Is a Tesla cleaner than a dirty diesel?

If you are a Minnesota driver, the answer is yes, but not by much it would not meet federal emissions standards for light duty vehicles. Including energy generation, a Tesla will produce slightly far fewer NOx emissions per mile than a “defeated” VW. But a Tesla in Minnesota is far three times dirtier, in terms of NOx emissions, than the average car on the road, or than federal emissions standards allow for light duty vehicles.

(I was kind of surprised by these results. If you think f I’m missing something here, let me know and I’ll make corrections to calculations and notes, as appropriate.)


Tesla is running this ad, apparently (UPDATE: this may not be an ad, but may have been produced by that twitter user. If you know the source, let me know), and Elon Musk says cars should be tested at random to see if they meet emissions requirements. This is in response to the Volkswagen Scandal.

It’s fairly clear at this point, that in many locations with cleaner electricity sources, EVs have a carbon benefit over ICE vehicles (this EPA calculator gives you results for your area). Minnesota is one of those places, even though our electricity still comes mostly from fossil fuels.

chart (3)

But what about nitrogen oxide (NOx), the pollutant at the heart of the VW Scandal? (Here’s a rundown of the bad stuff NOx does to things that breath air) A VW with a “defeat device” could emit up to 35 times the federal emissions limits for NOx. That could be as high as 2.45 grams or 0.0054 pounds per mile. So how does an EV, like the Tesla, compare?

A Tesla uses 33 kWh for every 100 miles traveled (or 0.33 kWh per mile). According to federal statistics, Minnesota power plants emit 1.4 pounds of NOx pollution for each MWh of electricity generated (Xcel Energy shows a similar 1.5 lbs/MWh in its reporting). That means a Tesla is responsible for 0.210 grams of NOx per mile. So, to fact check the ad: yes, driving a Tesla (in Minnesota) is slightly quite a bit cleaner than a diesel VW with a defeat device.

However, that Minnesota Tesla is responsible for emitting 1.5 times more NOx per mile than the dirtiest cars allowed on the road by federal emissions standards, and 3 times more NOx per mile than the allowable fleet average NOx emissions. So it’s too early for Tesla owners to get smug about their impact on the environment. In fact, they are squarely in dirty-diesel territory. (UPDATE: my math was off in the preceding calculation, it has been corrected). You might interpret that ad another way: Tesla has actually defeated emissions testing – by moving the tailpipe from the car to a distant power plant (UPDATE: again, not sure this is a real ad).

Electric vehicles can be better for NOx – if we kick coal

If you were driving a Tesla in Washington State, which has a NOx emissions rate of 0.3 lbs/MWh, you’d actually be emitting 35 percent less NOx per mile than the average light duty vehicle. Why? Because Washington residents get most of their electricity from emissions-free hydropower.

The break-even point seems to be 0.46 pounds of NOx/MWh, which is about equal to the emissions rate for the state of Oregon. This is what their electricity sources look like:

chart (4)

Oregon only gets 6 percent of its electricity from coal. Minnesota is currently at 50 percent. It’s clear what we need to do to make EVs cleaner: reduce Minnesota’s use of coal for making electricity.

We probably need to electrify transportation to meet the State’s aggressive climate action goals. However, we don’t want to just trade carbon benefits for dirtier air and all the associated impacts (asthma, deaths). We need to simultaneously begin the transition to electric vehicles, AND rapidly decarbonize and de-coal our electricity grid.


  1. Brendan Jordan asserts via twitter that 50 percent of Minnesota EV owners use wind power for their EV charging. This means they either buy, or their utility supplies, wind “credits” to supply the amount of electricity they use for EV charging. If the credit tracking system works, this wind is “additional”, and does in fact reduce emissions. If it’s true (I haven’t seen the data) that’s great, and just another argument for decarbonizing the whole grid. However, this is not the “default” when you plug your vehicle into a charging station at home or at work. Also free wind for EV charging is not offered by Xcel Energy, Minnesota’s largest utility.

22 thoughts on “Is a Tesla cleaner than a dirty diesel?

  1. Moving the ‘tailpipe’ to distant power plants also means reducing the number of people exposed to the NOx, though, and since the worst effects of NOx are experienced within the airshed, this matters. To really compare apples to apples, you’d need to take into account this spatial aspect, though obviously, that gets a lot more complicated.

      • And why do you think the EPA wants even stricter controls on the outputs of these plants? Because its easier to control there than on a million tailpipes, and oh, I forgot, the Tesla doesn’t have one of those. It doesn’t matter where you live, the Tesla is much better for you than any ICE mobile. And if the power plants move to NG all the better, the NOx emissions come way down then. Tell the power companies to clean up their act and we’ll all be better off!!

  2. You appear to have a factor of ten problem. Your number for the VW is 2.45g/m. For the Tesla your number is 0.209g/m which is less than a tenth as much. That is not ‘slightly cleaner’.

  3. ICE engines are at best 20% efficient in stop and go traffic. Even a coal burning generator running at one speed is much more efficient than that.

  4. Since our solar PV array generates more power than our home uses most of the year, and since our two Tesla’s do almost all of their charging at home, our Tesla’s contribute virtually no NOx to the atmosphere.

    • That’s great, and illustrates my point: EVs can be much cleaner, if the electricity used to power them is clean! Do your Teslas charge at night? If so, then there NOx emissions are actually equal to the emissions rate from your local utility at nighttime, not your solar.

  5. Have you ever checked how much (dirty) electricity is required to refine 1 gallon of Diesel (more than 4 kWh)! Or the other processes for generating diesel. Please take these into account and re-do your math!

    • This was only an analysis of local NOx impacts. Fuel refining could have additional upstream impacts as you point out, but it could also happen many states distant from where the fuel was burned. A full upstream analysis would also have to include the extraction, processing and transportation of the coal, natural gas and other fuels that went into creating the electricity. The NOx impacts of those could also be very distant from the vehicle.

      • Yeah, but then you would need to calculate the fuel and energy consumed to extract, process, and transport the oil/gasoline as well. Electric comes out on top when all things are considered. And I don’t see how you can justify dismissing the impacts of those processes as distant if your intentions are, in fact, humanitarian.

        • I’m not dismissing them – they are real! They aren’t quantified in this post. If you’ve got data sources on upstream criteria pollutant impacts, please leave them here in comments and I’ll try to develop that as a future post.

  6. you’re failing to compare apples to apples.

    ie, if you’re going to incorporate production of the fuel (electricity) for a tesla into your calculations, you should also incorporate the production (and delivery) of the fuel for gasoline powered vehicles.

  7. Setting aside whether your numbers add up, I’ll make the high level point that our electricity system is getting steadily cleaner (both for “conventional” pollutants and GHG pollutants. Average gasoline (upstream and downstream) is becoming more polluting for GHG emissions, may or may not be for conventional. And EV use is almost immeasurably small right now as a percent of electricity use (but likely to grow fast). The goal right now is deploying a technology that can have a measurable impact in the future when we can bring it to scale. The right question is what impact will EVs have in 5 or 10 or 20 years when we have lots more of them on the road? If the impact is likely to be positive, then we should be promoting the technology. Everyone from NGO activists to utility CEOs agree that our electricity system will look a lot cleaner in 10-20 years.

    I’m much more conversant in GHG lifecycle accounting, not so much on questions related to conventional pollutants. But I’ll throw out a few questions and comments:
    -We surveyed EV owners, got nearly 80 responses. More than half are enrolled in a utility green power program (e.g. Windsource from Xcel, Wellspring from GRE). Not a scientific survey, but informative nevertheless.
    -That number will likely go up, because GRE and Xcel both upgrade EV owners to green power for no extra charge.
    -EVs can enable more renewable power deployment by charging at night when there is currently more wind production and less demand. There is a lot of debate about the grid mix for a new nighttime kWh is right now, but its pretty clear that as coal plants shut down and there is less excess capacity (this is happening now), the marginal nighttime KWh will look more like wind power balanced by NG, rather than a ramping coal plant.
    -It seems like you are leaving out upstream emissions from petroleum production, transportation, and refining. At least for GHG accounting, upstream emissions are not such a big deal for power plants – most emissions are smokestack. But they are a big deal for transportation fuel. And we do refine a lot of the fuel we use in the metro area.
    -How important are spatial dynamics? It is my understanding that NOx impacts are a big deal along busy highways. Are they as big of a deal even coming out of Sherco?
    -Are the EPA numbers up to date? A number of new air rules (e.g MATS) have been forcing older, dirtier coal plants to shut down. Is that reflected in the “average” per kWh NOx numbers?

    Minor quibble – the Tesla “meme” feels a bit sensationalistic to me. You can easily run a Tesla on 100% renewable power, and many people do. Can you do that with a VW diesel, even w/o disabling the emissions controls? Comparing a Tesla w/ VW’s deliberate fraud is worse than misleading. I know you’re being snarky, but people tend to misinterpret things like that…

    • Many good points. A few responses:

      – I hope it is clear in the post I think EVs are the way to go (I say this at the end). However, one argument we are hearing in Minnesota right now (at least from the political realm) that it is more cost effective/pollution-preventing to focus on a transition to EVs than to continue to reduce our coal use. The point of this analysis is to show that this is a false choice. If we transition to EVs without cutting coal, we may actually be creating more pollution (at least in terms of NOx and SOx). We need to be doing both at the same time. While NGOs and utility executives might agree, this is still an active debate in the political arena (at least publicly).

      -I noted your comment on green power purchasing in the post. This is also an argument for continuing to decarbonize the grid!

      -This analysis does not include upstream fuel refining emissions. It also doesn’t include emissions from the mining of coal, extraction of gas, or transport of the fuel to power plants. I’d love to do a post on this as well, but the criteria pollutant impacts of these upstream sources are not obvious to me. If you’ve got sources, let me know! Your point is well taken that in the Twin Cities at least, some of the emissions from fuel refining may be felt locally given Flint Hills location.

      -The EIA figures on NOx lbs/MWh are from 2013. So, they do not show the impact of new air rules like MATS. This relates to my first point though – we can’t champion EVs without continuing to clean the grid.

      -I’ll admit that the VW angle is a bit sensationalist, but this is a blog! Also, the main point stands: in MN and many other states, a Tesla or other electric vehicle (without green power) is responsible for emitting three times the legal limit of NOx in the vicinity that an ICE vehicle is allowed under federal rules. Vehicle emissions are easier for people to understand, because most own a car. Power plants are distant and not familiar to most. I think a full accounting is important so people understand the trade-offs and impacts. While we’ve made great progress cleaning the grid, we’ve got a long way to go, and additional rapid improvements are not a given!

  8. The only conclusion I get from this article is that Minnesota’s electricity production sucks…. At least with an electric cars you have options.

  9. Pingback: How many battery-electric vehicles are there in Minnesota? | Net Density

  10. Pingback: What are the upstream impacts of internal combustion and electric vehicles? | Net Density

  11. Pingback: Upstream Impacts of Electric Vehicles Depend on Kicking Coal |

  12. Pingback: Sunday Summary – October 4, 2015 |

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s