Where to start … cleaning-up the haze created by VW’s “clean” diesels

Post by Ethan Sprague, Kevala, Sales and BD

In September, the ICCT released research that launched an EPA investigation into Volkswagen’s “clean diesel” vehicle fleet. It has been shown that VW installed “defeat devices” to enable the cars to game the emissions testing system. In doing so, the cars actually released emissions including NOx and particulate matter at levels up to 40 times the permitted levels. The defeat device software is on VW Group models from as early as model year 2009 covering roughly 482,000 vehicles in the United States. The software works by changing the operating characteristics of the engine based on when it detects the conditions one would expect while the vehicle is being tested for emissions; in essence bringing it into compliance only when it thinks it’s being tested.

There are many impacts on multiple levels that raise several questions:
– How much damage is being done
– How to compensate for local and environmental damage
– How or if to remove the vehicles from the road.
– With over 68,000 of these vehicles registered in California – how do we begin to assess remedying this situation?

Around the office we’ve been discussing this issue extensively, in part because I purchased one of the non-compliant VW diesel’s over a year ago deciding, after extensive research, that “clean diesel” was the best option. But we’ve also been discussing it because Kevala itself is based on the premise that better decisions will be made if there is broader access to information which models and quantifies the impact of an action and who bears the costs or benefit from it.

Unsurprisingly, I’ve been the recipient of a lot of good-natured ribbing at work and at home. My kids, having witnessed my extensive research in finding a “clean” car that wasn’t a gas-guzzler like my wife’s crossover SUV, have now nicknamed my car “Poison”. At the time of our purchase decision we had seriously considered an electric vehicle but in light of this “Clean Diesel” option, combined with our EV range anxiety and lack of quick-charge stations, we made what turned out to be an ironically regrettable decision. A decision I am reminded of daily when I go to pick up my kids from school.

Non-Compliant Volkswagens Near Me

Non-Compliant Volkswagens Near Me

Talk about a defeat device!

At Kevala we decided to do what we do. We gathered and layered transparent and verifiable data to see where the non-compliant VWs were located and where these unwitting polluters were likely to be driving.

You can check it out here: https://keva.la/vw.

The “where” part is key and is why mapping information is so powerful. If we can see the damage being done we are less likely to allow it. If it happens to others and is hard to identify, we question whether or not it is happening at all. Mapping the impacts is a strong vaccine to combat our out of sight, out mind mentality.

Sunlight, we say, is the best disinfectant.

In approaching this issue we decided to first identify the problem and then look at solutions that address that problem. The first step is to correlate the typical commute patterns of non-compliant VW diesels with the areas in California with higher concentrations of pollution and poverty. We’ve done this by first mapping where VW diesel owners are located and where they typically commute. We then analyze what kinds of communities those commuters drive through.

Quantifying how to address the harm is more complicated. We decided that knowing whether there were (or were not) EV charging stations, was a good place to start. If emissions from vehicles are the problem, looking at non-emitting vehicles might be a solution. Looking at electric vehicle infrastructure is a good first pass in determining where there may be structural limits to increased penetration of electric vehicles. Lastly we included calculations on typical asthma cases and asthma costs each year. Select a county or political district to get this calculation. The costs of asthma are staggering. The numbers are not VW specific, they’re just describing the spectrum of costs associated with the harm.

EV Chargers and Commute Patterns from my ZIP

EV Chargers and Commute Patterns from my ZIP
(dark blue are areas of highest pollution and poverty)

We’re still crunching numbers over here, but on this public version of the tool, you can make some inferences about where the non-compliant VWs are being driven. We use route calculation software to map the commute patterns – simply click where you want to see the top commute routes to calculate and render them. To see the affected communities – those who are bearing the real costs of the increased emissions – select the Pollution and Poverty layer.

We have made this data available for people to explore in the hopes it will help inform the dialog around appropriate courses of action to address the harm caused by this situation. Perhaps those remedies will involve investing in more electric vehicle charging infrastructure, directly subsidizing EV vehicle purchases, or funding electric buses in affected communities. Maybe there will be some other inspired solution. If there is, expect us to model it.

In the interim, VW apparently isn’t going to fix my model until the end of 2016. I can’t sell the car and it would be morally improper to do so. I used to love driving it (I’ve heard the fix limits its power) and now every time I get into it and start the engine I cringe. I too just want the problem to go away, but we can’t drive away from it, we must drive into it, as the one thing we don’t have the luxury of is time. The impact just continues to get worse with every passing mile. 

We’d like to thank CalCEF (http://calcef.org/) for funding some of our data acquisition related to this project and Stop Fooling California (http://stopfoolingca.org/) for funding the integration of commute data. Neither entity was involved in the underlying analysis, but they provided us with the ability to make this research freely available to the public.

If you’ve got some ideas and want us to model more solutions we’d be happy to hear from you (email us at ideas@kevalaanalytics.com).

About the map

Non-compliant VWs

The orange dots represent the 68,000 non-compliant VWs in the zip codes to which they are registered (scattered randomly). We obtained the VW data from Hedges and Company, an automotive industry analytics firm.

Pollution and Poverty

Next we added a Pollution and Poverty Index, called the California Enviro Score (see http://oehha.ca.gov/ej/ces2.html for more information). This is a state metric correlating where pollution and poverty are the highest. The deeper purple areas have the highest levels of both pollution and poverty.

Electrical Vehicle (EV) Charging Stations

The green dots are EV charging stations. A subset of those charging stations were a result of recently a settlement with NRG related to the ill gotten gains of a company NRG acquired after the California energy crisis called Dynegy. The settlement directed NRG to invest a specified amount of money ($100 million) in electric vehicle charging infrastructure. The result of this settlement is NRG’s eVgo commercial vehicle charging business. Unlike Dynegy’s actions during the energy crisis, which equally affected all Californians, the effects of VW’s efforts are actually localized. NOx and particulate emissions are localized pollutants: Where those non-compliant VWs are driven determines who is harmed by them. Accordingly we should expect better localization of chargers than eVgo in light of the localized harm associated with VWs

Commute Patterns

If you click on a zip code in the map it shows you where the top 5 zip codes destinations (commute pattern) the from the original zip code. In this way you can get a sense of where the VWs are likely to be driving. This is based on routing data developed from the 2000 Census (see http://www.transtats.bts.gov/Tables.asp?DB_ID=630 for the full data set).

Searching by County and Political District

You can enter a county of state political district into the window. The map focuses and shows you the selected area. In the data bar on the bottom is counts the expected cases of asthma and expected costs. The count of non-compliant VWs registered to zip codes in that district are also estimated.

Follow us on Twitter: @KevalaAnalytics