An Ontological Analysis of Water Features

Abstract

GIScience 2016 Short Paper Proceedings AN ONTOLOGICAL ANALYSIS OF WATER FEATURES Boyan Brodaric 1 , Torsten Hahmann 2 , Michael Gruninger 3 Geological Survey of Canada, Ottawa, Canada K1A0E9 Email: boyan.brodaric@canada.ca School of Computing and Information Science, University of Maine, Orono, ME 04469, USA Email: torsten.hahmann@maine.edu University of Toronto, Toronto, Canada M5S3GB Email: gruninger@mie.utoronto.ca Abstract Water features are understood and represented heterogeneously in a wide variety of settings, including in data standards, polices and regulations, and amongst different cultures and languages. Ontologies aim to reduce this heterogeneity by representing commonalities across such settings. In this paper we build upon existing work in hydro ontologies and philosophical ontology to enhance the conceptualization and representation of water features. This results in a new taxonomy for water features, which helps identify and organize their essential parts. The results are represented as a first-order logic extension of the DOLCE ontology as well as an independent ontology fragment, and these are intended to serve as a reference ontology for the hydro domain as well as an aid to data interoperability. 1. Introduction Water features are entities that are essentially composed of water and variably other things. Prototypical examples include lakes, rivers, puddles, and clouds, but can also include aquifers. They play a key role in many human activities, such as those related to health, climate and weather, agriculture, energy, recreation, and transportation. Research and operations in these domains are heavily dependent on digital representations of water features, but the inherent conceptualizations can vary widely. Examples of heterogeneity abound, and can be found when comparing international water data standards (Boisvert & Brodaric 2012; Dornblut & Atkinson 2013; INSPIRE 2013; 2014), national catalogs of hydrographic features (Duce & Janowicz 2010), ontological considerations (Galton & Mizoguchi 2009; Santos et al. 2005; Sinha et al. 2014; Wellen & Sieber 2013), and database structures (Maidment, 2002; Strassberg, et al., 2011). This is problematic as it inhibits some uses, especially their integration, which is typically an important precursor to regional scientific analysis such as water availability, or complex societal decision-making such as water allotment. At the heart of the problem is a disparity about the fundamental nature of a water feature, as different aspects are variously emphasized in distinct conceptualizations. These aspects include most notably the water body, its water matter, its container or void (the space it occupies), or even an immaterial spiritual entity (Mark et al. 2007; Wellen & Sieber 2013). The emphases exist perhaps to enable diverse uses, for example, reasoning about the presence of a water body facilitates navigation of rivers that might have wet or dry segments; reasoning about the constitution and flow of water matter informs contamination scenarios, as does reasoning about the permeability of the container; and reasoning about the container’s void informs storage and overflow scenarios. Yet, it is still somewhat surprising that an entity of such significance is so widely construed and often vaguely defined. In this paper we undertake an ontological analysis of water features and develop a new conceptualization and representation that encompass the key aspects. This is achieved by extending and uniting two significant approaches to physical ontology, namely Hayes’ ontology of liquids (1978) and Fine’s theory of parts (1999). The results contribute to the design of the HyFO reference