Offshore drilling operations generate rock cuttings and spent drilling fluids wastes during the well drilling phase. The wastes contain toxic substances that are harmful to the marine ecosystem. Discharge limitations and restrictive guidelines in different jurisdictions of the world are under development. Pollution prevention by encouraging synthetic based fluids (SBFs) in place of traditional oil based fluids has been supported in recent years. Despite having environmentally benign characteristics. SBFs associated wastes still have a certain amount of pollutants due to barite and contamination with formation oil.

The main aim of this study was to develop a risk management framework for determining the best drilling waste discharge scenario for disposal in the marine environment. The specific objectives of this research were:​ (1) development of probabilistic contaminant fate modeling methodology using fugacity and aquivalence based approaches;​ (2) development of an ecological risk assessment methodology using probabilistic concepts;​ (3) development of human health cancer and non-cancer risk assessment methodologies using probabilistic concepts;​ (4) development of a fuzzy composite programming framework for risk management by integrating environmental risk, cost estimates and technical feasibility for various treatment options;​ and (5) an application of the developed risk management framework to a hypothetical case study.

Fate modeling in this research was performed using fugacity and aquivalence based concepts. A chemical specific approach was employed for contaminant fate modeling. A steady state non- equilibrium water and sediment interaction model with probabilistic inputs was used to determine the contaminant concentrations in the water column and pore water. The uncertainty and variability in the model inputs were expressed by the statistical distributions. The concentrations in the water column and pore water were estimated using Latin Hypercube sampling (LHS) based Monte Carlo (MC) simulations. The concentrations in the water column and pore water followed lognormal distributions. The estimated parameters of lognormal distribution for known discharge conditions were used for performing multiple regression analyses. The highest 95th percentile was used as the predicted environmental concentration (PEC). The uncertainties in the PEC were expressed by the coefficients of regression models.

The PEC values were converted into exposure concentrations (EC) by adjusting for bioavailability and probability of exposure. The whole ecological community was defined as assessment endpoints. The toxicity assessment analyses were based on the lognormally distributed predicted no effect concentrations (PNEC). The lowest 10th percentile on PNEC distributions was used as a safety level or PNEC criteria value. Bootstrapping was performed on original PNEC data to determine the uncertainty in the PNEC criteria values. The hazard or risk quotients (HQ/RQ) were calculated by dividing EC with PNEC criteria values. The CHARM model's approach was used to convert HQ/RQ into risk estimates for each contaminant. The composite ecological risk for drilling waste was determined by integrating the individual risk estimates assuming statistically independent events.

The human health risk methodology was based on the consumption of contaminated seafood. A probabilistic framework for human health risk assessment was developed for cancer and non- cancer risk estimates. The chronic daily intake rate (CDI) was established based on fish ingestion rates, lipid content, bioconcentration factors, exposure duration, exposure frequency, and averaging time. The LHS based MC simulations were performed to estimate the CD I. Arsenic was the only proven human carcinogen in the drilling waste stream. The composite hazard index for non-cancer risks was calculated by simple addition for a given exposure scenario.

A risk management methodology using fuzzy composite programming (CP) was developed. The costs for treatment, drilling fluid loss due to discharge and ecological and human health damages were estimated. The technical feasibility of various solid control devices was also studied from a performance viewpoint. The environmental risk reductions, cost saving and technical feasibility indices were grouped using CP methodology. A double weighting scheme was employed in CP. The final utility and centroidal values of the system improvement indices were calculated through Chen and Yager fuzzy ranking methods, respectively to determine the best management alternative.

The risk management framework was applied to a hypothetical case study on the East Coast of Canada. Five discharge scenarios, or management alternatives including 10.0%, 8.5%. 7.0%, 5.5% and 4.0% attached base fluids to wet cuttings, were selected for the analysis. The 4% attached base fluid option was found to be the best management option for the first trial when risk and cost were given equal weight and technical feasibility was allotted one-third of the weight. The 7.0% discharge option was the second best management option and that was followed by the 5.5%, 8.5% and 10.0%. respectively. The sensitivity analysis was performed using four different weighting schemes to account for human subjectivity. It was concluded that 4.0% was the best management alternative for the discharge of drilling waste in the marine environment. The other better management alternatives were the discharge options of 7.0% and/or 5.5% attached base fluid to the wet cuttings (approximately 5-6% of dry cuttings).

This study has introduced a new concept of integrating probabilistic fate modeling with ecological and human health risk assessment methodologies within a risk management framework to determine the best management alternative under conflicting objectives. This study has provided a framework for a decision support system for the selection of the best drilling waste marine discharge option under any known regulatory and technical constraints.