PreprintPDF Available

The Climate Impact of High Seas Shipping

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Strict carbon emission regulations are set with respect to countries’ territorial seas or exclusive economic zone shipping activities to meet their climate change commitment under the Paris Agreement. However, no shipping carbon mitigation policies are proposed for the world high sea regions, which results in carbon intensive shipping activities on high seas. In this paper, we propose a Geographic-based Emission Estimation Model (GEEM) to estimate shipping GHG emission patterns on high sea regions. The results indicate that annual shipping CO2-e emissions on high seas reach 211.60 million metric tonnes in 2019, accounting for about one-third of all shipping emissions globally and exceeding annual GHG emissions of countries such as Spain. The average emission from shipping activities on high seas is growing at approximately 7.26% per year, which far surpasses the global shipping emission growth rate of 2.23%. We propose the implementation policies on each high seas region with respect to the main emission driver of each high seas identified from our results. Our policy evaluation results show that carbon mitigation policies could reduce 25.46 and 54.36 million tonnes CO2-e in the primary intervention stage and overall intervention stage respectively, with 12.09% and 25.81% reduction rates in comparison to the 2019 annual high seas shipping GHG emission.
Content may be subject to copyright.
The Climate Impact of High Seas Shipping
Shouyang Wang ( sywang@amss.ac.cn )
Academy of Mathematics and Systems Science, Chinese Academy of Sciences
Yuze Li
Questrom School of Business, Boston University
Peng Jia
Collaborative Innovation Center for Transport Studies, Dalian Maritime University
Shangrong Jiang
University of Chinese Academy of Sciences https://orcid.org/0000-0002-2780-9075
Haijiang Li
Collaborative Innovation Center for Transport Studies, Dalian Maritime University
Haibo Kuang
Collaborative Innovation Center for Transport Studies, Dalian Maritime University
Yongmiao Hong
Academy of Mathematics and Systems Science, Chinese Academy of Sciences
Xueting Zhao
Collaborative Innovation Center for Transport Studies, Dalian Maritime University
Dabo Guan
Department of Earth System Science, Tsinghua University, Beijing, China.
Physical Sciences - Article
Keywords: High seas, International shipping, GHG emission, Policy evaluation, Emission drivers
Posted Date: January 28th, 2022
DOI: https://doi.org/10.21203/rs.3.rs-1300753/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License. 
Read Full License
1
The Climate Impact of High Seas Shipping
2
3
Yu z e L i 1,#, Peng Jia2,3,4,#, Shangrong Jiang4,#, Haijiang Li2,3, Haibo Kuang2,3, Yongmiao Hong 4,5,6, Shouyang
4
Wang4,5,6,*, Xueting Zhao2,3, Dabo Guan7,8,*
5
1 Questrom School of Business, Boston University, Boston Massachusetts 02215, United States
6
2 Collaborative Innovation Center for Transport Studies, Dalian Maritime University, Dalian 116026, China
7
3 School of Maritime Economics and Management, Dalian Maritime University, Dalian 116026, China
8
4 School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190,China
9
5 Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China
10
6 Center for Forecasting Science, Chinese Academy of Sciences, Beijing 100190, China
11
7 Department of Earth System Science, Tsinghua University, Beijing 100080, China
12
8 School of International Development, University of East Anglia, Norwich NR4 7TJ, UK
13
# These authors are co-first authors and contributed equally: Yuze Li, Peng Jia, Shangrong Jiang
14
15
16
Abstract
17
18
Strict carbon emission regulations are set with respect to countries’ territorial seas or exclusive economic zone
19
shipping activities to meet their climate change commitment under the Paris Agreement. However, no shipping
20
carbon mitigation policies are proposed for the world high sea regions, which results in carbon intensive
21
shipping activities on high seas. In this paper, we propose a Geographic-based Emission Estimation Model
22
(GEEM) to estimate shipping GHG emission patterns on high sea regions. The results indicate that annual
23
shipping CO2-e emissions on high seas reach 211.60 million metric tonnes in 2019, accounting for about one-
24
third of all shipping emissions globally and exceeding annual GHG emissions of countries such as Spain. The
25
average emission from shipping activities on high seas is growing at approximately 7.26% per year, which far
26
surpasses the global shipping emission growth rate of 2.23%. We propose the implementation policies on each
27
high seas region with respect to the main emission driver of each high seas identified from our results. Our
28
policy evaluation results show that carbon mitigation policies could reduce 25.46 and 54.36 million tonnes
29
CO2-e in the primary intervention stage and overall intervention stage respectively, with 12.09% and 25.81%
30
reduction rates in comparison to the 2019 annual high seas shipping GHG emission.
31
32
Keywords: High seas; International shipping; GHG emission; Policy evaluation; Emission drivers
33
34
* Corresponding authors e-mails: sywang@amss.ac.cn; guandabo@tsinghua.edu.cn
2
Over the past decades, accelerated international and regional maritime trading activities have boosted
35
worldwide ocean-going shipping industry development. The associated shipping greenhouse gas
36
emission, however, has gradually become a non-neglectable issue against worldwide decarbonation
37
and climate change goals1. As estimated by the International Maritime Organization (IMO) voyage-
38
based method2, international shipping accounts for 755 million metric tonnes annual GHG emissions
39
in 2018. In order to improve emission reduction in maritime industry and to meet climate change
40
commitment under the Paris Agreement, many countries have submitted concrete plans and
41
implemented strict carbon emission regulations for shipping activities in their territory waters or
42
exclusive economic zones, such as emission controlled areas, alternative fuel substitution, electric or
43
nuclear propulsion adoption and renewable energy propulsion assistance etc.3,4,5 Previous maritime
44
related studies focus on the territory waters and exclusive economic zones carbon mitigation policy
45
effectiveness evaluation and improvement6,7. While the Paris Agreement clearly outlines the emission
46
reduction plans for each country’s territorial sea, little attention has been paid to the fast-growing
47
emission on the international high seas. Due to the non-sovereign property of high seas, no signatories
48
are directly responsible for high seas carbon emission reduction under the Paris Agreement and thus,
49
there have been no carbon mitigation policies or environmental regulations proposed or implemented
50
in these regions. As a result, ships travelling on the high seas often operate in economically efficient
51
manners, such as utilizing heavy fuel oil and travelling at high speeds, without any environmental
52
concerns8,9. Since the high seas account for more than two-thirds of the world ocean regions, the
53
carbon intensive shipping activities on the high seas could become a potential barrier against the
54
worldwide carbon mitigation and sustainability efforts.
55
56
In comparison to the top-down approaches utilized by previous maritime GHG inventory estimation
57
studies, bottom-up approaches enable more accurate estimation results by summing up detailed
58
individual ship emission outputs10,11. There are currently two bottom-up approaches used in the
59
existing literature to calculate shipping GHG emission for certain regions or countries, namely vessel-
60
based and voyage-based. However, these methods heavily rely on their assumptions: the vessel-based
61
method assumes that vessels with similar type and age have uniform shipping behaviors; the voyage-
62
3
based method assumes that international shipping emissions are those occurring on a voyage between
63
two ports in different countries2,12. In this paper, we propose a new Geographic-based Emission
64
Estimation Model (GEEM) to estimate international shipping GHG emission patterns on high seas
65
regions. By incorporating the IHS Market Maritime & Trade vessel technical specification data and
66
Automatic Identification System (AIS) data as our GEEM static and dynamic datasets respectively,
67
our GEEM method can be viewed as a bottom-up approach that identifies GHG inventories through
68
the broadly covered individual ship navigation information. In comparison to the existing two bottom-
69
up GHG emission approaches, our GEEM method utilizes real-time geographic coordinates data,
70
which enables more accurate and robust high seas emission estimation by relaxing the
71
aforementioned assumptions. In particular, we exclude international vessels that navigate between
72
two countries’ exclusive economic zones from high seas GHG emissions, and we include domestic
73
vessels whose route cover high seas regions in high seas GHG emissions. In essence, our GEEM
74
approach collects shipping navigation and emission data for those only occurring on high seas
75
geographically, and accordingly estimate our high seas shipping GHG emission results.
76
77
Our GEEM results indicate that the annual shipping CO2-e emissions on high seas reach 211.60
78
million metric tons per year (Mmt/yr) in 2019, accounting for about one-third of all shipping
79
emissions globally. The high seas shipping emission in 2019 exceeds annual greenhouse gas
80
emissions of countries such as Spain, Argentina and United Arab Emirates. More alarmingly, the
81
average emission from shipping activities on the high seas is growing at approximately 7.26% per
82
year, which far surpasses the global shipping emission growth rate of 2.23% per year. By classifying
83
the worldwide high seas into 8 geographic regions and incorporating detailed vessel dynamic data of
84
all ships from all routes, we find that there exists a great degree of heterogeneity in key factors that
85
drive the shipping emission patterns across different high seas regions. As suggested by commonly-
86
adopted maritime carbon mitigation regulations, we propose the primary implementation policies on
87
each high seas region with respect to the main emission driver of each high seas identified from our
88
results. Our evaluation results show that carbon mitigation policies could reduce 25.46 and 54.36
89
million tonnes CO2-e in the primary intervention stage and overall intervention stage respectively,
90
4
with 12.09% and 25.81% reduction rates in comparison to the 2019 annual high seas shipping GHG
91
emission. Indeed, to regulate high seas shipping activities via the global maritime industry effort,
92
international high seas shipping can contribute to the world trading and economic growth in a more
93
environmentally-friendly manner.
94
95
96
Fig. 1| High seas classification and shipping allocation, according to the United Nations Convention on the
97
Law of the Sea and the International Hydrographic Organization division standard (www.marineregions.org),
98
we divide the worldwide high seas into 8 geographic regions and calculate their shipping miles respectively,
99
namely North Pacific Ocean High Seas (a), South Pacific Ocean High Seas (b), North Atlantic Ocean High
100
Seas (c), South Atlantic Ocean High Seas (d), Arctic Ocean High Seas (e), Southern Ocean High Seas (f),
101
Indian Ocean High Seas (g), and Other High Seas (h). The yellow bars in each subfigure indicate the annual
102
shipping miles (million miles) of each high seas from 2015 to 2019.
103
104
105
117.71 131.19 138.05
178.62
196.72
0
50
100
150
200
250
2015 2016 2017 2018 2019
Other High Seas
256.90
283.31
314.11 309.75 326.24
0
50
100
150
200
250
300
350
2015 2016 2017 2018 2019
High Sea of the Indian Ocean
0.13 0.18
0.76 0.75
0.59
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
2015 2016 2017 2018 2019
High Sea of the Southern Ocean
0.10 0.09
0.19
0.23
0.39
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
2015 2016 2017 2018 2019
High Sea of the Arctic Ocean
280.68
299.67
309.50
314.24
319.54
260
270
280
290
300
310
320
330
2015 2016 2017 2018 2019
High Sea of the North Pacific Ocean
a
a
b
b
c
c
d
d
e
e
f
f
g
g
h
h
5
Shipping Activities and Emissions on High Seas. In this study, we follow the high seas division
106
standard provided by the International Hydrographic Organization and classify the worldwide high
107
seas into 8 geographic regions, namely the North Pacific Ocean High Seas, South Pacific Ocean High
108
Seas, North Atlantic Ocean High Seas, South Atlantic Ocean High Seas, Arctic Ocean High Seas,
109
Southern Ocean High Seas, Indian Ocean High Seas, and Other High Seas. Using an AIS-based
110
method, we first calculate the annual shipping activities of each region from 2015 to 2019 by adding
111
up the distance travelled (in nautical miles) from all routes for all ships within the respective areas.
112
Figure 1 illustrates the longitudinal and latitudinal boundaries and the respective shipping activities
113
of each high seas region. Of the total shipping activities in 2019, about 23.43% were from the North
114
Atlantic Ocean High Seas, which reflects the high traffic between European, North American, and
115
South American countries. As shown in Figure 2, the shipping activities between the United States
116
and the United Kingdom, Brazil and Spain, as well as United States and Brazil are amongst the busiest
117
routes in the region. The Indian Ocean High Seas accounted for approximately 20.98% of the total
118
shipping activities, which can be mainly attributed to the shipping activities along Asian countries. In
119
particular, the shipping route between China and Australia is responsible for nearly 30% of the traffic
120
in the region. The North Pacific Ocean High Seas is another region with significant shipping activities,
121
contributing about 20.55% as most of the shipping routes between Asia and North America passes
122
through this area. In particular, the shipping route from China to the United States, Korea to the
123
United States, and Japan to the United States exhibits the most amount of traffic in the region. Overall,
124
the resulting total shipping mileage in the high seas has exceeded a total of 1.56 billion nautical miles
125
(nm) in 2019 with an average annual growth rate of 6.81% in the past 5 years. The review of maritime
126
transport by UNCTAD indicate that international maritime trade expanded at 4.7%-6.7% annually
127
from 2015 to 2019, with total volumes amounting to 11 billion tons in 201813. The accelerated
128
international maritime trade volumes boost shipping activities on the high seas, especially for those
129
rapidly developed shipping routes such as South–South trade, Belt and Road Initiative by China,
130
Panama Canal and Suez Canal related seaborne trade13,14,15.
131
132
6
133
Fig. 2| High seas shipping mileage distribution, based on our GEEM calculation results, a-f report the average top 10
134
shipping mileage routes for North Pacific Ocean High Sea (a), South Pacific Ocean High Sea (b), North Atlantic Ocean
135
High Sea (c), South Atlantic Ocean High Sea (d), Indian Ocean High Sea (e), and Other High Seas (f). The specific
136
percentage in a-f indicate shipping mileage proportion of each high seas region during our GEEM sample period.
137
138
We collect and process 79613 vessels’ high seas shipping records in total, which represent a majority
139
of the international shipping fleet. We also utilize the 3-minute frequency AIS data spanning from
140
January 2015 to December 2019, which accounts for a total of 5.03 TB raw AIS dataset. Using the
141
detailed AIS messages transmitted by all ships from all routes, we calculate CO2, SO2, NO2,
142
particulate matter 2.5 (PM2.5), particulate matter 10 (PM10), CO, non-methane volatile organic
143
compounds (NMVOCs), CH4 and N2O emissions generated by all vessels for each of the eight world
144
high seas regions from 2015 to 2019. We calculate the carbon dioxide equivalent (CO2-e) by
145
switching other pollutants to CO2 to standardize the climate impact of high seas GHG emission. The
146
results indicate that the rapid development of shipping activities has resulted in a significant increase
147
in emissions in the identified high seas regions. During this time, the total CO2-e emissions rose from
148
163.98 Mmt/yr to 211.60 Mmt/yr, accounting for about one-third of all shipping emissions globally.
149
2.92%
3.13%
3.47%
4.22%
4.69%
4.90%
5.85%
6.19%
6.60%
9.72%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00%
Australia to Japan
US to Korea
US to US
Japan to Australia
US to Japan
China to Canada
US to China
Japan to US
Korea to US
China to US
1.89%
2.08%
2.08%
2.46%
3.03%
4.16%
4.54%
9.08%
9.27%
13.05%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00%
Chile to Chile
Panama to Chile
Chile to Peru
Argentina to Peru
New Zealand to China
Peru to Chile
New Zealand to Australia
Australia to New Zealand
Australia to Japan
Japan to Australia
1.72%
2.15%
2.49%
2.57%
3.00%
3.09%
3.69%
4.20%
5.15%
5.41%
0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00%
UK to US
Germany to US
the Netherlands to Canada
Brazil to the Netherlands
US to Spain
Canada to the Netherlands
US to Brazil
Spain to Brazil
Brazil to Spain
US to Uk
1.87%
2.98%
3.15%
3.32%
3.40%
3.83%
4.59%
4.68%
5.78%
12.51%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00%
UAE to Brazil
China to Brazil
Brazil to South Africa
South Africa to Brazil
Spain to Brazil
China to Brazil
Brazil to Spain
India to Brazil
Brazil to China
Singapore to Brazil
1.42%
1.57%
2.34%
3.00%
3.05%
3.20%
4.16%
4.21%
11.88%
14.17%
0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 16.00%
India to Uruguay
Indonesia to Australia
South Africa to India
South Africa to China
China to Brazil
India to Brazil
India to South Africa
Singapore to Brazil
China to Australia
Australia to China
1.22%
1.68%
1.74%
2.58%
2.59%
3.05%
3.21%
3.47%
3.99%
5.72%
0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00%
US to UAE
Saudi Arabia to US
South Africa to India
Japan to UAE
China to Saudi Arabia
UAE to India
New Zealand to Australia
Australia to New Zealand
Saudi Arabia to Singapore
Saudi Arabia to China
a b
c d
e f
7
At the international level, the total high seas shipping-related CO2-e emission in 2019 exceeds the
150
total annual greenhouse gas emissions of countries such as Spain, Argentina and United Arab
151
Emirates (countries’ emission data are available at www.globalcarbonatlas.org). More alarmingly, the
152
average emission from shipping activities on the high seas is growing at approximately 7.26% per
153
year, which far surpasses the average total global emission growth rate of 0.09% per year and the
154
global shipping emission growth rate of 2.23% per year1. That is to say, international shipping GHG
155
emission on high seas could become an increasing barrier against the worldwide carbon mitigation
156
and sustainability efforts.
157
158
Figure 3 illustrates the changes in CO2-e emission output and carbon emission intensity across all
159
high seas regions from 2015 to 2019. In terms of total emissions, the results reflect, to some degree,
160
the intensity in shipping activities over the different high seas regions. Although the North Pacific
161
Ocean High Seas ranks third in total mileage travelled, it is responsible for more than 25% of the total
162
emissions generated. In fact, the North Pacific Ocean High Seas has the highest carbon emission
163
intensity at 0.1481 tonne/nm, which suggests that the shipping activities on the North Pacific Ocean
164
High Seas are more carbon intensive. The North Atlantic Ocean High Seas share is approximately
165
22%, while the Indian Ocean High Seas contributes about 18%. In terms of the average emission
166
growth rate, the top three fastest growing emitting regions are the Other High Seas (14.56%), the
167
North Atlantic Ocean High Seas (9.90%), and the South Pacific High Seas (9.11%). In particular, the
168
North Atlantic Ocean High Seas ranks in the top three regions for both total emissions generated and
169
emission growth rate, which indicates that the emission problem in one of the most heavily polluted
170
regions has become increasingly severe. Although the South Atlantic High Seas only contributes 13%
171
to the total emissions, the emission intensity there is among the highest, reaching 0.1450 tonne/nm
172
for CO2 emissions. Indeed, Figure 3 shows the different GHG emission patterns of each high seas
173
shipping activities in terms of total emission, emission growth rate and intensity. We provide more
174
in-depth investigation on emission drivers of high seas shipping in the next Section.
175
176
8
177
Fig. 3| High seas shipping emission outputs and operational efficiencies, a-h provide the monthly carbon
178
dioxide equivalent emissions and energy efficiency operational indicator for North Pacific Ocean High Seas
179
(a), South Pacific Ocean High Seas (b), North Atlantic Ocean High Seas (c), South Atlantic Ocean High Seas
180
(d), Arctic Ocean High Seas (e), Southern Ocean High Seas (f), Indian Ocean High Seas (g), and Other High
181
Seas (h). The carbon intensity indicator (CII) is defined by IMO as carbon dioxide emissions per actual cargo
182
mile. The annual emission and efficiency growth rate are shown on the x-axis, with blue and red arrows
183
respectively.
184
185
Emission drivers on High Seas. By incorporating detailed vessel and dynamic data of all ships from
186
all routes, we find that there exists a great degree of heterogeneity in key factors that drive the
187
shipping emission patterns across different high seas regions. As shown in Figure 4, the degree of
188
heterogeneity is mainly driven by the differences in key characteristics of ships active on each high
189
seas regions, namely, the ship type, capacity, age and engine usage. Since the emissions in the Arctic
190
0
0.05
0.1
0.15
0.2
0.25
0.3
0
1
2
3
4
5
6
2015
2016 2017 2018 2019
11.80%
-7.82%
8.83%
2.82%
2.19%
0.01%
-3.17%
-14.45%
0
0.05
0.1
0.15
0.2
0.25
0.3
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
2015
2016 2017 2018 2019
6.04%
-8.91%
18.89%
4.70%
4.59%
32.21%
3.47%
3.55%
0
0.05
0.1
0.15
0.2
0.25
0.3
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
2015
2016 2017 2018 2019
8.43%
-10.32%
12.27%
0.47%
8.50%
2.56%
9.51%
3.37%
0
0.05
0.1
0.15
0.2
0.25
0.3
0
0.5
1
1.5
2
2.5
3
3.5
2015
2016 2017 2018 2019
8.16%
5.44%
4.68%
0.34%
-4.71%
1.07%
5.45%
-4.38%
-0.1
0
0.1
0.2
0.3
0.4
0.5
-0.001
0.001
0.003
0.005
0.007
0.009
2015
2016 2017 2018 2019
-14.69%
-17.70%
105.1%
55.70%
30.42%
1.25%
19.88%
-14.22%
-0.1
0
0.1
0.2
0.3
0.4
0.5
-0.001
0.004
0.009
0.014
0.019
0.024
0.029
2015
2016 2017 2018 2019
30.60%
40.39%
114.9%
1.74%
118.3%
26.12%
-45.19%
-35.14%
0
0.05
0.1
0.15
0.2
0.25
0.3
1
1.5
2
2.5
3
3.5
4
2015
2016 2017 2018 2019
10.92%
-16.24%
9.39%
-1.84%
-1.29%
0.00%
4.49%
-3.21%
0
0.05
0.1
0.15
0.2
0.25
0.3
0.5
1
1.5
2
2.5
3
2015
2016 2017 2018 2019
15.31%
-18.82%
6.34%
1.00%
20.04%
-7.90%
7.51%
-4.81%
a b
c d
e f
g h
Monthly carbon dioxide equivalent emission (left axis, mill ion tonnes )
Carbon intensity indicator (right axis, tonne/mile)
9
and Southern Ocean High Seas are nearly negligible (account for 0.02% and 0.08% of total high seas
191
GHG emission in 2019 respectively), we disregard them in the emission driver analysis. As a result,
192
the results are shown for the six emission significant high seas regions.
193
194
In terms of ship type, although there are nineteen official types of vessels defined by the IMO, the
195
top six most dominant ship types were responsible for nearly 80% of the total emissions while the
196
other thirteen types combined together for the rest. Thus, for the ease of analysis, we classify each
197
vessel into one of seven major types based on its usage, namely bulk carrier, chemical tanker,
198
container, general cargo, liquefied gas tanker (LG Tanker), oil tanker, and others. According to the
199
Energy Efficiency Operational Indicator (EEOI) defined in the IMO study, general cargo, liquefied
200
gas tanker and container are the top three most carbon-intense ship types2,16. Figure 3(a) illustrates
201
the distribution of CO2 equivalent emissions across different ship types in each high seas region. We
202
find that the emission patterns of the North Pacific Ocean High Seas and North Atlantic Ocean High
203
Seas are alarmingly different from those of the other high seas regions. In particular, the top 3 carbon
204
intense ship types were responsible for 62% and 54% of the emission in the aforementioned two
205
regions, respectively, while they were only responsible for 19-25% in the rest of the regions. The high
206
proportion of emission from the carbon intense ships can be mainly attributed to the rapid growth in
207
emissions of containers and liquefied gas tankers in these regions. Specifically, in the North Pacific
208
Ocean High Seas, the emission contribution from liquefied gas tankers spiked from 2% to 13%
209
between 2015 and 2019, resulting in a 175% growth in its amount of emissions in the region.
210
Moreover, the annual emission growth rate for containers in the North Atlantic Ocean High Seas is
211
25%, which is much higher than the 5% average in the other regions. In contrast, we find that in
212
regions such as the South Atlantic Ocean High Seas and Indian Ocean High Seas, low carbon intense
213
ships such as bulk carriers and oil tankers are the dominant source of emissions.
214
215
In terms of ship capacity, we classify all the vessels into five dead-weight tonnage (dwt) groups as
216
defined by the IMO, namely, 0-25000 dwt, 25000-50000 dwt, 50000-75000 dwt, 75000-100000 dwt,
217
and 100000+ dwt. It is important to note that according to the EEOI defined in the IMO study and
218
10
previous studies, smaller ships (under 50000 dwt) typically exhibit a higher carbon intensity than
219
larger ships2,17. Figure 3(b) illustrates the distribution of CO2 equivalent emissions across different
220
ship capacities. Out of all the high seas regions, the emission pattern in the South Atlantic High Seas
221
stands out from the rest. In particular, the smaller, carbon intense ships under 50000 dwt are
222
responsible for over 21% of the total emissions in the South Atlantic High Seas. In fact, the relatively
223
small-sized shipping pattern is consistent with the short shipping routes and varied goods demand for
224
berthing ports in the South Atlantic High Seas. In contrast, the proportion of emissions attributed to
225
ships of the same weight classes is only 10% in regions such as the North Pacific Atlantic High Seas.
226
227
In terms of ship age, we classify all the vessels into five age classes 0-5 years, 5-10 years, 10-15
228
years, 15-20 years, and 20+ years. It is a fact that most new-build ships install engines with a better
229
EEDI and specific fuel consumption than ships with older construction year2. As illustrated in Figure
230
3(c), there are two distinct emission patterns distributions across all the high seas regions. On the one
231
hand, the North Pacific, South Atlantic, Indian and the Other High Seas have relatively newer ships
232
as the dominant emission source. In particular, ships with service age less than 10 years contribute
233
42% of emissions for the North Pacific Ocean High Seas, 48% South Atlantic High Seas, 43% for the
234
Indian High Seas, and 42% for the Other High Seas. Moreover, the oldest ships (20+ years) only
235
account for an average of 10.71% of emissions in these regions over the five-year period. On the
236
other hand, emission sources in the South Pacific Ocean High Seas primarily consist of older ships
237
with service age greater than 15 years. In particular, the ships with service age over 20 years contribute
238
26% of emissions for the South Pacific, which is much higher than the average contribution rate in
239
the other four high seas regions. In fact, in comparison to the top 3 carbon emission high seas regions,
240
shipping emission regulations on territory water and EEZ for countries along the shipping routes in
241
the South Pacific Ocean are relatively loose due to less significant emission output. From a policy
242
regulation perspective, shipping companies would allocate their older ships on the South Pacific
243
Ocean and new-built ships on other Oceans, which results in the ships with service age over 15 years
244
the major emission sources in the South Pacific High Seas.
245
11
246
Fig. 4| High seas shipping carbon emission drivers, the carbon dioxide equivalent emission compositions
247
and structural changes of each high seas region are identified and classified by ship type (column a), ship
248
capacity (column b), ship age (column c) and ship engine (column d). The dark blue bars indicate the annual
249
individual high seas shipping emissions of carbon dioxide equivalent (million tonnes). The bars in bright colors
250
(labeled in the legend) represent the annual emission contribution of the associated unit categories, which are
251
noted with their respective emission outputs.
252
Bulk Carrier
Chemical Tanker
Container
General Cargo
LG Tanker
Oil Tanker
0-25000
25000-50000
50000-75000
75000-100000
100000+
0-5
5-10
10-15
15-20
20+
Main Engine
Auxiliary Engine
Boiler
North Pacific Ocean
High Sea
South Pacific Ocean
High Sea
NorthAtlantic Ocean
High Sea
South Atlantic Ocean
High Sea
Arctic Ocean
High Sea
Southern Ocean
High Sea
Indian Ocean
High Sea
Other High Seas
a) Ship Type b) Ship Capacity c) Ship Age
Others
0.15
0.49
0.76
0.12
0.29
0.01
0.59
0.70
-0.18
3.78
0.00
2.88
0.06
0.22
-0.64
-0.10
-0.71
-0.21
1.65
-0.30
1.49
0.22
0.37
0.73
0.01
1.59
-0.24
-0.98
40
42
44
46
48
50
52
54
56
58
60
2015
2016 2017 2018 2019
0.23
0.01
0.13
0.01
0.02
0.12
0.13
0.04
0.01
0.78
0.00
0.02
0.15
-0.001
-0.17
0.01
0.28
0.01
0.07
0.04
0.17
-0.07
0.02
-0.03
-0.01
0.13
0.01
0.01
5
5.5
6
6.5
7
7.5
2015
2016 2017 2018 2019
-0.46
-2.27
0.20
0.53
0.59
1.96
1.27
0.96
0.34
1.94
0.17
0.58
0.37
0.43
0.00
-0.19
1.80
-0.05
0.49
-0.36
0.38
-0.76
0.31
0.13
0.08
1.94
1.01
0.43
30
32
34
36
38
40
42
44
46
48
50
2015
2016 2017 2018 2019
1.77
0.05
0.47
-0.03
0.03
0.40
-0.14
1.64
-0.03
-0.89
0.07
0.01
0.63
-0.03
-0.94
-0.09
-0.30
0.01
-0.30
0.01
0.12
0.01
0.10
0.02
0.02
0.06
0.67
-0.01
25
26
27
28
29
30
31
32
33
2015
2016 2017 2018 2019
0.0000
0.0000
0.0003
0.0009
0.0000
-0.0001
-0.0015
0.0019
0.0018
0.0045
0.0025
0.0000
0.0033
0.0000
0.0029
-0.0006
-0.0008
-0.0010
0.0000
-0.0012
0.0090
-0.0014
-0.0003
-0.0027
0.0006
0.0000
0.0037
0.0061
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
2015
2016 2017 2018 2019
-0.0002
0.0000
0.0003
-0.0006
0.0000
0.0000
0.0065
0.0188
0.0042
0.0089
0.0064
0.0000
0.0000
0.0414
0.0028
0.0008
-0.0033
-0.0010
0.0000
0.0000
0.0176
0.0127
0.0062
0.0021
-0.0002
0.0000
0.0000
0.0220
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
2015
2016 2017 2018 2019
1.59
0.12
1.05
0.01
0.27
0.95
-0.06
1.71
0.10
-0.20
-0.16
0.38
1.49
0.14
-0.74
-0.08
-0.09
0.22
0.09
0.20
-0.13
-0.05
-0.02
0.13
-0.02
-0.25
0.88
0.10
30
32
34
36
38
40
42
2015
2016 2017 2018 2019
0.80
0.21
0.80
0.12
0.46
0.76
0.15
-0.03
0.03
0.94
-0.01
0.12
0.20
0.03
1.06
0.19
1.31
0.06
0.81
0.58
0.28
0.40
0.18
-0.66
0.10
0.23
0.86
0.14
14
16
18
20
22
24
26
28
30
2015
2016 2017 2018 2019
1.11
0.12
1.29
0.10
-0.21
-0.16
0.31
2.77
3.24
1.19
-0.87
-0.17
-0.18
2.44
1.62
-0.15
-0.08
1.73
0.08
0.13
40
42
44
46
48
50
52
54
56
58
2015
2016 2017 2018 2019
0.26
0.14
0.18
0.12
-0.36
-0.27
0.07
0.17
0.86
0.14
-0.01
0.04
-0.55
0.32
0.60
0.00
-0.11
0.12
-0.06
0.12
5
5.5
6
6.5
7
7.5
2015
2016 2017 2018 2019
0.63
1.03
-0.17
1.50
-1.17
0.51
1.80
0.97
2.20
-0.67
0.08
-0.46
-2.00
3.70
0.76
-0.72
-0.27
1.77
1.37
0.99
30
32
34
36
38
40
42
44
46
48
50
2015
2016 2017 2018 2019
1.19
0.32
-0.15
0.59
0.60
-1.26
0.75
0.61
1.18
0.13
-0.05
-0.53
-1.36
0.74
-0.29
0.05
-0.29
0.62
0.09
0.39
25
26
27
28
29
30
31
32
33
2015
2016 2017 2018 2019
0.0000
-0.0006
-0.0015
0.0000
-0.0002
0.0024
0.0023
0.0017
0.0072
0.0005
-0.0017
0.0034
0.0048
-0.0028
0.0046
0.0012
-0.0025
0.0032
-0.0031
0.0073
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
2015
2016 2017 2018 2019
-0.002
-0.006
0.0004
0.0002
0.0064
0.0111
0.0129
0.0252
0.0090
0.0214
-0.0064
0.0059
-0.0007
-0.0056
0.0248
-0.0008
-0.0151
-0.0200
0.0045
0.0744
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
2015
2016 2017 2018 2019
1.46
0.94
0.09
1.18
0.44
-1.50
0.85
1.90
0.80
1.41
-0.06
-0.10
-0.48
-0.07
0.19
-0.02
-0.43
0.28
1.19
-0.25
30
32
34
36
38
40
42
2015
2016 2017 2018 2019
0.87
0.28
-0.03
0.75
1.13
-0.81
0.23
1.15
-0.07
0.78
0.03
0.67
0.42
2.34
0.85
-0.03
0.33
0.28
0.44
0.23
14
16
18
20
22
24
26
28
30
2015
2016 2017 2018 2019
-0.56
3.48
0.83
0.47
-0.88
1.67
3.77
1.39
0.84
0.84
2.13
0.75
-1.15
0.10
-0.13
0.68
-0.38
1.38
-1.18
-0.50
40
42
44
46
48
50
52
54
56
58
60
2015
2016 2017 2018 2019
-0.05
0.25
0.15
0.09
-0.09
0.16
0.52
0.05
0.12
0.14
0.23
-0.05
0.04
0.00
0.17
0.32
0.10
-0.19
-0.15
-0.01
5
5.5
6
6.5
7
7.5
2015
2016 2017 2018 2019
-0.80
0.50
0.73
0.65
0.72
0.95
3.03
0.53
1.56
-1.27
2.17
-0.39
0.38
0.39
-0.48
1.11
0.48
0.28
0.08
1.20
30
32
34
36
38
40
42
44
46
48
50
2015
2016 2017 2018 2019
0.11
2.24
0.89
0.24
-0.71
1.24
0.54
-0.19
-0.12
0.07
1.59
-1.00
-1.45
-0.31
-0.30
0.96
-0.73
-0.12
0.24
0.52
25
26
27
28
29
30
31
32
33
34
2015
2016 2017 2018 2019
0.0000
0.0001
0.0010
-0.063
0.0031
0.0023
0.0029
0.0056
0.0009
0.0023
-0.007
0.0012
-0.019
0.0054
0.0043
0.0033
0.0019
0.0037
-0.0041
0.0012
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
2015
2016 2017 2018 2019
0.0003
-0.0002
0.0020
0.0003
0.0038
0.0056
0.0177
0.0271
0.0107
0.0184
0.0045
-0.0042
0.0095
0.0015
0.0061
0.0068
-0.0149
-0.0322
-0.0511
0.0883
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
2015
2016 2017 2018 2019
0.11
2.62
0.44
0.18
-0.46
1.50
1.25
0.93
0.04
-0.26
1.95
-1.22
-1.06
0.15
-0.35
0.06
-0.25
-0.33
0.09
1.21
30
32
34
36
38
40
42
2015
2016 2017 2018 2019
0.00
1.97
0.84
0.22
-0.05
0.87
0.27
0.11
0.10
-0.08
1.69
0.66
0.79
0.78
0.39
0.43
-0.21
-0.08
0.12
1.00
14
16
18
20
22
24
26
28
30
2015
2016 2017 2018 2019
2.16
0.02
0.22
6.28
0.64
0.41
0.51
0.23
0.43
1.05
0.38
0.28
40
42
44
46
48
50
52
54
56
58
2015
2016 2017 2018 2019
0.62
0.19
0.78
0.72
0.13
0.13
0.23
-0.04
0.21
0.12
0.03
-0.08
4.5
5
5.5
6
6.5
7
7.5
2015
2016 2017 2018 2019
0.91
0.89
0.02
4.37
0.40
0.03
1.59
0.46
0.02
2.20
0.47
0.48
30
32
34
36
38
40
42
44
46
48
50
2015
2016 2017 2018 2019
2.42
0.12
0.02
1.41
0.03
0.02
1.35
0.07
0.20
0.61
0.02
0.27
18
20
22
24
26
28
30
32
34
2015
2016 2017 2018 2019
0.0005
0.0005
-0.0033
0.0148
-0.0008
0.0001
0.0010
0.0025
0.0048
0.0095
0.0008
-0.0042
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
2015
2016 2017 2018 2019
0.0054
0.0006
0.0000
0.0765
0.0022
0.0008
0.0166
0.0003
0.0000
0.0441
-0.0006
-0.0006
0.01
0.03
0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
2015
2016 2017 2018 2019
2.58
0.06
0.03
3.20
0.19
0.07
0.58
0.02
0.05
0.60
0.15
0.02
30
32
34
36
38
40
42
2015
2016 2017 2018 2019
2.71
0.24
0.05
1.15
0.10
0.02
3.27
0.55
0.47
1.00
0.19
0.06
14
16
18
20
22
24
26
28
30
2015
2016 2017 2018 2019
d) Ship Engine
12
253
In terms of ship engine usage, the majority of CO2 equivalent emissions are associated with main
254
engines18. However, the Other High Seas stand out from the rest of the regions as the growth rate in
255
main engine emission contribution reaches 10.2% annually, which is more than twice the average
256
growth rate in emission contribution in other regions. Moreover, the auxiliary engine emission
257
contribution in the Other High Seas is growing at 12.89% annually, far exceeding the 4.94% annual
258
growth rate in other regions. This reflects the fact that the cruise distances in the Other High Seas
259
region are much shorter than the rest of the high seas regions. Thus, the growth in emission share
260
attributable to auxiliary engines is much higher.
261
262
Shipping carbon mitigation policies evaluation. The Fourth IMO GHG Emission Study and
263
previous works evaluate territory water or EEZ shipping GHG emission mitigation policies
264
effectiveness with respect to the mid-term (2030) and long-term (2050) reduction targets2,19,20.
265
Following the current mid- and long-term policy settings on territory water or EEZ, we examine the
266
high seas GHG reduction effectiveness in both primary policy implementation stage and overall
267
implementation stage. We propose the primary carbon mitigation policies to be implemented on each
268
high seas region based on its main emission driver identified in our results (in Table 1). Specifically,
269
in the ship type (ST) policy, we target the top 3 carbon-intense ship types (general cargos, liquefied
270
gas tankers, and containers) by substituting the heavy fuel oil (2.43% Sulfur content) used in these
271
ships with alternative fuel such as marine diesel oil and marine gas oil (0.13% Sulfur content)21. Since
272
the carbon emission contribution and emission growth rate of the carbon-intense ships in the North
273
Pacific Ocean High Seas and North Atlantic Ocean High Seas far exceed those in other high seas
274
regions, we directly implement the ST policy in these two regions; for the ship capacity (SC) policy,
275
we intend to improve the average international shipping capacity by shifting the shipping activities
276
conducted through ships with relatively small capacities (<50000 dwt) to ships with larger
277
capacities22. Since the proportion of emissions attributed to small-capacity ships in the Indian Ocean
278
High Seas and South Atlantic High Seas are more than twice as much as that of other high seas regions,
279
we directly implement the SC policy in these two regions; for the ship age (SA) policy, we raise the
280
in-service ship standard by implementing compulsory scrapping of active ships with service age
281
13
greater than 20 years and substitute them with newly-built ships for high seas shipping activities23.
282
Since the emission contribution rate of ships with service age over 20 years in the South Pacific far
283
exceeds that in other high seas regions, we directly implement the SA policy in this region; for the
284
ship engine (SE) policy, we set speed reduction at 10% for all ship types on high seas shipping for
285
main engine carbon emission mitigation24. Since the emission growth rate of main engines used in
286
the Other High Seas is significantly higher than that in other regions, we directly implement the SE
287
policy in this region.
288
289
Tabl e 1| High seas shipping carbon mitigation policies and implementation strategies
Ship
categories
Intended carbon
mitigation policies
Measures
Primary high seas area
Ship type
High emission ship
type supervision
Alternative fuel adoption for
general cargo, liquefied gas
tankers and container
North Pacific Ocean
High Seas; North
Atlantic Ocean High
Seas
Ship capacity
Shipping capacity
intervention
Improve the average
international shipping
capacity
Indian Ocean High Seas;
South Atlantic Ocean
High Seas
Ship age
New-build ships
substitution
Raise the in-service ship
standard; Compulsory
scrapping policy for old ship
South Pacific Ocean
High Seas
Ship engine
Main engine
improvements
High sea shipping speed
reduction
Other High Seas
Note: Extensive shipping carbon mitigation policies are proposed by International Maritime
Organization and previous works. We collect the specific policies that target to foster maritime
industry decarbonization through ship type, capacity, age and engine categories. We define the primary
implementation policy with respect to the main shipping carbon emission driver of each high seas
derived from our results. As a result. we evaluate the high seas carbon mitigation policy effectiveness
in both primary implementation stage and overall implementation stage.
290
291
14
292
Fig. 5| High seas shipping carbon mitigation policy effectiveness evaluation, a-f provide the estimated
293
carbon mitigation effectiveness of North Pacific Ocean High Seas (a), South Pacific Ocean High Seas (b),
294
North Atlantic Ocean High Seas (c), South Atlantic Ocean High Seas (d), Indian Ocean High Seas (e) and
295
Other High Seas (f). The primary and overall stage indicate the primary and overall policy implementation
296
indicated in Table 1. The carbon reduction amount (million tonnes) and percentage of each high seas region
297
are presented above each bar respectively.
298
299
The policy effectiveness assessment is conducted through two stages. In the primary stage, each of
300
the carbon mitigation policy is implemented separately in the target high seas regions. After 2030, we
301
enter the overall stage, where the carbon mitigation policies are implemented together across all the
302
high seas regions. As illustrated in Figure 5, our evaluation results indicate that implementing tailored
303
carbon mitigation policies in different high seas regions could reduce 25.46 and 54.36 million tonnes
304
of CO2 equivalent emission in the primary intervention stage and the overall intervention stage,
305
respectively, with 12.09% and 25.81% reduction rates in comparison to the 2019 annual high seas
306
0
2
4
6
8
10
12
14
16
Primary stage Overall stage
North Pacific Ocean High Sea
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
Primary stage Overall stage
South Pacific Ocean High Sea
0
1
2
3
4
5
6
7
8
9
10
Primary stage Overall stage
North Atlantic Ocean High Sea
0
2
4
6
8
10
12
Primary stage Overall stage
Indian Ocean High Sea
0
1
2
3
4
5
6
7
8
Primary stage Overall stage
South Atlantic Ocean High Sea
0
1
2
3
4
5
6
Primary stage Overall stage
Other High Seas
6.79
(12.79%)
13.49
(25.49%)
0.46
(6.96%)
1.68
(24.94%)
5.33
(11.43%)
9.51
(20.41%)
3.06
(10.38%)
6.87
(23.28%)
5.31
(13.82%)
10.77
(28.01%)
1.46
(5.75%)
5.17
(20.34%)
a b c
de f
Ship type policy Ship capacity policy Ship age policy Ship engine policy
15
shipping GHG emission. In particular, it is worth noting that the regions with the greatest emission
307
reduction rate in the primary stage are the regions with the greatest high seas shipping emission
308
contribution rate overall, namely the Indian Ocean High Seas (13.82%), North Pacific Ocean High
309
Seas (12.79%), and North Atlantic Ocean High Seas (11.43%). Moreover, the tailored carbon
310
mitigation policy implemented through the primary stage shows the greatest emission reduction
311
percentage in each of the high seas regions (at average 46.84% of the total emission reduction), which
312
indicates that it is the most effective policy in reducing emissions in the particular region compared
313
to other policies. Thus, by identifying the key factors driving the emission patterns in different high
314
seas regions and accordingly designing tailored carbon mitigation policies for each region, it allows
315
international high seas shipping to contribute to the world trading and economic growth in a more
316
environmentally-friendly manner.
317
318
Discussion
319
Climate change is a global issue that requires international cooperation and solutions. To alleviate the
320
negative impact of climate change, the Paris Agreement was signed by world leaders in 2015 to foster
321
global greenhouse gas emission reduction across national borders25. Signatories are committed to
322
reduce GHG emission from all industries and human activities, and all the countries’ climate efforts
323
are monitored and reviewed by United Nations every 5 years26. In terms of maritime industry, strict
324
carbon emission regulations are set with respect to countries’ territory water or exclusive economic
325
zone shipping activities, such as emission controlled areas, alternative fuel substitution, electric or
326
nuclear propulsion adoption and renewable energy propulsion assistance etc27,28. For example,
327
emission controlled areas are proposed to limit SO2, NO2 and particulate matter emissions in world
328
major countries’ territory sea shipping29; Ships are required to use low Sulfur content fuel such as
329
marine diesel oil and marine gas oil to reduce carbon emissions when the shipping activities occur at
330
berth or at exclusive economic zone of US and East Asia area30,31. As a result, the Fourth IMO GHG
331
Emission Study suggests that the annual international shipping GHG emission growth gradually slow
332
down at a 2.23% average annual rate from 2015 to 2018 due to the associated maritime and
333
international shipping carbon mitigation policy interventions.
334
335
16
However, current shipping carbon mitigation policies are effective and implemented only in territory
336
seas and exclusive economic zone as part of certain country’s carbon reduction policies. Due to the
337
non-sovereign property of high seas regions, no shipping carbon mitigation policies or environmental
338
regulations are proposed or implemented for the world high seas regions, regarding that no signatories
339
are responsible for high seas carbon reduction under the Paris Agreement32,33. For a lower operational
340
cost of international shipping, ships usually navigate in an economically efficient manner on high
341
seas by utilizing heavy fuel oil and travelling with high speed without environmental concern. As a
342
result, our GEEM estimation results indicate that the carbon intensive shipping activities have
343
resulted in a significant increase in emissions on high seas regions. The total CO2-e emissions reach
344
211.60 Mmt/yr in 2019, which exceeds the total annual greenhouse gas emissions of countries such
345
as Spain, Argentina, and United Arab Emirates. In addition, the average emission from shipping
346
activities on the high seas is growing at approximately 7.26% per year, which far surpasses the global
347
shipping emission growth rate of 2.23% per year. In essence, without policy intervention,
348
international shipping GHG emission on high seas could become a tragedy of the commons in the
349
global maritime industry: individual ships behave on their own interests to maximize their high seas
350
shipping profits, ignoring the negative externality and climate change impact of their carbon intensive
351
shipping patterns on high seas.
352
353
Utilizing our GEEM bottom-up vessel dynamic statistics of all ships from all routes, we find
354
heterogeneity in key factors that drive the shipping emission pattern across different high seas regions
355
in terms of ship type, capacity, age and engine categories. In order to evaluate the effectiveness of
356
different carbon mitigation policies on high seas shipping, we collect the specific policies that target
357
to foster maritime industry decarbonization in territory seas and exclusive economic zones and
358
propose the primary implementation policy with respect to the main emission driver of each high seas
359
region identified from our results. Specifically speaking, we set and evaluate ship type policies on the
360
Other High Seas, ship capacity policy on the Indian High Seas and South Atlantic High Seas, ship
361
age policy on the South Pacific High Seas, ship engine policy on the North Pacific High Seas and
362
North Atlantic High Seas. The evaluation results indicate that carbon mitigation policies could reduce
363
17
25.46 and 54.36 million tonnes CO2-e in the primary intervention stage and overall intervention stage
364
respectively, with 12.09% and 25.81% reduction rates in comparison to the 2019 annual high seas
365
shipping GHG emission.
366
367
Although our evaluation results indicate that the carbon mitigation polices can effectively reduce
368
shipping GHG emissions on the high seas, implementing these policies may be difficult as no
369
signatories are directly responsible for these regions. To facilitate international cooperation and the
370
development of targeted regional high seas emission control agreement between countries, we also
371
identify the major emission-contributing shipping routes and the key signatories involved in each
372
high seas region as shown in Figure 2. Specifically, the trading activities between China, the United
373
States, Korea, Japan, and Canada generate contribute to nearly 50% of the emissions generated in the
374
North Pacific Ocean High Seas. In the South Ocean Pacific High Seas, the shipping routes between
375
Japan, Australia, and New Zealand are the most carbon emission intensive, generating 36% of the
376
emissions in the region. In the North Atlantic Ocean High Seas, the trading routes with the heaviest
377
traffic and carbon emissions involve countries such as the United States, the United Kingdom, Brazil,
378
Spain, Canada, the Netherlands, and Germany. In the South Atlantic Ocean High Seas, the top 3 most
379
carbon-intensive trading route is from Singapore to Brazil, Brazil to China, and India to Brazil,
380
contributing 13%, 6%, and 5%, respectively. In the Indian Ocean High Seas, the shipping routes
381
between China and Australia are especially carbon intensive, contributing to over 35% of the
382
emissions in the region. In the Other High Seas Region, some of the most carbon intensive trading
383
routes involve countries such as China, Saudi Arabia, Singapore, Australia, New Zealand, United
384
Arab Emirates, and South Africa. By identifying these emission-intensive trading routes, it can
385
promote the key countries involved to form regional high seas emission control agreement and impose
386
the carbon mitigation policies discussed above on the vessels operating along these trading routes.
387
After all, by regulating high seas shipping activity via the international cooperation and global
388
maritime industry effort, international high seas shipping can contribute to the world trading and
389
economic growth in a more environmentally-friendly manner in the future.
390
391
18
References
392
[1] Cisneros-Montemayor, A. M., Moreno-Báez, M., & Reygondeau, G., et al. Enabling conditions
393
for an equitable and sustainable blue economy. Nature. 591(7850), 396-401(2021).
394
[2] Fourth IMO Greenhouse Gas Study 2020 (International Maritime Organization, 2020).
395
[3] Bednar, J., Obersteiner, M., & Baklanov, A., et al. Operationalizing the net-negative carbon
396
economy. Nature. 596(7872), 377-383 (2021).
397
[4] Luderer, G., Madeddu, S., & Merfort, L., et al. Impact of declining renewable energy costs on
398
electrification in low-emission scenarios. Nat. Energy. 1-11 (2021).
399
[5] Sofiev, M., Winebrake, J. J., & Johansson, L., et al. Cleaner fuels for ships provide public health
400
benefits with climate tradeoffs. Nat. Commun. 9(1), 1-12 (2018).
401
[6] Meng, Q., Wang, S., Andersson, H., & Thun, K. Containership routing and scheduling in liner
402
shipping: overview and future research directions. Transp. Sci. 48(2), 265-280 (2014).
403
[7] Zis, T. P., & Cullinane, K. The desulphurisation of shipping: Past, present and the future under a
404
global cap. Transp. Res. D: Transp. Environ. 82, 102316 (2020).
405
[8] Gan, Y., El-Houjeiri, H. M., Badahdah, A., Lu, Z., Cai, H., Przesmitzki, S., & Wang, M. Carbon
406
footprint of global natural gas supplies to China. Nat. Commun. 11(1), 1-9 (2020).
407
[9] Liu, J., Law, A. W. K., & Duru, O. Reducing emissions of atmospheric pollutants along major
408
dry bulk and tanker routes through autonomous shipping. J. Environ. Manag. 302, 114080 (2022).
409
[10] López-Aparicio, S., Tønnesen, D., Thanh, T. N., & Neilson, H. Shipping emissions in a Nordic
410
port: Assessment of mitigation strategies. Transp. Res. D: Transp. Environ. 53, 205-216 (2017).
411
[11] Nunes, R. A., Alvim-Ferraz, M. C. M., Martins, F. G., & Sousa, S. I. Environmental and social
412
valuation of shipping emissions on four ports of Portugal. J. Environ. Manag. 235, 62-69 (2019).
413
[12] Mujeeb-Ahmed, M. P., Seo, J. K., & Paik, J. K. Probabilistic approach for collision risk analysis
414
of powered vessel with offshore platforms. Ocean Eng. 151, 206-221 (2018).
415
[13] Review of Maritime Transport 2019. (United Nations Conference on Trade and Development,
416
2019)
417
[14] Wang, X. T., Liu, H., & Lv, Z. F., et al. Trade-linked shipping CO2 emissions. Nat. Clim. Chang.
418
11(11), 945-951 (2021).
419
[15] Liu, H., Meng, Z. H., & Lv, Z. F., et al. Emissions and health impacts from global shipping
420
embodied in US–China bilateral trade. Nat. Sustain. 2(11), 1027-1033 (2019).
421
[16] Perera, L. P., & Mo, B. Emission control based energy efficiency measures in ship operations.
422
Appl. Ocean Res. 60, 29-46 (2016).
423
[17] Walsh, C., & Bows, A. Size matters: exploring the importance of vessel characteristics to inform
424
estimates of shipping emissions. Appl. Energy 98, 128-137 (2012).
425
[18] Faber, J., Nelissen, D., Hon, G., Wang, H., & Tsimplis, M. (2012). Regulated slow steaming in
426
maritime transport. An assessment of options, costs and benefits.
427
[19] Qi, X., & Song, D. P. Minimizing fuel emissions by optimizing vessel schedules in liner shipping
428
with uncertain port times. Transport. Res. E Log. 48(4), 863-880 (2012).
429
[20] Bugnot, A. B., Mayer-Pinto, M., & Airoldi, L., et al. Current and projected global extent of
430
marine built structures. Nat. Sustain. 4(1), 33-41 (2021).
431
19
[21] Yang, Z. L., Zhang, D., & Caglayan, O., et al. Selection of techniques for reducing shipping NOx
432
and SOx emissions. Transp. Res. D: Transp. Environ. 17(6), 478-486 (2012).
433
[22] Ammar, N. R., & Seddiek, I. S. Enhancing energy efficiency for new generations of containerized
434
shipping. Ocean Eng. 215, 107887 (2020).
435
[23] Ling-Chin, J., & Roskilly, A. P. Investigating the implications of a new-build hybrid power
436
system for Roll-on/Roll-off cargo ships from a sustainability perspective–A life cycle assessment
437
case study. Appl. Energy 181, 416-434 (2016).
438
[24] Psaraftis, H. N., & Kontovas, C. A. Speed models for energy-efficient maritime transportation:
439
A taxonomy and survey. Transp. Res. Part C Emerg. Technol. 26, 331-351 (2013).
440
[25] Fagerholt, K., Laporte, G., & Norstad, I. Reducing fuel emissions by optimizing speed on
441
shipping routes. J. Oper. Res. Soc. 61(3), 523-529 (2010).
442
[26] DeConto, R. M., Pollard, D., & Alley, R. B., et al. The Paris Climate Agreement and future sea-
443
level rise from Antarctica. Nature. 593(7857), 83-89 (2021).
444
[27] Rogelj, J., Huppmann, D., & Krey, V., et al. A new scenario logic for the Paris Agreement long-
445
term temperature goal. Nature. 573(7774), 357-363 (2019).
446
[28] Sala, E., Mayorga, J., & Bradley, D., et al. Protecting the global ocean for biodiversity, food and
447
climate. Nature. 592(7854), 397-402 (2021).
448
[29] Psaraftis, H. N., & Kontovas, C. A. Balancing the economic and environmental performance of
449
maritime transportation. Transp. Res. D: Transp. Environ. 15(8), 458-462 (2010).
450
[30] Wang, C., & Corbett, J. J. The costs and benefits of reducing SO2 emissions from ships in the
451
US West Coastal waters. Transp. Res. D: Transp. Environ. 12(8), 577-588 (2007).
452
[31] Xiao, Q., Li, M., & Liu, H., et al. Characteristics of marine shipping emissions at berth: profiles
453
for particulate matter and volatile organic compounds. Atmospheric Chem. Phys. 18(13), 9527-
454
9545 (2018).
455
[32] Sala, E., Mayorga, J., & Costello, C., et al. The economics of fishing the high seas. Sci. Adv. 4(6),
456
eaat2504 (2018).
457
[33] Ardron, J., Gjerde, K., Pullen, S., & Tilot, V. Marine spatial planning in the high seas. Mar. Policy.
458
32(5), 832-839 (2008).
459
460
Acknowledgement
461
This work was supported by grants from the National Natural Science Foundation of China
462
(71988101, 72174035, 71774018 and 71831002), and LiaoNing Revitalization Talents Program
463
(XLYC2008030).
464
465
Competing Interests
466
The authors declare no competing interests.
467
468
20
Methods
469
470
GEEM static database. In this study, a new Geographic-based Emission Estimation Model (GEEM)
471
static database is created using the vessel technical specification dataset provided by the IHS Market
472
Maritime & Trade and AIS static database. The IHS & AIS static database contains ships
473
characteristics for ships as of 2020. The ships range from 100 GT fishing ferries and service vessels
474
to the largest bulk carriers and cargo ships, covering both ships that engage in international as well as
475
domestic navigation. In this study, the combined IHS & AIS static database contains all the data
476
collected and updated to 2020. Thus, we checked each vessel’s status against a timestamp of the most
477
recent change in status separately to ensure that only “in service” vessels are included in our GEEM
478
static database.
479
480
The GEEM static database provides detailed ship characteristics including the IMO number, vessel
481
type, build year, length, width, height, capacity, design speed, fuel type, installed engine power,
482
engine RPM, maximum draught, dead weight tonnage (dwt) etc. This wide range of metrics is
483
essential for estimating fuel consumption and emission from ships. However, the data is sometimes
484
incomplete so that one or more technical information was found to be missing for some ships in the
485
IHS dataset. In our data, 0.06% of the ships are missing capacity, 2.4% of the ships are missing build
486
year, 3.0% of the ships are missing fuel type, 20.7% of the ships are missing design speed, and 26.4%
487
of the ships are missing engine RPM. Simply excluding those particular ships with missing technical
488
information from our calculation or assigning default values to the missing property will lead to
489
significant computation inaccuracies. To correct the data and address the uncertainty, we designed a
490
robust method to infill these missing technical specifications. Following the guidelines recommended
491
by the Fourth IMO GHG Study34, we create a multilinear regression for each ship type by taking into
492
account individual vessel’s known design parameters such as beam, draught and capacity. Since both
493
beam and draught serve as the basis in the estimation of other metrics, the missing values for these
494
metrics are first filled based on median values per type and size category. After this essential
495
information is infilled, we apply individual regressions on each of the other metrics. Finally, for
496
21
individual ships that could not be infilled due to too many missing entries, we replace the missing
497
information with the median values of their respective type and size class.
498
499
In this analysis, the technical specification data were collected and pre-processed for 79613 vessels,
500
which represents a majority of the international shipping fleet. The fleet scale is reasonable compared
501
with the previous literature35,36. The original vessels collected are categorized into 19 vessel type
502
categories according to the IMO ship types classified in the Fourth IMO GHG Study. In addition, the
503
vessels collected are also classified into 4 age groups based on their build years. Supplementary Table
504
1 lists the classified vessel types, and the number of vessels counted in each category. Supplementary
505
Table 2 lists the classified engine tier groups, vessel capacity group and the number of vessels counted
506
in each group.
507
Supplementary Table 1| GEEM static dataset vessel types
508
Ship type
This study
Bulk Carrier
11981
Oil Tanker
7403
Container
5366
General Cargo
13196
Chemical Tanker
5145
Ro-Ro
1162
Cruise
475
Refrigerated Bulk
684
Liquefied Gas Tanker
2011
Other Liquids Tankers
34
Ferry - Pax Only
1272
Ferry - Ro-Pax
2120
Veh i c l e
842
Service - Tug
5172
Service - Other
4799
Miscellaneous Fishing
8942
Miscellaneous - Other
1727
Offshore
5068
Yac h t
2214
509
Supplementary Table 2| Sample vessel classification and statistics
510
22
Vessel capacity category
Vessel type
capacity
count
Bulk Carrier
0-9999
524
10000-34999
1603
35000-59999
2912
60000-99999
3305
100000-199999
1049
200000-+
538
Chemical Tanker
0-4999
1274
5000-9999
777
10000-19999
1595
20000-39999
1058
40000-+
2066
Container
0-999
72
1000-1999
75
2000-2999
82
3000-4999
197
5000-7999
205
8000-11999
209
12000-14499
49
14500-19999
213
20000-+
1074
General cargo
0-4999
2777
5000-9999
838
10000-19999
4058
20000-+
1359
Liquefied gas tanker
0-49999
4816
50000-99999
2964
100000-199999
132
200000-+
0
Oil tanker
0-4999
1394
5000-9999
70
10000-19999
50
20000-59999
364
60000-79999
684
80000-119999
2761
120000-199999
1621
200000-+
2337
Others
0-+
34511
23
511
512
513
514
515
Dynamic Ship Movement Database. One of the advantages of this study is the superior quality
516
Automatic Identification System (AIS) data in the high seas regions across the globe. In 2002, the
517
AIS was introduced by the IMO International Convention for the Safety of Life at Sea (SOLAS) to
518
improve maritime safety. Acting as a dynamic tracking and monitoring system, the AIS provides
519
broad coverage and delivers detailed real-time information on the ship, including a ship’s identity,
520
position, speed, and draught at a given timestamp. Following the mandate set by IMO SOLAS, all
521
ships over 300 GT engaged in international voyages, cargo ships over 500 GT engaged in national
522
voyages and all passenger ships are required to install AIS transceiver. According to the most recent
523
study37,38,39, the number of ships equipped with AIS and the number of AIS messages transmitted per
524
year has experienced a significant growth over the years, which suggests that the introduction of the
525
automatic vessel position reporting system has significantly reduced the uncertainty concerning ship
526
activities and their geographical distribution.
527
528
Supplementary Table 3| AIS message broadcast frequency
529
Transponder Type
Vessel's Moving Status (Transponder ON)
Transmission Rate
Class A
Anchored / Moored
Every 3 Minutes
Class A
Sailing 0-14 knots
Every 10 Seconds
Class A
Sailing 14-23 knots
Every 6 Seconds
Class A
Sailing 0-14 knots and changing course
Every 3.33 Seconds
Class A
Sailing 14-23 knots and changing course
Every 2 Seconds
Class A
Sailing faster than 23 knots
Every 2 Seconds
Class A
Sailing faster than 23 knots and changing course
Every 2 Seconds
Class B
Stopped or sailing up to 2 knots
Every 3 Minutes
Class B
Sailing faster than 2 knots
Every 30 Seconds
530
Engine tier category
Tier 0 (engine construction date before 2000)
29058
Tier 1 (engine construction date between 2000-2010)
21614
Tier 2 (engine construction date between 2011-2015)
15695
Tier 3(engine construction date after 2016)
13246
24
This study utilizes AIS dataset to construct our GEEM dynamic database. That is to say, our GEEM
531
dynamic database includes metrics that are essential for the analysis of vessel movement and activity,
532
such as the IMO identification number, Maritime Mobile Service Identify (MMSI) code, vessel
533
coordinate (longitude and latitude), vessel actual speed, voyage draught, and time information. As
534
shown in Supplementary Table 3, the AIS transmission rate is consistent with vessel’s moving status
535
and transponder type. All the AIS data are transmitted with a broadcast frequency of one message no
536
more than 3 minutes. As a result, the study utilizes the full year 3-minute frequency AIS data spanning
537
from January 2015 to December 2019, which accounts for a total of 5.03 TB raw AIS dataset.
538
539
Research domain identification. As suggested by the United Nations Convention on the Law of the
540
Sea, high seas can be defined as all parts of the sea that are not included in the exclusive economic
541
zone (EEZ). In this study, we first identify the worldwide high seas by subtracting the world country’s
542
exclusive economic zones from the world sea boundary. In addition, following the high seas division
543
standard provided by International Hydrographic Organization, we divide the worldwide high seas
544
into 8 geographic regions and define each high seas region’s longitude and latitude based on
545
Geographic information system (GIS) database (namely the North Pacific Ocean High Seas, South
546
Pacific Ocean High Seas, North Atlantic Ocean High Seas, South Atlantic Ocean High Seas, Arctic
547
Ocean High Seas, Southern Ocean High Seas, Indian Ocean High Seas, and Other High Seas).
548
Supplementary Fig. 1 presents the research domain of this study and the geographic boundary of each
549
high seas region. GHG emission of vessels are collected and aggerated to respective high seas region
550
by 0.05° grid box based on their AIS messages. By dividing the worldwide high seas into 8 geographic
551
regions, we are able to investigate the different GHG emission pattern and emission driver of each
552
high seas, and the policies can be designed and evaluated in a more specific and effective manner.
553
554
25
555
Supplementary Fig. 1| High seas boundary and classification. According to the United Nations Convention
556
on the Law of the Sea and the International Hydrographic Organization division standard and, by subtracting
557
the world country’s Exclusive Economic Zones from the world sea boundary, this study divides the worldwide
558
high seas into 8 geographic regions, namely North Pacific Ocean High Seas (a), South Pacific Ocean High
559
Seas (b), North Atlantic Ocean High Seas (c), South Atlantic Ocean High Seas (d), Arctic Ocean High Seas
560
(e), Southern Ocean High Seas (f), Indian Ocean High Seas (g), and Other High Seas (h).
561
562
Previous studies use bottom-up vessel-based or voyage-based methods to calculate shipping GHG
563
emission for certain regions or countries34,40,41. In essence, these methods heavily rely on their
564
assumptions: the vessel-based method assumes that vessels with similar type and age have the
565
uniform shipping behaviors; the voyage-based method distinguishes international and domestic
566
shipping emissions as those which occurred on a voyage between two ports in different or same
567
countries.
568
569
In this paper, we propose a new geographic-based method for estimating the high seas shipping GHG
570
inventory. As illustrated in Supplementary Fig. 2, no matter the vessel type, age or the shipping
571
destination, AIS messages are collected only when shipping occurred on high seas geographically.
572
Based on the bottom-up approach, this paper uses the spatial join method in GIS spatial superposition
573
analysis to identify ship trajectory points located in different high seas regions. We then obtain high
574
seas emissions by accumulating ship trajectory emissions in each high seas region. In comparison to
575
the existing two bottom-up GHG emission approaches, the geographic-based method enables more
576
accurate and robust high seas emission estimation due to the following aspects: international vessels
577
navigate between two countries’ exclusive economic zone are excluded from high seas GHG
578
inventory (such as Port A Country A to Port A Country B in Supplementary Fig. 2); domestic shipping
579
a
b
c
d
e
f
g
h
26
would be accounted as high seas GHG emission if the shipping route covers high seas region (such
580
as Port A Country A to Port C Country A in Supplementary Fig. 2).
581
582
583
Supplementary Fig. 2| Geographic-based high seas shipping allocation. In comparison to the existing
584
bottom-up shipping estimation approaches such as vessel-based and voyage-based shipping calculation
585
method, we collect AIS messages for both international and domestic shipping routes and all kinds of vessel
586
types if and only if the AIS messages are reported from high seas regions (yellow bar in Supplementary Fig.
587
2). As a result, exclusive economic zones AIS messages and shipping emission (grey bar in Supplementary
588
Fig. 2) are excluded in this study.
589
590
Geographic-based Emission Estimation Model. The technical strategy of our proposed
591
Geographic-based Emission Estimation Model (GEEM) is illustrated in Supplementary Fig. 3. As
592
discussed above, we utilize the IHS raw database and vessel AIS messages to construct our GEEM
593
static and dynamic database. We design a geographic-based AIS messages collection method to
594
classify and categorize high seas shipping route and associated GHG emission. In this section, we
595
demonstrate the detailed GEEM high seas emission calculation method, the specifications of emission
596
equation settings and the updated emission factors used in this study.
597
598
Country
A
Country
B
Port A
Port B
Port C
Port A
Port B
Territorial water AIS messages
High sea
High sea AIS messages
High sea AIS messages
27
599
Supplementary Fig. 3| Technical strategy for Geographic-based Emission Estimation Model (GEEM)
600
601
In international shipping, the GHG emission of each vessel is produced by three types of vessel
602
engines, namely the main engine (propulsion engine), the auxiliary engine and the boiler. Since the
603
main engine and the auxiliary engine are the moving power source of shipping, the emission
604
intensities of the main and auxiliary engine are determined by a variety of shipping characteristics
605
such as the vessel movement modes, instantaneous load factors, and maximum continuous rated
606
power etc. Boiler is used for hot water production and fuel heating. Its emission intensity is mainly
607
associated with vessel fuel types. We next show how our GEEM approach calculate and collect the
608
GHG emission of the above three engines and formulate the total emission of high seas from a bottom-
609
up approach.
610
611
Data
collection and
database
construction
Technical Strategy for Geographic-based Emission
Estimation Model
IHS&AIS raw database AIS messages
lNavigational status
lSpeed over ground
lPosition coordinates…
GEEM static data GEEM dynamic data
lInstantaneous speed
lVessel activity modes
lLoad factor…
High sea emission allocation and
classification
Bottom-up estimation Updated emission factor
Policy design and evaluation
lIMO number
lCall sign
lDraught…
lVessel types
lEngine tiers
lFuel types…
lGeographic-based AIS message
collection
lHigh seas region classification
Research
domain
identification
lMain engine emission
lAuxiliary engine emission
lBoiler emission…
lSpeed factor
lFuel type factor
lAdjusted load factor…
Emission
estimation
lEnergy-saving techniques
lrenewable energy adoption
lSpeed reduction…
Policy
analysis
28
The GHG emission of each vessel
!
can be calculated as the sum of vessel main engine, auxiliary
612
engine and boiler emission:
613
" # "$%&' ("%)*&+&%,- ("./&+0,11111111111111111111111111111111111111111111111111111234
614
615
In this study, we formulate the total GHG emission
5"
of each high seas region by summing up the
616
individual vessel’s emission occurring in high seas region geographically:
617
5"&#
6
"&7
'
&89 :11111111111111111111111111111111111111111111111111111111111111111111111111112;4
618
In terms of the vessel main engine, the GHG emission of main engine can be expressed as follows:
619
"$%&' #<=> ?"@$%&' ?
6
A@7?B@7?C57
$
789 :111111111111111111111111111111111111111112D4
620
where MCR is the maximum continuous rated power;
"@$%&'
represents the emission factor of
621
vessel main engine;
A@7
is the instantaneous load factor at time j,
B@7
is the emission adjustment
622
factors when the vessel’s instantaneous load factor is lower than 20%.
C57
is the time span of the
623
two adjacent AIS messages. It is worth noting that the emission factor for vessel main engine and
624
auxiliary engine provided by the previous studies lack the High Speed Diesel (HSD) and Slow Speed
625
Diesel (SSD) emission factors, respectively. In this study, we update the overall vessel main and
626
auxiliary engine emission factors based on IMO Fourth GHG Emission Study. In addition, we also
627
calculate the Tier 3 (vessel construction date after 2016) vessel’s GHG emission intensities for both
628
the main and auxiliary engine. The emission factors for main engine are reported in Supplementary
629
Table 4.
630
631
Supplementary Table 4| Emission factors for vessel main engine (g/kwh)
632
Engine
Typ e
Fuel
Typ e
Tier
Model
Yea r
PM2.5
PM10
NOx
SO2
CO
NM
VOC
CO2
N2O
CH4
SSD
HFO
(2.43%
Sulfur)
Tier
0
1.28
1.39
18.1
8.751
0.54
0.632
576
0.031
0.012
MSD
1.28
1.39
14.0
9.224
0.54
0.527
607
0.034
0.01
HSD
1.29
1.4
10.0
9.697
0.54
0.527
638
0.030
0.01
SSD
Tier
1
2000-
2010
1.28
1.39
17.0
8.278
0.54
0.632
545
0.031
0.012
MSD
1.28
1.39
13.0
8.751
0.54
0.527
576
0.034
0.01
HSD
1.28
1.39
9.8
9.224
0.54
0.527
607
0.030
0.01
SSD
1.28
1.39
14.4
8.593
0.54
0.632
545
0.031
0.012
1999£
29
MSD
Tier
2
2011-
2015
1.28
1.39
10.5
9.084
0.54
0.527
576
0.034
0.01
HSD
1.28
1.39
7.7
9.575
0.54
0.527
607
0.030
0.01
SSD
Tier
3
1.28
1.39
14.4
8.890
0.54
0.632
545
0.031
0.01
MSD
1.28
1.39
10.5
9.398
0.54
0.527
576
0.034
0.01
HSD
1.28
1.39
7.7
9.906
0.54
0.527
607
0.030
0.01
SSD
MDO/
MGO
(0.13%
Sulfur)
Tier
0
0.17
0.18
18.1
0.508
0.044
0.632
561
0.030
0.012
MSD
0.17
0.18
14.0
0.537
0.046
0.527
593
0.030
0.01
HSD
0.17
0.18
10.0
0.551
0.54
0.527
609
0.034
0.01
SSD
Tier
1
2000-
2010
0.17
0.19
17.0
0.479
0.044
0.632
529
0.030
0.012
MSD
0.17
0.18
13.0
0.508
0.046
0.527
561
0.030
0.01
HSD
0.17
0.18
9.8
0.537
0.54
0.527
593
0.034
0.01
SSD
Tier
2
2011-
2015
0.17
0.19
14.4
0.446
0.044
0.632
529
0.030
0.012
MSD
0.17
0.18
10.5
0.473
0.046
0.527
561
0.030
0.01
HSD
0.17
0.18
7.7
0.500
0.54
0.527
593
0.034
0.01
SSD
Tier
3
0.17
0.19
14.4
0.231
0.044
0.632
529
0.030
0.01
MSD
0.17
0.18
10.5
0.245
0.046
0.527
561
0.030
0.01
HSD
0.17
0.18
7.7
0.259
0.54
0.527
593
0.034
0.01
Otto
LNG
Tier
0-2
<2016
0.03
0.03
1.3
0.003
1.3
0.5
457
0.018
8.5
Otto (MS)
LNG
Tier
3
0.02
0.02
1.3
0.003
1.3
0.500
457
0.02
5.5
Otto (SS)
0.02
0.02
1.3
0.003
1.3
0.500
457
0.02
2.5
Diesel
0.01
0.01
3.4
0.003
1.04
0.400
457
0.03
0.2
Note: 1) SSD: Slow Speed Diesel; MSD: Medium Speed Diesel; HSD: High Speed Diesel; HFO: Heavy Fuel Oil;
MDO: Marine Diesel Oil; MGO: Marine Gas Oil; LNG: Liquefied Natural Gas; Otto: Otto-cycle LNG-fueled engine.
2) According to the Fourth IMO GHG Study, we set Heavy Fuel Oil (HFO) average sulfur content at 2.43%, Marine
Diesel Oil (MDO) and Marine Gas Oil (MGO) at 0.13%. In addition, we collect and update all types of emission factors
(PM2.5, PM10, NOx, SO2, CO, NMVOC, CO2, N2O, CH4) reported in Fourth IMO GHG Study.
633
The instantaneous load factor
A@7
in Equation (4) at time j can be formulated as follows:
634
A@7#
E
F7
<GH
I
J:111111111111111111111111111111111111111111111111111111111111111111111111112K4
635
where
F7
is the vessel’s instantaneous speed at time j and MDS is the vessel’s maximum designed
636
speed. The base emission factors for vessel main engine would decrease by about 20 percent load. As
637
a result, according to the adjusted emission factor statistics of Energy and Environmental Analysis,
638
Supplementary Table 5 shows the adjusted emission factors
B@7
for vessel main engines at low loads.
639
640
2016³
1999£
2016³
2016³
30
Supplementary Table 5| Adjusted emission factors for vessel main engines at low loads (%)
641
Load Factor
(%)
PM
NOx
SO2
CO
NMVOC
CO2
N2O
CH4
2
7.29
4.63
1
9.7
21.18
1
4.63
21.18
3
4.33
2.92
1
6.49
11.68
1
2.92
11.68
4
3.09
2.21
1
4.86
7.71
1
2.21
7.71
5
2.44
1.83
1
3.9
5.61
1
1.83
5.61
6
2.04
1.6
1
3.26
4.35
1
1.6
4.35
7
1.79
1.45
1
2.8
3.52
1
1.45
3.52
8
1.61
1.35
1
2.45
2.95
1
1.35
2.95
9
1.48
1.27
1
2.18
2.52
1
1.27
2.52
10
1.38
1.22
1
1.97
2.18
1
1.22
2.18
11
1.3
1.17
1
1.79
1.96
1
1.17
1.96
12
1.24
1.14
1
1.64
1.76
1
1.14
1.76
13
1.19
1.11
1
1.52
1.6
1
1.11
1.6
14
1.15
1.08
1
1.41
1.47
1
1.08
1.47
15
1.11
1.06
1
1.32
1.36
1
1.06
1.36
16
1.08
1.05
1
1.24
1.26
1
1.05
1.26
17
1.06
1.03
1
1.17
1.18
1
1.03
1.18
18
1.04
1.02
1
1.11
1.11
1
1.02
1.11
19
1.02
1.01
1
1.05
1.05
1
1.01
1.05
20
1
1
1
1
1
1
1
1
642
In terms of vessel auxiliary engine, the GHG emission of auxiliary engine can be expressed as follows:
643
"%)*&+&%,- #"@%)*&+&%,- ?
6
L%)*7? C57
$
789 :111111111111111111111111111111111111111111112M4
644
Where
"@%)*&+&%,-
is the emission factor of vessel auxiliary engine at certain fuel types,
LNOPQ7
is
645
the auxiliary engine power output at time j, and the
C57
is the time span of the two adjacent AIS
646
messages. Emission factors for auxiliary engine are reported in Supplementary Table 6. In fact, the
647
vessel auxiliary engine
LNOPQ7
and boiler have different power outputs when the vessel movement
648
mode changes. Vessel movement modes can be categorized into 4 types by their speed and maximum
649
continuous rated power (MCR), namely At berth (speed less than 1 knot), At anchorage (speed
650
between 1 knot-3 knot), Maneuvering (speed great than 3 knots and less than 20% MCR) and At sea
651
(speed above 20% MCR). As a result, Supplementary Table 7 provides adjusted auxiliary engine and
652
boiler power outputs for different vessel movement modes.
653
654
Supplementary Table 6| Emission factors for auxiliary engine (g/kwh)
655
31
Engine
Typ e
Fuel
Type
Tier
Model
Yea r
NOx
CO
NM
VOC
CO2
N2O
CH4
SSD
HFO
(2.43%
Sulfur)
Tier 0
1.29
1.4
11.2
9.697
0.54
0.421
638
0.040
0.01
MSD
1.29
1.4
11.2
9.697
0.54
0.421
638
0.040
0.01
HSD
1.29
1.4
11.2
9.697
0.54
0.421
638
0.040
0.01
SSD
Tier 1
2000-2010
1.28
1.39
11.2
9.224
0.54
0.421
607
0.040
0.01
MSD
1.28
1.39
11.2
9.224
0.54
0.421
607
0.040
0.01
HSD
1.28
1.39
11.2
9.224
0.54
0.421
607
0.040
0.01
SSD
Tier 2
2011-2015
1.28
1.39
11.2
9.575
0.54
0.421
607
0.040
0.01
MSD
1.28
1.39
11.2
9.575
0.54
0.421
607
0.040
0.01
HSD
1.28
1.39
11.2
9.575
0.54
0.421
607
0.040
0.01
SSD
Tier 3
1.28
1.39
11.2
9.906
0.54
0.421
607
0.040
0.01
MSD
1.28
1.39
11.2
9.906
0.54
0.421
607
0.040
0.01
HSD
1.28
1.39
11.2
9.906
0.54
0.421
607
0.040
0.01
SSD
MDO/
MGO
(0.13%
Sulfur)
Tier 0
0.17
0.18
11.2
0.551
0.54
0.421
609
0.036
0.01
MSD
0.17
0.18
11.2
0.551
0.54
0.421
609
0.036
0.01
HSD
0.17
0.18
11.2
0.551
0.54
0.421
609
0.036
0.01
SSD
Tier 1
2000-2010
0.17
0.18
11.2
0.537
0.54
0.421
593
0.036
0.01
MSD
0.17
0.18
11.2
0.537
0.54
0.421
593
0.036
0.01
HSD
0.17
0.18
11.2
0.537
0.54
0.421
593
0.036
0.01
SSD
Tier 2
2011-2015
0.17
0.18
11.2
0.500
0.54
0.421
593
0.036
0.01
MSD
0.17
0.18
11.2
0.500
0.54
0.421
593
0.036
0.01
HSD
0.17
0.18
11.2
0.500
0.54
0.421
593
0.036
0.01
SSD
Tier 3
0.17
0.18
11.2
0.259 <