Effects on the total annual emissions in the BaySpeciesDomestic vessels International Total Emissions Trang 4 Fig.. The variation in annual emissions for SO2 and PM species can be attri
Master Dissertation The Influence of Reduced Fuel Sulfur Content on Emissions from Maritime Operations in Vietnam: a case study of Ha Long Bay September 2023 The Graduate School Chung-Ang University Major in International Logistics Department of International Trade & Logistics VU THI DUNG Data collection and processing method summary The AIS information and IMO identification numbers were used to gather activity data for international vessels, sourced from Marinetraffic.com In contrast, data for domestic vessels relied on secondary data from previous research, segregating them into four groups: passenger, fishing, rubbish collection/transport (referred to as 'environmental vessels'), and cargo The first three groups consisted of vessels registered locally, while the domestic cargo group included vessels registered in other Vietnamese provinces Domestic vessels, typically small and lacking IMO identification numbers and AIS information, did not have available AIS data, such as average and maximum speed, navigational status, and position These collected data were utilized to calculate engine power in kilowatt-hours (kWh), derived from the product of three variables within the Equation (Eq.1): Ei,j,k = (LFj,k ∗ EPj ∗ Tj,k) ∗ EFi,j ∗ 10−6 (1) where, Ei,j,k: Emissions (tonnes) of pollutant species i; from engine j (ME or AE); at the corresponding mode of operation k (at sea mode or hoteling mode) of each ship on a trip; LFj,k: Average load factor of engine j, at operation mode k; EPj: Power of engine j (kW); Tj,k: Running time of engine j, at operation mode k (h); EFi,j: Emission factors of species i, for engine type j (g kWh−1); and 10−6: Factor for unit conversion (gram to tonne) Additionally, relevant secondary data for local vessels from governmental organizations in the province, such as the Department of Natural Resources and Environment, Fisheries Department, and Ha Long Bay Management Department of Quang Ninh province, were incorporated The engine working time, a crucial activity parameter for emissions calculations in Equation (1) (where Tj,k represents time spent on engine j at operation mode k), required the determination of working hours for both sailing and hoteling operation modes and for both Main Engine (ME) and Auxiliary Engine (AE) types These working hours were calculated based on the traveling distance and average vessel speed, with the average speed of each domestic vessel group (passenger, fishing, domestic cargo, and environmental vessels) obtained from GPS survey data For international vessels, the average speed data were sourced from online tracking data The time spent by a vessel on sailing and hoteling per trip was determined as below: Ts =Dr/V (2) Th = T − Ts (3) where: Ts: Time for sailing (h); Dr: Distance (km); V: Average vessel speed (km h−1); Th: Time for hoteling (h); and T: The total time for one trip (h) Annual emissions in 2015 vs 2018 are presented in Table Species NOx SO2 CO CO2 PM10 PM2.5 BC OC NMHC CH4 N2O Domestic vessels Passenger Fishing 2018 2018 2015 864 864 17.2 85.8 118 118 54,691 54,69 13.6 19.1 12.5 17.6 3.88 5.46 2.50 3.52 41.2 41.2 0.84 0.84 2.81 2.81 International 69 1.3 10 4252 1.1 1.01 0.31 0.20 3.39 0.07 0.22 201 59 5.7 8.7 363 1.3 1.20 0.37 0.24 2.90 0.06 0.19 Domestic cargo Environmental Total local 2018 2015 2018 2015 2018 2015 155 3.1 26 9754 155 15.3 26 9754 1.29 0.03 0.22 81 1.29 0.13 0.22 81 1089 22 155 68,778 2.4 2.25 0.70 0.45 8.20 0.17 0.50 3.4 3.15 0.98 0.63 8.20 0.17 0.50 0.02 0.02 0.01 0.004 0.07 0.001 0.004 0.03 0.03 0.01 0.01 0.07 0.001 0.004 17 16 4.9 3.2 53 1.1 3.5 1079 107 153 68,16 24 22 6.8 4.4 52 1.1 3.5 122 89 9.6 5905 11.4 10.5 3.26 2.07 4.16 0.08 0.30 Results and discussions a Effects of lowering sulfur content in fuel on shipping emissions i Effects on the total annual emissions in the Bay Total Emissions 2018 2015 1210 110 164 74,68 28.5 26.3 8.2 5.2 57 1.2 3.8 1200 196 163 74,070 35.3 32.5 10.1 6.5 57 1.2 3.8 Fig Annual emissions for the relevant species impacted by the alteration in sulfur content, comparing the years 2018 and 2015 The variation in annual emissions for SO2 and PM species can be attributed to changes in the sulfur (S) content of diesel oil (DO) used by domestic vessels This is because the number of domestic vessels between 2015 and 2018 saw only a minor increase, with an addition of fishing vessels and environmental vessels in 2018 The total estimated emissions for NOx, SO2, CO, CO2, PM10, PM2.5, BC, OC, NMHC, CH4, and N2O in 2018 from all shipping activities in Ha Long Bay were 1210, 110, 164, 74,683, 28.5, 26.3, 8.2, 5.2, 57.0, 1.2, and 3.8 tons per year, respectively The annual emissions in 2015 showed significantly higher amounts of SO2, PM10, PM2.5, BC, and OC due to the still high sulfur content in DO used by domestic vessels In 2018, with a sulfur content in DO reduced to 0.05%, the total emissions of SO2 decreased by approximately 44% to 110 tons per year from 196 tons per year in 2015 when the sulfur content was around 19% Similarly, the emissions of PM10 reduced from 35 to 29 tons per year, PM2.5 dropped from 33 to 26 tons per year, BC decreased from 10.1 to 8.2 tons per year, and OC reduced from 6.5 to 5.2 tons per year It is noteworthy that despite the sulfur content in DO decreasing by times for domestic vessels (from 0.25% to 0.05%), the total annual SO2 emissions in the Bay only decreased by around times This is because annual SO2 emissions were predominantly influenced by international vessels operating in the Bay, assumed to be constant in both years The assumptions regarding the consistent number of international vessels and their fuel sulfur content (hence similar emission factors) may introduce uncertainty, especially considering potential increases in international vessel arrivals as projected by the Ministry of Industry and Trade of Vietnam However, the changes in emissions from domestic vessels related to sulfur content, the main focus of this study, would remain unaffected by these assumptions Yau et al (2012) illustrated that adopting fuel-switching, specifically to reduce sulfur (S) content, is an effective measure for controlling sulfur oxide (SOx) emissions from ships Jiang et al (2014) also suggested various measures for reducing sulfur dioxide (SO2) emissions, such as installing seawater scrubbers and transitioning from heavy fuel oil to marine gas oil, resulting in potential reductions of 98% and 90% in SO2 emissions, respectively Ushakov et al (2013) emphasized the significance of lowering S content in marine fuels for particulate matter (PM) emissions, demonstrating that a reduction from 1% to 0.1% S content led to an 80% reduction in SOx and a 67% reduction in PM emissions Zetterdahl et al (2016) conducted measurements on a ship in the Baltic Sea, confirming an 80% reduction in SOx and a 67% reduction in PM emissions with decreased S content Our findings in the Bay align with these outcomes, underscoring the effectiveness of transitioning to lower S content in domestic vessels to control both SOx and PM emissions Furthermore, it is crucial to establish regulations regarding S content for international vessels operating in the Bay For instance, enforcing the use of cleaner fuels upon entering the Bay could contribute to reducing primary PM emissions and secondary PM pollution, given the lowered emissions of a significant SOx precursor (Seinfeld and Pandis, 2016) The transition to lower S content across all ships in the Bay promises multiple advantages for air quality, including health benefits stemming from reduced SOx emissions, coemitted PM species, and secondary sulfate particles It is recommended to assess the impact of reduced S content on ambient air quality, particularly SO2 concentrations in the Bay and coastal areas Although there is currently a lack of SO2 monitoring data for such an evaluation, employing air quality modeling tools with emissions data from this study is suggested for future investigations ii Impacts on Emission Distribution between Domestic and International Fleets Figure illustrates that, in the year 2018, emissions from domestic vessels ranged from 20% to 94% of the total emissions, depending on the pollutant type The utilization of high sulfur fuel resulted in international vessels predominantly contributing to SO2 emissions (80%), as well as substantial shares of PM10, PM2.5, BC, and OC emissions (40% each) The international vessels' contribution to emissions of other gaseous pollutants, unaffected by sulfur content in the fuel, remained relatively modest (6%-10%) Among passenger vessels, this group was accountable for 48% of PM10 and PM2.5, BC, and OC emissions, 16% of SO2 emissions, and the majority (>71%) of emissions from other gases The domestic cargo group contributed between 3% and 16% of the total emissions, surpassing the contribution of the fishing group, which ranged from 1% to 6% depending on the pollutant species Fig Emissions share by vessel group in Ha Long Bay, 2018 b Spatial distribution of emissions Spatial distribution maps illustrating emissions in the study area, generated using ArcMap (v.10.5) with a resolution of km × km grid, are depicted in Figure for PM2.5, NOx, and SO2 (panels a, b, and c, respectively) Corresponding maps for other species (PM10, BC, OC, CO, CO2, NMHC, CH4) demonstrating similar distribution patterns Notably, NOx, CO, and NMHC emissions predominantly originated from domestic passenger vessels along their six routes (see Figure S2a, Supplementary Material), resulting in concentrated emissions hotspots along these routes and near the overnight cruiser sleeping points (Ti Top, Sung Sot, Lom Bo, and Trong Cave) Elevated emissions were also observed along the four routes of fishing vessels (see Figure S2b, Supplementary Material) SO2 emissions were primarily attributed to international vessels (80%) and passenger vessels (16%), resulting in heightened emission intensities along their travel routes (see Figure S2e, Supplementary Material) The most significant SO2 emission intensity was observed along international vessel routes and their anchorage areas (Hon Gai and Hon Net), as well as in Ti Top and Sung Sot sleeping points of overnight cruisers A similar spatial distribution was noted for PM2.5 (and also for PM10, BC, and OC), with a slightly more prominent influence from passenger vessels (contributing 48% of total PM2.5 emissions) compared to international vessels (contributing 40% of total PM2.5 emissions, as shown in Figure 2) In general, high emissions intensities were evident along the route connecting Tuan Chau Marina tourist port to tourist attractions and the routes of international vessels The average emissions intensity in kg y−1 km−2 for 2018 was 16.9 for PM2.5, 798 for NOx, and 72.2 for SO2 Table presents the average emissions intensity from shipping activities in Ha Long Bay in 2018, with domestic vessels contributing 10.1 for PM2.5, 720 for NOx, and 15.1 for SO2, reflecting their respective shares in total annual emissions in the Bay The spatial distribution of shipping emissions provides valuable input data for future dispersion modeling studies, enabling the assessment of current and future emissions associated with control strategies in the Bay The emissions intensities obtained in this study for 2018, presented in Table S6 (Supplementary Material), are compared with shipping emission intensity results from different regions, including Asia (Tianjin port, Yangtze River Delta, and Hong Kong port), the ASEAN region, and the top 10 largest ports in the world (World Shipping Council, 2019) The emissions intensity in the Bay for CO2, NOx, SO2, PM10, and PM2.5 in 2018 were 52.2, 0.8, 0.07, 0.02, and 0.02 t y−1 km−2, respectively (refer to Table S6, Supplementary Material) Notably, the emissions intensity of the three large ports in Asia (Tianjin port, Shanghai port in Yangtze River Delta, and Hong Kong port) was significantly higher than the results for Ha Long Bay, attributed to more intensive shipping activities in these ports The lower sulfur content of distillate oil used for domestic vessels in Ha Long Bay also contributed to the lower ship emissions intensity in 2018 Comparisons with emissions intensities from the Yangtze River Delta port cluster, Hong Kong, and the ASEAN region underscore the importance of accounting for domestic vessels in emission inventory studies within local domains The emissions intensity for the international vessels in the ASEAN region, reported by the Norwegian Maritime Authority (2018), was approximately 10 times lower than the results for the Bay, indicating the substantial contribution of domestic vessels, which was not considered in the ASEAN study This highlights the significance of including domestic vessels in emission inventory assessments for accurate local domain results Fig Spatial distribution of PM2.5 , NOx & SO2 emissions in 2018 (kg km−2 ), grid size is km × km (circles/dots are the port locations) In general, satellite observations would be valuable for analyzing shipping emission inventory (EI) results on a large scale, encompassing regional or global coverage (Johansson et al., 2017) However, for the Bay area, characterized by its relatively small size and surrounded by several significant inland emissions source regions with substantially larger emissions intensity from industry, traffic, and residential combustion, the utilization of satellite products for a comprehensive emissions analysis was hindered Notably, the Dinh Vu port, the largest port in Northern Vietnam, situated approximately 30 km southwest of the Bay, contributes intensive international shipping activities, dominating emissions in this coastal zone Given this limitation, the consideration of prevalent wind fields (directions and speeds) governing the study area is crucial for tracking emissions plumes While satellite observations may face challenges in this specific context, future studies could explore the use of inverse modeling tools that leverage monitoring concentrations and meteorological data to calculate emissions intensities, providing a potential avenue for verification (Brasseur and Jacob, 2017) c Monthly Variation of Emissions The variation in emissions from passenger vessels in Ha Long Bay for 2018 is depicted in Figure S5 (Supplementary Material) Two distinct emission peaks were observed during the summer (June) and winter (November and December) holidays, closely mirroring the activity patterns of passenger vessels The highest number of domestic tourists typically occurs in the summer (May to August), aligning with the summer holiday period in Vietnam Conversely, winter (November to February of the following year) is a busy season in the Bay, driven by an influx of international tourists In fact, the number of international visitors to Vietnam in December 2018 was approximately 1.2 times higher than in June (VNAT, 2018, 2019) The lowest emissions occurred in February and September, corresponding to decreased domestic tourist activity during the Lunar New Year celebration (February) and the end of the summer vacation (September) As mentioned earlier, the monthly distribution in this study was primarily influenced by variations in the activities of passenger vessels, which were major contributors to emissions of most pollutants in the Bay (refer to Figure 2) Monthly trip data for passenger vessels in 2018, obtained from the Maritime Administration of Quang Ninh, indicated significant fluctuations in activities While this study provides insights into temporal variations, further surveys are warranted to capture the nuanced temporal dynamics of shipping activities and associated emissions for all vessel groups in the Bay d Limitations of This Study and Uncertainty in Emissions Results Despite significant efforts invested in gathering local data to develop a bottom-up emission inventory (EI) database, several challenges introduce uncertainties into the emission results The uncertainties in this study arise from both activity data (including the number of vessels, installed engine power, and working hours) and the emission factors (EFs) employed for emissions calculation Primarily, uncertainty in activity data was encountered due to the absence of online tracking information for domestic vessels through the Automatic Identification System (AIS) Estimating the travel route length had inherent uncertainty, particularly for fishing vessels, which lack fixed operation routes The route length for small and medium-sized fishing vessels was roughly estimated based on the distance to the farthest points of their operation areas, determined by the engine size range Moreover, detailed information on Auxiliary Engine (AE) power was unavailable for all domestic vessel groups, necessitating the calculation of installed AE engine power from Main Engine (ME) power using AE/ME ratios For international vessels, the AE/ME ratio was derived from literature sources Additionally, the assumption of a constant load factor for the engine throughout the entire trip, despite potential regular changes due to weather conditions, introduces uncertainty Advanced EI methods, such as employing STEAM (Johansson et al., 2017), could be explored in future studies to mitigate uncertainty, but this approach requires AIS data not currently available for domestic vessels in the Bay Maneuvering emissions were disregarded in this study under the assumption that they occur over a short period, with a relative contribution deemed insignificantly lower than other operational modes This assumption is particularly applicable to domestic vessels, most of which lack AE and generally operate at low speeds, resulting in negligible speed changes upon entering the Bay While maneuvering mode has a lower EF for NOx, emissions of incomplete combustion products like PM and HC are higher compared to the at-sea mode The exclusion of emissions from this mode may not significantly impact the accuracy of emission results for Ha Long Bay Studies in different locations, such as Izmir Bay, Turkey, and China, reported contributions of 4%–8% and less than 1%, respectively, from maneuvering mode to total air pollutant emissions from shipping activities The EFs for domestic vessels were calculated based on specific fuel oil consumption and load factor data is secondary data, using average values for estimation In reality, these parameters can vary significantly based on operating conditions, engine age, and ambient conditions, contributing to emissions variation Relying on average values introduces uncertainty into emission estimates, highlighting the importance of continuous on-board emissions monitoring for future studies to address this source of uncertainty The data of domestic vessels, utilizing the quota sampling method, resulted in a relatively small sample size due to challenges in securing respondents from vessel owners Future studies should employ statistical probability sampling designs, such as stratified sampling, to mitigate bias Additionally, the wide age span and diverse maintenance and operation conditions of domestic vessels operating in the Bay suggest the need for longer-term surveys to produce a larger dataset addressing these factors adequately For international vessels, the study employed a sulfur (S) content of 2.7% in residual oil (RO) for ME, which represents the upper range reported globally (0.1%–2.7%) However, it remains below the global limit of S content, up to 3.5% To improve EI results, more precise information on the S content in the fuel used by international vessels in the Bay is required The data for international vessels was for a short period, and results were extrapolated for annual data, introducing uncertainty While monthly marine transport data for the province where the Bay is located demonstrate stability, AIS data for the entire years of 2015 and 2018 would enhance the accuracy of emission estimates It is noteworthy that adherence to local regulations on fuel S content by international vessels entering the Bay could significantly reduce shipping emissions in the domain Given the mentioned uncertainties, this study provides central values for emission estimates Future studies should aim to minimize uncertainties and provide a range of estimates (low, best, and high) for each species Conclusions The results from the emission inventory (EI) for the years 2018 and 2015 clearly highlight the positive impact of improving fuel quality by reducing sulfur (S) content on mitigating shipping emissions in Ha Long Bay In 2018, the annual emissions from all shipping activities in the Bay, including NOx, SO2, CO, CO2, PM10, PM2.5, BC, OC, NMHC, CH4, and N2O, amounted to 1210, 110, 164, 74,683, 28.5, 26.3, 8.2, 5.2, 57, 1.2, and 3.8 t y−1, respectively The decreased S content in diesel oil (DO) used by domestic vessels led to significant reductions in 2018 compared to 2015, notably a 44% reduction in SO2 and 19% reductions in particulate species (PM10, PM2.5, BC, and OC) Despite these improvements, domestic vessels still accounted for the highest contributions (60%–94%) for all other gaseous species (excluding SO2) emitted from shipping activities in the Bay in 2018 The international fleet, with higher S content in fuels, remained a dominant contributor, particularly for SO2 (45%) and particulate species (32%) in 2015, and even more prominently in 2018 (80% for SO2 and 40% for particulate species), emphasizing the need for stringent regulations on fuel S content for international vessels in the region to further reduce emissions Within the domestic fleets, the passenger group emerged as the primary emitter in 2018, contributing 71%–73% of gas species emissions (except for 16% for SO2) and 48% of particulate species emissions The domestic cargo group was the second-largest contributor, sharing 9%–16% of total emissions for every species except SO2 (3%) Contributions from other domestic groups were comparatively small, with fishing vessels (1%–6%) and environmental vessels (