The transport of ship emissions in the Strait of Malacca using a high-resolution WRF simulation and low-resolution GDAS data coupled with HYSPLIT

This research describes and quantifies the role of deep convection within the Strait of Malacca (hereafter referred to as the “Strait”) on the transport of ship emissions. The Strait is located in Southeast Asia between the Indonesian island of Sumatra and the Malay Peninsula and is one of the world's major shipping lanes. The vessels emit large quantities of air pollutants. The Strait also is a region of intense convection that can transport the pollutants from the surface to great heights where winds are stronger than near the surface. These winds can lead to long-range transport of the ship emissions. This study utilizes both the Weather Research and Forecasting (WRF) Model with a 2 km horizontal grid spacing and the HYbrid Single Particle Lagrangian Integrated Trajectories model (HYSPLIT_4) to simulate the vertical and horizontal transport of the ship-borne pollutants. Results from the higher-resolution WRF simulations are evaluated against the coarser-resolution (1° horizontal grid spacing) Global Data Assimilation System (GDAS) data provided by the Air Resources Laboratory (ARL). World Wide Lightning Network (WWLLN) observations reveal that the Strait has a pronounced diurnal cycle of lightning with a nighttime (1900–0700 LT) maximum that is 2–3 times greater than the daytime (0700–1900 LT) maximum over the surrounding landmasses. WWLLN observations also reveal that the Strait region has a seasonal cycle of deep convection that is influenced by the Intertropical Convergence Zone and is out of phase with the Asian monsoon. April is the month with the most lightning, followed by October. Conversely, February is the month with the least amount of lightning. Therefore, these three months are the focus of this study. The Emissions Database for Global Atmospheric Research v4.2 is used to determine the average emissions rate of CO from ships within the Strait. A mass is assigned to each HYSPLIT particle to produce a three-dimensional representation of CO concentrations.

HYSPLIT results based on WRF as the meteorological input reveal that more CO is transported to the upper troposphere/lower stratosphere (UTLS) during April than any other month. Efficient transport to the UTLS also occurs during October, but in smaller concentrations than during April. CO transport during February is primarily to the lower to middle troposphere. The effect of model resolution on transport is shown by comparing WRF-derived trajectories to GDAS-derived trajectories. The coarser-resolution GDAS-derived trajectories remain close to their point of release after 120 h. The higher-resolution WRF-derived trajectories exhibit more horizontal and vertical transport than those from GDAS. This has major implications on simple dispersion forecasts. These results show that the coarser-resolution GDAS data, which are suggested for use by ARL, do a poor job of particle dispersion in the meteorologically and topographically complex region of the Strait of Malacca, and likely do so in other similarly complex regions. This can lead to inaccurate forecasts of air quality for interested users, as well as misrepresentation of the impact of local anthropogenic emissions (such as from ships) throughout the depth of the atmosphere.