In this paper, we aim to explore methodologies of developing multilingual metadata and thesauri for digital archives through case studies of two research projects. We first analyze the Chinese thesaurus and the English thesaurus to construct a methodological framework for Chinese-English lexical and semantic interoperability, proposing four main modules, including translation, mapping, localization, and creation. The next step is to examine the four steps involved in transcribing the Chinese metadata into the English form, including mapping, selection, translation, and vocabulary control. This paper further looks into several key issues faced by the two research projects, such as the varying degrees of semantic equivalence between Chinese and English vocabulary, metadata translation, and the quality of metadata.