GSA Connects 2021 in Portland, Oregon

Paper No. 91-10
Presentation Time: 9:00 AM-1:00 PM


ROSTAMI, Masoud A., Reno, NV 89507; Department of Biology, University of Nevada, Reno, Reno, NV 89557, FRONTALINI, Fabrizio, Department of Pure and Applied Sciences, Università degli Studi di Urbino “Carlo Bo”, Urbino, NV 61029, Italy, GIORDANO, Patrizia, Istituto di Scienze Polari, Consiglio Nazionale delle Ricerche, Bologna, NV 40129, FRANCESCANGELI, Fabio, University of Hamburg, Institute for Geology, Centre for Earth System Research and Sustainability, Hamburg, NV 20146, Germany, VIRGINIA ALVES MARTINS, Maria, Rio de Janeiro State University (UERJ), R. São Francisco Xavier, 524, LabMicro 4037F, Rio de Janeiro, 20550-900, Brazil, DYER, Lee, Department of Biology, University of Nevada, Reno, Reno, NV 89557 and SPAGNOLI, Federico, School of Science and Technology, Geology division, University of Camerino, Camerino, 62032, Italy

In recent decades, anthropogenic activities have increasingly affected community structure of marine organisms including benthic foraminifera. Although, numerous studies on benthic foraminifera as bio-indicators have been published, very few of these studies have determined thresholds over which a parameter significantly induces changes on benthic foraminiferal assemblages, on species or on ecological indices.To discover main environmental variables affecting the regional distribution, assess magnitude of species response and to evaluate their influence on ecological indexes, we applied novel machine learning approaches including Gradient forest and Random forest modeling to understand the relationship between abiotic (environmental) and biotic (foraminifera) variables in a costal marine area in the central Adriatic Sea (Italy). The analysis was accomplished on 26 species of benthic foraminifera and a set of 38 environmental variables. The result shows the variables including silt, Pollution Load Index (PLI), sand and clay have the highest impact on the distribution of benthic foraminifera in this area. In addition, we identify the most important environmental variables and their thresholds that strongly influence ecological indexes (i.e., Exp(H`bc), Foram Stress Index, Tolerant Species index, Foram-AMBI) used for the EcoQS determination. Our findings define which variables play a major role and at which extent on the selected indexes in the area and therefore support their applications to set the sediment quality and environmental standard for marine conservation. We belive the application of such approach represents a useful tool for policymakers to survey the diversity of marine organisms and to improve the ability to protect and restore marine ecosystems by identifying environmental drivers and defining their thresholds.