Three-Dimensional Quantification of Copepods Predictive Distributions in the Ross Sea: First Data Based on a Machine Learning Model Approach and Open Access (FAIR) Data

Abstract

Zooplankton is a fundamental group in aquatic ecosystems representing the base of the food chain. It forms a link between the lower trophic levels with secondary consumers and shows marked fluctuations in populations with environmental change, especially reacting to heating and water acidification. Marine copepods account for approx. 70% of the abundance of zooplankton and are a target of monitoring activities in key areas such as the Southern Ocean. In this study, we have used FAIR-inspired legacy data (dating back to the 1980s) collected in the Ross Sea by the Italian National Antarctic Program at GBIF.org. Together with other open-access GIS data sources and tools, it allows one to generate, for the first time, three-dimensional predictive distribution maps for twenty-six copepod species. These predictive maps were obtained by applying machine learning techniques to grey literature data, which were visualized in open-source GIS platforms. In a Species Distribution Modeling (SDM) framework, we used machine learning with three types of algorithms (TreeNet, RandomForest, and Ensemble) to analyze the presence and absence of copepods in different areas and depth classes as a function of environmental descriptors obtained from the Polar Macroscope Layers present in Quantartica. The models allow, for the first time, to map-predict the food chain per depth class in quantitative terms, showing the relative index of occurrence (RIO) in 3Dimensions and identifying the presence of each copepod species analyzed in the Ross Sea, a globally-relevant wilderness area of conservation concern. Our results show marked geographical preferences that vary with species and trophic strategy. This study demonstrates that machine learning is a successful method in accurately predicting the Antarctic copepod presence, also providing useful data to orient future sampling and the management of wildlife and conservation.

A Beacon in the Dark: Grey Literature Data Mining and Machine Learning Enlightening Historical Plankton Seasonality Dynamics in the Ligurian Sea

“…In this work, we applied the methodology described in Grillo et al (2022) [57] to the grey literature data from Fabiano et al (1988) [48], leveraging pre-existing scripts in the scientific literature [64] and environmental data obtained from both the original technical report [48] and the Copernicus Marine Service to produce distribution maps. The inclusion of environmental descriptors sourced from open-access outlets enriched the predictive capacity of the models, which were evaluated by using a Random Forest machine learning algorithm approach (regression method).…”

“…Additional proxies to increase the resolution and accuracy of the model were retrieved from the Copernicus Marine Service (https://marine.copernicus.eu/, accessed on 23 January 2024), NASA Earth Observations (https://neo.gsfc.nasa.gov/, accessed on 23 January 2024) and General Bathymetric Chart of the Oceans (https://www.gebco.net/, accessed on 23 January 2024) (See Supplementary Files S2-S4 for details). In our investigation, we employed the methodology proposed in Grillo et al ( 2022) [57] with modifications. We utilized open-source software including QGIS [58], R [59] and Ocean Data View (ODV) [60] data exploration, visualization, mapping and modeling, incorporating basemaps where necessary.…”

Vertical structure characterization of acoustically detected zooplankton aggregation: a case study from the Ross Sea

“…2000; Pane et al, 2004;Guglielmo et al, 2015;Smith et al, 2017;Bonello et al, 2020;Grillo et al, 2022;Kim et al, 2022), representing in austral summer more than 70% of the total community, while the remaining species are pelagic amphipods (Minutoli et al, 2023), Limacina helicina antarctica (Accornero et al, 2003;Manno et al, 2010), postlarval and juvenile stages of Pleuragramma antarcticum (Guglielmo et al, 1997;Granata et al, 2000;Granata et al, 2002;Granata et al, 2009), and calyptopis and furcilia stages of Euphausia crystallorophias (Guglielmo et al, 2009).…”

With super SDMs (machine learning, open access big data, and the cloud) towards more holistic global squirrel hotspots and coldspots

“…After removing duplicates (utilizing “removing duplicates” function in MS Excel), we also removed all records without a geographic location and a described species name 18 , after which the dataset was saved as a CSV file and imported in the data directory to be accessible for the cloud hardware. This data preparation is necessary for Maxent's algorithm, which sets it apart from more advanced and deep-learning methods such as boosting (TreeNet) or bagging workflow etc., that are better able to work with raw and messy data within which the corresponding Machine Learning algorithm seeks patterns 19 , 20 . This resulted in 665,529 final occurrence points which have been mapped and presented in Fig.…”