On the Probabilistic Distribution of External Pitting Corrosion Rate in Buried Pipelines

External pitting corrosion constitutes the degradation mechanism responsible for about 66% of the incidents reported in the last decade for oil and gas pipelines in Mexico. Thus, major efforts are underway to improve the characterization and modeling of pitting corrosion of buried pipelines. Special attention is devoted to estimate the average corrosion rate and corrosion rate variance because they are the key parameters in the estimation of the trend in pipeline reliability. This work presents the results of field and simulation studies in which soil and pipe data were gathered together with the maximum depth of external corrosion pits found at more than 250 excavation sites across southern Mexico. The distributions of parameters such as chloride, bicarbonate and sulfate levels, resistivity, pH, pipe/soil potential, humidity, redox potential, soil texture and coating type have been used to predict the distribution of pitting corrosion rate of pipelines in contact with clay, clay-loam and sandy-clay-loam soils. The time dependence of the pitting corrosion rate was fitted to a power law through a multivariate regression analysis with the maximum pit depth as the dependent variable and the pipeline age and the soil and coating properties as the independent variables. Monte Carlo simulations were conducted in which random values drawn from the distributions fitted to the field data were used to evaluate the power law model proposed for the corrosion rate. For each soil type, the distribution that best fitted the corrosion rate data was found. The results of this study will provide reliability analysts with a more accurate description of the growth rate of external corrosion pits. It is expected that this information will positively impact on integrity management plans addressing the threat posed by this damage mechanism.