Reliable and relevant policies towards a sustainable management and conservation of marine resources must take into account biogeography. This is inherently a dynamic reality driven by multiple abiotic and biotic variables. In a context of increased pressure over marine ecosystems, understanding the natural partition of the ocean including established and emerging patterns is of paramount importance.
Oceanic thermal fronts are important mesoscale regions of maximum horizontal gradients with variable subsurface expression. They are simultaneously natural boundaries and meeting places, driving enhancements in productivity and separating communities. In the context of biogeography, fronts constitute natural boundaries, which reflect real discontinuities across the oceanic domain.
The physical and ecological characteristics of fronts make them especially meaningful features for a wide variety of disciplines and operational activities. Understanding the location, magnitude, and drivers of frontal activity promotes a better comprehension of the spatial heterogeneity of marine ecosystems. Designing Marine Protected Areas, identifying Potential Fishing Zones, and comprehending migration and foraging behaviors by a multitude of marine organisms all depends on the implementation of frontal activity cartography.
Traditional direct observation techniques are inadequate to this end given the dynamic nature of ocean and coastal areas. Earth Observation data can effectively address these challenges and provide synoptic views of the ocean and mesoscale features including fronts at a global and local level.
However, and despite the multiple applications possible using available technology, frontal mapping is still limited. Simultaneously, most works provide only incomplete analysis of broad areas using low resolution (spatial and temporal) datasets.
This work presents a dense time series (2003-2014) of frontal activity over an Eastern Boundary Current System (Western Iberian Peninsula). To this end, a set comprising 3450 MODIS daily images was processed to extract the location of thermal fronts from Sea Surface Temperature data. The methodology of Cayula and Cornillon (1992) was adapted to meet the requirements of the MODIS sensor. Daily, monthly, seasonal and yearly maps were produced and compared against regional topography and oceanographic features.
The spatial distribution of fronts was also statistically analyzed to identify meaningful patterns correlated with regional biogeography. Spatial autocorrelation in particular was tested using Moran’s Global I.
Additionally, an Empirical Orthogonal Function analysis was applied to the frontal activity dataset to reduce redundancy and identify meaningful spatial and temporal patterns in front distribution.
Results demonstrate in a significant way the spatial patterns of frontal activity are not random in the study area and are strongly correlated in with the seasonal variability of important currents and coastal features, such as upwelling and freshwater lenses. Topographic effects are also relevant in the establishment of enhanced frontal activity regions, persistent throughout the year. These correlations are explored in detail, highlighting measured intra- and inter-annual variability.
The findings present important insights into frontal spatial and temporal distribution, with a detail level previously unavailable for this important system. A comparison of the findings with those of other Eastern Boundary Current Systems (e.g. California) and a discussion of optimum methods and datasets for scalable frontal monitoring are presented.
Operational monitoring of these features for a broad set of stakeholders is discussed including the requirements, opportunities, and barriers to the successful implementation of the solution. Converting frontal activity maps into actionable information is still challenging, requiring the intervention of multidisciplinary teams and robust dissemination platforms. Ongoing efforts towards the deployment of such a system are mentioned in the present work as a contribution to dynamic marine biogeography.

Intertidal habitats rank amongst the most productive ecosystems providing bird habitat and feeding areas, commercial fish nursery grounds, as well as other ecosystem services including nutrient cycling and coastal protection. Seagrasses, mud flats and saltmarshes are some of the world’s most protected habitats. However, this protection has failed to prevent accelerating global loss due to over exploitation, direct damage and numerous other stressors including algal mats, pollution and marine litter. What is more, poor availability of relevant in-situ data, makes the monitoring of intertidal habitats challenging, costly and inefficient.
Deimos Space UK, in collaboration with the University of Portsmouth and Deimos Engenharia aims to develop Total Ecosystem Management of the InterTidal Habitats (TEMITH project/ESA). Within this project, high resolution optical Earth Observation data (such as Sentinel-2, Planet, WorldView and UAV) in conjunction with Convolutional Neural Networks and Deep Learning based feature extraction tools will be applied to assess the threats that have significant impact on the function and quality of intertidal habitats. The main threats to be analysed within the project are: sediment disturbance caused by human activities (e.g. dredging, fishing, inshore boat use etc.), sewage plume characteristics, litter accumulation points and algal mats detection (to be distinguished from seagrass beds).
The project will focus on the Solent Region of the South of England (separating mainland England with the Isle of Wight). This region is one of the most protected and exploited intertidal regions in the world. It only covers 150 km2, but it is an example of the multi-user and complex management structures typical of intertidal habitats across the globe.
The TEMITH system will utilise proven elements of the architecture already existing in Deimos. For visualisation purposes, elements of the SIMOcean (System for Integrated Monitoring of the Ocean) platform developed by Deimos Engenharia will be used, whilst EO product creation will be performed on the Service4EO platform developed by Deimos Space UK.

Coastal ecosystems are regions of remarkable primary and secondary productivity, biodiversity, and high accessibility. Apart from supporting numerous physical and biological processes, they also act as recreational, leisure, and tourism centres and overall provide a wide range of valuable ecosystems services. Italy comprises ∼8000 km of coastline of which approximately 25% consists of low-lying plains, some of which are subsiding with a potential for flooding. Over 9% of the coastline is man-made, comprising works close to the shore (3.7%), port structures (3%) and other structures superjacent to the coastline (2.4%) (ISPRA, 2016).
Timely, accurate, and spatially resolved environmental information from local to national scales are therefore relevant to support effective management for such landscape. In such a context, satellite remote sensing has been successfully applied for mapping coastal waters for about 50 years. Recently, the Italian Space Agency funded the costeLAB project (Contract Number 2017-I-8.0), which is aimed to develop an operational EO based service to support both the monitoring of the ecosystems and the response in case of natural and human-made hazards as well. The primary goal of the project is the creation of a virtual laboratory to analyze, develop and test applications and products for monitoring and managing risk coastal. The platform, equipped with all the tools to carry out these activities, also serve as a point of access to consolidated services in the pre-operational environment for the same thematic area. Aside the operational purposes, costeLAB also includes the development of research activities, which address challenges for assessment, quantification, and monitoring near-shore; e.g. the development of techniques for the retrieval of wind and wave fields based on SAR aiming to deal with issues related to high variability of winds and sensitivity of the wave systems to the bathymetry for shallow water.
Within this work, the principal outcomes of research activities will be presented by presenting algorithms development for multiple purposes such us:
- detection of Posidonia of Pianosa Island for limiting erosion processes;
- mapping of dune volumes from High resolution Cosmo SkyMed imagery of the biggest dune of the country nearby Piscinas (Sardinia);
- wind field and sea state mapping of the Northern Adriatic Sea using SAR images provided by Sentinel-1 in support to model forecasts;
- the evaluation of the wind field at high spatial resolution (~1 km) to investigate the interaction between the coastal orography and the wind flow, so important to model the surface ocean transport;
- monitoring of coastal ecosystem dynamics in terms of vegetation community type evolution, spanning from salt marsh to sandy dune habitats, in Venice Lagoon;
- mapping of suspended sediment pattern in the nearshore zone for monitoring coastal dynamics.
These themes are developed by using the latest generation of optical (Sentinel-2 and 3) and radar (Cosmo-SkyMed, Sentinel-1) satellite sensors are mostly used in this project, while imaging spectrometry support the development of algorithms for the ASI-PRISMA mission scheduled for next year.

Aquaculture, the farming of aquatic organisms, such as fish, shrimp and molluscs, is one of the fastest growing food production sectors, supplies more than half of the fish consumed today and is a primary source of animal protein and essential fatty acids for millions of people. With an average annual growth rate of 7.2 %, world aquaculture production increased from 13 million t in 1990 to more than 80 million t in 2016 and is currently valued at USD 231 billion. Rising demand for fish and seafood, stagnating capture fishery production and the development of new technologies and management practices fostered the rapid growth of aquaculture. Aquaculture makes an important contribution to the supply of protein-rich foods in human consumption of the growing world population, ensuring global food security. Land-based pond aquaculture is the predominant culture system in coastal regions and widely spread along coastal environments of East and South-East Asia, where new challenges emerge from global climate change, land use changes and environmental degradation.
We present a novel approach for the estimation of aquaculture production utilizing earth observation data and statistical data. An object-based image processing chain was developed to detect and map rectangular shaped aquaculture pond in coastal regions utilizing high-resolution Copernicus Sentinel-1 SAR time series, image segmentation methods and pond object shape features. Production and yield statistics were collected and harmonized at national, regional and local level to link earth observation based findings with production estimations based on a regression model. The combined methodological approaches include a large-scale assessment of land-based pond systems, regression analysis, and pond production estimates. It enables a quantification of aquaculture area at large spatial scale and provides valuable information on productivity of the derived systems. Space-borne data can assist aquaculture producing countries by enabling improved estimation, particularly for non-reporting regions. With the world’s growing population and rising demand for fish and seafood, this framework can be used as a basis to gain more knowledge with regard to the analyses of coastal land use dynamics and food security.

Sentinel-2 (S2) offers great potential coral reef mapping, with a 10 m spatial resolution which is comparable to the scale of heterogeneity in reef habitats, and data acquired in several spectral bands useful for discriminating the variety of colors of sand, algae, corals and the overlying water. With a revisit time of five days from two satellites in orbit, S2 may enable the observation of ecosystem processes such as coral bleaching, coral mortality or shifts to macroalgal domination.

The aim of the Sen2Coral study was to assess the capability of S2 for coral reef applications and to develop a freely available processing chain in ESA’ SNAP, to support not only coral reef monitoring and research but also for mapping of other shallow-water habitats from space. Implemented algorithms encompass image-based methods (e.g. pseudo-invariant-features) and physics-based approaches, such as optimization techniques for bio-optical models inversion. The algorithm performances were demonstrated for selected test areas (e.g. Lizard Island and Heron Island, Australia and Lampi Island, Myanmar) where reference field data have been collected in coincidence with S2.

Our findings confirm that S2 has excellent performance for meeting several essential coral reef scientific and monitoring objectives, including colonization patterns of corals, sea grass and marine habitats, and bathymetry mapping. In particular, the 2017 coral bleaching event on the Great Barrier Reef was clearly visible in two successive S2 images, and is one of the clearest demonstrations of bleaching visibility from a satellite to-date. Nevertheless, the acquisition plan needs adjustments to include coral reefs globally in remote oceanic reef environments, particularly those in European governed territories.

In studies of upwelling usually data from infrared and optical sensors are used which provide information on the sea surface temperature (SST) and the chlorophyll-a (Chl-a) concentration. In this paper, it is shown that also synthetic aperture radars (SAR) images can also give valuable contribution to such studies. Upwelling regions become detectable by SAR because they are associated with a reduction of the radar backscatter due to 1) a change of the stability of the air-sea interface or/and 2) the presence of biogenic slicks. While the change of the stability of the air-sea interface due to the presence of cold surface water in upwelling regions causes only small reductions of the radar backscatter, biogenic surface films cause strong reductions, usually of more than 10 dB. In areas of strong upwelling, the biological productivity is usually high due to increased nutrient supply from lower water levels. The biota living in this area secrete surface active material that ascends to the sea surface and forms there biogenic slicks. Although the biogenic surface films are only one-molecular layer thick, they damp the short-scale surface waves, which are responsible for the radar backscattering, as strongly as mineral oil films. The areas covered with biogenic surface films become visible on SAR images as areas of strongly reduced radar backscatter (visible on SAR images as dark areas) and thus can serve as indicators of upwelling. In particular, surface films often get entrained in the current field of oceanic eddies and thus render them visible on SAR images. This paper focusses on upwelling events at the west and south coast of Africa and at the south coast of Sicily, where SAR images provide additional information on upwelling. In particular, we address the issue how the kinetic energy generated by upwelling is dissipated. In the decaying stage of an upwelling event often filaments and small-scale eddies are formed. These phenomena are studied by using SST, Chl-a, and SAR data acquired over the waters off the southern coast of Sicily.