Article in the Wash Examiner Electric cars may be worse for the environment than gas"

[Zarwin]

As others have stated in this thread, ignoring the production of gasoline invalidates the study. Check out this pic from the Nissan Drive Electric roadshow back in 2010.

Complex table representation of required analysis components:

	Electricity	Gasoline
Pollution from use of fuel	A	B
Pollution from production of fuel	C	D

Almost all of the comparisons I have seen ignore D, “Pollution from production of Gasoline”, as if it doesn't exist. With A being “none”, they focusing on comparing B to C. Some go into tremendous detail and analysis, leaving out one of the four fundamental factors.

 

[Electroman]

I can't believe we have a "science" study that somehow purports that distributed pollution over hundreds of cities in your face is better then a concentrated discharge in a few power plants where you have much more control.

I am sorry, to me this classic example of junk science. There is a reason why toilets and sewage treatment plants were engineered.

This somehow reeks of a similar study that concluded Hummer is greener than Prius.

The problem with this kind slanted research is that there is huge constituent that would amplify this by repeating just the headlines without the context or background.

Expect this one to be picked by the usual oil shills.

 

[Gregg_G]

As others have stated in this thread, ignoring the production of gasoline invalidates the study. Check out this pic from the Nissan Drive Electric roadshow back in 2010.

View attachment 85231

Complex table representation of required analysis components:

	Electricity	Gasoline
Pollution from use of fuel	A	B
Pollution from production of fuel	C	D

Almost all of the comparisons I have seen ignore D, “Pollution from production of Gasoline”, as if it doesn't exist. With A being “none”, they focusing on comparing B to C. Some go into tremendous detail and analysis, leaving out one of the four fundamental factors.

Wait! That picture is from 2010! The refining of gasoline has become much more efficient since then. Quit cherry picking your Internet data pictures!

Lets face it, if we wanted to be "green" we would've bought a Prius and spent the remaining 50k on new windows/insulation and geo-thermal heating/cooling for our houses.

Can't wait to pick up my 70D on Saturday!

Gregg

 

[Ben W]

To be fair, that is 6kWh of energy/gallon, not just electricity. e.g. Heat from directly burning fuels. A direct comparison assumes you can convert all energy 100% to electricity, which you can't. But large-scale plants should still be able to get 50% efficiency from heat for electricity generation, which would still then vastly tilt the scale in favor of the E.V.

Granted. But extremely dirty forms of fossil fuel (the cheapest leftovers of the oil refining process) are being burned to create most of this energy, resulting in considerable pollution and CO2 emissions. On the other hand, electricity for EV’s is obtainable from much cleaner sources. As the power grid gets cleaner in the future, the balance will tilt further in favor of EV’s. Whereas ICE’s will stay dirty no matter what.

 

[ThosEM]

Granted. But extremely dirty forms of fossil fuel (the cheapest leftovers of the oil refining process) are being burned to create most of this energy, resulting in considerable pollution and CO2 emissions. On the other hand, electricity for EV’s is obtainable from much cleaner sources. As the power grid gets cleaner in the future, the balance will tilt further in favor of EV’s. Whereas ICE’s will stay dirty no matter what.

I guess I'm leaning toward taking this seriously. The study does consider the effect of cleaning up the grid and shows that the result is to turn around the study results, in the last table of the paper. That isn't a featured result, for sure, and in fact there is quite a bit of verbiage about why that isn't the bottom line AT THIS TIME.

Of course, all of us would make that a featured result, and insist on a full accounting of the energy and emissions cost of preparing fuel for the pump or the power plant. And we would point out that if we continue with ICE business as usual, no amount of hybridization would ever allow for a future with far lower emissions as the grid is cleaned up.

 

[techmaven]

A lot to respond to! About the comments from GregRF and ReddyLeaf. Take a look at Table 5 in our paper. On average in the US, electric cars cause 2.5 cents
per mile damages, of which 1.62 cents come from local pollutants (S02, NOx, etc) and the rest from the global pollutant (C02). On average in the US, gas cars
cause 2.0 cents per mile damages, of which 0.54 come from local pollutants and the rest from C02.

So we found that, on average, electric cars are better on C02 and worse on local pollution. But again, our paper is really about how much this
can vary from place to place. In the best county for EVs they only cause 0.67 cents per mile damages. In the best county for gas vehicles, they cause 1.13 cents per mile
damages. The worst county for EV's is a bit worse than the worst county for gas vehicles (4.72 vs 4.47)

Right, so the mapping to damage in terms of $ is extremely important to the conclusions of this study, even given the gross negligence of skipping the emissions of upstream processes. Which is more important, increased GHG emissions or non GHG emissions? It seems that the mainstream press can't distinguish between the two. What is the effect of just getting all the coal plants to install advanced pollution control equipment? Stepping outside of transport, isn't it important for our health to have the 40% of electric generating units (EGUs) that do not have this equipment install it as soon as possible? Once that is done or if the plant is too old anyways and is retired, then what is the resulting impact on transport?

Cleaner Power Plants | Mercury and Air Toxics Standards (MATS) for Power Plants | US EPA

It seems the problem is that the Clean Air Act hasn't been fully implemented and the various interests have successfully fought the regulations and the regulators to delay its effects. We have been paying for that with our health and with climate change in the meantime, regardless of the issues in transport.

Further, there should be a premium placed on the ability to shift emissions from the point of the vehicle, out of where most of us live (hence ZEV credits). The study ignores this in the economic model. The study also ignores the fact that combustion inside these vehicles is heavily dependent on the tuning of the vehicle and worsens with age, while it takes real world emissions data from many old power plants.

Also, to put more color onto the difference in vehicle emissions with better data, look at this comparison on fueleconomy.gov:

Compare Side-by-Side

Switch to the "Energy and Environment" tab and select tailpipe and upstream GHG which uses the Argonne National Labs GREET model for CO2 emissions and using annual electricity production mix. Here's the results:

2015 Toyota Prius: 218 g/mile
2015 BMW 535i: 453 g/mile
2015 Audi RS7: 572 g/mile

Now, a Model S 85 kWh in:
San Francisco, CA: 150 g/mile
Los Angeles, CA: 150 g/mile
New York, NY: 150 g/mile
Boston, MA: 160 g/mile
Seattle, WA: 170 g/mile
Washington, DC: 190 g/mile
Miami, FL: 250 g/mile
Dallas, TX: 250 g/mile
Atlanta, GA: 270 g/mile
Chicago, IL: 300 g/mile

Now, since this data uses annual production mix, it doesn't reflect the reality that in many of the power generation districts, the super-off-peak production is dominated by hydro, nuclear, and wind power. So the actual emissions from charging at super-off-peak is far less. Even in Chicago, where the power generation in that district is dominated with coal production, the CO2 emissions is lower than a BMW 535i by a significant margin. A family choosing a Model S over a BMW 535i, even in Chicago, will have a minimum of a 33% lower CO2 emissions level after 5,000 or so miles. Now, to compare it against a Prius, it is higher if one charges a lot during the day from the grid. Of course, if one has solar panels and is charging off them during the day, then emissions are again far lower.

 

[abc20000]

Another major set of comments is centered around the concern that we only address life cycle considerations (emissions from refining gas, mining coal) as a caveat. The general consensus is that electric cars offer a significant benefit in this regard. This consensus is at odds with the published research on the topic. See for example "Valuation of plug-in vehicle life-cycle air emissions and oil displacement benefits", Michalek et al, Proceedings of the National Academy of Scieinces, Vol 101, 2011. They find that the differences between a gas car and an EV are small.

 

[deonb]

A lot to respond to! About the comments from GregRF and ReddyLeaf. Take a look at Table 5 in our paper. On average in the US, electric cars cause 2.5 cents
per mile damages, of which 1.62 cents come from local pollutants (S02, NOx, etc) and the rest from the global pollutant (C02). On average in the US, gas cars
cause 2.0 cents per mile damages, of which 0.54 come from local pollutants and the rest from C02.

So we found that, on average, electric cars are better on C02 and worse on local pollution. But again, our paper is really about how much this
can vary from place to place. In the best county for EVs they only cause 0.67 cents per mile damages. In the best county for gas vehicles, they cause 1.13 cents per mile
damages. The worst county for EV's is a bit worse than the worst county for gas vehicles (4.72 vs 4.47)

Ok, so I think you've heard our objections of the global effects of CO2 and the incorrect argument about ignoring the upstream effects of refining & transportation on CO2.

However, even if you focus on the non-CO2 effects, I believe it would still be wrong to ignore the impact of refining. There are 137 oil refineries in the U.S, spread over 32 states. Those refineries also create SO2, NOx etc. and pollutes those into their respective environments.

Let's look at one point specifically. When you're saying (from 4.3 - Exporting pollution: Full and native damages):
"In contrast, only fifty seven percent of gasoline damages are exported at the county level and only nineteen percent at the state level."

I can't find a source for that to look up the raw data, but just visually looking at the following graph, it doesn't appear that it could be accurate:

Figure 2 Panel A


There are no refineries in Fulton county, where the source of pollution is marked. There is however one in Savannah, so at the very best I expect there to be a single red circle around that area as well. Realistically though it will probably highlight multiple spots around the East Coast in other states as well. This seems to be completely ignored, or am I wrong?


I have a question in general about the green areas in the graph: Is this just indicating air & water distribution of pollutants originating from Fulton county, or does it take any other externalities into account that originate in the surrounding areas? E.g. electricity generation to supply the gas station with energy for pumping. (I know that one specifically is minor - just want to know exactly what is being indicated).

 

[abc20000]

Thank you for all the comments on our study. I am sorry I was not able to respond to all of them. You have given me lots of interesting things to think about.

 

[MichFin]

Rule of thumb when an article leads with "May", "Some", "Possible" it's not based on real facts.

 

[ThosEM]

Another major set of comments is centered around the concern that we only address life cycle considerations (emissions from refining gas, mining coal) as a caveat. The general consensus is that electric cars offer a significant benefit in this regard. This consensus is at odds with the published research on the topic. See for example "Valuation of plug-in vehicle life-cycle air emissions and oil displacement benefits", Michalek et al, Proceedings of the National Academy of Scieinces, Vol 101, 2011. They find that the differences between a gas car and an EV are small.

That's a bad reference. It's actually Michalek, J. J., Chester, M., Jaramillo, P., Samaras, C., Shiau, C.-S. N., & Lave, L. B. (2011). Valuation of plug-in vehicle life-cycle air emissions and oil displacement benefits. PNAS, 108(40), 16554–8.

And it can be viewed here online: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3189019/

It's clear from the abstract that they concede the clear benefits of plug in vehicles over hybrids, but consider the extra cost to be unjustified by the benefits. This devolves into opinion and the valuation of benefits quite rapidly. IMHO, they don't properly value the impact of CO2 on environment, which is key. On the other hand I do agree that it will be easier to justify the switch to plug in EVs when their costs are comparable to HEVs.

"We assess the economic value of life-cycle air emissions and oil consumption from conventional vehicles, hybrid-electric vehicles (HEVs), plug-in hybrid-electric vehicles (PHEVs), and battery electric vehicles in the US. We find that plug-in vehicles may reduce or increase externality costs relative to grid-independent HEVs, depending largely on greenhouse gas and SO2 emissions produced during vehicle charging and battery manufacturing. However, even if future marginal damages from emissions of battery and electricity production drop dramatically, the damage reduction potential of plug-in vehicles remains small compared to ownership cost. As such, to offer a socially efficient approach to emissions and oil consumption reduction, lifetime cost of plug-in vehicles must be competitive with HEVs. Current subsidies intended to encourage sales of plug-in vehicles with large capacity battery packs exceed our externality estimates considerably, and taxes that optimally correct for externality damages would not close the gap in ownership cost. In contrast, HEVs and PHEVs with small battery packs reduce externality damages at low (or no) additional cost over their lifetime. Although large battery packs allow vehicles to travel longer distances using electricity instead of gasoline, large packs are more expensive, heavier, and more emissions intensive to produce, with lower utilization factors, greater charging infrastructure requirements, and life-cycle implications that are more sensitive to uncertain, time-sensitive, and location-specific factors. To reduce air emission and oil dependency impacts from passenger vehicles, strategies to promote adoption of HEVs and PHEVs with small battery packs offer more social benefits per dollar spent."

 

[deonb]

Check out this pic from the Nissan Drive Electric roadshow back in 2010.

I really wish it was as slam dunk, but it's not. I believe Nissan stopped citing that as well.

Those figures come from the fact that oil refineries are 85% efficient, that you get 45 gallons of gasoline from a 42 gallon barrel of oil, and that oil has a raw energy of 132'000 BTU. So applying the formula of:

It gives you 21'000 BTU's of energy loss, which is the equivalent of 6kWh.

But it's not quite accurate. You can't actually take 0.15 gallon of crude oil and put it in any kind of device and get 6 kWh of electricity from it. If you could, we wouldn't need solar for the next 100 years.

You could however say, if you lose 15% of oil energy during refining, that has to be going somewhere, and that's a lot of carbon to just swipe under a rug, so it goes into the air instead. I wish this was the argument from the beginning, but now the message is a bit lost.

 

[techmaven]

Another major set of comments is centered around the concern that we only address life cycle considerations (emissions from refining gas, mining coal) as a caveat. The general consensus is that electric cars offer a significant benefit in this regard. This consensus is at odds with the published research on the topic. See for example "Valuation of plug-in vehicle life-cycle air emissions and oil displacement benefits", Michalek et al, Proceedings of the National Academy of Scieinces, Vol 101, 2011. They find that the differences between a gas car and an EV are small.

That's not what that paper actually says. You ignored both total life cycle emissions which includes manufacturing/production emissions as well as upstream emissions during usage. The paper you referenced, published in 2011, uses very old data that includes manufacturing and production emissions. For a real world check, see this Mercedes environmental impact statement on the Mercedes B-class electric that has a Tesla powertrain:

http://media.daimler.com/Projects/c2c/channel/documents/2582749_final_UZ_B_Kl_ED_engl_15_12.pdf

That PDF specifically compares a B-class 180 versus the electric version. The power train, including motor, inverter, and batteries are assembled in CA by Tesla and then shipped to Germany for final assembly at the same place as the internal combustion version of the B class.

If you go back to the Michalek 2011 paper, the CV case versus the BEV240 case, you can see there are similarities, but the reality is very different. The EU energy mix has a substantial amount of coal. SO2 emissions is higher but pretty much everything else is lower. So the "US Avg Grid Mix" base case in the Michalek 2011 paper, ignoring the Military/Monopsony/Disruption bars, the BEV240 should be lower than the CV. The SO2 is not that much higher - 125%, not something like 500% as shown. The CO2 is also 22% lower with the electric version using the EU production mix. The Mercedes statement covers only a lifetime of 100,000 miles while the Michalek paper 12 year base case, so the results are even better for the electric version when stretched to a 12 year lifecycle. Again, look at the Mercedes chart on CO2 emissions, comparing EU grid mix, hydro, and internal combustion. After production, we're talking 11.9 EU grid mix, 0.2 hydro, and 20 internal combustion. That's with 100,000 miles, and we know that the pack is likely to last more like 200,000 miles. Many of us are charging off wind, hydro, and nuclear at night. It would be interesting to know just how many electric cars could be charged in each county or grid region before any coal plants need to increase power generation even one iota between 1 am and 4 am.

The primary problem for a number of these studies it the use of assumed weight x assumed materials = large assumption errors. It's a reasonable first approximation, but when the output differs by, say 20% and the assumed weight is off by, say, 100% or in some cases an order of magnitude or so, or the material itself is wrong, the results are garbage. What goes into a Tesla/Panasonic NCA battery is not the same as a Nissan Leaf lithium manganese battery or an i3 NMC battery. And none of them even come close to what is in the modeling of some of these papers, especially the ones widely quoted by the media.

 

[deonb]

Another major set of comments is centered around the concern that we only address life cycle considerations (emissions from refining gas, mining coal) as a caveat. The general consensus is that electric cars offer a significant benefit in this regard. This consensus is at odds with the published research on the topic. See for example "Valuation of plug-in vehicle life-cycle air emissions and oil displacement benefits"...

If you look at the supplemental material for that study, Table S25, it lists the effect of Gasoline production on GHGs:

CV GHGs (overall $2010/vehicle):
Vehicle production: 316
Battery production: 12
Gasoline production: 290
Electricity production: 0
Vehicle operation: 1408
Total: 2025

Hence gasoline production alone account for 14.3% of the overall Lifecycle GHGs of the conventional vehicle, and 17% of the direct fuel GHGs of the vehicle. This isn't insignificant.

 

[paulwesterberg]

The Washington Examiner is a conservative rag.

The National Bureau of Economic Research is a private conservative think tank with questionable funding.

 

[AudubonB]

Ad hominem attacks are rarely useful. In that one of the study's authors has been both gracious and brave enough to be present in this thread, that kind of post is even less appropriate.

 

[EarlyAdopter]

I really wish it was as slam dunk, but it's not. I believe Nissan stopped citing that as well.

Those figures come from the fact that oil refineries are 85% efficient, that you get 45 gallons of gasoline from a 42 gallon barrel of oil, and that oil has a raw energy of 132'000 BTU. So applying the formula of:

It gives you 21'000 BTU's of energy loss, which is the equivalent of 6kWh.

But it's not quite accurate. You can't actually take 0.15 gallon of crude oil and put it in any kind of device and get 6 kWh of electricity from it. If you could, we wouldn't need solar for the next 100 years.

You could however say, if you lose 15% of oil energy during refining, that has to be going somewhere, and that's a lot of carbon to just swipe under a rug, so it goes into the air instead. I wish this was the argument from the beginning, but now the message is a bit lost.

I don't know Nissan's data source, but it must include other factors as they cite 7.5kWh per gallon, not 6kWh.

Here's a stab at quantifying the other factors:

15% loss in refining, 6kWh. Assume it's burned at the refinery at 25% efficient power production. 1.5kWh of electricity through this cogeneration is retained for running refinery operations and 4.5kWh of waste heat is retained for catalyzing reactions.

Now, we know that refineries cannot meet all of their electric power needs through cogeneration alone and have to draw additional power from the grid. This is evident by looking at the large switching stations present at every refinery. Let's be generous and assume they meet 50% of their needs through cogeneration. This would then mean an additional 1.5kWh from the grid per gallon.

This brings us to 3kWh of electricty per gallon refined, with their related emissions (1.5kWh worth of emissions from the grid supplied power, and 1.5Kwh worth of bottom of the barrel oil burning emissions from cogeneration).

I ignore upstream costs of exploration, extraction, and transportation of oil feedstock because we're staying with the worse case comparison to EVs powered by the standard power mix, which has similar costs upstream of the power plant.

However, we do need to estimate the fossil fuel emissions from transportation from refineries to gas stations, the electric power needs of refining those gallons burned transporting, and the electricity needed for running gas stations. These are not insigificant values, but I have no data to go on to estimate these. Anyone have references?

I will say its not inconceivable to estimate another 3kWh of electricity per gallon finally pumped, with their emissions from the standard grid mix, plus additional emissions from transportation, to get us to 6kWh.

Factor in costs upstream of the refinery, and I could see 7.5kWh.

 

[deonb]

Anyone have references?

Unfortunately I haven't seen any up to date studies that analyzed it to a better extent.

This article has a few links:
http://greentransportation.info/guide/energy/electricity-to-refine-gallon-gasoline.html


However, if we go with the Michalek et. al study that's based on GREET 1.8d GHG factors (which is comprehensive cradle to grave), it puts a max theoretical cap of 17% for production pollution vs. operating pollution. So that would be 5.7 kWh.

It's still not the best study since it aggregates CO2 into general GHGs. At some point I'll download the GREET calculator and try to redo this table myself.

GHGs for the purpose of this table is some mix of CO2, CH4 and N2O.

 

[dsm363]

Article in the Wash Examiner Electric cars may be worse for the environment t...

Was the production of gas included in their projections?

 

[digitaltim]

Well 2 wheel is the only appropriate measure.