An Observations Framework for data interoperability: Enabling descriptive regional data for soil and land management in the Australian context

Abstract

Regional resource management organisations and researchers require consistent soil data for effective knowledge transfer and monitoring. However, the data collected by soil practitioners can fall outside the data permissible in national and international interoperability efforts. This study fills a gap by delivering diverse data from farming groups, catchment managers, and university researchers via a unique ‘Observations Framework’ based on the observation and measurement (O&M) standard, not implementing soil domain model constraints.

One hundred and sixty-five individual soil properties were successfully described, resulting in the delivery of interoperable, region-specific soil data. A significant amount of data fell outside existing national soil model implementations, with only 30 % of chemical observation procedures related to the Australian Soil and Land Survey Handbook Series (ASLSH) standard, and 38 % lacking procedure descriptions altogether. Procedures for soil physics and biology measurements entirely outside the ASLSH. Contextual information, including sampling protocol, was generally absent or challenging to align with FAIR (Findable, Accessible, Interoperable, and Reusable) vocabularies.

Concluding lessons for delivering diverse regional FAIR soil data include the need to 1) Support standard metadata schema framework for FAIR soil data, understanding the value of context and data quality elements to end-users; 2) Develop and support well-governed FAIR vocabularies for soil through expert and community input, particularly on properties and sampling; and 3) Investigate the potential for AI in the delivery of soil data and information, including where graph technology and domain models may be beneficial, and where soil domain expertise is critical.