Socioeconomic and Environmental Impacts of Charcoal Production Activities of Rural Households in Mecha District, Ethiopia
Socioeconomic and Environmental Impacts of Charcoal Production Activities of Rural Households in...
Tassie, Kassahun;Misganaw, Berihun;Addisu, Solomon;Tesfaye, Ermias
2021-06-16 00:00:00
Hindawi Advances in Agriculture Volume 2021, Article ID 6612720, 16 pages https://doi.org/10.1155/2021/6612720 Research Article Socioeconomic and Environmental Impacts of Charcoal Production Activities of Rural Households in Mecha District, Ethiopia 1 2 2 1 Kassahun Tassie , Berihun Misganaw , Solomon Addisu , and Ermias Tesfaye Bahir Dar University, College of Agriculture and Environmental Science, Department of Agricultural Economics, Bahir Dar, Ethiopia Bahir Dar University, College of Agriculture and Environmental Science, Department of Natural Resource Management, Bahir Dar, Ethiopia Correspondence should be addressed to Berihun Misganaw; berihunmisganaw1@gmail.com Received 24 November 2020; Accepted 4 June 2021; Published 16 June 2021 Academic Editor: Volkan Okatan Copyright © 2021 Kassahun Tassie et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Ethiopia is one of the largest charcoal-producing countries in Africa where its urban consumers burn over 3 million tons per year. +e purpose of this study was to measure the amount of charcoal produced and its related environmental and socioeconomic impact in the study area. A total of 305 respondents were selected by using a simple random sampling technique. +e amount of greenhouse gas emissions from charcoal production was analyzed based on the Intergovernmental Panel on climate change quantification techniques, and the impact of charcoal production on households’ income was analyzed using propensity score matching. +e results revealed that the annual charcoal production rate and emission of carbon dioxide equivalent have an increasing trend at an alarming rate in the study area. From propensity score matching analysis, the economic impact of charcoal production has a positive difference of 0.43813162 as compared to nonproducers. Socioeconomic factors like land size, eucalyptus coverage, agricultural extension, market distance, and the number of oxen have a highly significant effect but variables like sex, family size, education status, credit services, and marital status had no significant effect on charcoal production. In general, even though charcoal production is economically having a positive impact on households’ annual aggregate income; it has dis- proportionality adverse effect on the environment like air pollution in addition to sophisticated respiratory health problems. +erefore, responsible institutions and planners should have focused on the multidimensional effect of traditional charcoal production on environmental issues and sophisticated health problems especially on employed laborers and nearby residents. livelihood and environmental degradation in rural areas of 1. Background of the Study the African continent [3]. Production of traditional charcoal Globally, charcoal production trends between 1965 and 2005 serves as a lifeline for the increasing populations in less show increasing production levels with Africa topping the developed countries in general and SSA nations in partic- chart [1], and approximately 1.5 billion people in developing ular. Due to low cost as compared to other energy sources, countries drive more than 90% of their energy from charcoal the traditional energy alternative charcoal remains the most for cooking and heating. Africa’s charcoal production has used, and it is expected that about 2.4 billion people rely on it moved from about 18.5 million tons in 1965 to about 49 [4]. million tons in 2005. Africa is closely followed by Latin Charcoal consumption in the majority of many Sub- America and the Caribbean, producing less than five million Saharan Africa (SSA) is expected to double by 2030 and tons in 1965 and about 5.5 million tons in 2005 [2]. Charcoal fuelwood consumption used for charcoal production is is the primary urban fuel supply and a major source of estimated to be 544.8 million m and 46.1 million tons, 2 Advances in Agriculture socioeconomic impacts related to charcoal producers respectively [5, 6]. And traditional charcoal production has been a growing concern due to its threat of deforestation, during extraction and production phases. +ere is little information concerning the socioeconomic land degradation, and climate change impacts [7]. Its sat- uration ranges from 54 to 71% in urban areas, and it is the and environmental impacts of charcoal production activities main fuel for more than 1 million families in SSA [8]. +ey of rural households in Mecha district, Ethiopia. Further- added that because of the increasing use of charcoal in many more, the effects of charcoal production on human health countries; it is critical to assess and develop long-range and environments are not clearly understood, and also the charcoal policies for African and other developing countries assessment of charcoal production in the Mecha district is [9]. Charcoal production is an important economic activity not well studied. Perhaps, the most serious of these is the harmful effect on the environment and climate changes, by providing a considerable amount of employment and also serves as a source of livelihood for most rural households [3]. which both have consequences for human health and pol- lution. +erefore, this study needs to address the effect of +e rapidly increasing demand and urbanization in many developing countries have not seen an increase in modern charcoal production on the environment and human health in Mecha district, Ethiopia. +erefore, the reason that the fuels (kerosene, LPG, natural gas, and electricity) supplied over the period, making traditional energy source use un- researchers want to conduct this study is to provide answers avoidable. As a result, traditional and indigenous energy to the following research questions: why do charcoal pro- sources like charcoal have been humankind’s first source and ducers engage in charcoal production? and are charcoal the most used [4]. producers willing to patronize alternate sources of income? In most developing nations, charcoal makers use tra- Answers to these questions provide insight into how to deal ditional and inefficient means; they build temporary earthen with the environmental problems presented by charcoal production, which ensure the sustainable management of kilns for earth batch [10]. +e wood from natural forest and farm clearing is stacked compactly in a pit, and the stack is the environment and reduce the health impact of charcoal production. covered with straw or other vegetation types and then buried under a layer of soil. After some arrangement, it was kindled with burning embers introduced at one or more points at the 2. Objectives of the Study bottom of the stack. +e task of the charcoal maker throughout the ensuing burn is to open and close a suc- +e specific objectives were designed as follows: cession of vent holes heating the wood while burning as little (i) To quantify the amount of carbon emitted from of it as possible [11]. charcoal production Most environmentalists agreed and feel that the tradi- (ii) To assess the environmental impact of charcoal tional method of charcoal production should be stopped production because of its destructive nature as presently practiced in most nations [12]. However, Arnold and Persson [13] (iii) To estimate the socioeconomic impact of charcoal asserted that both rural and urban residents in less devel- production on households of the study area oped nations have a strong appetite for charcoal use. +erefore, attempts to ban the production or the use of 3. Materials and Methods charcoal will be mostly inefficient, and it has the ability to cause adverse health effects such as the respiratory illness of 3.1. Description of the Study Area. +e study was conducted women and children and indoor air pollution at a significant in Mecha district, which is one of the 106 districts of the level [14]. Charcoal producers can use free raw materials Amhara Regional State and found in the West Gojjam Zone. collected from the forest or other sources and turn them into Mecha district is one of the 15 districts of West Gojjam a marketable commodity in high demand. Moreover, this Administrative Zone (see Figure 1). It is bordered in the traditional production of charcoal has a result of forest loss, south by Awi zone and Sekela district, in the west by south and this threatens to reduce the ability of the forest to Achefer District, in the northwest by the north Achefer provide unlimited and multidimensional vital services [15]. District, in the northeast by the Bahair Dar zuria, and in the For example, in Ethiopia, about 230000 tons of charcoal are east by Yilmana Densa district. +e administrative center of used per year for domestic purposes [16]. the district is Merawi town. Merawi town is found 525 Charcoal production in Ethiopia has limited success kilometers far from the Capital City of Ethiopia (Addis stories to offer because charcoal producers do not follow Ababa) and 35 kilometers far from the regional capital any standardized methods or technology. In Ethiopia, a (Bahir Dar city). From the total 43 rural Kebeles in the study careful assessment of charcoal production through the area, more than 18 of them have access to permanent rivers, traditional techniques revealed an average 24% loss of and the district has huge groundwater potential [19]. timber and nontimber forest products [16, 17] while no +e study district has 43 rural Kebeles. +e total area of studies have investigated in depth the health, environ- the district is about 156,027 hectares. From this, 72,178 mental, and social risks associated with the production of (46.25%) hectares (nearly half) are used for cultivation. this highly demanded energy source. Smith [18] noted Forestland and grazing land cover 18,547 (11.88%) hectares that health-related impacts associated with charcoal and 15,591 (9.99%) hectares, respectively [20]. Mecha dis- ° ° production have focused on effects from their con- trict is globally located between 11 5′N 11 38′ latitude and ° ° sumption; however, little is known about the health and 36 58′ 37 22′E longitude. +e altitude of the district ranges Advances in Agriculture 3 240000 260000 280000 300000 320000 340000 360000 0 5 10 20 km 240000 260000 280000 300000 320000 340000 360000 Ambo mesk Enamirt Merawi town Mecha_district Figure 1: Map of the study area. from 1,800 to 2,500m (2137m with an elevation) above sea Kebele. So, respondents of charcoal producers and non- level. charcoal producers and employed laborers on charcoal production took part in the survey, and the total producers, nonproducers, and workers were selected randomly from 3.2. Data Types, Source, and Methods of Data Collection. two Kebeles, respectively. +e population had taken as the +e study was relying on both primary and secondary data. total number of households in Mecha district, which is 39992 +e primary data on the sociodemographic characteristics of [21]. Accordingly, sample households from two sample respondents, health, and safety problems arising from Kebeles were selected by using the lottery method. +e charcoal production and other variables which were relevant sample size was determined using the formulas used to to the study were collected using a pretested structured determine sample households to be taken for household questionnaire and semistructured questionnaires through a interviews used when the population is >10000 [22]. household survey. Secondary data were collected from the Sample households are the main primary data sources of Mecha district agriculture office, land administration office, this study. But determining research sample size is a trade office, environmental protection office, market in- function of different factors like resource, time, the purpose formation, research papers, land use regulation policy of the study, characteristics of the population, etc. So, to document, demographic and socioeconomic profiles, and determine the sample size, we used scientific formula, and a health office which were used as data source. Secondary data critical component of sample size formulas is the estimation collected from the above institutions were used to quantify of variance in the primary variables of interest in the study carbon emission from charcoal production activities in the [23]. For the categorical dependent variable, 5% margin of study area and to assess infectious disease prevalence related error is acceptable, and for the continuous dependent to charcoal production. variable, 3% margin of error is acceptable [24]. +e formula is given as 3.3. Sample Size Determination and Sampling Techniques. Z PQ (1) n � , Based on the 2007 national census conducted by the CSA of Ethiopia, the district has a total population of 292,080, with men and women having 34858 and 5134 households, re- where n is the required sample size (when the population is spectively, in 43 rural Kebeles. Ambomesk and Enamert >10,000), Z is the confidence interval at 95% which is 1.96, Kebeles were selected purposively because of the greater P � 0.72 (P refers to the proportion of the population who quantity of charcoal produced than other Kebeles in the produce charcoal), Q �0.28 ((Q �1-P) refers to the pro- district. +e sample households from two Kebeles were portion of the population who do not produce charcoal) selected by using a simple random sampling technique based based on previous research findings, and d refers to the on the sample frame adopted from respective administrative desired precision of the estimates (within a range of plus or 4 Advances in Agriculture minus 5%). So, using the above scientific sample size de- methods. Finally, the conclusion and recommendation were termination formula, one gets formulated based on findings. (1.96) × 0.72 × 0.28 3.8416 ∗0.72 ∗0.28 n � � n � � 310. 3.5. Impact Assessment: Propensity Score Matching Model 0.0025 0.05 Analysis. According to Caliendo and Kopeinig [28], there (2) are steps in implementing PSM. +ese are an estimation of the propensity scores, choosing a matching algorism, +erefore, n �310 was the minimum sample size of checking on common support conditions, testing the charcoal producer households for reliable results. Finally, by matching quality, testing the standard error, and testing using the proportional allocation method, the researchers sensitivity analysis. +e first step in the PSM method is to were decided to take sample households from two selected estimate the propensity scores. When estimating the pro- Kebeles. Standing on this, a total sample of 310 households pensity score in the binary treatment and control groups, from which 150 charcoal producers and the remaining 160 binary logit and probit models usually yield similar results nonproducers were selected randomly from generated strata except the assumption of error terms, and the choice is based sampling frame from roasters of each sample Kebele ad- on the researchers’ preference and simplicity for presenting ministration office and calculated proportionally based on the result [29]. +erefore, the logit model was applied to the number of producers and nonproducers, respectively predict propensity scores for the PSM method in this study. (see Table 1). 3.6. Specification of the Binary Logit Model. +e logit dis- 3.4. Data Analysis Methods. +e data gathered from both tribution has more density mass in the bounds, and it is the primary and secondary sources were analyzed using qual- best model to predict the probability of a household to be itative and quantitative methods. Qualitative analysis usually influenced by urban expansion, i.e., to predict propensity relies on inductive reasoning processes to interpret and scores, based on which, the affected and control groups have structure the meanings that can be derived from gathered to been matched using the nearest neighbor matching information [25]. +e data collected from the Mecha district (bandwidth 4) estimator. In estimating the logit model, for agriculture office that was the amount of charcoal produced impact analysis, the dependent variable is a producer that annually from the district at the hole to quantify the amount takes a value of 1 and is a nonproducer that takes the value of of greenhouse gas (GHGs) emissions from the production of 0. Yang [30] and Nguyen [31] also noted that the logit model charcoal were analyzed based on IPCC 2006: which has more density mass in the bounds could be used to estimate the propensity score p(x). Mathematically, the E � Fc × EF , (3) GHGi i GHGi model can be expressed as [32]: the logistic function is given after some mathematical work: where E is the quantity of GHG , Fc is the quantity of GHGi i i fuel type combusted (net), calorific value quantity �energy ln � β + β X + U . (4) 0 i i i input (T), and EF is the emission factor for a certain GHGi 1 − P i�1 GHG . +e emission of GHGs from charcoal involves two In determining the impact of charcoal production using phases: the production phase and the consumption phase. PSM, an impact assessment must estimate the counterfac- One kilogram of charcoal assumes to be produced from 6kg tual; that is, what would have happened if engagement in the wood, so sample charcoal would be measured and multiplied charcoal production was never undertaken or what other- by 6 conducted to estimate the amount of firewood pro- wise would have been. To determine the counterfactual, it is duced by the households. CH gas has a GWP of 21 times essential to net out the effect of charcoal production from greater than CO and that of N O is 310 times greater than other factors. +e choice of a good counterfactual is, 2 2 CO [26]. Emission factors for three major GHGs from therefore, crucial in impact assessment. +is is done through charcoal (CO , CH , and N O) due to household energy the use of control and treatment groups, and the only dif- 2 4 2 consumption are given in IPCC guidelines [27]. ference between the two groups is that treatment groups Environmental impact of charcoal production in the participate in charcoal production only. Propensity scores study area like climate variability, mean rainfall, and tem- are an alternative method to evaluate the effect of receiving perature variability was analyzed using Microsoft Office treatment when a random assignment of treatments to Excel Software, and simple descriptive statistics like the subjects is not feasible. Propensity score matching states the pairing of handling and control elements with comparable percentage and mean of the data collected, employed la- borer, and health problem faced were analyzed with chi- values on the propensity score, and possibly other covariates, square analyzed using SPSS version 20. After data sets are and the discarding of all unmatched units. It is principally collected, the researcher encodes them into STATA version used to compare two groups of themes but can be applied to 13. +e socioincome impact of charcoal production was analyze more than two groups. As [33] suggests, PSM works analyzed by using propensity score matching/PSM/. +e well as long as the survey instrument used for measuring findings from the analysis were presented by using de- outcomes is identical for treatment and control participants. scriptive statistics which include meaning, frequency dis- Hence, the success of PSM depends on the quality of data tribution tables, percentages, and standard deviation available and the variables used for matching. Advances in Agriculture 5 Table 1: Population and sample size of each sample Kebele. Sample HHs Sample Total number of Estimated number of Estimated number of Total Kebeles households estimated charcoal producer HHs noncharcoal producer HHs sample Producers Nonproducers Ambomesk 1125 513 612 77 92 169 Enamert 931 482 449 73 68 141 Total 2056 995 1061 150 160 310 As a charcoal production impact evaluation technique, probability. In this study, the PSM was used to evaluate the PSM is based on the idea of comparing the aggregate annual impact of charcoal production on the income of charcoal- income of charcoal producers with the aggregate annual producing farm households. If it denotes the potential income of “equivalent” charcoal producers. Since the two outcome conditional on charcoal production and denotes groups are comparable on all observed characteristics with the potential outcome conditional on charcoal nonpro- the exception of participating in charcoal production, the ducing, the impact of charcoal production is given by differences in the aggregate annual income are attributed to Δ � Y − Y . (6) 1 0 the charcoal production activity. +e estimated PSM for subject e (x ) (i �1, . . ., N) is the conditional probability of being assigned to a particular treatment given (participated in charcoal production) and is a vector of observed cova- 3.6.1. Estimating the Propensity Score (PS). +e propensity riates x : score is defined as the conditional probability of receiving e x � pr z � 1| x , income from charcoal production given pretreatment i i i characteristics [37]. +e propensity scores were computed (5) zi 1− zi pr zi, . . . , x , . . . , x � e(xi) 1 − e(xi) , using binary logit regression models given as 1 n i�1 1 D (7) P(X) � PrD � � E , where z �1 for charcoal producers; z �0, otherwise; x is the X X i i i vector of observed covariates for the subject. where D �(0, 1) is the indicator of exposure to charcoal +e treatment indicator takes either 1 or 0, but none of production characteristics (dependent variable). +at is, these outcomes were observed for the same individual at the D �1 if it is exposed to charcoal producer farm households, same time. In the urbanization program evaluation litera- and D �0 if it does not participate in charcoal production ture, these phenomena are commonly known as missing until this survey was conducted; X is the multidimensional data problems [34]. In a random charcoal production vector of observed characteristics (explanatory variables). participation assignment, the average treatment effect can be +ese explanatory variables are expected to jointly de- computed by taking the difference in means of the outcome termine the probability to participate in charcoal production variable between those who participate in the production in the study district. +e explanatory variables considered in and those who do not [35]. However, this procedure cannot this study were based on theory and from the review of be applied in our present case because the urbanization studies. process in the study area followed a nonrandom process. Under this condition, an impact evaluation is usually per- formed by applying a more suitable nonexperimental 3.6.2. Matching the Unit Using the Propensity Score. method than an experimental method even if it has merits After the propensity score is estimated and computed for and demerits [36] [34]. +e charcoal production income each unit, the next step is the actual matching. +e nearest impact in the study district administration, for instance, did neighbor matching uses the propensity score of similar not collect baseline data on the outcome variable of interest individuals in the treated and control groups to construct and other preintervention characteristics of displaced farm the counterfactual outcome. Because of this, the nearest households because of charcoal production and not pro- neighbor matching method was used to match. +e main ducing charcoal by farm households in the study area. (In benefit of this approach is the lower change which is the presence of both baseline and follow-up survey data, a reached for more data is used. +e matching estimator is more robust procedure for impact evaluation would be a given as difference-in-difference (DID) or DID in combination with PSM.) +us, we have to rely on a PSM that can identify M T c τ � iET Y − iETW Y i ij j comparable treatment and comparison observations using (8) cross-sectional data [37]. T c +e propensity score is a probability range in values from � iETY − iET iETW Y . ij j 0 to 1. +erefore, if PSM was used in a randomized ex- periment comparing two groups, then the propensity score for each respondent in the study would be 0.50. +is is (i) i, E, T, N denote the numbers of charcoal producers because each respondent would be randomly assigned to matched with observation and define the weights either the treated or the control group with a 50% W � 1/N ifjEC(i), and W � 0, otherwise ij i ij 6 Advances in Agriculture (ii) M stands for the nearest neighbor matching, and the probability of participating in the charcoal production. number of units in the treated group is denoted by Given that the Conditional Independence Assumption and N the common support assumption hold, then we estimate the mean effect of the charcoal production on annual income One of the major advantages of this method is that the through the mean difference in the outcomes of the matched absolute difference between the estimated propensity scores pairs: for the control and treatment groups is minimized. N N 1 0 ⎛ ⎝ ⎞ ⎠ ATT � y − w y , (10) 1i ij 0j 3.6.3. Estimating the Impact (Average Treatment Effect on the i�1 j�1 Treated (ATT). Heckman et al. [38] present several essential where w E[0,1] and w � 1, N is the number of ij j�1 ij 1 preconditions in order to get reliable and low-bias impact producers, N is the number of nonproducers, i is the index estimates using PSM. +ese preconditions included the of producers, j is the index of nonproducers, and W is the ij following: (1) data are collected using identical question- weights. naires for both groups during the same period, (2) treatment and comparison observations share a comparable socio- economic, demographic, and agroecological setting, and (3) 3.7. Definition of Variables and Expected Sign. Based on relevant variables related to treatment and outcome are reviews of empirical studies, like any other economic ac- included in the propensity score function. +e dataset used tivity, the household’s participation in charcoal production in our study clearly meets precondition (1) because an is expected to be affected by household sociodemographic identical survey instrument is used to elicit data from both factors, attitudes of households toward the determinants of control and treatment groups. Precondition (2) is also met charcoal production, and other factors. Based on our em- because as noted earlier the survey data for our study comes pirical review findings, the following explanatory variables from households (both displaced and nondisplaced farm are identified for this study, and the hypotheses of their households) residing in the same study area. To meet pre- expected signs are defined as shown in Table 2. condition (3), the propensity score is estimated by using sample households ‘observable characteristics which are relevant for both participation in the urbanization program 4. Results and Discussion and the outcome variable of interest (see Table 1) [34, 38]. 4.1. Descriptive Statistics Analysis of Sample Households. Propensity scores were estimated by a logit model with the From the total sample of the study, 5 responses from dependent variable coded as 1 for displaced households nonproducers were dropped out because of incompleteness because of urbanization and 0 for nondisplaced households, and errors, and the analysis was made on 305 samples. +e and independent variables comprised of several pre- simple descriptive statistical analysis results of the socio- intervention characteristics. After the propensity scores were economic characteristic of charcoal producers and non- estimated, a kernel matching estimator was used to compute producer households in the Mecha district are presented in the average impact of the program among IFSP households. Table 3. Some of the characteristics discussed here are sex, (+e choice of a matching method is a difficult exercise and marital status, age, educational level, household size, land largely depends on the data at hand. +e quality of matching size, eucalyptus tree coverage, market distance, number of can be compared using different statistical tests. In this oxen, credit service, and extension service. paper, kernel matching with a bandwidth of 0.25 was chosen As represented in Table 3, considering the age range of based on different criteria discussed in the Results section.) the people carrying out the charcoal production, it was In particular, the average treatment effect on the treated was discovered that people who are between the age range of computed using the following equation: 35–45 years have the highest percentage of 52.7% which ATT � E(Δ | D � 1, X) � E Y − Y | D � 1, X 1 0 implies that they are more involved in the production of (9) charcoal than those who are of other age ranges in Mecha � E Y | D � 1, X − E Y | D � 0, X , 1 0 district, and the next set of people are those who are in the where D �1 denotes program participation in charcoal age range of 46–55 years which have the record of 24.7% production, and Χ is a set of conditioning variables on which followed by those who are less than 35 years of age with the percentage of 21.3%, and the least age range involved are the the subjects were matched. Equation (10) would have been easy to estimate except people above >55 years of age having a percentage of 1.3%. +is result is in line with [39] findings that reported for the equation Ε (Y | D �1, X). +is is the mean of the counterfactual and denotes what the outcome would have that 20 (59%) of the respondents are in the ages of 20–39 been among participants on charcoal production had they years; this is not surprising since this is generally the active do not participate in the charcoal production age group in human life; in particular, the activity is an (nonproducer). energy exacting one and different from practices in other PSM provides a way of estimating this equation. A parts of the world most especially in Asia as reported by unique advantage of PSM is that instead of matching sub- [40] where most of the people involved in charcoal pro- jects on a vector of characteristics, we only need to match on duction in Asia are those who are in the active age range of a single item the propensity score that measures the 50–60 years. Advances in Agriculture 7 Table 2: Definition and measurement of independent variables in PSM. Variable Definition Measurement Hypothesis Sex Sex of household head 1 male, 0 female +Positive Education Educational level of the household head Level category −Negative Family size Household family size Number +/− Land size Household agricultural landholding size hectares −Negative Eucalyptus coverage Owned farm size covered with eucalyptus hectares +Positive Age Age of household head years −Negative Marital status Marital status of household Level category +Positive No. of oxen Number of oxen per household head Number −Negative Credit service Access to credit service of household head 1 �yes, 2 �no +Positive Extension service Educational level of the household head Level category −Negative Market distance in km Market distance in km of household head Number −Negative Table 3: Age distribution of the household heads of charcoal producers and nonproducers. Nonproducer Producer (N �150) Total (N �305) (N �155) Variables Age category N % N % N % <35 22 14.2 32 21.3 54 17.7 35–45 83 53.5 79 52.7 162 53.1 Age 46–55 39 25.2 37 24.7 76 24.9 >55 11 7.1 2 1.3 13 4.3 Source: survey result (2019). agricultural extension service since most charcoal producer As represented in Table 4, it was, therefore, observed that the percentage of the married people involved is very much farmer’s primary occupation was charcoal production rather higher than the other group of people involved in the than farming activity. Charcoal producer households’ access charcoal production as it is about 87.3%; the percentage of to agricultural extension service is not statistically significant another group of people involved is very minimal; the singles as the results are not significant at 5% (P >0.102). But involved are about 4%, the widowers are about 5.3%, and the noncharcoal producer households’ access to agricultural divorced people are about 3.3%. +is result is similar to the extension service is statistically significant as the results are report of [41] who reported that the percentage of the significant at 1% (P ≤0.001). As represented in Table 6, also married people involved is very much higher than the other 93.3% of charcoal producer households and 91.0% of noncharcoal producer households access credit service from group of people involved in the production as it is about 83.0%, the singles involved are about 7%, the widowers are the formal and informal financial institutions; charcoal producer and nonproducer households’ access to credit about 5%, and the widows and divorced people are about 4% and 1%, respectively. Reference [39] also found similar service is statistically significant at 1% (P ≥0.001). results where 73% of charcoal producers in Borgu local As represented in Table 7, only 4% of farmers incor- government area of Niger State, Nigeria, are married while porated in this study were females who are involved in the 24% are single. production of charcoal in the study area, but the remaining As represented in Table 5, the educational status of the and the largest portion of charcoal producer farmers in- people having the charcoal production was also taken into corporated in this study were males accounting for 96.0% consideration in this study, where it was observed that the which means that female household heads mostly partici- highest percentage of the people has a reading and writing pated in farming activities rather than participating in charcoal production. (informal education) education, about 52.0% of them, fol- lowed by those with illiterate (no formal education), having a +is result is in line with of results of [39] who reported percentage of 45.3% and those who have primary school and that 34 respondents are involved in charcoal production in secondary school education with the percentage of 2.0% and the study area and most of the respondents (31(91%)) are 0.7%, respectively. males; this is as a result of the tedious nature of commercial +is result is in line with the report of [41] who reported charcoal production which requires a lot of energy. +e that charcoal production mostly involves people with no study [42] reported that the producers are mostly male education; from his report, he affirmed that junior class (69.20–85.10%). dropouts had the highest number of frequencies in his +e result in Table 8 shows that statistically there is a research. significant difference between a charcoal producer and As represented in Table 6, only 56.7% of charcoal pro- noncharcoal producer in terms of household eucalyptus coverage, number of oxen, and amount of aggregate annual ducer households access agricultural extension service which means that the remaining 43.3% had not addressed income, and market distance. Furthermore, there is no 8 Advances in Agriculture Table 4: Marital status of household heads of nonproducers and charcoal producers. Nonproducer Producer (N �150) Total (N �305) (N �155) Variables Category N % N % N % Married 133 85.8 131 87.33 264 86.5 Single 3 1.9 6 4 9 3 Marital status Widow 15 9.7 8 5.33 23 7.5 Divorced 4 2.6 5 3.33 9 3 Source: survey result (2019). Table 5: Literacy level of heads of noncharcoal producers and producer households’ heads. Nonproducer Producer Total (N �305) (N �155) (N �150) Variables Category N % N % N % Illiterate 84 54.2 68 45.3 152 49.75 Reading and writing 70 45.2 78 52.0 148 48.60 Education status of HH Primary school 1 0.6 3 2.0 4 1.30 Secondary and above 0 0.0 1 0.7 1 0.35 Source: survey result (2019). Table 6: Extension and credit service access of heads of nonproducer and producers. Nonproducer Producer (N �150) Total (N �305) (N �155) Variables Category N % N % N % No 25 16.1 65 43.3 90 29.5 Extension service Yes 130 83.9 85 56.7 215 70.5 Chi significance 0.000 0.102 No 14 9.0 10 6.7 24 7.9 Credit service Yes 141 91.0 140 93.3 281 92.1 Chi significance 0.000 0.000 Source: survey result (2019). Table 7: Sex of heads of noncharcoal producer and charcoal producer households. Nonproducer Producer (N �150) Total (N �305) (N �155) Variables Category N % N % N % Male 139 89.7 144 96.0 283 92.8 Sex Female 16 10.3 6 4 22 7.2 Source: survey result (2019). statistically significant difference between a charcoal pro- which accounted for 1.9. Furthermore, in Table 8, the av- ducer and noncharcoal producer in terms of household land erage land size in a hectare of charcoal producer farmers was size and family size. Accordingly, household eucalyptus 1.63 whereas the average land sizes of noncharcoal producer coverage, number of oxen, and amount of aggregate annual farmers were 1.7 hectares. In line with this, as we get data income are significant at a 1% significance level, while from Mecha district agricultural office, the average land- holding at the district level is 1.5ha per household and market distance is significant at a 5% significance level. As represented in Table 8, considering the family size of ranges from 0 to 3ha among the farmers in the area. the people carrying out the charcoal production, it was discovered that people have average an family size of 4.96, and it was less than the average family size of noncharcoal 4.2. Carbon Emission Quantification from Charcoal Production. +e researchers accessed organized data on the producers, which accounted for 5.23. As represented in Table 8 also, people carrying out the charcoal production annual amount of charcoal production from Mecha district agricultural office starting from 2014 and are enforced to use have an average number of oxen of 1.7, which is slightly less than the average number of oxen of noncharcoal producers the available data only because of the inaccessibility of the Advances in Agriculture 9 Table 8: Summary statistics and mean difference test on continuous variables. Nonproducer Producer Total (N �305) Mean difference (N �155) (N �150) Explanatory variables T-value Mean (SD) Mean (SD) Mean (SD) Mean (SD) Family size 5.23 1.506 4.96 1.583 5.0983 1.5486 0.27 −0.077 .202ns Land size 1.713 0.578 1.629 0.517 1.6721 0.5497 0.084 0.061 .289ns ∗∗ Eucalyptus 0.469 0.318 0.812 0.256 0.63793 0.33661 −0.343 0.062 0.000 ∗∗ No. of oxen 1.883 0.44 1.686 0.58 1.786 0.5229 0.197 −0.14 0.000 Market distance 5.23 1.23 4.97 1.00 5.1054 1.1303 0.26 0.23 0.046 ∗∗ Annual income 42496.7 7971.6 65230 11497 53677 15052 −22733 −3525.4 0.000 ∗∗ ∗ Source: survey result (2019); significant at 1%, at 5%, and ns denotes nonsignificance levels, respectively. Note: means and standard deviations (SD) are adjusted for sampling weights. amount of annual charcoal production data from 2011 to Table 14 shows a total of 164648.2 tons of charcoal 2013. +e five-year data (2014–2018) amount of charcoal produced by households in the Mecha district from 2014 to produced annually from the Mecha district as a whole was 2018. On this basis, total greenhouse gas emissions due to collected from the Mecha district agriculture office. +e charcoal production and consumption for the last five years data were analyzed based on IPCC 2006 as expressed in amount to 266244 tons of CO e and on average 53248.8 tons equation (3). +erefore, quantity of GHGi �quantity of of CO e emitted per year. Generally, from these five years’ charcoal (tons) ×emission factors for three major GHGs data, the quantity of charcoal production increases year to from charcoal (CO , CH , and N O) given in IPCC year, and the same direction changes in the total emission of 2 4 2 guideline [27]. To quantify the total emissions of carbon CO e. dioxide equivalent (CO e), CH and N O are multiplied by 2 4 2 21 and 310, respectively, but CO is taken as it is since CH 2 4 4.3. Environmental and Health Impact of Charcoal Production gas has a GWP of 21 times greater than CO and that of N O is 310 times greater than CO [26]. +e amounts of in the Study Area. To realize the environmental impact of 2 2 charcoal production on local temperature and rainfall greenhouse gas (GHG) emissions from charcoal produc- tion were quantified based on the above formula and variability of the study area, rainfall and temperature data discussed in Tables 9–14. were provided from Bahir Dar Regional metrology service agency. +e rainfall and temperature data obtained from As stated in Table 9, 15096 tons of charcoal produced by households in the Mecha district per annum in 2014 is Bahir Dar Regional metrology service agency was analyzed using a Microsoft Excel spreadsheet and stated in Figure 2. estimated to have been produced from 90576 tons of wood (15096 6). On this basis, total greenhouse gas emissions due As represented in Figure 2, the annual minimum tem- perature has almost a constant trend before 2011 but has an to charcoal production and consumption per year amount to 24411.02 tons of CO e. increasing trend especially after 2011 since charcoal pro- duction started hugely in 2011 in the study area. As stated in Table 10, 20282.2 tons of charcoal produced by households in the Mecha district per annum in 2015 is As represented in Figure 2, the annual maximum temperature has also an increasing trend especially after estimated to have been produced from 121693.2 tons of 2011 since charcoal production started massively in 2011 in wood (20282.2 6). On this basis, total greenhouse gas the study area. emissions due to charcoal production and consumption per As represented in Figure 3, the average annual rainfall year amount to 32797.37 tons of CO e. As stated in Table 11, 25060 tons of charcoal produced by had a fluctuated trend from 2011 to 2015 and an increasing trend especially after 2016 in the study area. +is implies that households in the Mecha district per annum in 2016 is estimated to have been produced from 150360 tons of wood charcoal production affects local temperature negatively but the annual rainfall trend is not affected by the amount of (25060 6). On this basis, total greenhouse gas emissions due to charcoal production and consumption per year amount to charcoal produced in the study area. As represented in Table 15, 86% of respondents said that 40523.32 tons of CO e. As stated in Table 12, about 53270.0 tons of charcoal the reason behind engaging in charcoal production is the primary occupation rather than other livelihood systems, produced by households in the Mecha district per annum in and higher income was the generation from charcoal pro- 2017 is estimated to have been produced from 319320 tons of duction rather than other livelihood systems in the study wood (53270.0 6). On this basis, total greenhouse gas area. +e remaining 10% of respondents said that they emissions due to charcoal production and consumption per year amount to 86140.36 tons of CO e. should change unwanted branches, waste, and tree pieces from the eucalyptus tree trade to charcoal and to get ad- As stated in Table 13, 50940.0 tons of charcoal produced by households in the Mecha district per annum in 2018 is ditional income. About 6% of respondents said that higher market demand for charcoal production in the study area estimated to have been produced from 305640 tons of wood (50940.0 6). On this basis, total greenhouse gas emissions and near market enforce them to produce charcoal. As represented in Table 16, it shows that the majority due to charcoal production and consumption per year amount to 82371.91 tons of CO e. (66%) of the charcoal producers get wood for charcoal 2 10 Advances in Agriculture Table 9: Annual GHG emissions due to charcoal production and consumption by 2014. Quantity of charcoal (tons) Type of GHG EF Emissions (year) (tons) CO e emissions (year) (tons) 15096 CO 1.5130000 22840.248 22840.248 15096 CH 0.0041400 62.49744 1312.44624 15096 N O 0.0000552 0.8332992 258.322752 Total emissions (CO e) 24411.02 Table 10: Annual GHG emissions due to charcoal production and consumption by 2015. Quantity of charcoal (tons) Type of GHG EF Emissions (year) (tons) CO e emissions (year) (tons) 20282.2 CO 1.5130000 30686.9686 30686.9686 20282.2 CH 0.0041400 83.968308 1763.334468 20282.2 N O 0.0000552 1.11957744 347.0690064 Total emissions (CO e) 32797.37 Table 11: Annual GHG emissions due to charcoal production and consumption by 2016. Quantity of charcoal (tons) Type of GHG EF Emissions (year) (tons) CO e emissions (year) (tons) 25060 CO 1.5130000 37915.78 37915.78 25060 CH 0.0041400 103.7484 2178.7164 25060 N O 0.0000552 1.383312 428.82672 Total emissions (CO e) 40523.32 Table 12: Annual GHG emissions due to charcoal production and consumption by 2017. Quantity of charcoal (tons) Type of GHG EF Emissions (year) (tons) CO e emissions (year) (tons) 53270.0 CO 1.5130000 80597.51 80597.51 53270.0 CH 0.0041400 220.5378 4631.2938 53270.0 N O 0.0000552 2.940504 911.55624 Total emissions (CO e) 86140.36 Table 13: Annual GHG emissions due to charcoal production and consumption by 2018. Quantity of charcoal (tons) Type of GHG EF Emissions (year) (tons) CO e emissions (year) (tons) 50940.0 CO 1.5130000 77072.22 77072.22 50940.0 CH 0.0041400 210.8916 4428 50940.0 N O 0.0000552 2.811888 871.68528 Total emissions (CO e) 82371.91 Table 14: Summary of annual GHG emissions due to charcoal 29.5 y = 0.0994x – 172.26 R = 0.2089 production from 2014 to 2018. 28.5 Year Total charcoal production Total emissions (CO e) 27.5 2014 15096 24411.02 2015 20282.2 32797.37 26.5 2016 25060 40523.32 25.5 2017 53270.0 86140.36 2018 50940.0 82371.91 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 Year Figure 2: Trend of annual maximum temperature from 2000 to production from their plantation and by purchasing from another farmer, 22% of the producers used wood from their eucalyptus plantation, and the remaining 12% of the pro- reason behind choosing Eucalyptus tree species was the fast- ducers used wood by purchasing from another farmer who growing nature of Eucalyptus tree species and being easily had enough eucalyptus plantation. available than other tree species. Acacia species accounted It was revealed in Table 17 that Eucalyptus tree species for 2.67% of wood sources for charcoal burning, and the accounted for more than 94% proportion of tree species remaining proportion of tree species which were Cordia preferred for charcoal production in the study area. +e africana and others accounted for 2% and 1.33%, Maximum temperature Advances in Agriculture 11 2089.2 1919.6 1779.2 1638.1 1628.2 1586.8 1569.2 1558.6 1601.8 1557.9 1520.6 1378.9 1284.7 1622.1 1142.2 1488.2 1454.5 1308.2 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year Figure 3: Trend hyacinths of annual rainfall from 2000 to 2018. Table 15: Major reason behind choosing charcoal production rather than other activities. Variable Frequency Percentage Higher demand for charcoal 6 4 Higher income generation from charcoal product 129 86 To change eucalyptus tree trade waste to charcoal 15 10 Total 150 100 Table 16: Source of wood for charcoal production. Table 17: Tree species used for charcoal burning. Source of wood for charcoal production Variable Frequency Percentage Response Total Purchased and Eucalyptus trees 141 94 Inherited Purchased Acacia species 4 2.67 inherited Cordia africana 3 1.33 Frequency 33 18 99 150 Others 2 2 Percentage 22 12 66 100 Total 150 100 respectively. +is result is different from Larinde and Ola- could not care for natural resource near to the kiln but few of supo [43] who discovered that the majority (80%) of the them produce charcoal on their own land. respondents get wood for charcoal production from the As represented in Table 18, the respondents used more natural forest, while about 20% of the respondents get their than one laborer in their production. About 32.7%, 40%, wood from forest reserves in Oyo State, Nigeria. In Nigeria, 15.3%, and 12% of the respondents have laborers of 2, 3, 4, for charcoal production, with 73% of charcoal producers in and 5, respectively, to themselves in the production. the four villages preferring to use this species, it was revealed As represented in Table 19, employed workers on charcoal production faced skin irritation/skin roasting that Prosopis africana is widely used because it is hardwood; they further revealed that hardwoods give higher charcoal problem by fire since the significance level of the chi yield than softwoods [39]. analysis result is 0.002. Also, employed laborers significantly More than 60% of the respondents accounted for those faced respiratory illnesses like pneumonia, bronchitis, se- whose carbonization lasted for 6–7 days while less than 30% nescence, and other acute respiratory infections since the accounted for those whose carbonization lasted for 8–10 significance level of the chi analysis result is 0.000. But they days. Carbonization is subject to the level of dryness of the faced a slight problem of eye irritation since the significance wood and the quantity of wood used for charcoal produc- level of the chi analysis result is 0.572. +is result is in line tion. As shown in Figure 4, the only production method with the results of [42] who reported that dirty of their bodies practice in the study area was traditional earth-mound kilns and houses, sicknesses, generating smoke, and ash dust were methods which are a mound and are usually triangular and the forms of social-economic impact being encountered. circular in shape. Charcoal is produced in mounds igniting As represented in Figure 5, pneumonia prevalence had the kiln and allows carbonization under limited air supply as an increasing trend from 2010 to 2012 and showed slight shown in the photo. decrease in trend from 2012 to 2014 and then an increasing +e majority of charcoal-producing communities who trend from 2016 to 2018. Generally, pneumonia prevalence participated in this study did not own the land where had an increasing trend which implies that charcoal pro- production took place. In most cases, the land was owned by duction had its own impact on pneumonia prevalence. a third party, and they rent from them, and the remaining As represented in Figure 6, the acute respiratory in- were producing on common lands even near to social in- fection had an increasing trend from 2010 to 2012 and stitutions like schools and health centers, roadsides, and showed also an increasing trend from 2014 to 2016 and then settlements that exposed those parts of communities to smog an increasing trend from 2016 to 2018. Generally, the acute pollution and critical health and safety problems, and they respiratory infection had an increasing trend which implies Annual rainfall 12 Advances in Agriculture Figure 4: Charcoal production sites and process (source: observations (2019)). Table 18: Number of laborers employed in charcoal production. Response Frequency Percentage 2 persons 49 32.7 3 persons 60 40 Number of laborers employed in charcoal production 4 persons 23 15.3 5 persons 18 12 Total 150 100 Table 19: Employed laborer health problem faced analyzed with chi-square. Response Skin irritation Eye irritation Respiratory illness Yes 36 27 38 Do you face those health problems? No 14 23 12 Total 50 50 50 X 0.002 0.572 0.000 that charcoal production had its own impact on acute respi- y = 706x + 12218 2 2 R = 0.1348 ratory infection since R � 0.935. Employed laborers on 17642 charcoal production were closely proximate to extreme tem- perature kilns during the production phase especially at night since the average distance between the sleeping area and burning kiln at night is only 5.2 meters. +at exposed those laborers to adverse health risks related to charcoal production. Based on the information gathered from the key in- 2010 2012 2014 2016 2018 formant interview, charcoal production started in the study Year area in 2011. As we get data from the district agriculture Figure 5: Pneumonia prevalence from 2010 to 2018 E.C. office, there is no licensed charcoal producer but they use tree trade license for charcoal production. Charcoal pro- traffic accident because of roadside crowdedness by a piece duction kilns are also near to living house schools and of wood and charcoal, dust and aerosol problem, welting of roadsides as researchers observed in the study area. +e fruits and vegetables, and rainy season fluctuation. observed environmental problems because of charcoal production in the study area as checked by observation and concluded from key informant interview were smog pol- 4.4. Impact of Charcoal Production on Households’ Livelihood. lution, vision obstruction, roof rusting, soil and soil mi- +e logistic regression model was employed to estimate croorganism burning at kiln site deforestation, cause for propensity scores for matching charcoal producer Number of patients Advances in Agriculture 13 y = 4931.2x + 2805.6 R = 0.935 30000 23307 2010 2012 2014 2016 2018 Year Figure 6: Acute respiratory infection from 2010 to 2018 E.C. households with noncharcoal producer households. For with eucalyptus and then produce charcoal. As eucalyptus coverage increases, the probability of participating in estimating propensity scores, only those variables which affect both the likelihood of charcoal production partici- charcoal production activity increases (see Table 20). pation and the income impact were included. +e pseudo-R value of 0.4075 shows that the overall explanatory variables 4.4.3. =e Number of Oxen. +e number of oxen has a included in the regression explain the dependent variable statistically significant negative effect on charcoal produc- (socioeconomic impact of charcoal production) of about tion, and the result is significant at 1% (Pvalue <0.001). Farm 40.75% which is fair for nonlinear econometric regression. households who have a large number of oxen engaged in +e logistic regression model specified in equation (8) was farming activity would increase especially in irrigation work employed to estimate propensity scores for matching dis- rather than producing charcoal but farmers who have rel- placed households with control or nondisplaced households. atively small numbers of oxen highly engaged in charcoal +e dependent variable in this model was a dummy variable production rather than farming activity. As the number of indicating whether the household has been displaced which oxen increases, the probability of participating in charcoal takes a value of 1 and 0, otherwise. For estimating propensity production activity declined (see Table 20). scores, only those variables which affect both the likelihood of displacement and the outcomes of interest were included. +e estimated regression results show in Table 20 that 4.4.4. Extension Service. Extension service has a statistically the probability of charcoal production participation is sig- significant negative effect on charcoal production, and the nificantly and negatively affected by household land size, the result is significant at 1% (P value <0.001). As the extension number of oxen, extension service, and market distance service access increases, the probability of participating in from households’ home, and these results are significant at charcoal production activity declined (see Table 20). 1% and 10% probability level, respectively. Likewise, it is positively affected by eucalyptus coverage which is signifi- 4.4.5. Market Distance. Market distance has a statistically cant at an equally 1% probability level. significant negative effect on charcoal production, and the Numberof observations � 305, result is significant at 5% (P value <0.012). As the market 2 distance is far apart, the probability of participating in LR chi (10) � 172.27, charcoal production activity declined (see Table 20). (11) Prob >chi � 0.0000, Loglikelihood � −125.23307, 4.5. Distribution of Propensity Score Matching. +e distri- bution of propensity scores of both charcoal producers’ and Pseudo R � 0.4075. nonproducer’s observations is presented in a graph shown in Figure 7. As represented in Figure 7, most charcoal producer households are found on the right side of the distribution, 4.4.1. Household Land Size. As the model result indicates, whereas most of the nonproducer households are found the variable household land size had negatively and sig- partly in the center and partly on the left side of the dis- nificantly influenced the charcoal production participation tribution. On the other hand, charcoal producer propensity at less than 1% (P value< 0.001) probability level. +is score distribution was skewed to the right while it was finding indicates that those farm households with higher skewed to the left side for nonproducer households. From farmland size are more likely to decrease charcoal than the figures, one can observe that there is a wide area in which households with less land size (see Table 20). the propensity scores of both the treatment and the control groups are similar. Hence, it is possible to match the two groups using the common support region. 4.4.2. Eucalyptus Coverage. Eucalyptus coverage has a sta- tistically significant positive effect on charcoal production, and the result is significant at 1% (P value <0.001). Even 4.6. =e Impacts of Charcoal Production on Farm Household though charcoal producers have relatively lower land sizes Income. +is section presents evidence as to whether or not than noncharcoal producers, they covered most of their land charcoal production has brought significant changes to the Number of patients 14 Advances in Agriculture Table 20: Logistic regression model result for charcoal production participation. Variables Coef. Std. err. Z P >|z| [95% conf. interval] Sex 1.0861 0.9058607 1.20 0.231 [−0.6893542–2.861555] Age −0.2527549 0.3104656 −0.81 0.416 [−0.8612563–0.3557465] Family size 0.1283652 0.157903 0.81 0.416 [−0.1811189–0.4378494] Education status 0.2491024 0.3127596 0.80 0.426 [−0.3638952–0.8621001] ∗∗ Land size −1.44551 0.4236064 −3.41 0.001 [−2.275763–0.6152568] ∗∗ Eucalyptus coverage 7.287438 0.9916262 7.35 0.000 [5.343887–9.23099] ∗∗ Number of oxen −1.471818 0.3810437 −3.86 0.000 [−2.21865−0.7249858] ∗∗ Extension service −1.533715 0.3889231 −3.94 0.000 [−2.29599−0.7714395] Credit service 0.8906354 0.6276396 1.42 0.156 [−0.3395157–2.120786] Market distance −0.4654044 0.1858764 −2.50 0.012 [−0.8297155−0.1010933] Marital status −0.0139243 0.2785831 −0.05 0.960 [−0.5599371 0.5320885] Constant 1.462284 1.655139 0.88 0.377 [−1.781728–4.706296] ∗∗ ∗ Source: own survey (2019) �significant at 1%, �at 5%, and ns �not significance levels, respectively. Kernel density estimate 1.5 0.5 0 0.5 1 psmatch2: propensity score Sample households Charcoal producing households Control households kernel = Epanechnikov, bandwidth = 0.0977 Figure 7: Common support estimation with kernel density. Table 21: Impacts of urban expansion displacement on farming community. Outcome variables Charcoal producer Nonproducer Difference S.E T-value HH livelihood outcome 2.561822 2.123690612 0.43813162 .038137446 −2.057 Source: estimation result, 2019. annual income of the rural farming communities. After annual maximum temperature also increase after charcoal controlling the other characteristics, the propensity score production started in that area but the annual rainfall had matching model using the nearest neighbor matching es- some fluctuation. +e producers were using more than one timator result (bandwidth 4) indicates an urban expansion laborer in their production, and also, some of them were effect from the propensity score matching estimation (in child laborers. +e social and public health risks associated Table 21) which shows that there is a significant difference in with charcoal production were clearly identified in this the annual income of charcoal production farming com- study. Most laborers employed in charcoal production not munity by 0.43813162 as compared to the nonproduced used personal protective equipment that increases the risk of farming community. exposure to health and safety problems. socioeconomic factors like land size, eucalyptus coverage, agricultural ex- tension market distance, and the number of oxen have a 5. Conclusion statistically significant effect but other factors like sex, age, family size, education status, and credit service have no +e majority of the charcoal producers get wood for charcoal production both from their plantation and by purchasing statistically significant effect on charcoal production. from another farmer. +e reason behind choosing charcoal Charcoal production was economically profitable in the production as the primary occupation was the higher in- study area, while charcoal production is economically come generation from charcoal production. +e total annual valuable; it has disproportionately adverse effects on envi- charcoal production, rate, and total emission of CO e had ronmental degradation and local air quality contamination increased at an alarming rate (24411.02 by 2014, 32797.37 by in addition to respiratory health problems for producers and 2015, 40523.32 by 2016, 86140.36 by 2017, and 82371.9 by nearby residents. So, the utilization of high-efficiency kilns 2018). As a result, the annual minimum temperature and made of metal and brick can significantly lower the impacts Density Advances in Agriculture 15 [12] P. M. Fearnside and W. F. Laurance, “Tropical deforestation associated with an industry that is vital to the national and greenhouse-gas emissions,” Ecological Applications, economy. In light of the 2015 Sustainable Development vol. 14, no. 4, pp. 982–986, 2004. Goals and numerous findings across the continent, the re- [13] M. Arnold and R. Persson, “Reassessing the fuelwood situ- sults of this study and others suggest that an exclusive focus ation in developing countries,” International Forestry Review, on modern energy services should be redirected to consider vol. 5, no. 4, pp. 379–383, 2003. these traditional and widespread energy sources that have [14] I. M. Amugune, Technological Factors Influencing the Quality significant economic, environmental, and social impacts. and Quantity of Charcoal Produced in Western Mau Forest, Kericho County, Kenya, Africa Nazarene University, Nairobi, Kenya, 2020. Data Availability [15] V. Chiteculo, B. Lojka, P. Surovy,´ V. Verner, D. Panagiotidis, and J. Woitsch, “Value chain of charcoal production and Data and publications from this project will be of open implications for forest degradation: case study of bie´ province, access and available via an online repository. Angola,” Environments, vol. 5, no. 11, p. 113, 2018. [16] Y. Alem et al., Household Fuel Choice in Urban Ethiopia: A Conflicts of Interest Random Effects Multinomial Logit Analysis, 2013. [17] H. Geller, R. Schaeffer, A. Szklo, and M. Tolmasquim, +e authors declare that they have no conflicts of interest. “Policies for advancing energy efficiency and renewable en- ergy use in Brazil,” Energy Policy, vol. 32, no. 12, pp. 1437–1450, 2004. References [18] K. R. Smith, “Indoor air pollution in developing countries: recommendations for research+,” Indoor Air, vol. 12, no. 3, [1] W. A. Onekon and K. O. Kipchirchir, “Assessing the effect of pp. 198–207, 2002. charcoal production and use on the transition to a green [19] M. A. Mulugeta, Linking Fiscal Decentralization and Local economy in Kenya,” Tropical and Subtropical Agroecosystems, Financial Governance: A Case of District Level Decentraliza- vol. 19, no. 3, pp. 327–335, 2016. tion in the Amhara Region, Ethiopia, University of Western [2] A. Ghilardi and F. Steierer, “Charcoal production and use: Cape, Cape Town, South Africa, 2014. world country statistics and global trends,” in Proceedings of [20] M. Getahun, Characterisation of Agricultural Soils in Cascape the Symposium on the Role of Charcoal in Climate Change and Intervention Woredas of Amhara Region, 2015. Poverty Alleviation Initiatives, Arushai, Tanzania, June 2011. [21] M. D. A. Office, =e Population of the District Based the [3] B. Okello, T. O’connor, and T. P. Young, “Growth, biomass Ethiopia Central Statistical Agency Census Conducted in 2010, estimates, and charcoal production of Acacia drepanolobium in Laikipia, Kenya,” Forest Ecology and Management, vol.142, [22] W. G. Cochran, “+e estimation of sample size,” Sampling no. 1-3, pp. 143–153, 2001. Techniques, vol. 3, no. 1, pp. 72–90, 1977. [4] E. Chidumayo, “Environmental impacts of charcoal pro- [23] W. G. Cochran, Sampling Techniques-3, Wiley, Hoboken, NJ, duction in tropical ecosystems of the world,” in Proceedings of USA, 1977. the Meetings of the Association of Tropical Biology and [24] R. V. Krejcie and D. W. Morgan, “Determining sample size for Conservation, Arusha, Tanzania, April 2011. research activities,” Educational and Psychological Measure- [5] R. T. Kappel and E. K. Ishengoma, Economic Growth and ment, vol. 30, no. 3, pp. 607–610, 1970. Poverty: Does Formalisation of Informal Enterprises Matter?, [25] S. K. Antwi and K. Hamza, “Qualitative and quantitative German Institute for Global and Area Studies (GIGA), research paradigms in business research: a philosophical Hamburg, Germany, 2006. reflection,” European Journal of Business and Management, [6] L. C. Zulu and R. B. Richardson, “Charcoal, livelihoods, and vol. 7, no. 3, pp. 217–225, 2015. poverty reduction: evidence from sub-Saharan Africa,” Energy [26] IPCC III, W., =ird Assessment Report. Summary for Poli- for Sustainable Development, vol. 17, no. 2, pp. 127–137, 2013. cymakers, IPCC, Geneva, Switzerland, 2001. [7] A. Ellegard ˚ and M. Nordstrom, ¨ “Deforestation for the poor?” [27] IPCC, 2006 IPCC Guidelines for National Greenhouse Gas Renewable Energy for Development, vol. 16, no. 2, pp. 4–6, Inventories, IGES, Kanagawa Japan, 2006. [28] M. Caliendo and S. Kopeinig, “Some practical guidance for the [8] M. O. Dioha and A. Kumar, “Exploring sustainable energy implementation of propensity score matching,” Journal of transitions in sub-Saharan Africa residential sector: the case of Economic Surveys, vol. 22, no. 1, pp. 31–72, 2008. Nigeria,” Renewable and Sustainable Energy Reviews, vol. 117, [29] B. E. Frye and J. E. Bartlett, “Performing the PSM analysis: an Article ID 109510, 2020. applied example,” Journal of Applied Research in the Com- [9] S. Lurimuah, =e Economic and Environmental Effects of munity College, vol. 24, no. 1, pp. 43–58, 2017. Commercial Charcoal Production in the Upper West Region of [30] S. Yang, “Propensity score weighting for causal inference with Ghana, 2011. clustered data,” Journal of Causal Inference, vol. 6, no. 2, 2018. [10] J. Mohammed, I. Tetteh, and M. Mills, “Fuelwood to charcoal [31] N. L. P. Nguyen, Essays on Propensity Score Methods for production in the new juaben municipality of the eastern Causal Inference in Observational Studies, Duke University, region of ghan challenges and opportunities in seeking livelihood,” Researchers World, vol. 4, no. 4, p. 51, 2013. Durham, North Carolina, 2018. [32] A. Gujarati, Basic Econometrics, +e McGraw− Hill compa- [11] V. Bustamante-Garc´ıa, A. Carrillo-Parra, H. Gonzalez- ´ Rodr´ıguez, R. G. Ram´ırez-Lozano, J. J. Corral-Rivas, and nies, Pennsylvania, NY, USA, Fourth edition, 2004. [33] J. J. Diaz and S. Handa, “An assessment of propensity score F. Garza-Ocañas, “Evaluation of a charcoal production pro- cess from forest residues of Quercus sideroxyla Humb., and matching as a non experimental impact estimator: evidence from a Mexica’s PROGRESA program,” =e Journal of Hu- Bonpl. in a Brazilian beehive kiln,” Industrial Crops and Products, vol. 42, pp. 169–174, 2013. man Resources, vol. 41, no. 2, pp. 319–345, 2004. 16 Advances in Agriculture [34] J. A. Smith and P. E. Todd, “Does matching overcome LaLonde’s critique of nonexperimental estimators?” Journal of Econometrics, vol. 125, no. 1-2, pp. 305–353, 2005. [35] J. J. Heckman, H. Ichimura, and P. Todd, “Matching as an econometric evaluation estimator,” Review of Economic Studies, vol. 65, no. 2, pp. 261–294, 1998. [36] R. Blundell and M. Costa Dias, “Evaluation methods for non- experimental data,” Fiscal Studies, vol. 21, no. 4, pp. 427–468, [37] P. R. Rosenbaum and D. B. Rubin, “+e central role of the propensity score in observational studies for causal effects,” Biometrika, vol. 70, no. 1, pp. 41–55, 1983. [38] J. J. Heckman, H. Ichimura, and P. E. Todd, “Matching as an econometric evaluation estimator: evidence from evaluating a job training programme,” =e Review of Economic Studies, vol. 64, no. 4, pp. 605–654, 1997. [39] O. Adeniji, O. S. Zaccheaus, B. S. Ojo, and A. S. Adedeji, “Charcoal production and producers’ tree species preference in Borgu local government area of Niger State, Nigeria,” Journal of Energy Technologies and Policy, vol. 5, no. 11, pp. 1–8, 2015. [40] J. Bhattarai, E. Akiyama, H. Habazaki et al., “+e passivation behavior of sputter-deposited W-Ta alloys in 12 M HCl,” Corrosion Science, vol. 40, no. 4-5, pp. 757–779, 1998. [41] J. Ogara, “Preliminary studies on charcoal production and producers’ knowledge of environmental hazards in Nasarawa State, Nigeria,” PATS, vol. 7, pp. 68–75, 2011. [42] A. V. Goanue, Status of Renewable Energy in Liberia, Pre- sentation of Rural and Renewable Energy Agency, 2009. [43] S. Larinde and O. Olasupo, “Socio-economic importance of fuelwood production in Gambari forest reserve area, Oyo State, Nigeria,” Journal of Agriculture and Social Research (JASR), vol. 11, no. 1, 2011.
http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png
Advances in Agriculture
Hindawi Publishing Corporation
http://www.deepdyve.com/lp/hindawi-publishing-corporation/socioeconomic-and-environmental-impacts-of-charcoal-production-Ev7jm2eZIs