervals a modern n = 20,495) and cold intervals (9,600-8,200 BP, n = 175; 6,300-5,900 BP, n = 25; 3,800-2,900 BP, no data; and 1,500-500 BP, n = 330). Archeological and modern data for the last warm interval (500 BP to present) were treated separately to compare modern data to data from the last 10,000 years.
To compare mean shell sizes across time and SST periods, we conducted tests for homogeneity of variances using the Levene tests, and separate one-way analyses of variance (ANOVAs), followed by Games-Howell post hoc tests to determine if the F I G U R E 1 Location of the Santa Barbara Channel region, southern California, and the Northern Channel Islands. Inset map of San Miguel Islands and the locations of archeological sites and modern monitoring locations used in this study differences between each temporal period and/or each SST interval were statistically significant. To estimate the amount of variation in abalone shell sizes that were due to differences among the three temporal periods of the Holocene (Early, Middle, Late) versus simple natural variation within each temporal period, we estimated the magnitude of effects (ω 2 ) associated with each factor according to Graham and Edwards (2001). Here, we considered temporal period a fixed factor and used the average sample size for the three periods in the calculations. We then estimated how the importance of temporal period changed by repeating this procedure while including shell sizes from the Historic Period. To compare population distributions through time, we conducted Kolmogorov-Smirnov (KS) tests for each pair of periods to assess differences in population distribution across temporal and SST periods (Zar, 2014). Data were plotted on histograms to visually compare distributions when significant differences were identified. Data also were evaluated on the basis of juveniles versus adults. We considered all abalone shells <50 mm long to be juveniles and those >50 mm to be adults (Miner, Altstatt, Raimondi, & Minchinton, 2006

| RE SULTS
Unlike several other high-ranked shellfish species, black abalone sizes (mean length) remained relatively stable throughout the Native American fishery of the Holocene (Erlandson et al., 2008; see Table 1, Figure 2 Figure   S1, Table S1), visual inspection of the data suggested no clear trends were evident among periods (Supporting Information Figure S1). We believe this was due to our large sample sizes (100s) and the ability to statistically resolve very small differences in shell sizes. We Specifically, they were the least variable (σ = 19.315) and skewed toward larger abalone sizes, peaking between ~125 and 150 mm.
Lastly, when we examined abalone during the Modern Period, we found them to be the most variable (σ = 29.945) and appeared more uniform in distribution, although they did tend to skew slightly toward larger sizes, peaking between 100 and 150 mm (Figure 3). For a summary of data organized by archeological site and/or locus, see Erlandson et al. (2008).
Black abalone sizes also varied among SST intervals, with larger sizes observed during warmer SST periods (Figure 4). These  Table 2). Further, abalone size distributions also differed significantly between each successive SST interval, except for the first two (KS tests: p < 0.01; Supporting Information Table   S3). Specifically, black abalone skewed toward smaller sizes, with peaks in abundance between 50 and 75 mm during each cold-water interval, but were largely skewed toward larger sizes, with peaks in abundance between 100 and 150 mm, during each warm-water interval ( Figure 5). The relative abundance of adult versus juvenile black abalone was dependent on temporal period and SST interval (contingency table: p < 0.01; Table 3). However, this appears more closely influenced by SST than temporal periods, with juveniles making up a higher percentage of the total abalone population during cold intervals than warm ones ( Figure 6).
During both the Historic and Modern periods, sea otters were locally extirpated and oceanographic conditions were on a warming trend, making comparisons of these datasets important.
Archeological research also has documented that historical Chinese abalone fishermen focused on adult black abalone (see Braje, 2016).
We, therefore, excluded individuals <50 mm in length from the historical and modern data, and modern data were then separated into 5-year intervals for detailed time-series comparison. This identified significant differences in abalone size between the Historic Period and each 5-year interval of the Modern Period (Games-Howell: p < 0.01; Table 4). Specifically, abalone sizes decreased from the Historic Period through the 1985-1989 modern interval, and then steadily increased since that time (Figure 7).

F I G U R E 3
Histograms depicting black abalone population size structure during each temporal period. Note: Systat automatically bootstraps histogram data so that the scales are the same for each and can be compared visually  Table S4). In particular, sizes observed in the historical data peak between 100 and 150 mm and are skewed toward larger individuals, as are modern data for 1985-1989 and 1990-1994. However, beginning with the 1995-1999 modern interval, the data shift to peak around 100 mm, which continues for the 2000-2004, 2005-2009, and 2010-2013. For these later intervals, the data are no longer skewed toward larger individuals, but instead trend to a more normal distribution.

| D ISCUSS I ON AND CON CLUS I ON S
Similar to previous studies (Braje et al., 2007;Erlandson et al., 2008), we identified little to no obvious temporal trends in mean black abalone size on San Miguel Island during the Early, Middle, and Late Holocene periods. Although we did detect statistically significant differences among these three time periods, we believe these were likely due to our large sample sizes (i.e., 100s), as differences among periods explained very little (15%) of the variation in abalone sizes while natural variability within each period explained 85% of the variation. Indeed, abalone sizes appeared relatively similar across the Holocene, with average shell lengths  (Erlandson et al., 2011(Erlandson et al., , 2008 during cold-water intervals than during warm internals (>100 mm).
Consistent with this, the contribution of juveniles versus adults is larger for cold-water intervals than warm intervals, but the removal of juveniles from the dataset did not alter our results. This finding is surprising as modern abalone show decreased growth rates during warm-water periods, such as ENSO events, due to decreased supplies of giant kelp (Day & Fleming, 1992;Edwards & Estes, 2006). As Day and Fleming (1992) noted, however, studies of SST on abalone growth are often impaired by short time periods and a number of factors that may influence growth such as algae and kelp availability, temperature, spawning periods, and densities of conspecifics, competitors, and predators may have played significant roles over longer time intervals.
Over longer time spans, the effects of SST on abalone growth and size may be more equivocal. Increases in abalone sizes during short cold-water intervals may result from less interspecies competition for space and food, as food consumption in some abalone species is positively related to water temperature (Britz, Hecht, & Mangold, 1997). Macroalgae are generally more productive during cold SST periods when nutrient levels are high, and abalone grow faster and are more abundant due to increased algal drift (Tissot, 1990). This can lead to their aggregation in larger groups, however, where they exhibit stacking that reduces mobility and increases competition for food (Duros, 1987). As a result, such abalone can exhibit lower growth rates and smaller overall sizes (e.g., Lloyd & Bates, 2008). In contrast, during long periods of warmer SSTs over centennial to millennial scales, some competitors and predators may be better equipped to cope with elevated temperatures than other intertidal species (e.g., Fly, Monaco, Pincebourde, & Tullis, 2012;Sanford, 2002;), though they also may be more susceptible to disease (Ben-Horin, Lenihan, & Lafferty, 2013). It is also possible that other intertidal species may be more productive during prolonged cold-water periods and present a more enticing prey than abalone, leading to decreased predatory pressure and increased size. In addition, during warm SST periods there may have been a shift in prey species due to increased availability. Study of abalone growth rates and survivorship under typical short-term biological experiments and observations is unlikely to identify such patterning. Our data indicate very different trends when observing changes in black abalone size over longer timescales, underscoring the importance of understanding the aggregation of black abalone, in addition to individual size, for interpreting population health.
Although there seems to be a shift in modern black abalone size structure toward distributions similar to those in ancient times, modern populations still skew in comparison to prehistoric patterns, which appear to have been relatively stable for ~10,000 years until population collapse in the late twentieth century. Our results suggest that baselines constructed from historical commercial catch records alone are not sufficient for the long-term management of abalone. Abalone populations increased greatly during the historical period, reaching sizes and densities not seen during the previous, and seemingly more stable, deep history of Native American black F I G U R E 7 Mean black abalone size (length, mm) through Historical Period and modern 5-year intervals abalone fishing. These data provide a measure against which modern population distributions can be evaluated and offer an opportunity to build management protocols based on human-abalone ecodynamics over the longue durée.

Thanks to Chris Foote and Johan Eklöf and the production team at
Ecology and Evolution for all their help in the review, revision, and publication of our manuscript. We also thank two anonymous reviewers for their helpful suggestions on an earlier version of our manuscript and our home institutions for their continued support of our research.

CO N FLI C T O F I NTE R E S T
None declared.

AUTH O R CO NTR I B UTI O N S
HH and TB conceptualized and drafted the manuscript. ME, JE, and SW rewrote and drafted sections, edited, and shared essential data.