Abstract
In this paper we conduct a scoping review of current household energy use patterns and trends in rural Latin America (LA), with the objective of identifying strategies that help promote just and fair transitions in the region. We reviewed a total of 143 publications covering 13 countries within the period from 1996 to 2021. The review shows: (a) fuelwood (FW) continues to be a very important, resilient—and in many countries—the dominant cooking fuel for rural LA households, both exclusively and increasingly stacked (combined) with liquefied petroleum gas (LPG); (b) FW is mostly used in open fires and rustic stoves, with a total toll of 59 000 premature deaths. Interventions have centered on the dissemination of improved woodburning chimney cookstoves and increasing access to LPG through top-down government programs. These programs have focused mostly on single-fuel and stove combinations, and on the number of devices installed with little or no follow-up with local users. As a result, success has been limited and open fires have not been fully displaced in most programs. We conclude that renewed efforts are needed to ensure a sustainable and just household energy transition in the LA region. These efforts should promote integrated portfolios of options including improved practices (drying wood, use of pressure cooker), and the stacking of devices (stoves, water heaters, space heating) and fuels (biomass, other). Specifically, improved chimney woodburning stoves need to be integrated with and be an important component of these programs. Programs should adopt a user-centered perspective, beginning with the understanding of users' needs and priorities and tailoring solutions to their socio-environmental context. Innovation should be fostered through participatory methods, developing tests adapted to local circumstances and enforcing national standards. Implementation programs should focus on the adoption and sustained use of clean(er) devices and the displacement of traditional fires. Public policies should be more integrated and intersectoral seeking synergies between health, environmental, social development, and economic objectives.
Export citation and abstract BibTeX RIS
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
1. Introduction
Approximately 2.6 billion people, mostly rural (90%) and poor in the world, currently use fuelwood (FW) and charcoal, as their principal energy source for cooking, water heating, space heating and other household energy needs. Globally, these two woodfuels account for 10% of primary energy or 31–50.5 EJ are used mostly in open fires and rustic traditional stoves, almost 50% of total wood harvesting and 2% of global greenhouse gas (GHG) emissions (Masera et al 2015). The reliance on polluting devices to meet household energy needs is a leading cause of household air pollution (HAP), which results in millions of deaths worldwide and almost 59 000 deaths in Latin America (LA) and Caribbean alone for the year 2019 (http://ihmeuw.org/5fuo), while WHO's estimation for year 2016 shows almost 78 500 deaths for this region (WHO 2021). Exposure to HAP is a major avoidable health hazard that increases the risk of several communicable and non-communicable diseases (Lee et al 2020).
Assuring a just and fair household energy transition for these households is therefore imperative. By 'just and fair' energy transitions we mean a process that leaves no household behind and that everyone can relate to—with the outcome of providing universal access to clean and affordable energy services. To reach this goal it is essential that the benefits and costs of the transition are distributed equally, a participatory process that engages all stakeholders in the decision making, and recognizing multiple perspectives rooted in social, cultural, ethical and gender differences. The discussion on just, fair, and sustainable energy transitions has been stated as a global priority, being the subject of several Sustainable Development Goals (SDGs), specifically SDG 7 ('Energy for all') but also connected to SDGs 3 (Good health and well-being), 5 (Gender equality), and 15 (Life on land), among others. National and international programs that are designed to address these concerns have focused on transitioning households towards cleaner energy practices by encouraging access to improved woodburning cookstoves (ICS), and increasingly by promoting exclusive use of other fuels, such as liquefied petroleum gas (LPG), electricity, biogas, and other options. However, efforts have not been entirely successful—both in terms of the financial resources committed globally and the program's effectiveness. As a result, the targets are not expected to be realized in 2030 as planned, particularly within the world's poorest regions and populations (World Bank 2021).
In this paper we review the status of the so-called residential energy transition within rural LA households. The paper is timely and needed for several reasons. First, despite a long and rich experience of household energy programs within LA there has not been a recent review covering the whole region (see for example (Wang et al 2013) for Central America and Mexico). Second, LA is unique in that—while not comprising a large share of the global population—the energy-transition is much more advanced than in other regions. So, lessons from LA could be of great value to other regions. Also, biomass cooking technology has centered on the development of chimney stoves, some of which comply with WHO targets in terms of health impacts. There has also been a rich array of experiences, detailed studies on health, energy and environmental impacts of stove interventions, and public policies from which to reflect and learn to improve future actions and programs.
In the next sections we will review the current situation of household energy use in LA, discuss the main findings of existing studies on health and environmental impact of residential solid biomass use, and review the main cookstove implementation programs conducted within the region. We will then propose a series of strategic actions to cope with the present challenges.
2. Methods
We performed a scoping review to provide an overview of the available research evidence with a focus on the range of content identified. We examined the literature on the health and environmental impacts of household FW use for cooking in rural LA and the documented household energy programs. We identified relevant studies through an electronic database search (language English), adding other publications identified in reference lists and hand searches, and the authors knowledge of gray literature as part of the Red Latinoamericana y del Caribe de Cocinas Limpias (languages Spanish and English). We searched the electronic database PubMed using the following terms: fuels and devices (cooking, HAP, solid fuel, FW, stove, improved stove, stacking, LPG, gas), process and impact areas (implementation, adoption, exposure, health, emission, climate change, performance) and countries (Mexico, Guatemala, Honduras, El Salvador, Nicaragua, Belize, Panama, Costa Rica, Colombia, Venezuela, Peru, Bolivia, Ecuador, Brazil, Paraguay, Chile, Guyana, Argentina, Uruguay). We selected the papers that presented a comparison between fuels and/or devices and/or a specified intervention. The papers not including such a comparison were dropped. Most of the gray literature refers to the description of the country's household energy use profile or the implementation of household energy programs. We selected the main topic for each selected publication (implementation, performance, adoption, use, exposure, health, emissions, climate change, deforestation). The expert authors for each topic reviewed the publications (AS, BO, JAE, LAS, MS, VB, VR) and extracted the information in a previously designed data charting form. All authors reviewed the data charting form to discuss the findings after completing information when necessary. As shown in figure 1, we extracted information from 124 articles and 19 reports (supplementary information in Schilmann 2021).
Figure 1. Flow diagram of articles selected for the scoping review.
Download figure:
Standard image High-resolution image3. Current household energy use patterns
FW is still an important household energy source for cooking, water and space heating and other uses in LA. Specifically, rural communities are highly reliant on biomass, exclusively or in conjunction with LPG, to cover their main energy needs. There are no reliable statistics about the extent of FW users in LA, existing estimates range from 70–90 million exclusive users (SEGIB 2021) to 160 million users (GACC 2014) including both exclusive FW users and users that combine FW with other fuels for cooking and space heating. Rural FW users in several LA countries account for 80%–100% of their rural population, with large differences among countries—ranging from almost 100% of users for Bolivia and Nicaragua, to 55% of users for Panama and 25% of users for Costa Rica (Wang et al 2013). Peri-urban FW use, generally mixed with LPG, is also reported in the region ranging from 20% to almost 40% of this sector (Serrano-Medrano et al 2014, Ruiz-Mercado and Masera 2015, Garland et al 2018, Gould et al 2020b). FW is generally burned in traditional devices such as three-stone fires (TSFs), U-shaped stoves, and in poorly ventilated kitchens (Serrano-Medrano et al 2014, Garland et al 2018, Williams et al 2020a, Gould et al 2020b).
FW use has proved very resilient either as exclusive fuel or increasingly in combination (or stacking) with LPG despite the prolonged access to this latter fuel in some countries (Serrano-Medrano et al 2014, Ruiz-Mercado and Masera 2015, Gould et al 2018). Also, the scattered nature of rural settlements and the low purchasing power of most rural households has limited the penetration of LPG. Additionally, to cost and access restrictions, reliance on FW use patterns is associated with culinary, convenience, economic and cultural practices. For instance, the preparation of traditional meals based on corn like tortillas in Mexico, or the preparation of potato-based dishes in South American countries like Peru, Bolivia, and Ecuador (Ruiz-Mercado and Masera 2015, Gould et al 2018, 2020b).
Furthermore, space heating is also an important end use in rural communities located in cold regions and living space heating is the main FW use in countries such as Chile and Argentina (Schueftan et al 2016, Cardoso and González 2019). The use of FW to heat water for bathing and cooking food for animals, lighting of the home, drying of clothes, smoking of food, discarding of waste, keeping away insects and other animals from households, as well as when cooking for large number of people have also been widely reported in the literature (Ruiz-Mercado and Masera 2015, Gould et al 2018, 2020b).
The intensity, importance and studies regarding FW use within the residential sector varies among LA countries as shown in table 1 and figure 2.
Figure 2. Studies reporting rural household energy transitions in Latin America.
Download figure:
Standard image High-resolution image
Table 1. General characteristics of household fuelwood use in selected Latin American countries.
| Country | Number/percentage of users/households using fuelwood | Dominant fuel | Secondary fuel | Fuelwood consumption (kg cap−1 yr−1) | Observations |
|---|---|---|---|---|---|
| Brazil (Gioda 2019) | About 30 million people depend on firewood as a household energy source. Fuelwood is used by 17%–87% of households in the Brazilian rural sector (especially by rural families in the Northeast and indigenous communities) but it can also reach up to 38% of households in specific cities. | -LPG (National level and urban sector) -Fuelwood (Rural—Northeast and Indigenous communities) | -Fuelwood (National level) -LPG (Rural—Northeast mainly) | 600–780 | Additionally, a high percentage of households use LPG and firewood simultaneously (60%–90%). Fuelwood is mostly used for cooking. |
| Ecuador (Gould et al 2018, Gould et al 2020b) | Despite more than 98% of the households in the study reported using LPG for cooking, about 40% also reported using firewood as a secondary option to LPGb. | -LPG (In the study region) | -Fuelwood (In the study region) | N/A | Firewood use has proved resilient despite the existence of long-term subsidies to facilitate the transition from cooking with biomass to cooking to LPGa. |
| Guatemala (Pachauri et al 2018) | About 90% of the rural households, and 50% of urban households use solid biomass and stoves. | -Fuelwood (National level and rural sector) | -LPG (National level) | 1033 | The residential sector demands about 98% of the biomass energy consumption in the country. |
| Honduras (Garland et al 2018) | 52% of the population use firewood mainly for cooking and heating. Firewood for cooking is more used by the rural population (89%) than by its urban counterpart (24%). | -Fuelwood (National level and rural sector) | -LPG (National level) Fuelwood (Urban sector) | 1000 (rural) 500 (urban) | N/A |
| Mexico (Serrano-Medrano et al 2014, INEGI 2018) | About 28 million people use fuelwood as a primary or secondary fuel for cookinga. Country wise, 67% of rural households and 16% of urban households cook with fuelwood. | -Fuelwood (Rural sector) -LPG (National level and urban sector) | -LPG (Rural sector) -Fuelwood (National level and urban sector) | 770 (550–1100) | Fuelwood accounts for almost 40% of total residential energy consumption in the countryb. |
| Nicaragua (Pachauri et al 2018) | 97% of the rural population in Nicaragua (for cooking). | -Fuelwood (National level and Rural sector) | N/A | 927 | Nicaragua is one of the poorest countries in Central America. |
| Peru (Pollard et al 2018) | Over 80% of rural households use biomass as their primary fuel source to meet residential energy needs. | -Fuelwood (Rural sector) | LPG (Rural sector) | N/A | N/A |
N/A: not available information or no additional observations from the reviewed article.
4. Household energy transition programs and their impacts
4.1. Implementation programs
In LA, programs to foster the transition to cleaner and more sustainable energy for cooking have adopted various models and implementation scales. Improved woodburning cookstoves (ICS) and clean fuels have been promoted mostly to reduce FW consumption and deforestation, to mitigate GHG emissions and to reduce health risks in rural areas aiming to improve the life quality of local people (Troncoso et al 2007, Berrueta et al 2008, 2017, Ghilardi et al 2009, Wang et al 2013).
In this review we identified 30 programs from 11 countries, half of them implemented at the national or regional level, the other 11 at the local level and the remaining 3 represent research projects (table 2). Most programs have disseminated ICS of various models: on-site construction, portable or semi-portable; five have promoted the use of LPG through devices and subsidies; Haiti and Guyana have simultaneously disseminated two technologies: ICS + LPG stoves; and solar stoves + ICS respectively; Ecuador has a program of induction electric stoves (Gould et al 2018).
Table 2. Improved cookstoves and clean fuel implementation programs in Latin America.
| Country | Program | Type of program (implementer) | Program financer | Level of intervention | Stove type | Num. of stoves | User participation | Training | Follow up and maintenance |
|---|---|---|---|---|---|---|---|---|---|
| Bolivia (PAHO 2015a) | Cookstoves for a better life; 100 000 smoke-free households in Bolivia | Government program/Private companies/NGOs | Government/private companies/NGOs | National | ICS, in situ construction | 82 500 | Yes | Yes | Yes |
| Colombia (PAHO 2015b) | Programa Nacional de Estufas eficientes para cocción con leña | Government program/NGOs | Government/private companies/NGOs | National | ICS, several models | 28 238 | NI | Yes | Yes |
| Ecuador (ADRA 2013) | Energización Rural en Comunidades de Guano y Pujilí, a Través de la Implementación de Cocinas Mejoradas | Government program/private companies/NGOs | Government/private companies/NGOs | Community | ICS, plancha | 800 | Yes | NI | NI |
| Ecuador (Gould et al 2018) | National LPG Subsidy Program | Government program | Government Program Subsidy | National | LPG stoves/LPG | NI | NI | NI | NI |
| Ecuador (Gould et al 2018) | Energy Efficiency Program for Cooking | Government program | Government Program Subsidy | National | Induction stoves | NI | NI | NI | NI |
| El Salvador (PAHO 2012a) | El Salvador Ecocina | Private companies/NGOs | Private companies/NGOs/users | National | ICS, portable | 11 170 | Yes | NI | NI |
| El Salvador (PAHO 2012a) | El Salvador Turbococina | Government program/private companies/NGOs | Government/private companies/NGOs | Community (schools) | ICS, portable | 1200 | Yes | Yes | NI |
| Guatemala (Bruce et al 2004) | SIF (Social Investment Fund) | Government program | Government/international agencies | National | ICS | 90 000 | Yes | Yes | No |
| Guatemala (Alvarez et al 2004) | Tezulutlán | NGOs, international agency, Government | International agency | Community | ICS, in situ construction | 4129 | Yes | Yes | Yes |
| Guatemala (Alvarez et al 2004) | Intervida | NGOs | NGOs | Village | ICS, in situ construction | 9000 | Yes | Yes | Yes |
| Guatemala (Ludwinski et al 2011) | Field experiment in Guatemala | NGOs | NGOs | Village | ICS, portable | 28 | Yes | Yes | Yes |
| Guatemala (PAHO 2015c) | Onil stoves dissemination | NGOs | NGOs/government/microcredits | National | ICS, portable | 110 000 | NI | NI | NI |
| Guatemala (El-Saghir Selim 2013, Thompson et al 2018a) | NACER II | Research project | Small scale | LPG stoves/LPG | 50 | Yes | Yes | Yes | |
| Guatemala (World Bank 2011) | RESPIRE | Research project | Community | ICS | 534 | Yes | Yes | ||
| Guyana (PAHO 2012b) | Government/international agencies | International agencies | Community | Solar cooking stoves, ICS | 600 | NI | NI | NI | |
| Haiti (PAHO 2015d, Phanol and Pierre 2015) | Improved Cooking Technology Program (ICTP) | International agencies/private companies/government | Users/microcredits | National | ICS/LPG stoves | NI | NI | NI | NI |
| Honduras (PAHO 2016) | Proyecto Mirador | Government program/NGO | NGOs/users | National | ICS, in situ construction | 85 000 | Yes | Yes | Yes |
| Honduras (Clark et al 2010) | Adhesa | Government program/NGO | NGOs/users | National | ICS, in situ construction | 30 000 | Yes | Yes | Yes |
| Mexico (INSP 2016) | Programa de estufas ecológicas | Government program | Government | Provincial | ICS, several models | 60 000 | No | No | No |
| Mexico (García-Frapolli et al 2010, Masera et al 2007) | Patsari Project | NGO | NGO/users/microcredits | Village | ICS, in situ construction | 1500 | Yes | Yes | Yes |
| Mexico (PAHO 2012c) | Plan Especial de Cambio Climático/Programa de empleo temporal/Programa de atención a Zonas Prioritarias | Government program/NGOs | Government | National | ICS, several models | 561 000 | No | No | No |
| Mexico (Troncoso et al 2019) | Ciudades Rurales de Chiapas | Government program | Government/users subsidies | Village | LPG stoves/LPG | NI | NI | NI | NI |
| Mexico (Troncoso et al 2019) | Municipality program | Government program | Government | Village | ICS | NI | NI | NI | NI |
| Nicaragua (Terrado and Eitel 2005, PAHO 2015e) | ESMAP | NGO | NGO/users | Village | ICS | 1300 | NI | NI | NI |
| Nicaragua (PAHO 2015e) | Mifogón Program | Government program/NGO | NGO/users/government | National | ICS several models | 25 000 | Yes | NI | No |
| Peru (Checkley et al 2021) | Intervention project | Research project | Project/users subsidies | Community | LPG stoves/LPG | 180 | Yes | Yes | Yes |
| Peru (Fitzgerald et al 2012) | Juntos National Program | Government program | Government | National | ICS | NI | NI | NI | NI |
| Peru (PAHO 2015f) | Por un Peru sin humo | Government program | Government/carbon credits | National | ICS | 360 000 | NI | NI | NI |
| Peru (PAHO 2015f) | Fondo de inclusión social energético | Government program | Government/NGOs/users subsidies | National | LPG stoves/LPG | 491 000 | NI | NI | NI |
| Peru (Hartinger et al 2016) | Optima Program | NGO | Project | Community | ICS | 93 | Yes | Yes | Yes |
NI: no information.
National programs for the transition to clean energy for cooking were identified in seven countries: National Program of Efficient Stoves for cooking with FW in Colombia (PAHO 2015b); National LPG Subsidy Program in Ecuador (Gould et al 2018); Social Investment Fund (SIF) in Guatemala (Bruce et al 2004); Mirador and Adhesa Project in Honduras (PAHO 2016); special climate change plan/program of attention to priority areas in Mexico (PAHO 2012c); Mifogón in Nicaragua (PAHO 2015e); National Program Together (Fitzgerald et al 2012), for a Peru without smoke and the Social Inclusion Fund for energy in Peru (PAHO 2015f). Of these, five programs are implemented by the government and six are implemented by the government and NGOs.
Of the fifteen small-scale programs, five have been implemented with the exclusive participation of NGOs (Alvarez et al 2004, Masera et al 2007, García-Frapolli et al 2010, Ludwinski et al 2011, PAHO 2015c, 2015e, Hartinger et al 2016); five by NGOs in collaboration with governments, international agencies and/or private companies (Alvarez et al 2004, PAHO 2012a, 2015a, ADRA 2013); three are research projects (Smith et al 2010, 2011, Thompson et al 2018b, Checkley et al 2021) and two have been implemented exclusively by the government (Troncoso et al 2019).
Of all the programs, 13 report carrying out training on the use of technologies prior to installation, of which 11 also report performing follow-up and maintenance (Alvarez et al 2004, Masera et al 2007, García-Frapolli et al 2010, Smith et al 2010, 2011, Ludwinski et al 2011, PAHO 2015a, 2015b, 2016, Hartinger et al 2016, Thompson et al 2018b, Weinstein et al 2020, Checkley et al 2021).
Local Central American entrepreneurs have been very important in stove innovation, oftentimes with external support. However, as in the whole LA region, efforts in Central America have not transcended into a scale economy or self-sustaining market for improved stoves; indeed, the region is still far from forming a large-scale commercial stove market (Wang et al 2013).
4.2. Stove and clean fuel adoption and stacking
While most programs have focused on the number of stoves installed, no stove program can achieve its goals unless people adopt the stoves and continue using them on a long-term basis (Ruiz-Mercado et al 2011). We identified 53 articles with relevant information regarding the adoption and stacking of stoves and fuels in Argentina, Brazil, Chile, Colombia, Ecuador, Guatemala, Honduras, Mexico, Peru, Paraguay, Uruguay and Venezuela (supplementary information in Schilmann 2021). Of these, 39 articles gave specific data on adoption and all of them reported stacking of fuels and technologies (table 3).
Table 3. Stove and fuel stacking patterns within Latin American households.
| Country | References | Type of study | Type of stacking | Percentage of households stacking fuels and stoves (%) | Explanatory comments |
|---|---|---|---|---|---|
| Brazil | (Gioda 2019, Mazzone et al 2021) | QL/QN | TSF/ICS/LPG Electrical | 60%–90% | A high percentage of households use LPG and fuelwood simultaneously (60%–90%). The introduction of 'ready-meals' and processed food also contributes to the increased usage of LPG over firewood. |
| Chile | (Shupler et al 2020) | QN | ICS/LPG | 36% | Fuelwood was the primary fuel in 91% of rural households surveyed, and its main use is for space heating. Stacking with LPG was present in 36% of cases. Manufactured chimney stoves were most prevalent. |
| Colombia | (Shupler et al 2020) | QN | TSF/LPG | 39% | In the rural case study surveyed, fuelwood was the primary fuel in 65% of households, and LPG in 35%. Stacking was present in 39% of households. Open fires were the main traditional cooking device. |
| Ecuador | (Gould et al 2018, Shankar et al 2020, Gould et al 2020a, Gould et al 2020b) | QN | TSF/LPG ICS/electrical | 36%–81% | LPG is the main rural cooking fuel and TSF (and to a lesser extent induction stoves) are used for specific purposes. In the studies reviewed stacking was found in 36%–81% of households. |
| Guatemala | (Albalak et al 2001, Bruce et al 2004, Schei et al 2004, Mccracken et al 2011, Thompson et al 2018a) | QL/QN | TSF/ICS/LPG TSF/ICS TSF/ICS/LPG/electrical | 10%–30% | Fuelwood is the main cooking fuel in rural areas. The studies reviewed reported higher acceptance of ICS than LPG; TSF are used for more intensive tasks, followed by ICS. LPG used for quick tasks. Stacking ranged from 10% to 30% of households depending on the study and consisted in different combinations of fuels and technologies. |
| Honduras | (Young et al 2019) | QN | TSF/ICS/LPG | 42% | Fuelwood is the main cooking fuel; stacking was present in 42% of households surveyed (18% TSF/LPG and 24% ICS/TSF). |
| Mexico | (Masera et al 2005, Zuk et al 2007, Armendariz et al 2008, Romieu et al 2009, Pine et al 2011, Schilmann et al 2015, Ruiz-Mercado and Masera 2015, INSP 2016, Catalán-Vázquez et al 2018, Schilmann et al 2019, Troncoso et al 2019, Estévez-García et al 2020) | QL/QN | TSF/ICS/LPG/microwaves | 10%–80% | Fuelwood is the dominant rural cooking fuel, with increasing penetration of LPG (10%–80% of rural households depending on the region). Most people cook with open or semi-closed fires. Making tortillas—including the preparation of nixtamal—represents up to 50% of total household fuelwood use. There is an increasing—but still limited—adoption of ICS. LPG complements rather than substituting fuelwood and is used mostly for the less intensive cooking tasks. Stacking between TSF/ICS; ICS/LPG, LPG/TSF or even the three stoves together is very common; TSF are rarely completely abandoned. There is an increasing use of microwaves for warming food in the wealthiest households. |
| Paraguay | (Troncoso et al 2018, Tagle et al 2019) | QL/QN | TSF/metal braziers for charcoal/LPG/electric—hot plate | n.a. | The use of woodfuels, particularly charcoal, for cooking is very common in rural Paraguay, with estimates ranging from 40% to 74% of rural households depending on the study. Both fuelwood and charcoal are used on open kitchens and on rustic devices without chimneys (84% of households). Stacking with LPG is common, while electricity is seldom used as the main cooking fuel. |
| Peru | (Pollard et al 2014, Hartinger et al 2016, Wolf et al 2017, Pollard et al 2018, Díaz-Vásquez et al 2020, Shankar et al 2020, Williams et al 2020a, Williams et al 2020b, Checkley et al 2021) | QL/QN | TSF/ICS/LPG | 20%–100% | Fuelwood is the dominant cooking fuel in Peru. Penetration of LPG has been increasing, particularly due to government programs that have provided different types of subsidies to this fuel. Stacking is common, ranging from 20% to 100% of households, depending on the case study. In intervention studies devoted to promoting LPG and ICS it was observed that all households stacked LPG with TSF; 85% stacked clean stoves; and more than 50% stacked ICS with TSF after the intervention. ICS were found to be preferred and more intensively used than LPG. |
| Venezuela | (Kraai et al 2013) | QN/QL | TSF/LPG | 30% | In the case study of a Native American village in Venezuela, fuelwood was reported as the main cooking fuel. 30% of households stacked TSF/LPG; 20% cooked with LPG alone. |
Factors that influence adoption include sociocultural aspects (n = 19), like traditions, symbolic aspects related to food or to local uses of FW (Mazzone et al 2021); addressing users' preferences and needs (n = 11), like perceived differences in flavor and nutrition of food prepared in different stoves (Hollada et al 2017); follow up after stove installation (n = 13), to guarantee optimum stove performance and also to actually get the benefits of stove implementation programs (Masera et al 2005); the previous use of other technologies or fuels (n = 6), for example, in the case of the Patsari Stove an important factor for adopting the stove was user's previous experience with elevated stoves (Romieu et al 2009); and functional aspects of the technology (n = 14), i.e. including 'add-on' benefits to the stove that could produce small amounts of electricity to charge cell phones, sanitize water, or power compact fluorescent lights (CFL) (Bielecki and Wingenbach 2014) (table 4). A third of studies show economic variables (n = 13) as important predictors of adoption. These studies emphasize that economic reasons—like fuel or stove price, household incomes—are one of the main reasons for using traditional open fires (Thompson et al 2011) and therefore, economic incentives are recommended to facilitate the adoption of efficient stoves (Masera et al 2005). Regarding LPG, it is found that income is not a good predictor of adoption but rather of sustained use (INSP 2016) and that the high cost of LPG is one of the main barriers to its adoption (Wang et al 2013, Troncoso et al 2019). Table 4 lists other factors that are also associated with stove adoption.
Table 4. Factors related to adoption and use of clean energy options in Latin American households.
| Main factors | Examples | Countries (references) |
|---|---|---|
| Sociocultural factors | Traditions and symbolic aspects related to fire, food, and local uses of fuelwood, dominant gender roles (e.g. who decides about the purchase of new fuel/stove). | Latin America (Córdova and Castro 2012); Central America (Wang et al 2013); Brazil (Mazzone et al 2021); Ecuador (ADRA 2013); Guatemala (Schei et al 2004, Bielecki and Wingenbach 2014, Thompson et al 2018a, Williams et al 2020b); Mexico (Pine et al 2011, Ruiz-Mercado and Masera 2015, INSP 2016, Catalán-Vázquez et al 2018, Schilmann et al 2019, Estévez-García et al 2020); Peru (Hartinger et al 2013, Rhodes et al 2014, Hollada et al 2017, Wolf et al 2017, Williams et al 2020b) |
| User preferences and needs | Perceived differences in taste and nutrition associated to food prepared in different stoves; ranking of fuel savings, vs savings in cooking time vs smoke reduction in the kitchen; ease in lighting the fire or repair the stove. | Latin America (Córdova and Castro 2012); Central America (Wang et al 2013); Guatemala (Bielecki and Wingenbach 2014, Williams et al 2020a); Mexico (Masera et al 2005a, Ruiz-Mercado et al 2011, Ruiz-Mercado and Masera 2015, Catalán-Vázquez et al 2018); Peru (Rhodes et al 2014, Hollada et al 2017, Williams et al 2020b) |
| Follow up after stove installation | Visits to users to assuring adequate stove performance, answering users doubts or identifying problems not realized during stove installation. | Latin America (Córdova and Castro 2012); Ecuador (ADRA 2013, Gould et al 2020b); Guatemala (Bielecki and Wingenbach 2014, Williams et al 2020b); Mexico (Masera et al 2005, Ruiz-Mercado et al 2011, Smith et al 2011, Ruiz-Mercado and Masera 2015, Catalán-Vázquez et al 2018); Peru (Wolf et al 2017, Díaz-Vásquez et al 2020, Williams et al 2020b) |
| Previous use of other fuels and technologies | Families using TSF and LPG adopted ICS more easily than families using only TSF; also, women used to cook on elevated TSF adopted ICS more easily than those cooking kneeling on the floor. | Mexico (Romieu et al 2009, Pine et al 2011, Ruiz-Mercado and Masera 2015, Catalán-Vázquez et al 2018); Peru (Hollada et al 2017, Wolf et al 2017) |
| Functional aspects of the technology | Versatility of the proposed stove to satisfy the different user's needs; ability to provide additional benefits (e.g. stoves that could produce small amounts of electricity to charge cell phones, stoves that can provide hot water using residual heat). | Chile (Gómez et al 2014); Guatemala (Albalak et al 2001, Bruce et al 2004, Schei et al 2004, Bielecki and Wingenbach 2014, Thompson et al 2018a, Williams et al 2020b); Mexico (Ruiz-Mercado and Masera 2015, INSP 2016, Catalán-Vázquez et al 2018, Estévez-García et al 2020); Peru (Rhodes et al 2014, Wolf et al 2017, Williams et al 2020b) |
| Economic variables | Fuel and stove price relative to household incomes, subsidies to stove/fuels; financial incentives or facilities to purchase stoves (the relative weight of each factor depends on the type of stove and local circumstances). | Chile (Gómez et al 2014); Ecuador (ADRA 2013, Gould et al 2020b); Guatemala (Thompson et al 2011, Rajkumar et al 2018, Williams et al 2020a); Mexico (Masera et al 2005, Troncoso et al 2019); Peru (Hartinger et al 2013, Hollada et al 2017, Wolf et al 2017, Williams et al 2020a, Williams et al 2020b) |
| Focalized messages for different population groups (women/men/other family members) | Stove adoption increased when messages about health benefits of clean stoves were clearly stated and understood by all family members. | Guatemala (Thompson et al 2018a, Thompson et al 2018b); Mexico (INSP 2016); Peru (Hollada et al 2017) |
| Comprehensive approaches | Programs including multiple stove and fuel options, including options to cover the diverse uses of open fires (cooking, water heating, space heating), or improved cooking practices (e.g. using pressure cookers). | Brazil (Mazzone et al 2021); Guatemala (Bielecki and Wingenbach 2014); Mexico (Masera et al 2005, Estévez-García et al 2020); Peru (Rhodes et al 2014, Hartinger et al 2016) |
| Participation | Involvement of users—specifically local women—in the different phases of stove dissemination programs, from the design, implementation to follow up. | Latin America (Córdova and Castro 2012); Brazil (Mazzone et al 2021); Ecuador (ADRA 2013); Mexico (INSP 2016, Estévez-García et al 2020) |
Recommendations to encourage stove adoption include: that program participants should contribute with a payment or in-kind contribution (ADRA 2013, Gómez et al 2014), that the benefits from cleaner cooking should be clearly explained to women and men or other members of the household (Hollada et al 2017, Thompson et al 2018a, 2018b), that the approaches should be comprehensive rather than individual (Masera et al 2005, Bielecki and Wingenbach 2014, Rhodes et al 2014, Hartinger et al 2016) and that user participation is fundamental (Córdova and Castro 2012, ADRA 2013, Mazzone et al 2021).
As Mazzone et al (2021) states: 'The ethical and symmetrical energy transition requires decentralized strategies to understand, consider and include the cultural capital of local communities and their direct participation in the decision-making processes of the energy transition.'
4.3. Global environmental impacts
While important in specific areas or 'hot spots' in terms of forest degradation, household FW use in the region is mostly renewable and, in principle, solid biomass resources can be managed sustainably, particularly if actions are taken to reduce FW demand in the most critical areas. Bailis et al (2015) performed a spatial explicit assessment of pan-tropical woodfuel supply and demand to estimate the extent in which woodfuel demand surpasses regrowth (Bailis et al 2015). They estimated that in LA between 19% and 31% of woodfuel harvested was unsustainable, a figure which was later confirmed by national studies such as in Serrano-Medrano et al (2019). Woodfuel is mostly locally available, and it is extracted not only from the surrounding forest but from shrubs, agriculture, pruning residues, dead wood and from commercial wood harvesting residues. Very rarely this extraction is done by totally clearing the forest areas such as most commercial timber practices.
Regarding global environmental impacts, we identified only eight articles with detailed information on greenhouse and other air pollutants emissions for chimney cookstoves and open fires in LA. Mexico and Honduras are the countries where field and laboratory measurements on pollutant emissions have been carried out, including measurements of carbon monoxide (CO) and particulate matter (PM), and more recently measurements of black carbon and methane (Padilla-Barrera et al 2019). Johnson et al (2008, 2009), estimated a 25% reduction in products from incomplete combustion comparing improved cookstoves with the traditional open fire, and also estimated that methane emitted from open fires contributes to 45% of CO2e emissions (excluding CO2). Plancha-type stoves commonly used in Mexico and Central America have been estimated to lead to a reduction of 44%–55% in CO and PM2.5 emissions in controlled water boiling tests, and 65% in a typical cooking cycle test for the Mexican Highlands, with regards to TSF (EkouevI and Tuntivate 2012, Medina et al 2015, 2017). In addition to a reduction in total emissions, plancha-type chimney stoves ventilate on average 95 ± 3% of PM2.5 and 99 ± 1% of CO emissions (Ruiz-García et al 2018). Unfortunately, at the moment, there are no more studies about fugitive emissions from chimney cookstoves used in LA, which are essential to determine the actual contribution of these stoves to indoor air pollution. Scenarios modeling the country-scale implementation of ICS and LPG programs within Mexican rural areas have shown a potential GHG emissions mitigation ranging from 30% to 35% of business as usual (BAU) emissions from open fires by the year 2030 (Serrano-Medrano et al 2018).
Table 5a summarizes CO and PM2.5 total emission factors for cookstoves estimated in Central America and Mexico. Table 5b shows that emission rates from ICS can meet the WHO Air Quality Guidelines for both pollutants. So far, only Mexico has included the measurement of fugitive and chimney CO and PM2.5 emissions from cookstoves within the national stove testing standard. The lack of specialized testing equipment within the existing regional Stove Test Centers is one of the reasons why local country standards and regulations have not included the measurement of GHG gases and short-lived climate pollutants.
Table 5a. CO and PM2.5 total emission factors during water boiling test (WBT), controlled burning cycle (CBC), controlled cooking test (CCT) and uncontrolled cooking test (UCT).
| Total emission factor | ||||
|---|---|---|---|---|
| Test type | Description | PM2.5 (g kg−1) fuelwood | CO (g kg−1) fuelwood | Reference |
| CCT | Patsari simulated kitchen (n = 6) | 1.7 ± 0.1 | 47.0 ± 2.1 | [1] |
| CCT | U-type simulated kitchen (n = 6) | 3.0 ± 0.3 | 62.0 ± 14.6 | [1] |
| CBC | Patsari simulated kitchen (n = 5) | 1.6 ± 0.4 | 46.7 ± 3.1 | [1] |
| CBC | U-type simulated kitchen (n = 5) | 6.0 ± 0.8 | 70.0 ± 5.2 | [1] |
| WBT | Open fire simulated kitchen (n = 6) | 5.4 ± 0.4 | 39.7 ± 1.9 | [2] |
| WBT | Mud–cement Patsari simulated kitchen (n = 6) | 5.3 ± 0.9 | 81.7 ± 9.5 | [2] |
| WBT | Open fire in-home (n = 7) | 4.1 ± 0.9 | 25.7 ± 4.4 | [2] |
| WBT | Mud–cement Patsari in-home (n = 7) | 3.1 ± 0.5 | 58.3 ± 7.1 | [2] |
| WBT | Brick Patsari in-home (n = 4) | 2.3 ± 1.4 | 16.3 ± 8.2 | [2] |
| UCT | Open fire in-home (n = 8) | 9.7 ± 1.2 | 81.7 ± 4.9 | [2] |
| UCT | Mud–cement Patsari in-home (n = 9) | 5.9 ± 0.8 | 65.3 ± 3.9 | [2] |
| UCT | Brick Patsari in-home (n = 4) | 1.8 ± 1.0 | 18.7 ± 12.8 | [2] |
| UCT | Chimney cookstoves (n = 27) | 4.5* | 76.0* | [3] |
| UCT | Traditional cookstoves (n = 13) | 8.2* | 118.0* | [3] |
The devices were tested in Mexico and Honduras, Medina et al (2017) [1], Johnson et al (2008) [2], Roden et al (2009) [3]. Notes: the results apply only to plancha-type stoves used in Mexico and Central America. Variability is expressed as ±SD. * Variability is not available.
Table 5b. CO and PM2.5 fugitive emission rates during water boiling test (WBT).
| Fugitive emissions | ||||
|---|---|---|---|---|
| Test type | Stove | PM2.5 (mg min−1) | CO (mg min−1) | Reference |
| WBT | Onil in lab (n = 15) | 2.1 ± 0.3 | 12 ± 3.1 | [4] |
| WBT | Ecostufa in lab (n = 15) | 3.5 ± 0.5 | 5 ± 1.3 | [4] |
| WBT | Mera-Mera in lab (n = 15) | 2.4 ± 0.4 | 20 ± 5.2 | [4] |
| WBT | Patsari in lab (n = 15) | 3.9 ± 0.5 | 11 ± 2.8 | [4] |
| WBT | Cookstoves chimney-type (n = 60) | 3 ± 0.2 | 12 ± 1.5 | [4] |
| All | Unvented intermediate emission rate target for meeting AQG (24 h, CO) and AQGs (IT-1, PM2.5) | 1.75 | 350 | [5] |
| All | Vented intermediate emission rate target for meeting AQG (24 h, CO) and AQGs (IT-1, PM2.5) | 7.15 | 1450 | [5] |
| WBT | Target for fugitive emission rate | 2.7 | 133 | [6] |
The devices were tested in Mexico, Ruiz-García et al (2018) [4], WHO (2014) [5], ISO (2018) [6]. Notes: the results apply only to plancha-type stoves used in Mexico and Central America. Variability is expressed as ±SD. * Variability is not available.
4.4. Exposure and health impacts
We reviewed 71 articles with information on exposure and/or health outcomes carried out in 13 countries in LA as shown in figure 2 (supplementary information in Schilmann (2021)). Guatemala had the highest number of studies (n = 23), followed by Peru (n = 15), Mexico (n = 13), Honduras (n = 10), Colombia and Paraguay (n = 2), and Bolivia, Brazil, Ecuador, Chile, Costa Rica, Nicaragua, and Venezuela had one publication each.
Different improved cookstove models and technologies were evaluated in LA, predominantly wood burning stoves such as Plancha, Patsari, and Justa. In addition, there were ten studies carried out on the use of LPG stoves and one on the use of electric induction stoves.
As shown in Table 6, a total of 62 studies reported direct (n = 46) and microenvironmental (n = 38) exposure measurements of different air pollutants. Direct exposure measurements were carried out using personal monitors and quantifying biomarkers in urine, blood, or exhaled air. In the micro-environment mainly PM of different sizes (n = 44), and CO (n = 28) were measured in a fixed point in the household (mainly the kitchen). In addition, the measurements of other pollutants such as black carbon, polycyclic aromatic hydrocarbons, volatile organic compounds (VOCs), and NO2 were reported.
Table 6. Summary of exposure and health impacts for ICS studies in Latin America.
| Outcome | Countries (references) | n (%) studies report a significant difference in the outcome |
|---|---|---|
| Exposure | ||
| (a) Micro-environmental | ||
| PM (TSP, PM10, PM3.5, PM2.5), CO | Chile (Shupler et al 2020), Colombia (Shupler et al 2020, Martínez Vallejo et al 2021), Costa Rica (Park and Lee 2003), Guatemala (Naeher et al 2000a, Albalak et al 2001, Naeher et al 2001, Bruce et al 2004, Neufeld et al 2004, Northcross et al 2010, Smith et al 2010), Honduras (Clark et al 2009, Clark et al 2010, Benka-Coker et al 2018, Rajkumar et al 2018, Rajkumar et al 2019, Young et al 2019, Benka-Coker et al 2020, Benka-Coker et al 2021), Mexico (Brauer et al 1996, Riojas-Rodríguez et al 2001, Masera et al 2007, Zuk et al 2007, Armendariz et al 2008, Estévez-García et al 2020), Nicaragua (Clark et al 2013a ), Paraguay (Tagle et al 2019), Peru (Li et al 2011, Fitzgerald et al 2012, Eppler et al 2013, Hartinger et al 2013, Commodore et al 2013a, Commodore et al 2013b, Pollard et al 2014, Helen et al 2015, Checkley et al 2021, Fandiño-Del-Rio et al 2020) | 34 (87%) |
| Other (BC, BTX, NO2) | Colombia (Martínez Vallejo et al 2021), Honduras (Walker et al 2020), Peru (Helen et al 2015, Checkley et al 2021, Fandiño-Del-Rio et al 2020, Kephart et al 2021) | 6 (100%) |
| (b) Direct | ||
| PM (TSP, PM10, PM3.5, PM2.5), CO | Bolivia (Alexander et al 2014), Brazil (da Silva et al 2012), Chile (Shupler et al 2020), Colombia (Shupler et al 2020), Ecuador (Gould et al 2020b), Guatemala (Naeher et al 2000b, Bruce et al 2004, Neufeld et al 2004, Mccracken et al 2007, Northcross et al 2010, Smith et al 2010, Mccracken et al 2011, Thompson et al 2011, Guarnieri et al 2014, Thompson et al 2014, Guarnieri et al 2015, Heinzerling et al 2016, Grajeda et al 2020, Weinstein et al 2020), Honduras (Clark et al 2009, Clark et al 2010, Benka-Coker et al 2018, Rajkumar et al 2018,, Rajkumar et al 2019, Young et al 2019, Walker et al 2020, Benka-Coker et al 2020), Mexico (Riojas-Rodriguez et al 2011), Nicaragua (Clark et al 2013b), Peru (Li et al 2011, Eppler et al 2013, Commodore et al 2013a, Commodore et al 2013b, Helen et al 2015, Checkley et al 2021, Fandiño-Del-Rio et al 2020) | 41 (91%) |
| Other (BC, eCO, %HbCO, PAHs, VOCs, BTX, NO2) | Guatemala (Diaz et al 2007b, Guarnieri et al 2014, Guarnieri et al 2015, Lucarelli et al 2018, Weinstein et al 2020), Mexico (Torres-Dosal et al 2008, Riojas-Rodriguez et al 2011, Pruneda-Álvarez et al 2012, Ruiz-Vera et al 2019), Peru (Li et al 2011, Adetona et al 2013, Helen et al 2015, Li et al 2016, Checkley et al 2021, Fandiño-Del-Rio et al 2020, Kephart et al 2021) | 12 (100%) |
| Children health | ||
| (a) Respiratory and other symptoms (asthma symptoms, pneumonia, acute upper and lower-respiratory infections, symptoms related to sleep apnea) | Guatemala (Schei et al 2004, Harris et al 2011, Smith et al 2011), Mexico (Riojas-Rodriguez et al 2011, Schilmann et al 2015), Paraguay (Troncoso et al 2018), Peru (Castañeda et al 2013, Accinelli et al 2014) | 6 (67%) |
| (b) Lung function (spirometry and peak expiratory flow rates) | Guatemala (Heinzerling et al 2016), Honduras (Rennert et al 2015) | 2 (100%) |
| (c) Other (low birth weight, perinatal death and stillbirth) | Guatemala (Thompson et al 2011, Thompson et al 2014, Patel et al 2015) | 1 (33%) |
| Women health | ||
| (a) Respiratory and other symptoms | Brazil (da Silva et al 2012), Guatemala (Díaz et al 2007a, Harris et al 2011, Lucarelli et al 2018), Honduras (Clark et al 2009), Mexico (Romieu et al 2009, Riojas-Rodriguez et al 2011), Paraguay (Troncoso et al 2018), Venezuela (Kraai et al 2013) | 12 (86%) |
| (b) Lung function (spirometry, PEF) | Brazil (da Silva et al 2012), Guatemala (Guarnieri et al 2015), Honduras (Clark et al 2009, Rennert et al 2015), Mexico (Romieu et al 2009), Peru (Checkley et al 2021) | 4 (57%) |
| (c) Other (quality life scores, hemoglobin, blood pressure, self-rated health, ST-segment depression, gene expression airway inflammation; vascular inflammation regulators, urinary stress markers, eNO, eCO, eHbCO, SpO2 SpHbCO) | Bolivia (Alexander et al 2014), Guatemala (Neufeld et al 2004, Mccracken et al 2007, Díaz et al 2008, Ludwinski et al 2011, Mccracken et al 2011, Guarnieri et al 2015), Mexico (Torres-Dosal et al 2008, Ruiz-Vera et al 2019), Nicaragua (Clark et al 2013a), Peru (Eppler et al 2013, Commodore et al 2013b, Pollard et al 2014, Li et al 2016) | 18 (75%) |
Description: BC: black carbon; BTEX: benzene, toluene, ethylbenzene and xylene; CO: carbon monoxide; %COHb: blood carboxyhemoglobin; CO2: carbon dioxide; eCO: exhaled carbon monoxide; eNO: exhaled nitric oxide; eHbCO: carboxyhemoglobin measured from exhaled breath; LPG: liquefied petroleum gas; NOx : nitrogen oxides; PAHs: polycyclic aromatic hydrocarbons; PM: particulate matter; RSP: respirable suspended particles; SpO2: oxygen saturation; SpHbCO: carboxyhemoglobin measured from pulse co-oximetry.
HAP exposure studies recognizing that ICS showed significant reductions in pollutant exposure compared to open fires were first published in Mexico (Brauer et al 1996) and Guatemala (Naeher et al 2000b, Albalak et al 2001), In most post-intervention measurements (n = 56), the ICS showed significant reductions in PM, CO, and other pollutant levels compared to open fires. However, these concentrations were above the WHO air quality guidelines (AQGs), and the reductions in indoor concentrations were lower when the ICS is in poor condition (Clark et al 2013b).
The health of children was assessed only in 13 papers evaluating different respiratory outcomes (n = 8), lung function (n = 2) and perinatal outcomes (n = 3). Two thirds (n = 9) of these studies reported that ICS had a significant effect on the health outcomes. Guatemala was the country with the highest number of children studies.
The health of women was assessed in 32 papers evaluating respiratory symptoms (asthma, cough, phlegm, chest wheezing, and dyspnea n = 14) and lung function (n = 7). Twenty studies reported other health outcomes: blood pressure, exhaled CO, carboxyhemoglobin, eye irritation, headache, backache, diabetes, metabolic syndrome, ST-segment depression, vascular inflammation regulators, and urinary oxidative stress DNA biomarkers.
These papers present results under different study designs including randomized controlled trials (n = 12), observational studies (follow up and cross-sectional n = 42), before-and-after studies (n = 17), and program impact evaluation (n = 1). The randomized controlled trials are the experimental epidemiological designs to evaluate the effectiveness of an intervention but can be biased if there is a differential adherence to the intervention, as has been described in sections 3 and 4.2. There are randomized controlled trial reports assessing the impact of ICS conducted in Guatemala, Honduras, Mexico, and Peru. The studies were carried out among women, children, or both population groups.
In Guatemala, the Randomized Exposure Study of Pollution Indoors and Respiratory Effects (RESPIRE) followed by the Chronic Respiratory Effects of Early Childhood Exposure to Respirable Particulate Matter Study, under the leadership of Kirk Smith, showed the benefits of the Plancha ICS on exposure (Northcross et al 2010), children (Heinzerling et al 2016) and women (Díaz et al 2007a, Diaz et al 2007b, Mccracken et al 2007, 2011) health outcomes, and also presented some negative results for pneumonia in children (Smith et al 2011), low birth weight (Thompson et al 2011, 2014) and women lung function (Guarnieri et al 2015).
In Mexico, the comprehensive evaluation of the Patsari ICS Project showed exposure reductions (Masera et al 2007, Zuk et al 2007, Armendariz et al 2008), and benefits for children and women in a randomized controlled trial analyzed considering the reported use of the cooking device (Romieu et al 2009, Schilmann et al 2015).
In Honduras, exposure reductions were reported for a stepped-wedge randomized trial evaluating the Justa ICS (Benka-Coker et al 2020). In Perú, two recently LPG stoves randomized trials (Checkley et al 2021, Kephart et al 2021) assessed exposure and health outcomes after the intervention.
5. Discussion: what have we learned?
Facilitating universal access to environmentally clean and healthy residential energy, requires considering the needs of the local population and providing comprehensive options (GACC 2014). Evidence from our review shows that, to be successful, policies and programs for improving access to clean cooking must be adapted to local economies, household fuel use patterns, traditions and users' needs and preferences (Pine et al 2011, Ruiz-Mercado et al 2011, Ramirez et al 2014, Catalán-Vázquez et al 2018, Shankar et al 2020). Finding the right combinations locally has been documented to accelerate scaling and thus contributing to making a difference globally (Urmee and Gyamfi 2014). The experience in LA shows that FW users respond well when ICS and other options meet the needs of a specific circumstance: when FW is purchased and is becoming increasingly expensive; when health issues are clearly understood by the whole family; when incentives are provided to lower the upfront costs of stoves; when ICS are tailored to local cooking practices, resulting in tangible fuel and time savings; and when they do not involve major changes in the dimensions of FW and cooking habits and appeal to the 'modernity' aspirations of users (Wang et al 2013). Results from studies carried out in India also indicate that stove adoption requires the availability of spare-parts for stove repair and maintenance, clearly communicating stove health, economic and environmental benefits to local users, and, in many circumstances, some financial incentives (Bhojvaid et al 2014, Pattanayak et al 2019).
Also, cookstove programs in almost all cases promote only one stove model. This approach prevents learning and improvement through competition and denies consumers choice. A focus on community participation and local capacity building, particularly among women, improved cookstove program outcomes and created buy-in of beneficiaries. Most cookstove programs to date have lacked 'systematic community feedback, monitoring and evaluation'.
Household energy projects and ICS programs show that households' decisions to adopt or not a stove includes their perception of stove durability and the mid-and long-term needs of maintenance, repair, or replacement to support sustained use (EkouevI and Tuntivate 2012). A follow-up study carried out in Mexico to evaluate sustained use almost a decade after an ICS program, showed that Patsari ICS had a 50% survival time of four years. After this time, more than half of the stoves installed during the initial trial failed to be used, surpassing their useful lifespan and its well-functioning, failed to reduce the exposure to HAP and consequently people went back to using the traditional stove (Wolf et al 2017, Schilmann et al 2019).
As in other World regions, access to clean fuels in rural LA mostly leads to a diverse pattern of fuel and device stacking, where traditional open fires are seldom entirely replaced. Also, as LPG is not always an affordable fuel for the rural poor, chimney ICS together with improved cooking practices constitute a more realistic and effective approach for communities with low purchasing power. Also, high subsidies to LPG distort markets, preventing consumer feedback from reaching manufacturers and retailers, and thwarting efforts at sustainable commercialization. It is remarkable that only few studies have assessed best ways of disseminating stoves, and none have explicitly addressed the possibility of 'clean stacking options'. We argue that good implementation strategies should embark on context evaluations—identifying the needs and habits of the target groups—and co-creating ICS. This means that there is no one-size-fits-all approach. Furthermore, public awareness needs to be created, demonstrations about correct use of the ICS should be given and maintenance should be assured as shown in other regions (Thakur et al 2019).
While there is very limited data regarding GHG and aerosol emissions from residential solid biomass use in the LA region, our review shows that replacing an open fire with a well-designed chimney ICS may reduce from three to five times overall aerosol emissions and from 95% to 99% fugitive emissions (Johnson et al 2009, Ruiz-García et al 2018). The use of chimney ICS for cooking could also represent a solution to mitigate short-lived climate pollutants like methane and black carbon. Large GHG emissions savings could be obtained by replacing TSF with chimney ICS in rural areas of LA, and additional health benefits if ICS are stacked with LPG (Serrano-Medrano et al 2018, Medina et al 2019). Environmental and health implications depend on the specific stacking options in each region (Medina et al 2019). Locally assessed emissions factors and the development of new standard lab tests that better represent in-field stove performance for specific regional contexts will help to estimate more accurately the regional and country annual CO2-e and fuel savings that could be achieved with different interventions (Johnson et al 2009, Medina et al 2017, Serrano-Medrano et al 2018).
Results from our review indicate that chimney ICS have shown to be effective in reducing HAP and improving health in research settings but achieving these benefits on a large scale has been challenging. The range of health benefits that have been achieved in the region through clean cooking programs, includes acute problems such as headaches and conjunctivitis mainly in women, to other benefits such as improvements in lung function. In children, a decrease in the frequency and duration of respiratory symptoms, the main cause of demand for medical attention, has been demonstrated, although the pneumonia risk reduction has not been demonstrated. Other less evaluated impacts are cardiovascular outcomes and other chronic diseases such as cancer because long-term studies are required. The documented benefits are undoubtedly linked to the decrease in the concentrations of different toxins well represented by respirable particles in addition to gases.
Although there are not many cohort studies carried out in the region, there is a significant amount of pre and post intervention studies. These studies increasingly have a sufficient follow up time to assess the magnitude of the impacts. It is desirable that more studies of this type be carried out to quantify the benefits more accurately when they exist. Regarding poor communities that rely heavily on solid biomass we find that chimney ICS interventions contribute significantly to the construction of healthier environments, to increase the quality of life and to reduce the time that especially children and women remain ill (García-Frapolli et al 2010).
6. Conclusions and recommendations
FW is still the dominant rural cooking household fuel within most LA countries, and by far, continues to be used on open fires and rustic stoves. While in the last 20 years there has been an increasing penetration of LPG—very important in countries like Ecuador, Brazil, and to a lesser extent Mexico and Peru—FW use has only been partly displaced because of stacking. Also, there is still a large rural population who does not have the economic means to access LPG or electricity, even on a partial basis. We have also shown that some of the new ICS chimney stove models disseminated in the region could provide tangible health and environmental benefits, as the stoves result in large GHG savings and PM2.5 indoor concentrations with regards to TSF. Under these circumstances, regional programs and policies to promote clean and sustainable cooking should include modern solid biomass devices, such as chimney ICS, within their portfolio of options.
To be successful, programs promoting clean cooking should move from just installing or selling stoves to favoring adoption and the understanding of families' priorities rather than just focusing on behavior changes. It is essential to promote participatory innovation cycles that depart from the study of traditional practices and technologies, co-develop pilot models, disseminate the different options including follow up with users, and monitor the program. Use of trials, quality certification, consultations with stove users, and the training of stove builders can help ensure stove quality and durability (Barstow et al 2016, INSP 2016).
NGOs and communities should play important roles in promoting stoves at the local level, including building capacity, facilitating distribution and installation, and contributing to subsidies at the household level. Smaller subsidies can be devised to keep stoves affordable while promoting commercialization.
Goals for ICS dissemination need to be clearly stated and national ICS plans launched and designed as part of the overall regional mandate; it is necessary to provide an enabling institutional environment, to support the development of new and advanced products, and to increase efficiency and scale for ICS dissemination. Governments should prioritize household biomass use on their agenda and designate a national coordinating authority that has oversight of energy, health, environment, and gender issues related to household biomass use. It is also important for the region to remove trade barriers related to ICS dissemination and to develop regional ICS standards together with testing and M&E protocols. A country-based regional campaign is necessary to make sure the general population knows why ICS and clean fuels are important, including fuel savings, health, and quality of life for women and children, as well as environmental sustainability (Wang et al 2013).
Clean cooking options should help freeing time, opening educational, economic, and social opportunities in which men and women can have equal access for the control and enjoyment of benefits. The involvement of women can increase the effectiveness of the project and help increase the adoption of products and services, while in turn impacting their own livelihoods (GACC 2014). Women help catalyze the market as clean cookstove entrepreneurs, they can drive large-scale distribution as well as the distribution of quality after-sales services which in turn will contribute to the creation of a thriving global market. Also, women can take advantage of their existing networks to encourage the adoption of these new technologies and use their own experiences to promote solutions.
The studies reviewed show that health benefits derived from the use of chimney ICS are clear. However, keeping these benefits on a long-term basis is directly linked to the sustainability of the interventions. Achieving WHO recommendations on healthy air, depends not only on the stove, but also on the social acceptance of the intervention and the technical characteristics and maintenance of the device. Despite all the evidence built and despite efforts in specific countries, the involvement of the health sector both in research and clean cooking interventions remain insufficient. Taking into account the large benefits of clean cooking, it would be expected that health ministries more actively support the development of programs for the promotion, intervention, and evaluation of clean cooking programs.
Finally, clean cooking programs must go hand in hand and be integrated into larger projects aimed at reducing poverty and inequalities in rural areas, since these last are the driving force that prevents universal access to clean household energy.
Acknowledgments
This research was supported by UNAM, INSP, PAPIIT IA105820, PAPIIT IG101121, SENER CONACYT 2014246911 Clúster de biocombustibles sólidos para generación térmica y eléctrica and the Clean Cooking Implementation Science Network (ISN).
Data availability statement
The data that support the findings of this study are openly available at the following URL/DOI: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/IGDP9B.