Perspectives on validation in digital soil mapping of continuous attributes—A review

We performed a systematic mapping of validation methods used in digital soil (DSM), order to gain an overview current practices and make recommendations for future publications on DSM studies. A search screening procedure, largely following the RepOrting standards Systematic Evidence Syntheses (ROSES) protocol, was carried out. It yielded database 188 peer-reviewed studies from past two decades, all written English presenting raster map continuous property. Review full-texts showed that most (97%) included some type validation, while just over one-third (35%) estimated uncertainty. Most commonly, combination multiple (existing) sampe datasets resulting maps were validated by single data-splitting or cross-validation. common essential information be lacking method descriptions. This is unfortunate, as lack sampling design (missing 25% studies) sample support 45% makes it difficult interpret what derived metrics represent, compromising their usefulness. Therefore, we present list details should provided also provide detailed summary 28 published studies, how values obtained whether can compared between attributes.