Factors Affecting Fuelwood Consumption and CO₂ Emissions: An Example from a Community-Managed Forest of Nepal

Abstract

Fuelwood is the primary source of energy in Nepal, where 87.1% of the total energy is derived from wood, making it the major source for carbon emissions. This study explores the factors affecting the fuelwood consumption, the amount of carbon emissions including the potential for carbon sequestration in community forests, taking a case study of Kankali Community Forest Users Group (CFUG) of Chitwan district of Nepal. Interviews with 217 households revealed that 60% of the households still depend on fuelwood for cooking, which apparently emits approximately 13.68 tons of carbon dioxide annually. The emission, however, varies with the economic status of the households; poor households rely exclusively on fuelwood for cooking and therefore emit greater amount of carbon. Similarly, the carbon emission was also found to be directly proportional to the family size and livestock holding, and inversely proportional to landholding and per capita income. A more conservation-oriented forest management along with activities to support livelihood has contributed to lower carbon emissions. Interestingly, the poverty-energy trap seemed to have a distinct gender dimension. We argue that CFUGs need to invest in income-generating activities for local users, and especially for women of low-income households, in order to reduce current carbon emission.

1. Introduction

Global warming has been a topic of discussion for the last three decades, where household energy consumption is one of the major anthropogenic contributing factors [1]. About 2.9 billion people in developing countries are still dependent on fuelwood for cooking and heating [2,3,4] and this has become one of the primary drivers of climate change. The sustainable development goals (SDGs) put the sustainable use of plant-based biomass as one of its central goals [4]. Therefore, it is important to assess the factors which encourage households to reduce the use of biomass, and shift towards more efficient fuels [4]. Forests provide the main sources of energy for the people living in rural parts of developing countries, and Nepal is not an exception. This has led to deforestation (e.g., annual deforestation rate in the Terai region is 0.44%) [5] and forest degradation. The energy mix pattern of Nepal shows that approximately 87% of the total household energy comes from fuelwood [6]. However, evidences on emission of CO₂ from fuelwood consumption are limited.

Understanding household energy consumption helps to understand carbon emission pattern which then can lead to effective policy formulation [3]. “energy ladder is reported as a common model to describe the household fuel choices in developing countries introduced by [3], where primitive fuels such as fuelwood and agricultural waste are replaced by transition fuels, such as kerosene and then advanced fuels, such as liquefied petroleum gas (LPG) and electricity in the processes of urbanization/development [3]. Albeit studies show that energy transition does not occur as a series of simple, discrete steps as the “energy ladder” implies. Instead, “energy stack” is more common, where with increasing income, households adopt new fuels and technologies that serve as partial rather than perfect substitutes for the more traditional ones [3]. Knowledge of how changes in economic disparity affect the pattern of energy use can help to understand the carbon emission pattern and identify the group where intervention is required.

The long-term community-managed forests (CMF) in Nepal have been contributing to various aspects of society including poverty alleviation [7,8,9]. Such community managed forests have not only enhanced the economy of low-income households but also contributed to the energy consumption pattern among different economic classes. Such practices have proliferated well in Terai and also in few hilly areas that lie close to market and road-head, which has led to the promotion of alternative energy sources, such as LPG. However, for large population, mainly rural, in Nepal, fuelwood is still a major source of energy.

In 1998, fuelwood derived from forests constituted the largest proportion of the total fuelwood consumption (78%) in Nepal, whereas this dependency has reduced to 64% in 2013 [10]. In rural areas, one-third of the forests are being managed by communities as CMFs in which 60.9% of all households still depend on the fuelwood [11]. The main goal of this study is to assess the energy consumption patterns and factors affecting carbon emission by exploring the situation in a CMF from the Terai region of Nepal. The objectives to meet this goal are to:

(1).

Assess the extent of fuelwood consumption and carbon emission from the community managed forest

(2).

Identify factors affecting carbon emission, such as income, household size, literacy and gender composition in the community managed forest

(3).

Examine the carbon balance situation in the community managed forest

Additionally, feminization of poverty has been an established fact [12], however, the links of gender-poverty and energy use (carbon emission) have not been adequately established [13]. Therefore, in addition to the specified objectives of energy consumption at all socio-economic class, in the study, the data is further disaggregated in the lines of gender to highlight a preliminary link between gender, economic class, and CO₂ emission.

2. Methodology

2.1. Study Site

The Kankali community managed forest that is situated in the Terai region of Nepal was selected for this study (Figure 1). The total area of forest is 749 ha and the total number of households in its Community Forest User Group (CFUG) is 2065. A long history of community-based forest management (since 1995) and income generation by selling forest products were two main criteria for the selection of the CMF. Also, the site is established as a permanent research site by the Institute of Forestry (IOF), Nepal and has been closely observed for a decade. The associated CFUG generates fund by selling forest products both to its members and the outsiders. The fund is ploughed back through direct and indirect investment in the forest conservation, community development, community awareness programme, also through subsidies on alternative energy and livelihoods improvement activities.

2.2. Data Collection

The selected CMF and CFUG is taken as a case to understand energy consumptions by the households within a year. Primary data were collected from the CFUG and complemented by the secondary data from the CFUG records and inventory of 2013 maintained by IOF. The unit of analysis for the social data is individual household and for the biophysical data, sample plot. The households were surveyed to gather details on family occupation and their dependency on forest resources or other alternative sources for the energy. Following the methodology described in [14,15], participatory well-being ranking was carried out as listed in the following steps:

• Using snowball sampling, key informants such as municipality secretary, teachers, local leaders, and CFUG executive members were selected for well-being ranking.
• Using the DFID framework and considering five different assets - physical, social, financial, natural and human [16] a well-being categorization was done for each household.
• The key informants classified each household in the CFUG into three categories; low, medium and affluent. The low-income households (n = 590), medium income households (n = 1150), and affluent income households (n = 325).

A stratified random sampling was carried out based on the well-being ranking of the households. Referring to [17], we estimated a sample size of 217 households where we interviewed 80 male and 137 women. We assumed a prevalence rate of 50% to allow maximum variability, with an allowable error of 10% at a 95% confidence interval. We divided the sample size (low income 60, middle 122 and affluent 35) based on population probability of the size of each well-being category. Women respondents accounted for more than half of the total respondents.

Respondents were interviewed with the help of a structured pre-tested questionnaire. The questionnaire was prepared by considering socio-economic variables such as the number of family members, education, asset ownership, household income, livestock number, and quantity of fuelwood used. The questionnaire was translated into the local language (Nepali) and tested before the interview. The first author (SB), along with trained field enumerators (two male and one female), carried out data collection between August 2016 and September 2017. The interview was carried out either in the morning or in the evening as per the convenience of the respondents so that the maximum amount of time could be invested in data collection and a maximum number of household members could be involved. Such measures were adopted to extract in-depth information and allow self-triangulation. We were able to capture detailed demographic and socio-economic information through a household survey.

In addition to structured interviews, three focus group discussions (FGDs) were also conducted with the participation of total of 47 CFUG members representing different economic groups. FGDs with each group were conducted separately to understand the role of community forests and how community forestry is promoting the use of alternative energy. Likewise, 14 key informants (six female, eight male), including political leaders, school-teachers, business persons, and executive committee members, were also interviewed. After completion of the household survey, information was cross-checked and verified with randomly selected households in each settlement to avoid discrepancies and inconsistencies in data. The open-ended questions were coded before the information was entered into the computer. Data analysis was performed in SPSS (version 24).

The forest inventory data of 2013 were taken from the recorded data of IoF. By following the basic plot design of [18], the data for 2016 was collected by the authors (SB, BB and KG) and the enumerators, which includes data of three nested sub-plots. Trees with a diameter at breast height (DBH) of above 10 cm were measured. The parameters included species identification, positioning, DBH, and height of trees from 57 plots.

2.3. Data Analysis

2.3.1. Socio Economic Model Specification

Before analyzing the household consumptions of different sources of energy, we needed to unpack various parameters which depend on household consumption, such as financial and social status of the households [19].

We hypothesized that socio-economic factors and carbon emissions have a strong relationship. Nevertheless, the pervasive role of economic factors in the consumption of fuelwood cannot be ignored. We wanted to identify the most influential parameters for the consumption, such as land ownership [20], literacy (number of educated family members), income [21], livestock number [22], family size, awareness, alternative energy use and livestock unit. We hypothesized fuelwood consumption (Fwc) is measured on the basis of socio-economic variables represented in Equation (1) and

Table 1. Explanatory variables expected direction of relationship with response descriptions.

Independent Variables	Explanation	Expected Sign	References
Well-being	Categories of household according to economic class (poor = 0, medium = 1 and affluent = 2)	-	Baral et al., 2019 [23], Rezitis and Ahammad, 2016 [24], Rao, 1990 [25]
Family size	Number of household members	+	Rao, 1990 [25]
Literacy	% of educated household members above five years	-	Van der Kroon et al., 2013 [3], Chhetri et al., 2012 [26]
Per capita income	Total/gross household income (salaries or cash-in-hand/ad-hoc) of all family members	+	Charlery and Walelign, 2015 [21] Rao, 1990 [25]
Landholding	The total area of land owned by the household, including renting out and barren land (hectares)	-	Filmer and Pritchett, 2001 [20] Rao, 1990 [25] Van der Kroon et al., 2013 [3]
Livestock Unit	The number of livestock units owned by the household. (Adult female buffalo is considered as 1, adult male buffalo as 0.76, adult cow as 0.69, adult ox as 0.89, adult male sheep/goat as 0.23 and adult female sheep/goat as 0.20. c.f. (HMGN/ADB/FINNIDA, 1989 cited in [27].)	+	Schellenberg et al., 2003 [22]
Forest-based income	The total income (i.e., permanent or temporary job) from forest	-	Walelign and Jiao, 2017 [28]; Walelign, 2016 [8]; Gauli and Hauser, 2011 [29]
Awareness	Awareness regarding alternative energy (Yes or No)	-	Toft et al., 2015 [30]

Note: + for the positive relation and − for the negative/inverse relation between fuelwood use and variables.

.

Fwc_i = a₁ well-being + a₂familysize_i + a₃literacy_i + a₄Percapita + a₅landhol + a₆LSU + a₇incomefbemp + a₈awareness + e_i

(1)

We used a linear regression model to examine the relationship between fuelwood used from the community-managed forest in relation to carbon emissions. Enter regression method was used in the regression model with assumption that all the explanatory variables have some influence on dependent variables. We considered the quantity of the fuelwood use (in kg) as a dependent variable and all the associated socio-economic variables as independent variables (see Table 1 for details).

2.3.2. Estimation of the Potential of Carbon Emissions and Sequestration

Carbon sequestration is calculated by exploring the differences of the growing stock within a year. To estimate the potential carbon sequestration forests were classified into two classes: shorea robusta dominated forests and other Terai hard wood forests. The differences in the growing stock of 2016 and 2013 was analyzed using the volume functions derived by [31] assuming that the calculated average increment can serve as a proxy for the above ground carbon.

Carbon consumption of a household is estimated from the annual household fuelwood consumption. To compute carbon dioxide (CO₂) emissions, we used the conversion factors for fuelwood into CO₂ equivalents based on the Intergovernmental Panel on Climate Change (IPCC), 1996 revised report on Guideline for National Greenhouse Gas Inventory. One bhari (back load) of fuelwood is equivalent to 40 kilograms. The biomass of fuelwood is converted into carbon by the following relation;

Carbon dioxide emission (CO₂ e) = biomass of fuelwood * 0.47 (carbon) * 3.67 (CO₂ equivalent)

(2)

Likewise,

Carbon sequestration (CO₂ s) = growing stock increment of forest per year * 0.47 (carbon) * 3.67 (CO₂ sequestration equivalent)

(3)

3. Results

3.1. Socio-Economic Characteristics of the Sample CFUG

Table 2. Socio-economic parameters of the sampled community.

	Minimum	Maximum	Mean	Std. Error
Well being	0	2	0.88	0.04
Family Size	1.00	12.00	5.0553	0.13
Per capita income	5000	233,333.3	52,758.5	2770.20
Fuelwood	0.00	5760.00	1720.9217	85.77
Literature	0.00	1.00	0.7024	0.012
Landholding	0.00	1.91	0.4129	0.024
Livestock unit	0.00	8.00	1.2568	0.08
Awareness	0.00	1.00	0.4516	0.03
Forest-based income	0.00	12,400.00	1796.1290	112.05

Source: Household survey 2016–2018.

shows that, in the studied community, the family size ranged between 1 to 12, with an average family size of 5.05. (Standard Error (SE) 0.13). Likewise, the per capita income of households ranged from NRs 5000 to 233,333.3 with an average income of NRs 52,758.5 (SE 2770.2). Annual household fuelwood consumption varied from 0 to 5760 kg per household, with an average of 1720.9 kg (SE 85.7). The high standard errors of both per capita income and fuelwood consumption indicate that there is high variation. The maximum livestock(Livestock refers for the cow and buffaloes) holding is eight whereas minimum livestock holding is 0, with an average of 1.2 (SE 0.089). Likewise, landholding ranged from 0 to 1.91 ha, with an average of 0.41 ha (SE 0.024).

3.2. Status of Energy Consumptions by Households

Figure 2 shows that more than one-third of all households depend solely on LPG as an energy source, followed by a combination of LPG and fuelwood. There are no households that exclusively used biogas, as it is supplemented with fuelwood or LPG, or both. Very few households (i.e., 14.7%) depended solely on fuelwood for energy. Nevertheless, fuelwood remains a primary source of energy across surveyed households.

The results of Pearson correlation test between fuelwood consumption and other socio-economic variables, is shown in

Table 3. Correlation between low carbon emission and other socio-economic variables.

Variables	Pearson Correlation	P-Value
Well-being	−0.198	0.003
Family size	0.247	0.000
Literacy	−0.155	0.023
Ln per capita income	−0.275	0.000
Ln forest-based income	0.116	0.087
Landholding	−0.066	0.334
Livestock unit	0.134	0.043
Awareness	0.098	0.149

. Well-being has significant negative correlation with fuelwood consumption indicating that the affluent households consume lesser amount of fuelwood than poor households. Family size has significant positive correlation with fuelwood consumption, indicating that households with bigger family size consume more fuelwood. Likewise, literacy shows an inverse relationship with fuelwood use suggesting the higher number of educated family members, lesser was the amount of the fuelwood consumed. Per capita income is highly significant and was inversely related to fuelwood consumption; the higher the income lesser the dependency on forest resources. Results also show that households possessing a greater number of livestock consume more fuelwood.

3.3. Factors Affecting Carbon Dioxide Emission

The results from the multiple linear regression analysis (

Table 4. Factors affecting carbon dioxide emission.

Variables	Coefficients	Standard Error
Constant	10.635 ***	2.611
Well-being	−0.064	0.344
Family size	0.348 ***	0.096
Literacy	−1.584 *	0.905
Per capita income	−0.565 ***	0.213
Income from forest	0.197	0.123
Land holding	−0.269	0.464
Livestock unit	0.348 **	0.188
Awareness	0.188	0.366

Significance: * 10%; ** 5%; *** 1%.

) shows that the family size, per capita income and livestock income and family literacy were found to be significant determinants of fuelwood consumption (p = 0.000; 0.009; 0.002; 0.082, respectively). The positive coefficient value indicates that larger the number of family members, higher the fuelwood consumption. This could be attributed to the family needs and the availability of time to engage in the collection of fuelwoods. Likewise, households possessing a higher number of livestock consume more fuelwood, which is related to the fact that additional fuelwood is needed to prepare food for the available livestock. Generally in winter season the households provide kudo (Kudo is a type of cattle-feed prepared by cooking cereal flour in water with little salt, which is then diluted in warm water.) to livestock. However, no fuel wood is used exclusively for household heating purpose. Cooking also results to indoor heating. The negative coefficient value of per capita income shows that increase in family income decreases the fuelwood consumption. People tend to switch to other alternative energy sources particularly LPG, when the income starts to rise. Similar is true for the relationship between literacy and fuelwood consumption. Educated households use less fuelwood. This could be related to the observation that better-educated households know about the health impact of indoor pollution of fuelwood burning and seek for alternative energy such as LPG. As the value of Variation Inflation Factor (VIF) of all the variables is less then 10, collinearity problems among the explanatory variables do not exist.

3.4. Fuelwood Consumption and Carbon Emissions

The per capita carbon emission from the poor, medium and affluent households are 4.12, 2.63 and 2.15 tons, respectively (

Table 5. Economic classwise household fuelwood consumptions and CO₂ emissions.

Economic Class	Fuelwood Use (kg per Year)	CO2 (Tons per Year)
Low income	2392.0	4.12
Medium	1526.5	2.63
Affluent	1248.0	2.15

Source: Household survey 2016–2018.

). This calculation shows that low income households emit nearly two-fold CO₂ compared to affluent ones.

The inventory data reveals that carbon sequestration is 17.56 ton/ha while that of emission is 13.68 ton/ha, indicating that current emission is lower than sequestration. This is mainly because of control of fuelwood harvesting from the community forests and a shift to other alternative sources of energy.

3.5. Intra-Household Gender Analysis of Energy Provisioning

Women from low-income households are responsible for provisioning fuelwood for everyday use, which situates them at the bottom of the energy ladder. Our data highlights the highly gendered nature of poverty-energy trap which distinctly affects women from low-income group the highest. While 64% of women from the low-income group collect fuelwood for daily provision, only 36% of men from the same income group are involved (Figure 3). Non parametric Chi square test of independent of attributes shows that there is significance association between involvement of people for fuelwood collection with respect to gender on diffenent well being groups (p < 0.05). Further more, participation of females are higher than that of males in each well being groups. Hence, in an intra-household situation, because of their disproportionate share in fulewood collection and traditional gender roles, women from low-income group are the biggest users of fuelwood and hence highest carbon emitters. Despite their higher dependency on fuelwood, representation of the low-income group in the CMF is the lowest among economic classes. Furthermore, women who are primarily responsible for daily household energy provision from the CMF, have negligible presence in decision making bodies such as CFUG executive positions. With 31 men from low-income class represented, only one woman from the same class was found to be in the executive committee which thereby restricts/limits their possibility to ascend in the energy ladder. The representation of women, however, increases moving up the economic ladder.

4. Discussions

Among the various energy sources utilized by human society to fulfill their energy needs biomass still continues to be the primary energy source in rural areas of developing countries as Nepal [32]. About one billion people in Asia depend on biomass as their main source of energy [33]. Our findings also show that although the households from studied CFUG used energy from different sources, the majority of them still depend on biomass (i.e., fuelwood in this case). Nepal is one of the highest traditional fuel consuming countries in Asia because of its high dependency on traditional biomass fuels, mostly fuelwood [34]. This further resonates with the findings of several researchers [33,34,35].

The results showed that 60% of the households were dependent on multiple types of fuel for cooking primarily fuelwood followed by LPG, biogas and electricity, which also resonates with the finding of [36,37] that globally 40% of the population is fully dependent on biomass for cooking. Ref. [38] argue that in most of the developing countries people are trying to explore the low emission biomass energy for cooking. With respect to the economic classes, the household dwelling size per capita has a significant negative effect on fuelwood consumption which is line with the finding of [39] that wealthier households tend to consume less wood.

The regression analysis shows that there are various underlying factors such as per capita income, literacy, family size, livestock and landholding which determine the use of fuelwood as the household energy source, and similar findings were also found by previous studies elsewhere, e.g., [24,25,40]. For example, the relationship between household income and fuelwood consumption indicates that the higher the income, the lesser the fuelwood consumption [40]. As income increases, the use of alternative energy sources, such as kerosene and LPG, is gaining ground in Nepal and is reducing the need for fuelwood [41]. Similarly, family size is one of the underlying factors, as the fuelwood use increases with an increase in family size. Our study shows the low economic households have tendency of greater family size, which results in more fuelwood use. Ref. [42] illustrates similar finding in a mountain area of Bangladesh where family size, income, amount cooked, and burning hours significantly affected the amount of fuelwood used per family per year. Taking into account different family sizes, our study found that 2392 kg fuelwood per family per year is consumed by low-income households which is almost two times higher compared to the affluent households.

Landholding size has been identified as another factor influencing fuelwood consumption. Households having higher landholding and higher income are found to extract lesser quantities of fuelwood from CMF [43]. In our case, more the landholding, less the household fuelwood need since land holding is also a marker of economic status. Generally, the households with higher landholding size are affluent households, hence their fuelwood consumption is less. Furthermore, middle and affluent classes have diversified energy use and hence they do not extract as much as fuel wood from CMF in comparison to low-income groups. In addition, an increase in the number of livestock units also increases fuelwood consumption because many people use fuelwood for preparing food for their livestock; this is the most common practice in low economic families.

This study also found that the relationship between the economic group and fuelwood energy consumption has very distinct gender characteristics. In an intra-household analysis, it was observed that the male-female share of fuelwood collection tilted more towards women. While 70% of women from low-income households are involved in fuelwood collection for daily use, only 30% of men from the same economic class are involved for the same task. Our finding is consistent with those of the previous studies, which show women (and children) as primary collectors of fuelwood from forests on daily basis [44,45]. Due to the gender division of labour within households, women across different economic classes, have the responsibility of not only meeting energy requirements of the household by collecting fuelwood from the forest but also wisely using them in the kitchen and deciding on energy use [46]. thereby making them the biggest user and highest emitter in an intra-household scenarioThere also exists gender-based differences in collecting the types of fuelwood. Women collect branches and twigs for everyday use, whereas men carry logs, which require tree felling, which happens occasionally in all CMF. The data shows that the number of bhari of wood carried by men across all economic classes is larger than that of women. Furthermore, women’s role in energy provisioning is tedious compared to men, which implicates time poverty and a loss of opportunity to engage in other productive activities that generate income and hence pull them out of poverty trap. Moving up the energy ladder is necessary for moving higher in the economic status [47]; however, our case study suggests a possible vicious energy-poverty cycle where women of low economic classes, lack agency to make a change in their energy use. The situation of women from this economic class is further exacerbated by their low representation in CFUG executive positions.

The CMF plays a vital role in promoting energy alternatives by offering subsidy on installing biogas in individual households. In addition to that, the CMF conducts awareness-raising programme which supports the changes in the energy consumption behavior of users. Growing researches around the world have also found similar findings, e.g., [9,35]. Since it is women who bear the disproportionate share of fuelwood energy use [11], such interventions from CMFs have to be targeted at women for the desired result of reducing CO₂ emission. In our study site, the representation of women in the CFUG executive committee is very low compared to men which further decreases while moving down the economic classes. Hence, women, who are highly dependent on the CMF to meet their everyday energy provision, have very limited chances of influencing resource distribution and allocation through CMF-related activities. Livelihood activities of CFUGs that contribute towards decreasing of fuelwood among women of low-income group not only contribute towards decreasing the global CO₂ emission but, as a more direct benefit, improve the air quality of the immediate environment preventing any associated negative effects on the health and overall family wellbeing, providing better chances of moving up the economic ladder [48]. Our study reveals that the CO₂ emissions are low compared to the potential for carbon sequestration. This retrospectively suggests that CMF is contributing towards green economy development [49].

5. Conclusions

The energy consumption of the studied CFUG is similar to other developing parts of the world. A majority of households depend on biomass (fuelwood) for energy. However, the dependency on energy sources vary over time and with the socio-economic conditions; high-income households relying on alternate source of energy such as the LPG, while the poor continue to use fuelwood. The dependency on CMF for the fuelwood collection is still high, mainly for the low economic class households with low landholding, and large family sizes. The energy consumption pattern also has a distinct gender dimension as women disproportionately bear the drudgery of collecting fuelwood from CMF for daily use are causing additional pressures and gender imbalances trapping them in poverty–energy cycle. The current assessment reveals that the community had a positive carbon balance, which reveals that community forestry can contribute to a shift towards a green economy perspective. The use of forest resources, such as fuelwood can be reduced by increasing the level of income per capita. Focusing on such activities, especially on women of the low-income class can help to break the vicious poverty-energy cycle. In addition, promotion of alternative energy-efficient provisions can support low carbon emissions which can address various challenges related to carbon emission and sustainable management of energy in developing counties such as Nepal. It is the responsibility of the federal government to develop a effectivepolicy in order to promote low carbon investment and attain economic growth at national, provincial and local levels.