Sustainable Environment Research 28 (2018) 32e38 Contents lists available at ScienceDirect Sustainable Environment Research journal homepage: www.journals.elsevier.com/sustainableenvironment-research/ Original Research Article Emissions and fuel use performance of two improved stoves and determinants of their adoption in Dodola, southeastern Ethiopia Fikadu Mamuye a, Bekele Lemma b, *, Teshale Woldeamanuel c a Wondo Genet College of Forestry, Hawassa University, Shashemene, P.O. Box 128, Ethiopia Department of Chemistry, Hawassa University, Hawassa, P.O. Box 5, Ethiopia c Regional REEDþ Coordination Office, Hawassa, P.O. Box 1952, Ethiopia b a r t i c l e i n f o a b s t r a c t Article history: Received 6 April 2017 Received in revised form 29 June 2017 Accepted 25 September 2017 Available online 30 September 2017 Improved cook stoves (ICS) have perceived to exert a significant impact on households' economy, human health, and global climate change. There are few studies on ICS emissions and fuel use performance and on the factors that affect their adoption in Ethiopia. Thus, the objectives of this study were assessing: (a) the emissions of CO, CO2 and fine particulate matter (PM2.5) of improved Merchaye and Lakech charcoal stoves in comparison with traditional metal stoves; (b) specific fuel consumption (SFC) of the two ICS; and (c) the factors that affect their adoption. Data were collected using the Water Boiling Test in a laboratory and household survey. The results showed the Merchaye stove reduced emission of CO, CO2 and PM2.5 by 28, 22 and 27% respectively in comparison to a traditional charcoal stove. Whereas, the Lakech stove reduced emission of CO, CO2 and PM2.5 by 15, 8 and 13%, respectively. In non-sustainable fuel wood harvest circumstances, the annual emission reduction potential for individual Merchaye stoves was 0.33 t CO2e and Lakech stoves 0.14 t CO2e yr1. The SFC of Merchaye and Lakech were reduced by 222 and 164 g d1, respectively. The two ICS also reduced the time required for cooking. Regarding the status of adoption of ICS, 43.7% the sample households were adopters of Merchaye stoves and 31.3% Lakech, stoves. Whereas the non-adopters comprise 25% of the sample. Adoption of ICS was influenced by household head age, sex, education level and income. The results may have implication for mitigation of climate change, forest degradation and household workload. © 2017 Chinese Institute of Environmental Engineering, Taiwan. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Keywords: Merchaye Lakech Emission Adoption Climate change 1. Introduction In developing countries, biomass is still the predominant cooking fuel [1] and currently there are a wide variety of stove technologies and designs. Among biomass fuels, charcoal is the predominant cooking fuel in sub-Saharan Africa's cities and towns [2], and in Ethiopia charcoal stoves are commonly used in urban and semi urban settings. Inefficient fuel combustion in traditional stoves release gaseous products with a higher global warming potential than carbon dioxide, such as carbon monoxide [3]. Traditional stoves are still the most prevalent way of cooking in the developing countries regardless of their inefficiency and risks associated to human health and the environment [4]. * Corresponding author. E-mail address: Bekelelelemma@gmail.com (B. Lemma). Peer review under responsibility of Chinese Institute of Environmental Engineering. The main reason for the development of improved stoves is their environmental, health and socioeconomic benefits. Zhang et al. [5] have indicated that improved cook stoves (ICS) reduce the emission of health-risky pollutants in the short term and reduce greenhouse gases (GHG) emission in the long term. A study in China found that adoption of ICS reduced fuel wood consumption, wood collection time, and tree felling by 40.1, 38.2 and 23.7%, respectively [6]. In Guatemala the ‘Plancha’ ICS saved wood consumption by 39%, decreased time spent for wood collection and reduced indoor air pollution levels [7]. Pine et al. [8] asserted that ICSs reduced particulate matter (PM) by 74% and carbon monoxide (CO) concentrations by 78% in Mexico. The adoption of ICS (Patsari) has significantly contributed to improvements in living conditions through wood savings, and reducing indoor air pollution [9]. The adoption of ICS (patsari) improved womens' respiratory systems and eye comfort in Mexico [10]. In Gambia, ICSs saved fuel wood consumption by 40% and reduced indoor air pollution up to 90% [11]. Similarly, in Tanzania the adoption of ICSs saved fuel wood https://doi.org/10.1016/j.serj.2017.09.003 2468-2039/© 2017 Chinese Institute of Environmental Engineering, Taiwan. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 consumption, reduced women's workload by reducing time required for fuel collection, and created self-employment for the stove producers [12]. In developing countries like Ethiopia, whose energy supply is heavily dependent on biomass fuels, technical advances in energy efficiency are critical. In order to reduce pressure on forests and the adverse impact of indoor air pollution, the government of Ethiopia is trying to increase the availability of fuel saving technologies such as ICSs [13]. In this regard, non-governmental organizations, mainly GIZ, have been working on afforestation programs and dissemination of more efficient ICS technologies [14]. Although ICSs have magnificent contributions in reducing GHG emissions and PM, little attempt to quantify GHG emission from ICSs have been made in Ethiopia. Despite the fact that ICSs are a better option than traditional stoves, studies indicate that adoption of ICS has fallen behind expectations [15]. To our knowledge, there are no studies on improved charcoal stoves in Ethiopia. The objectives of this study, therefore, were to: (a) assess the CO, CO2 and PM2.5 emission reduction potential of Merchaye and Lakech charcoal stoves; (b) analyze their fuel and time saving efficiency, and (c) assess the factors that determine their adoption at Dodola, South East Ethiopia. 2. Methodology 2.1. Study area The study was conducted in Dodola town, Oromia National Regional State, southeastern Ethiopia. It is located 320 km southeast of Addis Ababa in the Adaba-Dodola forest priority area. The town has a population of 26,176 and 3842 households. Cooking accounts for the bulk of domestic fuel consumption. Preparing sauce (commonly known as ‘wot’), boiling water, making coffee and similar activities involve burning a fire several times a day. Electricity and petroleum products are also available energy sources in this town. 2.2. Selection of the study area Dodola was purposively selected as the study site due to (1) the accessibility of different types of charcoal stoves to the inhabitants, and (2) the town's close proximity to the Dodola Adaba forest reserve. Among the various available ICS, the researchers purposively selected Lakech, Merchaye as well as the traditional metal charcoal stove that is used by a large proportion of the inhabitants. The traditional stove was used as the control for comparison. 2.3. Description of charcoal stoves The traditional metal charcoal stove (Fig. 1) is square shaped with removable grates and weighs approximately 1.5 kg. Typical dimensions 9.5 cm deep, an upper surface area of 441 cm2, and a combustion area of 180 cm2. Evenly distributed holes are located at the bottom of a square charcoal container. The cooking pot sits directly on the charcoal in the chamber. The cost of the stoves was 60 birr (USD 3) in July 2015. The Lakech charcoal stove weighs 1.9 kg with combustion area of 179 cm2 and depth of 8.5 cm. It has also an upper surface area of 400 cm2. Pieces of charcoal are combusted in a bowl shaped combustion chamber. The stove's grates have 0.5 cm diameter holes. The pot sits on the stove's pan seat which is fixed to the metal part of the combustion chamber. The primary air metal entrance allows air to enter into the combustion chamber. The cost of a Lakech stove in June 2015 was 70 birr (USD 3.5). 33 In comparison to the Lakech, the Merchaye stove is lighter, with a smaller combustion area, depth, and upper surface area. It weighs 1.8 kg with a combustion area of 169 cm2, depth of 7.8 cm and upper surface area of 324 cm2. Charcoal is burned in a bowl shaped combustion chamber. The grates have 1e2 cm diameter holes. The pot sits on the stove's pan seat which is fixed to the metal side of combustion chamber. The primary metal air entrance enables air to enter into the combustion chamber. In June 2015, the cost of a Merchaye stove was 140e150 birr (USD 7e7.5). Merchaye and Lakech stoves are made from clay and sheet metal while the traditional stove is made from only sheet metal. 2.4. Water boiling test The water boiling point test (WBT version 4.2.2) was conducted in the Addis Ababa laboratory of the Ethiopian Ministry of Water Irrigation and Energy to determine the performance of the stoves [16]. Although WBT was originally designed for wood-stoves, it has been adapted for charcoal stoves, with three phases e a cold-start phase, hot-start phase, and a simmering phase [16]. In a cold-start phase the tester begins with the stove at room temperature and boils 2.5 L of water in a 3 L pot without a lid. In the hot-start phase, water is boiled beginning with a hot stove to identify differences in performance between a hot and cold stove body. The tester then simmers the remaining water at approximately 3 C below boiling for 45 min. These stove performance measurements help to simulate the process of cooking food. In order to estimate daily fuel consumption 2.5 L is multiplied by 3 (for morning, midday and night cooking time). Each stove's CO and CO2 emissions data were measured using IAQ-CALC meter (Model No. 7545 instruments, Onset Computer Corporation, Bourne, MA, USA) while fine particulate matter (PM2.5) data were collected using an indoor air pollution meter (IAP Meter-5000-Series, Aprovecho research center, 2008). Both the IAQ-CALC meter and IAP Meter stored the data on data logger minute-by-minute over the entire measurement period. The test was done three times for each stove type and data on CO2, CO and PM2.5 emissions were collected three times following the standard WBT version 4.2.2 in a controlled laboratory setting [16]. Background emissions were also accounted for by measuring concentrations of CO2, CO and PM2.5 before and during the test. The air temperature was 17.8e18.8 C, the local boiling point was 91 C, and the relative humidity was 64%. The charcoal used in this study was produced from an indigenous Podocarpus falcatus. Its moisture content was 9% and its pieces used in the study have a size of roughly 5e6 cm in diameter. 2.5. Household survey Data on determinants affecting ICS adoption were collected by means of a household survey. The standard statistical equation was applied to determine the total sample size needed for this study [17]. As a result, 40 samples households, who did not adopt improved charcoal stoves, and 120 households who adopted improved charcoal stoves were selected randomly from the town's 3842 household inhabitants. The major issues included in the household survey were, the types of stoves adopted by the household, status of adoption of the ICS, the factors that contributed for the differences due to adopter and non-adopter households, the type of fuel wood used for cooking, the amount used per day, etc. The household survey questionnaires were pretested before the actual survey and, based on the results, and were revised avoiding ambiguity and terms of cultural sensitivity. Data collectors were employed for the household survey after training them on how to handle the interview. In addition a supervisor was assigned to follow the data collection in unannounced time of interviews 34 F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 Fig. 1. Pictures of (a) Mirchaye, (b) Lakech and (c) traditional stove. within a day during the survey and, in approximately 5% of the households, to minimize interviewer bias. cis e CO2, CO, PM2.5 emission or SFC of improved stoves. c ts e CO2, CO, PM2.5 emission or SFC of improved stoves. 2.6. Calculations 2.6.1. The CO2 equivalents (CO2e) calculation The global warning potential of CO, CO2 and PM2.5 is different. So, the CO2e of CO and PM2.5 were calculated as follows to determine the total emission in CO2e. CO2e ¼ GWPi GHGi where GWPi is a global warming potential of each gas (relative to CO2). GHGi is the quantity of each greenhouse gas emitted. 2.6.2. Specific fuel consumption (SFC) and time SFC (g) is the charcoal required to boil 2.5 L of water, which was calculated following the equation: SfC ¼ fd ðphf pÞ where fd e fuel consumed (g). phf e weight of pot with water after test (g). p e weight of pot (g). Time needed to boil 2.5 L of water was calculated as the difference between start and finish times: Dt ¼ tf ti where Dt e total time (min) to boil. tf e time at the end of the test. ti e time at the start of the test. 2.6.3. Emissions and specific fuel wood reduction calculation Calculation of a specific emission reduction potential and fuel consumption reduction for each improved stove was accomplished by comparing with the corresponding values of the traditional metal stoves using the formula below. creduction ¼ cts cis 100 cts where, creduction e CO2, CO, PM2.5 emission reduction or SFC reduction of improved stoves. 2.6.4. Statistical analysis The statistical differences in SFC and emissions of CO2, CO and PM2.5 among the stoves were computed by one way Analysis Of Variance (ANOVA) using SPSS statistical software version 20 at 5% level of significance. The least significant difference test was conducted for mean separation of significant differences. The data from the household survey were analyzed using descriptive statistics: frequency, percentage, means and standard deviation. 3. Results and discussion 3.1. Improved charcoal stoves and CO2, CO and PM2.5 emission Both Merchaye and Lakech emitted significantly lower CO2 (P < 0.001) than the traditional metal stove. Fig. 2 shows the Merchaye stove emitted the least CO2. The CO2 emission per 2.5 L of water for the Merchaye, Lakech and traditional metal stove was 531, 625 and 681 g, respectively. The CO2 emission of this study's Merchaye stove was comparable with the Ghanaian Gyapa charcoal stove [18] which emitted 536 g per 2.5 L of water. However, the CO2 emission from the Lakech stove was higher than the Gyapa stove. With respect to CO emission, Merchaye and Lakech emitted significantly lower CO (P < 0.001) than the traditional metal stove while the Merchaye stove emitted significantly lower CO (P < 0.001) than Lakech stove. Merchaye, Lakech and traditional stoves emitted 66, 79 and 92 g of CO, respectively per 2.5 L of water. The CO emission reductions of the Merchaye and Lakech stoves were 28 and 15%, respectively (Fig. 2). Although both Merchaye and Lakech stoves emitted relatively less CO than the traditional metal stove, the emissions from these ICS were above the proposed benchmark value of 20 g [19]. This is because the international bench mark value is the average of wood and charcoal stoves. Charcoal would normally have a higher CO level than wood [19] which may explain this study's higher CO emissions. The PM2.5 emissions from the Merchaye, Lakech and traditional metal stoves were 275, 325 and 375 mg, respectively (Fig. 2). The Merchaye and Lakech emitted significantly lower PM2.5 (P < 0.001) than the traditional metal stove. The Merchaye stove emitted the least PM2.5 (P < 0.001; Fig. 2). The results show that PM2.5 emission reduction by Merchaye and Lakech stoves were 27 and 13%, respectively. The fine PM emissions from both improved stoves and the traditional metal stove were quite below the proposed benchmark value of 1500 mg [18]. The bench mark value was also computed for both charcoal and wood stoves. F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 Table 1 Total Global Warming Potential (TGWP, grams CO2e per 2.5 L of water) and CO2e emission per year of Merchaye, Lakech and traditional stoves. 700 600 Stoves Emissions (g 2.5 L1) CO CO2 PM2.5 CO2e CO2e Merchaye Lakech Traditional 66 79 92 531 625 681 0.28 0.33 0.38 917 1082 1213 1.00 1.19 1.33 500 WBT CO2 (g) 35 400 300 * TGWP (g 2.5 L1) CO2e emission per year (t yr1) GWP of CO ¼ 3, GWP of CO2 ¼ 1 and GWP of PM ¼ 680. 200 emission of CO, CO2 and PM2.5 of Merchaye, Lakech and traditional metal stoves, were 1.0 t, 1.2 t and 1.3 t CO2e, respectively (Table 1). Although this laboratory study should not be used to specifically predict real-world performance, it is interesting to project the potential emission reductions in CO2e, per stove, per year. Merchaye stoves can potentially mitigate 0.33 t CO2e yr1 and Lakech 0.14 t CO2e yr1 emission per stove. Thus, the Merchaye stove reduced total emissions of the studied GHG by 25% and the Lakech 11%. Improved charcoal stoves were estimated to reduce 20% of emissions produced from incomplete combustion [20], and in the present study the Merchaye ICS had emission reduction close to this value. If biomass is harvested sustainably, then the CO2 released in combustion is theoretically reabsorbed by the biomass growing to replace it. In the non-sustainable fuelwood harvesting circumstances in Ethiopia [21], the CO2 released is contributing to the build-up of CO2 in the atmosphere [19]. The results of this study show that shifting from traditional metal stoves to Merchaye and Lakech stoves could mitigate CO2, CO and PM2.5 emissions with Mechaye being superior in its performance. 100 0 WBT CO (g) 80 60 40 20 3.2. SFC and time required for cooking by ICS 0 400 350 WBT PM 2.5 (mg) 300 250 200 150 100 50 0 Merchaye stove Lakech stove Traditional Metal stove Fig. 2. Average CO2, CO and PM emission per 2.5 L of water of Merchaye, Lakech and traditional stoves. Therefore, the difference could be attributable to the higher PM emission from wood stoves than from the charcoal stoves. High amounts of PM precursor are removed during the charcoal production process which leads to lower levels of PM emissions from charcoal stoves [19]. The Merchaye and Laketch stoves had a lower potential of emitting GHGs with GWP as compared to traditional stoves. This becomes apparent when GWP is applied to all emissions and combined into the same scale of CO2e (Table 1). The annual Fig. 3 shows that cooking 7.5 L of water in a day, the fuel consumption of a Merchaye stove was 478 g while the Lakech charcoal stove required 536 g of charcoal. In comparison, the traditional stove had a fuel consumption of 700 g d1. The values of fuel consumptions were significantly different (P < 0.001) among the three stoves. Thus the Merchaye stove used 222 g (32%) less fuel and the Lakech stove 164 g (23%) less fuel, per day than the traditional stove. The SFC reduction in the present study was in the range for Ceramic Jiko ICS in Kenya which was 20e50%. According to EPA [22], ICS can save up to 25% over traditional stove. In the present study, the Lakech stove showed results close to this estimation while the Merchaye stove had even better results. In subSaharan Africa, fuel wood and charcoal accounts for 75% of total wood harvest, contributing to rapid deforestation in hotspot areas including Ethiopia [21]. Thus, the present results have implications concerning forest degradation since the use of ICS can reduce pressure on forests. The difference in SFC among the various stoves could be attributed to the difference in design [23] and the materials from which they were made. Design and materials are the most important variables that affect stove performance [24]. Even though Merchaye and Lakech stoves were made from the same materials, the design is different as is the performance. Traditional stove loses more heat as compared to ICS due to the use of only sheet metal in their construction. The Mechaye charcoal stove took 220 min for cooking per day with the Lakech stove 224 min. The two ICS significantly reduced (P < 0.01) cooking time when compared to traditional metal stove. Traditional metal stove was the slowest, taking 236 min d1. This implies that using these ICS stoves can save 13e17 min cooking time per day. In Uganda, studies found that the average cooking 36 F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 Table 2 Descriptive statistics of age and family size with respect to adoption of improved charcoal Stoves. 700 600 Variables Groups Min Max Mean St. dev Age Non-adopters Adopters Non-adopters Adopters 23 19 1 1 69 67 9 9 40.2 32.4 5.05 5.13 10.6 10.7 2.1 2.0 500 SFC (g) Family size 400 300 200 100 0 250 Time (min) 200 150 100 50 0 Merchaye stove Lakech stove Traditional Metal stove Fig. 3. Average SFC per day and mean time required for cooking. time per household was reduced by 27 min d1 when using the Rocket Lorena stove [25]. Although the time saved per day is relatively small, the accumulated time can be a benefit to the family's wellbeing. 3.3. Adoption of ICS and their determinants The household survey result indicates that the majority of the respondents (75%) have adopted ICS. Specifically, out of the 160 respondents, 43.8% of the respondents were using the Lakech stove and 31.3% the Merchaye stoves. The remaining 40 respondents (25%) had not adopted ICSs. In this regard, socioeconomic characteristics of the households such as sex of the household head, education level of household head, household income and family size were considered important variables that affected adoption of the improved charcoal stoves. Comparison of the adopter and non-adopter households with respect to their age shows that the former households are younger than the later. The average age of sampled households was 36.3 and the average age of adopter households was 32.4 which is less than this average age of all sampled households. However the average age of non-adopters (40.2) was higher than the average of adopter households as well as the average of all sampled households (Table 2). This shows that households that adopted ICS are younger than the non-adopters. This could be because the younger household heads were more eager to adopt ICS technologies than older household heads. In contrast to the present study, Gebreegziabher et al. [26] found that older household heads were more willing to use the improved Mirt stoves than younger household heads. The present study was consistent with Dawit [27] who indicated that younger household head were better adopters of Mirt stoves. A review on the adoption of improved stove by Lewis and Pattanayak [28] indicated that the age of household head influenced the adoption of ICS significantly and negatively. The younger household heads make adoption decision superior to older ones and older household heads were more conservative to use the traditional stoves. The mean family size of both improved cooking stove adopters and non-adopters was very close. The mean family size of adopters was 5.13 and that of the non-adopters was 5.05 with family size ranging between 1 and 9 for both groups (Table 2). This implies that the decision to adopt ICS was not influenced by household size. In contrast to this finding, family size was a significant factor in determining the adoption of ICS in Mexico [8]. Another Ethiopian study had similar findings [26]. Pine et al. [8] explained that households with larger family size consumed larger amounts of fuelwood leading to the adoption of ICS. However, this argument was not supported by the present finding. Table 3 shows that out of 160 households, males led 98 of these households. Among these male-led households, 64.3% were adopters of ICS while 35.7% were non-adopters. Whereas, out of the 62 female-lead households, 92% were improved stove adopters and only 8% were non-adopters. The majority of households who did not adopt improved charcoal stoves were male-headed households. Female-headed households were more likely to adopt ICS as compared to married women of male-headed families. One plausible explanation for this could be that female household heads had greater power to make economic decision as compared with females in male-headed households. In a patriarchal society such as Ethiopia, economic decisions are most often made by a husband in a male-led household [29,30]. The adoption of ICS can be affected by failure to recognize and target women in the ICS dissemination activities. Education level of the household head is an important variable found to influence adoption of the ICS. The results showed that 70.6% of the households attended formal education, and of these, 62.5% were improved charcoal stove adopters (Table 3). On the other hand, 29.4% of the households did not attend formal education and only 12.5% of them were ICS adopters. The proportion of households with formal education adopted ICS more than those who did not attend formal education. This shows that educated households have a higher probability of using improved stoves. This could be attributable to the awareness that education brings concerning fuel cost comparisons and the health issues with traditional stoves [31]. Another study has also indicated the key role of education in shifting household cooking preferences from traditional to improved cooking devices [32]. Consistent with the present study, a review of different studies by Lewis and Pattanayak [28] showed that the household head's education level was a significant factor determines ICS adoption. Jan [30] also indicated that knowledge F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 Table 3 Distribution of households by gender and education between improved charcoal stoves adopter and non-adopters. Variables Household head sex Education Categories Female Male No formal education Formal education Adopter Non adopter Total No. % No. % No. % 57 63 20 100 35.6 39.4 12.5 62.5 5 35 27 13 3.1 21.9 16.9 8.1 62 98 47 113 38.8 61.2 29.4 70.6 about the different financial instruments can be increased by education which minimizes the perceived expensiveness of ICS. Other studies in Ethiopia have also confirmed that the household head's education influenced the decision of improved stove (Mirt) adoption [26,27,29,33]. In this study, adoption of ICS was also affected by household income which is defined as the annual earnings of a household obtained from all sources (e.g., crop production, livestock and livestock products, salary, etc.). The majority of adopter households (95%) earned more than 1501 birr per month (Table 4). For the nonadopters, 75% earned less than 1500 birr per month and are grouped in the low income categories. Arthur et al. [4] indicated that higher socio-economic status affected a household's ICS adoption decision positively and significantly. Other studies also showed that household income was a significant factor that affected adoption of ICS in Ethiopia and elsewhere [8,26e30]. Since women are almost exclusively responsible for the cooking, they are more likely to make decision on cook stove adoption. But their decisions are constrained by the inadequate financial resources in the area of their decision-making [34]. The influence of income on ICS adoption was also supported by data concerning stove pricing. The majority (95%) of non-adopter respondents stated that the price of improved charcoal stoves was “expensive” and they could not afford to buy it. Whereas only 2.5% of the respondents indicated that the price was “cheap.” This implies that price of ICS affected their decision not to purchase the stoves. Regarding the adopter households, 34.2% of respondents said that the price is “cheap” and 33.3% of the respondents replied the price was “expensive” with 32.5% “fair” (Table 5). This may indicate that adoption of ICS is comparatively easy for adopter households who have relatively high income (Table 4). This finding was similar with a study by Axen [35] that indicated the cook stove's price was an important factor that affects the adoption decision. A study by Levine et al. [36] also found that the cost of ICS was an important adoption barrier. The poor cannot afford to buy ICS because of the relatively high cost and thus, the cost of ICS was a determinant factor of adoption [37]. In contrast, the poor often have to spend a substantial amount of an already constrained household finance on fuel wood. Thus the enabling of a more efficient fuel use may thus be an important strategy in poverty alleviation [38] in addition to other benefits. Table 4 Distribution of household monthly income (Birr) with respect to improved charcoal stoves adopters and non-adopters. * Monthly income (birr) Adopters Frequency Percent Frequency Percent < 1000 1001e1500 1501e2000 2001e2500 > 2501 Total 3 3 39 40 35 120 2.5 2.5 32.5 33 29.2 100 20 10 6 3 1 40 50 25 15 7.5 2.5 100 Note: birr is Ethiopian currency. Non adopters 37 Table 5 Opinion of households' on Improved Charcoal Stove Price. Opinion on stove's price Cheap Fair Expensive Total Adopters Non adopters Frequency % Frequency % 41 39 40 120 34.2 32.5 33.3 100 1 1 38 40 2.5 2.5 95 100 The current study shows that adoption of ICS is not only affected by household socioeconomic characteristics but also by fuel source. The majority of the households (96.7%) purchased their cooking fuel while 3.3% obtained free fuel from their own farm. Most high income households in the study area are more likely to afford ICS. However, purchasing fuel wood from market than obtaining for free could also be a major motivation for using energy saving ICS devices. When households have free access to free fuelwood, they are less likely interested in purchasing ICS. Several other studies also found that households who purchase fuel wood adopted ICS more than those who could get free fuelwood [29,30,35]. Similarly Geary et al. [39] showed that the availability of free fuelwood is a factor leading non-adoption of ICS. A study by Pine et al. [8] found that access to open forest land is negatively correlated with ICS adoption. Axen [35] and Troncoso et al. [40] reported a positive correlation between lack of open forest access and ICS adoption and vice versa. 4. Conclusions The findings from this study show that there is differential emission and fuel use performance among cook stoves. Merchaye and Lakech ICS emitted less CO, CO2 and PM2.5 as compared to the traditional metal stove. The Merchaye stove was better in reducing CO, CO2 and PM2.5 emissions compared to the Lakech stove. The annual emission reduction potential of the Merchaye stove is 0.33 t and the Lakech stoves 0.14 t CO2e. This is a daily SFC reduction of 32% for the Merchaye and 23% for the Lakech stove. The differential SFC reduction of the Merchaye and Lakech stoves could be attributed to the difference in design and the type of materials used for making the stoves. The potential reduction of CO, CO2 and PM2.5 emission and the reduction of SFC of these ICS in the present study may have implication for mitigation of climate change and forest degradation. The two ICS stoves also reduced the time required for cooking which will have impact on women's household workload. Results shows that three fourth of the households were found to be adopters of Merchaye and Lakech ICS, while one fourth of the households were non adopters. From the factors considered in the present study, household-head education, sex, age, income and stove price determined household ICS adoption. Whether the fuel is freely collected or purchased was also an adoption factor. The adoption of ICS technology is critical to reduce GHG emissions, forest degradation, and household workload in developing countries that heavily depend on biomass energy and where fuel wood and charcoal account for the rapid deforestation. The dissemination of ICS is, therefore, vital in the developing countries like Ethiopia. The contribution of ICS to the reduction of GHG, forest degradation and household workload can be augmented by: (i) increasing the capacity of the households to adopt the ICS by providing the ICS through credit and other means, (ii) recognizing the importance of women and targeting them in the dissemination activities of ICS, and (iii) improving the designs of biomass stoves. 38 F. Mamuye et al. / Sustainable Environment Research 28 (2018) 32e38 Acknowledgements We would like to thank the Ethiopian Ministry of Water Irrigation and Energy for allowing access to their laboratory facilities. We are grateful to Robert Sturtevant for editing the English of the manuscript and two anonymous reviewers as well as the Editor for their invaluable input that improved the quality of the manuscript. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References [1] Bruce N, Perez-Padilla R, Albalak R. The Health Effects of Indoor Air Pollution Exposure in Developing Countries. Geneva, Switzerland: World Health Organization; 2002. [2] Mwampamba TH, Ghilardi A, Sander K, Chaix KJ. Dispelling common misconceptions to improve attitudes and policy outlook on charcoal in developing countries. Energy Sustain Dev 2013;17:75e85. [3] Ramanathan V, Carmichael G. Global and regional climate changes due to black carbon. Nat Geosci 2008;1:221e7. [4] Arthur MFSR, Zahran S, Bucini G. On the adoption of electricity as a domestic source by Mozambican households. Energy Policy 2010;38:7235e49. [5] Zhang J, Smith KR, Ma Y, Ye S, Jiang F, Qi W, et al. Greenhouse gases and other airborne pollutants from household stoves in China: a database for emission factors. Atmos Environ 2000;34:4537e49. [6] DeWan A, Green K, Li X, Hayden D. Using social marketing tools to increase fuel-efficient stove adoption for conservation of the golden snub-nosed monkey, Gansu Province, China. Conserv Evid 2013;10:32e6. [7] Bielecki C, Wingenbach G. Rethinking improved cookstove diffusion programs: a case study of social perceptions and cooking choices in rural Guatemala. Energy Policy 2014;66:350e8. n-Mares A, Riojas[8] Pine K, Edwards R, Masera O, Schilmann A, Marro Rodríguez H. Adoption and use of improved biomass stoves in rural Mexico. Energy Sustain Dev 2011;15:176e83. [9] García-Frapolli E, Schilmann A, Berrueta VM, Riojas-Rodríguez H, Edwards RD, Johnson M, et al. Beyond fuelwood savings: valuing the economic benefits of pecha region of Mexico. introducing improved biomass cookstoves in the Pure Ecol Econ 2010;69:2598e605. [10] Masera O, Edwards R, Arnez CA, Berrueta V, Johnson M, Bracho LR, et al. n, Impact of Patsari improved cookstoves on indoor air quality in Michoaca Mexico. Energy Sustain Dev 2007;11:45e56. [11] Jacob NJ. Promotion and use of improved cook stoves in the conservation of biomass resources and biomass briquettes from solid wastes in the Gambia. ISESCO J Sci Technol 2013;9:17e26. [12] Bwenge NS. The Effects of Adopting Improved Wood Stoves on the Welfare of Rural Women: A Case of Kibaha District in Tanzania [Master's Thesis]. Leeuwarden (Netherlands): Van Hall Larenstein Univ.; 2011. € hlin G, Hyde WF. Fuelwood, forests and community management [13] Cooke P, Ko e evidence from household studies. Environ Dev Econ 2008;13:103e35. [14] Gebreegziabher Z, van Kooten GC, van Soest DP. Land degradation in Ethiopia: what do stoves have to do with it?. In: Annual Meeting of International Association of Agricultural Economists. Gold Coast, Australia; 2006 Aug 12e18. [15] Haider MN. Success Without Subsidy: A Case Study of the Fuel-Efficient Smokeless Stoves Project. Islamabad, Pakistan: United Nations Development Programme; 2002. [16] GACC. Water Boiling Test Version 4.2.3. Washington, DC: Global Alliance for Clean Cookstoves; 2013. [17] Kothari CR. Quantitative Techniques. 2nd ed. New Delhi, India: Vikas Publishing House; 2004. [18] Roden CA, Bond TC, Conway S, Pinel ABO. Emission factors and real-time optical properties of particles emitted from traditional wood burning cookstoves. Environ Sci Technol 2006;40:6750e7. [19] MacCarty N, Ogle D, Still D, Bond T, Roden C. A laboratory comparison of the global warming impact of five major types of biomass cooking stoves. Energy Sustain Dev 2008;12:56e65. [20] Barrett H, Christopher B. Field Guide of Appropriate Technology. 1st ed. Cambridge, MA: Academic Press; 2003. [21] Jagger P, Jumbe C. Stoves or sugar? Willingness to adopt improved cookstoves in Malawi. Energy Policy 2016;92:409e19. [22] EPA. The Third National Report on the Implementation of the UNCCD/NAP in Ethiopia. Addis Ababa, Ethiopia: Ethiopia Environmental Protection Authority; 2004. [23] Tryner J, Willson BD, Marchese AJ. The effects of fuel type and stove design on emissions and efficiency of natural-draft semi-gasifier biomass cookstoves. Energy Sustain Dev 2014;23:99e109. [24] Yohannes S. Design and Performance Evaluation of Biomass Gasifier Stoves [Master's Thesis]. Addis Ababa (Ethiopia): Addis Ababa Univ.; 2011. [25] Habermehl H. Economic Evaluation of Improved Household Cooking Stove Dissemination Programme in Uganda. Eschborn, Germany: German Agency for Technical Cooperation; 2007. € hlin G. Urban energy transition [26] Gebreegziabher Z, Mekonnen A, Kassie M, Ko and technology adoption: the case of Tigrai, northern Ethiopia. Energy Econ 2012;34:410e8. [27] Dawit W. Fuel efficient technology adoption in Ethiopia: evidence from improved “mirt” stove technology: a case in selected kebeles from “Adea” wereda. Ethiop J Econ 2008;17:77e107. [28] Lewis JJ, Pattanayak SK. Who adopts improved fuels and cookstoves? A systematic review. Environ Health Perspect 2012;120:637e45. [29] Beyene AD, Koch SF. Clean fuel-saving technology adoption in urban Ethiopia. Energy Econ 2013;36:605e13. [30] Jan I. What makes people adopt improved cookstoves? Empirical evidence from rural northwest Pakistan. Renew Sustain Energy Rev 2012;16:3200e5. € hlin G. Biomass Fuel Consumption and Dung Use as Manure [31] Mekonnen A, Ko Evidence from Rural Households in the Amhara Region of Ethiopia. Addis Ababa, Ethiopia: Environment for Development Discussion Paper Series 08e17; 2008. [32] Chambwera M, Folmer H. Fuel switching in Harare: an almost ideal demand system approach. Energy Policy 2007;35:2538e48. [33] Makame MO. Adoption of improved stoves and deforestation in Zanzibar. Manag Environ Qual Int J 2007;18:353e65. [34] Muneer SET. Adoption of biomass improved cookstoves in a patriarchal society: an example from Sudan. Sci Total Environ 2003;307:259e66. [35] Axen GJ. Fuel Efficient and Efficient Aid: An Analysis of Factors Affecting the Spread of Fuel Efficient Cooking Stoves in Northern Tanzania [Bachelor's €derto €rn Univ.; 2012. Thesis]. Stockholm (Sweden): So [36] Levine D, Beltramo T, Blalock G, Cotterman C. What Impedes Efficient Product Adoption? Evidence from Randomized Variation in Sales Offers for Improved Cookstoves in Uganda. Berkeley, CA: University of California; 2013. [37] Fullerton DG, Bruce N, Gordon SB. Indoor air pollution from biomass fuel smoke is a major health concern in the developing world. Trans R Soc Trop Med Hyg 2008;102:843e51. [38] Bensch G, Peters J. A Recipe for Success? Randomized Free Distribution of Improved Cooking Stoves in Senegal. Essen, Germany. Ruhr Economic Papers, no 325. 2012. [39] Geary CW, Prabawanti C, Aristani C, Utami P, Desa YD. A Field Assessment of Adoption of Improved Cookstove Practices in Yogyakarta, Indonesia: Focus on Structural Drivers, 360. Durham, NC: Family Health International; 2012. [40] Troncoso K, Castillo A, Masera O, Merino L. Social perceptions about a technological innovation for fuelwood cooking: case study in rural Mexico. Energy Policy 2007;35:2799e810.