Getting a Measure on Particulates in Stuttgart

There’s a big to-do going on right now in Germany over particulate-matter air pollution. Stuttgart, Germany’s “motor city” and one of Dante’s seven circles of Hell during rush hour, had the nation’s first-ever air pollution alert last year. Cities are considering banning older diesel cars outright. So far, Stuttgart’s no-driving days have been voluntary, and the change of the seasons has helped a lot as well. But that doesn’t mean there’s not a problem.

But how big is the issue? And where is it localized? Or is particulate pollution localized at all? These questions would benefit from a distributed network of particulate sensors, and the OK Lab in Stuttgart has put together a simple project(translated here) to get a lot of networked sensors out into the wild, on the cheap.

The basic build is an ESP8266 with an SDS011 particulate sensor attached, with a temperature and humidity sensor if you’re feeling fancy. The suggested housing is very clever: two 90° PVC pipe segments to keep the rain out but let the dust in through a small pipe. The firmware that they supply takes care of getting the device online through your home WiFi. Once you have it running, shoot them an e-mail and you’re online. If you want help, swing by the shackspace.

We love these sort of aggregated, citizen-science monitoring projects — especially when they’re designed so that the buy-in is low, both in terms of money spent and difficulty of getting your sensor online. This effort reminds us of Blitzortung, this radiation-monitoring network, or of the 2014 Hackaday-Prize-Winning SATNOGS. While we understand the need for expensive and calibrated equipment, it’s also interesting to see how far one can get with many many more cheap devices.

48 thoughts on “Getting a Measure on Particulates in Stuttgart

    1. The ESP8266 is programmed either via flashing the bin you probably found or by compiling & uploading the source with the Arduino IDE. The .ino file can be found in the following repo along with the description of the dependencies:

      The configuration (wifi settings, upload targets, used sensors) can be done once it is flashed by opening the ESP’s web interface in hotspot mode or by changing the config file. See here for more details:

  1. What’s interesting is that in Belgium they did a study on fine dust pollution and they found that speed bumps are a huge contributor since each time a car goes over one it seems to shake loose a huge amount of it.
    The interesting part is that this study was completely ignored by all the euro cities with fine dust issues.

    Meanwhile we already invented speed bumps that actually go flat if you drive at the slow speed expected and only are bumps if you speed over them. So the solution is available.

      1. Forgot to add, how much particulate from various volcanic eruptions and earthquakes are taken into account, or are they just ignored? Strikes me that with the rise in geologic activity and changing wind patterns a lot more particles are spewing into the air from sources other than cars.

          1. Annie
            Along with lava, ash clouds are ejected in the form of volcanic column plumes. Pyroclastic flow is hot gas and rock tephra that is expelled at high speeds and at high temperature during volcanism. Ash and other volatiles from pyroclastic flow can extend up to 50 km into the atmosphere at a speed of 700 km/h (Bryant, 2005). This fast moving cloud can reach up to temperatures of around 1,000ᵒC. Water, within the magma, provides the explosive potential of an eruption. Figure 2 illustrates the explosive potential of eruptions; as explosive potential increases so does the concentration of silica, oxygen, and of course water. Outgassed ash clouds extend up into the upper troposphere and into the lower stratosphere, also known as the UTLS; the height at which the volcanic plume reaches depends on the pressure within the volcanoes conduit tube and, as mentioned previously, depends on the amount of silica and oxygen within the magma (Krotkov, 2010). The extent at which the volcanic ash cloud extends at the time of eruption depends on particulate size, wind speed and direction, and eruption type. With the increase in particulate size there is a decrease in the distance of which the ash cloud disperses; smaller particulates allow for the ash cloud to extend further up and around the earth’s atmosphere. Wind speed affects how far and at what rate the ash cloud extends while wind direction influences where the ash cloud is guided. Eruption type, as seen in Figure 2, is the initial governing factor of what drives the establishment and the movement of the ash cloud (Lamb, 2008). Hawaiian, Strombolian, Vulcanian, Plilian, Lava Dome (Pelean eruption), and Surtseyan (Icelandic eruption) are eruption types (Schmincke, 2004).
            A Hawaiian eruption is characterized by it’s fluid basaltic lava that is expelled into the air from jets derived from fissure vents. In this eruption type due to the expulsion of lava through jets, the duration of eruption time is extensive; jets offer a steady continual lava flow allowing for drawn out eruption time. A Hawaiian volcano eruption time can last from anywhere from 1 hour to a couple day. The Kilauea volcano in Hawaii is characterized as having a Hawaiian eruption type. Strombolian eruption is another eruption type; characterized as having a bursting like expulsion of lava, a strombolian eruption is usually basaltic or basaltic andesite in origin. Unlike the Hawaiian eruption time that has a constant releases of lava, the Strombolian expels lava at minute intervals. This eruption time also can reach up to hundreds of meters and eject lava and pyroclastic flow onto earth’s surface by spatters. Vulcanian eruption occurs when violent, short, and, minimal amounts of lava is released during a single eruption event; ash clouds also are highly explosive reaching up to 350 meters per minute. A lava dome and an umbrella like ash cloud is produced. The most extensive and violent eruption type is the Plinian; ash columns have the ability to reach up to 50 km and speeds of about 100 meters per second. Ash clouds can disperse out to a distance of around 100 to 1,000 miles. A Plinian eruption has a ash column similar to a mushroom. These eruptions are tremendously destructive, ash particulates and lava bombs extend miles away from the volcano; the 1980 eruption of Mount St. Helens is a Plinian eruption. Icelandic eruption also known as the surtseyan eruption originates underwater; this eruption time is characterized as being hydromagmatic where magma or lava becomes explosive when it comes in contact with water. The Surtsey volcano is one example of an Icelandic eruption type volcano. (Ball, 2005

            The eruption of Mount St. Helens in 1980 saw a release of a massive plume that covered an area of 600² km. The massive gas cloud was laden with sulphur dioxide; figure 8 shows the concentration of sulphur dioxide in tones per day releases from Mount St. Helen outgassed from 1980 to 2005. Up to 3750 tons per day of sulphur dioxide was released into the atmosphere. The prevailing winds moved 520 million tons of ash eastward across the United States; the ash cloud migrated across the United States in three days and reached around the world in 15 days. The eruption of Mount St. Helens in 1980 resulted in an international pollution event; figure 9 is a photograph taken from Ephrata, Washington that shows the gas cloud produced by Mount St. Helens.(USGS, 2008 and 2012)

            Volcanism is a natural uncontrollable source of pollution, unlike anthropogenic derived pollution, we cannot control or put a cap on the amount of pollution expelled.

            Here are some highlights very fine book about the 1883 Krakatau eruption by Tom Simkin and Richard Fiske (Simkin, T., and Fiske, R.S., Krakatau 1883: The volcanic eruption and its effects: Smithsonian Institution Press: Washington, D.C., 464 p.) that should give you and idea about how far ash can travel during a large eruption.

            Ash fell on Singapore 840 km to the north, Cocos (Keeling) Island 1155 km to the SW, and ships as far as 6076 km west-northwest. Darkness covered the Sunda Straits from 11 a.m. on the 27th until dawn the next day.
            Blue and green suns were observed as fine ash and aerosol, erupted perhaps 50 km into the stratosphere, circled the equator in 13 days.
            Three months after the eruption these products had spread to higher latitudes causing such vivid red sunset afterglow that fire engines were called out in New York, Poughkeepsie, and New Haven to quench the apparent conflagration. Unusual sunsets continued for 3 years.
            The volcanic dust veil that created such spectacular atmospheric effects also acted as a solar radiation filter, lowering global temperatures as much as 1.2 degree C in the year after the eruption. Temperatures did not return to normal until 1888.

  2. “We love these sort of aggregated, citizen-science monitoring projects”
    Good advocate for WEST COAST radiation monitoring because after Fukushima damn near ALL gov detectors were decommissioned.
    Fukushima is oh exponentially worse than Chernobyl. Many food items as far as England were banned for sale until just recently due to contamination. And I believe that German Wild Boars are still banned for consumption due to build up from their food source.

    1. The Boar-part is true for some parts of Germany (south), but this is due to Tschernobyl, not Fukushima. see

      [quote]Seit der Atomkatastrophe von Tschernobyl muss in südlichen Regionen der Bundesrepublik auch auf radioaktive Belastung durch 137Cs untersucht werden, sofern das Fleisch in den Handel verbracht wird.[29] Laut einem Bericht des Telegraph ist 2014 die Strahlenbelastung der Wildschweine in Sachsen immer noch so hoch, dass 297 von 752 erlegten Tieren den Grenzwert von 600 Bq/kg überschritten und vernichtet werden mussten.[30][/quote]

      1. Dude
        Fukushima is FAR worse than Chernobyl and continues unchecked today.
        I was linking the TRAVEL distance of radiation from Chernobyl and the LENGTH of effect.
        Particles enter the food chain and accumulate. Animals eat plants which are radioactive, people eat animals.
        Much like the limitations on eating Tuna due to Mercury poisoning.
        Too much tuna, too much mercury

        For more than a decade, federal agencies have said that women of childbearing age and young children should limit their weekly consumption of albacore (white) tuna. That’s because it contains three times more mercury, on average, than canned light tuna; even just a few sandwiches can expose some people to too much. Now a federal committee is suggesting that the warning be eliminated, a move that Consumer Reports’ experts strongly oppose.

        Mercury can damage the brain and nervous system, especially when exposure occurs in the womb. That’s why we recommend that pregnant women not eat tuna and any other high-mercury fish, such as shark and swordfish. High-mercury seafood can pose health risks to other vulnerable groups as well. So we also recommend that young children, women of childbearing age, and anyone who eats 24 ounces or more per week of any fish limit their tuna consumption, especially those kinds that are high in mercury, such as yellow­fin and other species used in sushi.

        The importance of that advice was underscored earlier this year by a study that found that mercury levels in yellowfin tuna had increased at an annual rate of almost 4 percent from 1998 through 2008. Rising mercury levels in oceans because of pollution from coal-fired power plants and other industrial sources are to blame, the study suggested.
        Take into account any other fish you eat when deciding whether it is safe for you to eat more tuna in any given week or month, eating no more than a total of 12.5 or 14.5 ounces of low-mercury fish or seafood per week, and eating less if you consume fish that are higher in mercury. Avoid king mackerel, shark, tilefish, orange roughy, marlin and swordfish because they are all very high in mercury. Don’t consume more than 18 ounces per month of Chilean sea bass, grouper, bluefish, an Spanish or gulf mackerel, as these are also high in mercury. If you eat more than the recommended amount of fish one week, you should eat less fish the next week or two to keep your mercury levels from getting too high, recommends the U.S. Food and Drug Administration, because it can take more than a year for the mercury to leave your system.

        1. calm down… I only repeated what is written in the german wikipedia and clarified that this problem exists for a long time already (since Tschernobyl). I agree that Fukushima certainly did not make things better… :-(

          1. some guy
            The problem is stories pass down the rabbit hole fairly rapidly these days. It doesn’t help that Google censors not only for promotional purposes but also for content they pick and choose.
            I’m East coast and while there may have been wind driven radiation spikes I’m more worried about how it becomes concentrated in the food supply. Especially fish I eat.
            Recently there was an article about lead in the water supply, inspired by Flint Mi. Of the counties tested one about 20 miles away in a different state had a 36% elevated Lead levels among children.
            Infrastructure in my area, well just about everywhere, at about the same time.
            So how many children suffer from elevated Lead where I live? If they don’t get tested no one knows.
            To GOV we are simply statistics.

    2. Wild boar is not banned in Bavaria, but they _are_ still being tested for radiation. Ground -> trees -> acorns -> boar-meat is a slow cycle, and there’s likely to be some excess radiation for a long while. Still, it’s not like there’s any shortage. Wild boar ragout is on the menu nearly everywhere here, and I’m not sad that it’s passed a radiation test.

      This is not to say that Chernobyl was anything other than a disaster.

  3. The reason people drive diesels in Germany is the price. 1 liter diesel is ~20 cents cheaper than the cheapest gasoline (E10). Even though the road tax for diesel is higher, it’s still more economical if you drive more than 10000 km per year.

    1. Wookhash
      Bought a diesel f350 superduty because if need arises it can be converted to biofuel.
      Some of the Ford 150’s can be converted to NATGAS.
      There are actually many NATGAS stations around so it’s almost as easy to find as a gas station.

    2. And still most people buy them even though they never go over 10000kms. Which makes the whole pollution thing even worse, just by the engines design. Any combustion engine is just absolultely filthy as long as it’s not up to working temperature. This is especially a big issue with diesels, because they’re a bit more efficient and also a bit more heavy. More mass to heat and less heat produced by the combustion process. Catalytic converters simply don’t work when they’re not up to temperature. Same case with DPF – Diesel Particulate Filters.

      1. Watched (most of) that video. A very good resource on the fine-particle detector. The best part is when he correlates his values with the official stats taken at a gov’t measurement site “10 km” away. They agree very well, both in absolute magnitude and in variation over time, which gives some confidence in these $25 modules doing what they’re supposed to.

    1. jarek319
      Sarah Connor: [voiceover] Watching John with the machine, it was suddenly so clear. The terminator, would never stop. It would never leave him, and it would never hurt him, never shout at him, or get drunk and hit him, or say it was too busy to spend time with him. It would always be there. And it would die, to protect him. Of all the would-be fathers who came and went over the years, this thing, this machine, was the only one who measured up. In an insane world, it was the sanest choice.

  4. I don’t get the “ban old cars” hullabaloo. The EU recently demanded all new diesel cars should have a “soot filter” so that instead of a cloud of large particles that get stuck in the upper airways and settle down quickly we get an even larger cloud of nano-particles that embed themselves deep in the lungs and float around forever. But hey, we don’t know what harm that does to us yet, so it must be better right? And since most measuring equipment doesn’t measure particels that small, they don’t exist, so problem solved!

    1. Problem with old cars – or old engines to be correct: NOx emissions. NOx gasses tend to react with all kinds of stuff to form nitric acid vapor and related particles. Which in turn can penetrate lung tissue, destroying it and in some cases causing premature death.

      The “soot filter” is actually called DPF for “Diesel Particulate Filter” and it’s there to filter out the fine particles. Now you might think all is fine and dandy with a filter for fine particles, right?

      No it isn’t. Because fine particles aren’t just caused by combustion. And cars aren’t the only thing emitting fine particles.

  5. Well, there is a reason why the cost of the professional PM2.5 and PM10 sensors, used in the certified air pollution metering stations, can by higher than 25K $ (you should also think about 1-2K $ per year for maintenance…). And the reason is – such 20-100$ (aliexpress prices) PM sensors cannot be a source of data for a real, scientific research. They show “something”, but this “something” is quickly going into “no reliable data” area. After few weeks/months readings are totally unreliable (even if they are somehow correlated to the data shown by real-deal equipment at the beginning).
    World Air Quality Index ( is making good work assessing the true value of cheap sensors as measurement instruments… e.g. this is the look of SDS011 measurement chamber after 6 months of work:
    And here you can also check how current readings from two SDS011 in compare the reference data: (Real Time Data chapter).
    There are many initiatives last months, similar to this Stuttgart project. The hard question is – what is in fact measured and what is real value of such project when readings are so unreliable?…

    1. At least here in germany, the pollution is meassured on a day time basis (whole pollution over one day and only PM10 is meassured). As far as I know, the SDS011 results are still very similar to the official data collected by expensive devices (like these: and they had the sensores tested by a university to check the error rate. But I don’t think there is a long time study yet. The oldest sensors are from october 2016.

      1. I am almost sure, that the ‘official’ measurements in Germany are far more complex that only PM10, especially in big cities like Stuttgart. All UE countries are subjects to conditions of ‘Directive 2008/50/EC of the European Parliament and of the Council of 21 May 2008 on ambient air quality and cleaner air for Europe’. According to this law EU countries are obligated to measure SO2, NO2/NOx, CO, PM2.5 and PM10, benzene and lead air pollutions in regular basis (what, when and in what measurement ‘fashion’ – depends on specific regulations and local circumstances).
        According to mentioned law, my metropolitan area (Poznan, Poland, about 600K people in city and about 300K in whole area) is obligated to have 2 certificated, automatic and one manual air pollution control stations (I am sure, that SO2, NO2/NOx, PM2.5 and PM10 are measured). I don’t think so that in Germany this is looking different.
        I am a bit familiar with air pollution and various gas sensors. There is a real problem with mid-price-range instruments, not only able to ‘measure something’, (well if correlated with certified readings, at least for some time), but also capable to be calibrated and maintained to keep readings reliable. In case of the PM measurements we have cheap, but not really long term reliable sensors (like these SDS011) and certified instruments with very high price tag – and nothing in between. I would not try to prove in long term a quality of measurements, pollution maps, simulations based on “aliexpress sensors”. Maybe such systems have some informational role, but unfortunately it can be disinformation in the same time.

        1. Thanks for your response. I was not aware of this EU law!
          Do you know any resources for long term tests on sensors like the SDS011? Would be nice to see how the error rate changes over time and to approximate how long one of those SDS011 can be used with a certain error rate.

          1. For now the only known to me source of valuable information about performance of cheap air pollution sensors is:
            Maybe this is not perfect source, but at least someone is trying to gather facts and evaluate these sensors in scientific way – that is why this is great work and far more interesting and valuable information than typical marketing-blurred statements from the various seller’s websites.

            Air pollution topic is very popular. There is much politics around and this also generated a market for commercial solutions (cheap ones in the first place). This is making the already difficult topic even more blurred.
            For example, I know two companies in Poland that are trying to sell they wireless solutions for air quality measuring, based on Plantower PMS 5003 (or similar) sensor – the same class like mentioned SDS011. At least in one case the company is advertising their system as solution for cities, public communities and administration authorities. This is very misleading in my opinion. If an official administration unit would purchase air quality measurement solution based on such unreliable source of information (“aliexpress sensors”), it can be very easy and officially accused of mismanagement of public funds… Public administration just cannot depend on such kind of measurements in their reports and decisions about environmental management. So we have a difficult question: “Well… so, who can? What is the value of such, possibly misleading information for society?”…

        2. ” In case of the PM measurements we have cheap, but not really long term reliable sensors (like these SDS011) and certified instruments with very high price tag – and nothing in between.”

          There is this PM2.5 instrument:

          cheap enough to send up on balloons knowing they will never get it back. Not sure what ‘cheap enough’ is though.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s