The effects of marine vessel fuel sulfur regulations on ambient PM2.5 at coastal and near coastal monitoring sites in the U.S.

Abstract

In August of 2012 the U.S. began implementing fuel sulfur limits on certain large commercial marine vessels within 200 nautical miles (nm) of its coasts as part of a North American Emissions Control Area (NA-ECA). The NA-ECA limited fuel sulfur use in these vessels to below 1% in 2012 and to below 0.1% starting in 2015. This work uses ambient PM2.5 monitoring data from the U.S. IMPROVE network and Positive Matrix Factorization (PMF) receptor modeling to assess the effectiveness of the NA-ECA at reducing ambient PM2.5 from high-sulfur residual fuel oil (RFO) use. RFO combustion emissions of PM2.5 are known to have a fairly unique vanadium (V) and nickel (Ni) trace metal signature. To determine if IMPROVE sites were affected by residual fuel oil combustion, V and Ni data from 65 IMPROVE sites in coastal States of the U.S. were analyzed from 2010 to 2011, the two years prior to NA-ECA implementation. 22 of these IMPROVE sites had a V and Ni correlation coefficient (r2) greater than 0.65 and were selected for further analysis by PMF. The slopes of the correlations between V and Ni at these 22 sites ranged from 2.2 to 4.1, consistent with reported V:Ni emission ratios from RFO combustion. Each of the 22 IMPROVE sites was modeled independently with PMF, using the available PM2.5 chemical speciation data from 2010 to 2015. PMF model solutions for the 22 sites contained from 5 to 9 factors, depending on the site. At every site a PMF factor was identified that was associated with RFO combustion, however, 9 sites had PMF factors where RFO combustion was mixed with other aerosol sources. For the remaining 13 sites, PM2.5 from RFO combustion was analyzed for three time periods; 2010–2011 representing the time period prior to the NA-ECA implementation (pre-NA-ECA), 2013–2014 representing the time period where fuel sulfur was limited to 1.0% (NA-ECA 1.0% S), and 2015 representing the time period where fuel sulfur was limited to 0.1% (NA-ECA 0.1% S). All 13 sites indicated statistically significant reductions in the contribution of RFO combustion to PM2.5 between the pre-NA-ECA period and the two periods of fuel sulfur control. The average decrease in annual average PM2.5 from RFO combustion from the pre-NA-ECA to NA-ECA 1% S period was 50.2% (range, 29.0%–65.4%) and from the pre-NA-ECA to NA-ECA 0.1% S period was 74.1% (range, 33.0%–90.4%).