Research

INTRODUCTION

Indigenous communities of the Pacific Northwest possess deeply rooted management systems for coastal resources spanning from mountaintops (Deur and Turner 2005, Turner et al. 2011) to mid-elevation coastal rainforests and meadows (Turner and Turner 2008), to coastal estuaries (Deur et al. 2013) and intertidal regions (Augustine and Dearden 2014, Lepofsky et al. 2015). Coastal communities in British Columbia steward traditional foods through active management activities in intertidal environments called sea gardens (here used to acknowledge the complexity of these features beyond clam production; Groesbeck et al. 2014, Mathews and Turner 2017). Sea gardens increase the area of suitable clam habitat and in some cases create soft substrate beaches over previously rocky intertidal habitats (Lepofsky et al. 2021). Sea gardens can be found from Alaska (Moss and Wellman 2017) along the Northwest Pacific Coast (Groesbeck et al. 2014, Deur et al. 2015) to the Southern Gulf Islands in British Columbia (Augustine and Dearden 2014). Sea gardens are characterized by a rock wall built in the lower intertidal zone (Lepofsky et al. 2015). The rock wall traps sediment, resulting in a higher rate of sediment accumulation compared to non-walled beaches (Neudorf et al. 2017). The sea garden rock wall is designed to be at the precise tidal height to support clams that are traditionally harvested for consumption. At the same time, the rock walls also reduce the slope of the beach (Lepofsky et al. 2015). These beach modifications have been correlated to significant impacts on the productivity of littleneck clams (Leukoma staminea) in Quadra Island, British Columbia, increasing clam abundance, biomass, and growth rate compared to nearby non-walled beaches (Groesbeck et al. 2014, Jackley et al. 2016, Cox et al. 2019). Additionally, infaunal community diversity increases compared to non-walled beaches (Cox et al. 2019). Based on lessons shared by traditional knowledge holders, active management is a vital aspect of sea gardens (Deur et al. 2013), but active management has decreased in recent decades (Deur et al. 2015, Cox et al. 2019). Active management includes the addition of shell material, tilling of sea garden sediments to increase aeration, size-selective bivalve harvest, and repairing and maintaining sea garden rock walls (Deur et al. 2015). Unfortunately, most sea gardens have seen decreasing management and disuse, as intertidal spaces are currently contended with settler usage (Deur et al. 2013, Silver 2014).

As the tide changes toward increased Indigenous cultural revitalization and reactivation, sea gardens are increasingly being actively managed again (Augustine and Dearden 2014). This Indigenous-led revitalization reconnects people with traditional practices through the application of traditional ecological knowledge, often in partnership with scientific monitoring programs and joint community-management efforts (Berkes et al. 2000). Partnerships between First Nations and the scientific community can help bring mechanistic understanding to how sea gardens increase the condition of bivalves. This can be done by using dietary biomarkers (specifically, stable isotopic and fatty acid analysis) to determine if sea garden bivalves are responding to different food availability or feeding opportunities.

Fatty acid analysis is a tool that can identify the dietary relationship between producers and consumers (Dalsgaard et al. 2003, Budge et al. 2006, Iverson 2009, Kelly and Scheibling 2012). This analysis is based on different primary producers having unique fatty acids and differing carbon back bone length and double bond patterns, which are then transferred to consumers who feed upon them. An advantage of this approach is that the relative proportion of fatty acids within a consumer integrates longer dietary timeframes (weeks to months) and has reduced sampling biases compared to traditional stomach-content dietary analysis (France 1995, Kang et al. 1999, Iverson et al. 2004). Fatty acid analysis has been shown to be effective in identifying use of different primary producers by bivalves (Shin et al. 2008, Zhao et al. 2013). Of particular note for sea gardens, different dietary markers are well established for different plankton groups (Dalsgaard et al. 2003, Zhao et al. 2013), bacteria (Zhao et al. 2013), and terrestrial organic matter (Budge and Parrish 1998), which are likely food sources for clams in intertidal habitats.

A complementary approach to fatty acid analysis is stable isotopic analysis. In this context, different primary producers can have unique ratios of Carbon-12 to Carbon-13 (when reported in reference to the standard BeeDeeBelimnite it is referred to as δ¹³C) and this ratio is passed on, relatively unaltered, to consumers. Similarly, the ratio of Nitrogen-15 to Nitrogen-14 (δ¹⁵N when reported in reference to the atmospheric ratio of these two isotopes) is also set by the primary producer but modified by each successive trophic step as the lighter isotope (N-14) is preferentially excreted. As a result, these δ¹³C and δ¹⁵N provide a measure of food web complexity, trophic level, and, increasingly, overall evenness and range of dietary sources. These have been increasingly quantified through the “Layman” metrics (Layman et al. 2007) that use an “isotopic niche” to quantify the overall species variability in dietary uses. In addition, the quantitative uptake of these isotopes has allowed application of stable isotope mixing models to estimate the relative contribution of different dietary sources in ecological studies (Parnell et al. 2013, Stock et al. 2018).

The purpose of this project is to take a diet-based approach to understanding the ecological mechanisms behind the repeated observations of increased biomass, growth rate, and density of bivalves in sea gardens (Groesbeck et al. 2014, Jackley et al. 2016). Specifically, this project will test the hypothesis that sea garden and non-walled bivalves will vary in shell-weight condition on the basis of differences in diet, trophic position, and environmental stressors.

METHODS

Study area

Kanish Bay, on the northwest side of Quadra Island, British Columbia, has a rich history of use by the We Wai Kai First Nation and the Laich-kwil-Tach Treaty Society, including, but not limited to, sea gardens (Groesbeck et al. 2014, Neudorf et al. 2017, Smith et al. 2019, Toniello et al. 2019). These sea gardens have been radiocarbon dated to approximately 3500 years before present (Smith et al. 2019). Sea gardens have a remarkable diversity of shapes throughout the Pacific Northwest (Lepofsky et al. 2015), including within Kanish Bay (Smith et al. 2019).

Field parameter collections

Eight beaches, including four sea garden sites, were selected for study on the basis of the presence of L. staminea throughout Kanish Bay. Two tidal heights, low (0.5–0.7 m) and high (1.0–1.5 m) relative to the lowest low water large tide, were sampled along 10 m transects at each of the beaches. Transect placement was based on optimal L. staminea habitat between 0.5 to 1.5 m (Groesbeck et al. 2014). We collected clams from each transect by digging a 1 m by 10 m trench to a depth of 0.25–0.5 m below the surface. The sampling goal for L. staminea was seven individuals per transect, and a total of 224 clams were collected and processed. These clams were used for measuring their shell-weight index and their fatty acid methyl ester and stable isotope concentrations. Minimum clam shell length was set to 40 cm, from the longest section of shell, utilizing traditional knowledge of minimum harvestable size by Indigenous community members.

Kanish Bay has a large proportion of sea garden sites, with as much as 35% of shoreline in the bay altered from as long as 3500 to 4500 years ago and maintained to the present day (Lepofsky et al. 2021) by Indigenous communities in the region. Sea garden sites for our study were selected randomly throughout the bay within that altered shoreline. Sea gardens sitting within ideal modern bivalve ranges today are indicative of recent anthropological use (Smith et al. 2019), with older gardens sitting outside bivalve tidal ranges. The site selection process additionally included checking that sea garden sites contained L. staminea within ideal bivalve tidal ranges (Groesbeck et al. 2014) today. If a randomly selected site fell out of range or did not contain target species, we re-selected from the available altered shorelines. Non-walled beaches were additionally selected randomly, after field verification that they contained no walled features and had a presence of L. staminea within ideal bivalve tidal ranges (Groesbeck et al. 2014). Selections were completed by using randomly generated points in ArcGIS Pro 2020 within known altered wall areas and outside of altered wall areas within the bay.

To characterize the biophysical environment, we sampled temperature as well as water and sediment parameters at each of the sites. Transect level temperature was determined from 24 May to 13 July 2019 by using Hobo Tidbit V2 (+/- 0.2 [°C]) temperature loggers recording at 6-minute intervals. Three sediment cores were collected randomly along the transects to characterize the chlorophyll-a (chl-a) concentration within the sediment. To do this a 5 ml core was collected. These samples were freeze dried and processed using acid-verification chlorophyll and phaeopigment analyses (Caspers 1985) by using a calibrated Turner Designs TD700 fluorometer with a daylight white lamp, 340–500 nm bandpass excitation filter and > 665 nm sharp cut emission filter. Fluorescence values from the chl-a extractions were converted into µg per milliliter of chl-a, as well as µg per milliliter of phaeopigments present in each sample. We characterized the water environment at high tide by collecting water temperature (°C), salinity (PSU), and dissolved oxygen (mg/L) using a calibrated YSI Pro2030 at 14 stations. Because we aimed to understand the environment at high tide, these were collected over inundated clam beaches (depth > 2 m) and at six near-shore sites (depth 10–43 m) in Kanish Bay (Fig. 1). We sampled at 0.1, 1, 2, and 3 m water depth at each near-shore site and at the surface (0.1 m) and bottom (variable by tidal height but generally around 1 m) at the sea gardens clam beds.

Potential dietary endpoints were quantified by collection of sediment, benthic algae, and particulate organic matter to parameterize a dietary proportion model; samples with similar stable isotope ratios and tidal height were grouped for increased resolution in this model to maximize ecological relevance (Fry 2006). High tide algae contained a combination of Fucus spp. and Ulva spp. Low tide algae sources were a combination of Nereocystis luetkeana, Odonthalia floccosa, and Laminaria nigripes.

We collected particulate organic matter (POM) to characterize a potential oceanic food source (including planktonic and detrital) for the clams. This was done by using a battery-operated peristaltic pump that filtered 1 L of seawater through a 47 mm combusted glass fiber filter (GFF). These samples were collected at 2 m depth for near-shore sites and at the surface and bottom of inshore sites. In addition, 500 ml of seawater was filtered through 47 mm GFFs at all sampling locations and depths to determine chl-a concentrations and analyzed as described above. For stable isotope characterization of the POM, GFFs were freeze-dried, pulverized, and subsampled to a target mass of 125–150 µg following University of California Davis Stable Isotope Facility processing protocols. Samples were processed for δ¹³C and δ¹⁵N isotopes by using a PDZ Europa ANCA-GSL elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer. Samples were run against in-house standards with a mean absolute accuracy for δ¹⁵N of ± 0.06‰ and δ¹³C of ± 0.04‰ (Sercon Ltd., Cheshire, UK).

Leukoma staminea collection and parameter processing

We used a shell-weight condition, which is the ratio of all soft tissue to shell, as a proxy for bivalve health. Shell-weight ratios as well as internal volume measurements are common condition indices for bivalves (Rainer and Mann 1992). We utilized a shell-to-dry tissue condition index for this study because of decreased error in repeated sampling and increased accuracy as determined by Rainer and Mann (1992), providing a metric to compare intraspecies clam health. This shell-weight condition index was calculated on the basis of Rainer and Mann (1992) and further tested by Filgueira et al. (2013) by using the following equation:

Shell-Weight Condition Index = 1000* (Dry Tissue Weight (g)/ Dry Shell Weight (g))

To identify the diet of the L. staminea, fatty acid methyl ester (FAME) composition was quantified for the L. staminea anterior adductor muscle. FAMEs were extracted by using a one-step extraction-transesterification method (sensu Lewis et al. 2000). Extracted samples were stored at -20 °C and analyzed by gas chromatography-flame ionization detector (GS-FID) on a Thermo 1310 with a TR-FAME (cis/trans isomers and fatty acid methyl ester) 10 m x 0.1 mm x 0.2 µm I.D. column. All sample chromatograms were corrected for any peaks that appeared in the blank. Peaks were identified in relation to the Supelco 37-FAME standard, which is a 37-component FAME (Sigma Aldrich) mix that can be used to identify key fatty acid methyl esters in many tissue sample types. Data were processed for relative percent area for all fatty acid peaks per clam sample. Stable isotope tissue samples were taken by removing 1.50 mg of the posterior adductor muscle after 24 hours of desiccation at 60 °C. Samples were analyzed at University of California, Davis, Stable Isotope Facility as specified above.

Trophic level metrics analysis

We used the Stable Isotope Bayesian Ellipses in R (SIBER) method (Jackson et al. 2011) in R Studio 2022.07.1 Build 554 to generate credible interval ranges around six Layman et al. (2007) metrics to quantify the clams’ dietary niche. We calculated the total dietary range as quantified by the δ¹⁵N range (NR) and δ¹³C range. Total area (TA) encompassed by stable isotope species in δ¹³C - δ¹⁵N biplot space is akin to niche space occupied per grouping variable and is related to food web trophic diversity (Layman et al. 2007). Mean Euclidean distance of each species to the δ¹³C - δ¹⁵N centroid (CD) per grouping variable relates to trophic diversity in samples. Nearest neighbor distance (NDD) indicates species density by grouping variables and standard deviation of nearest neighbor distance (SDNND) relates to evenness of species distributions (Layman et al. 2007). This method allowed for corrections of known layman metrics generated by unequal sample sizes (Jackson et al. 2011), because of differences in viable sample numbers between sea garden samples (n = 114) and non-walled samples (n = 123). Comparisons between groups were conducted through estimated posterior distribution of layman metrics and can be used to review probability of differences between convex hull area (TA) generated per group, to examine trophic space occupied per community.

Dietary proportion analysis

L. staminea δ¹³C - δ¹⁵N stable isotope ratios allowed for estimation of dietary proportion of generalized bivalve dietary sources. Dietary proportions were estimated by using stable isotope mixing models in R (SIMMR), using site type bivalve isotopic data and aggregated sources per trophic enrichment factor (TEF) scenario within a Bayesian framework (Markov Chain Monte Carlo), to give estimated contributions to littleneck clam (L. staminea) dietary proportion. We used two scenarios to capture the range of dietary proportions in each mixing model, focusing on trends between scenarios as main dietary proportion results. Scenario one used a whole-body TEF estimate (δ¹³C mean and SD: 1.85 ± 1.1; δ¹⁵N mean and SD δ¹⁵N: 3.79 ± 1.1).

Scenario two used a muscle-tissue specific TEF estimate (δ¹³C mean and SD: 2.90 ± 1.1; δ¹⁵N mean and SD δ¹⁵N: 2.90 ± 1.1) available for a filter-feeding bivalve (Crassostrea gigas; Yokoyama et al. 2008). The trophic enrichment factors for scenario two led most samples from one site (NW-01) to fall outside of the standard deviation food source polygon, and so this site was not included in scenario two.

Statistical analysis

We used ordination methodologies to explore the differences in clam physiology between sea garden sites and non-walled sites based on FAME profiles, condition indices, and oceanic and site sediment parameters. Non-metric multidimensional scaling (nMDS) analysis was used to assess site types differences in clam FAME profiles. Bray-Curtis semi-metric dissimilarities were used to quantify the differences in FAME variable profiles between each site, in order to account for percentage measurements for FAME profiles as well as account for abundances of variables in the ordination.

Fatty acid methyl esters that occurred fewer than five times in all clam samples (rare FAMEs) were removed from the nMDS analysis to limit undue weighting of rare species (Warton et al. 2012). Mann-Whitney U-test was used to test for differences between sea gardens and non-walled sites in relevant fatty acid biomarker groups’ relative percent areas, as a non-parametric alternative to a Student’s T-test (McKnight and Najab 2010). This was done for main fatty acid biomarkers (bacterial markers) and groupings by type (saturated fatty acids, monounsaturated fatty acids, and polyunsaturated fatty acids) to examine main profile differences that relate to diet and overall FAME profile shifts by site type.

Distance-based redundancy analysis (db-RDA), as discussed by McArdle and Anderson (2001), was used to assess how physiological and environmental variables correlated to condition indices. Included environmental variables in the db-RDA were minimum chl-a, minimum salinity, minimum dissolved oxygen (mg/l), median ranked percent algal cover (percentage), sediment chl-a range (µg/ml), mean sediment chl-a (ug/l), and mean phaeopigments (µg/ml). Distance-based redundancy analysis was run by using Gower’s distance, which generated a distance matrix that was run through a principal coordinates analysis (PCoA), from which eigenvalues can be extracted and run in an RDA (Legendre and Legendre 2012). The resulting RDA shows ordinations of environmental and physiological variables in the model and can be examined for significant correlations in axis and environmental variables.

RESULTS

Similarities between site types

Environmentally available fatty acids were captured through inshore and midshore samples (Fig. 1) to review potentially present fatty acid groups prior to uptake in bivalve tissues (Table 1). Inshore seawater samples were dominated by dinoflagellate fatty acid signatures, determined by C16:1(n-7) and C16:0 ratios less than 1 (Zhao et al. 2013), based on ratios of the 0.12 and 0.02 for sea gardens and non-walled beaches, respectively. Terrestrial biomarkers, FAME C18:2(n-6) and C18:3(n-3), made up less than 1% relative percent area of all FAMEs.

Dietary proportion models

Oceanic food sources were indicated to be the primary dietary source for all sites, as determined by 13C and 15N isotope ratios (mean ± SD) in both three-source mixing models generated (Table 2). Leukoma staminea δ¹³C and δ¹⁵N isotopic measurements included in the SIMMR models resolved dietary proportion estimates for both scenarios (scenario one, n = 194; scenario two, n = 71) out of 224 bivalve samples. Scenario two (muscle-tissue) results indicated food sources did not resolve as accurately with the current collected basal food webs compared to scenario one, with multiple bivalve samples falling out of source isotope ranges. A large proportion of samples were excluded from the model in scenario two, indicating limited fit and confidence in the model dietary proportion estimates additionally. Scenario two would benefit from increased food sources to address resolution challenges.

Bivalve condition indices

We did not find a significant difference between L. staminea condition indices from sea gardens versus non-walled sites (Mann-Whitney U-Test, U = 2061, P = 0.255, n = 137). This suggests that there are no significant differences in condition indices by site type. Sea garden bivalves had a smaller range of condition indices (range = 150.61) compared to non-walled bivalves (207.93). Mean condition indices were relatively similar by site type, with sea garden bivalves averaging 161.5 ± 4.70 (± SE), and non-walled bivalves averaging 167.8 ± 5.18.

Trophic isotopic differences and shifts by site type

Comparison of standard ellipse area by community (sea garden isotopic signatures against non-walled isotopic signatures from L. staminea tissues) found non-walled L. staminea to have a 76.7% probability of occupying a lower convex hull area than sea garden L. staminea (Fig. 2). Centroid difference (CD, Fig. 2) additionally showed increased trophic diversity in sea garden bivalve isotopes, potentially related to shifted diets. Last, in sea gardens, both mean nearest neighbor distance and standard deviation of nearest neighbor distance (Fig. 3) indicated less evenness between samples compared to lower value non-walled samples. Sea garden L. staminea had an increased δ¹⁵N range (0.82‰) compared to non-walled samples (0.12‰). Differences in the range of δ¹⁵N (‰) and δ¹³C (‰) in sea garden clams show that at walled sites there is greater overall variability, including individuals feeding at a higher trophic position (δ¹⁵N > 11.5‰) than at non-walled sites. Much of the difference in food web complexities are obfuscated when looking at the means and thus the ranges become increasing informative to identify differences between clam diets in the two environments.

Bivalve fatty acid markers

Sea garden bivalve fatty acid profiles showed decreased C16:1(n-7), C15:1, C18:0, and C18:1(n-9) FAME species and increased in C16:0, i16:0, i17:0 and C17:0 FAME species as main drivers for ordination differences (Table 3). Saturated fatty acids were found to be significantly higher in sea garden bivalves (n = 75) compared to those in non-walled sites (n = 62; Mann-Whitney U-Test, W = 3447, P ≤ 0.01; Table 3), whereas monounsaturated fatty acids were significantly lower in sea garden bivalves compared to non-walled bivalves (Mann-Whitney U-Test, W = 1202, P ≤ 0.01, a = 0.05).

These differences by fatty acid groups are important factors as visualized in an nMDS ordination, indicating partial site separation by fatty acids (Fig. 4a, b), although there were additional groupings in approximately -0.5 to -0.5 that appeared to separate a moderate number of samples by sampling month. However, the primary cluster of points within the nMDS within dimension 1 and 2 (Fig. 4a) were from all months in this study, and the smaller cluster around -0.5 to 0.5 contained samples from multiple sites, excluding an individual site effect. Overall, grouping appeared to be related to site type with a minor monthly effect, but proportion of effect is not clearly discerned.

Biomarkers indicated the potential presence of dinoflagellates in the diet of bivalves (Langdon and Newell 1990, Kharlamenko et al. 2001, Zhao et al. 2013) from both sites, with 16:1(n-7) and C16:0 ratios averaging less than 1, a ratio in sea garden bivalves of 0.09, and a ratio at non-walled site bivalves of 0.18, although there was a distinct lack of 22:6(n-3) to increase confidence in total dinoflagellate contribution to diet profiles. Increases in C16:1(n-7) toward the non-walled sites could be indicative of a mixed diet of diatoms or other photosynthetically derived algal food sources, which are considered diatom-dominated when 16:1(n-7) and C16:0 ratios are greater than 1 (Shin et al. 2008, Zhao et al. 2013). However, we do not know which specific food particulates affect the 16:1(n-7) and C16:0 ratios observed, since 16:1(n-7) and C16:0 are common marine fatty acid species present in many algae and planktonic sources in marine systems (Langdon and Newell 1990, Viso and Marty 1999, Dalsgaard et al. 2003).

Correlated physiological differences and environmental variables

Bivalve physiology and dietary metrics distinguished site differences with condition index, with sea gardens showing less variability correlated with increased condition. The RDA model included 137 clam samples for which we had data for all variables and individual samples included FAME profiles, shell-weight condition indices, and shell thickness. Distance-based redundancy analysis was found to generate a statistically significant model (Pseudo-F5, 131 = 2.84, P ≤ 0.05; Fig. 3) between bivalve condition indices, FAME species, and shell thickness against minimum salinity (PSU), minimum dissolved oxygen (mg/l), mean sediment chl-a (ug/l), and mean sediment phaeopigments (µg/l). Largest variation significantly explained by environmental variables were minimum dissolved oxygen (Pseudo-F1, 131 = 4.12, P ≤ 0.05), mean sediment phaeopigments (Pseudo-F1, 131 = 3.98, P ≤ 0.05), minimum salinity (Pseudo-F1, 131 = 2.57, P ≤ 0.05) and mean sediment chl-a (Pseudo-F1, 131 = 2.42, P ≤ 0.05). Bivalve condition index had a smaller pull in the ordination and was correlated in the ordination with higher minimum dissolved oxygen, higher minimum salinity, mean sediment phaeopigments, and higher minimum seawater chl-a. Overall, condition indices overlapped by site type, represented in Figure 3, but increased toward the centroid of sea garden bivalve species scores in the db-RDA.

DISCUSSION

This study identified moderate differences in L. staminea fatty acid and stable isotopic composition between non-walled and sea-garden sites. Diet is a primary driver in fatty acid and stable isotopic composition; however, differences in diet can be the result of physiological stress, inducing changes in feeding behaviors. For example, L. staminea under osmotic stress have been observed to have altered feeding behavior (Talkington 2015). Changes in feeding behavior between sites could explain the differences in fatty acid profiles within our db-RDA, and further study of these mechanisms should be reviewed as potential hypotheses generating differences between sea gardens and non-walled beaches. These results are supported by grouping within the nNMDS model (Fig. 2a, b), moderate separations within isotopic bi-plots (Fig. 5) and Bayesian posterior ellipses (Figs. 2 and 3), and the correlations resolved db-RDA model (Fig. 6), although site type appears to only partially separate sites across the dietary metrics examined.

Dietary FAME comparisons and differences by site

Both dietary proportion models indicated that ocean-based dietary sources make up a larger proportion of L. staminea diets for both site types, compared to tidal algae detritus sources. Dietary proportion is supported by our FAME profile results, indicating increased dinoflagellates and limited marine POM signatures (PUFAs), although non-walled sites potentially showed some increased diatoms in the ratio of C16:0 and C16:1(n-7).

The fatty acid species that differentiated clams in the two different sites are common FAME biomarkers in bivalves (Langdon and Newell 1990, Kharlamenko et al. 2001, Zhao et al. 2013). Overall, there appeared to be overlap between site-type dietary FAME profiles (Table 3). The main dietary FAME drivers were primarily dinoflagellate biomarkers. Importantly, the main saturated fatty acid (SFA) profiles and mono-unsaturated fatty acid (MUFA) profiles are statistically different, with sea gardens having higher overall SFAs and lower MUFAs compared with non-walled sites. Dinoflagellate consumption by L. staminea is supported by dinoflagellate associated biomarkers found in clams and the surrounding seawater, and matches known contributions of phytoplankton and flagellates to filter-feeding bivalve diets (Cranford and Hill 1999, Cranford et al. 2011).

Alternative drivers of biomarker patterns: stress, not food

Isotopic trophic shifts in marine species have been observed when switching feeding behavior (Fry 2006), changing life stages and size (Hentschel 1999), physiological constraints (Rossi et al. 2004), and availability of food particulates (Rossi et al. 2004). In this study, Layman metrics (Figs. 2 and 3) showed an overlap in both δ¹⁵N and δ¹³C profiles, although individual sea garden bivalves had overall higher δ¹⁵N values, indicating that certain individuals are feeding at a higher trophic niche. We saw no difference in available food sources in our seawater sample, indicating that the potential food sources are evenly distributed throughout the study region. Thus, bivalve FAME profile differences between sites are likely being driven by selective feeding or physiological stress. Although differences in FAME profiles can also be driven by ontogenetic factors, these were minimized in this study by selecting only individuals of similar size, which should minimize challenges associated with tissue turnover and changes in feeding selectivity not driven by the environment itself. In addition, although different species can also have divergent assimilation and biomarker profiles, all bivalves were positively identified by multiple investigators to ensure only L. staminea were included rather than similar looking species (e.g., Venerupis philippinarum).

Variation of food availability was not detected on the basis of FAME and stable isotope profiles; additionally, the study site is relatively small with strong tidal mixing. Potential diet profiles indicated by FAME profiles of ocean filters and estimated food proportion uptake in our three-source dietary mixing models do not support different food particulates occurring by site type. Instead, differences in trophic niche positioning between sea garden L. staminea samples and non-walled samples are expected to be the result of physiological constraints between site types, differences in feeding behaviors, or some combination of both.

Suggested mechanism for sea garden bivalve fatty acid and trophic shifts

Here we propose that the impact of sea gardens on clam diet is driven by altered feeding as a result of differential salinity and oxygen patterns, influenced by wall presence. Increased trophic positioning in sea garden L. staminea samples, as well as reductions in food diversity, signaled by increased δ¹⁵N and decreased δ¹³C levels, could suggest both feeding behavioral shifts and decreased physiological stressors in sea garden sites. Talkington (2015) found that L. staminea reduced key osmolytes involved in osmotic regulation under stress caused by low salinity. This reduction of L. staminea osmolytes under stress in turn caused lower available energy from catabolic processes. Salinity stress also has strong behavioral effects on L. staminea, including shell closing (Talkington 2015), which negatively affects optimal feeding and respiration in intertidal organisms (Vernberg 1969).

Consistent with our expectations, both salinity and dissolved oxygen vectors in our db-RDA were found to have moderate correlations with condition indices in sea gardens. L. staminea has previously been classified as a stenohaline species, with decreased cilia activity reduced under 20 parts per thousand (ppt), in comparison to 30 ppt salinity (Talkington 2015). L. staminea are sensitive to hypoxic conditions and exhibit increased mortality compared to other endemic bivalve species under low oxygen conditions (Allee 2010), and it has been suggested that synergistic stressors, such as low salinity, would lower hypoxic tolerance (Allee 2010).

Our results of FAME profile shifts of SFAs and MUFAs are consistent with salinity driven responses observed in other systems (Navarro and Gonzalez 1998, Tomanek 2012, Gonçalves et al. 2017). Sea garden bivalve samples were also ordinated more closely compared to spread non-walled sites, within the db-RDA (Fig. 6). This indicates that sea gardens are more similar to each other and generate similar dietary FAME proportions consistently over time compared to non-walled site bivalve samples, which are highly variable in their diet. It appears that sea gardens are dominated by a more consistent diet and environment, leading to expected stress-alleviated fatty acid proportions seen in marine invertebrate stress studies (Navarro and Gonzalez 1998, Tomanek 2012, Gonçalves et al. 2017).

Alleviation of stressors by increasing salinity and dissolved oxygen are expected to generate different feeding behaviors in sea gardens, as a response to more optimal conditions. We saw in the db-RDA (Fig. 6) that increased minimum chl-a and mean sediment phaeopigments were correlated more strongly with condition indices and with SFAs commonly found in marine plankton. The fatty acid species that positively correlated most strongly with minimum chl-a, mean sediment phaeopigments, and increased condition indices were C16:0 and C18:0 SFAs. Both are general planktonic fatty acid biomarkers, although C16:0 is associated specifically with dinoflagellates or diatoms, according to the ratio of C16:1(n-7) over C16:0. Sea gardens have been found to increase residence times for seawater (Salter 2018), which is related to increased growth rates in bivalves. Decreased salinity stress behaviors seen in L. staminea, such as shell closing (Talkington 2015), would allow them to feed for longer with greater efficiency in waters with higher residence times. In non-walled sites, there are also correlations with increasing sediment chl-a associated with benthic photosynthetic algae, such as diatoms, as well as increases in C16:1(n7). These correlations could be indicative of less optimal feeding behaviors, potentially related to increased stress as well as lower residence times seen over non-walled beaches (Salter 2018).

Both increasing stressors in non-walled bivalves and increasing optimal feeding behaviors in sea garden bivalves would appear as a trophic shift noted in this study. This study suggests that sea gardens are more environmentally stable, allowing for less variability in dietary intake, by decreasing known stressors for L. staminea and allowing for increased feeding efficiencies.

CONCLUSION

Differences in L. staminea bivalves of Kanish Bay indicated potential mechanistic explanations for partial trophic shifts by site type, supported by fatty acid marker differences by FAME group, although differences between site types had overlap. These differences appeared to match expected shifts generated by decreased stressors and potential behavioral shifts in sea garden bivalve feeding compared to non-walled bivalves, and future studies could focus on clarifying the hypothesis generated by this exploratory study.

Increased condition indices were strongly correlated with increased chl-a and increased sediment phaeopigments, and were moderately correlated with higher minimum salinity and dissolved oxygen. Correlated relationships with increased condition indices occurred specifically on sea garden sites, compared to non-walled sites, suggesting increased productivity and lower salinity stressors on sea gardens compared to non-walled sites. These related to previous physiological studies of sea gardens under stress, as well as observed increased seawater retention time in sea gardens in previous studies (Salter 2018).

We found reasonable correlations to suggest both feeding behavior shifts and reduced stress factors in sea garden bivalves compared to non-walled clams, which are supported by fatty acid shifts and trophic shifts matching previous literature. Overall, correlated relationships to significant environmental drivers within our db-RDA provide plausible avenues to explain observed fatty acid shifts in saturated fatty acids and monounsaturated fatty acids in sea gardens, as well as trophic shifts in δ¹³C and δ¹⁵N stable isotopes.

Future restoration programs should consider measurements of algae levels over sea gardens in relation to dietary availability as well as trophic shift measurements as ways to examine progress of restoration. Utilizing fatty acid measurements for dietary biomarkers as well as stable isotope ratios for dietary proportion and trophic positioning measurements on bivalves were shown to be effective in identifying correlated environmental drivers in this study. As restoration continues on sea garden sites, ecological driving forces leading to observed shifts in this study, as well as other mechanisms for clam stress, feeding, and recruitment strategies, should be examined to determine efficacy of restoration policies as well as create target restoration goals.

LITERATURE CITED

Allee, R. 2010. The effects of anoxic conditions on thermal tolerance and stress protein levels in four local bivalves. Thesis. Western Washington University, Bellingham, Washington, USA. https://doi.org/10.25710/mz7z-1740

Augustine, S., and P. Dearden. 2014. Changing paradigms in marine and coastal conservation: a case study of clam gardens in the Southern Gulf Islands, Canada. Canadian Geographer 58(3):305-314. https://doi.org/10.1111/cag.12084

Berkes, F., J. Colding, and C. Folke. 2000. Rediscovery of traditional ecological knowledge as adaptive management. Ecological Applications 10(5):1251-1262. https://doi.org/10.1890/1051-0761(2000)010[1251:ROTEKA]2.0.CO;2

Budge, S. M., S. J. Iverson, and H. N. Koopman. 2006. Studying trophic ecology in marine ecosystems using fatty acids: a primer on analysis and interpretation. Marine Mammal Science 22(4):759-801. https://doi.org/10.1111/j.1748-7692.2006.00079.x

Budge, S. M., and C. C. Parrish. 1998. Lipid biogeochemistry of plankton, settling matter and sediments in Trinity Bay, Newfoundland. II. Fatty acids. Organic Geochemistry 29(5-7):1547-1559. https://doi.org/10.1016/S0146-6380(98)00177-6

Caspers, H. 1985. Timothy R. Parsons, Yoshiaki Maita and Carol M. Lalli: A manual of chemical and biological methods for seawater analysis. Internationale Revue der gesamten Hydrobiologie und Hydrographie 70(6):903-904. https://doi.org/10.1002/iroh.19850700634

Cox, K. D., T. G. Gerwing, T. Macdonald, M. Hessing-Lewis, B. Millard-Martin, R. J. Command, F. Juanes, and S. E. Dudas. 2019. Infaunal community responses to ancient clam gardens. ICES Journal of Marine Science 76(7):2362-2373. https://doi.org/10.1093/icesjms/fsz153

Cranford, P. J., and P. S. Hill. 1999. Seasonal variation in food utilization by the suspension-feeding bivalve molluscs Mytilus edulis and Placopecten magellanicus. Marine Ecology Progress Series 190:223-239. https://doi.org/10.3354/meps190223

Cranford, P. J., J. E. Ward, and S. E. Shumway. 2011. Bivalve filter feeding: variability and limits of the aquaculture biofilter. Pages 81-124 in S. E. Shumway, editor. Shellfish aquaculture and the environment. Wiley-Blackwell, Oxford, UK. https://doi.org/10.1002/9780470960967.ch4

Dalsgaard, J., M. St. John, G. Kattner, D. Müller-Navarra, and W. Hagen. 2003. Fatty acid trophic markers in the pelagic marine environment. Advances in Marine Biology 46:225-340. https://doi.org/10.1016/S0065-2881(03)46005-7

Deur, D., A. Dick, K. Recalma-Clutesi, and N. J. Turner. 2015. Kwakwaka’wakw “Clam Gardens.” Human Ecology 43(2):201-212. https://doi.org/10.1007/s10745-015-9743-3

Deur, D., and N. J. Turner. 2005. Keeping it living: traditions of plant use and cultivation on the Northwest Coast of North America. University of Washington Press, Seattle, Washington, USA.

Deur, D., N. J. Turner, K. Recalma-Clutesi, A. Dick, and D. Sewid-Smith. 2013. Subsistence and resistance on the British Columbia coast: Kingcome Village’s estuarine gardens as contested space. BC Studies 179:13-37. https://doi.org/10.14288/bcs.v0i179.184182

Filgueira, R., L. A. Comeau, T. Landry, J. Grant, T. Guyondet, and A. Mallet. 2013. Bivalve condition index as an indicator of aquaculture intensity: a meta-analysis. Ecological Indicators 25:215-229. https://doi.org/10.1016/j.ecolind.2012.10.001

France, R. L. 1995. Differentiation between littoral and pelagic food webs in lakes using stable carbon isotopes. Limnology and Oceanography 40(7):1310-1313. https://doi.org/10.4319/lo.1995.40.7.1310

Fry, B. 2006. Isotope Chi. Pages 76-118. Stable isotope ecology. Springer, New York, New York, USA. https://doi.org/10.1007/0-387-33745-8_4

Gonçalves, A. M. M., D. V. Barroso, T. L. Serafim, T. Verdelhos, J. C. Marques, and F. Gonçalves. 2017. The biochemical response of two commercial bivalve species to exposure to strong salinity changes illustrated by selected biomarkers. Ecological Indicators 77:59-66. https://doi.org/10.1016/j.ecolind.2017.01.020

Groesbeck, A. S., K. Rowell, D. Lepofsky, and A. K. Salomon. 2014. Ancient clam gardens increased shellfish production: adaptive strategies from the past can inform food security today. PloS ONE 9(3):e91235. https://doi.org/10.1371/journal.pone.0091235

Hentschel, B. T. 1998. Spectrofluorometric quantification of neutral and polar lipids suggests a food-related recruitment bottleneck for juveniles of a deposit-feeding polychaete population. Limnology and Oceanography 43(3):543-549. https://doi.org/10.4319/lo.1998.43.3.0543

Iverson, S. J. 2009. Tracing aquatic food webs using fatty acids: from qualitative indicators to quantitative determination. Pages 281-308 in M. Kainz, M. T. Brett, and M. T. Arts, editors. Lipids in aquatic ecosystems. Springer, New York, New York, USA. https://doi.org/10.1007/978-0-387-89366-2_12

Iverson, S. J., C. Field, W. D. Bowen, and W. Blanchard. 2004. Quantitative fatty acid signature analysis: a new method of estimating predator diets. Ecological Monographs 74(2):211-235. https://doi.org/10.1890/02-4105

Jackley, J., L. Gardner, A. F. Djunaedi, and A. K. Salomon. 2016. Ancient clam gardens, traditional management portfolios, and the resilience of coupled human-ocean systems. Ecology and Society 21(4):20. https://doi.org/10.5751/ES-08747-210420

Jackson, A. L., R. Inge, A. C. Parnell, and S. Bearhop. 2011. Comparing isotopic niche widths among and within communities: SIBER - stable isotope Bayesian ellipses in R. Journal of Animal Ecology 80(3):595-602. https://doi.org/10.1111/j.1365-2656.2011.01806.x

Kang, C. K., P. G. Sauriau, P. Richard, and G. F. Blanchard. 1999. Food sources of the infaunal suspension-feeding bivalve Cerastoderma edule in a muddy sandflat of Marennes-Oléron Bay, as determined by analyses of carbon and nitrogen stable isotopes. Marine Ecology Progress Series 187:147-158. https://doi.org/10.3354/meps187147

Kelly, J. R., and R. E. Scheibling. 2012. Fatty acids as dietary tracers in benthic food webs. Marine Ecology Progress Series 446:1-22. https://doi.org/10.3354/meps09559

Kharlamenko, V. I., S. I. Kiyashko, A. B Imbs, and D. I. Vyshkvartzev. 2001. Identification of food sources of invertebrates from the seagrass Zostera marina community using carbon and sulfur stable isotope ratio and fatty acid analyses. Marine Ecology Progress Series 220:103-117. https://doi.org/10.3354/MEPS220103

Langdon, C., and R. I. E. Newell. 1990. Utilization of detritus and bacteria as food sources by two bivalve suspension-feeders, the oyster Crassostrea virginica and the mussel Geukensia demissa. Marine Ecology Progress Series 58:299-310. https://doi.org/10.3354/meps058299

Layman, C. A., D. A. Arrington, C. G Montaña, and D. M. Post. 2007. Can stable isotope ratios provide for community-wide measures of trophic structure? Ecology 88(1):42-48. https://doi.org/10.1890/0012-9658(2007)88[42:CSIRPF]2.0.CO;2

Legendre, P., and L. Legendre. 2012. Numerical Ecology. Pages 487-493 in B. Fath, editor. Encyclopedia of Ecology. Second edition. Elsevier, New York, New York, USA. https://doi.org/10.1016/B978-0-12-409548-9.10595-0

Lepofsky, D., N. F. Smith, N. Cardinal, J. Harper, M. Morris, E. White, R. Bouchard, D. I. D. Kennedy, A. K. Salomon, M. Puckett, and K. Rowell. 2015. Ancient shellfish mariculture on the Northwest Coast of North America. American Antiquity 80(2):236-259. https://doi.org/10.7183/0002-7316.80.2.236

Lepofsky, D., G. Toniello, J. Earnshaw, C. Roberts, L. Wilson, K. Rowell, and K. Holmes. 2021. Ancient anthropogenic clam gardens of the Northwest Coast expand clam habitat. Ecosystems 24:248-260. https://doi.org/10.1007/s10021-020-00515-6

Lewis, T., P. D. Nichols, and T. A. McMeekin. 2000. Evaluation of extraction methods for recovery of fatty acids from lipid-producing microheterotrophs. Journal of Microbiological Methods 43(2):107-116. https://doi.org/10.1016/s0167-7012(00)00217-7

Mathews, D., and N. J. Turner. 2017. Chapter 9 - Ocean cultures: Northwest Coast ecosystems and Indigenous management systems. Pages 169-199 in P. S. Levin and M. R. Poe, editors. Conservation for the anthropocene ocean. Academic Press, Seattle, Washington, USA. https://doi.org/10.1016/B978-0-12-805375-1.00009-X

McArdle, B. H., and M. J. Anderson. 2001. Fitting multivariate models to community data: a comment on distance-based redundancy analysis. Ecology 82(1):290-297. https://doi.org/10.1890/0012-9658(2001)082[0290:FMMTCD]2.0.CO;2

McKnight, P. E., and J. Najab. 2010. Mann-Whitney U Test. In I. B. Weiner and W. E. Craighead, editors. The Corsini Encyclopedia of Psychology. John Wiley & Sons, Hoboken, New Jersey, USA. https://doi.org/10.1002/9780470479216.corpsy0524

Moss, M., and H. Wellman. 2017. The Magoun Clam Garden near Sitka, Alaska: niche construction theory meets traditional ecological knowledge, but what about the risks of shellfish toxicity? Alaska Journal of Anthropology 15(1-2):7-24.

Navarro, J. M., and C. M. Gonzalez. 1998. Physiological responses of the Chilean scallop Argopecten purpuratus to decreasing salinities. Aquaculture 167(3-4):315-327. https://doi.org/10.1016/S0044-8486(98)00310-X

Neudorf, C. M., N. Smith, D. Lepofsky, G. Toniello, and O. B. Lian. 2017. Between a rock and a soft place: using optical ages to date ancient clam gardens on the Pacific Northwest. PLoS ONE 12(2):e0171775. https://doi.org/10.1371/journal.pone.0171775

Parnell, A. C., D. L. Phillips, S. Bearhop, B. X. Semmens, E. J. Ward, J. W. Moore, A. L. Jackson, J. Grey, D. J. Kelly, and R. Inger. 2013. Bayesian stable isotope mixing models. Environmetrics 24(6):387-399. https://doi.org/10.1002/env.2221

Rainer, J. S., and R. L. Mann. 1992. A comparison of methods for calculating condition index in eastern oysters Crassostrea virginica (Gmelin, 1791). Journal of Shellfish Research 11(1):55-58.

Rossi, F., P. M. J. Herman, and J. J. Middelburg. 2004. Interspecific and intraspecific variation of δC and δN in deposit- and suspension-feeding bivalves (Macoma balthica and Cerastoderma edule): evidence of ontogenetic changes in feeding mode of Macoma balthica. Limnology and Oceanography 49(2):408-414. https://doi.org/10.4319/lo.2004.49.2.0408

Salter, N. 2018. Ancient clam gardens magnify bivalve production by moderating temperature and enhancing sediment carbonate. Thesis. Simon Fraser University, Burnaby, British Columbia, Canada. https://summit.sfu.ca/item/18596

Shin, P. K. S., K. M. Yip, W. Z. Xu, W. H. Wong, and S. G. Cheung. 2008. Fatty acid as markers to demonstrating trophic relationships among diatoms, rotifers and green-lipped mussels. Journal of Experimental Marine Biology and Ecology 357(1):75-84. https://doi.org/10.1016/j.jembe.2008.01.002

Silver, J. 2014. Shellfish and coastal change: Pacific oysters and Manila clams in BC waters. BC Studies 181:83-103.

Smith, N. F., D. Lepofsky, G. Toniello, K. Holmes, L. Wilson, C. M. Neudorf, C. Roberts. 2019. 3500 years of shellfish mariculture on the Northwest Coast of North America. PLoS ONE 14(2):e0211194. https://doi.org/10.1371/journal.pone.0211194

Stock, B. C., A. L. Jackson, E. J. Ward, A. C. Parnell, D. L. Phillips, and B. X. Semmens. 2018. Analyzing mixing systems using a new generation of Bayesian tracer mixing models. PeerJ 6:e5096. https://doi.org/10.7717/peerj.5096

Talkington, T. B. 2015. Physiological and behavioral responses to changes in salinity in the invasive bivalve Nuttallia obscurata compared to the native bivalve Leukoma staminea. Thesis. Western Washington University, Bellingham, Washington, USA. https://doi.org/10.25710/6pfj-0445

Tomanek, L. 2012. Environmental proteomics of the mussel Mytilus: implications for tolerance to stress and change in limits of biogeographic ranges in response to climate change. Integrative and Comparative Biology 52(5):648-664. https://doi.org/10.1093/icb/ics114

Toniello, G., D. Lepofsky, G. Lertzman-Lepofsky, A. K. Salomon, and K. Rowell. 2019. 11,500 y of human-clam relationships provide long-term context for intertidal management in the Salish Sea, British Columbia. Proceedings of the National Academy of Sciences 116(44):22106-22114. https://doi.org/10.1073/pnas.1905921116

Turner, N. J., Ł. J. Łuczaj, P. Migliorini, A. Pieroni, A. L. Dreon, L. E. Sacchetti, and M. G. Paoletti. 2011. Edible and tended wild plants, traditional ecological knowledge and agroecology. Critical Reviews in Plant Sciences 30(1-2):198-225. https://doi.org/10.1080/07352689.2011.554492

Turner, N. J., and K. L. Turner. 2008. “Where our women used to get the food”: cumulative effects and loss of ethnobotanical knowledge and practice; case study from coastal British Columbia. Botany 86(2):103-115. https://doi.org/10.1139/B07-020

Vernberg, F. J. 1969. Acclimation of intertidal crabs. American Zoologist 9(2):333-341. https://doi.org/10.1093/icb/9.2.333

Viso, A.-C., and J.-C. Marty. 1993. Fatty acids from 28 marine microalgae. Phytochemistry 34(6):1521-1533. https://doi.org/10.1016/S0031-9422(00)90839-2

Warton, D. I., S. T. Wright, and Y. Wang. 2012. Distance‐based multivariate analyses confound location and dispersion effects. Methods in Ecology and Evolution 3(1):89-101. https://doi.org/10.1111/j.2041-210X.2011.00127.x

Yokoyama, H., Y. Ishihi, and S. Yamamoto. 2008. Diet-tissue isotopic fractionation of the Pacific oyster Crassostrea gigas. Marine Ecology Progress Series 358:173-179. https://doi.org/10.3354/meps07306

Zhao, L., F. Yang, and X. Yan. 2013. Stable isotopes and fatty acids as dietary tracers of intertidal bivalves. Fisheries Science 79(5):749-756. https://doi.org/10.1007/s12562-013-0645-2

Copyright © 2023 by the author(s). Published here under license by The Resilience Alliance. This article is under a Creative Commons Attribution 4.0 International License. You may share and adapt the work provided the original author and source are credited, you indicate whether any changes were made, and you include a link to the license.