Marine environments undergo rapid changes under the influence of various pressures (human footprint, climate change) and the monitoring of their ecosystem status becomes critical. Such a monitoring requires gathering data, to process them and to extract indicators summarizing the status of the environment that is otherwise too highly dimensional to be grasped by a human being. In recent years, the massive availability of data combined with powerful machine learning algorithms and the associated hardware led to significant advances in domains that were not even dreamed about in the last few years (image classification, automatic translation, text to speech, action selection, ...). Marine ecosystems, where progress has been made in collecting large amounts of data, could also benefit from these AI advances. However, the data in environmental sciences are often sparse either in time, space or relative to the measured variables, and imbalanced which constitute challenges for AI algorithms. This leads to the two directions followed in the SMART-BIODIV proposal: 1) harnessing the power of machine learning algorithms to complete and process sparse and imbalanced data that we often encounter in environmental sciences and 2) designing indicators to qualify the ecological status of the considered environments. Even if the data are scattered, there are several heterogeneous databases that constitute as many points of view that can be combined to build a coherent and complete state of the ecosystem. We will study the potential of interpolation algorithms in time and space as well as predictive models based on co-occurrences. We will also exploit the large image databases collected by the partners on marine plankton and make them available to the challenge participants. More prospectively, we will study the feasibility of including symbolic data, such as food webs, to constrain the evolution of the state of the ecosystem and inject this knowledge of the interdependencies between the dimensions of the state to improve its estimation. These data, grouped, merged and completed, will then serve as a basis for the calculation of taxonomic and trait-based indicators, which will be designed on the basis of our expertise in freshwater bioindication. To reach the challenge’s objectives, our consortium gathers complementary expertises in deep learning, computer vision, oceanography, plankton imaging, and freshwater bioindication. In addition, our experts in AI (GeorgiaTech, CentraleSupelec) and biodiversity (LOV, LIEC) have a strong record of fruitful interdisciplinary collaborations (co-supervised PhD, co-authored articles).

The biological carbon pump (BCP) annually supplies 5-10 Gt of carbon to the oceans’ interior, driving long-term carbon sequestration. However, the vast majority of sinking particulate organic carbon (POC) entering the mesopelagic zone (~100-1000 m) is remineralized. This dramatic vertical decrease in POC flux is modulated by multiple processes involving zooplankton and microbes. Despite multi-decadal research, there is no consensus on the controls and latitudinal trends in POC flux attenuation. A better understanding of the factors that control the efficiency of the BCP therefore requires to separate the processes that jointly set POC flux attenuation. To this end, I propose a ground-breaking approach to autonomously quantify mesopelagic microbial remineralization rates from profiling floats at high temporal resolution and over the full annual cycle. This novel observational approach developed in adhoC and the analysis of the dataset obtained will allow dissecting the BCP for the first time. The newly developed profiling floats will be deployed for >1 year in two contrasting ocean regions: the Northwestern Mediterranean Sea, and the area around the Kerguelen Islands. Over the seasonal cycle, these two regions offer the opportunity to explore wide-ranging ecological, biogeochemical and environmental settings needed to get insights into the drivers of the POC flux attenuation, their relative magnitude and controls. Concurrent estimates of the flux attenuation, and the associated microbial remineralization and zooplankton-mediated particle fragmentation rates will be used to determine the degree of influence (and timing) of these microbial and zooplankton-mediated processes in setting regional patterns in POC flux attenuation. Finally, these float-derived estimates will be combined with numerical model simulations to better constrain regional mesopelagic C budgets and global C sequestration.

During the previous H2020 framework program, we submitted in 2019 and in 2020 a European doctoral training network to the MSCA-ETN-ITN call. Our European Training Network aimed to train 15 early stage researchers (ESRs) at the interface between omics, modeling, and data sciences, to study the response of marine plankton to climate change and its impact on ecosystem services. Our consortium included 10 beneficiaries (including 2 non-academic) and 15 simple partners (including 11 non-academic). The building of this network was supported by the MRSEI Pré-Plankserv project in 2018-2019. Thus, our project had been very well evaluated during the previous calls, with a score of 87.60% for its first submission in 2019 and a score of 91.20% for its re-submission in 2020. With the implementation of the new Horizon Europe program, the H2020 MSCA-ITN call for will become the MSCA-Doctorates call, with significant modifications, and in particular a reduction of one third in the number of funded PhDs winthin the network, from 15 to 10 ESRs. Consequently, our doctoral training network will have to be modify and especially resized to respond to the future MSCA-Doctorates call. The MRSEI project "HorizonPlankton" therefore aims to support the re-submission of our European doctoral training network for the next Horizon Europe MSCA Doctorate call.

We aim at constraining the co-evolution of life and the environments on early Earth, targeting five milestones through life evolution (between 3.4 Ga – 400 Ma, Billion-Million years) linked with important changes in redox conditions and oxygenation. Identifying the fossils of these times has been limited by (1) morphological simplicity, (2) non-diagnostic organic carbon isotope ratio, (3) difficulty to correlate individual fossils with molecular biomarkers analyzed on bulk rocks, (4) difficulty to correlate fossils with geochemical metabolic/environmental proxies from bulk rocks. To overcome these limitations, we will use a combination of micro- to nanoscale characterizations of fossils. We will develop novel microscale molecular methods: Time-of-Flight Secondary Ion Mass Spectrometry (ToF-SIMS), and microscale Laser-desorption Laser-ionization Mass Spectrometry (µL2MS). Thanks to these innovative techniques we will be able, for the first time, to retrieve molecular information (biomarker and fossil biopolymer composition) on single fossil cells, and to distinguish adjacent cells as well as cell anatomy. These spatially-resolved analyses will identify possible in-lab and weathering contaminations. Complementary nanoscale analytical (spectro)microscopy will be used to analyze anatomy as well as mineral structures informing on post-mortem morphological modifications and biominerals. Metabolic signatures will be investigated using microscale and bulk-rock isotope analyses of organic matter and biominerals. The project builds on the gathered participants’ expertise in the fields of organic and isotope geochemistry, paleontology, nano-mineralogy, mass spectrometry and spectroscopy, and analytical developments. (A) Our selection of samples will allow us to address the effects of diagenetic and metamorphic transformations, as i) all fossils are preserved in quartz, ii) their age gradient is correlated with an increase in organic matter maturity and mineral matrix recrystallization, iii) 412 Ma to 1.6 Ga samples contain comparable microfossils (e.g. cyanobacteria, algae). (B) The 412-410 Ma samples will allow us to build a database of microscale molecular fingerprints on a large diversity of micro-organisms (cyanobacteria, algae, fungi) and specialized cells of macrofossils (plants, animals) correlated with nanoscale anatomical imaging. This will inform on cell structures and compositions in some of the earliest land plants (412-400 Ma) thus constraining the evolution of biopolymers including lignin, which triggered a rise in pO2 (O2 partial pressure). (C) We will compare morphologically identified microfossils and ambiguous morphospecies in 800-700 Ma old rocks with coupled molecular, textural and isotopic criteria. We hope to identify the relative importance of cyanobacteria and micro-algae in the primary photosynthetic production in order to constrain the role of the evolution of algae in the rise of pO2 of the Neoproterozoic Oxygenation Event, which resulted in the evolution of multicellular life. (D) ~1.6 Ga microfossil assemblages, coincident with the earliest eukaryotic fossils, will be characterized to constrain primary production during a period of reduced pO2 using microfossils of increased thermal maturity. (E) Microfossil assemblages of the Great Oxidation Event and its aftermaths between 2.45 and 1.8 Ga, will be studied to constrain metabolisms in environments characterized by important redox and pO2 fluctuations, with a focus on Fe-biomineralization associated with ferruginous conditions characterizing the ocean of this time period. (F) Organic microstructures (2.7 Ga) and enigmatic assemblages of small and large microfossils (3.0-3.4 Ga) will be studied to document primary production, methane and sulfur metabolisms associated with early anoxic and ferruginous environments together with possible early production of O2.

Assessing how the oceans sequester atmospheric carbon is a key challenge in biogeochemistry. Marine heterotrophic prokaryotes are key agents for this sequestration via the production of refractory dissolved organic matter (DOM) that persists in the ocean for years. This process, known as the microbial carbon pump, has particular importance in oligotrophic ecosystems, where an important fraction of fixed carbon is released as DOM. However, the processes that drive the microbial carbon pump are poorly known. MicroPump will assess the metabolic and environmental modulators of DOM production by marine prokaryotes, as well as its further bioavailability for natural microbial communities, in the Mediterranean Sea, an oligotrophic area highly vulnerable to climate change. Experimental work using model strains and natural communities, quantifying and characterizing DOM together with microbial diversity and gene expression, will be combined with a temporal study in the open NW Mediterranean sea as well as work with available Argo databases.