The study was conducted in Kisii town, the capital city of Kisii County which is one of the 47 Counties in the Republic of Kenya (shown in Figure 1). It shares common borders with Nyamira County to the North East, Narok County to the South and Homabay and Migori Counties to the west. The County lies between latitude 0˚40'38.4" South, and longitude 34˚34'61" East and covers an area of 1317.5 km². The county exhibits a highland equatorial climate resulting in to a bimodal rainfall pattern with average annual rainfall of 1500 mm. The long rains are between March and June while the short rains are received from September to November. Maximum temperature ranges between 21˚C and 30˚C while the minimum temperature range between 15˚C and 20˚C [24].

Kisii town is located 120 km South of Kisumu city and about 400 km West of Nairobi and covers an area of 35 km² out of which 15 km² falls within the central business district which is leasehold, while the rest consists of peri-urban settlements in freehold areas. It has an average population of 200,000 [25]. The town is located on a confluence of numerous valleys surrounded by gentle to steep hills. In the South West lies Nyanchwa Hills at an altitude of 1800 metres above sea level and to the North East are the Mwamosioma Hills rising to about 1800 metres above sea level. To the South is the mountainous Bobaracho and Gesarara rising up to 1950 metres above sea level. It is also drained by several streams which are tributaries of River Riana [24].

The study utilized both primary and secondary data. For primary data collection, two structured questionnaires containing both open ended and closed ended questions were developed and administered. The first questionnaire was administered on 436 households with valid responses received from 400 which accounts for 92% of the sample. The second questionnaire was administered on 40 wood fuel sellers with valid responses received from 95% of them. The questionnaire administered to households was structured into three parts. The first part comprised of the demographic information of the respondents such as age, sex, income, family size and educational level. The second section of the questionnaire gathered information on energy use profile. That is, types of fuel used by respondents and their costs per week, number of times they cook in a day, and whether or not their kitchen is ventilated. Finally, the third part of the questionnaire sought information on use of alternative biofuels and energy efficient stoves. The second questionnaire administered on wood fuel sellers (those selling charcoal and firewood) was divided into two parts. The first part requested for personal information of the respondents while the second part elicited information on the amount of wood fuel, source and quantity sold per week and lastly an inquiry on the quantity of charcoal that can be harvested from one fully grown tree.

The total population of Kisii Town in 2019 was 200,000 with averagely 4.1 people per household [26]. With this information, the Slovin’s formula was employed in calculating sample size for the household survey as follows:

Number of households = Total population of the Town Average number of households

Number of households = 200000 4.1 .

Number of households = 48,780.

The Slovin’s formula for calculating the sample size is:

n = N 1 + ( N × e 2 )

where

n = the desired sample population.

N = is the total population of interest.

e = is the error margin at 95% confidence level.

n = 48780 1 + ( 48780 × 0.05 2 ) = 396.746

Giving a non-response margin of 10%;

N 0 = n + 10 % n = 396.746 + 10 100 × 396.746 = 436.4186 ≈ 436

Therefore, the sample used = 436 households.

A systematic random sampling technique was applied in which the first household was randomly chosen and subsequent households chosen at intervals of three. The sampling interval was calculated by dividing the total number of basic sampling units (BSU) in the population by the number of sampling units needed for the sample. Priority was given to females because they are the ones majorly responsible for cooking in each household. For secondary data collection, relevant literature was reviewed including published and unpublished books, magazines, journals and reports relevant to this study. Government publications, thesis and internet were the other sources of secondary data.

Data analysis and presentation were done using SPSS and MS-Excel statistical tools. Frequency tables, graphs and charts have then been used to present the thus summarized data. Inferential statistics was applied to make inferences from the data collected using the Statistical Package for Social Scientists (SPSS). The study had three hypotheses. Analysis of variance (ANOVA) test was used to test the first hypothesis that, there was no statistically significant difference in extent of deforestation/biomass loss attributable to household consumption of traditional fuels. The second hypothesis that there was no statistically significant difference in the relationship between adoption of energy efficient stoves and income, was also tested using analysis if variance (ANOVA). Finally, regression analysis was employed to test the third hypothesis that, there was no statistically significant difference in barriers to the adoption of energy efficient stoves and alternative biofuels.

A majority of the respondents (72%) were females as shown in Figure 2. This can be explained by the fact women are more conscious of their new role and the implications for family dynamics [27] and decision-making regarding energy consumption [20]. Challenges were encountered during the survey, as most of the respondents were not willing to disclose their age. Nevertheless, from the responses obtained and presented in Table 1, respondents within the age group of 0 - 20 years had the least representation that is 1.0 percent of the sample. From the findings, the majority of those below 20 years or above 60 years could probably be school going children or the elderly hence less represented in the findings.

As shown in Table 2, most (42.5%) of household heads acquired tertiary education which conforms to previous reports from Kisii County Government of the high literacy levels at 10.45% [25]. Most (51.5%) of households’ size had 4 to 6 members (Table 3). [25] reported that in kisii town, the mean household size is 4 - 6 people. Most people (44%) had income ranging from Ksh 10,000 to 29,000 (Table 4) and 69.3% of interviewees cooked three times per day (Table 5). People with lower income tend to spend slightly more time on food preparation [28]. Majority (43%) of the respondents acknowledged trading as their main occupation.

As shown in Figure 3, charcoal was used for cooking by a majority of respondents followed by gas, firewood, electricity and kerosene. While about 88% of the respondents used charcoal, gas was used by 79.3%, firewood by 42%, electricity by

18.8% and kerosene by 10.3%. Thus, use of charcoal was most prevalent compared to other fuels. The use of different sources of fuels by the respondents interviewed was compared on the basis of average income, respondents’ occupation and the size of the respondent’s household. It was assumed that income level would influence the choice of fuel used. Findings (Table 6) showed that charcoal was the main energy form used by the respondents who earned an average income of between below Ksh 10,000 - Ksh 29,999. This could be interpreted to mean it was the most affordable in terms of cost in relation to their income. Kerosene and electricity were the least energy forms used by the respondents.

It was also observed that charcoal was used widely by respondents whose main occupation was trading (38.8%), followed by Civil servants and farming who accounted for 19.8% and 16.8% respectively (Table 7). Un-expectedly, the artisans constituted only 10.5% of the respondents who used charcoal. It was also assumed that the household size could influence the number of times meals were prepared in a household and ultimately reflected in the choice and amount of energy fuel used. Findings showed that highest amount of fuel consumed cut across all of the respondents. About 43.6% of the respondents whose household size comprised of 4 - 6 members consumed 1 - 5 tins of charcoal per week (Table 8). As earlier mentioned, respondents were asked to supply information on the number of times they prepared meals in a day to establish whether it had an effect in the choice and amount of energy fuel used. As shown in Table 9, most of the respondents cooked three times a day and charcoal was used by 62% of respondents, followed by gas (55%). Electricity and kerosene were the least

forms of energy used and accounting for 11% and 8% of the respondents in the sample study respectively.

Findings showed that the average weekly quantities of charcoal and firewood

used were 3.8 tins and 1.98 stacks respectively. The annual average resulted to 201.76 tins and 102.96 stacks respectively. These findings validate the report by [29] on the total annual charcoal consumption estimated at 1.6 million tonnes, generating an estimated annual market value of over US$ 427 m.

The respondents who sold wood fuel were asked to supply information about their sources of the wood fuel they sold. The respondents were also asked for information about the locality from where they obtained the wood fuel they sold. This information helped in the determination of whether the deforestation being investigated was within Kisii county or not. As shown in Table 10, only 10.3% of the respondents got their fuelwood, that is charcoal and firewood, supply from Kisii County while 89.7% got obtained fuel wood supply from other counties neighboring Kisii. These finding was helpful in attempting to answer the research question on whether deforestation leading to biomass loss could be happening in Kisii County as a result of household consumption of traditional fuels. It is evident that any ongoing deforestation in Kisii County might not be as a result of the prevailing fuelwood consumption as 89.7% of the fuel wood is obtained from other counties.

Previous studies have shown that charcoal and firewood consumption for cooking in urban areas had resulted in concerns over their environmental impact [30]. Annual biomass loss for this study was calculated as follows: According to the data obtained from the fuelwood sellers, the average Number of bags of charcoal that can be harvested from one fully grown tree is 5.65.

• Average number on bag from one full grown tree = 5.65 Bags.

• Amount of charcoal in (Tins) used per year for individual household = 201.76 Tins.

• Number of tins that make one bag is 50.

If 1 full grown tree gives 5.65 bags.

1 Bag = 50 tins what about 201.76 tins.

201.76/50 tins = 4.0352 bags of charcoal.

Number of trees cut per year per household therefore is 4.0352/5.65 = 0.7142 which is approximately one tree. Number of trees cut /year/households = 1.

Therefore, multiplying the number of trees cut/year/household with target population gives the estimated biomass loss. The target population, that is, the total number of households in the study area was 48,780. The number of trees in a hectare may vary from 1000 to 3000 [31] and average number of trees can be found by in a hectare is 1250. Thus 39 ha per year was found to be the actual the extent of biomass loss. The average amounts of charcoal and firewood used were calculated and the findings indicated a clearing of 39 hectares was required annually for charcoal production and consumption by the households in the study area. This is in agreement with the results obtained by [30] who reported that for the whole urban population, the forest area needed to meet consumption of charcoal and firewood would be from 34 ha to 274 ha. ANOVA test results established that there was a statistically significant overall difference in mean standardized weekly amount of charcoal and firewood consumed by households that contributed to deforestation/biomass loss F (14, 4424.191 = 1.848, p < 0.031). According to these results, consumption of traditional fuels at household level contributes to deforestation/biomass loss as [32] also stated that firewood and charcoal biomass are among the major causes of deforestation in the Sub-Saharan region of Africa.

The study also enquired from the respondents their willingness to adopt alternative biofuels and energy efficient stoves as well as the barriers encountered. A majority (63%) were willing to use alternative biofuels and energy efficient stoves if given an opportunity. The respondents were also asked to give reasons for their willingness to adopt alternative biofuels and energy efficient stoves. As shown in Figure 4, the reason given by a majority (52.4%) of the respondents was eco-friendliness, which accounted for previous studies showed that women, that are most of the time in charge of cooking, are very interested in improved cooking stoves [33]. Environmental factor is one of the major factors of biofuel adoption [34] [35]. On the adoption of energy efficient stoves, 24% of the respondents used Jikokoa, 17% used firewood cook stove, while 67.5% used ceramic

jiko stove. This is due to the affordability of ceramic jiko stove compared to the other energy efficient stoves. The study further investigated if there could be a relationship between adoption of energy efficient stoves and income. The results of ANOVA test show that the F statistic value is 2.732, with a p-value of 0.029 (which is less than the 0.05 alpha level) for Jiko koa. This means there is a statistically significant difference between the means of the different levels of average income of the respondents who either use or do not use Jikokoa. Thus, average income level was significant only in the case of adoption and usage of Jikokoa. This means that there is no statistically significant relationship between adoption of other energy efficient stoves and income. This study also sought to establish the barriers to the adoption of energy saving stoves and alternative biofuels among residents in Kisii Town. As shown in Table 11, the relatively high cost of purchase of energy efficient stoves was a major barrier, mentioned by 55% of the respondents, followed by inability to let go of traditional cooking practices mentioned by 35.3%. Indeed, cultural practices may be hard to break and such reluctance could greatly impact adoption of energy efficient stoves despite the benefits they come with. Other barriers mentioned included lack of knowledge on benefits of energy efficient stoves and lack of technical know-how for their production. The major reasons hindering the respondents from adopting alternative biofuels were cost (61%) and low energy content of some alternative biofuels (69.5%). Other barriers mentioned included the lack of technical know-how and, lack of government support. Generally, the cost factor cut across adoption of both alternative biofuels and energy efficient stoves and this may be attributed to the average income of the respondents.

According to [36] educational level of the household head and fuel processing requirement of energy efficient stove were found to be barriers for adoption. [37] also reported that the major barrier for adoption biofuel is affordability or financial capacities and the technology. Regression analysis was performed on the dataset in this study to further investigate whether the barriers specified by the respondents had any significant relation to the adoption of the use of biofuels.

Result obtained indicated a low positive degree of correlation between whether or not respondents used biofuels and the barriers specified by the respondents (R = 0.230). The R² (0.053) indicated that 23% of variation in biofuel use could be explained by the barriers specified by the respondents.

1) The findings of this study confirmed that charcoal was the main form of energy fuel used for cooking in Kisii town and that income level influenced the choice of energy fuel used by the households. It was established that charcoal mainly used by the respondents who earned an average income of between below Ksh 10,000 - Ksh 29,999 was the most affordable source of energy, while kerosene and electricity were the least affordable energy products for the respondents. The causality between level of income and energy consumption has been studied by many other scholars. [38] and [39] proved that the relationship between energy and income is not a neutral one, hence, energy consumption is influenced by the income variation.

2) Based on the sources of wood fuel sold by the respondents and the quantities of charcoal and firewood used, the extent of biomass loss was estimated at about 48,780 trees per year representing about 39 ha of forest land. However, it was established that only 10.3% of the respondents got their fuelwood from Kisii County while a whopping 89.7% got their fuel wood from other counties neighboring Kisii. Thus the attendant deforestation/biomass loss takes place outside Kisii County.

3) Majority of the respondents were willing to adopt the use alternative biofuels and energy efficient stoves largely due to their perceived eco-friendliness, availability and promotion of good health of users. However, affordability predicated on the cost factor is a major barrier to their adoption. It was thus established that households within Kisii town have knowledge of the various forms of alternative biofuels and energy efficient stoves even though the rate of adoption is still relatively low due largely to non-affordability.

4) Policy interventions and government sensitization towards adoption of alternative biofuels and energy efficient stoves in Kisii town are relatively weak and ineffective as most respondents mentioned the lack of government support among the barriers to adoption.

1) Kenyan Government should provide mass sensitization on the harmful health effects of unclean energy fuels and make plans to ensure these clean energy forms are affordable and available in all rural and urban areas of Kisii town.

2) The government and other stakeholders should promote local technologies for producing energy efficient stoves since they can be more affordable to the locals.

3) Civic education on the benefits of clean energy fuels to the environment and households should be promoted.