Research

INTRODUCTION

In this study, we evaluate how the consumption of different wildlife species in Amazonian towns is influenced by social processes, and how species-specific consumption may relate to taste preferences and food insecurity. First, we attempt to disentangle the ways in which the species consumed and preferred are shaped by diverse kinds of access, including social capital, practicing rural livelihoods, being a rural-urban migrant, and geographic context. Second, we explore the linkages between species preferences and their consumption. Third, we assess potential variation in the linkages between household food insecurity and consumption of different wildlife species; an overlooked issue yet vital for designing policy interventions that balance the trade-offs between biodiversity conservation and human welfare (Cawthorn and Hoffman 2015). This research engages with a strategic priority in wild meat research by evaluating the socially constructed and complex, potentially species-specific, role of wild meat in food systems in tropical forest regions (Ingram et al. 2021). We use the terms wildlife to refer to wild mammals, birds, reptiles, and fish and wild meat to refer to meat from wild mammals, birds, and reptiles (but not fish).

The world is urbanizing at an unprecedented rate, due to the rapid growth of towns and cities, particularly in low- and middle-income countries (LMICs) (United Nations 2019). This century, population growth is expected to be greater in small and medium-sized urban areas, rather than in megacities (Guneralp et al. 2018). Secondary towns are key to achieving development goals given that rural-urban migrants, urban poverty, and food insecurity are concentrated in these places (Gibson et al. 2017, Ingelaere et al. 2018, Lanjouw and Marra 2018). Understanding the food practices of people living in urban areas is recognized as fundamental for achieving food security and developing sustainable, resilient food systems (Tendall et al. 2015, Meybeck and Gitz 2017, El Bilali et al. 2019). Indeed, there is growing interest in the role smaller towns, and their rural-urban connections, play within food systems (Tacoli and Agergaard 2017). Recently, however, a Lancet Commission concluded that achieving sustainable, equitable food systems requires going beyond agriculture and elucidating the role of wild foods in urban diets (Willett et al. 2019). Not surprisingly, urban population growth is predicted to increase demand for natural resources, including wild meat (Guneralp et al. 2018).   

Urban demand for wild meat in Amazonia is already significant (Van Vliet et al. 2011, Van Vliet et al. 2019, El Bizri et al. 2020, Chaves et al. 2021). Urban consumption may pose significant risks to vulnerable species of large vertebrates given that urban demand can be concentrated on a few preferred or valued species (e.g., tapir and white-lipped peccary in Amazonas state, Brazil; Carignano Torres et al. 2022). However, at least in rural contexts in the forested tropics, wild meat can provide “natural insurance” in periods of food shortages, emergencies, or economic hardship (Brashares et al. 2004, Jambiya et al. 2007, Sunderland 2011) and contribute to food security (Williamson 2002, Cawthorn and Hoffman 2015, Fa et al. 2015), but this can vary seasonally, with declines in wildlife harvest being associated with greater food insecurity (Tregidgo et al. 2020). Food insecurity is the absence of reliable access to sufficient quality and quantity of food, affecting around 800 million people worldwide (FAO 2014, Schmeer and Piperata 2017). In some urban areas (e.g., Kisangani, in Democratic Republic of Congo), wild meat can be one of the cheapest, and relatively nutritious, animal-sourced foods available, which underlies its apparent importance for poor households (Van Vliet et al. 2012). Yet, even within the same society, wild meat can have heterogeneous and species-specific linkages with poverty and social circumstances. For instance, although wild pigeons in Samoa are prestigious and expensive, their market-based consumption by wealthier households occurs alongside subsistence hunting for domestic consumption (Stirnemann et al. 2018). Hence, even “expensive” species might be accessed by poor households if they could obtain these species outside of market exchange through direct harvest or social relations (gifting and reciprocity; Hyden 1983).

Consumption of wildlife and related sustainability risks are typically framed as a rural issue (Ingram et al. 2021) and, to our knowledge, no study has evaluated the relationship(s) between wild meat species-specific consumption, diverse access mechanisms, urban food insecurity, and conservation of threatened species. We know that urban consumption of wildlife is influenced by social and economic factors including income, wealth and social status, earlier life (e.g., rural origins), gender, and age (e.g., Drury 2011, Parry et al. 2014, Shairp et al. 2016, Chaves et al. 2021). Understanding the socio-cultural and economic determinants of urban consumption of wildlife is key to informing appropriate policy interventions to avoid over-exploitation and promote sustainable use (sensu Cawthorn and Hoffman 2015). Nonetheless, most studies focus on urban consumption of wildlife in general and overlook whether preferences or social determinants of consumption are species-specific. Species-specific vulnerability to overharvesting is well established in the ecological sciences and conservation legislation (e.g., ICUN red-listing). However, the ways in which the consumption patterns, which reflect food choices and constraints, of urban residents may vary by wildlife species are poorly understood.

Any social patterning (i.e., differentiation) in wild species consumption is likely to reflect different kinds of households’ preferences, opportunities, and choices to acquire different kinds of wildlife and other animal-sourced foods. Here, we investigate how these preferences and opportunities shape urban consumption of wildlife by combining Sobal et al.’s (2006) food choice model with Ribot and Peluso’s (2003) theorizing on access to natural resources. Sobal et al.’s model has been influential in public health research (e.g., in relation to food choice and social disadvantage and policies to prevent obesity; Devine et al. 2006, Hawkes et al. 2015) and includes three components: (1) a person’s life experiences related to the accumulation of eating experiences, food choice trajectories, and transitions through the life course; (2) the influences of culturally-learned ideals of what and how we should eat, resources including time, money, transportation, skills, social relationships and networks, and contexts (physical environment, social structures, political economy); and (3) personal factors, such as taste preferences, self-image, and identities. Linking across these components, research from Congo demonstrates that perceptions of wild meat (e.g., as natural, tasty, or healthy) are associated with social norms (e.g., as a luxury status symbol) to influence its consumption in urban areas (Chausson et al. 2019). According to Ribot and Peluso (2003), households derive benefits from (natural) things through diverse means, relations, and processes, mechanisms which together comprise a “bundle of powers.” Hicks and Cinner (2014) classified (and then quantified related measures) these mechanisms into four categories: (1) right of access, (2) access through knowledge, (3) economic access (e.g., to markets, capital, technology, labor), and (4) social and institutional access (i.e., access to authority, social relations, or through social identities).

We apply this hybrid conceptual framework of access and preferences to understanding species-specific consumption in towns in the Brazilian Amazon (Fig. 1), where these intersecting social and environmental challenges coalesce around the apparent importance of wildlife as food for a growing urban population (Parry et al. 2014, El Bizri et al. 2020, Chaves et al. 2021); ~73% of people in the Brazilian Amazon live in urban areas (IBGE 2018). We then evaluate associations between wildlife consumption and household food insecurity. We do so by asking the following research questions: (1) How do access mechanisms, earlier life, and place influence urban consumption of different wild species?; (2) How do access mechanisms, earlier life, and place influence wildlife food preferences?; (3) Are wildlife taste preferences associated with consumption of different species?; and, (4) How does food insecurity relate to consumption of different wild species? We expect that a household’s access mechanisms interlink with their specific taste preferences and consumption patterns. First, people will use their bundle of powers to obtain preferred foods. Second, a household’s particular bundle of powers will shape opportunities for acquiring different wild foods, and hence influence which foods they eat. The access mechanisms we assessed are horizontal social capital (i.e., level of social trust, interpreted as forming the basis of reciprocity and informal urban safety nets; Ruel et al. 1999); vertical social capital (i.e., level of trust in authority); rural visits by household members; direct harvest (i.e., access through rural knowledge and skills); and cash income (i.e., economic access through markets, or as an indicator of household members’ power and esteem). We also include place-based access (i.e., municipality, Brazil’s lowest level of local governance) because a natural resource’s availability, market presence, and cultural preferences can vary with context. Earlier life experience is represented by the rural origins (or not) of household heads.

Neither Access Theory nor Sobal’s food choice model explicitly account for social position: a limitation for our study given that social norms may underlie wildlife food preferences and practices. Therefore, we additionally draw on Pierre Bourdieu’s notion of habitus (slowly acquired, socially-learned habits of thinking and action), common to people of similar backgrounds, including class, education, and profession (Castree et al. 2013). We thus also explore the effect of social background on urban Amazonians’ wildlife food practices (as social actions) by using formal education (linked to social class) as an additional predictor. Any effect of habitus linked to education (and, hence, social class) is marginal to the effects of rural origin (early life experience of eating wildlife in rural areas will partly reflect context-specific social norms) and town (habitus also reflects local cultural norms which may be common to a specific town).

We assessed the consumption of 13 species, including many of the ecologically, culturally, and economically important wildlife species consumed in Amazonia, capturing a spectrum of market prices, levels of consumption, taste preference, vulnerability to overhunting, and legality of consumption (Bodmer and Robinson 2004, Castello et al. 2011, Castello et al. 2015, Isaac et al. 2015, Van Vliet et al. 2015, Carignano Torres et al. 2016, Chaves et al. 2019, El Bizri et al. 2020, Mayor et al. 2021). Some species may be overharvested because they have low reproductive rates (such as tapir; Tobler et al. 2014) or take many years to achieve reproductive age (e.g., 12–20 years for turtle species; Vogt 2008). Other species may be able to sustain harvest because they have high rates of population increase or achieve reproductive age within a few years or even months such as paca (El Bizri et al. 2019). Some of the mammal, bird, and reptile species included in this study figure among the most consumed species in the region (e.g., paca, tapir, white-lipped peccary; Chaves et al. 2018 [Table 2]; Carignano Torres et al. 2022 [Fig. 4]), but other species are less consumed (e.g., manatee; Chaves et al. 2018 [Table 2]). We expect that some of these species may be more consumed by food-insecure families than others.

We acknowledge that some fish species are likely to be more heavily consumed in the four study towns, compared to the three fish species included in our consumption survey. Non-surveyed species widely consumed in Amazonian urban areas are in the Characidae family (e.g., Semaprochilodus spp. [local name, jaraqui], Triportheus spp. [sardinha], Prochilodus nigricans [curimatã], Brycon amazonicus [matrinxã], Potamorhina latior [branquinha] , “Pacu” spp.), Cichlidae family (e.g., Cichla spp. [tucunaré], “Acará” spp.), and Siluriforme Pterygoplichthys pardalis [bodó] (Santos et al. 2006; Gandra 2010; Parry et al. 2014). However, because some Characidae and Cichlidae fish species are so widely consumed across the region, at least when seasonally abundant in a particular location, we considered them unsuitable candidate species for examining linkages between odds of consumption and a household’s diversity in access mechanisms, taste preference, or food insecurity. Therefore, to look at these relationships, we focused on consumption of selected wildlife species. Nonetheless, the three fishes in our survey are still significant in terms of their catches within the Amazonian fisheries. For instance, Tregidgo et al. (2021) assessed the catches of 22 rural communities (fishing for home consumption and urban markets) along the Purus river, finding that arapaima ranked third (9.9% of caught biomass); barred catfish ranked fifth (4.6%), and red-tailed catfish, ranked seventh (3.6%). Our approach does not intend to address reliance on the selected species. Rather, it aims to test how household consumption of different species relates to access mechanisms, taste preference, and food insecurity, and explore how a species-specific approach to managing wildlife harvesting and restrictions on market sale may be more appropriate than a general approach to wildlife management.

METHODS

Study sites and design

Urban areas in the Amazon have grown rapidly in recent decades, partly due to rural-to-urban migration (Parry et al. 2010, and supplementary material in Parry et al. 2014). As a result, many urban residents in provincial towns have rural origins and maintain access to rural areas (Parry et al. 2010, Parry et al. 2014, Dodd 2020, Chaves et al. 2021). We used a cross-sectional design to assess how access (household access mechanisms, rural origin, and place-based effect) influences wild meat (i.e., meat from wild animals) and fish consumption (Q1), and preferences for the taste of wild meat (Q2; Fig. 1) among urban populations of four municipalities, namely Caapiranga, Ipixuna, Jutaí, and Maués, in Amazonas state, Brazil (Fig. A1.1; Appendix 1). We also looked at associations between preferences for, and consumption of, wild meat (Q3) and between perceived food insecurity and consumption of wildlife (Q4; Fig. 1).

These four municipalities have >90% forest cover remaining, and their urban centers represent a gradient of remoteness within a hierarchical urban network (Prodes 2021). Caapiranga, Ipixuna, and Jutaí are small towns (<15,000 residents in the urban areas) and Maués is medium-sized (~ 35,000 residents; IBGE 2010). Their economies include services, small-scale commerce, employment by the municipal government (IBGE 2010), sale of agricultural produce, and natural resources (e.g., fishes, and non-timber forest products such as açai). Farming and natural resource harvesting (including in managed agro-forests) occurs in both peri-urban and more remote rural localities. Maués has more infrastructure and services than the smaller towns, including more grocery stores, markets, and diversity of labor (IBGE 2018). People in these municipalities mostly self-identify as being of mixed White, Black, and Indigenous (categorized as “pardo,” median 71% of the population), with smaller proportions of White, Black, and Indigenous people (IBGE 2018).

We randomly selected households, adjusting sampling density to the household density per census sector from the national population census of 2010 (IBGE 2010). We geolocated sampling points using Open Street Map (2014), Google Earth (Google 2014), and a purchased satellite image for Jutaí. We selected sampling points relative to the population density (i.e., density of households within each census sector; IBGE 2010) and limited to the potential inhabitable area of the towns, defined as located within a 20m radius of streets or river edge. We approached the nearest household at each location for an interview and registered the coordinates for all households. All research was approved by Brazil’s National Research Ethics Committee (CONEP/CNS; protocol 45383215.5.0000.0005) and Lancaster University’s Research Ethics Committee (S2014/126).

Data collection

We conducted household surveys using a standardized questionnaire (Appendices 2 and 3). We collected data during the dry season (fieldwork during August to December 2015), and wet season (fieldwork during March to July 2016), with a sample size of approximately 200 households per town (50% of these households in each season). Timing for dry and wet seasons was based on long-term seasonality trends we derived using data from Brazil’s National Water Agency (E.D. Figure 4 in Chacón-Montalván et al. 2021). In other words, we carried out fieldwork first in the town with the earliest dry-season onset, and so on. We obtained a final sample of 798 households. The questionnaire was piloted in a small town (Autazes) in Amazonas.

Household consumption

We surveyed heads of households (men or women) and assessed the date of last consumption of 13 wildlife species in the household (Table 1) (i.e., not including consumption that may have occurred elsewhere in town, or when visiting a rural area). For the purpose of this study, we analyzed consumption (yes/no) within the last 12 months. For Amazonian manatee, which had very low rate of consumption within the last 12 months, we considered household consumption in the last 60 months in order to have enough observations to generate a converging model.

Access

We examined how access (household access mechanisms, rural origin, and place-based effects) was associated with consumption of different wildlife species. Next, we describe how we measured access.

Horizontal and vertical social capital

Access Theory does not refer to social capital but emphasizes “relational access”, which Ribot and Peluso (2003) describe as relying on social relations of friendship, trust, reciprocity, patronage, dependence, and obligation. We measure a household’s relational access to food and other resources using the cognitive dimensions of social capital, which refer to the norms of trust and reciprocity which emerge from networks and institutions (Putnam and Putnam, 2000). We developed measures of horizontal and vertical social capital using Likert-type 5-point questions from Grootaert et al. (2004). Each score is based on factor analysis using psych package within R (R Core Team 2020). Horizontal social capital captures the level of social trust in other citizens. Our score is based on answers to 5 questions around bonding capital (trust toward similar people such as family and close friends) and bridging capital (trust within socially heterogeneous groups, potentially including other people in your neighborhood). The questions were, (i) Are the majority of people in your area trustworthy?; (ii) Would the majority of people in the neighborhood help you if you needed it?; (iii) How much do you trust your friends?; (iv) Do you feel safe walking alone at night in the street in front of your house?; and, (v) Do you think the residents in your neighborhood are united? Vertical social capital (akin to linking social capital) connects people across gradients in formal power and authority. We asked about trust in doctors and nurses; municipal employees; local elected councilors; local police; and “Do you think the town hall listens to what you and people like you request and suggest?” (Appendix 1 and Fig. A1.2. for more details about horizontal and social capital).

Rural origin and other means of access

In addition to horizontal and vertical social capital, we assessed (1) rural origin, which refers to a binary variable on whether at least one of the household heads was originally from a rural area set to one, otherwise set to zero; (2) direct harvesting, i.e., whether someone in the household fishes or hunts, depending on whether the consumption referred to fish or other wildlife, respectively; (3) household monetary income in the previous 30 days (e.g., salaries, conditional cash transfers, retirement pensions, rent or other types of remuneration); (4) rural visits, specifically the approximate number of days any household member had spent in rural areas in past the 12 months; and (5) town (fixed factor). In addition, we included formal education, specifically the highest level of education (number of schooling years) by anyone in the household, as a covariate in the model to capture class-based associations with food practices (Appendix 1 for more details about rural visits and education variables).

Taste preference

We asked participants to list their three most preferred types of animal source foods, which included wild meat (mammals, birds, and reptiles), fish, and domesticated meat (beef, chicken, pork, etc.). We used this information to create a binary variable (0/1) for households who ranked (or not) wild meat among their three most preferred food items. This information was collected only during the second year of fieldwork, in 2016, from 400 households. We looked at two questions regarding taste preference. First, we assessed how a taste preference for wild meat (i.e., whether wild meat was ranked in the top three preferred meats) was influenced by access mechanisms (direct harvesting, rural visits, income), earlier life (rural origin), and place (town). Second, we looked at associations between taste preference for wild meat and consumption of the 10 mammal, bird, and reptile species (i.e., excluding the 3 surveyed fish species).

Food insecurity

We measured perceptions of food insecurity using a questionnaire module modified from the Brazilian Household Food Insecurity Scale (EBIA) (Rivero et al. 2022). The EBIA was developed and validated in Brazil in 2003, building on the Household Food Security Survey Module (HFSSM) from the US Department of Agriculture (Pérez-Escamilla et al. 2004). The EBIA is therefore similar to the widely-used Food Insecurity Access Scale (HFIAS), which also originated from the HFSSM, and was designed by United States Agency for International Development (USAID) to be adapted for different cultural contexts (Coates et al. 2007).

We defined food insecurity levels using the definitions underlying the EBIA (PNAD 2013, p. 28). Mild food insecurity reflects anxiety about running out of food. Increasing severity indicates reduction of portion sizes (moderate) or skipping meals (moderate-to-severe). Severe food insecurity means going hungry, or not eating for an entire day due to lacking food or resources. Accordingly, we classified the level of each household by the number of related questions to which they responded “yes,” controlling for whether there were children in the household or not (Appendix 1 for more details about the food insecurity scale used).

Data analyses

All analyses were conducted in R Studio (R Core Team 2020) and all continuous variables were standardized (i.e., mean of zero and standard deviation of 1). There was no collinearity among the predictor variables used in this study.

Socioeconomic determinants of consumption and taste preference

We used a logistic regression model to assess factors associated with consumption of wildlife species (research question 1) and factors associated with taste preference for wild meat (research question 2). Specifically, to assess factors that influence consumption of individual species, we included the predictor variables: horizontal social capital, vertical social capital, rural origin, income, direct harvesting, rural visits, town (using Maués as the baseline), and education. To assess factors that influence taste preference for wild meat, we included the predictor variables: rural origin, income, direct harvesting, rural visits, town, and education. In both analyses of consumption and taste preference, results for each predictor variable correspond to marginal effects, after accounting for other variables in the model.

Association between consumption and preference and consumption and food insecurity

We used a log-linear model to look at correlations between consumption of individual wildlife species and preference for wild meat (research question 3). We also used a log-linear model to look at correlations between consumption of individual wildlife species and food insecurity (research question 4). We used a binary variable for food insecurity, with households that were moderately or severely food insecure set to one and households that were mildly food insecure or food secure set to zero.

RESULTS

How access mechanisms, place, and rural origin influence consumption of species

Seventy-three percent of households stated that they had consumed wild meat (i.e., at least one species of 10 species of wild mammal, bird, or reptile we included in this study) and 83% had consumed fish (at least one species of the three surveyed fish species) in the household during the past 12 months. Among mammal species included in the survey, paca was consumed by 42.7% of households, tapir by 33.5%, white-lipped peccary by 32.8%, agouti by 13.5%, and howler monkey by 7.2%. Curassow was consumed by 9.8% of households. Among reptile species, yellow-headed turtle was consumed by 42.7% of households, tortoise by 11.9%, and caiman by 10.2%. Among fish species, arapaima was consumed by 72.0% of households, barred catfish by 49.1%, and redtail catfish by 8.5%. Manatee was consumed by 8.14% over the last 60 months.

For each unit increase in horizontal social capital (which varies on a scale of 1 to 5), there was an increase of 25% in the likelihood of consuming paca (odds ratio [OR]=1.25; 95% Confidence Interval [CI] 1.03-1.51; p=0.02) and 32% increase in the likelihood of consuming white-lipped peccary (OR=1.32; CI 1.07-1.57; p=0.01; Fig. 2 and Table A1.1), after accounting for other variables such as income. Vertical social capital, on the other hand, was not associated with consumption of any species assessed. Direct harvesting by a household member significantly increased the odds of consumption for all species.

In addition, rural origin was associated with an increase in the odds of consuming howler monkey (78% more likely to consume; OR=1.78; 95% CI 0.93-3.61; p=0.09) and redtail catfish (69% more likely to consume; OR=1.69; CI 0.93-3.21; p=0.10), and a decrease in consumption of river turtle (26% less likely to consume; OR=0.74; CI 0.53-1.03; p=0.09; Fig. 2 and Table A1.1). Among rural out-migrant households, 9.6% stated that they consumed howler monkeys, 11.5% stated they consumed redtail catfish, and 39.1% stated they consumed turtles. Among non-migrant households, 3.8% stated they consumed howler monkeys, 4.7% stated they consumed redtail catfish, and 45.9% stated they consumed turtles.

Income was associated with increased odds of consuming Amazonian manatee, tortoise, and river turtle. As per capita income increased by one standard deviation (equivalent to an increase of 696.6 Brazilian reais), households were 35% more likely to consume Amazonian manatee (OR=1.35; CI 1.06–1.70; p=0.01), 37% more likely to consume tortoise (OR=1.37; CI 1.11–1.68; p=0.002), and 23% more likely to consume river turtle (OR=1.23; CI 1.05–1.45; p=0.01; Fig. 2 and Table A1.1). Also, as educational level increased by one standard deviation (equivalent to 3.7 grade level increase), households were 44% more likely to consume tortoise (OR=1.44; CI 1.12–1.88; p=0.01), 18% more likely to consume river turtle (OR=1.18; 1.00–1.39; p=0.05), and 24% less likely to consume caiman (OR=0.76; CI 0.61–0.97; p=0.02). As time spent in rural areas increased by one standard deviation (equivalent to 73 days increase per year), households were 28% more likely to consume curassow (OR=1.28; CI 1.02–1.59; p=0.03; Fig. 2 and Table A1.1). Finally, the odds of consuming wildlife and fish varied by town and by species. For seven species (agouti, howler monkey, paca, tapir, white-lipped peccary, curassow, and barred catfish), the odds of consuming were higher in small towns than in Maués (Fig. A1.3; Table A1.1). For three species (manatee, arapaima, and redtail catfish), odds of consumption were higher in Maués than in small towns. For three species (tortoise, turtle, and caiman), the odds of consumption varied across towns (Fig. A1.3 Table A1.1).

Associations between food preferences and access mechanisms

Wild meat (mammals and birds) was listed by 17% of 400 respondents as the first favorite food item, by 12% of respondents as the second favorite item, and by 10% of respondents as the third favorite item. Among the 39% of respondents who had wild meat in their “top three” animal foods, preferred species included white-lipped peccary (by 28% of those respondents), tapir (27%), paca (19%), and other species combined (26%). White-lipped peccary was listed by 5% of all respondents as the top preferred item, by 3.8% as the second preferred, and 2.8% as the third preferred. Tapir was listed by 3.8% as the top preferred item, 3.8% as the second preferred, and 3.3% as the third preferred. Paca was listed by 3% of respondents as their top preferred item, 2.3% as the second preferred, and 2.3% as the third preferred. Tortoises and freshwater turtles (including various Amazonian species) were listed by 6.5% of respondents as the top preferred item, by 7.3% as the second preferred, and by 5.0% as the third preferred. Among these 18.8% of respondents, the preferred species listed were the yellow-spotted river turtle (90%) and tortoise (10%).

Rural origin, direct harvesting, and place were associated with taste preference for wild meat. Households with rural origins were 90% more likely to have a strong preference for wild meat than other households (OR=1.90; CI 1.19–3.07; p=0.01; Table A1.2). Direct harvesting was associated with an 83% increase in the odds of having a strong preference for wild meat (OR=1.83; CI 1.10–3.10; p=0.02; Table A1.2). Households in the town of Jutaí were 2.05 times more likely to have a strong preference for wild meat than households from Maués (Table A1.2). The other variables included in the model (rural visits, education, and income) were not associated with taste preference.

Association between taste preference and consumption

Wild-meat preference was correlated with consumption of six wildlife species. Preference for wild meat was positively associated with consumption of manatee (OR=3.41; CI 1.70–6.98; p<0.001), paca (OR= 1.88; CI 1.23–2.89; p=0.004), tapir (OR=1.67; CI 1.09–2.58; p=0.02), white-lipped peccary (OR=1.82; CI 1.17–2.83; p=0.008), tortoise (OR=2.06; CI 1.09–3.85; p=0.02, and river turtle (OR=2.51; CI 1.63–3.90; p<0.0001), but uncorrelated for agouti, howler monkey, caiman, and curassow (Fig. 3 and Table A1.3).

Association between food insecurity and consumption

Of the 798 households surveyed, 50.8% were moderately or severely food insecure. Moderate and severe food insecurity, compared to mild food insecurity and food security, was positively correlated with eating howler monkey (OR=1.60; CI 0.93–2.82; p=0.10), barred catfish (OR=1.28; 0.97–1.70; p=0.08), and redtail catfish (OR=2.96; p<0.001), and negatively correlated with eating manatee (OR=0.58; CI 0.34–0.97; p-value=0.04) and river turtle (OR=0.72; CI 0.54–0.96; p=0.02; Fig. 4 and Table A1.4). Among households where participants stated that they consumed howler monkeys, barred catfish, redtail catfish, manatee, and river turtles, 60.4%, 73.5%, 54%, 38%, and 45%, respectively, were moderately or severely food insecure. There was no correlation between food insecurity and consumption of the other nine species.

DISCUSSION

We examined the species-specific role of wildlife in the Amazonian food system, with a focus on urban consumption. We used an empirical, yet theoretically informed, approach that included a hybrid framework to account for social understandings of food choices (Sobal et al. 2006) and the diverse means by which households can access natural resources (Ribot and Peluso 2003; Fig. 1). A main finding was that access mechanisms, earlier life, and place influenced consumption of different wildlife species and taste preferences for wild meat. An additional novel insight was that consumption was associated with taste preferences and with food insecurity, but the nature of these associations varied by species. Although long-standing assertions that wild meat underpins household food security (Milner-Gulland et al. 2003) are finally receiving empirical support (Nunes et al. 2019, Friant et al. 2020), we find that, at least in urban areas, the relationship between food (in)security and wildlife consumption may be species-specific. This specificity is perhaps unsurprising given that consuming a particular species reflects not only the ability to access it (e.g., through market exchange, harvesting, or social relations) but also food choices, which are embedded within preferences, identities, and cultural meanings (Sobal et al. 2006).   

Urban consumers’ access to wild meat is species-specific

Our research shows that access mechanisms, earlier life, and place all influenced wildlife consumption, but the type of access shaping consumption was species-specific. Some vulnerable species appear to enter illicit urban markets, based on our finding that their consumption was positively associated with higher monetary income (e.g., manatees, tortoises, and turtles). Although these same species were more likely to be consumed by urban households that hunted or fished, direct harvesting was positively associated with the odds of consuming all of the species we assessed. This suggests that those urban households which hunt or fish adopt (or maintain, if they have migrated from rural areas) livelihoods that draw on their ecological knowledge and harvesting skills in order to consume a broad variety of species.

Our findings demonstrate that culture, economic resources, and social class influence the choices around which wildlife species are consumed in urban households. Eating howler monkey or redtail catfish was more common among rural out-migrants, even after accounting for participation in harvesting livelihoods and resources (levels of social capital, income, or education). In other words, earlier life rural experiences, which presumably embed cultural ideals and taste preferences (Sobal et al. 2006), appear to influence the choice to eat (or avoid) these species. Conversely, eating yellow-spotted turtles was less likely in households with rural origins. It seems unlikely, though, that migrants are avoiding eating turtles due to particular rural cultural ideals or taste preferences, given that eating turtles is generally culturally desirable in Amazonas state (personal observations of the authors). Instead, perhaps rural-urban migrants tend to sell, barter, or gift turtles to others rather than consume turtles themselves. Some species (manatee, tortoise, turtle) were seldom eaten in poorer households and were more likely to be eaten in wealthier households. Albeit we cannot distinguish whether wealthier households had superior access to these species within the towns’ informal (often illicit) wildlife markets or because their money interplayed with greater power and esteem (sensu Drury et al. 2011), which could plausibly lead to receiving wildlife through gifting. The consumption of several species was linked to social class, which we proxied using formal education; caiman was more likely to be consumed in lower-class households whereas turtle was more likely to be consumed among upper social strata. Given that we accounted for household resources, such as rural origins and livelihoods, this variation between caiman and turtles may partly reflect class-based differences in habitus in these towns. Potentially, socially-learned food choices may reflect normative ideals of what wild species people in the region should or should not eat, which then plays out in terms of an individual’s taste preferences, self-image, and identity (Sobal et al. 2006). Perhaps related to these habits in thinking and actions, (relatively) upper-class strata households appear to purposefully avoid eating caiman (Swan et al. 2016), given they have the resources to acquire it within these towns, should they so wish (sensu Parry et al. 2014).

A major finding was that higher levels of social trust (i.e., horizontal social capital) translate into greater odds of consuming some species (two forest mammals; paca and white-lipped peccary). This demonstrates the role of relational forms of access to natural resources (Ribot and Peluso 2004), even among urban populations. This makes sense given that wild meat consumption in urban Amazonia partly depends on informal wild meat markets (van Vliet et al. 2015). Moreover, social trust is known to underlie social safety nets and hence our results also hint at a wild meat “economy of affection” (gifting and reciprocity; Hyden 1983). These kinds of socioeconomic relations in urban Amazonia had previously been identified for fruits and vegetables (WinklerPrins and Souza 2005), and fishes (Lee et al. 2018). The acquisition of wild meat outside of market exchange has previously been observed in both rural and urban areas in the region, where wild meat is gifted by relatives, friends, or neighbors (Chaves et al. 2019, Carignano Torres et al. 2021). Yet, to our knowledge, no other study has linked the consumption of different wildlife species with variation in household cognitive social capital. Based on our findings, wildlife consumption may have a role in maintaining and strengthening relationships, including social connections between rural and urban environments (Morsello et al 2015, Chaves et al. 2019, Carignano Torres et al. 2022). Any interventions or stricter control to reduce consumption of vulnerable wildlife in urban areas, need to carefully consider the potential impacts of such actions on poorer households who may rely on their social capital to access wildlife species. Or conversely, when poorer urban households may gift wild meat in order to strengthen their social connectedness. Participatory approaches that also engage urban stakeholders in the decision-making process around wildlife management may help address these concerns.

Our results support the argument that policy for wildlife conservation and management should be species-specific. Vulnerable species that appear to be mainly consumed by wealthier, relatively privileged urban Amazonians (manatees, tortoises, turtles) do not seem directly relevant to the consumption of other households, such as rural out-migrants. Thus, there is potentially a case for stricter controls on these species and for interventions in urban areas to reduce demand. However, stricter controls could conflict with strong cultural food preferences (e.g., the strong preference we found for yellow-spotted turtles; Chaves et al. 2018) or the welfare of relatively disadvantaged urban households if they tend to sell higher-value species they acquire, instead of eating them at home. For species that are mainly consumed by poorer households, there is a strong case for ensuring equitable access to these species. At present, in Brazil, we are unaware of any mechanism for allowing sustainable wildlife harvest to supply urban consumers, except for caiman harvesting under management projects (for sale) to major urban centers (CEMAAM 2011, SEPROR 2011, Marioni et al. 2020). However, caiman is not among the most consumed species in urban areas in Amazonas state (consumed within the previous 12 months by 28% of urban households surveyed by Parry et al. 2014). To better inform such interventions, we will need to understand the reliance of poorer households on different species, as a source of food, importance of maintaining social relationships of trust and reciprocity, and income.

Place-based access (differences among the four towns, marginal to the effects of other household-scale predictors) influenced the consumption of several species. These differences are likely due to a combination of status of the wildlife populations (i.e., some species being more depleted in some locations in central Amazonia, than others), landscape and physical characteristics (e.g., floodplain versus terra firme, or remoteness from large urban centers; Parry and Peres 2015), histories of exploitation related to market access (Antunes et al. 2016), and conservation and management efforts (Campos-Silva et al. 2018, Eisemberg et al. 2019). These characteristics may influence the availability of different species to urban households in particular municipalities. For instance, for tapir and white-lipped peccary, both classified as Vulnerable by the IUCN, consumption was higher in smaller towns than in Maués. It is possible that these species’ populations are relatively more depleted around the larger town of Maués and, therefore, less available to urban residents.

Taste preferences and consumption

Our work shows that different access mechanisms are associated with varied wild meat taste preferences, supporting our hybrid conceptual framework which considers wildlife-related food practices as social actions. For instance, rural origin (interpreted as a cultural influence given it is marginal to the effects of other socioeconomic factors) and direct harvesting were associated with a strong preference for the taste of wild meat. Households in the small town of Jutaí also had higher taste preferences for wild meat when compared to Maués. These results indicate that wild meat consumption is a rural tradition that continues to influence households’ preferences even when they live in urban areas, but urban context will also affect these preferences. In addition, greater preference for the taste of wild meat was associated with consumption of several species. Not surprisingly, consumption was positively associated with taste preference for species ranked by households to be the most preferred (peccary, tapir, paca, turtle, and tortoise). These results suggest that, although a large number of wildlife species are consumed in these towns, people may draw on their diverse means of access (i.e., their bundle of powers, according to Ribot and Peluso 2003) to acquire certain preferred species and invest less time and resources in acquiring non-preferred species. These results support our framework under which food choices are shaped by social and economic resources (because they underpin access; Ribot and Peluso 2003) and personal preferences (Sobal et al. 2006). Our results also concur with Schenck et al. (2006) in that people make distinctions among wild species instead of treating them as a generic source of food. Unfortunately, in our study, most of the preferred species are also more vulnerable to overharvesting, when compared to the less vulnerable species we assessed. We argue that conservation efforts to reduce pressure on wildlife in urban areas need to prioritize species that are more vulnerable to overharvesting and preferred. More fully understanding the socio-cultural bases of these preferences would require additional qualitative research in order to explore the role of social norms, family influences, food safety perceptions, among other factors (Chausson et al. 2019). For instance, a recent qualitative study investigating primate consumption in Amazonia found the decision to eat related to a complex array of factors including taste perceptions, concerns about conservation, religious background, customs, and health concerns (Lemos et al. 2021).

Species-specific associations between food insecurity and wild meat consumption

Finally, our work shows complex associations between urban food insecurity and wildlife consumption, hence making an important contribution to the very limited literature on this subject. Since Milner-Gulland et al.’s seminal (2003) paper, the conservation literature repeats plausible yet largely untested assumptions, particularly lacking evidence for urban areas, that wild meat plays an important role in supporting food security in the forested tropics. A recent Amazonian study found that food-insecure urban households use small-scale fishing as a coping strategy (Rivero et al. 2022), but we are unaware of any previous study into the food (in)security dimensions of consuming other wild taxa (i.e., mammals, birds, chelonians, caimans) in urban areas. In rural Nigeria, Friant et al. (2020) found an overall positive relationship between bushmeat consumption and household food security, with some taxa-specific nuances (e.g., eating rodents was strongly associated with food security, with the opposite true for carnivores).

Although our work does not assess how much households rely on the species consumed, our species-specific, urban-centric research provides a novel insight that some wildlife species are more consumed among food-insecure households than relatively food-secure households. That is, some species (e.g., turtles, manatee) are less likely to be consumed by food-insecure urban households whereas other species are more likely to be eaten by food-insecure households (e.g., howler monkey, barred catfish, and red-tailed catfish; Fig 4). For example, although howler monkeys and redtail catfish were only eaten by 7.2% and 8.5% of surveyed households, respectively, 61.4% of consumers of howler monkeys and 73.5% of consumers of redtail catfish were moderately or severely food insecure. These species-specific differences are likely to be context-specific given their consumption will be related to socially-constructed preferences and food practices. For example, in Amazonas State, there are social taboos against the consumption of catfishes, yet Parry et al. (2014) found they were consumed by the poorest (often rural out-migrant) urban households. Similarly, Lemos et al. (2021) found that some Amazonian people consider eating primates (including howler monkeys) as taboo, while others have the custom of consuming primates. Given we found that eating howler monkeys was positively associated with rural origin and direct harvesting, but not associated with variation in monetary income, we interpret howler monkey consumption in Amazonian towns mainly in terms of socio-cultural practices. Albeit we cannot rule out the possibility that some urban households consume (and perhaps hunt) these primates as a direct response to insufficient access to other foodstuffs. People’s lived experiences in rural communities shape their different perceptions, knowledge, and attitudes toward a range of wildlife species (Mikolajczak et al. in press), which may partly explain why migrant households are more likely to consume howler monkeys and red-tailed catfish, beyond the effect of directly accessing them by going hunting or fishing. People’s ecological knowledge, habits, and dispositions toward different species are socially constructed, emerging during the childhood socialization process in rural Amazonian communities (Menegaldo et al. 2013). In addition, although we did not investigate the different ways food-secure and food-insecure households may acquire wild meat, it is likely that vulnerable households (more food insecure) rely less on purchasing than on hunting and gifts and that may influence their access to different species. Higher reliance on hunting and being gifted wild meat was more likely for rural out-migrants than for non-migrant in the study region (Carignano Torres et al. 2022).

We found that higher-income households tend to favor the same species that appear less likely to be consumed by food-insecure households. Turtles were ranked as highly preferred, so a lower probability of their consumption among food-insecure households may relate to barriers in accessing them for food, rather than to preferences. Although our work does not assess how much food-insecure households rely on the species consumed (which Ingram et al. (2021) define as a research priority), our results suggest that food-insecure households have greater access to some species but not others. These results highlight the importance of species-specific approaches regarding wildlife management and conservation.

Limitations and future research

Our data included binary response variables. While we acknowledge that having relative consumption frequency or quantity consumed of different species would have provided more detailed results about food systems in this region, our results still provide important insights into the relationships between access, taste preferences, food insecurity, and species-specific consumption. In addition, although exploring intra-household dynamics of food allocation (e.g., using participant observation) could have furthered our understanding of food choices and wild meat’s linkages to food (in)security, it was beyond the scope of this paper. Although we looked at diverse forms of access, our survey design did not permit us to assess the potential importance of rights-based access mentioned by (Ribot and Peluso 2003). Furthermore, future Amazonian research could explicitly examine how wildlife consumption preferences may relate to self-image and identities, such as in relation to forest livelihoods and historical struggles for rights, or identifying more strongly with cattle culture (Gomes et al. 2012). Finally, our work did not assess reliance on different species. Understanding how much poor, food-insecure households rely on different species for food and income will also be an important step to develop strategies (e.g., species-specific management, alternative livelihoods) to avoid unintended consequences of reducing urban demand for wild meat.

CONCLUSION

Despite claims that wild meat is important for food security in tropical forest regions (Milner-Gulland et al. 2003, Ingram et al. 2021), our research shows that its role is more nuanced, at least in Amazonian towns. Some species appear to be consumed more by poorer households than other species, depending on people’s access to the species, which is linked to livelihoods, social networks, and markets. Drawing on Sobal et al. (2006), our results suggest that, within urban Amazonian households, decisions to acquire and eat different wildlife species reflect life course experiences (particularly, rural origins and practicing rural livelihoods) and cultural ideals (perhaps explaining low preferences, e.g., for catfish), personal and social factors (e.g., monetary income, educational attainment and horizontal social capital), and context (e.g., town). The interplay of these factors helps shape a household’s food choice strategies, which in this urban context included wildlife. Ribot and Peluso’s (2003) theory helps us understand these food strategies in terms of how a household’s diverse means of access both reflect their food choices and may limit which species they are able to acquire, given their knowledge, skills, and social and economic resources. For instance, we found that some species are accessed by households who are relatively wealthy or upper-class, the latter indicated by education. These same species appear less consumed by rural out-migrants and food-insecure households who may use harvesting and social capital as coping strategies. Other species are consumed more by rural out-migrants and food insecure households while not favored by wealthier households. Taken together, our results provide strong support for a species-specific approach on how to manage (from supporting sustainable harvest, or enforcing against harvest or exchange) the consumption of forest wildlife by Amazonia’s growing urban populations.

Instead of either pursuing a blanket ban on wildlife harvesting or assuming all species are important for food security, policy needs to aim at sustainably managing species that are important to poorer, food-insecure households in particular contexts and that are more resilient to harvesting pressure. However, decisions on which species fit such criteria should be data-driven (e.g., which species would be more resilient to harvesting and important for food security of poorer households) and part of participatory approaches that engage local stakeholders, including diverse kinds of town-dwellers. Importantly, with a growing urban demand for natural resources, sustainable harvest of resilient species to provide wild meat is unlikely to be sufficient to meet that demand. Likewise, relying on harvesting a few resilient species will not likely fulfill the needs of poorer, food-insecure urban households. Considering the importance of fisheries for the economy and as a food source in Amazonia (Rivero et al. 2022; Coomes et al. 2010), investing in the management of inland fisheries (especially in ways that are inclusive of urban fishers) could provide alternative food sources and livelihoods that are culturally appropriate to address the needs not met by sustainable harvesting (Ingram et al. 2021) while trying to avoid unintended consequences (e.g., exacerbating human malnutrition; Heilpern et al. 2021)

LITERATURE CITED

Antunes, A. P., R. M. Fewster, E. M. Venticinque, C. A. Peres, T. Levi, F. Rohe, and G. H. Shepard. 2016. Empty forest or empty rivers? A century of commercial hunting in Amazonia. Science Advances 2(10). https://doi.org/10.1126/sciadv.1600936

Bodmer, R., and J. G. Robinson. 2004. Evaluating the sustainability of hunting in the neotropics. In K. M. Silvius, R. E. Bodmer, and J. M. V. Fragoso, editors. People in Nature: Wildlife Conservation in South and Central America. Columbia University Press, New York, NY, USA. https://doi.org/10.7312/silv12782-019

Brashares, J. S., P. Arcese, M. K. Sam, P. B. Coppolillo, A. R. E. Sinclair, and A. Balmford. 2004. Bushmeat hunting, wildlife declines, and fish supply in West Africa. Science 306:1180-1183. https://doi.org/10.1126/science.1102425

Campos-Silva, J., J. Hawes, P. Andrade, and C. Peres. 2018. Unintended multispecies co-benefits of an Amazonian community-based conservation programme - Nature Sustainability 1:650-656. https://doi.org/10.1038/s41893-018-0170-5

Carignano Torres, P., C. Morsello, and L. Parry. 2022. Rural-urban mobility influences wildmeat access and consumption in the Brazilian Amazon. Oryx 56(6):864-876 https://doi.org/10.1017/S0030605321001575

Carignano Torres, P., C. Morsello, L. Parry, and R. Pardini. 2016. Who cares about forests and why? individual values attributed to forests in a post-frontier region in Amazonia. Plos One 11:e0167691. https://doi.org/10.1371/journal.pone.0167691

Carignano Torres, P., C. Morsello, L. Parry, and R. Pardini. 2021. Forest cover and social relations are more important than economic factors in driving hunting and bushmeat consumption in post-frontier Amazonia. Biological Conservation 253:108823. https://doi.org/10.1016/j.biocon.2020.108823

Castello, L., C. C. Arantes, D. G. Mcgrath, D. J. Stewart, and F. S. D. Sousa. 2015. Understanding fishing-induced extinctions in the Amazon. Aquatic Conservation: Marine and Freshwater Ecosystems 25:587-598. https://doi.org/10.1002/aqc.2491

Castello, L., D. G. Mcgrath, and P. S. A. Beck. 2011. Resource sustainability in small-scale fisheries in the Lower Amazon floodplains. Fisheries Research 110:356-364. https://doi.org/10.1016/j.fishres.2011.05.002

Castree, N., R. Kitchin, and A. Rogers. 2013. A Dictionary of Human Geography. Oxford University Press,Oxford, UK. https://doi.org/10.1093/acref/9780199599868.001.0001

Cawthorn, D.-M., and L. C. Hoffman. 2015. The bushmeat and food security nexus: A global account of the contributions, conundrums and ethical collisions. Food Research International 76:906-925. https://doi.org/10.1016/j.foodres.2015.03.025

CEMAAM. 2011. Conselho Estadual do Meio Ambiente do Estado do Amazonas. Resolução CEMAAM n.008, de 27 junbo de 2011. Estabelece procedimentos ténicos para o manejo de jacaré, oriundo de Unidades de Conservação de Uso Sustentavel do Estado do Amazonas.

Chacón-Montalván, E. A., B. M. Taylor, M. G. Cunha, G. Davies, J. D. Y. Orellana, and L. Parry. 2021. Rainfall variability and adverse birth outcomes in Amazonia. Nature Sustainability 4:583-594. https://doi.org/10.1038/s41893-021-00684-9

Chausson, A. M., J. M. Rowcliffe, L. Escouflaire, M. Wieland, and J. H. Wright. 2019. Understanding the sociocultural drivers of urban bushmeat consumption for behavior change interventions in Pointe Noire, Republic of Congo. Human Ecology 47:179-191. https://doi.org/10.1007/s10745-019-0061-z

Chaves, W. A., D. R. Valle, M. C. Monroe, D. S. Wilkie, K. E Sieving, and B. Sadowsky, B. 2018. Changing wild meat consumption: an experiment in the Central Amazon, Brazil. Conservation Letters 11(2):e12391. https://doi.org/10.1111/conl.12391

Chaves, W. A., D. Valle, A. S. Tavares, T. Q. Morcatty, and D. S. Wilcove. 2021. Impacts of rural to urban migration, urbanization, and generational change on consumption of wild animals in the Amazon. Conservation Biology 35:1186-1197. https://doi.org/10.1111/cobi.13663

Chaves, W. A., M. C. Monroe, and K. E. Sieving. 2019. Wild meat trade and consumption in the central Amazon, Brazil. Human Ecology 47:733-746. https://doi.org/10.1007/s10745-019-00107-6

Coates, J., A. Swindale, and P. Bilinsky. 2007. Household Food Insecurity Access Scale (HFIAS) for measurement of food access: Indicator guide, version 3. Food and Nutrition Technical Assistance Project, Academy for Educational Development, Washington, D.C., USA. https://www.fao.org/fileadmin/user_upload/eufao-fsi4dm/doc-training/hfias.pdf

Coomes, O. T., Y. Takasaki, C. Abizaid, and B. L. Barham. 2010. Floodplain fisheries as natural insurance for the rural poor in tropical forest environments: evidence from Amazonia. Fisheries Management and Ecology 17(6):513-521. https://doi.org/10.1111/j.1365-2400.2010.00750.x

Devine, C. M., M. Jastran, J. Jabs, E. Wethington, T. J. Farell, and C. A. Bisogni. 2006. “A lot of sacrifices:” Work-family spillover and the food choice coping strategies of low-wage employed parents. Social Science & Medicine 63:2591-2603. https://doi.org/10.1016/j.socscimed.2006.06.029

Dodd, L. M. M. 2020. Aspiring to a good life: rural-urban mobility, and young people's desires in the Brazilian Amazon. The Journal of Latin American and Caribbean Anthropology 25:283-300. https://doi.org/10.1111/jlca.12478

Drury, R. 2011. Hungry for Success: Urban Consumer Demand for Wild Animal Products in Vietnam. Conservation and Society 9(3):247-257. https://doi.org/10.4103/0972-4923.86995

Eisemberg, C. C., R. C. Vogt, R. A. M. Balestra, S. J. Reynolds, and K. A. Christian. 2019. Don't put all your eggs in one basket - Lessons learned from the largest-scale and longest-term wildlife conservation program in the Amazon Basin. Biological Conservation 238:108182. https://doi.org/10.1016/j.biocon.2019.07.027

El Bilali, H., C. Callenius, C. Strassner, and L. Probst. 2019. Food and nutrition security and sustainability transitions in food systems. Food and Energy Security 8:e00154. https://doi.org/10.1002/fes3.154

El Bizri, H. R., J. E. Fa, J. Valsecchi, R. Bodmer, and P. Mayor. 2019. Age at sexual maturity, first parturition and reproductive senescence in wild lowland pacas (Cuniculus paca): Implications for harvest sustainability. Animal Reproduction Science 205:105-114. https://doi.org/10.1016/j.anireprosci.2019.04.009

El Bizri, H. R., T. Q. Morcatty, J. Valsecchi, P. Mayor, J. E. S. Ribeiro, C. F. A. Vasconcelos Neto, J. S. Oliveira, K. M. Furtado, U. C. Ferreira, C. F. S. Miranda, C. H. Silva, V. L. Lopes, G. P. Lopes, C. C. F. Florindo, R. C. Chagas, V. Nijman, and J. E. Fa. 2020. Urban wild meat consumption and trade in central Amazonia. Conservation Biology 34(2):438-448. https://doi.org/10.1111/cobi.13420

Fa, J. E., J. Olivero, M. Á. Farfán, A. L. Márquez, J. Duarte, J. Nackoney, A. Hall, J. Dupain, S. Seymour, P. J. Johnson, D. W. Macdonald, R. Real, and J. M. Vargas. 2015. Correlates of bushmeat in markets and depletion of wildlife. Conservation Biology 29:805-815. https://doi.org/10.1111/cobi.12441

FAO. 2014. The state of food insecurity in the world 2014. FAO Forestry Paper. Food and Agriculture Organization, Rome.

Friant, S., W. A. Ayambem, A. O. Alobi, N. M. Ifebueme, O. M. Otukpa, D. A. Ogar, C. B. I. Alawa, T. L. Goldberg, J. K. Jacka, and J. M. Rothman. 2020. Eating bushmeat improves food security in a biodiversity and infectious disease “hotspot”. Ecohealth 17:125-138. https://doi.org/10.1007/s10393-020-01473-0

Gandra, A. L. 2010. O mercado do pescado da região metropolitana de Manaus. INFOPESCA.

Gibson, J., G. Datt, R. Murgai, and M. Ravallion. 2017. For India’s Rural Poor, Growing Towns Matter More Than Growing Cities. World Development 98:413-429. https://doi.org/10.1016/j.worlddev.2017.05.014

Gomes, C. V. A., J. M. Vadjunec, and S. G. Perz. 2012. Rubber tapper identities: Political-economic dynamics, livelihood shifts, and environmental implications in a changing Amazon. Geoforum 43(2):260-271. https://doi.org/10.1016/j.geoforum.2011.09.005

Grootaert, C., D. Narayan, V. N.Jones, and M. Woolcock. 2004. Measuring social capital: An integrated questionnaire. World Bank Working Papers. https://doi.org/10.1596/0-8213-5661-5

Guneralp, B., S. Lwasa, H. Masundire, S. Parnell, and K. C. Seto. 2018. Urbanization in Africa: challenges and opportunities for conservation. Environmental Research Letters 13:8. https://doi.org/10.1088/1748-9326/aa94fe

Hawkes, C., T. G. Smith, J. Jewell, J. Wardle, R. A. Hammond, S. Friel, A. M. Thow, and J. Kain. 2015. Smart food policies for obesity prevention. The Lancet 385:2410-2421. https://doi.org/10.1016/S0140-6736(14)61745-1

Heilpern, S. A., K. Fiorella, C. Cañas, A. S. Flecker, L. Moya, S. Naeem, S. A. Sethi, M. Uriarte, and R. DeFries. 2021. Substitution of inland fisheries with aquaculture and chicken undermines human nutrition in the Peruvian Amazon. Nature Food 2(3):192-197. https://doi.org/10.1038/s43016-021-00242-8

Hicks, C. C., and J. E. Cinner. 2014. Social, institutional, and knowledge mechanisms mediate diverse ecosystem service benefits from coral reefs. Proceedings of the National Academy of Sciences 111:17791-17796. https://doi.org/10.1073/pnas.1413473111

Hyden, G. 1983. No shortcuts to progress: African development management in perspective, University of California Press, Oakland, CA, USA.

IBGE. 2010. Cidades@ Brasília, Brasil: Institute for Geography and Statistics of Brazil. http://www.cidades.IBGE.gov.br/xtras/home.php

IBGE. 2018. Cidades@ Brasília, Brasil: Institute for Geography and Statistics of Brazil, https://cidades.IBGE.gov.br

Ingelaere, B., L. Christiaensen, J. De Weerdt, and R. Kanbur. 2018. Why secondary towns can be important for poverty reduction - A migrant perspective. World Development 105:273-282. https://doi.org/10.1016/j.worlddev.2017.12.025

Ingram, D. J., L. Coad, E. J. Milner-Gulland, L. Parry, D. Wilkie, M. I. Bakarr, A. Benítez-López, E. L. Bennett, R. Bodmer, G. Cowlishaw, H. R. E. Bizri, H. E. Eves, J. E. Fa, C. D. Golden, D. M. Iponga, N. V. Minh, T. Q. Morcatty, R. Mwinyihali, R. Nasi, V. Nijman, Y. Ntiamoa-Baidu, F. Pattiselanno, C. A. Peres, M. Rao, J. G. Robinson, J. M. Rowcliffe, C. Stafford, M. Supuma, F. N. Tarla, N. Van Vliet, M. Wieland, and K. Abernethy. 2021. Wild meat is still on the menu: progress in wild meat research, policy, and practice from 2002 to 2020. Annual Review of Environment and Resources 46. https://doi.org/10.1146/annurev-environ-041020-063132

Isaac, V. J., M. C. Almeida, T. Giarrizzo, C. P. Deus, R. Vale, G. Klein, and A. Begossi. 2015. Food consumption as an indicator of the conservation of natural resources in riverine communities of the Brazilian Amazon. Anais Da Academia Brasileira De Ciencias 87:2229-2242. https://doi.org/10.1590/0001-3765201520140250

Jambiya, G., S. Milledge, and N. Mtango. 2007. Conservation implications and livelihood implications of wild meat use in refugee situations in North-Western Tanzania. Traffic East/Southern Africa, Dar es Salaam, Tanzania.

Lanjouw, P., and M. R. Marra. 2018. Urban poverty across the spectrum of Vietnam’s towns and cities. World Development 110:295-306. https://doi.org/10.1016/j.worlddev.2018.06.011

Lee, G. O., P. J. Surkan, J. Zelner, M. Paredes Olórtegui, P. Peñataro Yori, R. Ambikapathi, L. E. Caulfield, R. H. Gilman, and M. N. Kosek. 2018. Social connectedness is associated with food security among peri-urban Peruvian Amazonian communities. SSM Popul Health 4:254-262. https://doi.org/10.1016/j.ssmph.2018.02.004

Lemos, L. P., L. F. Loureiro, T. Q. Morcatty, J. E. Fa, C. F. A. De Vasconcelos Neto, A. De Souza Jesus, V. C. Da Silva, M. L. De Oliveira Ramalho, A. De Matos Mendes, J. Valsecchi, and H. R. El Bizri. 2021. Social correlates of and reasons for primate meat consumption in central Amazonia. International Journal of Primatology 42:499-521. https://doi.org/10.1007/s10764-021-00214-6

Marioni, B., J. a. L. Barão‐Nóbrega, R. Botero‐Arias, F. Muniz, Z. Campos, R. Da Silveira, W. E. Magnusson, and F. Villamarín. 2021. Science and conservation of Amazonian crocodilians: a historical review. Aquatic Conservation: Marine and Freshwater Ecosystems 31:1056-1067. https://doi.org/10.1002/aqc.3541

Mayor, P., H. R. El Bizri, T. Q. Morcatty, K. Moya, N. Bendayán, S. Solis, C. F. A. Vasconcelos Neto, M. Kirkland, O. Arevalo, T. G. Fang, P. E. Pérez‐Peña, and R. E. Bodmer. 2021. Wild meat trade over the last 45 years in the Peruvian Amazon. Conservation Biology 36(2):e13801 https://doi.org/10.1111/cobi.13801

Menegaldo, L. R., H. D. S. Pereira, and A. D. S Ferreira. 2013. Interações socioculturais com a fauna silvestre em uma unidade de conservação na Amazônia: relações de gênero e geração. Boletim do Museu Paraense Emílio Goeldi. Ciências Humanas 8:129-151. https://doi.org/10.1590/S1981-81222013000100008

Meybeck, A., and V. Gitz. 2017. Sustainable diets within sustainable food systems. Proceedings of the Nutrition Society 76:1-11. https://doi.org/10.1017/S0029665116000653

Mikołajczak, K., J. Barlow, A. Lees, C. Ives, M. Strack, O. Almeida, A. Souza, L. Parry, and F. Sinclair. In press. Evaluating the influence of nature connection and values on conservation attitudes at a tropical deforestation frontier. Conservation Biology e14067. https://doi.org/10.1111/cobi.14067

Milner-Gulland, E. J., and E. L. Bennett. 2003. Wild meat: the bigger picture. The SCB 2002 Annual Meeting Wild Meat Group. Trends in Ecology & Evolution 18:351-357. https://doi.org/10.1016/S0169-5347(03)00123-X

Morsello, C., B. Yagüe, L. Beltreschi, N. Van Vliet, C. Adams, T. Schor, M. P. Quiceno-Mesa, and D. Cruz. 2015. Cultural attitudes are stronger predictors of bushmeat consumption and preference than economic factors among urban Amazonians from Brazil and Colombia. Ecology and Society 20(4):21. https://doi.org/10.5751/ES-07771-200421

Nunes, A. V., C. A. Peres, P. D. L. Constantino, B. A. Santos, and E. Fischer. 2019. Irreplaceable socioeconomic value of wild meat extraction to local food security in rural Amazonia. Biological Conservation 236:171-179. https://doi.org/10.1016/j.biocon.2019.05.010

Parry, L., and C. A. Peres. 2015. Evaluating the use of local ecological knowledge to monitor hunted tropical-forest wildlife over large spatial scales. Ecology and Society 20(3):15. https://doi.org/10.5751/ES-07601-200315

Parry, L., C. A. Peres, B. Day, and S. Amaral. 2010. Rural-urban migration brings conservation threats and opportunities to Amazonian watersheds. Conservation Letters 3:251-259. https://doi.org/10.1111/j.1755-263X.2010.00106.x

Parry, L., J. Barlow, and H. Pereira. 2014. Wildlife Harvest and Consumption in Amazonia's Urbanized Wilderness. Conservation Letters 7:565-574. https://doi.org/10.1111/conl.12151

Pérez-Escamilla, R., A. M. Segall-CorrêA, L. Kurdian Maranha, M. D. F. T. A. Sampaio, L. Marín-León, and G. Panigassi. 2004. An adapted version of the U.S. Department of Agriculture food insecurity module is a valid tool for assessing household food insecurity in Campinas, Brazil. The Journal of Nutrition 134:1923-1928. https://doi.org/10.1093/jn/134.8.1923

PNAD. 2013. Pesquisa Nacional por Amostra de Domicílios (PNAD)Segurança Alimentar, Instituto Brasileiro de Geografia e Estatistica - IBGE. http://biblioteca.IBGE.gov.br/index.php/biblioteca-catalogo?view=detalhes&id=291984

PRODES. 2021. Monitoramento do Desmatamento da Floresta Amazônica Brasileira por Satélite. Instituto Nacional de Pesquisas Espaciais. http://www.obt.inpe.br/OBT/assuntos/programas/amazonia/prodes

R Core Team. 2020. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Rhodin, A., J. Iverson, R. Bour, U. Fritz, A. Georges, H. Shaffer, and P. Dijk. 2017. Turtles of the world: annotated checklist and atlas of taxonomy, synonymy, distribution, and conservation status (8th edition). Chelonian Research Foundation and Turtle Conservancy, New York, NY, USA. https://doi.org/10.3854/crm.7.checklist.atlas.v8.2017

Ribot, J. C., and N. L. Peluso. 2003. A Theory of Access*. Rural Sociology 68:153-181. https://doi.org/10.1111/j.1549-0831.2003.tb00133.x

Rivero, S. L., O. T. D Almeida, P. C. Torres, A. de Moraes, E. Chacón-Montalván, and L. Parry. 2022. Urban Amazonians use fishing as a strategy for coping with food insecurity. The Journal of Development Studies 1-22. https://doi.org/10.1080/00220388.2022.2113063

Ruel, M. T., L. Haddad, and J. L. Garrett. 1999. Some Urban Facts of Life: Implications for Research and Policy. World Development 27(11):1917-1938. https://doi.org/10.1016/S0305-750X(99)00095-9

Santos, G., E. Ferreira, and J. Zuanon 2006. Peixes Comerciais de Manaus. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA).

Schenck, M., E. Nsame Effa, M. Starkey, D. Wilkie, K. Abernethy, P. Telfer, R. Godoy, and A. Treves. 2006. Why people eat bushmeat: results from two-choice, taste tests in Gabon, Central Africa. Human Ecology 34:433-445. https://doi.org/10.1007/s10745-006-9025-1

Schmeer, K. K., and B. A. Piperata. 2017. Household food insecurity and child health. Maternal & Child Nutrition 13:e12301. https://doi.org/10.1111/mcn.12301

SEPROR. 2011. Instrução Normativa nº 001/2011 da SEPROR, estabelece critérios técnicos de abate e beneficiamento de crocodilianos no estado do Amazonas.

Shairp, R., D. Veríssimo, I. Fraser, D. Challender, and D. Macmillan. 2016. Understanding Urban Demand for Wild Meat in Vietnam: Implications for Conservation Actions. Plos One 11:e0134787. https://doi.org/10.1371/journal.pone.0134787

Sobal, J., C. A. Bisogni, C. M. Devine, and M. Jastran. 2006. A conceptual model of the food choice process over a life course. In R. Shepherd, and M. Raats, editors. The Psychology of Food Choice. CABI Publishing, Guildford, UK. https://doi.org/10.1079/9780851990323.0001

Stirnemann, R., I. Stirnemann, D. Abbot, D. Biggs, and R. Heinsohn. 2018. Interactive impacts of by-catch take and elite consumption of illegal wildlife. Biodiversity and Conservation 27:931–946. https://doi.org/10.1007/s10531-017-1473-y

Sunderland, T. C. H. 2011. Food security: why is biodiversity important? International Forestry Review 13:265-274. https://doi.org/10.1505/146554811798293908

Swan, N., J. Barlow, and L. Parry. 2016. Expert elicitation as a method for exploring illegal harvest and trade of wild meat over large spatial scales. Oryx 51(2):298-304. https://doi.org/10.1017/S0030605315001167

Tacoli, C., and J. Agergaard. 2017. Urbanisation, rural transformations and food systems: the role of small towns. IIED, London, UK.

Tendall, D. M., J. Joerin, B. Kopainsky, P. Edwards, A. Shreck, Q. B. Le, P. Kruetli, M. Grant, and J. Six. 2015. Food system resilience: Defining the concept. Global Food Security 6:17-23. https://doi.org/10.1016/j.gfs.2015.08.001

Tobler, M. W., F. Hibert, L. Debeir, and C. Richard-Hansen. 2014. Estimates of density and sustainable harvest of the lowland tapir Tapirus terrestris in the Amazon of French Guiana using a Bayesian spatially explicit capture-recapture model. Oryx 48:410-419. https://doi.org/10.1017/S0030605312001652

United Nations. 2019. World Urbanization Prospects: The 2018 Revision (ST/ESA/SER.A/420). United Nations, Population Division, New York, NY, USA. https://population.un.org/wup/publications/Files/WUP2018-Report.pdf

Tregidgo, D., J. Barlow, P. S. Pompeu, and L. Parry. 2020. Tough fishing and severe seasonal food insecurity in Amazonian flooded forests. People and Nature 2(2):468-482. https://doi.org/10.1002/pan3.10086

Tregidgo, D., L. Parry, J. Barlow, and P. S. Pompeu. 2021. Urban market amplifies strong species selectivity in Amazonian artisanal fisheries. Neotropical Ichthyology 19(3). https://doi.org/10.1590/1982-0224-2021-0097

van Vliet, N., C. Nebesse, S. Gambalemoke, D. Akaibe and R. Nasi. 2012. The bushmeat market in Kisangani, Democratic Republic of Congo: implications for conservation and food security. Oryx 46:196-203. https://doi.org/10.1017/S0030605311000202

van Vliet, N., J. Muhindo, J. K. Nyumu, and R. Nasi. 2019. From the Forest to the Dish: A Comprehensive Study of the Wildmeat Value Chain in Yangambi, Democratic Republic of Congo. Frontiers in Ecology and Evolution 7:132. https://doi.org/10.3389/fevo.2019.00132

van Vliet, N., M. P. Quiceno, D. Cruz, L. J. N. De Aquino, B. Yagüe, T. Schor, S. Hernandez, and R. Nasi. 2015. Bushmeat networks link the forest to urban areas in the trifrontier region between Brazil, Colombia, and Peru. Ecology and Society 20(3):21. https://doi.org/10.5751/ES-07782-200321

van Vliet, N., R. Nasi, and A. Taber. 2011. From the Forest to the Stomach: Bushmeat Consumption from Rural to Urban Settings in Central Africa. Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-17983-9_6

Vogt, R. C. 2008. Tartarugas da Amazônia. Gráfica Biblos, Lima, Peru.

Willett, W., J. Rockström, B. Loken, M. Springmann, T. Lang, S. Vermeulen, T. Garnett, D. Tilman, F. Declerck, A. Wood, M. Jonell, M. Clark, L. J. Gordon, J. Fanzo, C. Hawkes, R. Zurayk, J. A. Rivera, W. De Vries, L. Majele Sibanda, A. Afshin, A. Chaudhary, M. Herrero, R. Agustina, F. Branca, A. Lartey, S. Fan, B. Crona, E. Fox, V. Bignet, M. Troell, T. Lindahl, S. Singh, S. E. Cornell, K. Srinath Reddy, S. Narain, S. Nishtar, and C. J. L. Murray. 2019. Food in the Anthropocene: the EAT-Lancet Commission on healthy diets from sustainable food systems. The Lancet 393:447-492. https://doi.org/10.1016/S0140-6736(18)31788-4

Williamson, D. 2002. Wild meat, food security and forest conservation. In S. Mainka, and M. Travedi, editors. Links between biodiversity, conservation, livelihoods and food security: The sustainable use of wild species for meat. IUCN, Gland, Switzerland.

WinklerPrins, A. M. G. A., and P. S. Souza. 2005. Surviving the city: urban home gardens and the economy of affection in the Brazilian Amazon. Journal of Latin American Geography 4:107-126. https://doi.org/10.1353/lag.2005.0033

Copyright © 2023 by the author(s). Published here under license by The Resilience Alliance. This article is under a Creative Commons Attribution 4.0 International License. You may share and adapt the work provided the original author and source are credited, you indicate whether any changes were made, and you include a link to the license.