Results

General description of activities over the duration of the project

Objectives 1- 4 were addressed in two field campaigns and dedicated lab experiments. The field campaigns took place at tidal flat sites in France (La Coupelasse, Baie of Bourgneuf, June 2017) and The Netherlands (Biezelingse Ham, Westerschelde estuary, June-July 2018). All partners (except P2 in 2017 and 2018, and P6 in 2017) participated to both campaigns.

Central to the 2017 campaign was a large-scale stable isotope probing (SIP) pulse-chase experiment carried out in two contrasting sediment types (mud vs sand). At t=0, sediments were sprayed with sodium 13C-bicarbonate. Samples were then collected each hour for 4 (mud) to 3 hrs (sand)(pulse period) and then again after 24, 48 and 120 hrs (chase period). Samples were taken for abiotic (T°C, salinity, sediment grain size, water content, porosity and nutrient pools, P1,3) and biotic parameters (TOC, bulk carbohydrates, EPS, pigments, fatty acids, spectral reflectance, P1,3-5) and biodiversity assessments (pigments, 16S and 18S rDNA and rRNA amplicon sequencing, meio- and macrobenthos, P1,3,5, BIO-LITTORAL).

SIP analyses were performed for DIC, TOC, meio- and macrobenthos, fatty acids, EPS and RNA (P1, 3- 5). Primary production (PP) and community respiration were measured using PAM fluorometry (P5) and CO2 flux measurements (P4). BMA vertical migration was measured using microscopy (lens- tissue), spectral reflectance and PAM (P1, 5). Sediment stability and erosion thresholds were measured using a cohesive strength meter (P1).

As the RNA-SIP approach used in the 2017 campaign was not successful (see 4.3), we opted for a detailed metatranscriptomic and metabolomic approach in the 2018 campaign. One muddy site was sampled throughout two consecutive low tides (day, night). Metabolomics was performed using an untargeted metabolomic approach for BMA biofilms developed by P4 (M11, fig 1).

Fig.1 : Pie charts showing the proportions (from the total number of detected compounds) of annotated compounds and their chemical family in the MeOH and CHCl3 fractions analyzed by GC-MS.https://www.frontiersin.org/articles/10.3389/fmars.2020.00250/full

The same abiotic and biotic parameters and biodiversity assessments (as in the 2017 campaign) were made by the same partners. In addition, light and oxygen profiles were established (P5), and vertical profiles of oxygen, pigments, nutrients and organic C were established to model organic matter degradation and nutrient remineralisation during a diurnal tidal cycle (P1,3). EFs include PP (P4,5), community respiration (P4,5), organic matter degradation, nutrient remineralisation (P1,3), sediment stability (P6), vertical migration (P1,5) and enzymatic activity (P4). The 2017 field campaign data sets are almost complete, but measurements for the 2018 campaign are still in progress. It was originally planned to hold a project meeting in early spring 2020 to discuss, integrate and prepare data for joint publications on the field campaign data, but the covid-19 situation has severely impeded progress (see also 4.5).

In addition to the two main campaigns, monthly samplings were carried out at the French site (muddy and sandy station, March 2017-July 2019) to determine how seasonal changes in BMA biodiversity affect PP, pigment diversity and EPS diversity [objective 2, P1,4,5, in collaboration with academic stakeholders Geslin and Kühl]. These data are currently being prepared for publication. This field work was complemented with experiments to assess the impact of sediment type and light on BMA PP and photo-regulation (P4,5, M13, fig.2).

Fig. 2 : Relative changes of optimal photosystem two light efficiency ϕII1. ϕII1 is displayed as percentage of the initial value prior to illumination (at T0). Each color has been fitted with a linear model where confidence interval (0.95) are represented in gray (n = 3).https://www.frontiersin.org/articles/10.3389/fmars.2020.00203/full

Furthermore, in collaboration with the two above-mentioned scientific stakeholders a manuscript will soon be submitted on the interaction between benthic diatoms and kleptoplastidic foraminifera, and a paper on the impact of epipelic diatom biodiversity on their photo-regulation mechanisms is being prepared.

Biotic interactions between BMA and bacteria, and how these affect ecosystem functioning, were studied in dedicated lab experiments (objectives 2-3, P1,2,4,5).

  • In a first set of experiments, the effect of bacterial isolates on the growth of individual diatom strains was studied in order to evaluate the nature and specificity of the diatom-bacteria interactions.

  • In a second set of experiments, the effect of bacteria on diatom community composition and vice versa, and how this influences the BEF relation, was investigated.

  • In a third experiment, an axenic diatom strain was exposed to a foreign and a familiar bacterial inoculum, after which the cocultures were allowed to develop over 8 consecutive (diatom) growth cycles.

At the end of the experiment, differential recruitment of bacterial assemblages and their effect on diatom growth were assessed (metabarcoding, ecophysiological characterization). In addition, metagenomics and transcriptomics were used to functionally characterize the diatom- bacteria interaction, and SIP was applied to study the C flux from diatom to bacteria. Finally, the impact of bacteria on C assimilation strategies (photo- vs heterotrophy) in different BMA species was addressed using nanoSIMS (P1,2,5). The main aim of the latter experiment was to study mixotrophic capacities of different BMA species by feeding them with 13C labelled NaHCO3 or EPS and using nanoSIMS to quantify species-specific uptake of these labels in natural communities (objective 3, P1,2,5). The results of these experiments have been published (M3, 5, fig. 3) or are being prepared for publication.

Fig. 3: The presence of bacteria steepens the algal diversity-productivity relation. (A) Diatom biovolume production in function of species richness and its slope (m) in the presence (non-axenic, black line) or absence of bacteria (axenic, gray line). (B) Algal biovolume production per diatom species and its slope for each diatom: S. robusta (yellow), C. closterium (red), and N. phyllepta (blue). The presence or absence of bacteria are, respectively depicted as dark- and light-colored lines. The contribution of the selection and complementarity effects to diversity-productivity relation are, respectively shown in (C,D) in the presence (black) or absence (gray) of bacteria.https://www.frontiersin.org/articles/10.3389/fmicb.2019.01255/full

The effect of meiobenthos diversity on C cycling and sediment stability via bioturbation and grazing was quantified using field (cf. above) and lab experiments (objective 4, P1,3). The results of the 2017 SIP field experiment were used to quantify C uptake through BMA grazing by both meio- and macrobenthos at high taxonomic resolution (species to genus level). These data are included in the linear inverse model (objective 6) to allow quantification of the effect of nematodes on benthic C flows. Different simulations in which the contributions of one or more species are removed and substituted by surplus contributions of the remaining taxa (as a simulation of species loss without concomitant loss in abundance) will shed light on the possible effect of biodiversity on the contribution of nematodes to tidal flat C flows. Three experiments were set up to study the effect of meiobenthos grazing and bioturbation on C cycling (P1,3). In a first experiment, the effect of a natural nematode community on the productivity and EPS production of an artificial BMA biofilm were studied using a new design of tidal microscosm (M1, fig.4). In a second experiment, in collaboration with a new academic stakeholder (Cnudde, UGent, B), we developed a novel approach using a combination of X-ray radiography and computed tomography scanning of intact sediment cores to characterize bioturbation activity of different nematode species (D’Hondt et al., in prep.). Finally, using this set-up we investigated the relationship between nematode abundance and (micro)bioturbation. Results of these experiments allow evaluating to what extent nematode functional diversity affects biofilm EFs.

Fig. 4: .a) Microcosm design. Side apertures (diam. 18 mm) and bottom are sealed with 1.2 μm pore size glass fibre filters. Microcosms are filled with sediment and experimental organisms. HTW = high-tide water level, LTW = low tide, Seda level = sediment surface. b) Tidal aquarium with microcosms. The pump system is connected to a timer and a sensor measuring the water levels. c) Experimental set-up - daily cycle. Microcosms were subjected to an 8-h photoperiod and a low tide period of 6-h during the photoperiod. Oxygen measurements were made half an hour after the initiation of the photoperiod and 1 h before low tide. PAM measurements were made half an hour before the end of the low-tide period.https://www.sciencedirect.com/science/article/abs/pii/S0141113618300709

Upon request of private stakeholder Benth’ostrea it was decided to test the impact of diatom diversity on sea urchin production instead of on oysters (objective 5). More specifically, the effect of diatom diversity on the recruitment of the larvae of the sea urchin Paracentrotus lividus was studied (P5, Benth’ostrea, M9, Fig. 5). This work was followed up by further experiments in 2019 with the diatom species that gave the best results, and also including the effect of antibiotics, oyster shell particles and a natural BMA biofilm obtained. This follow-up work is still ongoing but has been temporarily halted by the COVID-19 pandemic. P4 was involved in a study to assess the importance of BMA as a food source of sea cucumbers.

Fig. 5: Survival rate (%) of P. lividus 10 days post-metamorphosis (DPM). NL = N. laevis, NATURAL = natural biofilm, SHELL = broken oyster shells, GABA + MIX = GABA + N. laevis + H. coffeaeformis. Data are expressed as mean ± confidence interval 95% (n = 4).https://www.sciencedirect.com/science/article/abs/pii/S0044848619316588

In collaboration with P1, P5 assessed the effect of resuspended sediment on oyster feeding. Clearance rates of two diatom species by the oysters were studied at different sediment concentrations. The first results were inconclusive, and follow-up activities have been suspended because of COVID-19. In addition, the potential of exploiting benthic diatom diversity for the production of commercially interesting lipids (PUFA, ARA and EPA) was investigated (P5, M10, Fig. 6). Lipid production of selected promising species was studied in relation to light and nutrient stress. Realizing that existing photobioreactors are not ideally suited for growing benthic diatoms, a promising collaboration with a new private stakeholder (the biotech start-up Synoxis Algae) was initiated to design a modular photo-bioreactor that can be used to optimize growth of benthic diatoms.

Fig. 6: Total lipid (A,B,C), carbohydrate (D,E,F), and protein ratios (G,H,I) of E. paludosa, N. alexandrina, and Staurosira sp. for each treatment. * p < 0.05;**p < 0.01;***p < 0.001 for nitrogen conditions and $ p < 0.05;$ $p < 0.01;$ $ $p < 0.001 for light conditions.https://doi.org/10.1371/journal.pone.0224701.g006

Objective 6 was addressed by developing two highly resolved food web models using linear inverse modelling. The models were developed by P3 with benthic biomass data collected and fluxes measured by P1, 3-5 during the 2017 field campaign. The models incorporated organic matter, bacteria, several primary producers, microbenthos (i.e., ciliates), meiobenthos (nematodes on genus level, other meiofauna taxa on higher taxonomic resolution), and macrobenthos at species level (Stratmann et al., in prep.). In addition, the early diagenesis of organic matter during a tidal cycle at night and a tidal cycle during the day will be modelled using the OMEXDIA approach (developed by P3, with input data collected and measured by P1,3-5 during the 2018 field campaign). This will allow assessing how organic matter degradation and nutrient remineralization is affected by tides and the day-night cycle.

Remote sensing (RS) approaches in combination with modeling and experimentation were used to assess and map sediment type, BMA gross primary production (GPP) and BMA resuspension (objective 7, P5). The objective of mapping sediment type using multispectral imagery was not successful. Several existing models were tested on SPOT and Pléiades images to map the mud content of sediment. Mapping GPP using multispectral imagery (SPOT and Pléiades) and hyperspectral imagery (HySpex) was performed by coupling RS, modeling and experimentation (M12,1 Fig. 7 the application of hyperspectral imagery is being further developed (PhD M. Zhang). Mechanisms underpinning BMA resuspension at ecosystem-scale were investigated using a coupled physical- biological model forced by realistic meteorological and hydrodynamical forcings to simulate chronic (without any concomitant sediment resuspension) and massive (driven by bed failure) BMA resuspension. Modeled estimates of resuspended BMA Chl a were compared with (simultaneously acquired) Chl a field data and RS-derived Chl a estimates. RS-derived Chl a was computed using a semi-analytical inversion algorithm specifically developed for coastal and inland waters specifically tuned to turbid intertidal waters (Gernez et al., 2017).

Fig. 7: Hourly (mg C m–2 h–1) and daily (mg C m–2 d–1) averaged remotely-sensed GPP from the GPP-algorithm based on Eilers and Peeters (1988) P–E model in March (a,d), May (b,e), and July (c,f).

All objectives have been achieved. Some tasks have been slightly modified [e.g. we used a metatranscriptomic not an RNA-SIP approach in the 2018 field campaign (WP1); experiments were performed with sea urchins not oysters in WP4]. Fifteen papers have already been published, 1 is submitted and 19 papers are being prepared. The fact that no in person meetings could proceed because of the COVID-19 pandemic (e.g. planned meeting in March 2020 between P1,3 and 5, final project meeting, etc.) has caused a significant delay in the preparation of joint publications.