In order to facilitate descriptions of groundfish habitat over a large portion of the EBS shelf, the NMFS/AFSC has assembled a single comprehensive database of the point sample data (EBSSED; n=2,587) from all available sources. The database represents sediment variation over the study area with uncompromized (i.e., original) spatial detail. Textural data in the database are of two main types: 1) ... standardized statistics characterizing sample grain size distribution based on laboratory measurements (granulometric data), including percent composition by size grades (e.g., gravel, sand, mud) and size distribution parameters (e.g., mean size); and 2) sample descriptions from less exacting, more subjective visual/tactile observations, usually made in the field, establishing size-grade constituents. In addition the EBSSED database includes two descriptive fields which were each added to characterize sample grain size distribution by a single, standardized variable based on the original data. These fields classify samples according to gravel-sand-mud composition using high and low resolution schemes derived from Folk's (1954) classic ternary diagram. The high resolution scheme classifies 903 samples with detailed granulometric data into 15 textural classes, providing the greater detail regarding textural variation. The low resolution scheme (7 classes) is designed to allow unambiguous classification of nearly all samples (n=2457) including those with subjective visual/tactile descriptions. It represents the maximum number of samples according to a single common variable and thus provides the most spatially detailed data for the study area, albeit at the expense of some of the textural detail for samples analyzed in the laboratory. Overall, the EBSSED database is the most comprehensive and detailed source of information about surficial sediment textures in the EBS study area. Patterns observed in the data generally agree with large-scale textural maps and summaries by previous investigators, particularly a general pattern, with exceptions, of decreasing average grain size with increasing depth and distance from shore. However, those previous large-scale works spatially smoothed data for the study area from smaller, more sparsely distributed sets of samples. The EBSSED database preserves potentially important fine-scale variation.