The Crude Oil Analysis (COA) database contains the digital data compilation of 9,076 crude oil analyses from samples collected from 1920 through 1983 from the United States and around the world and analyzed by the United States Bureau of Mines (National Institute for Petroleum and Energy Research, 1995). Two laboratories (Bartlesville, Oklahoma, and Laramie, Wyoming) performed routine crude oil analyses by a standardized method, and the data were originally reported in more than 50 reports by the Bureau of Mines.

Analyses include specific gravity, API gravity, pour point, viscosity, sulfur content, nitrogen content, and color of the crude oil, as well as the bulk properties of the distillation cuts.

The data were digitized in the late 1970s and a database retrieval system was implemented in 1980 and made available to the public. The Department of Energy (DOE) updated this system in 1995-96 with public access through a dial-up bulletin board system. The database was operated by the National Institute for Petroleum and Energy Research (NIPER) in Bartlesville, Oklahoma. A stand-alone version of the database (COADB) was available in 1995 in the form of a series of tables in Foxpro (.dbf) format. In 1998, an updated version of COADB was available on the NIPER website that included a Microsoft Access 97 version of the database called "coadb.mdb". The file contains more tables than the original 1995 version but we believe the number of oil samples and the amount of raw data are the same. The additional tables contain text translations for codes used in other tables regarding color, county, laboratory, formation, geologic age, lithology, and state name. Sample location information is generally inadequate to identify the specific well in most cases. The sample location information lacks lease name and in many cases well number and section-township-range. In rare cases, the latitude and longitude are given. A 2002 version was provided by the National Energy Technology Laboratory.