Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

Fan Zhou; Yunli Fan

doi:10.1155/2021/6679682

Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

Zhou, Fan;Fan, Yunli 2021-02-09 00:00:00 Hindawi Advances in Meteorology Volume 2021, Article ID 6679682, 10 pages https://doi.org/10.1155/2021/6679682 Research Article Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model 1,2 1,2 Fan Zhou and Yunli Fan College of Information Engineering, Shanghai Maritime University, Shanghai, China Shanghai Engineering Research Center of Ship Exhaust Intelligent Monitoring, Shanghai, China Correspondence should be addressed to Fan Zhou; fanzhou_cv@163.com Received 13 November 2020; Revised 9 January 2021; Accepted 28 January 2021; Published 9 February 2021 Academic Editor: Antonio Donateo Copyright © 2021 Fan Zhou and Yunli Fan. -is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Since January 1, 2018, ships berthed at all ports of the three designated emission control areas (ECAs) in China are required to use fuel with sulfur content not exceeding 0.5% (m/m), excluding one hour postarrival and one hour predeparture. To understand changes in SO due to this policy, two observation stations were established on Waigaoqiao Dock in the Yangtze River estuary. -ree data types were collected from March 2018 to May 2018, namely, wind speed and direction, SO concentration, and ships’ arrival and departure times. -e statistics indicate that the wind direction changed little during the observation period and SO concentration was below 5 µg/m 77.47% of the time. Meanwhile, ships’ arrival and departure at the dock had a distinct inﬂuence on overall SO distribution, including occurrence of concentrations≥5 µg/m . -e three types of data were divided into six groups and a diﬀerence-in-diﬀerence model was used for analysis. -e result shows that SO concentration increases due to the use of high-sulfur fuel and is especially signiﬁcant when the wind is southwesterly. Furthermore, there was a positive correlation between increases in SO concentrations over 5 µg/m and the number of ships arriving or departing from the port. -is study reports the positive impact of fuel switching on air quality and can be used to evaluate adherence to the ECA policy. also included the establishment of emission control areas 1. Introduction (ECAs), where the FSC should not exceed 0.1% (m/m) as of With the rapid development of the shipping industry [1], air 2015 [9]. In addition, some countries and regions have set pollution caused by ship emissions has gained increasing their own ECAs. For instance, the Atmospheric Pollution attention in recent years [2–4] as it is a major contributor to Prevention and Control Law of the People’s Republic of both local and global air pollution [5, 6]. To limit ship China was promulgated in 2015 [10]. In the subsequent emissions to the environment and to reduce the risks to implementation plan, three domestic ECAs were set up by human health, the International Maritime Organization the Chinese government, namely, the Yangtze River Delta, (IMO) adopted MARPOL Annex VI in 1997 to prevent air the Pearl River Delta, and Bohai Rim (also referred to as the pollution caused by shipping emissions. -e main pollutants Beijing-Tianjin-Hebei Region) [11]. Under these regula- in ship exhaust are sulfur dioxide, nitrogen oxide, and tions, FSC cannot exceed 0.5% (m/m) during berthing particulate matter, with SO being one of the most prom- within ECAs, excluding one hour following the ships’ arrival inent pollutants. SO comes from the oxidation of sulfur in at the dock and one hour before its departure, January 1, fuel oil during combustion. Hence, high average sulfur in 2018, onwards. fuel oil leads to high SO emissions [7]. -e regulation of Yangtze River is the world’s third-largest river and IMO included a global cap of fuel sulfur content (FSC) such China’s most important inland waterway. Yangtze River that it should not exceed 3.5% (m/m) (as of 2012) and should Delta is within the ECAs. Located at the mouth of the be reduced to 0.5% (m/m) by 2020 [8]. -e IMO regulation Yangtze River, Shanghai is one of the most prosperous cities 2 Advances in Meteorology worldwide. At the end of 2017, Shanghai had a permanent urban area of Saint Petersburg. -e emission factors show resident population of approximately 24 million people [12]. compliance with the 1% FSC ECAs limit for 90% of the ships Container traﬃc through Shanghai increased each year from in 2011 and 97% in 2012. Murena et al. [26] assessed the 2010 to 2018 to become the busiest port in the world. As impact of cruise ship emissions in the port of Naples on air such, it is important to understand the impact ship pollution quality using a bottom-up procedure. -ey concluded that has on air quality in this area [13–15]. -ere are currently cruise ships’ contribution seems limited but nonnegligible. two major ports in Shanghai: Yangshan and Waigaoqiao. -ese studies show that land-based monitoring can monitor -e Waigaoqiao Port is only 20 km from the city center, and ship emissions accurately. Meanwhile, various studies have air pollution caused by ship emissions in this area directly been conducted to clarify the environmental impact of aﬀects the urban air environment and the health of China’s ECAs policy. -e research of Zhang et al. [27] in- Shanghai’s residents. A previous study indicated that 70% of dicated that the SO concentration in Shanghai decreased by the global emissions from ships are produced within 400 km at least 0.229 µg/m daily on average due to the imple- of coastlines and that these emissions can cause severe mentation of the ECA policy. Wan et al. [28] compared and environmental and health problems within these regions analyzed the changes of air quality in port cities within and [16, 17]. Filonchyk and Peterson (2020) evaluated air quality outside the ECAs during the implementation of the policy. during the COVID-19 lockdown of Shanghai and found that -e results showed that the ECA policy has some time lag, daily concentrations of PM2.5, PM10, SO , NO , and CO possibly because enforcement has gradually become more 2 2 during the lockdown period were reduced by 9%, 77%, stringent. Zhang et al. [29] conducted a measurement 31.3%, 60.4%, and 3%, respectively, compared to the same campaign (SEISO-Bohai) from December 28, 2016, to period in 2019 [18]. -erefore, the mitigation of ship exhaust January 15, 2017, at Jingtang Harbor, an area within China’s emissions and the establishment of sustainable port eco- ECAs. -e results from this study indicated a positive impact systems have become urgent tasks that require complex and from fuel switching on the air quality in the study region. comprehensive systematic approaches [19]. -us, as one of From these studies, it can be seen that the monitoring of ship the busiest ports worldwide, it is important to understand emissions in ECAs can help evaluate the implementation and control the pollution from ships in Waigaoqiao Port. eﬀect of the policy. Previous studies have indicated that ship emissions At present, the requirement of China’s ECAs is mainly observed at the monitoring stations in ECAs have decreased aimed at the limitation of sulfur content in ship fuel. Ship signiﬁcantly since the aforementioned restrictions were emissions of sulfur oxides (mainly SO ) come from the implemented [20–24]. Accordingly, this study monitored sulfur content of fuel. -erefore, by detecting the change of the SO concentrations at the dock within the China’s ECAs SO content in the dock in combination with the berthing 2 2 from March 2018 to May 2018. Ship owners are likely to use information of the ship, the inﬂuence of the restriction cheaper heavy fuel during arrival or departure at the dock to policy on the air quality in the port area can be judged. As an reduce fuel costs while staying within the bounds of the alternative to the observation position of the above research, policy. When a ship is berthed, light fuel is likely used in- the dock area is not only the heaviest traﬃc area but also the stead. -erefore, in this study, changes in SO concentra- main area from which the maritime department supervises tions were used to understand the impact of the and enforces laws. An understanding of ship emissions and Atmospheric Pollution Prevention and Control Law on air atmospheric processing in dock areas is required in order to quality. develop eﬀective regulations to better manage the envi- Regarding methods for monitoring ship emissions, land- ronmental impacts of shipping. -erefore, SO monitoring based measurements provide continuous observations and equipment was installed on the bridge cranes of Waigaoqiao have been widely used. Kattner et al. [21] reported large Dock. We were allowed to observe the changes in sulfur reductions of SO in ship plumes from September 2014 to dioxide in the intensive shipping area from March 2018 to January 2015 near the mouth of Hamburg Harbor on the May 2018. Ships’ arrival and departure times from the port River Elbe. -eir results show that the vast majority (95.4%) and the wind speed and direction were also collected for of all the ships complied with the regulation of 0.1% FSC. comparative analysis. -e main purpose of this research was Yang et al. [22] measured SO continuously from the Penlee to clarify the impact of ECA policy on air quality in port Point Atmospheric Observatory near Plymouth, United areas. However, as the port area is a relatively complex Kingdom, between May 2014 and November 2015; this environment, it is diﬃcult to accurately assess the impact of coastal site is exposed to marine air across a wide wind the policy based on existing observational data. In policy sector. -eir observations suggest a 3-fold reduction in ship evaluation research, the diﬀerence-in-diﬀerence model emitted SO from 2014 to 2015. Alfoldy et al. [24] measured (DID) is an eﬀective tool to extrapolate the causal eﬀects of the chemical composition of the plumes of seagoing ships policy [27, 28]. -e primary application of the DID model is during a two-week-long measurement campaign in the port to estimate the diﬀerence in the mean outcomes of the of Rotterdam, Hoek van Holland, Netherlands, in September treated and control units posttreatment and to isolate the 2009. -e average SO emission factor was found to be outcome diﬀerence that existed before the treatment [30]. approximately half of what is allowed in ECAs, and Meanwhile, ships’ arrival and departure times and wind exceedances of this limit were rarely registered. Beecken speed and direction are the two main factors aﬀecting SO et al. [25] carried out measurements from coastal sites near concentration which were also monitored. -e monitoring the island of Kronstadt and along the Neva River in the dataset is suitable for the comparison experiment using the Advances in Meteorology 3 DID model. -erefore, we used the DID model to analyze proportional to the concentration of the gas. -e SO the collected data. -is can be employed to identify the electrochemical sensor was eﬃcient in terms of its low power impact of reducing SO emissions in policy-aﬀected regions. consumption, small size, and lightweight and being ex- tremely precise. Additionally, these sensors were capable of measuring SO concentrations at a low ppb range [32], with 2. Methods a resolution level of 1 ppb, an accuracy of 20 ppb, and a measuring range of 5 to 10,000 ppb. Overall, this kind in- 2.1. Measurement Site. As shown in Figures 1(a) and 1(b), strumentation is convenient and lightweight, which is ap- the measurement station is located on the south bank of the propriate for installation on a bridge crane. Yangtze River estuary in the Waigaoqiao Port area, north of Shanghai. In December 2018, the Waigaoqiao Port handled 1,747,000 teU, 14.987,000 tons of cargo, and 3,553 berthing 2.3. Data Treatment. -ree types of data at the monitoring vessels. In previous studies [21, 25, 29], the monitoring dock were collected from March 2018 to May 2018: wind station has been located on the shore of the channel. Under speed and direction, the concentration of SO , and ships’ an appropriate wind speed and direction, the monitoring 2 arrival at and departure from the dock. SO concentration stations used in previous studies can measure the emissions was measured once per minute from the two monitoring of a single ship that sails through the port. However, in this points. -e data from March and April are continuous data, study, the monitoring equipment was located on the sus- with observation times of 743 hours and 720 hours, re- pension bridge on the dock, as shown in Figure 1(c). We spectively. Due to the receiver server failure, there is a certain deployed two sets of equipment on two diﬀerent working degree of missing data in May. Data were gathered from 14 : bridge cranes approximately 1 km apart. During the task of 00 on May 1 to 14 : 00 on May 4 and from 0 : 00 on May 10 to loading/unloading containers, the bridge crane moved three 24 : 00 on May 21. -e total observation time for May was dimensionally within a small area, thus the position of the 354 hours. monitoring equipment varied (30–40 m in height) though -e three sets of data were normalized using each hour’s this did not aﬀect the outcome. -e bridge crane is mostly data in order to facilitate a comprehensive analysis. Using surrounded by container ships berthing. -e research of this methodology, the 2 minutes of average wind speed and Chen et al. [31] indicated that the main contributors to ship direction data that were collected and the mean values of SO emissions are containers, followed by ﬁshing ships, oil 2 concentrations observed at the two monitoring stations were tankers, and bulk carriers. -erefore, the equipment on the each considered as one-hour observations. -e ships’ arrival bridge crane can clearly detect changes in SO from its and departure times at the dock were recorded within each location. -e complexity of the monitoring points within the hour as one-hour observations. -en, SO concentration surrounding environment is shown in Figures 1(a) and 1(b). 2 data were organized into six groups based on the month and Sources of airborne SO are the ships that are in berth as well the monitoring points, as follows: Monitor-1 in March, as ships that are sailing far from the dock; at the same time, Monitor-2 in March, Monitor-1 in April, Monitor-2 in container trucks on the docks are also a source of SO . In April, Monitor-1 in May, and Monitor-2 in May. -ese data addition, it can be seen from Figure 1(b) that the sur- were then analyzed to identify the SO emission charac- roundings of monitoring points 1 and 2 diﬀer, and there are 2 teristics and their relationship with wind speed, wind di- more ship berths to the west of monitoring point 2. Higher rection, and ships’ arrival and departure. concentrations of SO may be detected when there are Several aspects inﬂuenced the results of the analysis. As westerly winds. -erefore, in this study, we analyzed the can be seen from Figure 1, in addition to nearby berthing correlation between the factors (wind speed and direction ships, the monitoring point also was able to collect data for and ships’ arrival and departure at the dock) and the ob- ships sailing in the distance and container trucks passing served results (SO concentration) rather than the emissions under the bridge crane. -ere are also many chemical plants from individual ships. within 10 km of the dock; the SO emitted from these sources will have impacted the measurements, which can 2.2. Instrumentation. In this research, SO was measured increase uncertainty. Preliminary studies showed that 1–19% of the sulfur in the fuel is emitted in other forms, with instruments from the ETL3000 series, the sensors (Unitec SENS-IT) of which meet the European Standard possibly SO or SO [33–36]. Hence, the assumption that all 3 4 sulfur is emitted as SO yields an underestimation of the true 2008/50/EC. As shown in Figure 1(c), the monitoring equipment employed in this study had a measured weight of sulfur content in the fuel. In addition, the research of Yang et al. [22] shows that the residence time of SO in the marine 12 kg, a solar power supply system, and the data commu- atmosphere is approximately 0.5 d, with dry deposition nication mode was 4G cellular broadband. It included a data acquisition and processing unit, modular gas sensor, solar explaining about a quarter of the total SO sink. -erefore, the SO measured by the sensor may be either from the ship power supply system, and wireless data transmission module and a software system that was directly installed outdoors that is emitting it or from ship emissions within half a day. Unfortunately, there are no quantitative data for the without any cooling device or air-processing device. -e SO sensor used the electrochemical method, where the elec- abovementioned uncertainty. Although the background value can approximate from the observed value when there trochemical sensor determined the concentration of the gas through a redox reaction and produced an electrical signal is no ship, the endogenous problems arise when reverse 4 Advances in Meteorology (a) (b) (c) Figure 1: (a) Location of the measurement station. (b) Satellite view of the monitoring environment, with the location of the two survey stations shown in red (map source: lbs.amap.com). (c) Equipment installed at a suspension bridge on the dock. causality occurs between the variables or some variables are these data were analyzed in relation to the equipment in- missing. However, by using the DID model, the observation stallation location and surrounding environment (Figure 1), data are robust enough that they can provide useful infer- it could be concluded that the results of this experiment were ences. It can be deﬁned as follows: less aﬀected by the sailing ships and that the main source of SO is the discharge of berthing ships and the pollution 1,2 2 (1) Y � α + α du + a dt + a du ∗ dt + a Z + ε , it 0 1 2 3 4 it it sources on the shore. -e arrival and departure of ships at the dock are where subscripts i and t represent the i-th monitoring point considered to be a major factor aﬀecting air quality. and t-th month, respectively, and superscripts 1 and 2 -erefore, we counted the distribution of SO during these represent monitoring points 1 and 2, respectively. Z rep- two periods, respectively. -e results displayed in Figure 4 resents a series of control variables, ε is a random term (or show that SO concentration in the six observation groups is disturbance), and the interpreted variable is the monitored higher during the ships’ arrival and departure times than at SO value. times when there was no movement of ships in or out of the According to the ECA policy, ships are allowed to use dock. -erefore, all six sets of observations reﬂect the high-sulfur fuel within one hour of their arrival or departure. positive impact of fuel switching on air quality. Moreover, To analyze the inﬂuence of this factor on air quality, we used berths around Monitor-1 are only for ocean-going vessels, dt as the control parameter. dt � 1 means that a ship arrived while berths around Monitor-2 are for river vessels and or departed the dock during this hour; otherwise, dt � 0. ocean-going vessels. To a certain extent, this may have led to Wind direction is also a major inﬂuencing factor. du was a diﬀerence in the concentration of SO observed between used as the control parameter to classify the monitoring data the monitors. At Monitor-2, the concentrations of SO of downwind (wind direction: 180–270 degrees) as the measured were higher, but the diﬀerences between periods treatment group and the nondownwind data as the control of ship movement and periods of no ship movement were group. Because the treatment group and the control group smaller compared to that of Monitor-1. are measured by the same equipment at the same location, In Figure 5, we can see the diﬀerence in the overall they meet the parallel trend required by the DID model. distribution of the average SO concentrations per hour in When du � 1, the monitoring equipment is leeward of the two diﬀerent contexts. Within one hour of a ship’s arrival at berthing ships; that is, the discharge of the berthing ships is or departure from the dock, the time quantum for SO mainly measured by the monitoring equipment. Otherwise, concentrations exceeded 5 µg/m , which was higher than du � 0. -e DID model parameters are explained according that during no ship movement period. Meanwhile, the to the above deﬁnition as listed in Table 1. overall distribution of SO was also slightly larger when a ship arrived at or departed from the dock. In addition, there 3. Results and Discussion are more ship berths to the west of Monitor-2; therefore, results could be signiﬁcantly underestimated at this point, -e overall distribution of the average SO concentrations considering the prevalent wind direction. per hour observed at the two monitoring points is shown in From Figures 4 and 5, we can see that ships’ arrival and Figure 2. It can be seen that the concentration of SO is departure at the dock are a relatively obvious inﬂuence on air mainly distributed between 0 and 5 µg/m (77.47% within quality; however, there is no quantitative index. In addition, the observation time range). -erefore, 5 µg/m was used as wind speed and direction can also produce monitoring the standard for determining high or low SO concentration results which are diﬃcult to assess. -erefore, the DID in this study. As shown in Figure 3 (which includes data for model discussed in section 2.3 was used for analysis and the duration of the experiment), the wind mostly originated calculation, and the results are shown in Table 2. We from the southwest during the observation period. When Advances in Meteorology 5 Table 1: DID model parameters. No ship movement in or out of the dock Movement of ship(s) in or out of the dock Diﬀerence (dt � 0) (dt � 1) Downwind (treatment group, α + α α + α + α + α ΔY � α + α 0 1 0 1 2 3 2 3 du � 1) No downwind (control group, α α + α ΔY � α 0 0 2 2 du � 0) DID model estimation α α + α ΔY � α 1 1 3 3 Monitors 1 and 2, March to May 2018 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10 μg/m Sulfur dioxide concentration Monitor-1 in March Monitor-1 in May Monitor-1 in April Monitor-2 in March Monitor-2 in April Monitor-2 in May Figure 2: Overall distribution of the average SO concentrations per hour observed at the two monitoring points, where the interval is 1 µg/ m , and the number of observation hours in each concentration interval were counted. NNW NNE NW NE WNW ENE W E WSW ESE SW SE SSW SSE –1 Wind speed (ms ) 0-1 3-4 6-7 1-2 4-5 ≥7 2-3 5-6 Figure 3: Wind speed and direction per hour. -e numbers on the diagram are nonevenly spaced scale lines. Number of observation hours 6 Advances in Meteorology Monitor-2 in May Monitor-2 in April Monitor-2 in March Monitor-1 in May Monitor-1 in April Monitor-1 in March 0 1 23 4 56 Sulfur dioxide concentration μg/m No movement of ships in or out of the dock Movement of ship(s) in or out of the dock Figure 4: Average SO concentration within one hour of ships’ arrival or departure versus periods of no ship movement in or out of the dock. 0 3 µg/m 0-1 1-2 2-3 3-4 4-5 5+ Monitor-1 in March Monitor-1 in April Monitor-1 in May Monitor-2 in March Monitor-2 in April Monitor-2 in May (a) % 30 µg/m 0-1 1-2 2-3 3-4 4-5 5+ Monitor-1 in March Monitor-1 in April Monitor-1 in May Monitor-2 in March Monitor-2 in April Monitor-2 in May (b) Figure 5: Percentage distribution of average SO concentrations per hour for (a) periods of ship movement in or out the dock versus (b) periods of no ship movement in or out the dock. Advances in Meteorology 7 Table 2: DID model results. No ship movement in or out of the dock Movement of ship(s) in or out of the dock Diﬀerence (dt � 0) (dt � 1) Monitoring point 1, March 2018 Downwind (treatment group, 2.25 4.24 1.99 du � 1) No wind (control group, du � 0) 3.18 1.42 −1.76 DID model estimation −0.93 2.82 3.75 Monitoring point 2, March 2018 Downwind (treatment group, 5.06 5.35 0.29 du � 1) No wind (control group, du � 0) 4.02 5.22 1.2 DID model estimation 1.04 0.13 −0.91 Monitoring point 1, April 2018 Downwind (treatment group, 3.56 3.86 0.3 du � 1) No wind (control group, du � 0) 2.72 2.18 −0.54 DID model estimation 0.84 1.68 0.84 Monitoring point 2, April 2018 Downwind (treatment group, 4.33 4.61 0.28 du � 1) No wind (control group, du � 0) 4.11 3.95 −0.16 DID model estimation 0.22 0.66 0.44 Monitoring point 1, May 2018 Downwind (treatment group, 3.25 3.71 0.46 du � 1) No wind (control group, du � 0) 6.02 9.91 3.89 DID model estimation −2.77 −6.2 −3.43 Monitoring point 2, May 2018 Downwind (treatment group, 3.52 4.15 0.63 du � 1) No wind (control group, du � 0) 3.96 2.96 −1 DID model estimation −0.44 1.19 1.63 calculated the average SO concentration per period. All and its inﬂuence is especially signiﬁcant when the wind measurement data were divided into the treatment group direction is southwest. and control group, and the ships’ arrival and departure times In addition, we conducted statistics on the time period were taken as the control conditions. Based on the calcu- during which the sulfur dioxide concentration measured at lation results of the DID model, the inﬂuence of the two the two monitoring points exceeded 5 µg/m . As shown in factors (i.e., ships’ arrival or departure time and wind speed Figure 6, when the SO concentration exceeded 5 µg/m at and direction) on the concentration of SO was judged. only one monitoring point when measured at two points, In the treatment group, the eﬀect of arrival or departure both frequency ratios of the arrival and departure of ships time is obvious and signiﬁcant, whereas it is not obvious in were higher than that within the period in which the SO the control group. -is means that SO concentration rose concentration was less than 5 µg/m at both monitoring signiﬁcantly in the period of ship arrival or departure when points. In other words, there is a positive correlation be- the wind direction was southwest. In contrast, when the tween the high concentration of SO and the increased wind direction is not southwest, the variation trend of SO frequency of ships arriving and departing from the port area. concentration is not obvious in the period of ship arrival or -is result also reﬂects the positive impact of fuel switching. departure. Similarly, when there is no arrival and departure Several studies have monitored and assessed the impact of ships, the inﬂuence of wind direction on the changing of ECAs [21–24, 27–29]. However, the impact of individual trend of SO is not obvious. In contrast, when ships are ship activities on air quality is rarely monitored, making it moving in or out of the dock, the concentration of SO virtually impossible to assess the impact of ECA policies on increases signiﬁcantly when the wind is downwind. Finally, air quality in local areas such as wharves. -e activities of two sets of DID model results (ΔΔY) are negative, which can ships and personnel in dock areas tend to be relatively be attributed to the small number of data samples. In spite of frequent, and ship emissions directly aﬀect public health. this, four sets of DID model results (ΔΔY) are positive and -erefore, this study sought to establish monitoring data on present the impact of these two factors on SO concentra- ship emissions in one of Shanghai’s main ports by installing tions. On the whole, the calculation results of the DID model two monitoring stations at Waigaoqiao wharf to explore the mentioned above show that the SO concentration in the air relationships between SO concentrations, wind speed and 2 2 of the port area increases due to the use of high-sulfur fuel, wind direction, and ships’ arrival and departure times. 8 Advances in Meteorology 313 28 % 50 94 66 25 147 13 Both > 5 One > 5 Both < 5 Both > 5 One > 5 Both < 5 Both > 5 One > 5 Both < 5 March April May Movement of ship(s) in or out of the dock No movement of ships in or out of the dock Figure 6: SO concentrations during periods of ship movement in or out of the dock versus periods of no movement of ship(s) in or out of the dock expressed in percentage terms. For each month (March, April, or May), SO concentrations were divided into three categories: SO 2 2 3 3 concentrations at both monitoring points greater than 5 µg/m (both >5), SO concentrations greater than 5 µg/m at only one monitoring station (one >5), SO concentrations less than 5 µg/m at both stations (both <5). Considering the results presented in Figures 4–6, it can be concentration of SO was low (less than 5 µg/m ) 77.47% of concluded that ship emissions during the arrival or de- the time. -e distribution of wind speed and direction parture of ships signiﬁcantly aﬀect the airborne SO con- indicated that there was little change in the wind direction centration in the study area. -e results of the DID model during the study period. -e data observed at the moni- presented in Table 1 further indicate that SO concentration toring stations were mostly attributable to the pollution in the air increases signiﬁcantly due to the use of high-sulfur sources in the berth and on the shore. -e change in SO fuel when the wind direction is southwest. -is is signiﬁcant concentration during periods of ship movement in or out in the context of recently implemented policies which of the dock compared to when there was no ship activity at stipulate that the FSC cannot exceed 0.5% (m/m) during the dock is indicative of the contribution of a ship’s berthing within ECAs, excluding the ﬁrst hour postarrival emissions to the SO concentration in the air. -e result of and the last hour predeparture. -us, we were able to clarify the DID model shows that the concentration of SO in the the impact of ECA policy on air quality in port areas. We air of the port area increased due to the use of high-sulfur found that the recently implemented policies may need fuel and that its inﬂuence is especially signiﬁcant when the amendment in the future to consider our ﬁndings. wind direction is downwind. -is ﬁnding is relevant to the implementation of ECA policy, which requires that the FSC cannot exceed 0.5% (m/m) during berthing within the 4. Conclusions ECAs, except for the ﬁrst hour after arrival and the last -e implementation of the ECA policy can eﬀectively reduce hour before departure. -e results of this study can be used the harm caused by ship emissions. How to evaluate the as the basis for understanding ship emissions and atmo- impact of the implementation of this policy on the envi- spheric processes within dock areas under ECA policy. ronment is a research topic of interest. In this study, two SO Subsequently, more attention should be paid to the eﬀects monitoring stations were installed on the bridge crane in of sailing ship emissions. Waigaoqiao Dock, Shanghai, China, which allowed data to be obtained from March 2018 to May 2018. To analyze the Data Availability factors that caused a change in the SO concentration, we collected the data on the wind speed and direction as well as -e data used to support the ﬁndings in this study are the arrival and departure times of ships at the port in the available at https://data.mendeley.com/datasets/427wfrck7w/ same period and normalized them to one hour. -en, the draft?a�883bb7b4-b9f1-4549-b10c-9ad9939a3e61. DID model was used to evaluate the impact of policy implementation. Conflicts of Interest -e overall airborne distribution of SO at Waigaoqiao Dock was obtained. It was determined that the -e authors declare no conﬂicts of interest. Advances in Meteorology 9 Journal of Geovisualization and Spatial Analysis, vol. 2, no. 12, Acknowledgments pp. 1–18, 2018. [14] X. N Wang, Y. Shen, Y. F Lin et al., “Atmospheric pollution -e authors would like to thank Pudong Maritime Safety from ships and its impact on local air quality at a port site in Administration of the People’s Republic of China for their Shanghai,” Atmospheric Chemistry and Physics, vol. 19, no. 9, support in coordinating the ﬁeld measurements. -is re- pp. 6315–6330, 2019. search was supported by the National Natural Science [15] Y. Y Zhou, Y. Zhang, D. Ma et al., “Port-related emissions, Foundation of China (Grant nos. 41701523 and 61703271), environmental impacts and their implication on green traﬃc the National Key Research and Development Project of policy in Shanghai,” Sustainability, vol. 12, no. 10, p. 17, 2020. China (Grant no. 2020YFC1511901), and the Special De- [16] J. J. Corbett, P. S. Fischbeck, and S. N. Pandis, “Global ni- velopment Fund for China (Shanghai) Pilot Free-Trade trogen and sulfur inventories for oceangoing ships,” Journal of Zone. Geophysical Research: Atmospheres, vol. 104, no. D3, pp. 3457–3470, 1999. [17] Y. Gonzalez, ´ S. Rodr´ıguez, J. C. Trujillo, and R. Garc´ıa, References “Ultraﬁne particles pollution in urban coastal air due to ship [1] UNCTAD, “World seaborne trade by types of cargo and by emissions,” Atmospheric Environment, vol. 45, no. 28, group of economies, annual, United Nations Conference on pp. 4907–4914, 2011. Trade and Development,” 2017, https://unctadstat.unctad. [18] M. Filonchyk and M. Peterson, “Air quality changes in org/wds/TableViewer/tableView.aspx?ReportId=32363. Shanghai, China, and the surrounding urban agglomeration [2] V. Eyring, I. S. A. Isaksen, T. Berntsen et al., “Transport during the COVID-19 lockdown,” Journal of Geovisualization impacts on atmosphere and climate: shipping,” Atmospheric and Spatial Analysis, vol. 4, no. 22, pp. 1–7, 2020. Environment, vol. 44, no. 37, pp. 4735–4771, 2010. [19] X. Geng, Y. Wen, C. Zhou, and C. Xiao, “Establishment of the [3] A. Sorooshian and H. T. Duong, “Ocean emission eﬀects on sustainable ecosystem for the regional shipping industry aerosol-cloud interactions: insights from two case studies,” based on system dynamics,” Sustainability, vol. 9, no. 5, p. 18, Advances in Meteorology, vol. 2010, Article ID 301395, 9 pages, Article ID 742, 2017. [20] N. Molders, ¨ S. Gende, and M. Pirhalla, “Assessment of cruise- [4] C. Gencarelli, I. Hedgecock, F. Sprovieri, G. Schurmann, ¨ and ship activity inﬂuences on emissions, air quality, and visibility N. Pirrone, “Importance of ship emissions to local sum- in Glacier Bay National Park,” Atmospheric Pollution Re- mertime ozone production in the mediterranean marine search, vol. 4, no. 4, pp. 435–445, 2013. boundary layer: a modeling study,” Atmosphere, vol. 5, no. 4, [21] L. Kattner, B. Mathieu-Uﬃng, J. P. Burrows et al., “Moni- pp. 937–958, 2014. toring compliance with sulfur content regulations of shipping [5] H. Liu, M. Fu, X. X Jin et al., “Health and climate impacts of fuel by in situ measurements of ship emissions,” Atmospheric ocean-going vessels in East Asia,” Nature Climate Change, Chemistry and Physics, vol. 15, no. 17, pp. 10087–10092, 2015. vol. 6, no. 11, pp. 1037–1041, 2016. [22] M. Yang, T. G. Bell, F. E. Hopkins, and T. J. Smyth, “Attri- [6] G.-H. Yu, S. Park, S.-K. Shin, K.-H. Lee, and H.-G. Nam, bution of atmospheric sulfur dioxide over the English “Enhanced light absorption due to aerosol particles in ship Channel to dimethyl sulﬁde and changing ship emissions,” plumes observed at a seashore site,” Atmospheric Pollution Atmospheric Chemistry and Physics, vol. 16, no. 8, pp. 4771– Research, vol. 9, no. 6, pp. 1177–1183, 2018. 4783, 2016. [7] J. X. Zhou, S. Zhou, and Y. Q. Zhu, “Characterization of [23] L. Barregard, P. Molnar, J. E. Jonson et al., “Impact on particle and gaseous emissions from marine diesel engines population health of baltic shipping emissions,” International with diﬀerent fuels and impact of after-treatment technology,” Journal of Environmental Research and Public Health, vol. 16, Energies, vol. 10, no. 8, p. 14, Article ID 1110, 2017. no. 11, 11 pages, Article ID 1954, 2019. [8] IMO, “Sulphur oxides (SOx)–regulation 14,” 2017, http:// [24] B. Alfoldy, ¨ J. B. Lo¨ov, ¨ F. Lagler et al., “Measurements of air www.imo.org/en/OurWork/Environment/PollutionPrevention/ pollution emission factors for marine transportation in AirPollution/Pages/Sulphur-oxides-(SOx)-Regulation-14.aspx. SECA,” Atmospheric Measurement Techniques, vol. 6, no. 7, [9] IMO, “Emission control areas (ECAs) designated under MAR- pp. 1777–1791, 2013. POL Annex VI,” 2017, http://www.imo.org/en/OurWork/ [25] J. Beecken, J. Mellqvist, K. Salo et al., “Emission factors of Environment/PollutionPrevention/AirPollution/Pages/Emission- SO NO and particles from ships in Neva Bay from ground- 2 x Control-Areas-(ECAs)-designated-under-regulation-13-of-MAR based and helicopter-borne measurements and AIS-based POL-Annex-VI-(NOx-emission-control. modeling,” Atmospheric Measurement Techniques, vol. 15, [10] Standing Committee of the National People’s Congress, no. 9, pp. 5229–5241, 2015. “Atmospheric pollution prevention and control law of the [26] F. Murena, L. Mocerino, F. Quaranta et al., “Impact on air People’s Republic of China,” 2015, http://english.court.gov. quality of cruise ship emissions in Naples, Italy,” Atmospheric cn/2016-04/15/content_24565639.htm. Environment, vol. 187, pp. 70–83, 2018. [11] Ministry of Transport of the People’s Republic of China, [27] Q. Zhang, Z. Q Zheng, Z. Wan et al., “Does emission control “Marine emission control area plan for Pearl river delta, area policy reduce sulfur dioxides concentration in Shang- Yangtzy river delta, Bohai Rim area,” 2015, http://www.gov. hai?” Transportation Research Part D, vol. 81, Article ID cn/xinwen/2015-12/04/content_5019932.htm. 102289, 2020. [12] Shanghai Municipal Bureau of Statistics, “-e statistic com- [28] Z. Wan, X. J Zhou, Q. Zhang et al., “Do ship emission control munique of Shanghai on the 2017 national economy and areas in China reduce sulfur dioxide concentrations in local social development,” 2017, http://www.stats.gov.cn/english/ pressrelease/201802/t20180228_1585666.html. air? A study on causal eﬀect using the diﬀerence-in-diﬀerence model,” Marine Pollution Bulletin, vol. 149, Article ID 110506, [13] Y. Y. Huang, Q. W. Yan, and C. R. Zhang, “Spatial-temporal distribution characteristics of PM2.5 in China in 2016,” 2019. 10 Advances in Meteorology [29] Y. N Zhang, F. Y Deng, H. L Man et al., “Compliance and port air quality features with respect to ship fuel switching regu- lation: a ﬁeld observation campaign, SEISO-Bohai,” Atmo- spheric Measurement Techniques, vol. 19, no. 7, pp. 4899–4916, [30] M. S. Delgado and R. J. G. M. Florax, “Diﬀerence-in-diﬀer- ences techniques for spatial data: local autocorrelation and spatial interaction,” Economics Letters, vol. 137, pp. 123–126, [31] D. S. Chen, X. T. Wang, P. Nelson et al., “Ship emission inventory and its impact on the PM 2.5 air pollution in Qingdao Port, North China,” Atmospheric Environment, vol. 166, pp. 351–361, 2017. [32] A. W. E. Hodgson, P. Jacquinot, and P. C. Hauser, “Elec- trochemical sensor for the detection of SO in the low-ppb range,” Analytical Chemistry, vol. 71, no. 14, pp. 2831–2837, [33] H. Schlager, R. Baumann, M. Lichtenstern et al., “Aircraft- based trace gas measurements in a primary European ship corridor,” Proceedings of the TAC-Conference, pp. 83–88, [34] H. Agrawal, W. A. Welch, J. W. Miller et al., “Emission measurements from a crude oil tanker at sea,” Environmental Science and Technology, vol. 42, no. 19, pp. 7098–7103, 2008. [35] J. Moldanova, E. Fridell, O. Popovicheva et al., “Character- isation of particulate matter and gaseous emissions from a large ship diesel engine,” Atmospheric Environment, vol. 43, no. 16, pp. 2632–2641, 2009. [36] J. M. Balzani Loov, B. Alfoldy, L. F. L. Gast et al., “Field test of available methods to measure remotely SOx and NOx emissions from ships,” Atmospheric Measurement Techniques, vol. 7, no. 8, pp. 2597–2613, 2014. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Meteorology Hindawi Publishing Corporation http://www.deepdyve.com/lp/hindawi-publishing-corporation/determination-of-the-influence-of-fuel-switching-regulation-on-the-WJPNAbGbMk

Loading next page...

References (39)

N. Mölders, S. Gende, Michael Pirhalla (2013)
Assessment of cruise–ship activity influences on emissions, air quality, and visibility in Glacier Bay National Park
Atmospheric Pollution Research, 4
(2017)
e statistic communique of Shanghai on the 2017 national economy and social development
V. Eyring, I. Isaksen, T. Berntsen, W. Collins, J. Corbett, Ø. Endresen, R. Grainger, J. Moldanová, H. Schlager, D. Stevenson (2010)
Transport impacts on atmosphere and climate: Shipping
Atmospheric Environment, 44
Zheng Wan, Xujun Zhou, Qiang Zhang, Jihong Chen (2019)
Do ship emission control areas in China reduce sulfur dioxide concentrations in local air? A study on causal effect using the difference-in-difference model.
Marine pollution bulletin, 149
X. Geng, Y. Wen, Chunhui Zhou, Changshi Xiao (2017)
Establishment of the Sustainable Ecosystem for the Regional Shipping Industry Based on System Dynamics
Sustainability, 9
Jinxi Zhou, Song Zhou, Yuanqing Zhu (2017)
Characterization of Particle and Gaseous Emissions from Marine Diesel Engines with Different Fuels and Impact of After-Treatment Technology
Energies, 10
Yuyan Zhou, Yan Zhang, Dong Ma, Jun Lu, Wenbin Luo, Yu Fu, Shanshan Li, Junlan Feng, Cheng Huang, Wangqi Ge, Hong Zhu (2020)
Port-Related Emissions, Environmental Impacts and Their Implication on Green Traffic Policy in Shanghai
Sustainability
Yuanyuan Huang, Qingwu Yan, Chuanrong Zhang (2018)
Spatial–Temporal Distribution Characteristics of PM2.5 in China in 2016
Journal of Geovisualization and Spatial Analysis, 2
Yanni Zhang, Fanyuan Deng, Hanyang Man, Mingliang Fu, Zhaofeng Lv, Qian Xiao, Xinxin Jin, Shuai Liu, K. He, Huan Liu (2019)
Compliance and port air quality features with respect to ship fuel switching regulation: a field observation campaign, SEISO-Bohai
Atmospheric Chemistry and Physics
J. Beecken, J. Mellqvist, K. Salo, J. Ekholm, J. Jalkanen, L. Johansson, V. Litvinenko, K. Volodin, D. Frank-kamenetsky (2014)
Emission factors of SO 2 , NO x and particles from ships in Neva Bay from ground-based and helicopter-borne measurements and AIS-based modeling
Atmospheric Chemistry and Physics, 15
A. Hodgson, P. Jacquinot, P. Hauser (1999)
Electrochemical Sensor for the Detection of SO2 in the Low-ppb Range
Analytical Chemistry, 71
(2017)
World seaborne trade by types of cargo and by group of economies, annual, United Nations Conference on Trade and Development
J. Corbett, P. Fischbeck, S. Pandis (1999)
Global nitrogen and sulfur inventories for oceangoing ships
Journal of Geophysical Research, 104
B. Alföldy, J. Lööv, F. Lagler, J. Mellqvist, N. Berg, J. Beecken, H. Weststrate, J. Duyzer, Laszlo Bencs, B. Horemans, F. Cavalli, J. Putaud, G. Janssens‑Maenhout, A. Csordás, R. Grieken, A. Borowiak, J. Hjorth (2012)
Measurements of air pollution emission factors for marine transportation in SECA
Atmospheric Measurement Techniques, 6
H. Agrawal, William Welch, J. Miller, David Cockert (2008)
Emission measurements from a crude oil tanker at sea.
Environmental science & technology, 42 19
Xinning Wang, Yin Shen, Yanfen Lin, Jun Pan, Yan Zhang, P. Louie, Mei Li, Q. Fu (2019)
Atmospheric pollution from ships and its impact on local air quality at a port site in Shanghai
Atmospheric Chemistry and Physics
Mingxi Yang, T. Bell, Frances Hopkins, T. Smyth (2016)
Attribution of atmospheric sulfur dioxide over the English Channel to dimethyl sulfide and changing ship emissions
Atmospheric Chemistry and Physics, 16
Geun-Hye Yu, Seungshik Park, Sung-Kyun Shin, Kwonho Lee, Hyounggu Nam (2018)
Enhanced light absorption due to aerosol particles in ship plumes observed at a seashore site
Atmospheric Pollution Research
(2017)
World seaborne trade by types of cargo and by group of economies, annual
L. Barregard, Péter Molnár, J. Jonson, L. Stockfelt (2019)
Impact on Population Health of Baltic Shipping Emissions
International Journal of Environmental Research and Public Health, 16
Qiang Zhang, Zhong Zheng, Zheng Wan, Shiyuan Zheng (2020)
Does emission control area policy reduce sulfur dioxides concentration in Shanghai?
Transportation Research Part D-transport and Environment, 81
A. Sorooshian, H. Duong (2010)
Ocean Emission Effects on Aerosol-Cloud Interactions: Insights from Two Case Studies
Advances in Meteorology, 2010
Huan Liu, Mingliang Fu, Xinxin Jin, Yi Shang, D. Shindell, G. Faluvegi, Cary Shindell, K. He (2016)
Health and climate impacts of ocean-going vessels in East Asia
Nature Climate Change, 6
Michael Delgado, R. Florax (2015)
Difference-in-differences techniques for spatial data: Local autocorrelation and spatial interaction ☆
Economics Letters, 137
J. Moldanová, E. Fridell, O. Popovicheva, B. Demirdjian, V. Tishkova, A. Faccinetto, C. Focsa (2009)
Characterisation of particulate matter and gaseous emissions from a large ship diesel engine
Atmospheric Environment, 43
(2015)
Marine emission control area plan for Pearl river delta, Yangtzy river delta, Bohai Rim area
H. Schlager, R. Baumann, M. Lichtenstern, A. Petzold, F. Arnold, M. Speidel, C. Gurk, H. Fischer (2007)
Aircraft-based Trace Gas Measurements in a Primary European Ship Corridor
(2017)
Sulphur oxides (SOx)–regulation 14
(2017)
Emission control areas (ECAs) designated under MARPOL Annex VI
Y. González, S. Rodríguez, Juan García, J. Trujillo, R. García (2011)
Ultrafine particles pollution in urban coastal air due to ship emissions
Atmospheric Environment, 45
(2017)
The statistic communique of Shanghai on the 2017 national economy and social development
F. Murena, L. Mocerino, F. Quaranta, D. Toscano (2018)
Impact on air quality of cruise ship emissions in Naples, Italy
Atmospheric Environment
L. Kattner, Barbara Mathieu-Üffing, J. Burrows, A. Richter, S. Schmolke, A. Seyler, F. Wittrock (2015)
Monitoring compliance with sulfur content regulations of shipping fuel by in situ measurements of ship emissions
Atmospheric Chemistry and Physics, 15
J. Lööv, B. Alföldy, L. Gast, J. Hjorth, F. Lagler, J. Mellqvist, J. Beecken, N. Berg, J. Duyzer, H. Westrate, D. Swart, A. Berkhout, J. Jalkanen, A. Prata, G. Hoff, A. Borowiak (2013)
Field test of available methods to measure remotely SOx and NOx emissions from ships
Atmospheric Measurement Techniques, 7
C. Gencarelli, I. Hedgecock, F. Sprovieri, G. Schürmann, N. Pirrone (2014)
Importance of Ship Emissions to Local Summertime Ozone Production in the Mediterranean Marine Boundary Layer: A Modeling Study
Atmosphere, 5
(2015)
Atmospheric pollution prevention and control law of the People’s Republic of China
Mikalai Filonchyk, M. Peterson (2020)
Air Quality Changes in Shanghai, China, and the Surrounding Urban Agglomeration During the COVID-19 Lockdown
Journal of Geovisualization and Spatial Analysis, 4
Dongsheng Chen, Xiaotong Wang, P. Nelson, Yue Li, Na Zhao, Yuehua Zhao, Jianlei Lang, Ying Zhou, Xiurui Guo (2017)
Ship emission inventory and its impact on the PM2.5 air pollution in Qingdao Port, North China
Atmospheric Environment, 166
Feifan Ji, J. Xiang, Wuhua Li, Quanming Yue (2016)
A Feedback Passivation Design for DC Microgrid and Its DC/DC Converters
Energies, 10

Publisher: Hindawi Publishing Corporation
Copyright: Copyright © 2021 Fan Zhou and Yunli Fan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN: 1687-9309
eISSN: 1687-9317
DOI: 10.1155/2021/6679682
Publisher site: See Article on Publisher Site

Abstract

Hindawi Advances in Meteorology Volume 2021, Article ID 6679682, 10 pages https://doi.org/10.1155/2021/6679682 Research Article Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model 1,2 1,2 Fan Zhou and Yunli Fan College of Information Engineering, Shanghai Maritime University, Shanghai, China Shanghai Engineering Research Center of Ship Exhaust Intelligent Monitoring, Shanghai, China Correspondence should be addressed to Fan Zhou; fanzhou_cv@163.com Received 13 November 2020; Revised 9 January 2021; Accepted 28 January 2021; Published 9 February 2021 Academic Editor: Antonio Donateo Copyright © 2021 Fan Zhou and Yunli Fan. -is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Since January 1, 2018, ships berthed at all ports of the three designated emission control areas (ECAs) in China are required to use fuel with sulfur content not exceeding 0.5% (m/m), excluding one hour postarrival and one hour predeparture. To understand changes in SO due to this policy, two observation stations were established on Waigaoqiao Dock in the Yangtze River estuary. -ree data types were collected from March 2018 to May 2018, namely, wind speed and direction, SO concentration, and ships’ arrival and departure times. -e statistics indicate that the wind direction changed little during the observation period and SO concentration was below 5 µg/m 77.47% of the time. Meanwhile, ships’ arrival and departure at the dock had a distinct inﬂuence on overall SO distribution, including occurrence of concentrations≥5 µg/m . -e three types of data were divided into six groups and a diﬀerence-in-diﬀerence model was used for analysis. -e result shows that SO concentration increases due to the use of high-sulfur fuel and is especially signiﬁcant when the wind is southwesterly. Furthermore, there was a positive correlation between increases in SO concentrations over 5 µg/m and the number of ships arriving or departing from the port. -is study reports the positive impact of fuel switching on air quality and can be used to evaluate adherence to the ECA policy. also included the establishment of emission control areas 1. Introduction (ECAs), where the FSC should not exceed 0.1% (m/m) as of With the rapid development of the shipping industry [1], air 2015 [9]. In addition, some countries and regions have set pollution caused by ship emissions has gained increasing their own ECAs. For instance, the Atmospheric Pollution attention in recent years [2–4] as it is a major contributor to Prevention and Control Law of the People’s Republic of both local and global air pollution [5, 6]. To limit ship China was promulgated in 2015 [10]. In the subsequent emissions to the environment and to reduce the risks to implementation plan, three domestic ECAs were set up by human health, the International Maritime Organization the Chinese government, namely, the Yangtze River Delta, (IMO) adopted MARPOL Annex VI in 1997 to prevent air the Pearl River Delta, and Bohai Rim (also referred to as the pollution caused by shipping emissions. -e main pollutants Beijing-Tianjin-Hebei Region) [11]. Under these regula- in ship exhaust are sulfur dioxide, nitrogen oxide, and tions, FSC cannot exceed 0.5% (m/m) during berthing particulate matter, with SO being one of the most prom- within ECAs, excluding one hour following the ships’ arrival inent pollutants. SO comes from the oxidation of sulfur in at the dock and one hour before its departure, January 1, fuel oil during combustion. Hence, high average sulfur in 2018, onwards. fuel oil leads to high SO emissions [7]. -e regulation of Yangtze River is the world’s third-largest river and IMO included a global cap of fuel sulfur content (FSC) such China’s most important inland waterway. Yangtze River that it should not exceed 3.5% (m/m) (as of 2012) and should Delta is within the ECAs. Located at the mouth of the be reduced to 0.5% (m/m) by 2020 [8]. -e IMO regulation Yangtze River, Shanghai is one of the most prosperous cities 2 Advances in Meteorology worldwide. At the end of 2017, Shanghai had a permanent urban area of Saint Petersburg. -e emission factors show resident population of approximately 24 million people [12]. compliance with the 1% FSC ECAs limit for 90% of the ships Container traﬃc through Shanghai increased each year from in 2011 and 97% in 2012. Murena et al. [26] assessed the 2010 to 2018 to become the busiest port in the world. As impact of cruise ship emissions in the port of Naples on air such, it is important to understand the impact ship pollution quality using a bottom-up procedure. -ey concluded that has on air quality in this area [13–15]. -ere are currently cruise ships’ contribution seems limited but nonnegligible. two major ports in Shanghai: Yangshan and Waigaoqiao. -ese studies show that land-based monitoring can monitor -e Waigaoqiao Port is only 20 km from the city center, and ship emissions accurately. Meanwhile, various studies have air pollution caused by ship emissions in this area directly been conducted to clarify the environmental impact of aﬀects the urban air environment and the health of China’s ECAs policy. -e research of Zhang et al. [27] in- Shanghai’s residents. A previous study indicated that 70% of dicated that the SO concentration in Shanghai decreased by the global emissions from ships are produced within 400 km at least 0.229 µg/m daily on average due to the imple- of coastlines and that these emissions can cause severe mentation of the ECA policy. Wan et al. [28] compared and environmental and health problems within these regions analyzed the changes of air quality in port cities within and [16, 17]. Filonchyk and Peterson (2020) evaluated air quality outside the ECAs during the implementation of the policy. during the COVID-19 lockdown of Shanghai and found that -e results showed that the ECA policy has some time lag, daily concentrations of PM2.5, PM10, SO , NO , and CO possibly because enforcement has gradually become more 2 2 during the lockdown period were reduced by 9%, 77%, stringent. Zhang et al. [29] conducted a measurement 31.3%, 60.4%, and 3%, respectively, compared to the same campaign (SEISO-Bohai) from December 28, 2016, to period in 2019 [18]. -erefore, the mitigation of ship exhaust January 15, 2017, at Jingtang Harbor, an area within China’s emissions and the establishment of sustainable port eco- ECAs. -e results from this study indicated a positive impact systems have become urgent tasks that require complex and from fuel switching on the air quality in the study region. comprehensive systematic approaches [19]. -us, as one of From these studies, it can be seen that the monitoring of ship the busiest ports worldwide, it is important to understand emissions in ECAs can help evaluate the implementation and control the pollution from ships in Waigaoqiao Port. eﬀect of the policy. Previous studies have indicated that ship emissions At present, the requirement of China’s ECAs is mainly observed at the monitoring stations in ECAs have decreased aimed at the limitation of sulfur content in ship fuel. Ship signiﬁcantly since the aforementioned restrictions were emissions of sulfur oxides (mainly SO ) come from the implemented [20–24]. Accordingly, this study monitored sulfur content of fuel. -erefore, by detecting the change of the SO concentrations at the dock within the China’s ECAs SO content in the dock in combination with the berthing 2 2 from March 2018 to May 2018. Ship owners are likely to use information of the ship, the inﬂuence of the restriction cheaper heavy fuel during arrival or departure at the dock to policy on the air quality in the port area can be judged. As an reduce fuel costs while staying within the bounds of the alternative to the observation position of the above research, policy. When a ship is berthed, light fuel is likely used in- the dock area is not only the heaviest traﬃc area but also the stead. -erefore, in this study, changes in SO concentra- main area from which the maritime department supervises tions were used to understand the impact of the and enforces laws. An understanding of ship emissions and Atmospheric Pollution Prevention and Control Law on air atmospheric processing in dock areas is required in order to quality. develop eﬀective regulations to better manage the envi- Regarding methods for monitoring ship emissions, land- ronmental impacts of shipping. -erefore, SO monitoring based measurements provide continuous observations and equipment was installed on the bridge cranes of Waigaoqiao have been widely used. Kattner et al. [21] reported large Dock. We were allowed to observe the changes in sulfur reductions of SO in ship plumes from September 2014 to dioxide in the intensive shipping area from March 2018 to January 2015 near the mouth of Hamburg Harbor on the May 2018. Ships’ arrival and departure times from the port River Elbe. -eir results show that the vast majority (95.4%) and the wind speed and direction were also collected for of all the ships complied with the regulation of 0.1% FSC. comparative analysis. -e main purpose of this research was Yang et al. [22] measured SO continuously from the Penlee to clarify the impact of ECA policy on air quality in port Point Atmospheric Observatory near Plymouth, United areas. However, as the port area is a relatively complex Kingdom, between May 2014 and November 2015; this environment, it is diﬃcult to accurately assess the impact of coastal site is exposed to marine air across a wide wind the policy based on existing observational data. In policy sector. -eir observations suggest a 3-fold reduction in ship evaluation research, the diﬀerence-in-diﬀerence model emitted SO from 2014 to 2015. Alfoldy et al. [24] measured (DID) is an eﬀective tool to extrapolate the causal eﬀects of the chemical composition of the plumes of seagoing ships policy [27, 28]. -e primary application of the DID model is during a two-week-long measurement campaign in the port to estimate the diﬀerence in the mean outcomes of the of Rotterdam, Hoek van Holland, Netherlands, in September treated and control units posttreatment and to isolate the 2009. -e average SO emission factor was found to be outcome diﬀerence that existed before the treatment [30]. approximately half of what is allowed in ECAs, and Meanwhile, ships’ arrival and departure times and wind exceedances of this limit were rarely registered. Beecken speed and direction are the two main factors aﬀecting SO et al. [25] carried out measurements from coastal sites near concentration which were also monitored. -e monitoring the island of Kronstadt and along the Neva River in the dataset is suitable for the comparison experiment using the Advances in Meteorology 3 DID model. -erefore, we used the DID model to analyze proportional to the concentration of the gas. -e SO the collected data. -is can be employed to identify the electrochemical sensor was eﬃcient in terms of its low power impact of reducing SO emissions in policy-aﬀected regions. consumption, small size, and lightweight and being ex- tremely precise. Additionally, these sensors were capable of measuring SO concentrations at a low ppb range [32], with 2. Methods a resolution level of 1 ppb, an accuracy of 20 ppb, and a measuring range of 5 to 10,000 ppb. Overall, this kind in- 2.1. Measurement Site. As shown in Figures 1(a) and 1(b), strumentation is convenient and lightweight, which is ap- the measurement station is located on the south bank of the propriate for installation on a bridge crane. Yangtze River estuary in the Waigaoqiao Port area, north of Shanghai. In December 2018, the Waigaoqiao Port handled 1,747,000 teU, 14.987,000 tons of cargo, and 3,553 berthing 2.3. Data Treatment. -ree types of data at the monitoring vessels. In previous studies [21, 25, 29], the monitoring dock were collected from March 2018 to May 2018: wind station has been located on the shore of the channel. Under speed and direction, the concentration of SO , and ships’ an appropriate wind speed and direction, the monitoring 2 arrival at and departure from the dock. SO concentration stations used in previous studies can measure the emissions was measured once per minute from the two monitoring of a single ship that sails through the port. However, in this points. -e data from March and April are continuous data, study, the monitoring equipment was located on the sus- with observation times of 743 hours and 720 hours, re- pension bridge on the dock, as shown in Figure 1(c). We spectively. Due to the receiver server failure, there is a certain deployed two sets of equipment on two diﬀerent working degree of missing data in May. Data were gathered from 14 : bridge cranes approximately 1 km apart. During the task of 00 on May 1 to 14 : 00 on May 4 and from 0 : 00 on May 10 to loading/unloading containers, the bridge crane moved three 24 : 00 on May 21. -e total observation time for May was dimensionally within a small area, thus the position of the 354 hours. monitoring equipment varied (30–40 m in height) though -e three sets of data were normalized using each hour’s this did not aﬀect the outcome. -e bridge crane is mostly data in order to facilitate a comprehensive analysis. Using surrounded by container ships berthing. -e research of this methodology, the 2 minutes of average wind speed and Chen et al. [31] indicated that the main contributors to ship direction data that were collected and the mean values of SO emissions are containers, followed by ﬁshing ships, oil 2 concentrations observed at the two monitoring stations were tankers, and bulk carriers. -erefore, the equipment on the each considered as one-hour observations. -e ships’ arrival bridge crane can clearly detect changes in SO from its and departure times at the dock were recorded within each location. -e complexity of the monitoring points within the hour as one-hour observations. -en, SO concentration surrounding environment is shown in Figures 1(a) and 1(b). 2 data were organized into six groups based on the month and Sources of airborne SO are the ships that are in berth as well the monitoring points, as follows: Monitor-1 in March, as ships that are sailing far from the dock; at the same time, Monitor-2 in March, Monitor-1 in April, Monitor-2 in container trucks on the docks are also a source of SO . In April, Monitor-1 in May, and Monitor-2 in May. -ese data addition, it can be seen from Figure 1(b) that the sur- were then analyzed to identify the SO emission charac- roundings of monitoring points 1 and 2 diﬀer, and there are 2 teristics and their relationship with wind speed, wind di- more ship berths to the west of monitoring point 2. Higher rection, and ships’ arrival and departure. concentrations of SO may be detected when there are Several aspects inﬂuenced the results of the analysis. As westerly winds. -erefore, in this study, we analyzed the can be seen from Figure 1, in addition to nearby berthing correlation between the factors (wind speed and direction ships, the monitoring point also was able to collect data for and ships’ arrival and departure at the dock) and the ob- ships sailing in the distance and container trucks passing served results (SO concentration) rather than the emissions under the bridge crane. -ere are also many chemical plants from individual ships. within 10 km of the dock; the SO emitted from these sources will have impacted the measurements, which can 2.2. Instrumentation. In this research, SO was measured increase uncertainty. Preliminary studies showed that 1–19% of the sulfur in the fuel is emitted in other forms, with instruments from the ETL3000 series, the sensors (Unitec SENS-IT) of which meet the European Standard possibly SO or SO [33–36]. Hence, the assumption that all 3 4 sulfur is emitted as SO yields an underestimation of the true 2008/50/EC. As shown in Figure 1(c), the monitoring equipment employed in this study had a measured weight of sulfur content in the fuel. In addition, the research of Yang et al. [22] shows that the residence time of SO in the marine 12 kg, a solar power supply system, and the data commu- atmosphere is approximately 0.5 d, with dry deposition nication mode was 4G cellular broadband. It included a data acquisition and processing unit, modular gas sensor, solar explaining about a quarter of the total SO sink. -erefore, the SO measured by the sensor may be either from the ship power supply system, and wireless data transmission module and a software system that was directly installed outdoors that is emitting it or from ship emissions within half a day. Unfortunately, there are no quantitative data for the without any cooling device or air-processing device. -e SO sensor used the electrochemical method, where the elec- abovementioned uncertainty. Although the background value can approximate from the observed value when there trochemical sensor determined the concentration of the gas through a redox reaction and produced an electrical signal is no ship, the endogenous problems arise when reverse 4 Advances in Meteorology (a) (b) (c) Figure 1: (a) Location of the measurement station. (b) Satellite view of the monitoring environment, with the location of the two survey stations shown in red (map source: lbs.amap.com). (c) Equipment installed at a suspension bridge on the dock. causality occurs between the variables or some variables are these data were analyzed in relation to the equipment in- missing. However, by using the DID model, the observation stallation location and surrounding environment (Figure 1), data are robust enough that they can provide useful infer- it could be concluded that the results of this experiment were ences. It can be deﬁned as follows: less aﬀected by the sailing ships and that the main source of SO is the discharge of berthing ships and the pollution 1,2 2 (1) Y � α + α du + a dt + a du ∗ dt + a Z + ε , it 0 1 2 3 4 it it sources on the shore. -e arrival and departure of ships at the dock are where subscripts i and t represent the i-th monitoring point considered to be a major factor aﬀecting air quality. and t-th month, respectively, and superscripts 1 and 2 -erefore, we counted the distribution of SO during these represent monitoring points 1 and 2, respectively. Z rep- two periods, respectively. -e results displayed in Figure 4 resents a series of control variables, ε is a random term (or show that SO concentration in the six observation groups is disturbance), and the interpreted variable is the monitored higher during the ships’ arrival and departure times than at SO value. times when there was no movement of ships in or out of the According to the ECA policy, ships are allowed to use dock. -erefore, all six sets of observations reﬂect the high-sulfur fuel within one hour of their arrival or departure. positive impact of fuel switching on air quality. Moreover, To analyze the inﬂuence of this factor on air quality, we used berths around Monitor-1 are only for ocean-going vessels, dt as the control parameter. dt � 1 means that a ship arrived while berths around Monitor-2 are for river vessels and or departed the dock during this hour; otherwise, dt � 0. ocean-going vessels. To a certain extent, this may have led to Wind direction is also a major inﬂuencing factor. du was a diﬀerence in the concentration of SO observed between used as the control parameter to classify the monitoring data the monitors. At Monitor-2, the concentrations of SO of downwind (wind direction: 180–270 degrees) as the measured were higher, but the diﬀerences between periods treatment group and the nondownwind data as the control of ship movement and periods of no ship movement were group. Because the treatment group and the control group smaller compared to that of Monitor-1. are measured by the same equipment at the same location, In Figure 5, we can see the diﬀerence in the overall they meet the parallel trend required by the DID model. distribution of the average SO concentrations per hour in When du � 1, the monitoring equipment is leeward of the two diﬀerent contexts. Within one hour of a ship’s arrival at berthing ships; that is, the discharge of the berthing ships is or departure from the dock, the time quantum for SO mainly measured by the monitoring equipment. Otherwise, concentrations exceeded 5 µg/m , which was higher than du � 0. -e DID model parameters are explained according that during no ship movement period. Meanwhile, the to the above deﬁnition as listed in Table 1. overall distribution of SO was also slightly larger when a ship arrived at or departed from the dock. In addition, there 3. Results and Discussion are more ship berths to the west of Monitor-2; therefore, results could be signiﬁcantly underestimated at this point, -e overall distribution of the average SO concentrations considering the prevalent wind direction. per hour observed at the two monitoring points is shown in From Figures 4 and 5, we can see that ships’ arrival and Figure 2. It can be seen that the concentration of SO is departure at the dock are a relatively obvious inﬂuence on air mainly distributed between 0 and 5 µg/m (77.47% within quality; however, there is no quantitative index. In addition, the observation time range). -erefore, 5 µg/m was used as wind speed and direction can also produce monitoring the standard for determining high or low SO concentration results which are diﬃcult to assess. -erefore, the DID in this study. As shown in Figure 3 (which includes data for model discussed in section 2.3 was used for analysis and the duration of the experiment), the wind mostly originated calculation, and the results are shown in Table 2. We from the southwest during the observation period. When Advances in Meteorology 5 Table 1: DID model parameters. No ship movement in or out of the dock Movement of ship(s) in or out of the dock Diﬀerence (dt � 0) (dt � 1) Downwind (treatment group, α + α α + α + α + α ΔY � α + α 0 1 0 1 2 3 2 3 du � 1) No downwind (control group, α α + α ΔY � α 0 0 2 2 du � 0) DID model estimation α α + α ΔY � α 1 1 3 3 Monitors 1 and 2, March to May 2018 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10 μg/m Sulfur dioxide concentration Monitor-1 in March Monitor-1 in May Monitor-1 in April Monitor-2 in March Monitor-2 in April Monitor-2 in May Figure 2: Overall distribution of the average SO concentrations per hour observed at the two monitoring points, where the interval is 1 µg/ m , and the number of observation hours in each concentration interval were counted. NNW NNE NW NE WNW ENE W E WSW ESE SW SE SSW SSE –1 Wind speed (ms ) 0-1 3-4 6-7 1-2 4-5 ≥7 2-3 5-6 Figure 3: Wind speed and direction per hour. -e numbers on the diagram are nonevenly spaced scale lines. Number of observation hours 6 Advances in Meteorology Monitor-2 in May Monitor-2 in April Monitor-2 in March Monitor-1 in May Monitor-1 in April Monitor-1 in March 0 1 23 4 56 Sulfur dioxide concentration μg/m No movement of ships in or out of the dock Movement of ship(s) in or out of the dock Figure 4: Average SO concentration within one hour of ships’ arrival or departure versus periods of no ship movement in or out of the dock. 0 3 µg/m 0-1 1-2 2-3 3-4 4-5 5+ Monitor-1 in March Monitor-1 in April Monitor-1 in May Monitor-2 in March Monitor-2 in April Monitor-2 in May (a) % 30 µg/m 0-1 1-2 2-3 3-4 4-5 5+ Monitor-1 in March Monitor-1 in April Monitor-1 in May Monitor-2 in March Monitor-2 in April Monitor-2 in May (b) Figure 5: Percentage distribution of average SO concentrations per hour for (a) periods of ship movement in or out the dock versus (b) periods of no ship movement in or out the dock. Advances in Meteorology 7 Table 2: DID model results. No ship movement in or out of the dock Movement of ship(s) in or out of the dock Diﬀerence (dt � 0) (dt � 1) Monitoring point 1, March 2018 Downwind (treatment group, 2.25 4.24 1.99 du � 1) No wind (control group, du � 0) 3.18 1.42 −1.76 DID model estimation −0.93 2.82 3.75 Monitoring point 2, March 2018 Downwind (treatment group, 5.06 5.35 0.29 du � 1) No wind (control group, du � 0) 4.02 5.22 1.2 DID model estimation 1.04 0.13 −0.91 Monitoring point 1, April 2018 Downwind (treatment group, 3.56 3.86 0.3 du � 1) No wind (control group, du � 0) 2.72 2.18 −0.54 DID model estimation 0.84 1.68 0.84 Monitoring point 2, April 2018 Downwind (treatment group, 4.33 4.61 0.28 du � 1) No wind (control group, du � 0) 4.11 3.95 −0.16 DID model estimation 0.22 0.66 0.44 Monitoring point 1, May 2018 Downwind (treatment group, 3.25 3.71 0.46 du � 1) No wind (control group, du � 0) 6.02 9.91 3.89 DID model estimation −2.77 −6.2 −3.43 Monitoring point 2, May 2018 Downwind (treatment group, 3.52 4.15 0.63 du � 1) No wind (control group, du � 0) 3.96 2.96 −1 DID model estimation −0.44 1.19 1.63 calculated the average SO concentration per period. All and its inﬂuence is especially signiﬁcant when the wind measurement data were divided into the treatment group direction is southwest. and control group, and the ships’ arrival and departure times In addition, we conducted statistics on the time period were taken as the control conditions. Based on the calcu- during which the sulfur dioxide concentration measured at lation results of the DID model, the inﬂuence of the two the two monitoring points exceeded 5 µg/m . As shown in factors (i.e., ships’ arrival or departure time and wind speed Figure 6, when the SO concentration exceeded 5 µg/m at and direction) on the concentration of SO was judged. only one monitoring point when measured at two points, In the treatment group, the eﬀect of arrival or departure both frequency ratios of the arrival and departure of ships time is obvious and signiﬁcant, whereas it is not obvious in were higher than that within the period in which the SO the control group. -is means that SO concentration rose concentration was less than 5 µg/m at both monitoring signiﬁcantly in the period of ship arrival or departure when points. In other words, there is a positive correlation be- the wind direction was southwest. In contrast, when the tween the high concentration of SO and the increased wind direction is not southwest, the variation trend of SO frequency of ships arriving and departing from the port area. concentration is not obvious in the period of ship arrival or -is result also reﬂects the positive impact of fuel switching. departure. Similarly, when there is no arrival and departure Several studies have monitored and assessed the impact of ships, the inﬂuence of wind direction on the changing of ECAs [21–24, 27–29]. However, the impact of individual trend of SO is not obvious. In contrast, when ships are ship activities on air quality is rarely monitored, making it moving in or out of the dock, the concentration of SO virtually impossible to assess the impact of ECA policies on increases signiﬁcantly when the wind is downwind. Finally, air quality in local areas such as wharves. -e activities of two sets of DID model results (ΔΔY) are negative, which can ships and personnel in dock areas tend to be relatively be attributed to the small number of data samples. In spite of frequent, and ship emissions directly aﬀect public health. this, four sets of DID model results (ΔΔY) are positive and -erefore, this study sought to establish monitoring data on present the impact of these two factors on SO concentra- ship emissions in one of Shanghai’s main ports by installing tions. On the whole, the calculation results of the DID model two monitoring stations at Waigaoqiao wharf to explore the mentioned above show that the SO concentration in the air relationships between SO concentrations, wind speed and 2 2 of the port area increases due to the use of high-sulfur fuel, wind direction, and ships’ arrival and departure times. 8 Advances in Meteorology 313 28 % 50 94 66 25 147 13 Both > 5 One > 5 Both < 5 Both > 5 One > 5 Both < 5 Both > 5 One > 5 Both < 5 March April May Movement of ship(s) in or out of the dock No movement of ships in or out of the dock Figure 6: SO concentrations during periods of ship movement in or out of the dock versus periods of no movement of ship(s) in or out of the dock expressed in percentage terms. For each month (March, April, or May), SO concentrations were divided into three categories: SO 2 2 3 3 concentrations at both monitoring points greater than 5 µg/m (both >5), SO concentrations greater than 5 µg/m at only one monitoring station (one >5), SO concentrations less than 5 µg/m at both stations (both <5). Considering the results presented in Figures 4–6, it can be concentration of SO was low (less than 5 µg/m ) 77.47% of concluded that ship emissions during the arrival or de- the time. -e distribution of wind speed and direction parture of ships signiﬁcantly aﬀect the airborne SO con- indicated that there was little change in the wind direction centration in the study area. -e results of the DID model during the study period. -e data observed at the moni- presented in Table 1 further indicate that SO concentration toring stations were mostly attributable to the pollution in the air increases signiﬁcantly due to the use of high-sulfur sources in the berth and on the shore. -e change in SO fuel when the wind direction is southwest. -is is signiﬁcant concentration during periods of ship movement in or out in the context of recently implemented policies which of the dock compared to when there was no ship activity at stipulate that the FSC cannot exceed 0.5% (m/m) during the dock is indicative of the contribution of a ship’s berthing within ECAs, excluding the ﬁrst hour postarrival emissions to the SO concentration in the air. -e result of and the last hour predeparture. -us, we were able to clarify the DID model shows that the concentration of SO in the the impact of ECA policy on air quality in port areas. We air of the port area increased due to the use of high-sulfur found that the recently implemented policies may need fuel and that its inﬂuence is especially signiﬁcant when the amendment in the future to consider our ﬁndings. wind direction is downwind. -is ﬁnding is relevant to the implementation of ECA policy, which requires that the FSC cannot exceed 0.5% (m/m) during berthing within the 4. Conclusions ECAs, except for the ﬁrst hour after arrival and the last -e implementation of the ECA policy can eﬀectively reduce hour before departure. -e results of this study can be used the harm caused by ship emissions. How to evaluate the as the basis for understanding ship emissions and atmo- impact of the implementation of this policy on the envi- spheric processes within dock areas under ECA policy. ronment is a research topic of interest. In this study, two SO Subsequently, more attention should be paid to the eﬀects monitoring stations were installed on the bridge crane in of sailing ship emissions. Waigaoqiao Dock, Shanghai, China, which allowed data to be obtained from March 2018 to May 2018. To analyze the Data Availability factors that caused a change in the SO concentration, we collected the data on the wind speed and direction as well as -e data used to support the ﬁndings in this study are the arrival and departure times of ships at the port in the available at https://data.mendeley.com/datasets/427wfrck7w/ same period and normalized them to one hour. -en, the draft?a�883bb7b4-b9f1-4549-b10c-9ad9939a3e61. DID model was used to evaluate the impact of policy implementation. Conflicts of Interest -e overall airborne distribution of SO at Waigaoqiao Dock was obtained. It was determined that the -e authors declare no conﬂicts of interest. Advances in Meteorology 9 Journal of Geovisualization and Spatial Analysis, vol. 2, no. 12, Acknowledgments pp. 1–18, 2018. [14] X. N Wang, Y. Shen, Y. F Lin et al., “Atmospheric pollution -e authors would like to thank Pudong Maritime Safety from ships and its impact on local air quality at a port site in Administration of the People’s Republic of China for their Shanghai,” Atmospheric Chemistry and Physics, vol. 19, no. 9, support in coordinating the ﬁeld measurements. -is re- pp. 6315–6330, 2019. search was supported by the National Natural Science [15] Y. Y Zhou, Y. Zhang, D. Ma et al., “Port-related emissions, Foundation of China (Grant nos. 41701523 and 61703271), environmental impacts and their implication on green traﬃc the National Key Research and Development Project of policy in Shanghai,” Sustainability, vol. 12, no. 10, p. 17, 2020. China (Grant no. 2020YFC1511901), and the Special De- [16] J. J. Corbett, P. S. Fischbeck, and S. N. Pandis, “Global ni- velopment Fund for China (Shanghai) Pilot Free-Trade trogen and sulfur inventories for oceangoing ships,” Journal of Zone. Geophysical Research: Atmospheres, vol. 104, no. D3, pp. 3457–3470, 1999. [17] Y. Gonzalez, ´ S. Rodr´ıguez, J. C. Trujillo, and R. Garc´ıa, References “Ultraﬁne particles pollution in urban coastal air due to ship [1] UNCTAD, “World seaborne trade by types of cargo and by emissions,” Atmospheric Environment, vol. 45, no. 28, group of economies, annual, United Nations Conference on pp. 4907–4914, 2011. Trade and Development,” 2017, https://unctadstat.unctad. [18] M. Filonchyk and M. Peterson, “Air quality changes in org/wds/TableViewer/tableView.aspx?ReportId=32363. Shanghai, China, and the surrounding urban agglomeration [2] V. Eyring, I. S. A. Isaksen, T. Berntsen et al., “Transport during the COVID-19 lockdown,” Journal of Geovisualization impacts on atmosphere and climate: shipping,” Atmospheric and Spatial Analysis, vol. 4, no. 22, pp. 1–7, 2020. Environment, vol. 44, no. 37, pp. 4735–4771, 2010. [19] X. Geng, Y. Wen, C. Zhou, and C. Xiao, “Establishment of the [3] A. Sorooshian and H. T. Duong, “Ocean emission eﬀects on sustainable ecosystem for the regional shipping industry aerosol-cloud interactions: insights from two case studies,” based on system dynamics,” Sustainability, vol. 9, no. 5, p. 18, Advances in Meteorology, vol. 2010, Article ID 301395, 9 pages, Article ID 742, 2017. [20] N. Molders, ¨ S. Gende, and M. Pirhalla, “Assessment of cruise- [4] C. Gencarelli, I. Hedgecock, F. Sprovieri, G. Schurmann, ¨ and ship activity inﬂuences on emissions, air quality, and visibility N. Pirrone, “Importance of ship emissions to local sum- in Glacier Bay National Park,” Atmospheric Pollution Re- mertime ozone production in the mediterranean marine search, vol. 4, no. 4, pp. 435–445, 2013. boundary layer: a modeling study,” Atmosphere, vol. 5, no. 4, [21] L. Kattner, B. Mathieu-Uﬃng, J. P. Burrows et al., “Moni- pp. 937–958, 2014. toring compliance with sulfur content regulations of shipping [5] H. Liu, M. Fu, X. X Jin et al., “Health and climate impacts of fuel by in situ measurements of ship emissions,” Atmospheric ocean-going vessels in East Asia,” Nature Climate Change, Chemistry and Physics, vol. 15, no. 17, pp. 10087–10092, 2015. vol. 6, no. 11, pp. 1037–1041, 2016. [22] M. Yang, T. G. Bell, F. E. Hopkins, and T. J. Smyth, “Attri- [6] G.-H. Yu, S. Park, S.-K. Shin, K.-H. Lee, and H.-G. Nam, bution of atmospheric sulfur dioxide over the English “Enhanced light absorption due to aerosol particles in ship Channel to dimethyl sulﬁde and changing ship emissions,” plumes observed at a seashore site,” Atmospheric Pollution Atmospheric Chemistry and Physics, vol. 16, no. 8, pp. 4771– Research, vol. 9, no. 6, pp. 1177–1183, 2018. 4783, 2016. [7] J. X. Zhou, S. Zhou, and Y. Q. Zhu, “Characterization of [23] L. Barregard, P. Molnar, J. E. Jonson et al., “Impact on particle and gaseous emissions from marine diesel engines population health of baltic shipping emissions,” International with diﬀerent fuels and impact of after-treatment technology,” Journal of Environmental Research and Public Health, vol. 16, Energies, vol. 10, no. 8, p. 14, Article ID 1110, 2017. no. 11, 11 pages, Article ID 1954, 2019. [8] IMO, “Sulphur oxides (SOx)–regulation 14,” 2017, http:// [24] B. Alfoldy, ¨ J. B. Lo¨ov, ¨ F. Lagler et al., “Measurements of air www.imo.org/en/OurWork/Environment/PollutionPrevention/ pollution emission factors for marine transportation in AirPollution/Pages/Sulphur-oxides-(SOx)-Regulation-14.aspx. SECA,” Atmospheric Measurement Techniques, vol. 6, no. 7, [9] IMO, “Emission control areas (ECAs) designated under MAR- pp. 1777–1791, 2013. POL Annex VI,” 2017, http://www.imo.org/en/OurWork/ [25] J. Beecken, J. Mellqvist, K. Salo et al., “Emission factors of Environment/PollutionPrevention/AirPollution/Pages/Emission- SO NO and particles from ships in Neva Bay from ground- 2 x Control-Areas-(ECAs)-designated-under-regulation-13-of-MAR based and helicopter-borne measurements and AIS-based POL-Annex-VI-(NOx-emission-control. modeling,” Atmospheric Measurement Techniques, vol. 15, [10] Standing Committee of the National People’s Congress, no. 9, pp. 5229–5241, 2015. “Atmospheric pollution prevention and control law of the [26] F. Murena, L. Mocerino, F. Quaranta et al., “Impact on air People’s Republic of China,” 2015, http://english.court.gov. quality of cruise ship emissions in Naples, Italy,” Atmospheric cn/2016-04/15/content_24565639.htm. Environment, vol. 187, pp. 70–83, 2018. [11] Ministry of Transport of the People’s Republic of China, [27] Q. Zhang, Z. Q Zheng, Z. Wan et al., “Does emission control “Marine emission control area plan for Pearl river delta, area policy reduce sulfur dioxides concentration in Shang- Yangtzy river delta, Bohai Rim area,” 2015, http://www.gov. hai?” Transportation Research Part D, vol. 81, Article ID cn/xinwen/2015-12/04/content_5019932.htm. 102289, 2020. [12] Shanghai Municipal Bureau of Statistics, “-e statistic com- [28] Z. Wan, X. J Zhou, Q. Zhang et al., “Do ship emission control munique of Shanghai on the 2017 national economy and areas in China reduce sulfur dioxide concentrations in local social development,” 2017, http://www.stats.gov.cn/english/ pressrelease/201802/t20180228_1585666.html. air? A study on causal eﬀect using the diﬀerence-in-diﬀerence model,” Marine Pollution Bulletin, vol. 149, Article ID 110506, [13] Y. Y. Huang, Q. W. Yan, and C. R. Zhang, “Spatial-temporal distribution characteristics of PM2.5 in China in 2016,” 2019. 10 Advances in Meteorology [29] Y. N Zhang, F. Y Deng, H. L Man et al., “Compliance and port air quality features with respect to ship fuel switching regu- lation: a ﬁeld observation campaign, SEISO-Bohai,” Atmo- spheric Measurement Techniques, vol. 19, no. 7, pp. 4899–4916, [30] M. S. Delgado and R. J. G. M. Florax, “Diﬀerence-in-diﬀer- ences techniques for spatial data: local autocorrelation and spatial interaction,” Economics Letters, vol. 137, pp. 123–126, [31] D. S. Chen, X. T. Wang, P. Nelson et al., “Ship emission inventory and its impact on the PM 2.5 air pollution in Qingdao Port, North China,” Atmospheric Environment, vol. 166, pp. 351–361, 2017. [32] A. W. E. Hodgson, P. Jacquinot, and P. C. Hauser, “Elec- trochemical sensor for the detection of SO in the low-ppb range,” Analytical Chemistry, vol. 71, no. 14, pp. 2831–2837, [33] H. Schlager, R. Baumann, M. Lichtenstern et al., “Aircraft- based trace gas measurements in a primary European ship corridor,” Proceedings of the TAC-Conference, pp. 83–88, [34] H. Agrawal, W. A. Welch, J. W. Miller et al., “Emission measurements from a crude oil tanker at sea,” Environmental Science and Technology, vol. 42, no. 19, pp. 7098–7103, 2008. [35] J. Moldanova, E. Fridell, O. Popovicheva et al., “Character- isation of particulate matter and gaseous emissions from a large ship diesel engine,” Atmospheric Environment, vol. 43, no. 16, pp. 2632–2641, 2009. [36] J. M. Balzani Loov, B. Alfoldy, L. F. L. Gast et al., “Field test of available methods to measure remotely SOx and NOx emissions from ships,” Atmospheric Measurement Techniques, vol. 7, no. 8, pp. 2597–2613, 2014.

Journal

Advances in Meteorology – Hindawi Publishing Corporation

Published: Feb 9, 2021

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

Determination of the Influence of Fuel Switching Regulation on the Sulfur Dioxide Content of Air in a Port Area Using DID Model

References (39)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies