What is one advantage of hierarchical clustering over standard k-means?

A. It is unaffected by variable scale B. It always finds objectively true clusters C. It can use no distance measure D. It provides a nested structure without requiring one fixed k at the start