For a closer look at wellhead to refinery gate Greenhouse gas (GHG) emissions, take a look at these Life Cycle Emissions reports by the Alberta Energy Research Institute.

The Jacobs Consultancy was originally commissioned by the Alberta Energy Research Institute (AERI) to verify or disprove the full fuel cycle analysis for gasoline-from-oil sands feedstock that is used in the GREET model and the modified GREET model that the University of Berkeley team has used to inform the California Low Carbon Fuel Standard. The presentations at the website shows you Jacobs’ findings as well as a comparison of Jacobs findings to those of TIAX—the research on which the Berkley modified GREET full fuel cycle GHG estimates are based.

After that first round of research was published, I asked AERI why they had not also looked at the diesel fuel cycle.  After all, 78% of the full fuel cycle GHGs for gasoline occur at the tailpipe, and we all know that diesel tailpipe GHGs can be 10% to 15% lower (per ton mile of vehicle use) than tailpipe GHGs for gasoline-powered vehicles.  Based on some of my historical working experience, I also understood that refinery GHGs could be as much as 15% lower in a refinery that is designed to maximize diesel output, relative to a refinery that has a 35% gasoline fraction target slate.  

As a result, AERI asked Jacobs to look at the diesel full fuel cycle, but so far I have seen only preliminary conclusions from that analysis.  You might want to look at a new presentation, "Life Cycle Analysis and Technology to Decrease Green House Gas Emissions".

This new research suggests that the full fuel cycle GHGs associated with Canadian exports of heavy oil and refined products from oil sands feedstock is statistically zero.  In other words, the difference suggested in the slide deck is much smaller than the estimation error.  But let’s move on from there.

All of the analysis at this site still deals in average wellhead to refinery gate and transportation emission estimates. It shows that oil sands recovery and upgrading/refining emissions vary greatly depending on extraction process, whether or not the upgrader/refinery is co-generating heat and steam, and whether or not it uses coke to produce heat/steam. The upgrading/refinery emission estimates you see in this analysis are based on the consultants’ best estimates of refinery averages.

But there is one combined upgrader/refinery in Canada that is designed to maximize diesel output (or at least NOT designed to achieve a 35% gasoline fraction) from oil sands feedstock. It is the Scotford Upgrader/Refinery Complex, owned and operated by Shell International. This 25 year-old upgrader/refinery complex was subject to a major expansion starting in 2006.  

We do know the 2005 through 2008 GHGs for the Scotford and the rest of Shell’s operations in Alberta, but I don’t have an estimate of the upgrader’s and refinery’s throughputs at this time.  The upgrader has a rated processing capacity of 155,000 barrels per day (24,600 m-3/d), but has at times pumped out more than 200,000 bbl/d (32,000 m-3/d).  The refinery has a rated capacity of 100,000 bbls per day of synthetic crude feedstock.  

Actual Scotford throughput levels in 2006 and 2007 would not be representative years, due to the construction project and related plant downtime.  But based on old (pre-expansion) estimates of Scotford’s GHGs/MMBTU of product output ratio, I am guessing it should be lower than most other Canadian refineries and the GHG factors used in either the Jacobs or TIAX analysis.  

This is because: (1) the upgrader uses waste heat from the refinery, (2) the upgrader co-generates electricity and steam and (3) neither the upgrader nor the refinery use coke as a heating fuel. I think the Scotford numbers would suggest that the differential between the oil sands to diesel upgrader/refinery GHGs relative to GHGs from conventional crude-to-gasoline in a refinery that is targeting a 35% gasoline fraction—which differential appears to favour diesel-to-diesel over conventional oil-to-gasoline—is large enough to offset the higher extraction -to-upgrader GHGs for oil sands feedstock.

I also believe you will see that a wide range of opportunities exist to cut GHGs in the oil sands supply chain, while GHGs/MMBTU worth of conventional oil tend to increase as conventional oil reserves become depleted.  So the oil sands feedstock is also a superior platform for incremental GHG reductions in the liquid fuel supply chain.

Finally, a new diesel-focused upgrader/refinery is under construction in Alberta, being developed by Northwest Upgrading . The new upgrader/diesel refinery complex should be the "greenest" petroleum product refinery in the world.  It includes carbon capture where the flue gas stream will be injected to enhance oil recovery rates at a nearby semi-depleted conventional sweet crude reserve. This complex relies on gasification and does not have a coking unit.

Fuel Shifting and Tailpipe GHGs

I found the best way to get a picture of how important fuel shifting from gasoline to diesel has been in the EU27’s transport GHG emission profile is to look at the nations’ actual detailed GHG inventories as they appear in the CRF reports here.

Download the .zip file for any country you want to look at. Open the file for 2007, then go to Table 1.A(a) and look at, say, "Road Transportation". Then do the same for as many prior years as you deem necessary. Look at the difference over time.

I show you part of the tables for Denmark, comparing 1999 and 2007, below.  What this shows you is that fossil fuel consumption for road use increased 16.6% in Denmark from1999 through 2007, even though: (1) population grew only 3.2%, (2) cargo freight shipments fell 16.2% and (3) the tax-included price of diesel fuel increased 81% from 61.6 euros/1,000 litres to 111.3 euros/1,000 litres.

Since total cargo shipment (in tonne-miles) declined over the period, clearly 100% of the diesel fuel consumption increases reflect increases in passenger vehicle use and passenger vehicle fleet fuel-switching from gasoline to diesel. The dataset shows that Danish demand for transportation fuels actually grew over 42% between 1990 and 2007—even though population growth was under 6% and cargo freight shipments fell over this longer period.

With the gasoline and diesel consumption estimates from the UNFCCC, and the passenger and cargo freight estimates you can get from the US EIA and Eurostat, you can fairly easily calculate what the European transport sector GHG growth would have been in the absence of the passenger vehicle fleet  fuel switch from gasoline to diesel. And you could calculate what the US and Canadian transport sector GHG trends would have been had we experienced a comparable fuel switch, given our passenger and freight transport trends (our cargo freight shipments have increased, not decreased as in the Danish example).

As an aside, the fuel consumption and price data belie the mantra that simply "putting a price on carbon" will lead to reductions in fuel demand.  

If you  compare population density and ambient temperature (in heating degree days or HDDs) weighted transportation fuel consumption for the developed OECD nations, you will find that there is a very high correlation between per capita fuel demand and (1) population density and (2) average household income. While fuel demand is income elastic it appears quite price inelastic over other-than-very-short periods.

So my gasoline to plug-in electric-biodiesel hybrid passenger vehicle fuel shift—where biodiesel from algae reactors are added to petroleum refinery complexes and the biodiesel is run through the hydro cracker to ensure that it works well in cold weather—is a short to medium term transport sector GHG mitigation strategy. Achieving liveable and walkable nodes of high population densities in our cities, with nodes connected by electric rail, has to be a long-term priority objective.

Read the first part of “The oil sands should be shut down, right?”