Visualization of multiple attributes across a region is a complex problem for which several experimental tools have been developed. In some cases, the attributes correspond to events occurring at discrete locations; in others, the data are phenomena that vary continuously across a region. In the discrete case, the focus is often on using maps for geospatial referencing. Attribute blocks are a technique for visualizing continuously varying attributes across a region, letting you quickly locate patterns in large data sets. Attribute blocks can be displayed in a flat 2D projection without needing constant 3D dynamic rotations. This article describes a new technique based on attribute blocks, a dynamically configurable array of lenses through which users can visualize specific attribute values at given locations throughout a region. Users can exploit this ability to dynamically adjust the array's configuration as well as the attribute blocks' size and origin in this new exploratory tool for discovering patterns of interest in multivariate data sets