Single-Vessel Plume Dispersion Simulation: Method and a Case Study Using CALPUFF in the Yantian Port Area, Shenzhen (China)

Shubin Bai, Yuanqiao Wen, Li He, Yiming Liu, Yan Zhang, Yu Qi, Weichun MaInternational Journal of Environmental Research and Public Health 2020 · DOI

TL;DR

This paper introduces a novel method for simulating SO2 dispersion from individual ships using the WRF/CALMET/CALPUFF modeling system, leveraging Automatic Identification System (AIS) data for high-resolution emission inventories. The approach successfully reproduces observed SO2 concentrations in the Yantian Port area, demonstrating that single vessels can significantly impact local air quality, especially within a few kilometers of their trajectory.

What Problem Does This Paper Address?

Maritime transport is a major source of air pollution, particularly SO2, with significant implications for regional and global air quality, as well as local ecosystems and human health. While previous studies have modeled port-wide emissions, there's a need for more granular simulations that can capture the impact of individual vessels. This is crucial for understanding localized pollution hotspots and developing targeted emission control strategies, especially in busy port areas where multiple ships operate. The challenge lies in accurately representing the dynamic nature of ship movements and their associated emissions, which are often treated as aggregated sources in broader models.

Methods

The authors developed a single-vessel atmospheric pollution diffusion model by integrating several established tools:

  • WRF (Weather Research and Forecasting Model): Used to generate high-resolution meteorological fields, providing essential atmospheric conditions for dispersion modeling.
  • CALMET: A meteorological model that further refines WRF outputs, assimilating local weather station data to create a more accurate, spatially resolved meteorological field for the study area.
  • CALPUFF: A Lagrangian puff dispersion model adapted to simulate the movement and dispersion of SO2 plumes from individual ships. The key innovation here is modifying CALPUFF, originally designed for stationary sources, to handle moving sources by discretizing ship trajectories into a series of point sources.
  • Automatic Identification System (AIS) Data: Crucial for creating detailed, single-vessel emission inventories. AIS data provides real-time information on ship location, speed, and status, allowing for the calculation of pollutant emissions based on engine power, load factor, and emission factors.

The model calculates SO2 emissions from both main and auxiliary engines, considering factors like load factor and fuel control. The plume lifting height was also adjusted using a national standard to better match observed phenomena.

Datasets

  • Automatic Identification System (AIS) Data: Provided by the Shenzhen Yantian Maritime Safety Administration, detailing ship navigation information.
  • Ship Information: Basic ship data, including engine capacity and funnel parameters, collected through field surveys.
  • Meteorological Data:
    • Ground station data from Yantian Port, Beizijiao, and Shatoujiao.
  • 3D vertically stratified meteorological data from WRF.
  • Land-use Data: Global Land Cover Characteristics (GLCC) database for Eurasia-Asia.
  • Terrain Elevation Data: Downloaded from USGS.
  • Shore-based Monitoring Data: SO2 concentration measurements from a monitoring instrument located approximately 2 km off the coast of Shenzhen, collected during a field experiment from June 22 to July 5, 2018.

Key Insights and Ideas

  • Individual Vessel Impact: The study highlights that even a single ship's emissions can have a significant localized impact on air quality, particularly SO2 concentrations, within a few kilometers of its path. This underscores the importance of granular modeling for port areas.
  • AIS Data as a Game Changer: The successful integration of AIS data for creating high-resolution, single-vessel emission inventories is a critical advancement, enabling more accurate and dynamic pollution simulations.
  • CALPUFF Adaptability: Demonstrates the flexibility of CALPUFF in being adapted from stationary to mobile sources by discretizing trajectories, offering a practical approach for simulating ship emissions.
  • Meteorological Influence: Reinforces the critical role of accurate meteorological fields (wind speed and direction) in determining pollutant dispersion patterns and the timing of peak concentrations. Deviations in simulated wind speed directly impact the predicted timing of observed peaks.

Key Results

  • The CALPUFF model, when driven by WRF/CALMET meteorological data and AIS-based emission inventories, showed good agreement with on-site SO2 measurements.
  • The ground concentration contribution of SO2 from departing vessels ranged from 10 to 102 µg/m³.
  • The affected area by these emissions was approximately 4–26 km².
  • High SO2 concentration zones were primarily located within 1–2 km of the ship's track.
  • Simulated peak SO2 concentrations for Vessel B were 47.2 µg/m³, compared to a monitored peak of 38.7 µg/m³, with a simulated peak occurring 5 minutes earlier than the observed peak. This timing difference was attributed to wind speed discrepancies.
  • The study found that emissions from vessels contribute considerably to SO2 pollution around harbor areas, with ocean-going container vessels having an important short-term adverse impact on air quality near the port during departure.

Limitations and Future Work

  • Meteorological Model Errors: Inconsistencies in peak timing were primarily attributed to deviations between simulated and actual wind speed and direction. The authors suggest developing more accurate meteorological models, potentially using Computational Fluid Dynamics (CFD), to improve results.
  • Limited Scope of Emissions: The study focused on ocean-going container ships and did not consider emissions from other sources like tugboats, vehicles, or auxiliary port machinery, which could explain some observed small peaks.
  • Limited Receptor Network: The placement of shore-based instruments meant that only southerly and easterly winds could effectively disperse plumes to the monitoring site. The authors recommend adding more instruments in different directions to capture a wider range of scenarios.
  • Short-Term Observation: The experiment was conducted over a relatively short period. The authors suggest establishing long-term observations (e.g., quarterly, yearly) to study seasonal or interannual impacts and refine model results.
  • Complex Plume Dynamics: Further consideration is needed for factors like mechanical turbulence near the discharge port and the influence of funnel exit tilt on plume diffusion.

Future work could involve expanding the study to include other ship types and port activities, incorporating more sophisticated meteorological modeling, and conducting longer-term observational campaigns.

Summary updated

Abstract

To study the impact of vessel pollution on the atmospheric environment of the surrounding area, we present a numerical simulation method based on regional emissions inventories. The general spatial resolution is ≥1 km and the temporal resolution is ≥1 h; parameters which are suitable for the study of larger space-time scales. In this paper, the WRF/CALMET/CALPUFF model and Automatic Identification System (AIS) data are employed to develop a single-vessel atmospheric pollution diffusion model. The goal of this research uses existing meteorological models and diffusion models to provide a simulation technology method for studying the diffusion of SO<sub>2</sub> from a single ship. We take the outgoing phase of ocean-going container vessels in Yantian Port as an example. It can be used to set the position of sensitive receptors near the port area. Simulations are implemented with CALPUFF and the results are compared with data derived from on-site monitoring instrument. The CALPUFF modelling domain covers an area of 925 km<sup>2</sup> with a grid spacing of 500 m. The simulation results demonstrated agreement with the measured data. The ground concentration contribution value ranged from 10 to 10<sup>2</sup> μg/m<sup>3</sup>, while the affected area was about 4-26 km<sup>2</sup> and the high-value area of the ground concentration contribution was distributed within 1-2 km from the ship track. Emissions generated by the vessels represent a considerable contribution to SO<sub>2</sub> pollution around the harbor areas.
BibTeX
@article{bai2020single,
  author = {Shubin Bai and Yuanqiao Wen and Li He and Yiming Liu and Yan Zhang and Yu Qi and Weichun Ma},
  title = {{Single-Vessel Plume Dispersion Simulation: Method and a Case Study Using CALPUFF in the Yantian Port Area, Shenzhen (China)}},
  journal = {International Journal of Environmental Research and Public Health},
  year = {2020},
  doi = {10.3390/ijerph17217831},
}

Loading related content...

Figure