An efficient access technique is one of the most important requirements in web-based GIS. With levelized spatial data we can access spatial data efficiently, because of no access to the fully detailed spatial data. Previous spatial access methods have disadvantages, such as data redundancy and low search performance, for levelized spatial data. To solve it, a few spatial access methods for levelized spatial data, are proposed. However these methods support only a few kinds of levelized data, i.e, data through a selection operation and a simplification operation. For the effects, we propose a new spatial access method supporting all kinds of levelized spatial data. A new access structure is designed and implemented. Experiments are then performed on real data sets. The results show that our method offers both high search performance and no data redundancy.