Published by City Farmer, Canada's Office of Urban Agriculture


An Agro-Silvo-Pastoral System In The South-East Of Mexico City, And Policy Considerations For Peri-Urban Agriculture.

Losada, H., R. Bennett1, P. Zavaleta 2, F. De Leon2, J. Vieyra, J. Rivera and J. Cortés.

Animal Production Systems Area, Department of Biology of Reproduction, Division of Biological and Health Sciences. Universidad Autónoma Metropolitana- Iztapalapa. Av. Michoac án y la Purisíma. Col. Vicentina. Iztapalapa C.P. 09340 Mexico D.F. 1. Researcher from U.K. sponsored by the interchange CONACYT-British Council. 2. Area of Environment of Agricultural Systems. Department of Agricultural and Animal Production. Division of Biological and Health Sciences. Universidad Autónoma Metropolitana- Xochimilco. Av. del Hueso 11000. Col. Villaquietud. Tlalpan C.P. 04960. México D.F.

Telephone: 00 525 724 4701 ext 2532; Fax 00525
724 4930;
Correspondance: To H.Losada, as above


Given the increased recognition and enthusiasm for urban agriculture, an aspect of interest are the appropriate policy responses required for its support. Production activities in peri-urban areas in general, may be subject to a range of policies and programmes. Focussing on the mountainous area to the south-east of Mexico City, policies related to agriculture, urban development and environment are outlined, and criticised for assuming that the aims of each are mutually exclusive. The paper then describes the integration of maize cropping, sheep grazing and forest related activities in the zone, constituting an agro-silvo-pastoral system. Vegetation surveys and soil analyses were carried out in areas representative of the agro-silvo-pastoral system, and the results are presented along with data from surveys and interviews with producers in the region. The use of sheep manure to fertilise maize crops, and use of crop and vegetable wastes as food supplement for sheep grazed in natural pastures, constitute an efficient use of natural resources. Preliminary data from the vegetation survey results suggest that grazing prevents the development of a dense shrub layer in the forest, and a diverse herb strata is facilitated by the seasonal cropping of maize. Greatest soil organic matter content was found in agricultural spaces, closely followed by pastoral areas, suggesting the effects of incorporation of animal excreta. The discussion makes reference to the efficient use of space and natural resources, which is not reflected in the policies and programmes which impact on the area. As a means to open up discussion regarding policy responses in peri-urban areas in general, examples are taken of landscape conservation policies such as those applied in similar areas of Europe. While maybe not being directly applicable to this context, they do show a recognition of the integration of agricultural, environmental and urban planning aims which seems to be appropriate in the case of this peri-urban agro-silvo-pastoral system.

Key words: peri-urban, agro-silvo-pastoral, soil analyses, vegetation survey, policy considerations.


The recent interest in 'urban agriculture' in general, has made much of the social and resource use contributions that such activities can make towards sustainable development in urban areas(Smit and Nasser, Ellis and Sumberg 1998). In the case of peri-urban areas, an interesting theme is the effects of production activities on landscape, particularly given the distinction that is made between rural and urban zones in planning and policy. As in many cities in so-called less developed countries, Mexico City experienced a very rapid and uncontrolled growth in population and urban area in the second half of the twentieth century, such that urban planning and policy has been a step behind the reality, and imposed distinct classifications on land use, ('urban' or 'rural'), in generally unfruitful attempts to manage urban growth (Pezzoli 1998). This type of basic distinction between 'natural' and 'social' is increasingly seen as problematic in itself (Leff and Carabias 1993, Pezzoli, 1998), and interest in the apparently paradoxical phenomenon of urban agriculture illustrates this. It seems appropriate then to examine the nature of productive activities carried out in urban related contexts, as well as the policies and planning contexts within which they function. Following the description of an agrosilvopastoral system in the south west of the Ajusco zone of Mexico City by De Lucas et al (1993), the aim of this work was to use data from two previously published works on sheep and maize production systems (Losada et al 1996a and 1997), together with data gathered on soil and vegetation characteristics, in order to describe their integration in the southeast of the city. The policy and planning approaches to the management of similar agroecosystems in Europe are discussed with reference to the potential for their application in peri-urban areas in less developed countries.

Agroecological features

The southern hills of Mexico City fall within three delgations (political administrative structures) of the Federal District: Xochimilco, Milpa Alta and Tlahuac. Some 80% of the 496km2 area is considered mountainous. The mean altitude above sea level is 2570m (range of 2420-2750m) and gradients are from 20 to 60%. The climate is classified as 'sub-humid temperate' with low humidity (Garcia, 1989), and whilst there is a mean annual temperature of 160C, within a range of 8 to 46oC, frosts occasionally occur on the highest slopes in December and January. Seasonal streams are generated between the months of May and October, from 600-800mm of annual rainfall. The erosion prone, well drained, brown soils are volcanic in origin, and mostly between 25 and 50cm in depth, although there exist some deeper than 50cm. Soil texture ranges from medium clay to medium sandy in the highest areas.

Governmental Policies and Programmes in the Southern Hills of Mexico City.

Three strands of policies and programmes which impact on land use in the zone, are those related to urban development, agriculture, and the environment. As regards the first, the forested zone to the south of Mexico City, and the campesino livelihoods carried out within and around it, constitute a productive green belt, bordering the urban development and expansion which has been considerable since the 1950's. This role has been recognised by the government, as shown in the 1980 General Plan for Urban Development, which created a zonal classification of the city, consisting of urban, 'buffer' and ecological conservation (Pezzoli, 1998). Nevertheless, this policy has proved to be inadequate in its objective to constrain urban growth, as it attempts to 'plan' a solution to problems which have structural causes (demographic growth, immigration effects of macro-economic policies). In the case of agricultural policies and programmes, governmental support has followed a modern, technological approach (S ánchez, 1982). It has promoted dependance on external inputs, and reduction of labour inputs, via subsidised agrochemicals and machinery, and supported intensive, mechanised livestock production systems. Thirdly, a key element of the governments response to environmental concerns is the promotion of forestry; specifically, as an alternative to support for maize cultivation following 'free-trade' conditions placed upon the national staple crop (Encinas et al 1992). The overall effect of these three strands is that agricultural activities in this peri-urban zone are effectively discouraged, whether as human activities in the 'ecological' zone, 'inefficient' systems of agricultural production, or interfering with forest ecology. These approaches are infused by a fragmented perspective which sees the aims of agricultural development, urban planning and environmental conservation as mutually exclusive. Reliance on individual scientific disciplines for policy direction has been the basis for the long held myth of peasant environmental degradation (see Battersbury et al, 1997), and this is reflected in programmes in the south of Mexico City (for example, Grether, (1992) "Technical Proposal for Sustainable Regional Development in the South of the Federal District", claiming that campesino activities are responsible for resource degradation). Certainly, in view of the increased interest in urban and peri-urban agriculture, such claims need to be evaluated with detailed site specific research which takes into account the interdependence of social and natural processes in resource use, and upon which policy responses may be considered.


Data regarding sheep and maize production were obtained from previously published papers (Losada et al 1996a and 1997). The method used was application of questionnaires, tried extensively in the field before final application, and interviews with producers working with sheep, maize and in the forest. A total of 130 questionnaires were applied, 30 regarding sheep, and the remainder concerning maize, in a study comprising 13 towns, in which activities relevant to the agro-silvo-pastoral system are carried out.

The second part of the study involved journeys into the forest, where areas subject to 3 different management regimes were identified:

Components of the Agro-silvo-pastoral system

According to the literature, definition of an agro-silvo-pastoral systems tends to require several activities to be carried out in the same physical space (S ánchez, 1995, Kosarik 1997, Leng 1998), and their value lies in the possibility of reducing external inputs by using tree foliage as a source of dry matter and protein for ruminants, and encouraging soil nitrogen fixation (Speedy and Pugliese, 1992). In the mountainous zone to the south of Mexico City, the system is formed by the linked production processes of the latter three geographical spaces identified by Rivera et al (1998): Firstly, in the immediate surroundings of the town of Milpa Alta, land is appropriated by means of animal and crop production processes highly dependant on man, such as Nopal cultivation; secondly, there is a transitional border dominated by maize production in which sheep are periodically grazed, whilst the third and fourth spaces are summer pasture and forest. Table 1 shows the altitude of areas subject to soil and vegetation survey.

Table 1
Table showing altitude of land use zones in Milpa Alta subject to soil and vegetation surveys.
Land UseAltitude(masl)Soil and Vegetation Survey
Town . No
Nopal vegetable fields . No
Maize fields 2600 Yes
Reforested zone 2900 Yes
Forest with sheep pasture 3300 Yes


Table 2 shows the number and type of sheep owned by producers interviewed.

Table 2:
Number, breeds and type of sheep maintained per producer.

Type of animalTotal%Ave.per producer
Rams 66 4 2
Ewes 729 52 24
Female lambs (7-8 months of age) 198 13 6
Lambs 429 28 13
Castrated males 33 2 1
Total * 1518 100 46
*Breeds: %
Criollo 64%
Commercial Suffolk 21%
Commercial Merino 15%

The mean flock size was 46 animals, with ewes being most common, followed by lambs, and lastly, rams, although the mean size was skewed by the presence of a flock of 300 owned by one producer. Most sheep were criollo, the rest being Commercial Suffolk and Merine. At night, most producers kept sheep in their backyard, though some leave them in the fields. Manure is always collected, dried and used to fertilise field crops, and in some cases, a mobile pen is used to confine sheep for up to 16 hours a day, in fields later to be planted with maize: The pen remains in the same place for around 2 days, until there is a 4cm depth of manure covering the soil. Grazing is carried out in non-conventional areas such as roadside verges and stream banks, as well as in and around the forest. In these latter zones, grass species are mostly native, and a minority of farmers use improved pastures of introduced species (Ryegrass). Supplementary food sources are straw, hay, maize stubble, fresh lucerne and tortilla, and there is little use of commercial concentrates. The feeding regime for the sheep is shown in table 3.

Table 3
Feeding regime for the sheep in the areas studied.

Frequency (%)
Source Crop residues (when available)
Grazing: Roadside verger and steam banks
Forest and neighbouring areas*
*Types of grass found Introduced (Rye grass)
Native grass**
Supplementary feeds Tortilla***
Oat straw
Commercial concentrate
Maize stubble
Fresh lucerne

**Festuca tolucensis HBK, Muhlenbergia macroura HBK, Hitc and Agrostis perennans Walt

***Mexican bread made of maize


Maize is of the native 'chalqueño' type, with family plots generally being 100-100m2 in area. The general characteristics of the crop are shown in table 3.

Table 3.
General Characteristics of Maize Cultivation
Criteria Characteristic %
Form of cultivation Monocrop
Grown in association with one or more other crops
Method for obtaining seed Purchased
Selected from previous years
Seed selection critieria Size
Size and colour

Seasonal Calendar of Agro-silvo-pastoral System.

Figure 1 shows the principal activities of the agro-silvo-pastoral system at different times of the year.

Figure 1:
Seasonal Calendar of activities carried out in the agro-silvo-pastoral system.

JAN                      MAR                     JUNE                      SEP                      DEC

SEASON               DRY SEASON           R A I N Y           S E A S O N                 C O O L

LAMBING/SHEARING                             SHEARING

MAIZE PLOUGH+SOW                                           M A I Z E H A R V E S T I N G

             GRASS BURNING               FUNGI COLLECTION               FIREWOOD COLL.+SALE

           R E S I N S     C O L L E C T E D     A L L     Y E A R     R O U N D

           S H E E P     M E A T     S O L D     A L L     Y E A R     R O U N D


The activities relating to pasture, sheep, forest and maize production occur throughout the year as determined by the seasonal variations in ecological and climatic conditions. For lambing, sheep need adequate energy reserves for both the pregnancy and lactation, which are built up from the grass available during the rainy season. Also, reduction in photoperiod precedes the advance of winter months before the dry season which is less disease ridden, hence lambing occurs from January to March. Soil preparation and sowing of maize occur at the start of the rainy season, permitting greater grass growth and dry matter yield, and harvesting of the cobs, and, later, grain, is carried out during the period August to December. Collection of fungi occurs during the rainy season, when humidity favours fungal growth. The burning of grasses in the forest occurs during the dry months, in order to remove woody stems from pasture, and possibly to break the dormancy of seeds for growth the forthcoming rainy season. Activities such as gathering of wood are carried out throughout the calendar, with the exception of dry branches or trunks, while collection of resins ('ocote') occurs during the winter and dry seasons.

Vegetation, soil and forestry activities

Table 4 presents the results from vegetation surveys and soil analysis.

Table 4.
Main characteristics of the three sampled zones subjected to different management regimes in the south-east hills of Mexico City.

number of Species,
Variables Management Pattern . .
. Livestock Reforestation Arable land
Human action, Bianual burning, Replanted young trees five years after burning, Anual crops during the cycle of spring and summer,
Diversity of tree stratum, 1 2 5
Density of tree Stratum, trees/100 Square m. 4.8 Pine= 1
Aile= 1.2
Cover of the tree estratum, % 22 Pine= 48
Aile= 17
Diversity of shrub stratum,species, 5 3 7
Density of shrub stratum,individuals/ 100 square m. 4.8 149 28
Cover of shrub stratum, % 13 5.3 6
Diversity of herbaceous stratum, Number of species, 3 2 15
Density of herbaceous stratum, Per 100 square m. 270 bunches 141 bunches 5 individuals
Cover of herb áceous Stratum, % 75.0 54.0 20.0
Depth of surface layer of soil,cm 14.6 15.2 10.0
OM content,%* 12.18 11.8 12.45
Nitrates,ppm** 357.0 316.0 321.0
Mineral Content, ppm, 252.2 261.9 225.0
Altitude, masl 3 300.0 2 900.0 2 600.0

*Determined using method of Walkley Black (1946)
**Determined using method of Hommer and Parker (1988).

Based on the sample sizes used, few conclusions can be made from the data. Nevertheless, some tendencies can be seen which can be related to the management regimes imposed. Arable areas were found to have little vegetative cover, consisting of shrubs towards the boundaries, and fast growing grasses and wildflowers that develop out of the crop cycle, before competitive species become dominant and reducing diversity. Reforested areas, in which there is no grazing, had a tree stratum, as well as a dense layer of young shrubs though with small cover, and a herbaceous strata. The area of forest in which sheep were grazed had an adult tree stratum, sparse but well developed shrubbery, and a dense herbaceous layer of a few competitive species which flourish under the grazing regime. The reforested area showed greatest soil depth, followed by the pasture and cropped land, which may reflect increasing disturbance intensity although the differences are not significant. The same pattern is observed in mineral content, but reversed in the case of soil organic matter, which would reflect the incorporation of sheep manure in cropped lands and on the forest floor.

Effects of agro-silvo-pastoral activities on landscape

A biodiverse seasonal herb strata is facilitated by the cropping of maize, and grazing, while allowing a few shrubs to mature and provide cover, may prevent the development of a dense shrub layer, such as found in the ungrazed reforested areas. Pinto (1993) states that a dense shrub strata increases the risk of the spread of forest fire, so it may be proposed that the producers fulfill a landscape maintenance role in the southern hills of Mexico City, and this type of system appears to be a regional characteristic, following the description of an agro-silvo-pastoral system in the Southwestern hills by De Lucas et al (1993). The system has been created according the availability of local resources, both natural and from the household. The use of household and crop wastes to supplement the diet of the sheep, low use of commercial concentrates, as well as the use of sheep manure to fertilise maize crops, are examples of the close integration of activities in this system, and its efficient cycling of resources, as commonly found in urban agricultural production (Losada et al 1996c) in Mexico City. Figure 2 depicts the cycling of materials and functions between components of the agro-silvo-pastoral system:

Figure 2. Functions and interactions between components of agro-silvo-pastoral system.

                     Provide meat and wool

                    Consume household waste and

                      crop residues



Fertilise soil

Graze shrubs

Fertilise soil





Fungi, firewood and kindling

Food source

Research and policy considerations

In view of the role played by the campesinos in facilitating the agro-silvo-pastoral system, which the data suggest prevents scrub proliferation and potential fire risk, maintains soil organic matter, and encourages floristic diversity, it would appear wise for policies to acknowledge this role. On the contrary, policies have favoured intensive production, and accused campesinos of being responsible for forest degradation (Grether, 1992), though ignoring other activities liable to cause degradation of the forest such as collection of firewood and leaf mulch for urban middle class homes, in turn encouraging pests such as Dendroctonus adjuntus and Niodoprion spp. It is worthwhile to consider alternative approaches to policy in similar contexts, and consider their applicability in this (and other) peri-urban agricultural areas. Pinto (1993) describes the comparable 'montado' agro-silvo-pastoral system in southern Portugal, in which the activities of the rural population are necessary to prevent degradation of the landscape. In that case, support by authorities for rural activities to counter the depopulating effects of urban growth, is lent theoretical support from the ecological principle of disturbance in landscape maintenance (Green 1996). Support for rural upland populations was incorporated into countryside planning in the 1990's in Britain (Gilg, 1990), seeking to diversify activities in the face of the production oriented Common Agricultural Policy which disfavours marginal regions. There may be room to consider such 'post modern' landscape maintenance approaches, rather than the 'modernist' production oriented policies currently applied in Mexico City. At present, the narrow perspective of the activities of campesinos, in which livestock and arable production and forest conservation are seen as mutually incompatible, neither recognises the needs of the campesinos, nor reflects current countryside management thinking in developed countries. Since the agro-silvo-pastoral system provides rural based livelihoods for the population, in a region where urban expansion since the 1940's has brought enormous social and environmental costs, a re-orientation of policies would seem appropriate. More extensive research would be necessary, both in Mexico City and in peri-urban systems around the world, in order to formulate appropriate policy responses, but this research should not arrive at simplistic discriminatory conclusions by dissociating ecological processes from the actions and social contexts of the people who work within them.


The authors wish to thank the producers of sheep, maize and forest for information and allowing us to measure some parameters of vegetation and soil, the authorities of the Universidad Autónoma Metropolitana (Autonomous Metropolitan University) for facilities given to the research; and the students of Animal Production and Agronomy options who helped with the field work.


Alatorre, F G., 1993. Que es la agricultura sustentable?. Fundación Friedrich Ebert. México.

Daniels, W., 1984. Bioestadística. Ed. Limusa, México.

Daubenmire, R., 1968. Ecology of fire in grasslands. Avances en reservas ecológicas. V.S. Londres, pp209-254.

De Lucas, T J., Arbiza, A S., y Martínez, L P., 1993». Los sistemas trashumantes de producción ovina en Xalatlaco, Edo. De México. I Descripción. Memorias. VI Congraso Nacional de Producción Ovina. SLP. México. 251p

Ellis, F.,Sumberg, J.1998 Food Production, Urban Areas and Policy Responses. World Development. Vol.26,No2,pp213-225

Fenn, M E., 1996. De la limitación del nitrógeno al exceso: saturación de nitrógeno en bosques de NorteAmérica. II Simposio Internacional y III Reunión Nacional sobre Agricultura Sostenible: Una contribución al desarrollo agrícola integral. Memoria resœmenes. CP. UACh. México p4.

Firebaugh, F M., 1990. Sustainable agricultural systems: a concluding view. In: Sustainable agricultural systems. Soil and Conservation Society. USA. 664p.

Garcia, E., 1989. Modificaciones al sistema de clasificación de Kooepen. UNAM. México.

Green, B., 1996. Countryside Conservation. 3rd Edition. E& FN Spon.UK.

Grether, R., 1997. Propuesta técnica del programa de desarrollo regional sustentable de la

Harwood, R., 1990. A history of sustainable agriculture. In: Sustainable agricultural systems.Soil and Conservation Society. USA. p3.

Hommer, D Ch. and Parker, F P., 1988. Métodos de an álisis para suelos, plantas y aguas. Trillas 6» imp. México. p172.

Kosarik, M., 1997. La agroforestería en Argentina. FAO, Red Latinoamericana de Cooperación Técnica en Sistemas Agroforestales, Santiago, Chile, 50p.

Leff, E., Carabias, J.(coordinadores) 1993. Cultura y Manejo Sustentable de los Recursos Naturales. Vol1. Grupo Editorial Miguel Angel Porrua.

Leng, R A., 1998. Tree foliages, their role in ruminant nutrition. FAO Animal Production and Health Paper, Rome 105p (in press).

Losada H, Neale, M.,Vieyra, J.,Rivera, J., and Cortés, J., 1996b. Sheep management in the region of Xochimilco for supplying benefits to the local population. LRRD. 8:3, p6.

Losada H, Grande,D.,Vieyra, J., Arias, L.,Pealing, R., Rangel, J., and Fierro, A., 1996c. A sub-urban agro-ecosystem of nopal-vegetable production based in the intensive use of dairy cattle manure in the southeast hills of Mexico City. LRRD. 8:4 p66.

Losada H,Vieyra, J., Rangel, J., Zamudio, A., and Martínez, G., 1997d. The sub-urban agro-ecosystem of maize production in the south-east hills of Mexico City as an important cultural and sustainable crop resource. LRRD. 9:5 p1.

Pinto T,C., 1993. Threatened landscapein Alentejo, Portugal: the 'montado' and other agrosilvopastoral systems. Landscape and Urban Planning 24:43-48

Rivera ,J., Bennett,R., Rangel,J., Losada,H., Vieyra,J., Cortés, J., and Arias, L., 1998. Formas de uso del suelo en la comunidad peri-urbana de Villa Milpa Alta. Memorias.Seminario Mesoamericano sobre agrodiversidad en la agricultura campesina. CICA, FCA. UAEM. México. p112

S ánchez,M D., 1995. Integration of livestock with perennial crops. World Animal Review 82(1).

SEMARNAP.,1998 Programa de Reforestación de ēreas Forestales Afectadas por Incendios 1998. URL:

Smit, J.,Nasr, J. 1992 Urban Agriculture for sustainable cities: Using wastes and Idle land and water bodies as resources. Environment and Urbanisation 4 (2) 141-152.

Speedy, A., Pugliese, P L.,(Eds)1992. Legumes trees and other fodder trees as protein sources for livestock. A. FAO Animal Production and Health Paper 102, Rome.

Walkley-Black 1946 In: SARH.1978. Métodos para an álisis fisico químico de suelos, aguas y plantas. 2a ed. México. p201.

Search Our Site[new]

pointer Return to Contents' Page pointer

Revised Thursday, June 11, 1999

Published by City Farmer
Canada's Office of Urban Agriculture