Data from: Unraveling the multiple facilitative effects of consumers on marine primary producers
Data files
May 03, 2024 version files 11.04 KB
-
chl_f0_gain_20240430.csv
-
field_algal_recruitment_20240430.csv
-
field_grazer_experiment_20240430.csv
-
grazer_slime_ammon_20240430.csv
-
lab_algal_recruitment_20240430.csv
-
README.md
-
tegula_ammonium_pam_20240430.csv
Abstract
The loss of consumers threatens the integrity of ecological systems, but the mechanisms underlying the effects on communities and ecosystems remain difficult to predict. This is, in part, due to the complex roles that consumers play in those systems. Here, we highlight this complexity by quantifying two mechanisms by which molluscan grazers – typically thought of as consumers of their algal resources – facilitate algae on rocky shores. Initial observations in high-zone tide pools revealed that both water-column ammonium concentrations and photosynthetic biomass were higher in pools containing higher densities of grazers, suggesting that local-scale nutrient recycling by the grazers could be enhancing algal biomass. We assessed this possibility by experimentally manipulating grazer abundances at the level of whole tide pools but controlling access of those grazers to experimental plots within each pool. Contrary to predictions that algal biomass inside grazer exclusions would increase as grazer abundances in the pools increased, we found that algal biomass inside grazer-exclusion fences was unaffected by grazer abundances. Instead, the consumptive effects of grazers that were evident at low grazer abundances transitioned to facilitative effects as experimentally manipulated grazer abundances increased. This finding suggested that these positive interactions were associated with the physical presence of grazers and not just grazers’ effects on nutrient availability. Subsequent experiments highlighted the potential role of “slime” – the pedal mucous trails left behind as the mollusks crawl on the substratum – in promoting recruitment of algae and thereby mediating a spatial subsidy of new organic matter into the system. Furthermore, different grazer groups contributed disproportionately to ammonium excretion (i.e., turban snails) versus slime production (i.e., littorine snails), suggesting a potential role for grazer diversity. Our work highlights the complex ways in which consumers affect their resources, including multiple, complementary mechanisms by which these grazers facilitate the algae they consume.
README: Data from: Unraveling the multiple effects of consumers on marine primary producers
Data represent a series of observations and experimental results that highlight the roles of grazers on rocky shore primary producers. Observational datasets (chl_f0_gain_20230222.csv, tegula_ammonium_pam_20230222.csv) highlight relationships between variables based on field observations of algal biofilms, grazer abundances, and ammonium concentrations. Experimental datasets (field_algal_recruitment_20230222.csv, field_grazer_experiment_20230222.csv, lab_algal_recruitment_20230222.csv) highlight responses in manipulative field and lab experiments to evaluate effects of grazers on a variety of responses related to primary producer abundances. A final dataset (grazer_slime_ammon_20230222.csv) includes measurements of ammonium fluxes and pedal mucous "slime" accumulation associated with different types of grazers.
## Description of the data and file structure
This dataset includes the following files, each with a unique set of measured attributes, described below:
(1) chl_f0_gain_20230430.csv
a. gain = the gain setting on the WALZ Diving PAM pulse amplitude modulated fluorometer. Values = 3 or 9.
b. f0_avg = the average dark-adapted f0 value measured by the PAM fluorometer on the tile in fluorescence units.
c. chla = the extracted chlorophyll a concentration measured on a Turner Trilogy benchtop fluorometer in micrograms per m2.
(2) field_algal_recruitment_20230430.csv
a. F0_mean = the average of four dark-adapted f0 values measured on that tile using the PAM fluorometer.
b. treatment = the grazer manipulation treatment. Values = SNAIL (snails enclosed on the tile) or CONTROL (no snails enclosed on the tile).
(3) field_grazer_experiment_20240430.csv
a. tide_pool = a number code used to identify each tide pool on the shore. This number was etched into a label adhered to the rock adjacent to each pool and served as a unique identifier. Values are nonconsecutive (some pools were initally marked but could not be used for the experiment) and range from 1 to 23.
b. treatment = the grazer manipulation treatment. Values = AMB (ambient, grazer abundance unmanipulated), or CONTROL (no snails enclosed on the tile), NO (grazers removed and added to the matched addition pool), or ADD (grazers added).
c. mean_grazer_biomass = the average biomass of grazers in the tide pool (grams per m2).
d. mean_grazer_fence = the average difference in grazer biomass between fenced and unfenced tiles within a tide pool (grams per tile).
e. mean_fenced_chla = the average chlorophyll a value measured on each fenced tile (micrograms per tile).
f. initial_fenced_chla = the chlorophyll a value measured on each fenced tile at the beginning of the experiment (micrograms per tile).
g. mean_unfenced_chla = the average chlorophyll a value measured on each unfenced tile (micrograms per tile).
h. initial_unfenced_chla = the chlorophyll a value measured on each unfenced tile at the beginning of the experiment (micrograms per tile).
i. mean_chla_fence_effect = the average difference in the chlorophyll a values measured on the fenced and unfenced tiles within each tide pool (micrograms per tile).
j. initial_chla_fence_effect = the initial difference in the chlorophyll a values measured on the fenced and unfenced tiles within each tide pool (micrograms per tile).
k. height = the elevation of the tide pool on the shore (m above mean lower-low water).
l. vol = the volume of the tide pool (L)
m. sa = the surface area of the tide pool (m2)
n. sa_v = the surface-area-to-volume ratio of the tide pool (m2 per L)
(4) grazer_slime_ammon_20230430.csv
a. ammon = the per-biomass ammonium accumulation rate (micromol per hr per g).
b. slime = the uncorrected, per-biomass slime accumulation rate (units per hr per g).
c. grazer_group = the type of grazer (teg = turban snail, lit = littorine snail, lim = limpet, chi = chiton).
d. ammon_pct = the percentage of total ammonium accumulation attributable to that individual.
e. slime_pct = the percentage of total slime production attributable to that individual.
(5) lab_algal_recruitment_20230430.csv
a. F0_mean = the average dark-adapted f0 value measured on that tile.
b. treatment = whether snails had been present on that tile. Values = Snail or Control.
(6) tegula_ammonium_pam_20230430.csv
a. teg_m2 = the density of turban snails in the tide pool (number per m2).
b. hrs = the length of time elapsed since the pool was isolated by the receding tide.
c. ammon_flux = the change in ammonium concentrations in the tide pool (micromol per L per hr).
d. f0 = the average dark-adapted f0 value measured in the tide pool using a PAM fluorometer.
e. teg_g_m2 = the biomass of turban snails in the tide pool (grams per m2)
f. chla = the calculated chlorophyll amount on the tile (micrograms per m2)
## Sharing/Access information
Data are not archived or accessible elsewhere. For questions, please contact Matthew Bracken at m.bracken@uci.edu.
## Code/Software
This submission does not include code.
Methods
This work was conducted at Corona del Mar State Beach (33° 35' 16.3" N, 117° 52' 1.5" W) on the coast of southern California USA. All research was conducted under California Department of Fish and Wildlife Scientific Collecting Permit SCP-13405. Except for some collections and short-term experiments in nearby tide pools to avoid affecting long-term experimental treatments, all measurements, surveys, and experiments were conducted in n = 18 tide pools in the rocky reefs at this location. Pool volumes averaged 22.6 ± 3.6 (mean ± SE) L, and bottom surface areas averaged 0.46 ± 0.05 m2. The pools were high on the shore; tidal elevations, determined using a self-leveling laser level (CST/berger, Watseka, Illinois, USA), averaged 1.58 ± 0.04 m above mean lower-low water. Based on comparisons with a nearby tide gauge (Newport Bay Entrance, Corona del Mar; 1.1 km from our study location) and typical wave heights, these pools were either splashed or covered for ~ 30% of the time during our measurements and experiments there.
Initial observations focused on the largest, most conspicuous grazers in the tide pools, turban snails (Tegula funebralis and T. gallina). Subsequent work also included littorine snails (Littorina scutulata and L. plena), limpets (Lottia limatula, L. scabra, and L. strigatella), and chitons (Cyanoplax hartwegii and Nuttalina californica). Collectively, these grazing mollusks comprised almost all the invertebrates in the tide pools, though pools also contained occasional mussels, hermit crabs, and sea anemones. None of these “other” species were abundant enough, relative to the grazers, to appreciably modify nutrient availability (Bracken et al. 2018). Other, smaller invertebrates (e.g., copepods, amphipods) were rare in the pools. Macroalgae were also virtually absent, and the grazers primarily consumed periphyton, including bentic microalgae and microscopic forms of macroalgae.
We surveyed grazer abundances by spreading a piece of flexible nylon trawl net (10 cm × 10 cm squares) across the bottom of each pool (Foulweather Trawl Supply, Newport, Oregon, USA; Bracken and Nielsen 2004). The net was used to count grazers and measure surface area. Grazer abundances were converted to ash-free dry biomass using values in Bracken et al. (2018). Initial surveys and measurements (relationships between turban snail biomass density, ammonium accumulation, and biofilm photosynthetic biomass) focused on n = 10 tide pools, but our subsequent experiment included n = 18 pools for greater replication of treatments.
Water samples were collected from each pool as the pool was isolated by the receding tide and again ~3.5 hr (range 2.8 – 4.3, depending on the height of the pool) later. Samples were collected in acid-washed bottles, placed immediately on ice, and transported to the lab for ammonium (NH4+) analyses. We used the difference in NH4+ concentrations and the elapsed time between samples to calculate the rate of NH4+ accumulation in each pool (μmol L-1 hr-1). Samples were analyzed using the phenol-hypochlorite method (Solórzano 1969) on a UV-1800 benchtop spectrophotometer (Shimadzu, Carlsbad, California, USA).
Photosynthetic biomass associated with the benthic periphyton growing on the substratum was quantified non-destructively in each pool using pulse amplitude modulated (PAM) fluorometry. We covered each tide pool with an opaque plastic sheet for at least 30 min, then used a Diving-PAM-II underwater chlorophyll fluorometer (Heinz Walz GmbH, Effeltrich, Germany) to measure the dark-adapted minimal level of fluorescence (F0; Murchie and Lawson 2013). This parameter is a frequently used non-destructive surrogate for extracted chlorophyll a (chl a) values (Honeywill, Paterson, and Hagerthey 2002, LaScala-Gruenewald et al. 2016), and our measurements support this relationship. Chl a concentrations (mg per tile) extracted from travertine tiles placed in pools to quantify periphyton abundance were correlated with log(dark-adapted F0) measurements taken on those tiles in the field, regardless of the gain setting (gain 9: F1,58 = 104.9, P < 0.001, R2 = 0.64; gain 3: F1,31 = 92.0, P < 0.001, R2 = 0.75), and we used these relationships to convert PAM measurements to chl a. Reported values for each pool are the average of n = 5 measurements taken haphazardly across the rock substratum under the opaque covering. Because these pools lack macroalgae, the dark-adapted F0 measurements provided an estimate of the photosynthetic biomass in the biofilm in each pool.
Travertine tiles (10 cm × 10 cm) were similar in composition, color, and rugosity to the rocky reefs at the study location but provided a consistent substratum that was easily fenced to limit grazer access. Three tiles were surrounded by 5-cm-high fences made from stainless steel wire cloth mesh (8 × 8 mesh size, 0.7 mm wire diameter, 2.5 mm opening size; McMaster-Carr, Elmhurst, Illinois, USA) and attached to a wall anchor drilled into the bottom of each tide pool using lag bolts, fender washers, and neoprene washers for cushioning. Tiles were placed in each pool 2 months prior to the start of the experiment to allow microalgae to recruit and grow. At the beginning of the experiment, dark-adapted F0 values (gain setting 9) were measured on each tile, as described above, and one tile was randomly selected for removal and chl a extraction in 90% acetone. Extracted chl a was determined using a Trilogy benchtop fluorometer (Turner Designs, Sunnyvale, California, USA) to evaluate the relationship between initial chl a and F0 values.
Within each pool, the remaining two tiles were randomly assigned to either a “fenced” treatment – the stainless steel fence was left surrounding the tile – or an “unfenced” treatment – the fence was removed. Fences were effective at limiting access by grazers. Whereas there was no effect of the herbivore manipulation treatment on grazer biomass on fenced tiles (generalized linear model [gamma distribution, log link]: X2 = 3.4, d.f. = 2, P = 0.186), the biomass of grazers on unfenced tiles depended on the treatment (X2 = 13.3, d.f. = 2, P = 0.001). Accordingly, the difference in grazer biomass between fenced and unfenced tiles depended on the treatment; effects were most apparent in tide pools where grazers were added and least apparent where grazers were removed (X2 = 17.9, d.f. = 2, P < 0.001), as differences between treatments were minimized where there were few grazers in the pools.
We used a stratified random method to assign grazer manipulations at the level of tide pools. Pools were ranked by volume, generating 6 sets of n = 3 pools each. Within each set, pools were randomly assigned to one of three treatments: reduced, ambient, or addition. Every few days, all grazers were counted (and counts were subsequently converted to ash-free dry biomass [Bracken et al. 2018]), removed from each “removal” pool, and added to each “addition” pool. Grazers in “ambient” pools were removed from the substratum and replaced in the same pool to mimic disturbance associated with removals and additions. Herbivore manipulations were effective at modifying grazer biomass in tide pools (GENMOD: X2 = 13.0, d.f. = 2, P = 0.002). Relative to unmanipulated control pools, additions resulted in an increase in grazer biomass (GENMOD: z = 2.4, P = 0.049 after Tukey adjustment). Rates of NH4+ accumulation in each pool were evaluated as described above.
The experiment was maintained for ~ 6 weeks (26 April to 09 June 2016), and dark-adapted F0 values were measured on each tile (the average of 4 measurements, 1 in each quadrant) every 2 – 7 days using the PAM fluorometer. Initially, the fluorometer was set to a more sensitive gain setting, but as algal biomass increased on the tiles, we reduced the setting from 9 to 3 to avoid overflow errors due to high signal intensity. At the end of the experiment, we removed all tiles, extracted accumulated chl a in 90% acetone, and evaluated the relationship between the final dark-adapted F0 readings (gain 3) and the extracted chl a concentrations. We used these relationships to convert F0 to chl a so that units were comparable regardless of the gain setting.
We evaluated mechanisms underlying our experimental results by examining how the presence of grazers affected algal biomass. In the field, we anchored 10 cm × 10 cm travertine tiles (n = 10) in sets of two in a large tide pool adjacent to our experimental pools. Both tiles were enclosed in fences, but they were also covered with stainless steel wire cloth to prevent enclosed snails from escaping. One tile in each set was randomly assigned to a +Snail treatment (8 turban snails, corresponding to densities observed in nearby tide pools); the other ‑Snail tile in each set contained no snails. After seven days, we estimated photosynthetic biomass accumulation on each tile by covering the tide pool with black plastic sheeting for 30 minutes prior to measuring dark-adapted F0 values in situ using a PAM fluorometer.
To assess effects of slime in the absence of snails, we quantified effects of snails on algal recruitment in n = 8 square microcosms in the lab. Each microcosm contained a 10 cm × 10 cm travertine tile and 200 ml of saltwater (salinity of 35; Instant Ocean® Sea Salt, Blacksburg, Virginia, USA). Half of the microcosms (n = 4) contained littorine snails (Littorina scutulata and L. plena), and the other half contained no snails. We focused on Littorina because our measurements indicated that littorines produced more slime than other grazer groups (see Results). Snail abundances in snail-addition microcosms were determined by ash-free dry biomass (0.91 ± 0.01 g), consistent with abundances in our assessments of grazer ammonium excretion (Bracken et al. 2018) and grazer abundances at the site (M. Bracken, personal observation). Microcosms were covered with nylon mesh filter screens to prevent escapees and placed in a temperature-controlled (16°C) room under lighting conditions programmed to mimic natural sunrise and sunset cycles. After 24 hours, 44 ml of f/2 culture medium containing dense cultures of the phytoplankton Tetraselmis chui and Dunaliella salina (Algae Research and Supply, Carlsbad, California, USA) were added to each microcosm. Both phytoplankton species are found in nearshore ecosystems in the northeastern Pacific (Jensen and Tanner 1973; Hori et al. 1986), though we acknowledge that they are not perfect analogues of the species recruiting to tide pools. After 1 hour, we poured off the seawater and culture medium, removed the snails, and carefully added 200 ml of fresh seawater. We then covered the microcosms with an opaque plastic sheet for 30 minutes before measuring dark-adapted F0 values to estimate the photosynthetic biomass retained on each tile.
Ammonium excretion and slime production rates were evaluated in microcosms containing 200 ml of saltwater (salinity of 35; Instant Ocean® Sea Salt, Blacksburg, Virginia, USA). Each microcosm contained a travertine tile as substrate. To simulate the still-water conditions in tide pools, the water was not stirred or aerated. Temperatures were maintained at 20 °C, which is representative of field conditions (19.51 ± 0.01 °C). Grazers were collected from the field immediately prior to measurements. The same ash-free dry biomass of grazers (0.91 ± 0.01 g) was added to each microcosm at the beginning of each trial based on average ash-free dry mass measurements of representative individuals of each grazer group (Bracken et al. 2018). Because individuals of the different groups were characterized by different masses, maintaining a constant mass across grazer groups necessitated different numbers of individuals for each group in each microcosm: 135 littorine snails, 45 limpets, 15 chitons, or 5 turban snails. Microcosms were covered with nylon mesh filter screens to prevent escapes.
Trials were run for ~19 h in the lab, with a light-dark cycle (92 ± 3 mmol photons m-2 s-1) that mimicked outdoor conditions (13 hr light : 11 hr dark). This duration was between the median isolation time of experimental tide pools in the absence of wave splash (23.0 h) and after accounting for a 0.5 m swell height (8.3 h; Bracken et al. 2018). Initial water samples were taken from each microcosm prior to adding the grazers, and a second set of samples was taken at the end of the trial. Ammonium concentrations in the water samples were analyzed using the phenolhypochlorite method (Solórzano 1969) on a UV-1800 benchtop spectrophotometer (Shimadzu, Carlsbad, California, USA). Changes in ammonium concentrations in control microcosms without grazers were minimal and were accounted for when calculating ammonium accumulation rates. Initial ammonium concentrations averaged 0.9 ± 0.2 mmol/L.
As molluscan grazers move across the substratum, they glide on a thin layer of a glycoprotein network (hereafter, “slime”) that adheres them to the surface and allows them to crawl (Denny 1980). Because of the potential for slime to affect algal recruitment (Connor and Quinn 1984), we quantified accumulation of slime associated with the different grazer groups. After the final ammonium samples were collected from each microcosm, we removed and rinsed each tile with 25 ml of artificial seawater, saving the rinse water. We then used a razor blade to carefully scrape the surface residue from the tile, cleaning the razor blade in the rinse water. Finally, we rinsed the tile again with 25 mL of seawater, for a total volume of 50 ml, which we poured into a 50 ml conical tube and vortexed. The contents of the tube were then trapped on an IsoporeTM membrane filter (0.4 mm pore size hydrophylic polycarbonate membrane; MilliporeSigma, Merck KGaA, Darmstadt, Germany) and stained with Alcyan Blue (pH 2.5), a dye that complexes with acidic groups of polysaccharides and mucous glycoproteins (Hall et al. 1980, Passow and Alldredge 1995). Filters were rinsed with deionized water then placed in 5 ml of 80% sulfuric acid for 2-4 hr while mixing gently. The resulting solution was then read at 787 nm on a UV-1800 benchtop spectrophotometer (Shimadzu, Carlsbad, California, USA; Passow and Alldredge 1995). The method was calibrated using xanthan gum as a standard (Passow and Alldredge 1995), so all slime accumulation values are xanthan-gum equivalents. Ammonium and slime accumulation rates were calculated on both a per-individual (umol h-1 ind‑1 or mg h-1 ind‑1) and per-biomass (umol h-1 g‑1 or mg h-1 g‑1) basis based on the change in ammonium concentrations over time or the slime that had accumulated on the substratum in n = 4 microcosms for each grazer group.
To estimate contributions of each grazer group to ammonium and slime accumulation rates in the field, we combined lab-measured rates with field surveys of grazer abundances. Grazer counts in surveys of n = 18 tide pools were multiplied by the per-individual rates of ammonium and slime accumulation for each grazer group. These were then used to calculate the percentage of the total estimated ammonium and slime accumulation in each pool associated with each grazer group.
References
Bracken, Matthew E. S., and K. J. Nielsen. 2004. "Diversity of intertidal macroalgae increases with nutrient loading by invertebrates." Ecology 85 (10):2828-2836. doi: 10.1890/03-0651.
Bracken, Matthew E. S., Jill M. Oates, Alexander J. Badten, and Genevieve Bernatchez. 2018. "Predicting rates of consumer-mediated nutrient cycling by a diverse herbivore assemblage." Marine Biology 165 (10):165. doi: 10.1007/s00227-018-3422-z.
Connor, Valerie M., and James F. Quinn. 1984. "Stimulation of food species growth by limpet mucus." Science 225 (4664):843-844. doi: 10.1126/science.225.4664.843.
Denny, Mark. 1980. "The role of gastropod pedal mucus in locomotion." Nature 285 (5761):160-161. doi: 10.1038/285160a0.
Hall, Roderick L., Rosamund J. Miller, Anthony C. Peatfield, Paul S. Richarson, Isobel Williams, and Irvin Lampert. 1980. "A colorimetric assay for mucous glycoproteins using Alcyan Blue." Biochemical Society Transactions 8 (1):72. doi: 10.1042/bst0080072.
Honeywill, Claire, David Paterson, and Scot Hagerthey. 2002. "Determination of microphytobenthic biomass using pulse-amplitude modulated minimum fluorescence." European Journal of Phycology 37 (4):485-492. doi: 10.1017/S0967026202003888.
Hori, Terumitsu, Richard E. Norris, and Mitsuo Chihara. 1986. "Studies on the ultrastructure and taxonomy of the genus Tetraselmis (Prasinophyceae)." The Botanical Magazine, Tokyo 99 (1): 123-35. doi: 10.1007/BF02488627
Jensen, James B., and Sarah J. Tanner. 1973. “A preliminary check-list of the marine algae of the Moss Landing jetty: an annotated floristic compilation.” Contributions from the Moss Landing Marine Laboratories No. 38. Technical Publication 73-7, CASUC-MLML-TP-73-07. https://repository.library.noaa.gov/view/noaa/10891.
LaScala-Gruenewald, Diana E., Luke P. Miller, Matthew E. S. Bracken, Bengt J. Allen, and Mark W. Denny. 2016. "Quantifying the top-down effects of grazers on a rocky shore: selective grazing and the potential for competition." Marine Ecology Progress Series 553:49-66. doi: 10.3354/meps11774.
Murchie, E. H., and T. Lawson. 2013. "Chlorophyll fluorescence analysis: a guide to good practice and understanding some new applications." Journal of Experimental Botany 64 (13):3983-3998. doi: 10.1093/jxb/ert208.
Passow, U., and A. L. Alldredge. 1995. "A dye-binding assay for the spectrophotometric measurement of transparent exopolymer particles (TEP)." Limnology and Oceanography 40 (7):1326-1335. doi: 10.4319/lo.1995.40.7.1326.
Solórzano, L. 1969. "Determination of ammonia in natural waters by the phenolhypochlorite method." Limnology and Oceanography 14:799-801. doi: 10.4319/lo.1969.14.5.0799.
Usage notes
Data files are provided as comma-delimited (.csv) text files.