Oil-slimes appear on all the stages of oil mining and processing. As a rule, waste accumulation is performed in specially designed places and sludge tanks. Depositories of large petrochemical complexes occupy hundreds of hectares and present places of danger. During the accumulation and preservation of wastes in oil-slime depositories some natural transformations are observed. Usually, light liquid hydrocarbons concentrate in upper layers, while middle layers tend to contain the water. Heavy hydrocarbon fractions, tars and mineral particles gather in the bottom.

Structure investigation of depositories in connection with the oil-slime composition was carried out for several places of oil extraction, transportation and preparation in Samara region. Analysed depositories had different age and slime formation sources. Geometrical dimensions of depositories, thicknesses and densities of layers, as well as their chemical composition (water, diesel fraction, tars, chats and sulfur content) were determined. A quantitative characteristic describing quality changes of hydrocarbons is the ratio of diesel fraction to tars concentration. Gathered information, containing 34 variables, was analysed by means of Principal Component Analysis (PCA).

Four PCs were found to be enough to adequately describe the data. On the scores and loadings plots reveal the internal data structure that can be interpreted. One can distinguish groups of samples typical to the northern and southern clusters of oil fields of Samara region, as well as the areas typical to depositories of definite types. PC1 correlates with the sulfur content in all the layers and strippant, diesel content in the upper layer, layer density, middle layer thickness, mineral impurities content in all the layers and in the depository itself. PC2 is basically determined by water and tar content of the bottom layer, its thickness and age. PC3 is influenced by the ratio of diesel fraction to tars concentration. PC4 depends on the thickness of the bottom and upper layers and general water content. PCA and its interpretation was also carried out for each of the layers separately.

With the help of the developed model, unknown waste samples can be identified, e.g. assigned a geographic region or depositary type. The principal components, derived from the data analysis of various oil slimes, are put into the basis of a new oil-slime classification and discrimination system, suggested by the authors.