Establishing a low-cost air quality sensor network in Kigali, Rwanda with publicly accessible data empowering the public and policies.
To attack the present and growing problem of air pollution in East Africa, we plan to introduce an initial network of five portable air quality sensors in areas of Kigali (the capitol city of Rwanda) representative of different traffic patterns and air pollution sources. Beyond contributing to climate change, air pollution accounts for 1 in 8 deaths globally, more than malaria and HIV together (1,2). This project will inform government policies and educate civilians on the impact of vehicles on air quality and greenhouse gas emissions, thus ensuring a less polluting, healthier urbanization for Rwanda. Kigali is a perfect choice to serve as a pilot city for networked low-cost air quality monitoring and this work will fill a much-needed data gap on pollution emission measurements in East Africa.
Current model results suggest that African countries could be major global emitters of pollutants in 50 years, if growth and development trends continue without emission regulations(3). While Rwanda remains one of the least urbanized African countries, the government of Rwanda is aiming for 35% urbanization by the end of 2020 as a means of increasing economic development(4). This urbanization will change vehicular emissions in unknown ways, and the goal of this proposal is to study the variety of ways that development will impact the climate, air quality and human health.
With rapid urbanization come concerns about the environmental and climate change impact of that growth. Rwanda has demonstrated commitment to green practices by implementing a Green Growth Strategy in 2011(5). Furthermore, new vehicle emission regulations were enacted in January 2015. However, little data is available on the ambient air quality levels to inform policy and determine the impact of new regulations or of increased development as of now. Use and development of these novel, low-cost air quality sensors will increase pollution data accessibility in countries that need it most.
What actions do you propose?
We propose a project combining air quality measurements using novel low-cost air quality (AQ) sensors (including CO2 measurements), public outreach, data delivery to policy makers, and in-country capacity building to take place in Kigali, Rwanda as part of an effort to reduce vehicular emissions and improve air quality. Modeling work has suggested that by 2030 Africa could be responsible for up to 50% of global organic carbon emissions unless new emission regulations are enacted(3). Rwanda, while one of the less urbanized countries in Africa, is developing rapidly and Rwanda’s government has enacted new vehicular emissions regulations (2015). However, little ambient air quality data is available. Measurements taken in Feb. 2011 (dry season) in Kigali found daily average levels of PM10 (particulate matter 10 microns in diameter or less) to be around 1000 ug m-3, a factor of 20 above the average daily limit set by the World Health Organization(6) (50 ug m-3). Increased development will likely make this existing problem even worse. Another study found that the major sources of PM10 in Kigali are older diesel vehicles and cook fires, both sources of CO2 gas and black carbon aerosol(7), which is known to contribute to climate change (8).If this project is successful, it could serve as a model to introduce more ambient AQ measurements in the developing world. To that end, we propose to:
I. Establish a network of low-cost air quality sensors in 5 locations
Air quality sensors will be purchased from Aerodyne Research Inc. to measure ambient concentrations of CO, CO2, NO, NO2, and particulate matter less than 2.5 microns (PM2.5). Kigali will serve as the first African city to implement this novel sensor. Five locations will be used, including one measurement site outside of the city (Mt. Mugogo) and four inside the city of Kigali. The choice of these different sites with different emission sources and traffic patterns, will help ascertain the most important factor in reducing pollutants.
1). Mt. Mugogo: Currently, MIT and the government of Rwanda are collaborating to build the Rwanda Climate Observatory Project, currently located on the summit of Mt. Mugogo, a 2.5 km peak near Musanze. This location will serve both as a background site, collecting information on the baseline of pollution in Rwanda away from major anthropogenic activities, and as a calibration station for the low-cost AQ sensors. State-of-the-art instruments measuring CO/CO2/CH4 (Picarro Analyzer), Black Carbon aerosol (Magee Scientific 7-wavelength aethalometer), and ozone (Teledyne) are located at the station. By co-locating the low-cost AQ sensors alongside reference instruments for rotating periods of time, these sensors can be calibrated and data quality can be assured.
2). The Office, Rwanda. The Office is a four-story coworking space located in the Kiyovu area of Kigali, a wealthier residential area away from major traffic or bus routes. Buildings in this area tend to be lower to the ground, avoiding urban canyon effects trapping local pollution, and streets are lined with trees. The owner of The Office is interested in purchasing AQ monitoring equipment, which will allow another opportunity for instrument intercomparison.
3). The METEO Rwanda office in Nyambugogo. Nyambugogo is a majority-local neighborhood in Kigali, and the largest bus depot in Kigali. Traffic is heavy with frequent gridlock. An adjacent weather station will provide meteorological data to compare with pollutant data.
4). Downtown Kigali. Kigali Town is experiencing rapid development. Mototaxis, buses, and personal vehicles are concentrated in this area.
5). The College of Science and Technology (CS&T) campus. The campus is near the center of town but removed from direct traffic emissions. The Climate Observatory Project has their Kigali office on this campus, soon to be live streaming data from the Observatory. Placing an AQ sensor here will allow a comparison, by visitors, of pollution in Kigali and the countryside, driving home the importance of green growth in Rwanda’s cities.
**As technology develops, mobile units can be placed on motos equipped with gps to track pollution around the city and capture traffic plumes in motion.
II. Monitor sites for emission sources/traffic patterns
Periodic visits to each site will be performed to collect information on types of traffic, speed of traffic, presence of idling vehicles, presence of other emission sources (e.g., cook fires), and other factors that may affect pollution concentration in the immediate area. General and unusual emission sources can thus be connected to local levels of pollutants.
III. Testing of low-cost sensor technology, data quality control
While electrochemical (EC) sensors have been commercially available for many years (primarily monitoring pollutant concentrations in industrial workplace environments), use of EC sensors for ambient AQ monitoring is a relatively new area of research, enabled by recent improvements in the sensitivity of the devices (D.L. = 2-5 ppb). Accuracy is not yet at regulatory-level, a major drawback of these devices, and understanding of the response of EC sensors to RH and T is necessary for data interpretation. The market for low-cost AQ sensors is projected to dramatically increase (~5%/year) over the next 5 years resulting in over 400 million low-cost gas sensors in use by 2020. Combining multiple sensors into a compact, robust, field-deployable package allows simultaneous assessment of multiple pollutant concentrations, which can inform source attributions (biomass, fossil fuel, or other combustion sources). The air quality sensor systems developed by the Aerodyne-MIT team are a relatively new innovation in atmospheric research and field verification in Kigali will serve as an important demonstration of the quantitative potential of the devices. Packaged in 3D printed weather-proof plastic, these units are robust and able to be left in their monitoring location for several months to a year. To ensure the accuracy of the sensor systems spanning the environmental conditions of Kigali, short-term co-location experiments alongside reference instruments will be conducted to routinely (once/year, based on results from the MIT CLAIRITY network) check the sensor calibrations. Data inter-comparison will allow assessment of the reliability and performance of these new low-cost research tools.
Studying low-cost AQ monitor performance will increase possibilities for air pollution measurements in developing countries, enhancing knowledge in this data-poor area. Increased access to AQ measurements in developing nations will assist policy makers, modelers, and other researchers in fully understanding the impact of development on climate change. The goal if this research is to make low-cost AQ sensors a viable, accurate option for regulatory purposes.
IV. Deliver data to policy makers in Rwanda
Data obtained from the instruments will be analyzed, quality-controlled, and delivered to key members of the Rwandan Environmental Management Agency (REMA), who can enact new policies. These data can then be used to inform future policy proposals and speak to the effectiveness of current regulations on reducing vehicular emissions.
Major concerns include the prevalence of older diesel vehicles on the road, the use of inefficient cookstoves, agricultural burning and emissions (e.g., methane from rice fields), gridlock, and the large concentration of motos within Kigali. The contribution of cookstoves vs vehicular emissions to AQ is controversial, and using the CO to NOx (NO+NO2) ratios, measured by the sensors, will theoretically allow source apportionment of these emissions (as vehicles and cookfires emit different ratios of CO/NOx). These data will help policy makers focus on the heaviest emitters. Locating the sensor next to major roadways will allow pollution concentrations over the day to be observed and the 10 second time resolution of the sensors will allow individual emission plumes from passing vehicles to be captured. For example, public transport in Kigali is via older diesel minivans, known to be heavy emitters of pollutants with low fuel efficiency. Recently, newer buses have been purchased. By comparing pollution plumes from newer public buses versus older minivans, data can be collected to demonstrate the efficacy of introducing newer buses into rotation to reduce emissions in Kigali. This could be an effective way to demonstrate the need for cleaner buses, and encourage the government to reduce the tax barrier to investment in transport.
Other examples of potential policy are no burn days, increased bus coverage, lower building height to reduce urban-canyon effects, and an enforcement of a no-idling policy in bus parks. Mobile units on motos will help the government target areas for improved traffic flow (e.g., traffic light sensors, wider roads). Rwanda has recently adopted a low-sulfur fuel policy, but requiring higher purity fuel for vehicular use could also further emission reductions, and carpool networks can be set up and encouraged by tax breaks.
V. Public Outreach
A component of this project will involve public outreach. This will be achieved by a website of publicly archived data, tweeting of averaged air pollution in the city (morning and evening rush hours, noon, late evening), and real-time data displays of AQ at the Rwanda Climate Observatory Laboratory in the College of Science and Technology. The goal of the outreach will be to increase awareness of the health dangers, relation to climate change, and causes of air pollution, with a specific focus on driving behaviors (appropriate speed, reducing idling, good traffic flow, carpooling) to curb emissions and a discussion of why emissions testing is helpful to the greater community.
VI. Capacity Building
Four technicians have already been hired and trained by the Rwandan government and MIT for work at the Rwanda Climate Observatory, and would be trained on the proper use of these instruments. The University of Rwanda is also developing a Climate Science Master’s Program, to begin in September, 2015. By the fall of 2016, students will be ready will begin the internship portion of their degree program. These students could be involved in this initiative and take newfound knowledge of AQ measurements and vehicular emissions into different sectors of Rwanda after graduation.
VII. International Data Delivery
Air quality in Africa is still an open question in the international community(9) and injecting these data into a global platform will speak to this question. The website OpenAQ.org, currently in its development phase, plans to host both standard AQ measurements (historical and as close to real-time as possible) and low-cost sensor measurements in an easily downloadable format in the near future, and Rwanda’s measurements could give a much-needed snapshot of pollution in at least one major East African city. These data, if adequately quality controlled, will be useful for global atmospheric modeling and published in journal articles for international peer-reviewed journals.
Capturing and publicizing air quality data in a developing country as new regulations are enforced, along with a public outreach program, and delivering data to government policy makers would help regulations make the maximum impact on the future of Rwanda’s air quality. Air quality monitoring during this nascent stage of development and regulation will allow real-time tracking of the changing Rwandan atmosphere and the effectiveness of environmental policies. Developing countries often cannot afford standard AQ monitoring equipment (at $10,000 plus per instrument) and lack the capacity to use this equipment, yet often have high concentrations of pollutants, making them an ideal place to use low-cost monitoring equipment and build capacity on the use of this equipment and the understanding of emissions measurements.
Who will take these actions?
Dr. Langley DeWitt, Station Chief Scientist of the Rwanda Climate Observatory, will lead the acquisition, calibration, placement, and data quality control of the low-cost air quality monitoring sensors in Kigali. Currently, Langley is working with MIT and the Ministry of Education (MINEDUC) in Rwanda to establish the Climate Observatory, train technicians, and facilitate collaborations between the station and the University of Rwanda Climate Science Master’s Program. She is based primarily in Kigali and has contacts in MINEDUC and other ministries in Rwanda (REMA, Meteo Rwanda).
The data will be shared freely with REMA Rwanda, Meteo Rwanda, and MINEDUC. These government ministries are involved in policies and planning to mitigate air pollution and climate change impacts, and having access to ambient air quality information will greatly enhance their platforms as they push for more government regulation on greenhouse gas and particulate emissions.
Four Rwandan technicians, already trained to work at the Climate Observatory, will perform day-to-day instrument maintenance. A new Master’s program in Climate Science will start the next academic year at the University of Rwanda, and students and professors will have access to the data and the instruments for projects and hands-on lectures.
Dr. Eben Cross (Senior Scientist at Aerodyne Research) will support Dr. DeWitt throughout the project providing hardware and software support for the low-cost AQ sensor systems.
Where will these actions be taken?
These actions will happen in Rwanda, a developing nation in East Africa. Specifically, they will take place in Kigali, the rapidly growing capital city of this small, mountainous country. Little data on ambient pollution is available in East Africa and Rwanda, with a stable government, reasonable infrastructure, and a strong cell network is a good place to perform this type of research. The majority of Rwandans rely on agriculture for their work, and thus are particularly concerned about the impacts of climate change. Rwanda has also recently (2011) implemented a Green Growth strategy aimed at growing Rwanda with the least pollution and greenhouse gas emissions possible(5). MIT and the government of Rwanda already have a well-established Memorandum of Understanding (MoU), allowing MIT researchers associated with the Rwanda Climate Observatory to perform research within Rwanda and easily facilitating the sharing of data between MIT and the relevant Rwandan authorities and researchers (who will have priority on the data).
The lack of information on air quality in developing nations hinders efforts to enact and enforce vehicular emission regulations, thus creating a lag between development and air quality regulations that can lead to rapid increases in pollution (such as that being experienced now in China). Choosing Rwanda, a country in the early stages of development, as the site for an air quality sensor network will allow real-time data to be collected on this growing and changing nation as it develops and regulates air pollution emissions.
How much will emissions be reduced or sequestered vs. business as usual levels?
In Jan. 2015 Rwanda introduced new regulations on vehicular emissions. All vehicles are required to be equipped with a catalytic converter, motorcycles imported for commercial purposes (i.e., taxis) must be new, those for personal use <5 years old, and vehicles will be tested for emissions once/year. A recent paper(3) used growth and current emission factors to compare Africa’s projected emissions in 2030, with no changes in current regulations and with emissions after the implementation of emission regulations. They calculated a 1.6 and 2.0 x reduction in organic carbon and black carbon projected emission with reduced emission factors. Better emission standards for vehicles in Rwanda now could have a large impact on air quality in the future; however, without data to observe the effects of regulations on air quality, increasing the stringency of regulations and continuing to budget for emissions checks could be hard to justify.
What are other key benefits?
In addition to greenhouse gases, vehicles also emit particulate matter, particularly older diesel vehicles like those prevalent in Rwanda. One in every eight deaths is associated with pollution(1) and black carbon aerosol, associated with wood burning and diesel emissions, is known to have a positive (warming) radiative forcing on the atmosphere (10). Reducing particulate matter from vehicles will remove these negative health and climate effects from the city of Kigali.
Three-stone cook fires are abundant in Rwanda and are significant emitters of particulate matter and CO. Measuring air quality in areas with (lower-income, local neighborhoods) and without (wealthy neighborhoods) traditional cook fires will be another benefit to this study.
More testing on low-cost air quality monitors, a by-product of this study, will enhance their use globally and perhaps open air pollution measurements to developing countries with high pollution and little regulation or data.
What are the proposal’s costs?
The purchase of the air quality monitoring systems will cost around $10,000. Eventually a calibration station for Kigali, including standard instruments measuring CO/CO2, NOx, and PM2.5 will need to be purchased, with a cost of around $40,000. This station will also function as a ‘supersite’ of robust air quality measurements. For now, the present instrumentation at the Mugogo Climate Station will suffice for general calibration and data comparison.
The cost to enforce the new vehicular emission regulations is not trivial, nor is compliance, so future governments must make the choice to invest in environmental policy. This data will hopefully convince policy makers of the necessity of these costs. New motorcycles are significantly more expensive than used, a few hundred to a few thousand dollars more, depending on brand and year. If less mototaxis exist in the cities, due to higher cost barriers to entry, standard taxis cost 10x as much and may be too expensive for the majority of Rwanda’s citizens. The new, larger buses in Kigali are slightly more expensive—250 RWF per ride versus 200 RWF—and bus repair will be costly initially, as there are less of these buses in rotation now. Residents will need to tune their vehicles regularly, or purchase newer cars, to meet emission standards.
·Low-cost air quality monitors will be purchased and placed in Kigali.
·Instrument intercomparisons will be performed.
·Monitors will be networked to the Rwanda 3G network and initial data will be collected, quality controlled, analyzed.
·Achieve record of Kigali air quality as it develops and as new regulations are enforced over time.
·Additional air quality monitors to be purchased and placed in other major cities, along with mobile units on motos.
·Air quality monitors will have undergone extensive testing, quality control, and increased in use to develop an international scientific community of users, of which Rwanda will be a part. Widespread use of accurate low-cost monitoring equipment and building the capacity to use this equipment will be transformative in emission regulation and monitoring in developing countries.
·New cookstove initiative projects will be more complete, allowing a comparison of before/after ambient air quality to be observed.
· Development of more policy to guide Rwanda’s growth. Future policy, depending on initial experimental results, to regulate emissions could include 1) stricter emission regulations 2) regulations on the type of vehicle allowed to be imported into Rwanda (e.g., only vehicles with carburetors and particle filters) 3) purchase of newer, more-efficient and larger capacity buses and increased bus coverage 4) introduction of measures to control traffic flow (e.g., sensor-based traffic lights, wider streets) 5) regulations on cookstoves for purchase 6) bike lanes on major roads and mapped bike routes around the city 7) a push for the construction of energy-efficient buildings and use of alternative energy, among others.
· As Rwanda develops, it will develop in a greener, more sustainable way, avoiding the problems of heavy greenhouse gas emissions and poor air quality in cities that have plagued other countries in development.
No other related proposals from our team.
1.WHO. Health Effects of Particulate Matter: Policy implications for countries in eastern Europe, Caucasus and central Asia. World Heal. Organ. 15 (2013). at
2.Boucher, O. et al. in Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (eds. Stocker, T. F. et al.) 571–657 (Cambridge University Press, 2013).
3.Liousse, C., Assamoi, E., Criqui, P., Granier, C. & Rosset, R. Explosive growth in African combustion emissions from 2005 to 2030. Environ. Res. Lett. 9, 035003 (2014).
4.Urbanization and Rural Settlement Sector: Ministry of Infrastructure. at
5.Rwanda Green Growth Strategy 18 Nov 2011 http://www.uncsd2012.org/content/documents/364Rwanda-Green-Growth-Strategy-FINAL.pdf
6.Nsengimana, H., Bizimana, J. Pi. & Sezirahiga, Y. A Study on Air Pollution in Rwanda With Reference To Kigali City and Vehicular Emissions. (2011).
7.Henninger, S. M. When Air Quality Becomes Deleterious—A Case Study for Kigali, Rwanda.J. Environ. Prot. (Irvine,. Calif). 04, 1–7 (2013).
8.France, O. B., France, C. G., Germany, C. H. & Uk, A. J. IPCC Report: Clouds and Aerosols. (2014).
9. Galbraith, K. Measuring Africa’s Air Pollution. The New York Times (2014). at
10.Bond, T. C. et al. Bounding the role of black carbon in the climate system: A scientific assessment. J. Geophys. Res. Atmos. 118, 5380–5552 (2013).