Existing online hashing methods generally employ the Hamming distance for similarity evaluation, which leads to the information loss of data location. Candidate images may have the same Hamming distance but different similarity from the query, which reduces the retrieval accuracy. Especially for non-stationary data environments, concept drift problems are prevalent. The location information loss makes it more difficult for capturing the distribution changes in data environments. To alleviate above concerns, Incremental Hashing with Asymmetric Distance (ICHAD) is proposed in this paper for image retrieval in non-stationary environments. In ICHAD, the online asymmetric distance based on learned hash codes is employed for the similarity evaluation. It preserves the location information of data more accurately and is computed efficiently without accessing the old data. Experimental results show that ICHAD outperforms existing hashing methods in various non-stationary data scenarios with concept drift.