Indoor Air Pollution (IAP) Measurement

Combustion of wood

• Wood consist mainly of carbon, hydrogen and oxygen (CH2O)x
• Results of complete combustion: CO2, H2O and heat

BUT: 'Cooking stove never achieve 100% complete combustion'

There are more combustion products in incomplete combustion like:

• CO, NO2, Small particles
• Formaldehyde, Acrolein, Benzene, Toluene, Styrene, 1,3-Butadiene etc.
• Polyaromatic hydrocarbons
• CO and Particular matter (PM) have the biggest health impacts

Examples:

1. CO – low concentration: Mild headache, fatigue, nausea, dizziness
2. CO – high concentration: Death within one hour
3. PM: strong effects to respiratory system

Annualy X.X mio people die due to being exposed to incomplete combustion every day

IAP and improve stoves

• Rule of thumb:
  • “stoves with less fuel consumption emit less pollutants”
  • “chimneys reduce IAP signicant”

• EnDev relevant stove ( reduced wood consumption) should have reduced emissions, too
• Evidence of the improvement through CO and PM measurement:    

       --> Improvement against Baseline

IAP – Measurement Devices

So far:

• seperated devices for PM and CO measurement
• Data analysis difficult

Now:

• One device for PM and CO measurement
• Data analysis simple using MS Excel
• But interpretation needs to be done

Procedure in the field:

• Mount devices
• Switch on, close lid
• Perform test, take notes in parallel
• Open lid, switch off
• In parallel: fill protocol manual

Data analysis

• read data from SD card and press “Process data” button in MS Excel

See, analyse and interpret results

• Careful as calibration of device might be expired

Aprovecho IAP meter

• Demonstration
• 1st tests
• results

IAP measurement

• Understand results
• Check for negative values
• Be critical, don’t trust what you read without crosschecking

overview on results from projects

Evaluation of Changes in Cooking Practices in the Area of Lliipapuquio, Andahuaylas, Peru

The objective of the study was to determine the health effects of chronic exposure to smoke produced by burning biomass in traditional stoves and to see if there are any changes occurring after the installation of the improved Inkawasi stove . The Inkawasi stove is a built-in brick stove equipped with a chimney. It can be built with two different combustion chambers for different geographical regions and uses the principle of sunken pots to minimise emissions.

The study was conducted by the University of Peru at Cayetano Heredia in a village of the vicinity of the town of San Jeronimo in central Peru. In this area wood is the main fuel used by households for cooking purposes. In the past older (non-smoking) women frequently showed signs of chronic damages of the bronchi (cough, respiratory infections etc.). The methods used were interviews and medical investigations. No determination of kitchen levels of PM and CO was done, however, previous laboratory measurements have shown that emissions of both were greatly reduced by the Inkawasi stove.

A total of 64 families comprising 221 people were randomly divided into two groups. One group received an Inkawasi stove, whereas the other group used the traditional stove. One year later the traditional stove group received a Inkawasi stove too. There was no difference between the groups regarding age, sex, and time spent in the kitchen. However, there was a different in smoking with higher incidence in the Inkawasi stove group (3.7 vs. 1.9%). Differences in health status were determined using a questionnaire and by measurement of blood parameters (haematocrit, haemoglobin levels) and by determination of lung function (vital capacity). After 6 months changes in wood consumption assessed by interviews to assure that the Inkawasi stoves were properly used.

Women using the Inkawasi stove stated that it reduced consumption of wood and there was less smoke in the kitchen. This corroborates well with preliminary results obtained from lab tests, which show a reduction of average PM and CO by 75 and 81 % respectively. However comparable data from the project areas are not available yet. Baseline measurements performed in traditional kitchens showed an extreme variation of both PM and CO by the factor of 114 and 244. This suggests that e. g. kitchen management could also be a very important factor.

Regarding their status health the women reported less arousal at night due to coughing as well as less expectoration. In addition, olfaction and colour vision was improved. Laboratory measurements showed increased haematocrit and haemoglobin content of the blood compared to women using traditional stoves which the author relates to improved food preparation using the improved stove. All differences were statistically significant and the effects were directly related to exposure time and pollutant concentration. In contrast, lung capacity did not differ between the groups although there was a tendency for having higher capacities in the Inkawasi-group. Further, only 8 % of the users of improved stoves had to visit a health facility due to respiratory symptoms in contrast the 22 % of the control group. No correlation existed between health problems and poverty level indicating that the respiratory problems are not due to other factor linked to poverty but that they are caused by the use of the traditional stove.

The study shows that use of energy-efficient Inkawasi stoves significantly improves women’s health besides fuel saving. Its approach using a combination of questionnaires as well as medical measurements may serve as al model for future studies. However, the effects of energy-efficient stoves on chronic diseases caused by IAP such as chronic obstructive lung disease remain to be assessed.

Further reading

• Evaluacion de Cambio de Cocinas en el Centro Poblado de Lliupapuquio, Andahuaylas 
• Results of Testing And Existing Stove Design Recommendations -- Peru

Uganda: Indoor Air Pollution Measurements during PCIA Conference 2009

The report presents the results of preliminary measurements of particulate matter (PM) and carbon monoxide (CO) in four households and two schools in Kampala, Uganda. Three households were equipped with an improved Rocket Lorena Stove, whereas one household used a traditional stove. In addition measurements were performed in two school kitchens with traditional stoves (3-stone fire). No further details on the traditional stove type of schools are given. CO was measured using a GasBadge Plus device and a CO gauge for CO measurement and an UCB PM Monitor to determine particulate matter. Measurements were done by a team from Berkeley University together with a project officer.

Table 1: Results of Emission Monitoring in Households and Schools in Kampala

Stove Type		PM	CO
Improved	House 1	0.237	5.3
Improved	House 2	0.357	2.1
Improved	House 3		9.4
Traditional	House 4	1.746	38.2
Institutional	School 1	0.251	7.2
Institutional	School 2	1.213	47.7

The measurements show a reduction of both CO- and PM-levels in households with a Rocket-Lorena-Stove by 69 and 83 %, respectively. However, the results obtained in schools are less clear. In one school kitchen the level are high whereas in the other they are as low as in households using the Rocket-Lorena Stove. Unfortunately, no reason for these discrepancies is given. From the study the following recommendations emerge:

• Detailed IAP monitoring should be performed for the stoves promoted by the project as well as for the 3-stone-fire.
• Institutional stoves should be monitored more urgently because they consume more firewood than stoves in households.
• Based on the results emissions should be used as one criterion for stove selection.A detailed IAP study is in progress.

Summary of Stove-Testing Report from GTZ-SUN:E - Ethiopia

The goal of the study was to obtain data on fuel consumption and IAP-Pollutants particulate matter (PM) and carbon monoxide (CO) from traditional and energy-efficient stove models (open fire, Mirt stove, Approvecho stove). It was performed at the EREDPC und applied the controlled cooking test (CCT), an established testing protocol. Measurements were performed in a kitchen and care was taken that stoves and measurement devices were always installed at the same place. In addition fuel consumption was determined and was found to be about 50 to 60 % in the Mirt stove and the Approvecho stove as compared to the open fire. CO levels were measured using a HOBO CO logger.

Table 1: CO-Emissions in ppm (mean + standard deviation of 8 measurements)

	Open Stone Fire			Mirt Stove			Approvecho Stove
	C1	C2	C3	C1	C2	C3	C1	C2	C3
Average	78.90 +10.55	90.71 +17.81	88.97 +17.64	6.09 +7,50	6.97 +7.56	7.97 +6,78	8.37 +4.60	9.14 +3.57	7.89 +2.43
Average of baking session	88.84 +13.61	101.06 +21.61	99.11 +21.51	6.11 +9,35	7.02 +9.43	8.08 +8.8	8.70 +4.55	8.82 +4.49	7.58 +3.05
Maximum	124.30 +0	241.93 +43.73	239.5 +64.0	42.80+ 29.26	44.44 +29.53	42.00 +29.6	43.28* +17.16	43.21* +16.99	40.00* +17.14

The means given in the study are 79.3, 79.1 and 74, respectively.

However, from the table it is clear that these values represent individual measurements. Therefore means were recalculated. HOBO loggers were used to monitor CO Measurement of CO emissions was performed eight times presumably to improve accuracy of the mean levels. The results of three different channels given as C1 - 3 represent CO-levels below 125, up to 500 and above 500 ppm. In addition, for each test the average, the average of the baking session and the maximum values are given. In the open fire the maximum CO-levels for C1 are eight times the same, which is in marked contrast to the variation in other groups, which is sometime quite large. Further, no explanation is given, why the C2 mean was used for comparison. Generally, the description of the testing procedure is not sufficient.

Nevertheless, the results indicate higher emissions of both CO and PM in the open fire compared to the Mirt stove and the Approvecho stove. Particulate Matter PM concentrations were also measured eight times and means were calculated. However, from the individual values it becomes clear that that some means contain outliers which obscure the results. Therefore the means were recalculated after removing these data. However, no information is given whether these data could be due to problems with the measurement or with the test procedure (high concentrations during the lightning of the stove

Table 2: Particulate Matter Emission (mg/m³) (mean + standard deviation)

	Average		Average of Baking Session		Maximum Values
	n = 8	n = 7	n = 8	n = 7	n = 8	n = 7
Open fire Injera	0.98 + 0.45	0.83 + 0.17	1.04 + 0.50	0.87 + 0.21	8.91 + 7.69	6.23 + 1.49
Mirt stove	0.68 + 0.65	0.47 + 0.26	0.77 + 0.75		6.55 + 5.01
Approvecho Stove	0.51 + 0.7	0.27 + 0.09	0.73 + 1.24	0.29 + 0.09	3.60 + 2.33

According to the study average reduction is 27 % for the Mirt stoves and 46 % for the Approvecho stove compared the traditional fire. However, if the means are recalculated without outliers (figures in italics) PM reduction was higher being 44 % for the Mirt stove and 68 % for the Aprovecho stove. In contrast to CO-Concentration, PM-Levels are much stronger affected by the apparent problems of data collection or test procedure. Although the study contains important data it lacks essential information how the tests were performed and the problem of the numerous outliers is not discussed. Evaluation study on the performance of cook stoves to quantify air pollution, health impacts, energy efficiency and sustainability in Kisii By Indeginous Innovation Enterprises, Nairobi on behalf of Ministry of Agriculture and GTZ/PSDA- Kenya (2007)

Within the framework of a comprehensive assessment of promotion of the rocket stoves emissions of particulate matter (PM) of four different stove models in local households were measured.

The study was performed in Kembu division, located in Kisii district, Nyanza Province. The region is densely populated and has suffered from fuelwood problems. Households were randomly selected from three sub-locations in the Ibero location.

The selection criteria for the households that were considered for 24 hour monitoring of smoke level (PM10) where: - to have children below five years of age to ensure that cooking took place most of the time; - to have a built-in kitchen attached or detached from the main house; - household members who voluntarily accepted to take part in the study and were willing to use only one stove type (for kitchens with more than one stove) during the 24-hour monitoring period.

The selection of the households was guided by the stove monitors who were responsible for each sub-location as the safety of the sampling equipment.

36 households were included in the study. The stove types tested were the traditional 3-Stone-Fire, the Rocket Mud Stove, the Shielded Fire Stove and the Jiko Kinasa Stove. Particulate matter of less than 10 microns in diameter levels were determined by using filters fixed in cyclones and mounted on battery operated pumps left in a household for 24 hours. In order to emulate the sitting position of the cook, the cyclone was placed at 4 feet above the kitchen floor, and 4.5 feet horizontally from the hearth. Filters were weighted on a 6-digit balance, prior to insertion into the cyclone cassettes, labelled and dispatched to the field for air sampling. After return from the field, the filters (still in cassettes) were re-weighted. Filters were changed after 24 hours to conform to standard procedures employed in measurements of particulate matter. No information about on the manufacturer of equipment or statistical procedures is given.

All stove types reduced PM: by 33 % (Jiko), by 29 % (Shielded Fire) and by 69 % (Rocket Mud Stove). Consumption of wood was also lowered by 31 %, 53 and 32 %, respectively. However, no relationship exists between the two parameters. In addition, moisture content of the fuelwood was more than 50 % higher in households using a 3-Stone-Fire. The authors explain this difference by the fact that the cooks, who use an improved stove, are sensitized about the importance of drying wood prior to use. Further, the large variation of PM-emissions of improved stoves could be due to the lack of drying shelves for fuelwood in some of the households. Nevertheless, since the higher moisture content of the fuelwood adversely affects both fuel consumption and emissions it may augment the inferiority of performance of the 3-Stone-Fire.