Maintaining environmental context and geoprivacy protection in agriculture

To achieve sustainable agriculture and food security there is an urgent need to share agricultural data with a range of relevant stakeholders; however, to reduce the risk of identification, spatial data must be obfuscated prior to sharing. To-date, most obfuscation methods that have been developed do not consider a) the areal nature of field-level data and b) the differing environmental conditions at the original and obfuscated sites. To address these issues, we developed the Polygon-based Environmental Similarity Obfuscation Method (PESOM) to provide geoprivacy protection and guarantee that obfuscated data will retain the same environmental conditions as the original data. PESOM was developed using an unsupervised clustering algorithm and seasonal climate data, before being applied to the Nutrient Management Plan (NMP) online in Ireland. PESOM satisfied high level of geoprivacy protection and absolute environmental clustering preservation, with no false-identification and non-unique obfuscation risk. It provided a low level of distribution preservation and correlation preservation, large location displacement and subsequently low local analytical accuracy. PESOM is a significant advance on existing obfuscation techniques in agriculture data and will allow the sharing of data to be used widely for agri-environmental purposes, a current limitation of existing methods. The results of this research should be of wide interest to those working in agri-environmental research and computer science, and be of relevance to researchers, data managers, and practitioners.