2015 Progress Report: Quantifying the Climate, Air Quality and Health Benefits of Improved Cookstoves: An Integrated Laboratory, Field and Modeling Study

EPA Grant Number: R835438
Title: Quantifying the Climate, Air Quality and Health Benefits of Improved Cookstoves: An Integrated Laboratory, Field and Modeling Study
Investigators: Volckens, John , Johnson, Michael , Peel, Jennifer , Pierce, Jeffrey , Robinson, Allen
Current Investigators: Volckens, John , Johnson, Michael , Peel, Jennifer , Pierce, Jeffrey , Robinson, Allen
Institution: Colorado State University , Berkeley Air Monitoring Group , Carnegie Mellon University
EPA Project Officer: Keating, Terry
Project Period: September 1, 2013 through August 31, 2016 (Extended to August 31, 2017)
Project Period Covered by this Report: September 1, 2014 through August 31,2015
Project Amount: $1,520,000
RFA: Measurements and Modeling for Quantifying Air Quality and Climatic Impacts of Residential Biomass or Coal Combustion for Cooking, Heating, and Lighting (2012) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Global Climate Change , Tribal Environmental Health Research , Climate Change , Air


The primary objectives of this EPA STAR project are to (1) develop a Global Cookstove Emissions Dataset using a suite of laboratory and field measurements; and (2) model the impacts from a set of feasible cookstove intervention scenarios (and associated emissions reductions) on global climate, air quality, and health.

Progress Summary:

Objective 1: Development of a Global Cookstove Emissions Dataset

Laboratory Work. Substantial time has been devoted to analysis and interpretation of data collected during the 2014 intensive laboratory campaign. The 2014 study resulted in a total of 70 emissions tests. The results of these tests are being prepared for inclusion in the global cookstoves emissions dataset. The data from the study also are being prepared for use in peer-reviewed journal publications. Three publications based upon the laboratory data collected currently are under development.

Emissions data from the newly-developed firepower test protocol tend to reproduce field emissions data much better than current test protocols (i.e., the water boil test). Parameterizations also have been developed using data from the firepower sweep test method. Multi-linear regression analysis has shown that 66-95 percent of the variance in PM2.5 is explained when modified combustion efficiency and firepower are used as predictors. Parameterizations developed for individual stove/fuel combinations also have shown that firepower can be used independently to predict emissions. Firepower predicts a large amount of the variance in black carbon emissions from a given stove/fuel combination. Future work will address whether the parameterizations developed in the lab hold under field conditions.

A major finding of the laboratory study was that cookstove gas and particle emissions vary strongly with cookstove operating condition, as hypothesized.

We conducted three size-selected burning experiments using European White Birch bark as the fuel and a rocket stove operating in flaming mode. We size selected, using a differential mobility analyzer, three mobility diameters: 140, 180, and 210 nm respectively, each corresponding to a different experiment. The aim of these experiments was to characterize absorption and scattering efficiencies (per unit mass) of fresh and aged (with alpha-pinene secondary organic aerosol (SOA)) cook-stove emissions. Aging was done in a smog chamber (2 m3) and stepwise condensation of SOA, from the ozonolysis of alpha-pinene (using UV lights), coated the fresh emissions, reaching organic to black carbon mass ratios (OA:BC) of up to eight. A suite of instruments were deployed including two photo-acoustic extinctiometers (PAXs) operating at wavelengths of 405 and 532 nm), a single particle soot photometer (SP2), and a scanning mobility particle sizer spectrometer. We measured an average fresh mass absorption cross section (MAC) of 10.18 (±1.19) m2/g and 7.73 (±0.59) m2/g at 405 and 532 nm respectively, which is in agreement with the range suggested by Bond and Bergstrom (2006). As we coated the emissions, absorption enhancements of up to two were measured indicating lensing due to the SOA coating, in agreement with previous laboratory studies using BC from a diesel generator (coated with SOA from alpha-pinene ozonolysis) and from an Ethylene flame (coated with dioctyl sebacete). The mass scattering cross section (MSC) increased initially as coating thickness increased, but started to decrease with further increase in coating.

A portable sampler has been built to characterize combustion emissions from residential cookstoves operated in the field. The sampler is capable of measuring real-time PM2.5 mass and black carbon concentration as well as particle size distributions from 10–400 nm. The sampler also measures continuous concentrations of carbon monoxide (CO), carbon dioxide (CO2), oxygen (O2), sulfur dioxide (SO2), and volatile organic compounds (VOCs), in addition to relative humidity, temperature, and pressure (the latter three at exhaust and ambient levels). The sampler also is configured for filter collection, which allows for offline analysis of PM (mass, organic and elemental carbon). All components are battery powered allowing for remote sampling without electricity. The portable sampler has been tested both at the Carnegie Mellon University smog chamber lab and the Colorado State University combustion lab. Evaluation tests were performed in unison with a suite of standard lab instrumentation while sampling from a variety of cook stoves and fuels. The comparison between standard lab instrumentation and data collected with the portable sampler is being compiled for a journal article that will help correlate data collected in the field using the field sampler with similar lab experiments.

The portable sampler showed good agreement when compared with lab-instrument measurements of PM, black carbon, gases, and aerosol size distributions.

Field Work. To date, three of the four field campaigns (designed to capture real-world cookstove emissions data) have been completed. Field campaigns have been completed in Dong Wei, China; La Esperanza, Honduras; and Kampala, Uganda. The portable emissions sampler was used to characterize cookstove emissions in approximately 10 homes per site. In total, over 110 hours of intensive emissions measurements have been collected. Emissions were measured continuously in each household with the intent of capturing a full use cycle. Normal daily use activities varied by country. In China, daily use activities were dominated by home heating tasks, although about half of the stoves also were used to heat water for drinking and supplemental cooking needs. In Honduras and Uganda the stoves were used nearly exclusively for preparing meals. In each home, flue gas temperatures also were measured for approximately two weeks following the intensive emission sampling. These data will provide estimates of daily firepower cycles. Finally, surface temperature loggers also were installed in additional homes to provide a larger, more representative sample of stove usage patterns. These loggers record stove temperature for approximately 30 days. Analysis of filter samples (gravimetric and thermal-optical) and processing of real-time data from these field campaigns is currently underway. In total, measurements have been taken across 78 households.

A structured review of the techniques and approaches used during the first three campaigns also is being conducted to ensure that the lessons learned from this field campaign are translated into practical improvements for the final field campaign scheduled to be conducted in India in January 2016. Coordination and discussions with local partners in India are currently on-going to plan the upcoming field campaign. The field campaign in India will follow the same general structure and protocols used to date but also will seek to fill any potential data gaps needed for the development of the emissions inventory and modeling efforts.

The field studies have provided critical data needed for establishing cookstove usage predictions. The field data also will allow the team to apply field adjustments to the data collected in the laboratory.

Objective 2. Global Climate Modeling

To quantify the uncertainties in the aerosol climate effects from biofuel and cookstove combustion, we performed a sensitivity study on the direct radiative effect (DRE) and aerosol indirect effect (AIE) to uncertainties in biofuel emissions and model processes. This work was published in August 2015 in Atmospheric Chemistry and Physics (Kodros, et al., 2015). Briefly, we found the global-mean DRE ranges from -0.02 to +0.06 W m-2 and the global-mean AIE ranges from +0.01 to -0.02 W m-2. The DRE is strongly dependent on optical properties and emission composition (BC to OA ratio), while the AIE is strongly dependent on total emission mass and emission size distribution. These results were discussed in further detail in last year’s report, but the work was published during this year’s reporting period.

Major findings from this work are that the climatic effects of biofuel aerosols are largely unconstrained in that the overall sign of the aerosol effects is unclear due to uncertainties in model inputs.

In addition to climate effects, we want to quantify the impacts of combustion PM on health. Open combustion of domestic waste, a source similar to cookstoves, is a potentially significant source of pollutants in developing countries; however, it is not currently included in many emissions inventories, and no one has yet determined the impact of this open combustion on health. Globally, open waste burning emits 9.2 Tg yr-1 of organic aerosol and 0.6 Tg yr-1 of black carbon (Wiedinmyer, et al., 2014). The proximity to largely populated urban areas has created concerns for both local air quality and climate. We incorporate the Wiedinmyer, et al. (2014) open waste burning inventory into GEOS-Chem- TOMAS, a chemical transport model with online aerosol microphysics, and estimate excess mortality rates due to increased ambient PM2.5, as well as the direct radiative effect (DRE) and cloud-albedo indirect effect (AIE). We find that trash combustion leads to approximately 300,000 adult mortalities per year, mostly in Asia, Eastern Europe, and dense urban areas such as Mexico City and Rio de Janeiro. The global-mean climate effects are on the order of -0.001 W m-2 for both the DRE and AIE; however, in the source regions of Southeast Asia, Western Africa, and Central America the DRE and AIE range from -0.2 to -0.3 W m-2.

Experimentally measured optical properties (MAC and MSC) were compared with theoretical predictions from Mie and Rayleigh–Debye–Gans (RDG) models. We used the OA:BC ratio as a parameter (surrogate to coating thickness). Mie theory inputs include the core size distribution (SP2), coating thickness (from measured OA:BC ratio), the wavelength of incoming radiation (405 and 532 nm), the refractive index of BC (1.85+0.71i), and the refractive index of the shell (1.55+0i, i.e., non-absorbing). RDG model inputs are similar to Mie since we treated the monomers as non-interacting “Mie spheres”, and the monomer size distribution was obtained by transmission electron microscopy analysis of samples collected from two different experiments than the aforementioned experiments. Measured MAC was in good agreement with predictions from Mie theory (with less than 20% differences between the two) across the entire OA:BC range, however, RDG theory consistently underestimated the MAC of cook-stove emissions. Measured MSC of fresh and slightly coated emissions (OA:BC<4) followed RDG predictions, consistent with a fractal morphology, however, as the OA:BC ratio (coating thickness) further increased, measured MSC deviated from RDG scattering predictions and approached Mie theory predictions. This indicates a switch in optical regimes from Rayleigh scattering to Mie scattering of aged BC particles emitted form cook-stove emissions, suggesting a core-shell morphology under heavy coating thicknesses. Combining Mie modeled MAC and MSC, we carried out spectral simple forcing efficiency (SFE) calculations. Using this simple forcing metric we calculated a positive forcing (net warming) of cook-stove emissions for OA:BC<6, and a switch in forcing sign (net cooling) at higher OA:BC ratios (coating thicknesses), in good agreement with measurements.

This work suggests that the absorption and scattering properties of cookstove emissions can be reliably modeled using Mie theory for OA:BC > 4, which is a safe assumption for atmospherically relevant conditions.

Future Activities:

During the upcoming project period we shall focus on the following objectives:

  1. Apply firepower sweep parameterization to field data to develop a global cookstoves emissions inventory to be used in climate forcing simulations.
  2. Complete the India field campaign (Winter 2016).
  3. Revise and re-analyze climate forcing simulation models with updated data from the 2014 and 2015 measurement campaigns.
  4. Submit journal publications on (i) results of 2014 Front Range Cookstove Study, (ii) firepower sweep protocol, and (iii) optical properties of cookstove aerosols.


Bond TC, Bergstrom RW. Light absorption by carbonaceous particles: an investigative review. Aerosol Science and Technology 2006;40(1):27-67. Available at http://dx.doi.org/10.1080/02786820500421521. Exit

Kodros JK, Scott CE, Farina SC, Lee YH, L’Orange C, Volckens J, Pierce JR. Uncertainties in global aerosols and climate effects due to biofuel emissions. Atmospheric Chemistry and Physics 2015;15(15):8577-8596. Available at http://dx.doi.org/10.5194/acp-15-8577-2015. Exit

Wiedinmyer C, Yokelson RJ, Gullett BK. Global emissions of trace gases, particulate matter, and hazardous air pollutants from open burning of domestic waste. Environmental Science & Technology 2014;48(16):9523-9530. Available at http://dx.doi.org/10.1021/es502250z. Exit

Journal Articles on this Report : 1 Displayed | Download in RIS Format

Other project views: All 32 publications 6 publications in selected types All 6 journal articles
Type Citation Project Document Sources
Journal Article Kodros JK, Scott CE, Farina SC, Lee YH, L'Orange C, Volkens J, Pierce JR. Uncertainties in global aerosols and climate effects due to biofuel emissions. Atmospheric Chemistry and Physics 2015;15(15):8577-8596. R835438 (2015)
R835438 (Final)
  • Full-text: EGU Journals-Full Text PDF
  • Abstract: EGU Journals-Abstract
  • Supplemental Keywords:

    household air pollution, wood smoke, airborne particulate matter, radiative forcing, emissions, aerosol;

    Progress and Final Reports:

    Original Abstract
  • 2014 Progress Report
  • 2016 Progress Report
  • Final Report