Interoperability in Marine Sensor Networks through SWE Services: The RITMARE Experience

Introduction

The rapid growth and development in different fields related to sensors, such as in micro electro mechanical systems (MEMS), wireless communication, and digital electronics has, together with the huge increase of devices due to the decrease of their costs, led to a shift in environmental disciplines (Papp & Hakkesteegt, 2008): While in traditional monitoring data collections were not managed according to specific strategies, in sensor/processing networks more stages of data life cycle are devoted to trace data acquisition processes and distribution. Environmental Sensor Networks allow for increasing the number of observations and measurements that are made accessible: This improves the understanding of complex and fragile environments such as coastal zones. Indeed, in order to exploit the variety and volume of observations, it is necessary to adopt an integrated approach on distributed, mobile, fixed, and asynchronous sensors from different networks, as suggested by Hart and Martinez (2006). In the last few years, for the sciences of the environment, several initiatives were undertaken to allow for data discovery and accessibility, as ruled by standard protocols and application programming interfaces. In the past, the research work was characterized by a phase where the collected data were exchanged only after a painstaking formatting and harmonisation, according to a shared format. More recently, data integration is considered the prerequisite to advanced environmental monitoring and sustainable development, but the capabilities of different systems to work together has been realized only recently (Barnaghia, Ganza, & Abangara, 2011; Havlik et al., 2011) and, under many viewpoints, it is still a challenge. In particular, managing different and heterogeneous data sources, of several types, is an actual critical requirement for environmental monitoring, in particular in the field of marine sustainable development, for which the integration of information from different disciplines and domains (e.g. economic indicators, traditional GIS sources, output of oceanographic models, time-series of sensors monitoring the environment continuously) is a key point (Diviacco & Leadbetter, 2017, chapter 2 this book). In a vision for a Digital Earth in the modern era, Al Gore (1998) and, more recently, McKee (2015) argued for the creation of interoperable systems for environmental science, buisiness and policies and for free and open standards to implement them.

To achieve complete data integration, the actuation of concepts like interoperability are fundamental in order to realize a linkup among data and to enable across-the-board analyses for the advice of decision-makers and public stakeholders dealing with sustainable development policies. The dimensions of such linkup cover spatial (e.g. location, depth, footprint), temporal (e.g. time stamps, time zones), and thematic (e.g. parameter, measurement unit, domain, quality, methods) attributes. The involved domains and community of practices have their consolidated traditions in data management. The emerging sensor data deluge (Baraniuk, 2011) is requiring novel methods and tools for data management. Several marine observing systems (GOOS¹, EuroGOOS², IOOS³, IMOS⁴), and marine research initiatives (ODIP⁵, EuroFleets⁶, SeaDataNet⁷, FixO3⁸, NeXOS⁹, ScheMA¹⁰, JericoNext¹¹, EnvEurope¹², RITMARE¹³) have focused on the requirements for interoperable access to observations from marine sensors. This topic has become more and more important with the emergence of climate change as a world-class issue, because the detection of sudden anomalies is now a priority for early warning systems, and the monitoring of environmental parameters is fundamental in order to understand ongoing phenomena and to create sustainable development scenarios.