Demonstration of Approaches to Quantifying Ferry Particulate Matter Emissions and Uncertainty: A Case Study of a North Carolina Vessel

Abstract Description: Diesel-powered ferry engines emit particulate matter (PM), posing adverse health impacts. Quantifying ferry emissions is crucial for assessing environmental impacts and improving air quality. Additionally, quantifying the uncertainty in ferry emissions is crucial for understanding the potential range of emission estimates and ensuring their reliability. This study aims to demonstrate an approach for quantifying ferry PM emissions and their uncertainty using statistical simulations. A case study is used to illustrate the approach based on a specific North Carolina ferry. The vessel is operated with two main engines for propulsion and one auxiliary engine for onboard services. Ferry emissions were estimated as the sum of the product of emission factors and activity factors for each engine. Emission factors (g/kWh) were identified by matching key engine characteristics with the U.S. Environmental Protection Agency’s certification databases for marine and nonroad engines. Activity factors (kWh) were estimated by multiplying the engine rated power, percent load of rated power, and operating hours. Percent load data for the main engines were obtained from prior data collection on similar engines operating on the same route, conducted by the research team. These data were scanned and recorded from the engine’s electronic control module via a datalink scan tool. Percent load for the auxiliary engine was assumed to follow a uniform distribution between 40% and 60%, based on literature. Annual vessel operating hours for 2019 to 2023 were provided by the North Carolina Department of Transportation. Non-parametric bootstrap simulations are used to randomly sample emission factors and activity factors from their respective distributions for each engine. This approach is used to generate randomized non-parametric distributions of the annual emission estimates, from which the mean and uncertainties (e.g., 95% uncertainty ranges) can be quantified. Correlation analyses are conducted to assess the sensitivity of factors affecting annual emission estimates. Probabilistic results and sensitivity analysis for the example ferry will be presented. This study establishes an approach for estimating PM emissions and associated uncertainty for a typical vessel, which is applicable to other pollutants and vessels.