Foraging under extreme events: Contrasting adaptations by benthic macrofauna to drastic biogeochemical disturbance

Read the free Plain Language Summary for this article on Journal blog. Marine ecosystems and biogeochemical cycles are facing growing pressures from extreme natural events human activities, resulting in a rapid loss of biodiversity changes to functioning food webs (Rees, 2012). This is particularly true benthic webs, which often spatially fragmented inhabited by rare, endangered, or endemic faunal species (Ashford et al., 2019; Gianni 2016). Gaining better understanding how communities respond sudden unpredictable disturbances crucial evaluating their resilience vulnerability formulating effective conservation management strategies these habitats. Monitoring such most habitats challenging (Hall-Spencer 2008; Harada 2020), replicating large-scale environmental controlled experiments presents its own set challenges (Ledger 2013; Sagarin However, valuable insights into response can be obtained through longitudinal studies that take advantage naturally occurring areas prone earthquakes, typhoons, other disturbances. Shallow marine hydrothermal vents, also known as shallow offer unique opportunity study Benthic chemosynthetic rely both photosynthetic resources (Ho 2015; Reid Tarasov 2005; Wang 2015). Chemosynthetic production occurs when warm vent fluids rich sulphides, methane, minerals mix with seawater (Chan 2016; Chen 2017; Hall-Spencer Levin 2005). The availability dietary largely influenced intensity interaction surrounding ecosystem horizontal vertical transport pathways (Levin These turn affected various factors, fluid flow, elemental cycling, mixing Ramirez-Llodra 2011; Therefore, producers fuelling sensitive perturbations. ability animals use sources tolerate high levels toxicity, hydrogen sulphide (H2S), makes them model origin life Earth (Little & Vrijenhoek, 2003; Luther 2001; Martin Minic, 2009; Petersen Sellanes 2011). Despite importance, threats, including climate change, resource exploitation, pollution Suzuki 2018) our drastic disruptions limited. alter balance between primary production, potentially leading dire consequences consumer populations (Bell Kharlamenko 1995; Portail 2018; Wu 2021). In 2016, one studied systems world, Kueishantao (KST), Turtle Island off coast Taiwan, was hit M5.8 earthquake subsequent C5 typhoon within few weeks (12 May 2–10 July, respectively, Supporting Information Videos S1 S2). KST located at 24.843°N, 121.951°E depth water <30 m. site background information detailed Text I. triggered underwater landslides, buried active vents habitat caused near “shutdown” activity almost two years (Figure 1, Lebrato 2019). combined effect decreased toxicity discharging (Lebrato 2019), important factors structuring 2017). Before events, Yellow Vents (YV) semi-inactive White (WV) had distinct properties 2). hot toxic YV discharged gases CO2, (H2S) sulphur dioxide (SO2) (Chen contrast, WV low degassing no discharge II). After disturbance, change physical environment reduction discharges led dramatic biogeochemistry Specifically, significant venting flow observed along complete depletion dissolved elements (e.g., Cd, Ba, Pb, Fe, Cu, As, Cl, Mg), recovered pre-disturbance concentrations 2 later Most notably, previously abundant native yellow accretions completely disappeared 1c). Time series drone images region (Supporting II Video S3) show noticeable decline vibrant hue generated prevalent chemoautotrophic bacteria epsilonproteobacteria gammaproteobacteria) after may have shift source resources. only macrofaunal (i.e., body length 0.5–50 mm) inhabit hostile crab Xenograpsus testudinatus (Ng 2000). Its successful existence has been attributed physiological adaptation highly acidic Hu 2012; Yang To date, it unclear whether adapted crabs obligate feeders because particulate organic matter (POM) make up substantial proportion (6%–87%) diet (Chang 2018). opportunistically photosynthetically derived foods Ho Jeng 2004; 2014). For example, X. feeds vast array zooplankton killed sulphurous plumes during slack episodes (Jeng 2004). Gut content analyses detritus biofilms rocks (Wang Compared YV, less much higher diversity Besides testudinatus, hosts numerous generalist gastropods sessile Bostrycapulus gravispinosus, Thylacodes adamsii, mobile Ergalatax contrata) capable rapidly colonizing early succession using variety Vent-dwelling acquire (60%–80%) coexistence non-endemic before investigating different members conditions. understand 2016 altered web, we sampled three gastropod peripheral area 2017): snail adamsii (Mörch, 1859) gastropods, limpet gravispinosus (Kuroda Habe, 1950) contracta (Reeve, 1846), 2015 2018. pioneer species, fairly tolerant conditions feeding differ: T. B. filter E. carnivorous We analysed stable isotopes sulphur, carbon, nitrogen (δ34S, δ13C, δ15N) consumers potential diet. Sulphur powerful tracer distinguishing generally lower δ34S values than (Fry, 1988; Fry 1983, 1991). As δ13C chemo- more reductive allow methane used carbon energy consider oxidizers partially tricarboylic acid cycle (rTCA) CO2 fixation (Tang Blankenship, 2010). pathway discriminates against 13C otherwise dominant Calvin-Benson-Bassham (Hügler Sievert, explain why Chang al. (2018) found POM enriched (−18.2‰ vs. –23.4‰). Finally, δ15N usually an indicator trophic position due 15N enrichment transfer (McCutchan Jr Post, 2002). It uncertain diagnostic deep-sea fauna tend non-vent (Michel Given close association systems, hypothesize will decrease species. No sample permits were required fieldwork. Research cruises collection organisms carried out April each year 2018 avoid seasonal variation temperature direction predominant tides Kuroshio Current. landscape landslides shore. accumulated chimney destroyed original seabed new layers debris sediments. Vent activities increased again many smaller fumaroles. Hence, sampling conducted top old site, GPS landmarks. (16–18 m), farther away shoreline (6–10 thin layer sediment compared 2015, allowing us same collected around transect cruise SCUBA diving minimum scientific divers. did not collect outside areas, they remained mostly unaffected chemistry zone, seafloor typically reflect shelf A total 12 specimens X six WV. 2018, 16 collected, 10 divers samples every four meters 25 m Key (T. adamsii) motile (B. gravispinosu contracta), year. At least mollusc collected. All placed portable nets brought surface where aerated containers seawater. sampling, invertebrates transported laboratory National Taiwan Ocean University 80 × 50 60 cm (L W H) tank constant fresh filtered (~25°C, similar ambient KST) 24 h empty gut contents. Unfortunately, died preyed cleaning, final sizes per size n = 3 6 (Table 1). lack trained personnel dissect animals, all rinsed deionized then frozen −20°C drying oven 60°C 48 h. dry shipped back dissected Kiel (CAU). being water, molluscan Euthanasia desensitized placing ice half hour dissection. sexed according shape pleonal structures crabs, pooled claw muscle tissues. available tissue, foot abductor muscles small mass specimens. investigate macrofanua changed changes, measured except irregularly coiled adamsii. determined measuring distance protruding notches. shell height length, determine gravispinosu. data presented Table Although methods, is, freeze methods little isotope macroinvertebrates (Akamatsu 2016) III). assess basal fuel web POM, chemolithotrophic (Epsilonproteobacteria) epilithic green algae (Chlorophyta indet. sp.) trip October spoiled unexpected electrical outage. filling L Nalgene bottles (n 5) 5). fashion 15 jugs, but 3). mouth vent, 2) several 1 plume vent. kept ship pre-combusted (500°C) Whiteman GF/C filters immediately lab. Green 5 total). Zooplankton handheld plankton net (500 μm mesh) locations zone location, cast nets, waited min drawing nets. Then separated fractions passing finer mesh (100 μm). Under microscope, further removed fish larvae jellyfish samples. Food freeze-dried days. Institute Geosciences CAU, homogenized mortar pestle subdivided fractions, C N S analyses. extracted published work 2022). approximately 60–100 μg ratios continuous ratio spectrometer MAT253 (Thermo Finnigan) coupled analyser EA1108 (Carlo Erba Instruments) (EA-IRMS) Conflo III interface (ThermoFinnigan) facility La Coruña, Spain. expressed delta (δ) permil (‰): δ(‰) [(Rsample/Rstandard) – 1] 1000, R heavy light isotope. relative international standards: VPDB atmospheric air nitrogen. During analyses, reference materials measurement calibration (NBS 22, IAEA-CH-6, USGS24 IAEA-N-1, IAEA-N-2, IAEA-NO-3 δ15N). Analytical errors ±0.15 ‰ calculated based replicated standard acetanilide interspersed Samples EA-IRMS Iso Limited Inc, UK. Approximately mg analysis. quality control, internal lab standards (IA-R068, IA-R038, IA-R069, mixture IAEA-C7 IA-R-R046) runs. calibrated material IAEA-C-7 δ15N. bulk sequence yielded 1σ 0.03‰ 0.03‰, respectively. About 20% performed duplicate ensure reproductivity measurement. identical reported analytical uncertainties. C:N > 3.5, applied lipid correction tissue 49) following Logan (2008). perform extraction prior tissues affect (Svensson Iso-Analytical Limited, Tin capsules containing plus vanadium pentoxide catalyst loaded automatic sampler, dropped, sequence, furnace held 1080°C, combusted oxygen. control purposes, test IA-R061 (barium sulphate) 12), IAEA-SO-5 5), IA-R068 (soy protein) IA-R069 (tuna 3) batch analysis Internal standards, IA-R061, IA-R025 sulphate), IA-R026 (silver sulphide), calibrate correct 18O contribution SO+ ion beam. interlaboratory comparison (NBS-127, barium sulphate IAEA-S-1, silver sulphide) distributed International Atomic Energy Agency. deviation ± 0.10‰ (±0.32‰). (‰) Vienna Canyon Diablo Troilite (V-CTD); run results yield population investigated correlation abundance conducting video survey mouths (121.962°E, 24.834°N, Figure I), populous testudinatus. digital camera capture 88 cm2 days August consistent previous data, captured day, footage daylight hours density analysis, image adequate. count entered area, divided 10-min intervals, amounted 66 film frames estimated 26 9 cm2. counts baseline established 2004 2014 comparable occasions, standardized individuals m2 estimate densities. (2004) remarked method collecting 0.25 likely underestimated there hiding rocky crevasse. Chan (2016) surveyed broader KST. They quadrats (25 cm) stations 2014. number counted inside quadrat photograph; however, authors noted approach underestimate some moved placed. statistical version 4.1.0 (R Core Team, 2020) R-Studio interface. Isotope (δ13C, δ15N, δ34S) assessed homoscedasticity Fligner-Killeen tests visualized any deviations normality QQ plot. P 0.87 (Chi-squared 0.71, df 3), 0.24 4.2, 0.29 3.8, df-3) p-values >0.05, indicates isotopic variance homogeneous. evaluated function: cor) other, none parameters correlated another. On basis tests, results. first tested t-tests. if differences WV) sex (female male), multivariate (MANOVA) univariate (ANOVA) 10, while 8. sex, so MANOVA Pillai's trace inter-annual collections assumptions satisfied. identify dependent variables and/or values) significantly years, ANOVA output summary.aov). two-sample t-test events. (D 4) met assumption (De Winter, 2013). 3; De intra-annual variability performing full value ranges wider those Additionally, examine (before events). p-adjusted decomposition p.adjust function FDR method. Unless stated, significance p < 0.01. ellipse (SEA) bivariate space group, niche space, 95% confidence interval package SIBER (Jackson Because calculation requires five points group calculate covariance matrix 2011), report do discuss molluscs Likewise, Bayesian proportional contributions incomplete, Discussion I). Missing result inaccurate estimation (Weltje, 1997). figures made “ggplot2” (Wickham, put dataset global context, δ34S, DeepIso database (https://www.seanoe.org/data/00654/76595/; Michel 2021) (Erickson 1983). publicly assessable contents ecosystems. primarily gathered Pacific inactive cold seeps, well fauna. selected soft preserved samples, either frozen. mean 1. Individual S2. include 2022) terms algae, bacteria, ranged 5.0 7.2‰ −1.7‰. average 3.8‰, (t10 1.85, p-adj 0.09). contrary, 4.2‰ (t11 -7.8, 0.0001). non-POM −22.4 −19.6‰ years. From −23.4 −20.8‰ –10.9, <0.0001), −18.2‰ −22.2‰ 5.7, 0.00014). bacterial (4.8‰) (4.9‰) closer (1.1‰) (16.0‰), (20.6‰) detect (MANOVA 0.49, F3,12 3.80; 0.08, S3). sites (ANOVA 0.41, 0.03, 0.16, respectively). females carrying eggs time sexes male) (Pillai's 0.27, 1.49; S4). 0.8, F3,24 34.9; mainly driven (ANOVA, 0.003 0.0001 respectively; S5). Interestingly, differed. ~2.4‰, whereas ~2.1‰ statistically 0.18, S5), −15.3 2.1‰ −16.2 1.1‰ shifts suspension replicates species-specific average, event ~3‰ ~5‰ temporal greater intraspecies Their intraspecific variabilities substantially contracta, (two t-tests: t5 3.7, 0.01; t7 12.6 0.0001;