Textual Annotation and Provenance Ontology (TAPO) stores the results of NLP workflow processes and describes all the provenance information like tools used. TAPO is an extention of W3C Web Annotation Ontology dedicated to store the process that have generated the annotation. TAPO was first used to annotate French agricultural alert bulletins called : Bulletin de Santé du Végétal. This ontology was build during the D2KAB project.