The　acquisition　of　terminology　in　the　field　of　water　environment　is　the　key　to　constructing　the　ontology　of　related　fields,　and　it　is　also　an　important　part　of　information　extraction　and　information　retrieval.　This　paper　proposes　algorithms　based　on　rules　and　statistics　to　extract　water　environmental　terms.　Firstly,　use　the　N-gram　algorithm　to　segment　the　pre-processed　text.　Secondly,　use　relevant　rules　to　filter　the　vocabulary.　And　then,　use　improved　mutual　information　and　adjacency　entropy　to　filter　the　vocabulary　to　obtain　candidate　term　words.　Finally,　use　TFIDF　to　select　terms　related　to　the　field　of　water　environment.　Experiments　show　that　this　method　has　achieved　good　results　in　extracting　terminology　in　the　field　of　water　environment.　©　2020　ACM.