SciELO - Scientific Electronic Library Online

vol.8 issue3Changing technologies: innovation and (re)production of objects among the Asurini do XinguRecovering macroremains in Amazonian archaeological sites: a new methodological proposal for archaeobotanical studies author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Boletim do Museu Paraense Emílio Goeldi. Ciências Humanas

Print version ISSN 1981-8122

Bol. Mus. Para. Emílio Goeldi. Ciênc. hum. vol.8 no.3 Belém Sept./Dec. 2013 



Contributions of the old urban homegardens for food production and consumption in Rio Claro, Southeastern Brazil


Contribuições dos quintais urbanos antigos na produção e no consumo de alimentos em Rio Claro, Sudeste do Brasil



Mayra Teruya EichembergI; Maria Christina de Mello AmorozoII

IUniversidade do Estado de Santa Catarina. Chapecó, Santa Catarina, Brasil
IIUniversidade Estadual Paulista Júlio de Mesquita Filho. Rio Claro, São Paulo, Brasil





Urban homegardens are green areas of households within the city limits and they have the potential to provide families with a cheap alternative for diet improvement, and to complement the income of the families who sell cultivated products. This research analyzes the contributions of old urban homegardens on food consumption and household economy. Data related to homegardens composition were collected by interviews and by collecting cultivated plants. Diets were assessed through a retrospective method (last 24 hours food recall) and administered every two months, during a year, to include seasonal variations. The diet of the sampled population was found to be dependent on certain foods, indicating a narrow food niche (Levins index = 25.9; Levins standardized index = 0.23). Variations in interviewees' diet are related to the consumption of fruits and vegetables, which are partly supplied by homegardens. Spices and teas consumed were obtained from homegardens, revealing its importance in food consumption and health. Among the 98 species found in homegardens, only 38% appeared in the interviewees' diet, indicating an under-exploitation of these homegardens. Our study found that the main role of homegardens is to supply variation in the diet, contributing to the consumption of different types of products.

Keywords: Diet. Ethnobotany. Food niche. Urban agriculture.


Os quintais urbanos correspondem às áreas verdes domiciliares dentro dos limites urbanos e têm o potencial de fornecer às famílias uma alternativa barata para a melhoria da dieta e complementação da renda familiar, pela venda de itens produzidos nesse espaço. Este estudo analisa a contribuição dos quintais urbanos antigos na alimentação e na economia doméstica. Os dados sobre a composição dos quintais foram obtidos por meio de entrevistas e pela coleta de plantas presentes nos quintais. As dietas foram levantadas pelo método recordatório das últimas 24 horas, aplicado bimensalmente, durante um ano, para incluir variações sazonais. Verificou-se que a dieta da população amostrada é dependente de alguns alimentos, indicando um nicho alimentar estreito (índice de Levins = 25,9; Levins padronizado = 0,23). As variações nas dietas dos informantes estão relacionadas com o consumo de frutas e hortaliças, em parte fornecidas pelo quintal. Temperos e chás consumidos foram obtidos dos quintais, revelando sua importância para alimentação e saúde. Das 98 espécies alimentares encontradas nos quintais, apenas 38% apareceram na dieta dos entrevistados, indicando uma subexploração dos quintais. Em nosso estudo, a principal função relatada para os quintais foi proporcionar variações da dieta, contribuindo para o consumo de itens diversificados.

Palavras-chave: Dieta. Etnobotânica. Nicho alimentar. Agricultura urbana.




Homegardens can be defined as areas associated to households where small-scale mixed production helps residents meet food, aesthetic, medical and leisure needs (Zaldivar et al., 2002; Wezel and Bender, 2003; Kumar and Nair, 2004; Amaral and Guarim Neto, 2008). Their richness and diversity is strongly related to their age - old gardens are richer and more diverse, as new species are added along the time (Coomes and Ban, 2004; Pinho et al., 2011). Homegardens also represent a transition between rural and urban areas and between traditional and modern lifestyles that associates agrobiodiversity with the experience and knowledge transmitted by local traditions (Winklerprins, 2002; Semedo and Barbosa, 2007). This continuum between rural and urban areas is revealed by the fact that homegardens are a site of food production and conservation of the agrobiodiversity (Florentino et al., 2007; Kabir and Webb, 2008; Huai and Hamilton, 2009); a social network of the donation, exchange and sale of products, and a heritage endowed with aesthetic and social value (Winklerprins, 2002).

Urban homegardens are a type of urban agriculture (UA) where growing "crops and raising small livestock or milk cows for self-consumption or sale in neighborhood markets" (FAO, n.d.) take place within small areas in the city and offer availability of fresh, perishable food and non-market access to food for poor consumers. In fact, the production in these places can guarantee food to about millions of poor people, contributing for food security (FAO, n.d.).

Studies on urban homegardens have primarily emphasized their potential for food security. Because homegardens represent a source of nutrition, particularly from fruits and vegetables, and generate income through the sale of surplus products, they contribute to improving food security and household economy by reducing expenses and increasing income (Niñez, 1984; Rico-Gray et al., 1990; Padoch and De Jong, 1991; Caballero, 1992; Brito and Coelho, 2000; Kumar and Nair, 2004). Other studies have focused on the importance of growing condiments and medicinal plants (Amorozo, 1981; Da Silva and Silva, 1995; Murrieta, 2001).

In urban areas, homegardens can promote the spread of plant germplasm and play an important role in the in situ conservation of native and exotic plants from diverse vegetation communities (Amaral and Guarim Neto, 2008; Albuquerque et al., 2005; Eichemberg et al., 2009), particularly those rarely found in markets and stores, which provide different food sources of vitamins and nutrients as well as alternative medicinal resources (Amorozo, 1981; Madaleno, 2000; Semedo and Barbosa, 2007).

Based on these general considerations, this study assesses how these old urban homegardens in Rio Claro, Brazil, complement diets and generate additional income. It thus characterizes how the sampled homegardens contribute for food production and household economy.




Rio Claro is located in the center of the state of São Paulo (22º 24' S; 47º 31' W), Southeastern Brazil, 180 km from the state capital, São Paulo. The climate is subtropical, with two well-defined seasons (winter is the dry season); temperatures in the coldest months vary from 3 ºC to 18 ºC and presents the warmest month average temperature over 22 ºC. In this municipality, large scale industrialized agriculture (sugarcane and orange) is an important economic activity. Smaller crops include corn and rice. Cotton is also cultivated. The population of Rio Claro was 168,087 in 2002 of which 96% live in urban areas (Brasil em números, 2002).


Based on a map of the history of the urban space in Rio Claro (Troppmair, 1992), seven neighborhoods formed around 1945 were selected, making up 380 squares or city blocks, under the assumption that older homes with preserved gardens would be found in these neighborhoods. Sixty city blocks were randomly selected using a standard table of random numbers. Field observations revealed which dwellings had old homegardens. Only 22 buildings satisfied the criteria for the study: the presence of tree and shrub species that had been maintained for 25 years or more and had inhabitants over 50 years old. During the first two months of field work, five of the households covered their homegardens with cement or cut all the trees. This decision was made mainly because of the difficulty of keeping up the gardens by owners of advanced age, a lack of time to devote to activities in the backyard and a lack of interest among the younger generations. Thus, the final sample was reduced to 17 households.

Vouchers of collected specimens were deposited in the Herbarium of the Department of Botany (HRCB), Instituto de Biociências, Universidade Estadual Paulista Júlio de Mesquita Filho (UNESP), Rio Claro. Species identification was made in herbaria by experts, whose names are listed in the end of the article.


All informants agreed to participate in this study, and the consent was oral. The names were absent as a manner of privacy protection. Diets were assessed using the retrospective method, which consists of administering a food survey in which informants reveal what they have eaten in the last 24 hours. This method allows for the characterization of a group's diet and the monitoring of the consumption and origin of their food intake (Dwyer, 1994). The food items surveyed by the retrospective method were classified into six different food groups, following the food pyramid adapted from Welsh et al. (1992) and Shuette et al. (1996). They are: group 1 - cereals, starches and pastas (breads, rice, pastas, flours, cakes, snacks, cereals, and tubercles); group 2 - fruits (all fruits); group 3 - vegetables (all vegetables, except those already mentioned in the first group); group 4 - dairy products (milk, cheeses, and yoghurts); group 5 - meats and sausages (bovine, pork, chicken, fish, egg, and sausages); group 6 - fats, sugars and alcohol (oil, margarine, butter, sugars, candies, and honey).

The information obtained helped measure the food niche of the studied group, providing data on the variation in the diet and on the contribution of food items from the homegarden. These data were analyzed using the Levins index and the Levins standardized index (Krebs, 1989; as discussed by Nascimento et al., 2010). Levins proposed that a means of measuring the amplitude of the niche is in the following numerical relationship (Krebs, 1989):

B = 1[Σpj2]-1

B = Levins index for the niche amplitude;

pj = proportion of individuals who use the resource j.

In this study, pj is the proportion which a given item presents in the diet.

The Levins index value can range from 1 to n, where n corresponds to total number of consumed items. The closer to 1, the more specialist is the group studied, and, if the index gain maximum values, more evenly the population uses resources, i.e., more generalist it is.

In order to facilitate the process of direct comparison of the use of resources, it is common to standardize the values of niche amplitude, expressing them on a scale from 0 to 1 using the Hurlbert formula (Krebs, 1989):

BA = (B - 1) (n - 1)-1

BA = Levins index standardized for the food

niche amplitude;

B = Levins index for food niche amplitude;

n = number of consumed items.

The niche parameter assessed was the frequency of food items present in the diet (Krebs, 1989). In order to assess the importance of homegardens to the owners' budgets information on the frequency of sale of food or other products from the backyards was gathered along with the food surveys.

The food survey was administered during one week every two months from January to November 2002, totaling 42 feeding days and 714 food recall interviews. Our sample included five men and twelve women who were interviewed. All of these participants descended from Western European immigrants who came to Brazil to work on coffee plantations around 1840 and were born in the state of São Paulo. Most of the participants (77.3%) came from Rio Claro region or neighboring areas. The interviewees were either of rural origins or were born in urban areas but also had a rural heritage from their parents and grandparents.

Most interviewees (59%) were retired from the service and trade sectors, and the others were housewives (41%). The monthly family incomes of 29% of the interviewees were in class E (Brasil em números, 2002), and the incomes of 52% of the interviewees were in class D. Only 18% of the interviewees were in class C, the medium-to-low economic class. These data are in line with those found for the southeastern region of Brazil, where about 60% of the families are in class D (Brasil em números, 2002). Among the interviewees, education was low: 9% were illiterate, and 64% had not completed their basic education (one to eight years of schooling).



The homegardens studied varied from 428 m2 to 2,000 m2 and sheltered various plants, both native and exotic. The main uses of the plants were ornamental, as food or as medicine. In the food category, we found 98 species grown in homegardens, mainly fruits and, to a lesser degree, vegetables, for an average of 21 species per homegarden (Eichemberg et al., 2009).

Although the number of food species grown is high, only 28 of these appeared in the diet of the interviewees. Among different items consumed within the 24 hours before interviews, 106 were foods, and 30 were tea, coffee and condiments (136 items in total). This very high number indicates that a wide variety of foods are part of the interviewees' diet. However, if we take into account the frequency with which each of these items was cited, only 10% of the total items were consumed in 25% or more of the food recall interviews (Table 1).

The basic composition of the diet can be summarized into the following foods and frequencies: bread (80.3%), milk (79.4%), rice (73.8%), beans (60.7%) and beef (50.4%), all of which are acquired in stores and markets. The variations found in the interviewees' diet concern the consumption of fruits and vegetables, part of which are produced in homegardens.

The foods mentioned in our surveys were classified into six different food groups, following the food pyramid. The Levins index value for the total breadth of the food niche (Table 2) was 25.9, only 23% of the maximum breadth attainable for 106 food items. Of the 106 food items consumed, 28 (26.4%) were produced in homegardens. The most frequently homegrown food group was fruits. Among the 26 types of fruit consumed, 15 were grown in homegardens, accounting for 34% of the total frequency of consumption in that group. Approximately 41% of the times these 15 items were consumed, they came from homegardens (Table 3).

The consumption of homegarden fruit occurs as occasional snacks, either through the direct ingestion of fruit or in the form of smoothies or juice. Lemons, 'acerolas' (Malpighia emarginata) and oranges were among the favorites. 'Acerolas' were exclusively obtained from homegardens, and lemons were purchased only two of the 17 times they were consumed. 'Mamões' (Carica papaya), despite their low frequency of consumption, came from homegardens and were consumed whenever in season. They were common items in homegardens perhaps because they do not require special care; 'mamão' trees grow quickly and are easy to plant. 'Laranja' (Citrus sinensis) and 'banana' (Musa sp.) were the most consumed foods in this group, but they came from homegardens: 20% and 14% of the time, respectively. In addition to the fruits already mentioned, others were consumed in smaller proportions, among which 'jaboticaba' (Myrciaria cauliflora), 'ameixa japonesa' (Eriobotrya japonica), 'amora' (Morus nigra), 'goiaba' (Psidium guajava), and 'pinha' (Annona squamosa) came solely from homegardens. Several species of fruits grown in homegardens, such as 'goiabas', 'bananas', 'figos' (Ficus carica), 'peras' (Pyrus comunis), and 'mamões' are also made into jam. Others are widely used as ingredients to produce liqueurs, including 'jaboticaba' and 'acerola'.

Plant resources also appear to play an important role in the nutritional quality of the interviewees' diets. Vegetables are important elements in salads and to compose side dishes. In this group, among the 25 food items consumed, 11 came from homegardens, accounting for 23% of the total frequency of vegetable consumption. Vegetables came from homegardens approximately 51% of the times they were consumed (Table 3). All of the 'couve' (Brassica oleracea var. acephala) consumed, more than 60% of the 'abóbora' (Cucurbita moschata) and 'almeirão' (Cichorium intybus), 33% of the 'rúcula' (Eruca sativa), and 22% of 'abobrinha' (Cucurbita pepo) came from homegardens. Although often consumed, lettuce was bought in stores, despite the interest some interviewees showed in cultivating it.

The most consumed condiments from the old urban homegardens of Rio Claro were 'salsinha' (Petroselinum sativum), 'cebolinha' (Allium fistulosum), 'manjericão' (Ocimum basilicum), 'manjerona' (Origanum majorana), and 'alfavaca' (Ocimum sp.) (Table 4). Other items, such as 'louro' (Laurus nobilis) and 'pimenta' (Capsicum baccatum), were consumed less often.

Plants primarily used in herbal teas had a special place in homegardens and in the diets of the interviewees (Table 4). Teas can be prepared with one or more plants. This study shows that 'soothing teas' can be made with 'poejo' (Mentha pulegium), 'erva cidreira' (Cymbopogon citratus), 'solevante' (Mentha citrata), 'melissa' (Lippia alba), and 'hortelã' (Mentha sp.), together or individually, through infusion. Similarly, the "teas good for kidneys" used leaves of 'palitária' (Parietaria officinalis) along with 'salsinha' roots and 'quebra-pedra' (Phyllanthus tenellus), or only with leaves of 'cana-do-brejo' (Costus spiralis).

The sampled population presented seasonal variations in their consumption of foods obtained from homegardens during our survey (Figure 1). Although fruit consumption showed no seasonal variation, except for a small decrease in September, vegetable consumption was lower in May and July, which are typically dry months. Spice consumption also decreased in May and July, whereas the frequency of spice consumption increased in January.



This study found that ten food items produced in homegardens were sold. In 76% of these cases, they were vegetables or condiments: 'couve', 'cebolinha' and 'salsinha'. Among the fruit group only 'laranjas' and 'jaboticabas' were sold on three occasions. We observed that 52 items - 21 herbal teas and condiments, 19 fruits and 11 vegetables - produced in homegardens were donated to neighbors and/or relatives. On average, 19 items were donated every two months: eight condiments and teas, six fruits and three vegetables, particularly 'cheiro verde' ('cebolinha' and 'salsinha') and 'couve', which accounted for 28% of all the donated items. Fruits rich in vitamin C, such as 'limão' and 'acerola', were donated 17 and 11 times, respectively.

In the months we administered our surveys, 27 items were planted: 15 species of vegetables, seven fruits, four condiments and teas, and one flower species. The groups with the highest frequency of planted items were vegetables at 46% and condiments and teas at 21% of the total frequency. Only two fruit trees were cut because they were old and did not produce anymore, and one persimmon was replaced by another tree of the same species.



The Levins index shows that although the food items registered were numerous, only a few foods made up the staple diet of our subjects. Among the food groups, a low value of the Levins index was also observed, indicating a narrow food niche dependent on certain items. The production of food items in homegardens, particularly fruits, contributes to the