Project Overview

Small pelagic fish, like anchovies (Engraulis encrasicolus) and sardines (Sardina pilchardus), are a valuable marine resource that affects both humans and the natural environment greatly. These species are the linking factor between lower and higher trophic levels and play a major role in balancing the health of the marine ecosystem. Additionally, they are one of the major food resources for coastal environments, especially for the Mediterranean countries, holding both economic and cultural importance. Due to anthropogenic activities though these species are threatened. Climate change and overfishing are identified as the major pressures that these species are threatened by, and successful fishery management strategies are needed to protect these resources.

The Marine Remote Sensing Group has made efforts for the establishment of innovative tools to assist fishery management efforts with powerful monitoring tools. Through satellite remote sensing Essential Oceanographic Variables (EOVs) are utilized to map the potential spatial distribution of small pelagic fish in the Greek Seas. The trained models scope to produce daily results that can be distributed to the respective management authorities through a dedicated Geospatial Web Service.

The main objectives of the project are summarized as:

• Development of methodology for acquisition, processing, and production of information regarding small pelagic fish spatial distribution from satellite data.
• Model training for predicting the spatial distribution of target species (Spatial Distribution Modeling).
• Implementation of a Geospatial Service for publishing and disseminating the research outcomes.

A new approach in fisheries monitoring

Fish population distribution monitoring is a key aspect and the basis for an effective fisheries management strategy. Accurate and robust monitoring can assist in the evaluation of the health of the stocks and in turn, contribute to targeted decision-making for the conservation actions where they are needed. Traditional monitoring practices rely on field measurements, like acoustic biomass surveys and fishing vessel onboard observers, which even though precise have limited spatial and temporal coverage. As a rising technology in the field of oceanography in the last decades, remote sensing has been proven a powerful tool in marine monitoring. However in fisheries monitoring, operational applications usually rely on simple data visualization that does not convey the interactions between the target species and their environment, or they require non-expert data interpretation.

The proposed application aims to bridge the gap between scientific work and real-life applications. With a simple and user-friendly interface, users can access near real-time or time series datasets of small pelagic fish SDM outputs  and auxiliary satellite-derived data, benefiting both the scientific community and management authorities. The platform also incorporates weather and sea-state forecasting, which can contribute to maritime safety and planning.

The Data

The key data for modeling the spatial distribution of anchovies and sardines are obtained from the Sentinel-3 satellite mission. These data include environmental proxies such as chlorophyll-a concentration, Sea Surface Temperature (SST), and information on mesoscale ocean fronts. Chlorophyll concentration and primary productivity are pivotal proxies for monitoring the creation of the food chain base. SST monitoring helps identify areas of upwelling, crucial for pelagic fish behavior and breeding area preference. Mesoscale ocean fronts are mapped daily, using a gradient-based algorithm developed by the MRSG team (Spondylidis et al., 2020). Additionally, in-situ data from acoustic surveys conducted by the Hellenic Centre of Marine Research provide additional biomass measurements of anchovy and sardine populations in the North Aegean Sea, aiding in model training. All the datasets that are published in the Geospatial Web Service are gap-free, through an interpolation procedure (DINEOF), and are ready to use by non-GIS and remote sensing expert users.