Carlo Mantovani ^1* Jay Pearlman ² Anna Rubio ³ Rachel Przeslawski ⁴ Mark Bushnell ⁵ Pauline Simpson ⁶ Lorenzo Corgnati ¹ Enrique Alvarez ⁷ Simone Cosoli ⁸ Hugh Roarty ⁹

• ¹ Institute of Marine Sciences, National Research Council of Italy, Lerici, Italy
• ² Institute of Electrical and Electronics Engineers (France), Paris, France
• ³ AZTI, Marine Research, Basque Research and Technology Alliance, (BRTA), Pasaia, Spain
• ⁴ NSW Department of Primary Industries, Fisheries, Huskisson, Australia
• ⁵ CoastalObsTechServices LLC, Virginia Beach, United States
• ⁶ Intergovernmental Oceanographic Commission (IOC) of UNESCO, Oostende, Belgium
• ⁷ Mercator Ocean (France), Ramonville-Saint-Agne, France
• ⁸ University of Western Australia, Perth, Australia
• ⁹ Department of Marine and Coastal Sciences, Rutgers, The State University of New Jersey, New Brunswick, United States

Ocean practices, intended as a wide spectrum of methodologies supporting ocean-related research, operations, and applications, are constantly developed and improved to enable informed decisionmaking. Practices start from the idea of an individual or a group and often evolve towards what can be called good or best practices. This bottom-up approach may in principle result in different paths for the evolution of each practice, and ultimately generate situations where it is not clear to a new user how to compare two practices aiming at the same objective, and determine which one is best. Also, although a best practice is supposed to be the result of a multi-institutional collaborative effort based on the principles of evidence, repeatability and comparability, a set of individual requirements is not yet defined in literature for a practice to be considered a good, better, and ultimately a best practice. This paper proposes a method for addressing those questions and presents a new maturity model for ocean practices, built upon existing maturity models for systems and software, developed and adopted in the last decades. The model provides attributes for assessing both the maturity of the practice description and its implementation. It also provides a framework for analyzing gaps and suggesting actions for practice evolution. The model has been tested against a series of widely adopted practices and the results are reported and discussed. This work facilitates a common approach for developing and assessing practices, from which greater interoperability and trust can be achieved.