The use of formal ontologies can significantly enhance metadata descriptions of ecological data by providing richer and more consistent terminologies. Annotating data with ontology terms can both help users interpret data as well as enable advanced capabilities for data discovery and integration, e.g., by exploiting subsumption and part-of hierarchies as well as more formal constraints such as cardinality restrictions on properties and term equivalence. Many efforts are underway to engage scientists in the development of ontologies to represent common ecological and biodiversity concepts. Most of these projects leverage OWL (the W3C Web Ontology Language) as a standard XML syntax for representing and sharing ontologies. A key advantage of OWL is that it is supported by a wide range of generic tools, including editors, reasoning systems, query languages, and storage technologies. These tools, however, are primarily targeted at experts in knowledge engineering and software development. This is especially true with ontology editors (such as Prot?g?, OWLed, SWOOP, etc.), which allow for very detailed ontology specifications, but at the same time require a considerable amount of understanding of the underlying ontology formalisms and syntax. Thus, we see the lack of suitable ontology editing tools for scientists as one of the major barriers for more wide-scale adoption of ontologies in ecology.

Here, we present a novel approach for ontology creation that aims at being more intuitive for ecologists and that can be used to rapidly construct large ontologies for describing scientific data. Our approach is to allow scientists to use common spreadsheet tools to describe, in an intuitive way, different aspects of an ontology, and then to take these descriptions and convert them into full-fledged OWL ontologies using a software application called owlifier. An owlifier spreadsheet consists of a set of "blocks" that have a predefined template structure for users to fill in. Each non-empty row in an owlifier table constitutes a block, where blocks can be used to define ontology concepts (including sub-concepts), concept synonyms, overlapping concepts (i.e., that by definition are not disjoint), properties (including their domain and range), and inverse properties. We also provide blocks for plain-text descriptions of concepts and properties, and for referencing one or more existing ontologies (e.g., to extend an ontology or to define ontology articulations). Blocks can be sparse (inheriting from previous blocks), which simplifies the creation of large ontologies. Although not as expressive as full-fledged OWL ontologies, our approach can produce ontology structures essential for enhanced data discovery and integration, while at the same time providing a more accessible user interface for ecologists. Further, compared to many existing editors, our approach can be used to rapidly construct concept hierarchies, e.g., from long lists of keywords, within familiar spreadsheet software. We have field-tested the owlifier approach with ecologists and evolutionary biologists working with trait data, and found this application enabled them to quickly and easily comprehend and construct ontologies.