3 ways to calculate co2 emissions in shipping


COVID 19 has marked our time of unprecedented crisis with the global economic shutdown, questioning the way our society operates from an economic and environmental point of view, as daily global CO2 emissions decreased by –17% by early April 2020.

But COVID 19 has also allowed us to see how well human beings can adapt to situations that leave them no choice but to act and adapt. Continuing working remotely, driving our car less, attending a meeting via Zoom, not taking a flight, changing our health habits, shifting to a less carbon-intensive lifestyle, reveal the scope of what we can achieve in just days.

Why, then, is it so laborious to apply and impose new rules in terms of climate change? COVID 19 forced governments to step in and deal with the catastrophe in a way that is unprecedented, including supporting business and industry, and public and private infrastructure. Individual actions won’t be enough to cut CO2  emissions. Governments and industries must take over.

In the Maritime industry for instance, last December, an overwhelming report from SeaIntelligence stated “the carriers’ CO2  calculators are completely useless!”. A very first step could be to harmonize the way CO2 emissions are measured and reported by the Maritime and freight transportation stakeholders.

Few weeks ago, Hapag-Lloyd reported a 50% drop in CO2  emissions per teu/km in 2019 according to its sustainability report.  What does this result mean? How do they calculate their CO2  emissions? Shippers need to have a reliable and comprehensive CO2  emissions benchmark. In this new blog series, we’ll give you all the information to become a CO2 expert. This first blog post will help you have an overview of calculating CO2 emissions, and stay focused on the information you need to have accurate results.


Data and emissions factors, are core information to estimate CO2

In order to calculate accurate CO2  emissions, many parameters must be taken into account. The result depends, for example on the distance travelled, the type of vehicle (train, truck, ship but also more specifically cargo container, bulk ship, etc…), the particular vehicle (fuel type, motor power, age of the motors, etc…), the load of the vehicle, the speed of the vehicle compared to its design speed, the weather, etc. The more accurate data you have on these parameters, the more accurate your results will be.

All these parameters make it difficult to design an accurate calculation method.  Hence many approximation methods have been presented, from global approximations to more granular methods.

According to Hartmut Zadek & Robert Schulz (2010) “There are two possibilities to calculate the CO2  emissions of mobile sources: the fuel-based method and the distance-based method. Both methods use CO2  emission factors (EFs) for the calculation of CO2  emissions.”

An emission factor is a variable which allows you to transform data you get on a route such as distance, amount of goods transported or the quantity of fuel consumed to a quantity of CO2  consumed. Depending on the parameter you choose in your calculation, your CO2 emissions factor won’t be the same.


1st way: Distance based methodology


The easiest way to compute an approximation of CO2 emissions of a particular voyage is to associate to each transport mode an emission factor EFs in g/t.km. Here below, emissions factors recommended by ADEME ([1]) for container vessels.

Vessel capacity CO2  emission rate per ton transported & km
Container vessel – less than 1 200 TEU 32,5 g CO2  / t.km
Container vessel – 1 200 to 1 899 TEU 21,6 g CO2  / t.km
Container vessel – 1 900 to 3 849 TEU 20,1 g CO2  / t.km
Container vessel – 3 850 to 7 499 TEU 13,4 g CO2  / t.km
Container vessel – More than 7 500 TEU 10,1 g CO2  / t.km

The distance based methodology is based on ADEME [1] recommendations and takes into account the direct distance traveled and the cargo weight.

Let’s take as an example a maritime company which needs to calculate CO2  emissions based on the information below:

  • Cargo container: 4,400TEU
  • Freight weight:
    • We assume 1TEU = 10t, this means 4400 TEU * 10t = 44,000,000 kg
    • Vessels are full at 70% (according to IMO study 2009 [2] ).  So the weight the vessel is carrying is 44,000,000*0.7 kg = 30,800,000 kg or 30,800t 
  • Emissions Factor: 13,4 gCO2 /t.km
  • Distance: Hong Kong – Marseille = 14,790 km. You can get this information here searoutes.com

E= (14790km x 30,800t x 13,4g CO2 /t.km)/1 000 000 = 6,104 t of CO2. 

This method of calculation is based on the assumption that emissions can be accurately approximated based on distance and weight only and that actual emission factors are homogenous among the vehicles of the same type.


Pros and cons

This method is easy to use. But you have to keep in mind that it is an order of magnitude and you can’t use these results to report CO2  emissions properly.

Indeed for accuracy you would need one emission factor per vessel and the true distances sailed by the vessel (including ports of call) instead of direct distances. As mentioned previously, vessel speed also has an impact on CO2  emissions as it is directly related to fuel consumption. This conclusion leads us to the second method based on vessel fuel consumption.


2nd way: Fuel consumption based methodology


A more realistic CO2  emission calculation rests upon fuel consumption as it is the actual source of emissions and thus combines many parameters. Indeed, fuel consumption varies with speed, load, vehicle type. Moreover, emissions also vary depending on fuel type as reported below.


Did you know? 

The relationship between the fuel consumption and amount of CO2  produced comes from the chemical fuel composition with Hydrocarbons (Carbon(C) and Hydrogen (H)). When combusted, hydrocarbons react with oxygen (O2) and create CO2 . The carbon content varies according to the different types of fuel and thus has an impact on CO2  emissions

Type of Fuel Reference Carbon Content
Diesel/gas Oil ISO 8217 Grades DMX through DMC 0.875
Light Fuel Oil (LFO) ISO 8217 Grades RMA through RMD 0.86
Heavy Fuel Oil (HFO) ISO 8217 Grades RME through RMK 0.85
Liquefied Petroleum Gas (LPG) Propane 0.819
Butane 0.827
Liquefied Natural Gas (LNG) 0.75

In order to calculate CO2  emission from fuel consumption, one needs a conversion factor i.e mass of CO2  emitted by tonne of fuel consumed and type of fuel consumed. The conversion factors below can be found in the “Third IMO Greenhouse Gas Study”, conducted by the International Maritime Organization (IMO) [3]

Fuel type CO2  emission rate per g of fuel consumed
HFO eFbaseline CO2 = 3,114 g CO2 /g fuel
MDO eFbaseline CO2 = 3,206 g CO2 /g fuel
LNG eFbaseline CO2 = 2,750 g CO2 /g fuel

This calculation is based on the assumption that these factors are accurate enough for a vehicle type.

Let’s take the same example as above following

    • Cargo container: 4,400 TEU
  • HFO fuel consumption: 1,663.7 t
  • Emissions Factor: 3,114 g CO2 /g fuel (HFO)
  • Distance: Hong Kong – Marseille =14,790 km. You can get this information here searoutes.com

E= 1,663.7t x 3,114 gCO2 /g fuel = 5,181 tonnes of CO2 


Pros and cons

This methodology provides accurate CO2 emissions as it is directly related with fuel consumption. Reliable fuel data is therefore a key factor to use it. Moreover, to be useful, you need a continuous monitoring of fuel, so you know exactly how much emissions on each leg of a voyage. Unfortunately carriers are the only ones to own this data and this is not available to all.

Clean Cargo Working Group (CCWG) has undertaken an initiative to allow carriers to benchmark their carbon footprint and report the results to customers in a standard forma. However, they only report aggregated fuel consumptions, yearly, per trade lane, for only 80% of their fleet.

Although this method seems to be the most appropriate, it loses its value if the initial data are not specific and detailed enough. For this methodology to be used by shippers and freight forwarders, one would basically need fuel consumptions for each vessel as given in the noon reports. Needless to say that information is sensitive, and carriers are not giving that information away!

Finally, for this to be useful, one needs to cross reference it with other data points, on every leg of a voyage, with AIS, weather, vessel information etc.


3rd way: Engine consumption methodology


The previous method, although accurate, is difficult to apply as the data is not open to all. In the following methodology we assume that the engine stress level impacts the fuel consumption and therefore the CO2  emissions.

Indeed, the efficiency of a diesel engine is related to its load level or its load factor. The engine load factor is defined as the actual power output of the engine relative to its Maximum Continuous Rating (MCR).

As shown in the figure below, there is an optimal load factor  level where the amount of fuel consumed for each unit of power output is the lowest which directly impacts emissions.


3 ways to calculate co2 emissions in shipping
Figure 1: The engine Specific Fuel Consumption (SFC) as a function of the load factor


The main engine power output and load factor vary over time as a result of a ship’s operation and activity specifics: operational mode  (e.g.  at  berth,  anchoring,  manoeuvring),  speed,  loading  condition,  weather,  etc. This method takes into account at least the following parameters: the traveled distance, the average speed, the engine nominal power, the load factor of the engine and the emission factors.

Here below we made a comparison of the results between solutions based on the same methodology. The main difference comes from the vessel specificities (size, design speed and speed) and real distances (integrating ports of call) by this specific vessel.

As reported below, other solutions estimate CO2  emissions for a vessel size category (7,000-14,500 TEU) and use direct distances.


Pros and cons

This model is available to all as fuel consumption calculation is based on open source data. The more accurate data you integrate  such as vessel’s real speed, real distances, different engine and fuel types, impact of weather if you have reliable data the more accurate your fuel consumption and therefore CO2  emissions results will be.

However it is only as accurate as your fuel consumption estimates, accuracy depending on the qualitative and relevant data you have to refine your calculation. Discover Searoutes’ methodology secret.


Next steps?

Now that you know the 3 most common ways to calculate CO2 emissions, you ‘ll be able to better understand the existing solutions methodologies for your business to achieve your goals in CO2  emissions reporting or monitoring and therefore make better decisions to answer your sustainability issues .

As mentioned above, accurate data is key to estimate and report accurate CO2  emissions. If you’re looking to test out a few CO2  emissions calculations options, start by checking out our 14-day free CO2  emissions API trial or ask for more information.



[1] Information CO2 des prestations de transport, October 2012

[2] Second IMO GHG study 2009. / Buhaug, Ø & Corbett, James & Endresen, Ø & Eyring, Veronika & Faber, J. & Hanayama, S. & Lee, David & Lindstad, Elizabeth & Markowska, A.Z. & Mjelde, Alvar & Nelissen, D. & Nilsen, J. & Pålsson, C. & Winebrake, James & Wu, W.–Q & Yoshida, K.

[3] Third IMO Greenhouse Gas Study 2014. / Smith, T. W. P.; Jalkanen, J. P.; Anderson, B. A.; Corbett, J. J.; Faber, J.; Hanayama, S.; O’Keeffe, E.; Parker, S.; Johansson, L.; Aldous, L.; Raucci, C.; Traut, M.; Ettinger, S.; Nelissen, D.; Lee, D. S.; Ng, S.; Agrawal, A.; Winebrake, J. J.; Hoen, M.; Chesworth, S.; Pandey, A.

[4] EMEP/EEA air pollutant emission inventory guidebook 2019 / Carlo Trozzi, Riccardo De Lauretis, Kristin Rypdal, Anthony Webster, Erik Fridell, Gillian Reynolds, Jean-Pierre Fontelle, Kevin Lavender, Niels Kilde, Nikolas Hill, Roel Thomas, Morten Winther.

[5] The influence of different types of marine fuel over the energy efficiency operational index 2014 / European Geosciences Union General Assembly 2014, EGU 2014Nicoleta Acomia, Ovidiu Cristian Acomi.