Using AIS data to calculate emissions inventories for small commercial watercraft

Abstract

The shipping industry is heavily reliant on the use of fossil fuel and contributes significantly to global emissions of carbon dioxide (CO2), nitrogen oxides (NOx), sulphur dioxide (SO2) and particulate matter (PM) resulting in deleterious impacts upon the climate, human health and the environment. A large proportion of global fishing and other small commercial vessels (< 100 GT) are omitted from global shipping emissions inventories, leading to potentially significant underestimation of emissions from the shipping sector. Effective quantification of shipping emissions requires quality data and sophisticated methods. This thesis introduces a new method for the calculation of emissions inventories for small commercial vessels that utilises Automatic Identification System (AIS) data, a highquality source of activity data for modelling atmospheric emissions from ships. The methodology offers a novel approach to activity sampling for modelling the emissions of vessels that cannot be directly matched to AIS data. A new speed calculation methodology based on the AIS data is also developed. An approach is also introduced for the detection of pushing and towing operations of vessels such as dredgers and trawlers in order that corrected engine load estimates can be applied for these operations. A case study emissions inventory for the year from May 2012 to May 2013 is calculated for UK fishing vessels. This is compared with the annual emissions calculated using a fuel-based methodology. Fuel use calculated using the activity-based methodology is 270.8 kt, which is slightly higher than the fuel-based methodology which yielded results of 251.8 kt. The activity-based method produced a CO2 emissions estimate of 864.3 kt, compared to 803.3 kt for the fuel-based approach. An analysis of uncertainty and sensitivity shows that activity sampling and emission factor uncertainty produce significant but unbiased uncertainty in results. However, uncertainties in values used to parameterise engine load calculation are found to generate potentially significant bias in results, highlighting the importance of calibrating model input parameters to ensure that sensible results are produced. Overall uncertainties in fuel use and emissions calculated using the activity-based method are found not to exceed ±6% at the 95% confidence interval. The close alignment of the results of the fuel-based and activity-based methods and the relative stability of results shown by the uncertainty analysis indicates that an AIS-based methodology with activity sampling is a viable approach for the calculation of emissions from small commercial vessels. The finding that 43.5% of UK fishing fleet emissions are produced by small vessels (< 100 GT) supports the claim that omitting these vessels from emissions inventories could lead to a significant underestimation of shipping emissions.

More information

Identifiers

ORCID for Ian Williams: orcid.org/0000-0002-0121-1219

Catalogue record

Export record