COMMUNITY EMPOWERMENT AND SUSTAINABILITY OF THE COMMON FOREST OF CHERAN , MEXICO By Aldo Fabian Gonzalez Camacho A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Community Sustainability Master of Science 20 20 ABSTRACT COMMUNITY EMPOWERMENT AND SUSTAINABILITY OF THE COMMON FOREST OF CHERAN, MEXICO By Aldo Fabian Gonzalez Camacho In Mexico, approximately 80% of the forest is commonly owned by local communit ies and its management is decentralized. E vidence suggests that in many cases, decentralization is ineffective in reaching positive social and environmental outcomes due to a lack of power for local actors . This research aims to investigate how communi ty empowerment impacts the sustain abil ity of common forest . During the 2000 ´s , the indigenous community of Cheran lost a third of its forest to foreign log ge rs linked to criminal organizations. Given the lack of response by aut horities, in 2011 the community conducted an armed resistance against t he intruders and recovered control over their territory . Then, the y completed a legal process that ended in the establishment of a traditional self - governance system . I investigated three research questions: 1) What kind of power did the community gain in relation to natural resources management? 2) How have the institutions for the sustainable management of common pool resources (CPRs) evolved after 2011? and 3) How has for est land - cover changed from 200 5 to date? To answer the first two research questions , I conducted a qualitative case study . I found that the community in fact gained authority over the management of the forest and improved their institutions to better resemble Ostr o ciples for sustainable management of CPRs . To answer the third res earch question, I p erform ed a semi - automated land cover change assessment between 200 5 and 2020. The area covered with healthy forest reached its lowest point in Cheran in 2014 but has been recovering ever s ince. This study provides empirical evidence sugg esting that community empowerment in decentralized forest management c a n facilitate the creation of robust forest institutions and reduce deforestation. iii ACKNOWLEDGEMENTS I want to specially thank Dr. Maria Claudia Lopez for her unconditional support thr oughout studies and for never giving up o n me , I will forever be thankful . The thesis also got direct contributions from the committee members . Dr. John Kerr and Dr. Laurie Medina played a crucial role in shaping the research from the beginning , while Dr. Grant Gunn introduced me to the implement ation of GIS tools and helped me design the methodology for land cover change assessment . The monetary contributions of the d epartment of Community S ustainability , the Graduate School and the Center for Latin American and Caribbean Studies were fundamental for the completion of this research. My work simply consisted in reporting the outstanding achievements of the people of Cheran ; I want this thesis to be a tribute to their combativ eness, unity, and c ommitment towards the conservation of the forest . I w ant to thank the Consejo de Bienes Comunales for facilitating the field work ; specially Don Angel Chapina, Doña Nieves , Lupita, Oscar, Jose Luis , Doña Isabel and the Guardia Comunitari a who looked out fo r me during my visit to Cheran . Fernando Jeronimo and Humberto Hernandez provided the initial contact with the community. I could have not survived graduate school and finished my thesis without the support of my friends in Michigan : Laura, Natalia, Alai na, Ramiro, Florencia, Marco, Alan, Sjona , Brock , Kyl e , Timmy , Bethany, Obafemi, Taylor and many more that I regrettably fail to mention. Despite the distance, my family continu es to be my strength . I refer not only to my parents Antonia Camacho and José González, and my siblings Sarahi and Marco, but also my many aunts, uncles, and cousins. iv TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ............... v LIST OF F IGURES ................................ ................................ ................................ ............ vi 1. INTRODUCTION ................................ ................................ ................................ ............ 1 2. LITERATURE REVIEW ................................ ................................ ................................ . 5 2.1 Common p ool res ource management ................................ ................................ .............. 5 ........................ 8 2.3 Forest commons in Mexico ................................ ................................ ........................... 10 2 .4 Decentralization, empowerment and natural resource governance .............................. 13 2.5 Land cover change assessment u sing Google Earth Engine ................................ .......... 14 3. SITE DESCRIPTION: THE COMMUNITY OF CHERAN, MICHOACÁN ................ 18 4. METHODS ................................ ................................ ................................ ..................... 22 4.1 Qualit ative methods. ................................ ................................ ................................ ..... 24 4.2 Land cover change assessment ................................ ................................ ..................... 28 5. RESULTS ................................ ................................ ................................ ....................... 31 5.1 Origins of the cri sis and mobilization ................................ ................................ ........... 32 5.2 Forest empowerment ................................ ................................ ................................ .... 35 5.3 Forest management institutions ................................ ................................ .................... 39 5.4 Landcover Change ................................ ................................ ................................ ........ 53 6. DISCUSSION AND CONCLUSION ................................ ................................ .............. 55 LITERATURE CITED ................................ ................................ ................................ ...... 61 v LIST OF TABLES Table 1 : Bundles of property rights (Adapted from Schlager & Ostrom (1992)) .......................... 6 ................................ ....................... 9 Table 3: Ban d designation of Landsat 7 (NASA, 2019) ................................ ............................... 15 Table 4: Confusion matrix example ................................ ................................ .............................. 17 Table 5: The evolution of violence in Mexico (Adapted from Merin o & Cendejas, 2017) ......... 21 Ta ble 6: Qualitative data collection technique ................................ ................................ .............. 25 Table 7: Landsat Imagery ................................ ................................ ................................ ............. 29 Table 8: Change in power (Empowerment) after the mobilization of 2011 ................................ . 38 Table 9: Appropriati on rules ................................ ................................ ................................ ......... 43 Table 10: Fi rewood extraction rules ................................ ................................ ............................. 44 Table 11: Accuracy parameters of the supervised classifications. ................................ ............... 53 Table 12: Land cover change ................................ ................................ ................................ ........ 54 vi LIST OF FIGURES Figure 1: Facilitating conditions for CPR sustainable management (Agrawal 2001) .................... 8 Figure 2: NDVI temporal spectra for land cover classes of interest (Knight et al., 2 007) ........... 16 Figure 3: Locati on of the study site ................................ ................................ .............................. 19 Figure 4: Study area and training polygons ................................ ................................ .................. 31 Figure 5: Nesting of Enterprises ................................ ................................ ................................ ... 52 Figure 6: Land Cover Change within the Forestry Areas of Cheran ................................ ............ 54 Figure 7: Maps of land c over by class ................................ ................................ .......................... 55 1 1. INTRODUCTION Local communities own and manage 80 % of the forests in Mexico (Bray et al., 2003) . In 1992, the Mexican agrarian law was modified as a con sequence of neoliberal policies; communities gained the right of alienation 1 over the land and the forest management was decentralized (De Janvry et al., 2 001) . According to the literature on decentral ization, this policy was meant to promote a more effi cient, equitable and sustainable use of natural resources (Ribot, 2002) . However, some authors arg ue that neoliberal corrupt elites oftentimes take power away from local communities to serve their own interests, and fail to provide security and legality (Cortés Calderón, 2018) . In general, the decentralization of forest management i n Mexico has failed to deliver positive environmental and social outcomes (Deininger & Minten, 1999; FAO, 2010) . Between 1990 and 2000 Mexico lost an average of 354,000 hectar es of forest per year , occupying the 7 th place worldw ide in deforestation for that period (FAO, 2010) . In addition, a ccording to the World Bank (2005) , more than half of the Mexican population l ives under the poverty line and 60% of the poor live in rural areas. Deininger & Minten (1999) found that poverty is associated wi th higher l evels of deforestation in the country . In some countries, the decentralization process did not include an increase of stakeholder s power to create rules and make decisions over natural resources (Agraw al & Ribot, 1999) . Some em pirical evidence from Bolivia, Kenya, Mexico, and Uga nda suggest s that disempowered local actors are less likely to engage in effective collective action after decentralization of natural resource management (Colem an & Fleischman, 2012) . 1 The right to sell or lease the land (Schlager & Ostrom, 1992) 2 Em powerment is the process by which people gain power; meaning that they expand the authority and control over the resources and decisions that affect their lives (Narayan, 2007) . It is a mainstream concept in the development literature and central to the work of many development organizations; it is r ec ognized as an instrument for community development, p overty alleviation, public health and local governance (Alsop & Heinsohn, 2005; Hennink et al., 2012; Luttrel l et al., 2007) . In the context of natural resources , empowerment is central for sustainability and refers to the right to access, use and manage resources by individuals or communities (Hennink et al ., 2012) . The objective of this research is to use the c ase of the indigenous community of Cher a n, Mexico to study the relationship between community empowerment , resource governance and sustainability of the commons under decentralized management. The f orest group or comunidad de Cheran collectivel y own 17 ,000 ha of temperate forest in the State of Michoacán, Mexico. F orest users live in a town called Cher a n , which has approximately 18,000 inhabitants and it is the capital of the municipality with the s ame name. illegal logging became a regular practice in Cheran . The situation in the whole country was chaotic by the federal government . O utsiders linked to criminal organizations took advantage of the situ ation to rob wood from Cheran , and state authorities failed to stop them . C onsequen tly , the people fr o m Cheran organized an armed resistance against external loggers and recovered control of their territ ory (Luján - Álvarez et al., 2015; Alejand ra Velázquez, 2013) . Dissatisfied with the performance of State authorities, the community deci ded to get legal advice and succeeded to substitute the municipal government with an autonomous form of government recognized by the Federal Government of Mexico (Luján - Álvarez et a l., 2015; Alejandra Velázquez, 2013) u sos y nd cost o and it recognizes indigenous communitie 3 their authorities according to their traditions . This figure is widely used i n states such as Oaxaca where most inhabitants are indigenous (Valdiva Dounce, 2010) . The case of Cheran provides the opportunity to assess the relationship between empowerment , governance and environmental sustainability of the comm ons , in this case the forest . I explored three resear ch questions: 1) What kind of power did the community gain in relation to natural resour ces management? 2) How have the institutions for the management of common pool resources (CPRs) evolved after 2011? and 3) How has forest land - cover changed from 200 5 t o date? I applied a mix ed methods approach to answer the se research question s . Using qua litative research methods, I assess ed the process of empowerment and the evolution of forest management institution s to answer question 1 and 2 . The quantitative part o f this research consisted of a land cover change analysis performed using Google Earth Engine to answer question 3 . I found that the community gained considerable authority over the management of the for est. After 2011, the power to enforce rules , also kno wn as judicial power was completely shifted from the S tate authorities to the com munity. The power to create rules was not totally transferred to local authorities , but the community certainly gained inf luence over the creation of forest institutions . Befo re 2011 , the community already had de jure executive power; however, after the mobilization they got de facto power for executing the decisions over the fores t . In terms of common forest institutions, the community experienced several changes after 2011. d in endogenous and dependent (Ostrom 1990, Agrawal 2001). The first six principles are considered endogenous because they are de termined and affected only the user group, while the last two principles depend In fact , Agrawal (2001) states m ost 4 of Ost within the local context. Only two of the m , about leg al recognition of institutions by higher level authorities and about nested institutions, can be seen to express the relationship of a given group with other groups or authorities (p. 1652) . Endogenous institutions became more robust in Cheran after 2011. On the other hand, the community struggled to gain recognition of their rights to organi z e and their institutions are not fully nested with h igher levels of governance. In 2005, Cheran had 14,000 hectares of healthy forest. By 2014, that number fell to 11,000 hectares , 3,000 hectares of forest were lost . In 2020, after 9 years of the empo werment of the local community, the surface covered b y healthy forest is 13,000 hecta res or 92% of the surface they had in 2005. This thesis has both practical and theoretical contribution s which are linked . This research contributes to the literature of the commons, but more importantly to the one on decen tralization and how to devolve f unctions to local communities and that includes the devolution of power. On the other hand, a practical contribution is the fact Cheran can serve as an example on how pro viding power to communities can improve the governanc e of the forest. Cheran improved its forest through the design of institutions , this is a crucial lesson for other communities in Mexico c onsidering that the country is struggling with deforestation . The people of Cheran had to go to many difficulties and legal battles to get that right. It is fundamental to make decentralization real and easier for communities. The rest of this document is organized as follows: Section 2 contains a review of the re leva nt literature; Section 3 describes the study site in detail; Section 4 explains the research method; section 5 shows the findings of this research and Section 6 is a general discussion. 5 2. LITERATURE REVIEW 2.1 Common pool resource management tra (Hardin, 1968) describes the expected deg radation of common resources when many u sers have acc ess to them , because each user seeks to maximize his immediate profit by extracting as many units of the resource as possible. article was not supported by empirical evidence, his pa per was translated into the claim that o nly private p roperty or centralized management exerted by the State could avoid the degradation of ized control and privatizatio n policies , which resulted not only in t he dispos sess ion of stakeholders , but in the degradation of resources (Ostrom, 1990) . After Hardin s article, researche r s started to investigate case studies from around the globe where local communities had successfull y engage d in collective action , resulting in sustaina ble use of resources (Baland & Platteau , 1996; Ostrom, 1990; Wade, 1988) . These studies showed that the tragedy of the commons did not always occur . Furthermore, the stu dies show ed article failed to make the distinction between open access and common property , and that communit ies can create rules to manage their shared resources. This has led to the misconception that communities could not communally manage their resources in a successful way (Schlager & Ostrom, 1992) . In addition, Ostrom (1990) created the term esources (CPR) to refer to any natural or man - made resource system that is sufficie ntly large as to make it costly (but not 1990:30). CPRs have two c haracteristic s: subtractability and excluda bility (Ostrom, (1990) . Ostrom (1990) analyzed cases of local communit ies that successfully engaged in collective action 6 to manage their CPR s in a successful way. She proposed eight design principles that were the shared features of the robust user - organized systems who we re developing sustainable CPR management. Schlager & Ostrom (1992) noted that all regimes that were not private or state property were assumed to be of one single kind, disregarding the diffe rence between reg imes, situation that was leading to errors in policy analysis. In addition, they proposed a classification of the property - right regimes that include d authorized user, claimant, proprietor and owner. Eac h of those categories corresponds to a distinctive bu ndle of the five possible property rights: access, withdrawal, management, exclusion and alienation (Table 1). Table 1 : Bundl es of property rights (Adapted from Schlager & Ostrom (1992) ) Right Description Owner Proprietor Claimant Authorized User Access Enter a defined physical property Yes Yes Yes Yes Withdrawal Obtain the products of the resource Yes Yes Yes Yes Management Regulate internal use patterns and transfo rm the resource Yes Yes Yes No Exclusion Determine who will have an access right Yes Yes No No Alienation Sell or lease the above rights Yes No No No As Table 1 shows, a ccording to Schlager & Ostrom (Schlager & Ostrom, 1992) , o wners possess all five property rights; therefore, they have incentives to take long term actions for the conservation of reso urces. sers lack of authority to partici pate in collective action to to follow, they are less likely to agree to the necessity and legitimacy 7). T his 7 situation can lead to de struction of the resource unless mechanisms for monitoring and enforcement of rules are set in place (S chlager & Ostrom, 1992) . According to Gibson, McKe an, & Ostrom (2000, Chapter 6) there are three fundamental requirements for successful common pool resource management by local people: 1) u sers must highly value the natural resource, 2) they must ha ve property rights over the resource , and 3) they must be able to create institutions to regulate the use of the resource. These three requirements have been widely studied (Bremner & Lu, 2006; Ensminger, 1997; Anne M. Larson & Soto, 2008; Tucker, 1999; Velez, 2011; Velez et al., 2020) . Tucker (1999) found no differences in forest conservation lost in Honduras between two different property rights regimes: commo n and individual property rights . She attributed the success of common property to the existing institutions historically created by local communities . After reviewing the existing l iterature, Larsen & Soto (2008) argue that having the right to manage natural resources and excluding outsiders from using th em are essential conditions for sustainability. Velez et al (2020) found that formalizing property rights via collective tit l ing in Afro - Colombian communities significantly reduced deforestation when locals were involved in the creation of rules for the use of the resources an d also were able to exclude private companies from extracting timber. Agrawal (2001) classified the characteristics of common property identified by Wade (1988) into four categories: 1) Resource system characteristics, 2) Group characteristics, 3) Institutional arrangements and 4) External environment. Two additi onal categories contain the relationship between 1 and 2 (Resource system characteristics and Group ch aracteristics) and between 1 and 3 (Resource system characteristics and Institutional arrangements). Figure 1 shows a synthesis of Agrawal s classificatio n. 8 Figure 1 : F acilitating conditions for CPR sustainable management (Agrawal 2001) 2.2 Ostrom s design principles as an assessment tool for CPR management design pri nciples help to look at the CPRs governance structure s (Coria & Sterner, 2011; Cox, 2014) , knowing that the design were supposed to necessary conditions but not sufficient (Ostrom, 1990). The principles have been used as a f ramework to study resource sustainable management of a big variety of CPR systems at the local (Kamran & Shivakoti, 2013; Sattler et al., 2015; Trimble & Berkes, 2015; Vélez, 2009) , national (Lacroix & Richards, 2015; Lancaster et al., 2015; Raudla, 2010; Sarker et al., 201 5) and even global level (Johnson - Freese & Weeden, 2012) . Based o n the meta - analysis of 91 studies that evaluated Ostrom s design principles , Cox et al (2010) proposed some improvements to Ostrom initial list as shown in Table 2. 9 Table 2 (Cox et al (2010) Principle Description 1A User bo undaries: Clear boundaries between legitim ate users and nonusers must be clearly defined. 1B R esource boundaries: Clear boundaries are present that define a resource system and separate it from the larger biophysical environment. 2A Congruence with local conditions: Appropriation and provision rules are congruent with local social and environment al conditions. 2B Appropriation and provision: The benefits obtained by users from a common - pool resource (CPR), as determined by appropriation rules, are proportional to t he amount of inputs required in the form of labor, material, or money, as determin ed by provision rules. 3 Collective - choice arrangements: Most individuals affected by the operational rules can participate in modifying the operational rules. 4A Monitoring users: Monitors who are acc ountable to the users monitor the appropriation and p rovision levels of the users. 4B Monitoring the resource: Monitors who are accountable to the users monitor the condition of the resource. 5 Graduated sanctions: Appropriators who violate operational rules are likely to be assessed graduated sanctions (depending on the seriousness and the context of t he offense) by other appropriators, by officials accountable to the appropriators, or by both. 6 Conflict - resolution mechanisms: Appropriators and their officials have rapid access to low - cost local arenas to resolve conflicts among appropriators or between appropriators and officials. 7 Minimal recognition of rights to organize: The rights of appropriators to devise their own ins titutions are not challe nged by external governmental authorities. 8 Nested enterprises: Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises. In the case of forest, Johnson & Nelson (2004) u sed the design principle s to document forest status a fter severe fires in a national park in Mexico and analyzed the existing common functioning of the co mmon property and provides insights regarding the out (Johnson & Nelson, 2004, p. 728) . 10 Gautam & Shivakoti (2005) assessed forest governance system s in Nepal using the design principles ; the authors f ound that of the two case studie s they did , the forest with relatively good conditions was the one with institutions that better resemble d design principles . Van Ast, Widaryati, & Bal (2014) , most of the design principles w ere prese nt in the functioning of its traditional ins titutions. Moreover , the authors suggest ed that the presence of th ose design principles contributed to the sustainability of the forests over centuries (van Ast et al., 2014, p. 302) . 2.3 Forest commons in Mexico The Mexican R evolution (1910 - 1917) was led by peasants whose main demand was land redistribution because most of the land in the country was owned by few very rich families, while most of the population was landless and poor. As a result, the current C onstitution of Me xico (1917) recognizes three types of rural property: national lands, private property, and the agrarian reform sector or e jido sector. The Mexican Constitution granted long - term and stable usufruct rights over the land to the peasants; however, ultimate ownership was granted to the State, which means that stakeholders did not have the alienation right. By excluding this righ t, the State retained great control over the decision - making in the communities; a situation that De Janvry et al. (2001) called State - led governance. Between 1920 and 1940, president Cardenas led the agrarian reform that s (105 million hectares) tenure into local communities (both ejidos and communities) (Barnes, 2009; Bray et al., 2003) . In 1992, Article 27 was modified to promote community - led management. The new legislation allows privatiza tion of individual plots, association with private in terests outside the ejido and the use of land as collateral for credit. More importantly, decision - making was 11 decentralized , giving communities full control over the land. In that sense, ejido sector pro prietors became common owners after th at amendment to agrarian law (De Janvry et al., 2001) . The same year, the forestry policy was also modified to match the A rticle 27 amendment and the liberalization of the market enforced by the North American Free T rade Agreement (NAFTA). Forestry activities were dere gulated allowing ejido communities to esta blish commercial plantations in partnership with private enterprises. The reform was meant to integrate forestry communities to the global market. However, feder al policies were poorly designed and lacked incentive s for conservation that resulted in deteri oration of local institutions, impoverishment of the communities and forest degradation characterized by clandestine logging (Pérez, 2003) . B etween 1990 and 2000 Mexico lost an averag e of 354,000 hectares of forest per year , occupying the 7 th place worldwide for deforestation for that period (FAO, 2010) . Most of the deforestation happ ens in the S outh of the country where biodiversity is greater (A. Velázquez et al., 2002) . Estimates range widely , but the consensus is that d eforestation rate s ha ve been declining since 2000 (Kolb & Galicia, 2012) . Still, Bradshaw, Giam, & Sodhi (2010) ranked Mexico in 9 th place for natural forest lost in 2010 . Deininger & Minten (1999) investigated the contributi on of different socioeconomic and geographic factors to deforestation in Mexico using land use data and its changes; they did not find differences between common and private property. They also concluded that provision of technical assistance decreases defore station , while poverty is associated with higher leve ls of deforestation. According to the Worl d Bank (2001) , the ejido represents an opportunity for the implementation of poverty alleviation and forest conservation strategies because a large proportion 12 of the forest is owned by poor communities. However , the same repor t mentions that forest policy provides mini mal support to communal forest owners , and add that f orest policy focuses on control rather than incentives which makes compliance difficult and expensive (Wo rld Bank, 2001) . The excessive regulatory burden of forestry makes it less attractive than agri culture for land owners which is an incentive for them to clear the forest for crops (World Bank, 2001) . Still, the most functional and productive ejidos in Mexico are those that focus on forestry rather than a griculture (Barnes, 2009) . According to (Barsimantov et al., 2010) e ven though the reform of 1992 enabled owners to transition to private property, c ommunities whose main activity was forestry have been less likely to change from communal to individual property possibly because forest can be m anaged more effectively as a whole ; o the r factors that relate to the permanence of common property are strong governance and ethnic homogeneity . Despite the loss of forested area in the country as a whol e, Mexico has some remarkable cases of sustainable co mmunity - based forestry (Bray et al., 2003, 2005; Merino & Cendejas, 2017; Orozco - Quintero & Davidson - Hunt, 2010; Pérez, 2003; Tucker, 2004) . Under appropriate institutional condi tions and thanks to their high levels of social capit al, forestry communities have shown significant capacity to deliver positive social, economic and ecological outcomes. Orozco - Quintero & Davidson - H unt (2010) study a well - known examp le of such forestry communities is San Juan Nuevo in the Purepecha region of the State of Michoacán. The authors found enough empirical evidence to support the communi ng a positive socioeconomic impact t hrough forestry communal enterprise; all of this, thanks to very robust institutions . 13 2.4 Decentralization, empowerment and natural resource governance Decentralizati on is the devolution by central government of specifi c functions to regional and local agents that are independent of the center (Faguet, 2014) T he theoretical arguments in favor of decentralization include its potential for increasing accountability, citizen voice and governanc e as well as limiting abuses of power by central authorities (Faguet, 2014) . M any countries have d ecentralized forest management and the effects of such reforms have been studied widely ; according to the e mpirical evidence , decentralization social and environmental outcomes differ from country to country (Larson & S oto, 2008; Ribot, 2002; Ribot et al., 2006; Wright et al., 2016) . For example, Wright et al . (2016) studied decentralized forest management in Peru and Bolivia, and found that decentralized management can reduce deforestation but only when forest user gr oups are actively engaged in governance, and have mea ningful roles in decision - making, monitoring an d conflict resolution. Lars o n & Soto (2008) argue that the results of decentralizatio n are generally disappointing because the polices have been poorly implemented as they rarely change the power relations in natural resources management. In the same line of argument, Larson et al (2010) found that decentralization polices enacted around the world have legally transferred pr operty rights to forest user groups , but in practice the transference of power has been limited by regulations, competing claims and S tate actors seeking for the retention of control . According to Colem an & Fleischman (2012) , the ecological and socioecono mic outcomes of local collective action for the management of the commons in a decentralized environment are a function of empowerment and accountability. Decentralization initiatives that include strong accountability and loc al empowerment are more likely to achieve their stated policy goals: resource s sustainability and improving rural livelihoods (Coleman & Fleischman, 2012) . 14 Narayan (2007) defines empowerment as the expansion of assets and capabilities of vu lnerable groups; such expansio n results in people s increased power to control their li ves . According to Cornwall (2008) , empowerment is a continuing, dynamic and transformative form of participation where people are able to decide and act by themselves; citizen control or citizen power is the highest level of participatio n. Both definitions consider that empowerment is a p rocess whose outcome is an In the context of natural resources, empowerment refers to the ability of indivi duals or communities to access, use, and manage natur al resources towards sustainability (Hennink et al., 2012) . Coleman & Fleischman (2012) define empowerment in that context as the transfer of autho rity from central government to local actors so that they can make decisions over their resources . Theore tical and empirical evidence suggests that sustainable decentralized management of the commons necessarily requires effective empowerment in all its in stances to local representative authorities (Agrawal & Gibson, 1999; Anne M. Larson & Soto, 2008; Ribot, 2002; Wright et al., 2016 ) . Agrawal and Ribot (1999) distinguish three types of power involved in the decentralized management of CPR: 1) the power to create and modify rules or legislati ve power , 2) the power to make decisions a bout the us age of resources or executive power , and 3) the power to implement and ensure compliance of the rules and adjudicate disputes or judicial power . 2.5 Land cover change assessment using Google Earth Engin e Google Earth Engine (GEE) is a free - access potent t ool for the investigation of land cover changes (Nyland et al., 2018) . GEE is not only a repository of satellite im agery and geospatial datasets but also a cloud - based geospatial processing platform that utilizes a Javascript or Pyt h on 15 Application Programming Interface (API) which makes it s use very efficient for users with varying computing resources ( Google Earth Engine , 2020) . Landsat imagery For the las 4 decades, Landsat program has continuously provided optical sate llite imagery of the earth´s surface thanks to the series of 8 satellites launched between 1972 and 2003. Landsat imagery is cataloged using the Worldwide Reference System (WRS) which assigns sequential Path numbers (from east to west) to the 251 satellite tracks. As the satellite moves along the path , a raw number is assigned to the latitudinal center line of a frame. Paths and Rows form a matrix where each image can be identified by this notation ( The Worldwide Reference System. Lan dsat Science , n.d.) This study utilizes da ta acquired by Landsat 7 from 2005 to 2020. Landsat 7 was launched in 1999 and is equipped with an earth - observing instrument called the Enhanced Thematic Mapper Plus (ETM+) which incorporates hardware and pro cessing improvements over its predecessors , as well as improvements in spatial resolution in the th ermal bands . In 2008 the Landsat archive of imagery became free to the public ( The Worldwide Reference System. Landsat Science , n.d.) . Table 3 shows the band designation for the ETM+ instrument on board Landsat 7. Table 3 : Band designation of Landsat 7 (NASA, 2019) Band W avelength (micrometers) Resolution (meters) Band 1 0.45 - 0.52 30 Band 2 0.52 - 0.60 30 Band 3 0.63 - 0.69 30 Band 4 0.77 - 0.90 30 Band 5 1.55 - 1.75 30 Band 6 10.40 - 12.50 60 (30) Band 7 2.09 - 2.35 30 Band 8 .52 - .90 15 16 Normalized Difference Vegetation Inde x (NDVI) The Normalized Difference Vegetation Index (NDVI) is an indicator of the presence of healthy vegetation within a pixel of a satellite image, it is calculated using the values for visible and nea r - infrared light in each pixel using this formula: ND VI= (NIR Red)/(NIR + Red). Healthy vegetation absorbs most of the visible light that hits it and reflects a large portion of the near - infrared light. Unhealthy or sparse vegetation reflects more visibl e light and less near - infrared light. NDVI ranges fro m - 1 to 1; healthy vegetation presents NDVI values higher than .6 while grassland usually has low values. Negative NDVI values correspond to bare soil (Herring, 2000) . Figure 2 : NDVI temporal spectra for land cover classes of interest (Knight et al., 2007) As shown in figure 2, the NDVI values of coniferous are constant througho ut the year whereas the NDVI values of agriculture la nd and deciduous vegetation drastically drop during the dry season. The difference in NDVI values during the dry season (October - m arch) among classes make it easy for classification algorithms to differe ntiate them. Random F orest classification 17 Random F or est is a machine learning classification method that constructs a multitude of decision trees at the time of model training and pr oduces the class that is the of the classes of the individual trees ( Decision Tree Learning , n.d.) . Random decision forests corr ects for decision trees' habit of overfitting to th eir training set (Breiman, 2001) . When used in for the classification of imagery, the random forest builds an array of classification and regression - tree classifiers that use bootstrapping of pixels contained in training polygons for ea ch land cover class. The class for each pixel is dete rmined via a voting process of the trees where the most popular class at each node is selected (Nyland et al., 2018) . An accuracy report of a random forest classification usually includes a confusi on matrix, estimates of overall accuracy, kappa coefficient, user accuracy, and producer accuracy for each land cover class. A confusion matrix is a ta ble containing the actual classes in each column while rows contain the number of pixels predicted by the algorithm for each class (Foody, 2002) . Table 4 is an exa mple of a confusion matrix. In addition to regular ac curacy indicators (Overall, users and producer accuracy) the kappa coefficient is often included because it compensates for chance random agreement. Equations 1 - 3 define how the accuracy parameters are c alculated as described by Foody (2002) . Table 4 : Confusion matrix example Actual class Predicted class A B C D A n AA n AA n AA n AA n A+ B n BA n AA n AA n AA n A+ C n CA n AA n AA n AA n A+ D n DA n AA n AA n AA n A+ n +A n +A n +A n +A n 18 3. SITE DES CRIPTION: THE COMMUNITY OF CHERAN , MICHOACÁN Cheran is one of the 11 municipalities within the Purepecha region of the State of Michoacán, Mexico. The total area of the municipality is 222.80 km 2 and its population is 12,000. It is in a tropical area with an altitude of 7 , 874 feet above sea level , where conifer forest is the dominant vegetation type 2 . The surface of Cheran municipality is 24,000 ha and 83% of the total surface of the Cheran municipality i s commonly owned by 600 indigenous people (Rojas, 2012) . ancisco Cheran Cheran , which in 2010 had a total population of around 18,000 inhabitants, mostly Purepecha 3 indige nous people. Figure 3 shows the location of Cheran. 2 http://www.inafed.gob.mx/work/enciclopedia/EMM16 michoacan/municipios/16024a.html 3 Purepecha is the 13 th biggest indigenous group in Mexico with a population o f around 200,000 people. 19 Figure 3 : Location of the s tudy site During the Spanish colonial period in Mexico (1521 1810), original communities of the Purepecha region received common property title s over this land. After Mexican independence (1810), mestizo commu nities started to gain recognition as municipalities. Cheran municipality was established in 1868. During President Diaz administration (1876 1910) the original territory of Cheran was in gr eat part divided and transferred to private property; US owned companies initiated the industrialized extraction of wood thanks to the assistance of local authorities and the purchase and rental of extensive areas , situation that caused concentration of la nd (Rojas, 2012) . Through the Mexican revolution, Cheran went through a turbulent period as did the rest of the country , but the indigenous governance structures survived. During President Cardenas administration (193 4 - 1940) the agrarian reform was extensively applied , dist ributing land among the peasants with no land and creating the ejido sector. In Cheran , the agrarian reform was not 20 welcomed because it implied greater control by the S tate. Local groups in favor of the agrarian reform, common land indigenous original owne rs and private owners continued to dispute the control of the land until 1984 when the federal government solved the disputes establishing well - defined limits between the different kinds of property. About 80% of the land was defined as common property and the rest was individually owned. That arrangement persist s until today with one exception already mentioned, in 1992 the common property regime was given the right of alienation (Rojas, 2012) . According to Pérez (2003) , f orestry activities in the region were more or less successful thanks to the support of the S tate until the 90s when neoliber a l polic ies destroyed the local governance institution s, cut the public investment and opened the markets to international trade. This institutional reconfiguration caused a crisis of forest degradation and impoverishment in the communities (Pérez, 2003) . In this context of failed neoliberal policies , a sector of the population R (Revolutionary Institutiona l Party or PRI) demanding better live conditions and the recognition of their traditional indigenous authorities. Ho wever, the movement was eventually incorporated to he leading leftist party in Mexico at the time which diluted its purpose and power (Nelson, 2003) . During the last 30 years, the violence in Mexico has risen to levels that resemble that of war zones, forcing local communities like Che ran to organize and take control of their own safety (Merino & Cendejas, 2017) . In order to understand the context in which Cheran´s mobilization came a bout, it is crucial to know the evolution of violence in Mexico . , the product of a failed war on crime. Table 5 present s a summary of the hi story of violence related to organized crime in Mexi co according to Merino an d Cendejas (2017) . 21 Table 5 : The evolution of violence in Mexico (Adapted from Merino & Cendejas, 2017) Period Highlights 1940´s In response to the increasing demand in North America, Mexico started producing poppy an d marijuana in large - s cale Drug dealing was mostly tolerated by the local government 1980´s Mexico became the main area of transit of drugs coming from South America and Asia to the US The economic and social influence of the drug cartels increased 1990´s The US government promoted direct intervention of the a rmy a nd local paramilitary groups in the battle against dr ugs in Latin America. Anti - drug polic i es led to the professionalization of drug cartels in the form of greater military power and resources to corrupt authorities. 2000´s The power of the Mexican drug cartels increased even more after the dismantling of the Colombian drug cartels. 2000 - 2006 President Fox modified the anti - drug strategy of the country and for the first time directly confronted the drug cartels. 2006 - 2012 President Calderon toughened action against drug cartels by deploying a large frac tion of the national army into the streets. In response to the increased pressure, criminal groups diversified their activities attacking the populations and seeking control of the local govern ments. Th e number of murders increased by %240 in relation to previous federal government, more than 50,000 people were killed. Mexico was ranked among the countries with the worst Global Peace Index In response to the failure of the State to provide s ecu rity, seve ral social movements and organizations emerged to sel f - secure their territories. 2012 - 2018 President Peña maintained the war against drugs and formally instituted the army as the provider of security. The repression of the press grew stronger Social pro test was criminalized An estimated 450,000 people wer e directly employed by criminal organizations When the administration of President Calderon (2006 - drug cartels of Mexico , a crisis of crime and violence took over the s tate of Michoacán. Taking advantage of the chaotic si tuation, criminal groups stole wood from community forest on a regular 22 basis in Cheran . Around 3,000 hectares of its forest was lost to the illegal loggers (E spaña - Boquera & Campo - Jiménez, 2016) . Despite multiple complaints by the forest users group, local authorities were unable or unwilling to stop the illegal logging (Rojas, 2012; Ventura Patiño, 2012; Wainer, 2016) The community started a n armed resistance against the illegal loggers and an emancipation process i n 2011 . In April of that year, a small group of wome n stopped some trucks that carried illegally obtained wood from their forest. Th is event triggered a series of struggles between the community and loggers. The mobilization of the community increased dur ing the next days until a general assembly decided to disavow the municipal government. That same year, Cheran refused to participate in the S tate elect ions , arguing that the regional government never listened to their demands to stop illegal logging and v iolence. After a complicated legal process, Cheran ac hieved electoral independence and currently they are self - governed under a traditional scheme (Rojas, 2012; Alejandra Velázq uez, 2013; Ventura Patiño, 2012) . 4. METHODS For th is study, I followed a qualitative case - study approach in Cheran. A c ase stud y approach is widespread in social sciences including policy science and community studies where the researchers have little c ontrol over the circumstances of the object of stud y, for example in the study of real - life phenomena (Yin, 2009) . The case of Cheran is particularly important because in a relatively short period, the community went through a radical change in its governance structure s , and those impac ted the forest. The selection of methods in this case study aim s to gather information that help me understand what was happening with the community and in particular what institutions they created to govern the forest with the power they gained in 2011, a nd how those institutions were reflected in the ecolo gical sustainability of the forest. 23 After several years of lawlessness, the community organized and mobilized to stop criminal organizations form robbing their forest. Furthermore, the citizens of Chera n undertook a process that ended in the esta blishment of a traditional form of government , with considerable independence from the State . The gain in power by the people of Cheran was drastic and closely related to the defense of the commonly own ed forest, thus, the case study approach gives me a un ique oppo rtunity to study the relationship between community empowerment and sustainability of the commons . Not surprisingly, c ase stud ies are prominent in the common forest literature (Gautam & Shivakoti, 2005; Klooster, 1998; Tucker, 2004; Velez et al., 2020) , however many of the case stu dies only report social an d institutional data, not often do they also incl ude land cover assessment. As mentioned earlier, I aim to answer three research questions: 1) What kind of power did the community gain in relation to natural resources management? 2) How have the instituti ons for the sustainable management of common pool resources (CPRs) evolved after 2011? and 3) How has forest land - cover changed from 20 05 to date? In order to answer questions one and two , I conducted in - depth interviews, focus g roups and participant obse rvation and analyzed the collected data using the qualitative methods of case studies as described by Yin (2009) . Question three was answered using Google Earth Engine to run a semiautomated land cover change assessment from 200 5 to 2020 . The mixed methods approach allows me to determine the effectiveness of the with forest land cover as the main output of effective forest governance . The methods section is organized as follows, I will first describe t he methods for data collection a nd data analysis to answer questions 1 and 2, and then I will proceed to explain the methods to answer question 3. 24 4.1 Qualitative methods . I a sses s empowerment as the change in authority o ver forest management . The literature acknowledges three ki n ds o f power (Agrawal and Ribot, (1999) : a) the power to create rul es or legislative power, b) the power to make decisions or executive power and c) the power to enforce rules or judicial power. To facilitate the assessment of power change , I used this division in the collect ion and analysis of information . In reality, th os e powers overlap and change f rom country to country but the division still offers a solid analysis framework. For each of the three types of power, I collected information through interviews done with fores t authorities and forest users about who exerted the authority and how , both before and after the establishment of the autonomous government in 2011 . Th is allowed me to determine to what extent the community gained influence ov er the decisions, laws , regul ations and rule enforcement in relation t o their forest . The interviews also allowed me to collect information about the origins of the deforestation crisis and the mobilization to complement the existing literature and better understand the context of the empowerment process. In order to asse ss fores t governance, I looked at the development of institutions using s as a framework . I collected information about the compliance of each principle before and after the process of empow erment. Data collection Before arriving to the c ommunity of Cheran, I contacted the P resident of the C ommon G oods , the executive body of the forest users group, Mr. Angel Chapina and we agreed on the terms of my visit and research. When I arrived to Chera n, I was granted written permission by Mr. Chapin a to do interviews and collect data in the forest. Written permission is essential because locals mistrust outsiders and the forest rangers continuously survey the forest looking for illegal 25 loggers. In orde r to protect the identity of the informants, I ga ve all of them an Alias that I use in this document. All the field work was conducted during the months of June and July 2016 . I stayed in a cabin that the community built for visitors . This was fundamental for the success of my data collection because it allow e d me to become familiar with the different dynamics of the community. I conducted semi - structured interviews and focus groups involving key informants: forest users, local authorities, employees from s. The key informants were selected according to their area of expertise and experience. D ata collection consisted of a total of 25 interviews, 3 focus groups and p articipant observation w ere conducted on ke y activities like resin extraction and collection , f orest surveillance by the rangers and meetings of the forest authorities. Table 6 presents the number of interviews and focus groups done by type of key informant. All the forest users and the community e interviewed were m e n, bec ause traditionally they perform the field work. However, the forest , municipality, and state authorities I interviewed included a similar amount of m e n and wom e n. Table 6 : Qualitative data collection technique Key informants Intervi ews Focus Groups Local forest authorities 9 2 Municipal authorities 4 1 4 0 Forest users 6 0 State authorities 2 0 Total 25 3 Semi - structure interviews For the study, I conducted 25 interviews with 5 different types of informants: local forest authorities, municipal authorities, 26 authorities . The interviews were semi - structured in nature , which means that few ques tions were prepa red in advance depending of the e xperience of the informant and his area of expertise; however, the conversation could flow freely . All the interviewees were purposefully selected. Thanks to the interaction with Mr. Chapina , he introduced me to other local authorities and forest users. T his introduction was necessary to gain the trust o f the respondents. Then , I took a snowball sampling approach to interview people, which means that I identified additional subjects relevant to the investiga tion during the interviews and participant observ ations. I took notes in all interviews and got audio recordings when people gave me authorization to do so . In total, I audio recorded 6 interviews of 25 . After the interviews , I reviewed and extended my not es. The interviews lasted between 1 - 2 hours . Foc us groups I conducted two focus groups with local authorities , with four and five people respectably. I choose focus groups because the local authorities met regularly and they gave me to join these meetings to talk to them as a group. I took notes and rec orded audio . During the focus groups we talked about desig n principles in a semi - structured way, which means that I guided the participants with ques tions but for the most part the conversation flow ed freely. Participant observations I conducted participant observations of different activities associated with the forest. With authorization of the different participants, I took pictures, audio recorde d the observation and took notes . The activities were: Collection of fallen wood Processing of wood 27 Preparation of trees to produce resin Collection of resin Processing of resin Forest users meeting determine reforestation strategies Fence repa i rs Rangers patrolling Supervision of newly reforested areas Activists for the defense of the forest meeting Administrative tasks in the Co u ncil of Common Goods office C ollection of edible herbs and mushrooms in the forest Data condensation and data analysis . I do cumented interviews without aud io recordings and observations using field notes. For all the othe r interviews, t he transcriptions were complemented with expanded notes . All that constituted the raw information for the qualitative analysis . All the informa tion was collected and anal y zed in Spanish. I lat er translated the most relevant parts and some quotes to English for their inclusion in this document . I analyzed the qualitative information using the iter a tive and continuous system proposed by Miles, Hub erman and Saldaña (2014) . First, I devel oped a cod ing book (see appendix A) that containe d labels , definitions, rules, and examples for th e classification of information . The coding book was created based on the literature of empowerment , and . I classified the information about empowerment under the themes: legislati ve power, executive power, and judicial power. Fo r the data about forest management institutions, I used each of lso contains some topics that emerged during the interviews such as the mo bilization of 2011 . Once I had a first draft of the coding book, I used it on several interviews, and I modified it accordingly. Then, t wo researchers external but Spanish native s peakers applied the cod ing book to a sample of interview s to verify its co nsistency. After that process, the coding book wa s modified and finalized. Then , I proceeded 28 to code all transcripts. As I was reading the transcripts, I also highlighted other infor mation I consider ed important , but that did not fall into any of the preco nceived themes of the coding book. Then, I transf erred all the information under the same code into a single document, and I summarized the memos individually. Finally, I organized all the summaries in a matrix containing two horizontal cells for each them e that corresponded to the situation before and a fter the mobilization of 2011 in Cheran . 4.2 Land cover change assessment In order to assess land cover change in the forest of Cheran, I assembled a Time Series of classified Land Sat 7 image composites covering the years 2005 to 2020 in 3 - year interv als. I used Being a mountainous and rainy area, imagery from Cheran is often obstructed by clouds; therefor e, I used images obtained during the dry se ason ( October - March). The classifications can be performed effectively with this imagery because healthy evergreen forest maintains a high NDVI value throughout the year while the values for grassland and agricult ure land drop during the dry season making it eas ier to differentiate classes. I used GEE to perform supervised classifications of Landsat 7 Tier 1 top of the atmosphere Reflectance (TOA) imagery composites. The TOA product is preferred over the TOA radia nce because it contains a quality mask whic h prov ides information on a per pixel basis that describes what is contained in the pixel (e.g. cloud, cloud shadow, snow, ice, etc). The GEE algorithm creates a cloud - free stack with all the Landsat 7 available i magery for the months January - March of the studie d year. I performed the analysis in three - year intervals to omit short term land cover changes caused by factors different from forest governance such as rain level. A particularly rainy year for example, ca n show a higher forest land cover than a dr y one. The resulting time series consists of 29 6 3 - month composites. Table 7 shows the images GEE used to create the composites of each year of the study. A total of 30 Landsat 7 images were used for the analysis. Table 7 : Landsat Imagery The three - month image composit es were created by flattening the reflectance and NDVI values of the image collections. The median reflectance value for all the optical bands for each pixel was extracted and the maximum NDVI value was appe nded to the reflect ance composite as a band. The resulting image contains the median reflectance and the maximum NDVI value for Year (Number of Images) Stacked Images Acqu isitio n Date Satellite Path Row 2005 (4) LE07_028046_20050112 2005 - 01 - 12 L7 028 046 LE07_028046_20050128 2005 - 01 - 28 L7 028 046 LE07_028046_20050213 2005 - 02 - 13 L7 028 046 LE07_028046_20050317 2005 - 03 - 17 L 7 028 046 2008 (6) LE07_028046_20080105 2008 - 01 - 05 L7 028 046 LE07_028046_20080121 2008 - 01 - 21 L7 028 046 LE07_028046_20080206 2008 - 02 - 06 L7 028 046 LE07_028046_20080222 2008 - 02 - 22 L7 028 046 LE07_028046_20080309 2008 - 03 - 09 L7 028 046 LE07_0 28046_20080105 2008 - 01 - 05 L7 028 046 2011 (4) L E07_028046_20110113 2011 - 01 - 13 L7 028 046 LE07_028046_20110214 2011 - 02 - 14 L7 028 046 LE07_028046_20110302 2011 - 03 - 02 L7 028 046 LE07_028046_20110318 2011 - 03 - 18 L7 028 046 2014 (5) LE07_028046_2014 0105 2014 - 01 - 05 L7 028 046 LE07_028046_2014012 1 2014 - 01 - 21 L7 028 046 LE07_028046_20140206 2014 - 02 - 06 L7 028 046 LE07_028046_20140222 2014 - 02 - 22 L7 028 046 LE07_028046_20140326 2014 - 03 - 26 L7 028 046 2017 (5) LE07_028046_20170113 2017 - 01 - 13 L 7 028 046 LE07_028046_20170129 2017 - 01 - 29 L7 0 28 046 LE07_028046_20170214 2017 - 02 - 14 L7 028 046 LE07_028046_20170302 2017 - 03 - 02 L7 028 046 LE07_028046_20170318 2017 - 03 - 18 L7 028 046 2020 (6) LE07_028046_20200106 2020 - 01 - 06 L7 028 046 LE07_0 28046_20200122 2020 - 01 - 22 L7 028 046 LE07_0280 46_20200207 2020 - 02 - 07 L7 028 046 LE07_028046_20200223 2020 - 02 - 23 L7 028 046 LE07_028046_20200310 2020 - 03 - 10 L7 028 046 LE07_028046_20200326 2020 - 03 - 26 L7 028 046 30 each pixel for a three - month period. This arrangement provides extended information for the classification of the images. Agri culture land is sub jected to sudden changes in ve getation depending on human activity and environmental factors such as rain. For example, during an unusually rainy year, farmers can plant short cycle crops to utilize remaining moisture in the soil; the re sulting healthy veg etation has high NDVI values a nd can be erroneously classified as healthy forest by the algorithm. Therefore, the composites of Landsat Imagery were clipped to exclude the agriculture land and the urban area of Cheran which reduced noise . Then the clipped composites were classified usi ng the random forest classification method within GEE. This method requires training polygons for each class; in this case the classes were: 1) Bare soil, 2) Transition forest and c) Healthy forest. I perfor med some tests wher e the class Grassland was incl uded but it was difficult for the algorithm to distinguish bare soil from grassland during the dry season because when grassland has low NDVI values. The algorithm also requires validation polygons for each class for the calcu lation of accuracy parameters. The training and validation polygons were drawn using Google Earth´s high definition imagery archive to find areas whose class remained constant throughout the duration of the study. Figure 4 shows the stud y area with the pol ygon that the community claims , the excluded agriculture land and urban area, and the class polygons. 31 Figure 4 : Study area and training polygons The algorithm reports accuracy via a confusion matrix con taining the classification of the pixels within e ach validation polygon. Ideally, 100% of the pixels inside a validation polygon should fall into the polygons class. The overall accuracy is given by the kappa coefficient which is calculated as described by Foody (200 2) . The GEE algorithm was also programed to export the classified image and display the area of each class area in square kilometers. The algorithm w as applied 6 times with a three - year gap between each mea surement and the outputs were recorded to assembl e a Timeline Series. 5. RESULTS This section is organized as follows: first, I will present the results on the origins of the crisis and the mobiliz ation in Cheran in order to provide context to the follow ing sections; second, I will explain to what exte nt the community won authority over the forest resources. Then I will 32 explain how the management institutions changed after 2011, and finally, I will address the change in forest land cover. 5.1 Origins of the crisis and mobilization In the state of Mic hoacán commercial avocado plantations are historically very profitable; and most of the production is sent to the international markets. Therefore, landowners have monetary incentives for the establishment o f avocado plantations. As a r esult, large areas of forest were converted into avocado plantations, however, Cheran did not follow that trend. Avocado plants are b anned from the community since 2011. As the forested area has decreased in the State the sup ply of timber has shrunk, and buyers have incenti ves to buy wood in unconventional places like Cheran that was never a commercial producer of timber. In Cheran, historically, resin has been the most important commercial forest product; however, people als o extracted timber to make fur niture. The claiman ts were responsible for managing wood extraction within their assigned land. For the most part, only sick or old trees that did not produce enough resin were cut for timber. In the community, there is a comm only owned processing plant, a nd several carpentr y shops, together they covered the timber supply of the community. Given the increased demand for timber in the mid - 2000s, external loggers started buying trees in Cheran even though the Environment and N atural Resources Bureau (SEMAR NAT by its Spanish acronym) did not officially allow the commercial extraction of wood. SEMARNAT is the nationwide branch of the government that is in charge of regulat ing forests use. The executive body of the forest user gro up of Cheran called the Counci l of Common Goods ( CBC by its Spanish acronym) and the Municipal Government allowed the commercial extraction of timber. 33 Furthermore, according to some Cheran inhabitants, the municipal government benefited politically and com mercially from that illegal lo gging as described below: government very badly, they let the loggers grow and they used intimidation to Mr. Jaime During this s that destabilized the whole country as the cartels started fighting one another for the control of the territories. In the state of Michoacán, the police forces where ove rwhelmed by t he criminal organizations. The violence and corru ption permeated all sectors of society and the drug cartels diversified their criminal activities to include kidnaping, extortion and, illegal logging. The groups that initially purchased wood i llegally in C heran colluded with criminal organizations and st arted stealing and cutting trees. were selling wood but they paid only once. When the loggers (from outside) knew their way in, they started going into the forest [to cut wood]and t hey did not pay; they had big - ass guns in their hands! Do you think [the people] could Mr. Javier, Forest Authority from 2008 to 2011 According to the interviews, the CBC a ttempted to s top the deforestation, but they lacked enough peo ple and/or arms to effectively fight the external loggers. In addition, they did not have judicial authority; all they could do was to denounce the offenders to the State authorities. Some of th e interviewee s mentioned that in the rare occasions that the S A MARNAT caught loggers, they were released short ly after by paying a fine. Then, the members of the CBC became the target of imants decide d to sell their trees at low er price s before they were taken away violently. Under such circumstances of weakened institutions, Cheran lost approximately a third of its forest. 34 icle, they sh we needed backup from the State police [Because the loggers are heavily armed], we needed a hand Mr. Javier Before this, community members not belo nging to the involved in fo restry matters; however, they started suffering because of the violence and they became increasingly concerned about predatory deforestation. The situation became highly tense when the deforeste d area almost reached a water spring that provides a significa nt portion of the supply and such fear triggered the mobilization, as one of the leaders of the mobilization states: portion of the township water and when [the deforestation ]reached that point is when people said we will be left without water - Mr. Rafael, leader of the mobilizat ion of 2011 At the beginning of the armed resistance, during the first days of April 2011, a group of women confronted the driver of a truck that transported logs extracted illegally, the women demanded the intruders to stop extracting wood from the fore st, they rece ived insults for an answer. But the group did not give up and started encouraging their neighbors to act against the loggers. The neighbors supported those brave women and soon the whole community joined the violent resistance, detaining intr uders and bur ning their vehicles. The people used whatever res ources they had available to defend their territory, including rocks, machetes, fireworks and some firearms. In the following days, they were able to control the two main points of access to the community wi th barricades, but the community became isolated to the outside world . Meanwhile, the community received support (food, money, and arms among other things) from neighbor ing communities, as well as from people from Cheran living in the US. 35 Tha t same year, after achieving the security of their territory, the community members of Cheran refused to participate in the State elections and sued the Federal Electoral Institute (IFE) for their rights to elect their own authorities according to their tr aditional ind igenous customs. The lawsuit agains t the IFE was resolved by the State Electoral Tribunal which is the government branch in charge of settling election - related legal disputes in the State of Michoacan; they ruled against Cheran. Then the commu nity took the case to the Federal Electoral Tribunal which rul ed only as s supervisor of the process. The new governance system is more horizontal as they elect 12 people (four from each neighborhood) to perform the task s of a municipal president. The elections are a public ceremony in which supporters of the different candidates form a line behin d their candidate of choice and the four candidates with most supporters in li ne are elected within each neighborhood. The 12 e lected people serve for three years; at the time this research was conducted, the second period under this governance system was ongoing. Forestry management was at the core of the mobilization and contin ues to be an important issue for the communal ass embly. Despite many difficulties, the community has managed to undertake reforestation projects, they also organized an armed gro up for the protection of the forest. Furthermore, several communal enterprises have been created not only for forestry - related activities, but also for all kinds of business aim ing to bring money to the local economy. each activity, we make sure that everything is accomplished. The community has realized that we must work collectively instead of individually. The activities are efficient and Mr. Jaime 5.2 Forest empowerment As mentioned earlier, the new g overnance structure is more horizontal, and authors describe it as participato ry and decentralized, and where accountability is a priority of the new 36 administration (España - Boquera & Campo - Jim énez, 2016; Rojas, 2012; Ventura Patiño, 2012) . I was interested, however, in investigating the empowerment process in relat ion to the natural resources, more specifically, the forest. atural resources since 2011, particularly in terms of power to enforce rules. The mobilization in Cheran was a response to the lack of accountability and violence; therefore, rule compliance is a priority of the new government. Before 2011, the power to en and courts which, accor ding to the people of Cheran, were deficient. In 2011, the community created their own police force that in 2016 consists of 120 armed subjects that patrol both the forest and t he urban area. The police monitor the extraction of firewood and wood, but als o ensure the security of the territory. Offenders are referred to the department of Justice of the M unicipal government. 4 never engaged in their duties; it was their Mr. Javier nning of the movement, an [armed] group was creat ed for security gro Mr. Jaime The community of Ch e ran shares the power to create laws with the S tate and the Federal government. The local author ities create local laws that apply to Cheran, but forest use is also subject to federal Laws. For example, wood extraction projects must be submitted to CONAFOR for approval. In contrast, most of the power to enforce laws is decentralized, which means tha t the community is in charge of enforcing the law and punishing criminals. In Mexico there are two 4 PROFEPA ( Procuraduria Federal para la Proteccion del Ambiente) the judicial branch of the government in charge of enforcing environmental law nation wide 37 types of police groups, one belongs to the judicial branch of government and one belongs to the executive b ranch. The state of Michoacán unified all the pol ice forces of the State to increase accountability. However, Cheran refused to join that initiative and maintained an autonomous police force. The local governme nt of Cheran has a branch that sanction s law o ffenders and it consist s of a judge and a group o f rangers. The CBC also sanctions those who do not follow the forest rules. The community is so confident on their judicial system that they do not collaborate w ith the national army, the State judicial poli ce, the State courts, or the Federal Attorney of Environmental Protection (PROFEPA). Before 2011, the creation of rules for the forest management or legislative power resided in the PROFEPA. After the establ ishment of the autonomous government in Cheran , the general assembly of neighbors gained signif icant legislative power. The following quotes show that apart from resin and it was approved [by SEMARNAT], but the community, th e Mr. Roman, CBC st uff member Ms. Jose fina Forest Local Authority According to the interviews, all the legislation is democraticall y decided by the people of Cheran through the assembly and the execution of forest - related decisions is the duty of the Council of Common Goods (CBC). Before 2011, the executive power resided in the CBC a nd currently the Council is still in charge of ex ecuting the decisions that people make in relation to forest management. The Council authorize s the extraction of firewood through individual paid permits; they give permission to take wood from fallen and sick trees; and they administrate the reforestati on programs. When it comes to make more important decisions like large - scale wood extraction, the CBC must get the approval 38 of the municipal and state authoritie s. But the structure of the CBC radically chan ged since 2011, it became more horizontal and it is more integrated to the municipal government and the General Assembly of Neighbors (GAN). Although the CBC is independent from the municipal government, it is highly accountable to it and to the GAN. [the CBC] are in charge of designing projects an d we only approve Mr. Rafael community decided to have a representative from each neighborhood, each representative has the sa Mr. Jaime T able 8 presents the types of power (legislative, executive, and judicial) and summarizes how the community gain power in each of these aspects by comparing what was happening before and after the mobilizatio n of 2011. I symbolize a visible gain of power wi gain of pow From Table 8 we can conclude that the Community of Cheran gained overall power in relation to the forest, especially in terms of power to enforce law and exec ute decisions. Table 8 : Change in power (Empowerment) after the mobilization of 2011 Type of power Functions Actors Community empowerment Comments Before 2011 After 2011 Power to create laws Create laws that regulate the ext raction of resources at large scale for profit SE MAR N AT (De jure) SEMARNAT, GAN + The S E MARNAT continues to be the highest authority in relation to natural resources and environment law, however, Cheran has dramatically increased its bargaining power. GAN can block wood extraction even if SEMARNAT author izes it. Create appropriation and provision rules for the extraction of resources at domestic scale CBC (De jure) GAN + Before 2011 the CBC determined the rules for the extraction of firewood and resin fr om the common property. Nowadays the general asse mbly votes such rules 39 Table 8 (cont´d) Power to make and execute decisions Manage the extraction of fallen and sick trees No one CBC + This activity was approved by SEMARNAT and the GAN but the CBC executes it Manage the extraction of firewood No one CB C + The AG granted the CBC faculties to extend permits for the extraction of firewood Create reforestation programs No one CBC + The AG instructed the CBC to reforest the damaged areas. Resource monitor ing SEMARNAT (De jure) CBC ++ Now days the CBC is in char ge of monitoring the condition of the forest Design and execution of maintenance plans (Fencing, road maintenance, etc) CBC (De jure) CBC + Monitoring of users and boundaries State police (De jure) Cheran Rangers ++ Before 2011 the boundarie s were very loose. Now days the community has full control of the boundaries. Power to enforce laws Sanctioning offenders SEMARNAT (De jure) CBC, Cheran Department of Justice ++ Minor infractions are sanc tioned by the CBC whereas recurrent and serious i nfractions are remitted to the Cheran D epartment of Justice Conflict resolution CBC, State courts CBC, Cheran Department of Justice ++ The CBC settles disputes among forest user s and claimants; the dispu tes are usually about land. In the past , complica ted cases were taken to the state courts and notaries; nowadays the CBC settles all disputes with the help of legal advisors. 5.3 Forest management institutions Resource and Users Boundaries Perhaps, th e most pressing demand of Cheran inhabitants that mobilized in 2011 was the establishment of clear boundaries for both the forest and legitimate users; therefore, it is not surprising that this subject showed a significant improvement after the mobilizatio n. In 2011, the t well marked in the terrain. Not only the fences were obsolete, but the intruder s destroyed many of them to facilitate their illegal activities. Thus, o ne of the first projects that the new CBC undertook wa s the rehabilitation of the barriers. They review ed original 40 documents to determine the limits and hired local people to build fen ces. In the areas of the forest where intruders kept logging despite the barb wire fences, they dug trenches in order to preve nt trucks from trespassing, making it more diffic ult for outsiders to cut trees. Cheran has conflicts with three other communities in relation to the location of limits for an approximate total of 100 hectares. cess, the first thing we did was cleaning the bor der [external limits of the community] Mr. Jaime The de jure owner of the forest in Cheran is the whole community; however, the de facto owners before 2011 were the user group which is comprised of approxim ately 1200 people. In general, people outside the forest user group did not feel entitled to the forest and its resources. That ch anged during the 2011 mobilization in which the whole community participated in the armed resistance. Thus, after the mobiliza tion, the community has shown more interest in th e forest. The forest group of Cheran is integrated by claimants or posesionarios ; they have the right to extract resin from a n individual plot , but they ca nnot extract timber nor sell the land. The cl aimants most of the time inherit their rights to use the land; however, other members of the community can ask for the righ ts to a plot of land that no one is currently using, if the community approves the ask , they are given a responsibility to take care of the plot. Although the forest user group of Ch eran continues to be the claimants, the entire community exercises now de facto ownership of the forest. Mr. Jaime In ad dition to fencing the external boundaries of the forest, the internal divisions of the forest were marked with fences as well . Approximately 90% of the forest is claimed by members 41 of the community , the rem a ining 10% of the forest is commonly owned and no one has the right to withdraw resin from it . The tenure claimers hold entitles them to the exclusive right to manage the production and withdraw al of resin from the trees . H owever, the community as a whole retains the ownership of th e forest as well as the right to make de cisions about logging and change s in soil us e . The CBC extends certificates of tenure to the claimants when they require it; such certificate specifies that despite being granted the rights related to resin extraction, the ultimate ownersh ip of the land st ays with the community. Claimant s have the right to extract resin from a defined plot; however, t he individual plots within the forest are not usually fence d by barb ed wire because claimants verbally agree on the limits. However, when a re forestation progr am is implemented in an individu al plot, fences are constructed to keep livestock from entering the plots and prevent saplings to be harmed. a c knowledge that t rs, they Mr . Roman , member of the CBC st a ff in 2016 After 2011 , in Cheran all people from the community are considered forest u sers as they can withdraw firewood with a permit from the CBC. They can also collect herbs and mushrooms freely in all the forest. Community members can also have livestock ranging in the for est. Therefore, rangers allow the free transit of members of the community in the forest. However, the community patrols the forest looking for people that withdraw firewood without a permit or carry green wood. The rangers are e specially concerned with ou tsiders extracti ng anything from the forest or even trespassing i t . Even though the population of Cheran is numerous, the rangers are very confident that they can recognize members of the community f r om outsiders very effectively. 42 After 2011 there was an important enhanc ement of the user boundaries. Before the mobiliza tion, virtuall y everybody could enter the forest and extract as much of the resources as possible. At the worst point of the situation, loggers did not even hide their activities; they transported stolen woo d through the township during the day . The CBC tr ied to stop il legal logging, but they lacked the resources and the personnel to make an impact; the president and his assistants themselves oversaw patrolling. Even when they caught criminals, there were no consequences because they lacked the authority to incarcerate t he offenders and the State authorities kept ignoring the problem. There is also the presumption that local authorities were involved in the illegal logging either by complicity or by intimidati on. everywhere to extract wood, they did not respect the fences, they were all t orn Mr. Jaime At the core of the mobilization was the demand of stopping intruders from logging in Cheran. At the beginning of the conflict, a community guard group was set in place to de fend the territory from outsiders, with whom the community guard group had several violent encounters. The community guard survives to date and serves as police and forest rangers. Th ese armed guard s obey only local authorities and they patro l the perimeter of the forest and the urban area. group was created; they are part of the community and their duty is to mo nitor MR. Roman Thanks to the sec ured boundaries and the permanent surveillance, the number of intruders has dropped dramatically since 2011. In 2015 only isolated cases of illegal logging in Cheran´s forest by outsiders were registered. In ternal monitoring for the compliance of usage rul es is now the main activity of the forest rangers. Appropriation and provision rules 43 Appropriation rules were in place before 2011; however, they were improved after the empowerment process. Table 9 conta ins a description of the appropriation rules for each forest product before and after 2011. The National Forestry Commission (CONAFOR for their Spanish acro nym) grants extraction licenses for commercial purposes under a supervised forestry plan. Commercial extraction has not been allowed by the CONAFOR i n Cheran since the 80s. In 2012, the CBC got an authorization to extract fallen and ill trees to clean the forest; the wood was turned into boards and sold locally. No one besides the CBC is authorized to ex tract lumber in Cheran. Table 9 : Appropriation rules Product Rules before 2011 Enforcer (Compliance) After 2011 Enforcer (Complianc e) Lumber CONAFOR did not allow anybody to extract lumber. SEMARNAT (No) CONAFOR authorized the ex traction of ill and fallen trees; this activity i s carried out exclusively by the CBC. CBC, Cheran rangers (Yes) Firewood There were no rules for the extraction of firewood N/A Cheran inhabitants are granted permission for the ex traction of a certain amou nt of firewood in a weekly basis; they must pay a fee. CBC, Cheran rangers (Yes) Resin Only claimants can extract resin in their plots. Each claimant (Yes) Only claimants can extract resin in their plots. Each claimant, CBC, Cheran rangers (Yes) Mushroom s and herbs No rules N/A No rules N/A Grass (For livestock) Livestock can range freely in all the forest N/A Livestock can range freely in all the forest . Each claimant is responsible for the fencing of his plot to avoid dama ges to the young trees . Theref ore, fences are installed when the trees are youn g. CBC (Yes) 44 Pieces of wood under 35 cm in length are considered firewood. After 2011, a set of rules for firewood extraction were established. All inhabitants of Cheran over the age of 18 have the righ t to extract firewood from any place in fallen tr ees in the forest if they pay their permit . The CBC grants permit to extract firewood at different prices and frequencies, depending on the transportation method as shown in Table 10 . Table 10 : Firewood extraction rules Transport C ost Frequency of extraction Pick - up truck loaded with firewood. 15 pesos Twice per week Wagon (small animal - traction car that carries approximately half of the load of a pick - up truck) 10 pesos 3 times per week Equines (Horses and/or donkeys that carry firewood in their backs) 10 pesos 4 times per week Resin extraction has a long tradition in Cheran; however, this activity was in decay before 2011 due to the high deforestation rate . Thanks to improved s afety , a healthier forest and a rise in prices of the product, resin extraction became more profitable and claimants resumed the extraction activity. Only claimants are authorized to extract resin in their plots, and they are in charge of the management. A ll the resin is collected in a commonly owned ent erprise that separates the liquid and solid parts and sells both to the industry. The enterprise is owned by 100 forest users and administrated separately from the CBC. ducts and to work in the forest, if it is agricul ture land they can put crops, if it is forest they can extract resin but they cannot make any other kind of extraction like trees, sand or gravel, that requires the appro Mr. Roman Besides the main forest products (lu mber, firewood and resin), mushrooms and herbs are extracted from the forest; however, there are no appropriation rules for these goods, meaning that 45 anyone can extract them from the forest and as much as th ey wan t. Herbs and mushrooms are col lected by Ch eran inhabitants both for self - consumption and for commercialization. u cannot [extract], each claimant wor ks, makes gr ooves [ on the trees for resin extraction] and extracts [resin] Ms. Josefina, Member of the CBC stuff Provision rules refer to the way each common owner contributes to the group and the management of the forest . Before 201 1, reforesta tion and fencing activities were the responsibility of each claimant, while the CBC oversaw fixing the roads that allow transportation within the forest . Traditionally, all claimants are obligated to help st opping fires in the forest in case these occur . S ince 2011, the CBC oversees the maintenance of the forest; they manage the reforestation, fencing and maintenance of roads. The CBC obtains money from selling wood, construction materials from the mines 5 and trees from the plant nur sery; additional to reso urces that come from firewood permits and federal grants 6 . With that money the CBC pays for the wages of its members and provides employment to around 200 people from the community to work in reforestation a nd fencing programs. Col lective - Choice arrangemen ts As explained in the section on empowerment, the community gained a lot of authority over the governance of their natural resources after 2011; many of the decisions are now made by the 5 Besides timber, resin, herbs, mushrooms and cattle food, the forest of Cheran also produces construction materials from digging sites or mines. The CBC is in charge of the administration of such mines. 6 Mexico ha s several subsidy programs for the rural s ector. Initially, Cheran had difficulties accessing to such subsidies; however, thanks to a very good internal organization they now get federal resources for reforestation campaigns, soil conservation programs and environmental services payments. 46 community or they are negotiated with the National government. Insi de the community, decision making with regards to forest decisions has always been democratic in the sense that the general assembly of forest users is the highest local authority ; however, before 2011 the l ocal authorities had lit tle to no de facto power. After 2011, t he democratic mechanisms for decision making were strengthen ed and the general assembly of forest users gained more power. This assembly makes important decisions and elects an operative team (CBC) via elections . After 2011, the governing sy stem became more horizontal as it was described before . Instead of choosing a single president with its secretary and treasurer, the comm unity elects 12 people to conform the CBC with equal representation fr om each of the four neighborhoods of Cheran. All members of the CBC share equal authority although there is one person that serves as president for legal and administrative purposes. All bureaucrats in Cheran are required to come from the roots of the 2011 mobilization; they are elected by and hold accou regularly scheduled evaluation meetings. The appropriation and provision rules, sanctions and decisions about the reforestation activities are all submitted to vote in the general assembly. cided to have representatives from all neighborhoods. one pr al Mr. Tomás , CBC membe r Monitoring of the resource and the users Before the mobilization of 2011, no one was monitoring the condition of the forest even thou gh the law says that the SEMARN AT oversaw this activity. Estimations o f the total area deforested were made by the comm unity once they recovered control of the territory in 2011; the new CBC has personnel to monitor the forest and quantify the areas that needed to be reforested. 47 Once the communi ty recovered full control of the forest, they started to try to restore the da maged land. Thanks to their organization they were able to negotiate with CONAFOR, SEMARNAT and other governmental organizations to obtain federal funding for their reforestation programs. The community had to negotiate for more than one year before their rights were recognized, but nowadays they have a close relationship with the federal forest organizations. Applying for new funding for reforestation requires the compliance of a minimum of survival rate in the new plants; therefore, both the community and the federal government closely monitor this indicator. survival rate. They come and verify everything and based on that they renew the Mr. Roman In 2013, the CONAFOR surve illed the forest and determined that it was possible to implement a forest extraction plan without compromising the integrity of the forest; logging was planned for areas that we re not affected by the defore station during the 2005 - 2011 crisis. In case of i mplementing the plan, CONAFOR would have intensified the monitoring. However, the general assembly of Cheran municipality voted against the implementation of the plan and decided to focus only o n reforestati on. and non - timber extraction plan (on top of the resin) and it was authorized; but the community, the assembly Mr. Roman Users m onitoring was greatly enhanced af ter the empowerment of Cheran squad was created mainly to prevent outsiders from entering the forest, but nowadays they are also in charge of monitoring compliance of rules by claimants . The territory is constantly patrolled t o identify and arrest offende rs. People carrying firewood are required to show a valid 48 permit. No one besides the CBC personnel can cut trees of any type or transport pieces of wood bigger than 35 cm in length. Monitoring the extraction of pine and oak is very effective because these woods are valuable in big pieces and therefore ea sy to spot. More precious woods, on the other hand, a re easily hidden form the rangers, which makes them more vulnerable to logging for commercial purposes. Offenders are arrested and sent to the CBC, where they are given the corresponding sanction. Gradua ted Sanctions Before 2011, minor sanctions were appl ied by local authorities, while loggers punishment was a responsibility of State authorities. According to the testimonials of forest users, illegal logg ers were not held accountable by the SEMARNAT. Wi th the new governing structure, the community took co ntrol over the sanctioning and judicial system . Nowadays, if somebody is found extracting wood without a permit for the first time, the forest rangers co nfiscate the paying a fine of up to $100 USD to the CBC, the tools and the vehicle can be recovered. The second time someone is caught violating forest rules, firewood permits are denied for one month , in addition to a $ 100 USD fine. After a third time, people are sent to the department of justice, which is the judicial branch of the municipal government in Cheran. In such cases, the department of justice determines a more severe sanction such as jail time or permanent co nfiscation of the tools and vehicle. ngers, they confiscate the tools of the intruder and we write a report. We determine a sanction depending on the magnitude of the offence: the greener the wood the bigger the fine. If the logger has a record , the fine is bigger. The sanctions go form $5 to $100. When they pay they get their tools Mr. T omas 49 Conflict Resolution Mechanisms According to former authorities, adjudication of disputes was happening efficaciously before 2011; however, this tre nd has seen improvement since the mobilization. P reviously, the officials relayed in their own experie nce and local knowledge to settle disagreements between forest users and they had some success. After 2011, however, the CBC hired a specialized lawyer to assist with all legal issues . This lawyer interp rets the agrarian reform law for them. Most conflicts relate to property rights; although claimants do not own the land, they have disputes over the usufruct rights of forest plots. Sometimes, two different people claim to have the property rights over the same plot; the CBC reviews the existing documentatio n; determines who is the claimant and releases new documentation. Sometimes, the claimant of a plot a bandons it for some years; when he comes back som eone else is using it. We have the authority to solve this with the advice of a representative Mr. Tomas The community does not have written rules for settlement of disp utes. The CBC deliberates based on interpreting t he existing legal documents with the advice of t heir lawyer. The interpretation and validation of legal documents is important because in the past several legal entities had issued property rights documents to forest users. The old CBC sometimes settled di sputes and gave legal documents to claimants wit h no solid legal basis. Similarly, the State notaries had issued documents that entitled forest users to full property rights. Currently, the CBC issues proper ty rights documents that override the old ones an d the ones from the State notaries. In the new l egal documents, it is clearly stated that the ultimate owner of the land is the community and claimants only have usufruct rights over the land. 50 the plot] using a GPS, then we make a document wi th a map in which we clearly state that the comm Mr. Roman Recognition of Rights to Organize All over Mexico , Ejidos and comunidades have always had rights to organize to mana ge their natural resources. In Cheran, b efore 201 1, the forest user group elected their authorities according to the rules established by the agrarian reform bureau, the instance that deal s with legal issues related to land property. The law establishes th at all user groups must have a president, a secre tariat and a treasurer, as well as a mo nitoring body composed of three people. After 2011, the community of Cheran decided to substitute the governing structure with a more horizontal one as described earlie r . Since the new structure is not exactly the sa me as the one specified by the national law, the new CBC struggle d to gain recognition by higher levels of government. [single ] representative but not a council so we cannot r cannot give you a d Mr. Rafael funded programs for ref orestation and development. After a year of litig ation, the CBC was finally recognized, but every time the community elects new authorities, the general assembly must get together and ratify its validity under the supervision of the A grarian Re form B ureau. Currently, most organizations recognize the CBC ore, they have access to diverse federal programs. situation and finally being a legal f Mr. Tomas 51 Nested Enterprises Accordin g to Ostrom (1990) appropriation, provi sion, monitoring, enforcement, conflict resolution, and governance activities must be organized in multiple layers and be consistent among them. Before 2011, all instit utions were in theory fully nested as the forest user group was under the control of the municipal administration. Similarly, the municipal government was subordinated to the State and federal governments. After 2011, when Cheran established their autonomo us municipal government, the nesting of organizat ions went through several change s. Within the community , the organizations became more integrated after 2011 as the forest user group authorities work closely with the municipal authorities; both are elected through identical mechanisms by the general asse mbly of neighbors. The CBC manag es the forest and common enterprises (plant nursery, mines, and sawmill) except the resin processing plant which is run by a hired manager. The resin processing plant was func tioning well before the mobilization of 2011 and therefore remained outside the n ew governing structure. The nesting of municipal institutions within higher levels of government decreased after 2011. The municipality became independent from many State and federal instances, such as the federal electoral institute and the police. Chera with the federal army nor the State police. The State notaries do not coordinate with the CBC and even in some cases, they write contradic tory legal documents in relation to the property rights of the land. The lack of coordination often leads to conflicts. For example, one plot of land can have two separate sets of documentation that give property rights to different people leading to dispu tes among forest users. Initially, the federal e ntinites that dealt with forest matters such as CONAFOR and SEMARNAT did not cooperate with the community. However, thanks to the organizational 52 capacity that Cheran has shown, the community now has a close relationship with the federal government when it comes to forestry. Cheran is now portrayed as a model of sustainable community forestry system by the same federal organizations that refused to help the community during the deforestation crisis. Figure 5 shows a map of the many entities involved in the governance of Cheran and how they relate to each other. Inside Cheran municipality, all entities are fully nested and work with each other. In contrast, Cheran is not fully coordinated or nested with the fed eral and state government. In some respects, such as health administration, Cheran heavily depends on higher levels of government. In other aspects, local authorities do not coordinate with their federal counterparts; a good example of this is the armed fo rces. In figure 4 , two entities are fully nested when one is inside the other. Cheran is not fully nested within the federal government, therefore, the corresponding rectangles only overlap. Two entities are separated but perfectly coordinated when linked with a solid line; that is the case of the mill a nd the resin plant. A dashed line, however, means that two entities known about the existence of each other, but they do not coordinate at all. Such is the case of the federal armed forces and the rangers of Cheran; they recognize each other´s existence bu t do not communicate or coordinate at all. Figure 5 : Nesting of Enterprises 53 5.4 Land c over Change I applied the GEE algorithm to six different years between 2005 and 2020 and obta ined the accuracy indicators shown in Table 1 1 . T he values for validation Kappa ranged from 82.09% to 92.63% Table 11 : Accuracy parameters of the super vised classifications. Year Accuracy indicators 2005 Classification Data Bare So il Transition Forest User Accuracy Bare Soil 10 1 6 24 77.10% Transition 0 535 69 88.58% Forest 0 0 250 100.00% Producer Accuracy 100.0% 98.9% 72.9% Kappa: 82.09% 2008 Classification Data Bare Soil Transition Forest User Accuracy Bare Soil 101 6 24 77.10% Transition 0 535 69 88.58% Forest 0 0 250 100.00% Producer Accuracy 100.00% 98.89% 72.89% Kappa: 88.37% 2011 Classification Data Bare Soil Transition Forest User Accuracy Bare Soil 130 0 1 99.24% Transition 6 566 32 93.71% Forest 0 4 246 98.40% Producer Accuracy 95.59% 99.30% 88. 17% Kappa: 92.13% 2014 Classification Data Bare Soil Transition Forest User Accuracy Bare Soil 129 2 0 98.47% Transition 10 559 35 92.55% Forest 1 8 241 96.40% Producer Accuracy 92.14% 98.24% 87.32% Kappa: 92.63% 2017 Classification Data Bare Soi l Transition Forest User Accuracy Bare Soil 127 4 0 96.95% Transition 26 535 43 88.58% Forest 7 243 97.20% Producer Accuracy 83.01% 97.99% 84.97% Kappa: 85.64% 2020 Classification Data Bare Soil Tr ansition Forest User Accuracy Bare Soil 130 0 1 99.24% Transition 10 555 39 91.89% Forest 2 248 99.20% Producer Accuracy 92.86% 99.64% 86.11% Kappa: 90.56% 54 Table 1 2 and figure 6 show the area per class in square Km in the territory of Cheran ou tside the agriculture land and urban areas, from 2005 to 2020 in three - year intervals. Healthy forest reached its lowest point in 2014 and by 2 0 2 0 recovered to around 90% of its original ex tent . Bare Soil also reached its highest point in 2014. Table 12 : Land cover change Class 20 05 2008 2011 2014 2017 2020 Bare Soil 10.84 14.62 17.38 25.27 19.75 22.97 Transition Forest 24.50 25.54 32.02 39.22 35.14 23.58 Healthy Forest 140.60 135.78 126.53 111.46 121.05 129.39 Figure 6 : Land Cover Change within the Fo restry Areas of Cheran From Figure 6 we can learn that during the deforestation process, healthy forest was substituted by both bare soil and transition forest. However, during the most recent years, the m issing forest is mostly accounted by bare soil. T his might be the result of pla n ned cam paigns of timber extraction in which big plots of land are cleared but quickly recover to healthy forest. Despite the progress, healthy forest land cover is yet to come back to the level of 2005; 140.60 135.78 126.53 111.46 121.05 129.39 0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00 160.00 2005 2008 2011 2014 2017 2020 Square Km Year Bare Soil Transition Foerest Healthy Forest 55 have an explanation for that and a se cond visit to the field might be necessary to understand the processes behind it. Figure 7 shows the areas covered by bare soil, transition forest and healthy forest at different points in time. It is clear that big a reas of forest were substituted by bare soil and transition forest in 2011 and 2014. In 2020, there is an increased area of bare soil, however this are well defined regions which might mean that those areas were used for the planned extraction of timber. F igure 7 : Maps of l and cover by class 6. DISCUSSION AND CONCLUSION In the literature, empowerment is widely used and defined in several ways to the point that some authors conside r it a buzzword (Cornwall & Brock, 2005) . However, the case of Cheran demonstrates that empowerment is a phenomenon that can be observed and has implications for governanc e, environmental sustainability and the wellbeing of the community. In Cheran, r the management of natural 56 resources. After 2011, the forest user group gained power to create, execute and enf orce rules for the usage of the forest. Therefore , the case gave me the opportunity to empirically assess the process of empowerment and its pos itive consequences for the sustainability of the forest. This research work did not include a formal assessment of the empowerment´s consequences on social and e conomic aspects of sustainability; however, from my observations I can tell that the community was improved overall thanks to the process of empowerment. Therefore, future research on the topic of empowermen t, should look at its concrescences for sustainab ility in general. Mexico decentralized forest management in 1992; however, as shown in the cas e of Cheran, the policy implementation did not actually transfer powers to local communities. Before 2011, the f orest user group in Cheran held no judicial power and incomplete legislative power; their authority was basically limited to executive power. La rson & Soto (2008) argue that it is pr ecisely the lack of community empowerment t positive social and environmental outcomes. The case of Cheran presents solid evidence to support such argument because in many ways the lack of power by the com munity caused the deforestation crisis of 2005 - 2011. Because the community did not have judicia l power; they relied on the State authorities to hold illegal loggers accountable. However, according to the loc al authorities, the State judicial system did not sanction or incarcerate the intruders due to corruption. Certainly, other factors such as viol ence contributed to the deforestation of Cheran; however, the lack of local power played a critical role. Form this research we can conclude that decentralizati on policy not always includes an actual transfer of power to local communities which hinders it s potential to deliver positive environmental and social outcomes. In the literature, decentralization and empow erment are 57 sometimes defined similarly as both co ncepts are related to a change in power. I argue that decentralization is a top - down phenomenon while empowerment happens in a bottom - up fashion. Empowerment in Cheran happened because the community mobilize d to grab power and found from resistance from th e top. This antagonist relationship between empowerment and decentralization should be taken in to consideration in future research about the topics. Decentralization and community self - empowerment need to b e supported by the centralized power. Therefore, the success of self - empowered, community - based natural resource management depends on the governments. It is therefore a perfect example of polyce ntrism. But it does not need to be as difficult a s it was for the community of Cheran. Merino (2017) studied the process of pe ace building in Cheran and argues that even thoug h s ocial and com munity participation are indispensable for the defense, restoration and re - creation of diverse common (Merino & Cendejas, 2017, p. 924) . However, the particip ation of the community of Cheran became relevant only after the process of empowerment; therefore, I consider it essential to investigate t his process. According to Gibson et al (2000) , a fund amental condition for sustainable management of CPRs is that users must highly value the natural resource. This is co nsistent with what I observed in Cheran because a big sector of the local economy depends on the production of resin. Furthermore, the stak eholders reported that the mobilization was triggered by the possible destruction of their source of drinking water w hich is in the forest. Other communities of the s tate of Michoacán do not value the forest as much because they have substituted it with av ocado plantations. Nowadays , avocado plants are prohibited in Cheran but even before the mobilization 58 of 2011 no comm ercial plantations were e stablished in the territ ory , which gives us an idea of how much the people of Cheran value their forest. In addition to the high value that the forest represents for the community, high levels of social capital can explain the su ccess of the mobilization as theorized by some au thor s (Bray et al., 2003; Ellis & Porter - Bolland, 2008; Pérez, 2003) . Cheran has a strong indigenous identity and a history of subversive movements; additionally, most of the population lives in a single urban area which fa cili tates communication. I argue that all those characteristics of the community allowed them to coordinate their efforts in response to the obvious degradation of their forest; however, my research is focus ed in the consequences of the empowerment rather than its causes. The case of Cheran presents insights into the topic of institutional development which is us eful framework for the analysis of CPR institutio ns. I found that each of the principles was relevant for Cheran and the needs of the stakeholders. As noted by Agrawal (2001) , the first six of cal institutions whereas the last two focus on th e re lationships within the local context; that distinction is especially relevant in the case of Cheran. After 2011, es; on the other hand, the community struggled to est ablish robust relationships with higher levels of government. Now a days, Cheran is recognized as an example of sustainability in Mexico and the National Commission of Forestry is very supportive of the co mmunity. The recently elected President of Mexico , An dres Manuel Lopez Obrador is known for its leftist policy and willingness to support indigenous communities which most likely strengthened the relationship of Cheran with the federal government . 59 The a ssessment of land cover change reveals a scenario that matches the narrative of the people of Cheran. The area with healthy forest decreased drastically during the first half of the 2000s and reached its lowest point in 2014 when transition forest reached its high. Transition is being substituted by heal thy forest since 2014; therefore, in 2020, the area covered with healthy forest is at 90% of the 2005 level. The institutions that were established in 2011 have a clearly positive effect in the environmental sustainability of the forest in Cheran. It would be important to continue monitoring the institutions and the status of the forest to understand the effectiveness of this arrangement in the longer term. More so, to study other aspects of sustainability th at seem to be positive as well, as econom ic, and social sustainability. Violence in Mexico continues to be a problem which combined with big changes in the political scenario, could present important challenges to the current institutions of Cheran and tes t its endurance. GEE proved to be a powerful and flexible tool for the assessment of land cover change. Even though my assessment of land cover change was satisfactory for the purposes of this research, it can be improved by increasing the temporal resolu tion of the measur ements and the number of includ ed classes. The kappa values of the classifications are acceptable but not ideal. Increasing Kappa values can be achieved with specific training polygons for each year which is not a simple task given the la ck of high - resolut ion imagery for every year of t he study. Another factor affecting kapa values is the similarity between classes. The model can differentiate forest from bare soil very easily; however, it struggles to differe ntiate between healthy and tra nsition forest. After the community of Cheran m obilized to gain their autonomy, several neighbor ing communities have initiated similar processes. The relevance of Cheran as a case study is increased by this trend in self - empowerme nt movements. On the one hand, Cheran provides evidence suggesting that i ncreased local authority can result in more robust institutions and reduced 60 deforestation; in this sense, local self - empowerment movements are desirable. On the other hand, the case r eveals the deficiencies of natural resources policy in Mexico : even tho ug h the federal government finally recognized Cheran s traditional form of government, it was at a very high cost for the community to the point that they had to use violence to gain po wer . A more effective tr ansfer of power after forest management decentral ization could have prevented the violent difficulties to fully nest its institutions with in higher levels of govern ment suggests that natio nal policy is not polycentric enough to integrate local enterprises and this is a clear contradiction to the theoretical principles of decentralization and to their achievement . 61 LITERATURE CITED 62 LITERATURE CITED Agrawal, A. (2001). Common pr operty institutions and sustainable governance of resources. World Development , 29 (10), 1649 1672. https://doi.org/10.1016/S0305 - 750X(01)00063 - 8 Agrawal, A., & Gibson, C. C. (1999). Enchantment and disenchan tment: The role of community in natural resource conservation. World Development . https://doi.org/10.1016/S0305 - 750X(98)00161 - 2 Agrawal, A., & Ribot, J. C. (1999). Accountability in decentralization: a framework with South Asian an d West African cases. The Journal of Developing Areas , 33 (4), 473 502. htt ps://doi.org/10.2307/4192885 Alsop, R., & Heinsohn, N. (2005). Measuring empowerment in practice: Structuring analysis and framing indicators. World Bank Policy Research Working Paper 3510 , February , 123. ht tps://doi.org/10.1037/e596892012 - 001 Baland, J. - M ., & Platteau, J. - P. (1996). Halting Degradation of Natural Resources: Is There a Role for Rural Communities? In International Affairs Royal Institute of International Affairs 1944 (Vol. 72, Issue 4, p. 821) . Food and Agriculture Organization of the United Nations. https://doi.org/10. 2307/2624181 Barnes, G. (2009). The evolution and resilience of community - based land tenure in rural Mexico. Land Use Policy , 26 (2), 393 400. https://doi.org/10.1016/j.landusepol .2008.05.007 Barsimantov, J., Racelis, A., Barnes , G., & Digiano, M. (2010). Tenure, tourism and timber in Quintana Roo, Mexico: land tenure changes in forest ejidos after agrarian reforms. International Journal of the Commons , 4 (1), 293 318. Bradshaw, C. J. A., Giam, X., & Sodhi, N. S. (2010). Evaluatin g the relative environmental impact of countries. PLoS ONE , 5 (5). https://doi.org/10.1371/journal.pone.0010440 Bray, D. B., Merino - Pérez, L., & Barry, D. (2005). The community forests of Mexico: Managing for sustainable landscapes. In The Community Forests of Mexico: Managing for Sustainable Landscapes . https://doi.org/10.1659/0276 - 4741(2007)27[97:tcfomm]2.0.co;2 Bray, D. B., Merino - pérez, L., Negreros - castillo, P., Segura - warnholtz, G., Torres - rojo, J. M., & Vester, H. F . M. (2003). - Ma naged Forests as a Global Model for Sustainable Landscapes. Conservation Biology , 17 (3), 672 677. https://doi.org/10.1046/j.1523 - 1739.2003.01639.x Breiman, L. E. O. (2001). Random Forests . 5 32. Bremner, J., & Lu, F. (20 06). Common Property among Indigenou s Peoples of the Ecuadorian Amazon. Conservation and Society , 4 (4), 499 521. 63 Coleman, E. A., & Fleischman, F. D. (2012). Comparing forest decentralization and local institutional change in Bolivia, Kenya, Me xico, and Uga nda. World Development , 40 (4), 836 8 49. https://doi.org/10.1016/j.worlddev.2011.09.008 Coria, J., & Sterner, T. (2011). Natural Resource Management: Challenges and Policy Options. Annual Review of Resource Economics , 3 , 203 230. https://doi.or g/10.1146/ann urev - resource - 083110 - 120131 Cornwall Community Development Journal , 43 (3), 269 283. https://doi.org/10.1093/cdj/bsn010 Cornwall, A., & Brock, K. (2005). What do buzzwords do for developm ent policy: a critical look Third World Quarterly , 26 (7), 1043 1060. https://doi.org/10.1080/01436590500235603 Cortés Calderón, S. V. (2018). Security, Justice, and Forest Protection! An Indigenous eforestation and Organised Crime in Cherán, Mexico. In Green Crime in Mexico . https://doi.org/10.1007/978 - 3 - 319 - 75286 - 0_5 Cox, M. (2014). Understanding large social - ecological systems: Introducing the SESMAD project. Int ernational Journal of the Commons . C ox, M., Arnold, G., & Tomás, S. V. (2010). A review of design principles for community - based natural resource management. Ecology and Society , 15 (4), 38. De Janvry, A., Dutilly, C., Muñoz - Piña, C., & Sadoule t, E. (2001). Chapter 12: Liberal Reforms and Com munity Responses in Mexico. In Communities and Markets in Economic Development (pp. 1 19). Decision tree learning . (n.d.). https://en.wikipedia.org/wiki/D ecision_tree_learning Deininger, K. W., & Minten, B. (1999). Poverty, Policies, and Deforestation: The Case of Mexico. Economic Development and Cultural Change , 47 (2), 313 344. https://doi.org/10.1086/452403 Ellis, E. a, & Porter - Bolland, L. (2008). Is com munity - based forest management more effective than pr otected areas? Forest Ecology and Management , 256 (11), 1971 1983. https://doi.org/10.1016/j.foreco.2008.07.036 Ensminger, J. (1997). Changing Property Rights: Reconciling Formal and Informal Rights to La nd in Africa. In The Frontiers of the New Institution al Economics (pp. 165 196). España - Boquera, M. L. , & Campo - Jiménez, O. (2016). Proceso de deforestación en el municipio de Cherán, Michoacán, México (2006 - 2012). Madera y Bosques , 22 (1), 141 143. Faguet, J. - P. (2014). Decentralization and Governance. World Development , 53 , 2 13. https://doi.org/http://dx .doi.org/10.1016/j.worlddev.2013.01.002 64 FAO. (2010). Global Forest Resources Assessment 2010. FAO FORESTRY PAPER 163. In FAO FORESTRY PAPER . https://doi.o rg/ISBN 978 - 92 - 5 - 106654 - 6 Foody, G. M. (2002). Status of land cover classification accuracy assessment . In Remote Sensing of Environment . https://doi.org/10.1016/S0034 - 4257(01)00295 - 4 Gautam, A. P., & Shivakoti, G. P. (2005). Conditions for Successful Loca l Collective Action in Forestry: Some Evidence From t he Hills of Nepal. Society & Natural Resources , 1 8 (April), 153 171. https://doi.org/10.1080/08941920590894534 Gibson, C. C., McKean, M. A., & Ostrom, E. (2000). People and Forests: Communities, Instituti ons, and Governance . MIT Press. http://www.fao.org/do crep/005/ac694e/AC694E00.htm#TopOfPage Google Ear th Engine . (2020). https://earthengine.google.com/ Hardin, G. (1968). The tragedy of the commons. Science (New York, N.Y.) , 162 (3859), 1243 1248. https://doi.org/10.1126/science.162.3859.1243 Hennink, M., Ki iti, N., Pillinger, M., & Jayakaran, R. (2012). D efining empowerment: perspectives from international development organisations. Development in Practice , 22 (2), 202 215. https://doi.or g/10.1080/09614524.2012.640987 Herring, J. (2000). Measuring Vegetation (NDVI & EVI) . NASA Earth Observatory. https:/ /earthobservatory.nasa.gov/Features/MeasuringVegetation/measuring_vegetation_2.p hp Johnson - Freese, J., & Weeden, B. (2012). Application Governance of Common - Pool Resour ces to Near - Earth Orbit. In Global Policy (Vol. 3 , Issue 1, p. 72). Wiley Subscription Services, Inc. https://doi.org/10.1111/j.1758 - 5899.2011.00109.x Johnson, K. A., & Nelson, K. C. (2004). Common property and conservation: The potential for effective com munal forest management within a National Park in Mexico. Human Ecology , 32 (6), 703 733. https://doi.org/10.1007/s10745 - 004 - 6833 - z Kamran, M. A., & Shi vakoti, G. P. (2013). Design principles in tribal and settled areas spate irrigation management instituti ons in Punjab, Pakistan . In Asia Pacific Viewpoi nt (Vol. 54, Issue 2, pp. 206 217). Wiley Subscription Services, Inc . https://doi.org/10.1111/apv.12 020 Klooster, D. (1998). Institutional Choice, or a Process of Struggle? A Case Study of Forest Co - Manage ment in M xico . 28 (1). Knight, J. F., Lunetta, R. S., Ediriwickrema, J., & Khorram, S. (2007). Regional Scale Land Cover Characterization Using MODIS - N DVI 250 m Multi - Temporal Imagery: A Phenology - Based Approach. GIScience & Remote Sensing , 43 (1), 1 23. ht tps://doi .org/10.2747/1548 - 1603.43.1.1 65 Kolb, M., & Galicia, L. (2012). Challenging the linear forestation narrative in the Neo - tropic: Regional pattern s and processes of deforestation and regeneration in southern Mexico. Geographical Journal , 178 (2), 147 1 61. https ://doi.org/10.1111/j.1475 - 4959.2011.00431.x Lacroix, K., & Richards, G. (2015). An alternative policy evaluation of the British Columbia carbo common - pool resources . In Eco logy and Society (Vol. 20, Issue 2, p. 38). Resilience Alliance . https://doi.org/10.5751/ES - 07519 - 200238 Lancaster, D., Haggarty, D. R., & B s managem ent effectiveness . In Ecology and Society (Vol. 20, Issue 3, p. 41). Resilience Alliance . https://doi.org/10.5751/ES - 07815 - 200341 Larson, A .M, Barry, D., & Dahal, G. R. (2010). New rights for forest - based communities? Understanding processes of forest t enure reform. International Forestry Review , 12 (1), 78 96. https://doi.org/10.1505/ifor.12.1.78 Larson, Anne M., & Soto, F. (2008). Decentrali zation of Natural Resource Governance Regimes. Annual Review of Environment and Resources , 33 (1), 213 239 . https:/ /doi.org/10.1146/annurev.environ.33.020607.095522 Luján - Álvarez, C., Miguel Olivas - García, J., Guadalupe González - Hernández, H., Vázquez - Álvar ez, S., Ciro Hernández - Díaz, J., & Luján - Álvarez, H. (2015). Forestería comunitaria en México: modelo est ratégico para empoderamiento y competitividad en la globalización Community forestry in Mexico: strategic model for the empowerment and competitiveness in globalization . 49 (49), 665 696. Luttrell, C., Quiroz, S., Scrutton, C., & Bird, K. (2007). Understand ing and o perationalising empowerment . Overseas Development Institute. Merino, L., & Cendejas, J. (2017). Peace building from a commons perspective. In International Journal of the Commons . https://doi.org/10.18352/ijc.773 Miles, M. B., Huberman, M. A., & S aldaña, J . (2014). Qualitative Data Analysis. A Methods Sourcebook. Ze itschrift Für Personalforschung , 28 (4), 485 487. Narayan, D. (2007). Empowerment. Journal of Ambulatory Care Management , 30 , 120 125. https://doi.org/10.1097/01.JAC.0000264601.19629.5f N ASA. (201 9). Landsat 7 (L7) Data Users Handbook . 7 (November), 106. htt ps://prd - wret.s3.us - west - 2.amazonaws.com/assets/palladium/production/atoms/files/LSDS - 1927_L7_Data_Users_Handbook - v2.pdf Nelson, L. (2003). Decentering the movement: Collective action, p lace, and of radical political discourses. Environ ment and Planning D: Society and Space , 21 (5), 559 581. https://doi.org/10.1068/d348 66 Nyland, K. E., Gunn, G. E., Shiklomanov, N. I., Engstrom, R. N., & Streletskiy, D. A. (2018). Land co ver chang e in the lower Yenisei River using dense stacking of landsat imagery in Google Earth Engine. Remote Sensing , 10 (8). https://doi.org/10.3390/rs10081226 Orozco - Quintero, A., & Davidson - Hunt, I. (2010). Community - based enterprises and the commons: Th e case of San Juan Nuevo Parangaricutiro, Mexico. International Journa l of the Commons , 4 (1), 8 35. http://www.thecommonsjournal.org/index.php/ijc/article/view/URN:NBN:NL:UI:10 - 1 - 100200/100 Ostrom, E. (1990). The Evolution of Institutions for Collective Ac tion: Vol . pbk. (p. 280). Cambridge University Press. https://doi.org/ 10.1017/CBO9780511807763 Pérez, L. M. (2003). Los bosques de México, reflexiones en torno a su manejo y conservación. Ciencias , 072 , 59 67. Raudla, R. (2010). Governing budgetary commons : what ca n we learn from Elinor Ostrom? . In European Journal of Law a nd Economics (Vol. 30, Issue 3, pp. 201 221). Springer US . https://doi.org/10.1007/s10657 - 010 - 9187 - 6 Ribot, J. C. (2002). Democratic decentralization of natural resources: institutiona lizing po pular participation . World Resources Institute. http://msulib raries.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwY2AwNtIz0EUr EyyB1byxqXmiQaqlYRqwWk - 0SEs1S0kyME4xAdYRBqC9ylGORqFuJgHB5m5MDLCL6mCHOIK69bnQTiNs8ALU8 QHWTnXBxpag9pGFCWT_niHoFj5gp8TSHN S0gJ3qBOc D65fcYsS6I6S6xE2QgQW0v0 CIgSk1T5iBE7YxuFiEwcAlNT Ribot, J. C., Agrawal, A., & Larson, A. M. (2006). Recentralizing While Decentralizing: How National Governments Reappropriate Forest Resources. World Development , 34 (11), 1864 1886. https://doi.org/ 10.1016/j .worlddev.2005.11.020 Rojas, F. (2012). defensa del bosque . Universidad Autónoma Metropolitana Iztapalapa. Sarker, A., Ikeda, T., Abe, T., & Inoue, K. ( 2015). Design principles for managing coastal fisheries commons i n present - day Japan . In Ecological Economics (Vol. 117, pp. 32 38). Elsevier B.V . https://doi.org/10.1016/j.ecolecon.2015.06.019 Sattler, C., Schröter, B., Jericó - Daminello, C., Sessin - Dil ascio, K., Meyer, C., Matzdorf, B., Wortmann, L., de Almeida Sini sgalli, P. A., Meyer, A., & Giersch, G. (2015). Understanding governance structures in community management of ecosystems and natural resources: The Marujá case study in Brazil . In Ecosystem Services (Vol. 16, pp. 182 191). https://doi.org/10.1016/j.ecos er.2015.10.015 Schlager, E., & Ostrom, E. (1992). Property - Conceptual Analysis. Land Economics , 68 (3), 249 262. https://doi.org/10.2307/3146375 67 The Wo rldwide Reference System. Landsat Science . (n.d.). Retrieved Marc h 31, 2017, from https://landsat.gsfc.nasa.gov/the - worldwide - reference - system/ Trimble, M., & Berkes, F. (2015). Towards adaptive co - management of small - scale fisheries in Uruguay and Brazil: Maritime Stud ies U6 - ctx_ver=Z39.88 - 2004&ctx_enc=info%3Aofi%2Fenc%3AUTF - 8&rfr_id=info:sid/summon.serialssolutions.com&rft_val_fmt=info:ofi/fmt:kev:mtx:journal &rft.genre=article&rft.atitle=Towards+adapt ive+co - management+of+small - scale+fisheries+in+Urugu (Vol. 14, Iss ue 1, pp. 1 20). Springer Berlin Heidelberg . https://doi.org/10.1186/s40152 - 015 - 0032 - y Tucker, C. M. (1999). Private Versus Common Property Forests: Forest Conditions and Tenure in a Hondura n Community. Human Ecology , 27 (2), 201 230. https://doi.org/10.10 23/A:1018721826964 Butterfly Reserve. Society & Natural Resources , 17 (7), 569 587. https://doi.org/10.10 80/08941920490466143 Valdiva Dounce, M. T. (2010). ELECCIONES POR USOS Y COSTUMBRES EN EL CONTEXTO DE LAS REFORMAS ESTATALES OAXAQUEÑAS (1990 - 1998). , 23 (63), 247 263. van Ast, J., Widaryati, A., & Bal, M. (2014 ). es for Sustainable Management of Common Pool Resources. Conservation and Society , 12 (3), 294. https://doi.org/10.4103/0972 - 4923.1 45146 Velázquez, A., Mas, J., Díaz Gallegos, J., Mayorga Saucedo, R., Alcántara, P., Castro, R., Fernández, T., Bocco, G., Ezcurra, E., & Palacio, J. (2002). Patrones y tasas de cambio de uso del suelo en México. Gaceta Ecológica , 062 (January), 21 37. https://doi.org/10.4067/S0718 - 23762009000200004 Velázquez, Alejandra. (2013). RECONSTITUCIÓN DEL TERRITORIO COMUNAL . CIESAS. Velez, M. A. (2011). Collective Titling and the Process of Institution Building: The New Common Property Regime in the Colom bian Pacific . Human Ecology. https://doi.org/10.1007/s10745 - 011 - 9375 - 1 Vélez , M. A. (2009). Sistemas complejos de gobierno local. Reflexiones sobre la titulación colectiva en el Pacífico vallecaucano. Revista de Estudios Sociales , 32 , 74 85. https://doi.or g/10.1016/j.jebo.2008.02.008 Velez, M. A., Robalino, J., Cardenas, J. C., Pa z, A., & Pacay, E. (2020). Is collective titling enough to protect forests? Evidence from Afro - descendant communities in the Colombian Pacific region. World Development , 128 , 2000 2005. https://doi.org/10.1016/j.worlddev.2019.104837 68 Ventura Patiño, M. del C. (2012). Proceso de autonomía en Cherán. Movilizar el derecho. Espiral, Estudios Sobre Estado y Sociedad , XIX (55), 157 176. Wade, R. (1988). Village Republics: Vol. 40. (p. Cambr idge University Press). Cambridge University Press. http://msulibraries.summ on.serialssolutions.com/2.0.0/link/0/eLvHCXMwhV09C8IwFHw UHRQXP_GbLo5Km6ZNMopYHAU_Bhdp4quTDnb3t5vQBquDjoEQXiDk7kLuHkBAFt 78604wdkaZhoGgjIYhUpJo5PdkgoowlKnxKp - W5BDT7Y7FDthGdTbE0cj6WyE aS7FJTHOR5y4Qhh9xmvv33mzdcH - q0b9wqBrtELEic8eONe1t3L Wainer, A. (2016). Development in Practice Rural development and migration in Mexico . https://doi.org/10.1080/09614524.2013.772025 World Bank. (2001). Mexico Land Policy: A Decade after the Ejido Reform. World Bank Report 22187 - ME , 22187 , 71. World Bank. (2005). Income Gener Poverty in Mexico . https://openknowledge.worldbank.org/handle/10986/8305 Wright, G. D., Andersson, K. P., Gibson, C. C., & Ev ans, T. P. (2016). Decentralization can help reduce deforestation when user groups engage with local government. Proceedings of the National Academy of Sciences , 201610650. https://doi.org/10.1073/pnas.1610650114 Yin, R. K. (2009). Case Study Research: Des ign and Methods. In Essential guide to qualitative methods in organizat ional research (Vol. 5, Issue 5). https://doi.org/10.1097/FCH.0b013e31822dda9e