Characterization of key heavy petroleum feedstock properties is important for optimizing the production of clean transportation fuels. A correlation-based method has been developed to accurately estimate the aromatic carbon and total hydrogen content of heavy petroleum feedstocks. The new characterization methodology is significantly superior to existing methods in that it is applicable to all types of heavy petroleum feedstocks, uses only two (most) commonly measured bulk petroleum properties and is more accurate. The new method has been validated via analysis of an unprecedented number (three hundred and fifty four) of heavy petroleum feedstocks with an extremely broad range of properties. (C) 2008 Elsevier B.V. All rights reserved.