Agricultural intensification

The concept of agricultural intensification describes the changes in agroeeosystem structure, functions, management practices and purposes during evolution and change from "traditional" to "modernist" forms. The change most directly related to our present concerns is an increasing specialisation in the plant and livestock species that are cultivated (Figure 11.1). Specialisation is associated in the modernist mind with increased efficiency of production, although this is scientifically questionable. For instance, Rappaport (1971) showed that a multi-species home-garden was energetically more efficient than monoculture. Decrease in specics richness is commonly accompanied by increased utilisation of the same area of land and by intensified management intervention, resulting in greater pressure on both labour and natural resources. These resources are then substituted by externally purchased inputs of industrial origin such as fertiliser, pesticides and petrochemical energy. This results in a shift from internal (largely biological) control of agroeeosystem function to external (largely economic) regulation.

These later stages of agricultural development are therefore made possible by linkage with markets, and have often been accompanied by substantial subsidy from other sectors of the national or regional economy. This tends to obfuscate the second claim of modern agriculture, that it is economically more profitable than natural resource-based systems.

Whilst all these factors are part of agricultural intensification they do not necessarily occur together, and it has proved difficult to classify the great variety of agroecosystems into a few easily described categories. A grouping based loosely on the intensity of land-use and management intervention nonetheless provides a useful framework for discussing the relationship between biodiversity and ecosystem function. The following is a brief description of a range of different agroecosystem types along this gradient, including comment on their relative biodiversity. This is most easily described in terms of the diversity of plant (particularly crop) and livestock species, but it is useful to distinguish between this planned biodiversity and the lotal biodiversity, which includes all the associated but non-economically productive plants (e.g. weeds, cover species, agroforestry trees, etc.), animals (including pests and soil fauna) and microbes (e.g. pathogens, symbionts and soil microorganisms).

11.2.2 The variety of agricultural systems

Shifting agriculture Large areas of forest on all three tropical continents have for many centuries been the site of a variety of forms of "shifting agriculture", also known as "slash-and-burn agriculture", and by a variety of local names. The basic principle of this extensive form of agriculture is the alternation of short crop phases with long periods of natural or modified fallow vegetation. Yield is thus managed on a long-term basis, rather fhan by maximisation in the short-term (Ruthenberg 1980; Ramakrishnan 1992). Shifting agriculture systems traditionally maintain diversity in the cropping phase by utilising mixed cropping systems. Within this the economically important crops are largely annuals, the perennial shrubs and trees being separated in time and confined to the fallow regenerative phase of the forest. The number of crop species in the mixture may vary considerably, from six to over 40, depending upon the agricultural cycle and the social and economic background of the community concerned. This land-use system may also include animal components such as poultry and swine. The crop phase alone does not express the total diversity of the system; when the fallow phase is included, the plant species richness may run to several hundred.

Rotational fallow Increased pressure on land or shortage of labour can lead to intensified forms of slash-and-burn agriculture. Shortening of fallow periods and increasing sedentarism commonly results in practices where the same plot of land is cropped on a 3-10 year rotation. The fallow phase in such systems is often dominated by herbaceous "weedy" species which may be slashed and allowed to decompose in the surface layers of the soil before the land is prepared for cultivation. The diversity of this phase is thus much reduced. The cropping intensity may vary from place to place, ranging from a few food crops to highly diverse systems. Leguminous crops often form part of the mixed cropping system, together with traditional cereals such as rice or maize and lesser-known crops of food, medicinal and other value.

Home gardens One of the oldest traditional ways in which humans in the humid tropics imitate nature in their agricultural practices is through incorporation of trees and other perennials as components of an elaborately constructed home garden, a system prevalent in many regions of the world. Home gardens are small plots, usually of 0.5-2 ha, located close to the habitation and fertilised with household wastes. Home gardens have a rich plant species diversity (30-100 species), dominated by woody perennials and structurally stratified (Gliessman 1989; Nair 1989; Ramakrishnan 1992) with a mixture of annuals and perennials of varied habits - herbs, shrubs, trees and vines - as well as more conventional food crops. The farmer obtains food products, firewood, medicinal plants, spices and ornamentals, and some cash income all the year round. Further complexity comes from the common association of the home gardens with traditional animal husbandry systems such as poultry and swine. These self-sustaining systems are ecologically and economically very efficient, but are fast disappearing. As with shifting cultivation, the total biodiversity of these systems is on a par with that of many natural systems.

Compound Farms Traditional farming systems in the humid tropics often comprise a mixture of land-use systems under the control of the same household. The subsystems may range from home gardens through rotational fallow fields to fully sedentary and relatively specialised fields (see below). Such compound farms have been described for West Africa (Okigbo and Greenland 1976) as well as Asia (Ramakrishnan 1992). These systems possess both spatial complexity and high total biodiversity.

Mixed arable-livestock farming Whilst shifting cultivation has been characteristic of the humid tropics, agriculture in the drier savanna zones has traditionally been centred on livestock production. In many parts of Africa there was a strong cultural separation between the cattle people, who derived practically all their needs directly or indirectly (by barter of the products) from livestock, and the sedentary cultivators raising cereals (such as sorghum) and root crops (such as cassava) on a rotational or shifting basis.

In most areas, as political boundaries have changed and land pressures have grown, nomadic cattlc management has declined, and this separation has been replaced by more sedentary systems where livestock and food-crop production are closely integrated. This linkage is both economic and ecological. Cattle provide a resource for arable production in terms of draft power for ploughing and cartage and a source of manure for fertilisation. Nutrients are harvested from the pasture during grazing and transferred and concentrated on the crop fields (Swift et al. 1989; Campbell et al. 1996). This demands continuous management for the cattle and intermittently intensive management for the crop fields, often separated between the genders. These systems are complex in their range of landuse, and diverse in the full range of species utilised in production, particularly those from the grazing subsystems.

intensified agroforestry systems In all the systems described above trees are common components, either retained selectively from the natural vegetation or deliberately planted. In many parts of the world this feature has been further developed to produce more deliberately structured "agroforestry systems". Three broad categories of these systems have been identified based on their structural and functional attributes (Nair 1989): Agrisilviculture is the use of crops and trees, including shrubs or vines, on the same land; Silvopastoralism is a combination of pasture and trees; Agrisiivopastoralism combines food crops, pastures for livestock and trees. Modern research has drawn on these systems to produce a range of intensified agroforestry practices, the most well known of which is "alley farming". Trees are planted closely in rows to form hedgerows and the crops are grown in the "alleys" between them. It is thus similar to the intercropping system except that trees are regularly pruned to provide mulch or fodder and to reduce above-ground competition effects with crops. Extra labour is thus required for these activities, a factor which is critical in determining the acceptance of the system (Kang et al. 1990). The trees used in such systems are often chosen for their multi-purpose nature, but in practice the development of intensified agroforestry has led to a significant narrowing of the tree germplasm utilised in the system.

Intercropping and crop rotation Intensification of landuse may ultimately result in one or more practices of continuous cultivation of food crops. These practices vary in respect of the degree of diversity and complexity utilised. Intercropping is the practice whereby more than one crop is cultivated in the same land area through time. The number of species can be as high as 10-15, but is commonly low (2-6). Nonetheless, this can still represent a risk-spreading investment for the farmer in terms of the range of potential products and ecological strategies it can embrace (Francis 1986).

Species are planted in rows or dispersed randomly or along field margins, depending on the method of cultivation, farmers' needs and uses. A combination of cereals and legumes is common as the "core" of such systems (e.g. sorghum and pigeon pea, or pearl millet and groundnut, in semi-arid India), but cereal-cereal or cereal root crop combinations are also frequent in West Africa.

Crop rotation is a practice of similar intensity where different crops are planted in the same ground at different times, i.e. in sequential cropping seasons. Rotations are a common feature of intensive agriculture in many parts of the world, largely because of the need to avoid the build-up of diseases. The inclusion of a leguminous crop in a cereal rotation is well recognised as a means of fertility improvement, but a strict rotation sequence is seldom maintained in many tropical cropping systems.

Specialised cash-crop systems A common outcome of intensification is the increase in the proportion of specialised fields, some of them devoted to "high value" crops. These often form part of traditional economies, yielding products which can be bartered for other materials. As more structured markets develop, these crops may become important components of the cash economy. The traditional cash crops include a diversity of fruit trees, bananas, ginger, pineapples, yams and special products like broom grass (for broom making) or bamboo (for a variety of purposes). With the coming of the industrial revolution, small-scale plantation crops of rubber, cocoa, oil-palm or coffee were incorporated in farming systems in many parts of the world. These fields often require intensive management at certain times of the year, either in field preparation, pest management, harvest or post-harvest activities. Nonetheless they arc traditionally handled internally, labour coming exclusively from within the family household. Many of these systems also continue to emphasize recycling of organic residues, with minimal dependence upon inorganic fertilizers. The specialised crops are commonly interspersed with other plants. In traditional systems these fields or plantations are often very diverse (e.g. "jungle" rubber in southeast Asia), but modern plantations are kept much more "clean" of other species (see section 11.2.4.).

Modern farming systems Field specialisation in food crops, particularly cereals, typifies much of modern (i.e. post-World War II) agriculture in many parts of the world. These systems represent the ultimate reduction in biodiversity the genetically uniform, continuous cultivation of a monocrop. This form of agriculture relies on mechanised (petrol driven) tillage, crop management and harvest. Soil and pest management are chemically regulated, with consequent effects on the biodiversity of microbes and invertebrate animals both above and below ground (Figure 11.1). This level of intensification includes such systems as the intensive fruit plantations of the tropics, the intensive orchard systems of California and the Mediterranean, intensive cereal production throughout the world, and large-scale vegetable production.

11.2.3 Impact of agricultural intensification on biodiversity

Agroecosystems can be ordered approximately according to the intensity of their management. A generalised gradient might move from unmanaged vegetation (usually a forest or grassland) to "casual" management (including shifting cultivation, home gardens and nomadic pastoralism), to low-intensity management (including traditional compound farms, rotational fallow and savanna mixed farming), to middle-intensity management (including horticulture, pasture mixed farming, and alley farming), to high-intensity management (including crop rotation, multi-cropping, alley cropping and intercropping), and finally to modernism (plantations, orchards, and intensive cereal and vegetable production).

It is generally acknowledged that diversity decreases as habitats change from forest to traditional agriculture to modern agriculture (Altieri 1990; Hoiloway and Stork 1991; Pimentel et ai 1992). If we plot total biodiversity against the points along the intensity gradient, it is thus highly probable that the resulting relationship will be monotonie and decreasing. However, the exact form of the curve is uncertain, and four possible scenarios for the relationship are presented in Figure 11.2. Before considering the shapes of the hypothesised curves, we should note that the specific positioning of levels of intensification on the x-axis is approximate and non-quantitative. Clearly to explore the relationships more rigorously it would be necessary to derive one or a number of quantitative indices for intensification.

Curve I, hypothesising a substantial loss in biodiversity as soon as any human use and management is brought to bear on the ecosystem, probably represents the most commonly held prediction of the relationship. The other extreme. Curve II, in which diversity is only significantly affected under high intensity, would in contrast generally be regarded as less likely. Nonetheless we know of no data or theoretical argument that provides hard evidence for deciding between these extremes. Indeed, the small amount of data available on this issue (see below) might be interpreted as support for a Type II curve. We propose, however, that in most cases something in between these two states will be the pattern. Two intermediate forms of the relationship are represented by Curves III and IV.

Curve III is a "softer" version of the ecologists' expectation, and simply says that after an initial very dramatic loss in biodiversity, the further loss as management intensifies is relatively slight until it reaches the extreme of the truly modern systems. Curve IV is perhaps a more interesting hypothesis.

Unmanaged system (forest, grassland)

Casual management

(shitting cultivation.

nomadic pastoralism,

Low Intensity management (traditional compound farm, rotational fallow, traditional

Middle Intensity management (horticulture, pasture mixed (arming, traditional cash cropping)

High management (crop rotation, multicropplng, alley cropping, intercropping)

Unmanaged system (forest, grassland)

Casual management

(shitting cultivation.

nomadic pastoralism,

Low Intensity management (traditional compound farm, rotational fallow, traditional home gardens) agroforestry)

Middle Intensity management (horticulture, pasture mixed (arming, traditional cash cropping)

High management (crop rotation, multicropplng, alley cropping, intercropping)

Modernism (plantations and orchards, intensive cereal and vegetable production!

Figure 11.2 Hypothetical relationships between agricultural intensification and total agroecosystem biodiversity. Note that the jr-axis is non-quantitative. The four curves illustrate four different scenarios, representing differential effects of agricultural management on total biodiversity and with differing implications for conservation particularly with respect to its implications for biodiversity preservation per se. This case suggests that initiai stages of management have only a minor impact on total biodiversity, and thai further loss is gradual until some rather critical stage of management intensity is reached. In Figure 11.2 this is arbitrarily pictured as located between low intensity and middle intensity, but the critical stage at which biodiversity declines very rapidly might be at any other point along the intensification axis. If this relationship holds, then it would follow that planning activities for biodiversity conservation should be focused on maintaining management intensities below that critical point, rather than aiming at a zero management strategy. This can be considered an interesting application of the intermediate disturbance hypothesis (Connell 1978). Janzen (1973) has suggested, in support of this, that casually managed agroecosystems may actually promote more species diversity than their unmanaged counterparts. For example, forests showing scars of former subsistence agricultural activity seem to have higher species diversity than those in which such intervention is missing.

It is important to remember that the patterns of agricultural change are diverse. It is not, as might be implied by Figure 11.2, that intensification necessarily proceeds from unmanaged forest, through shifting cultivation, to sedentary intercropping, to intensive cereal cropping. Furthermore, there is also no reason why having moved in one direction along the intensification axis that intensification may not be reversed. Agroecosystems at a particular level of intensiveness may indeed be transformed in the opposite direction. For instance, former banana plantations in Costa Rica have recently been converted to small family subsistence production. Thus there is no particular pathway of change suggested in Figure 11.2, rather the simple observation that a qualitative scale of intensity of management can be approximately ordered, and that biodiversity follows as a functional response to that ordering.

Nonetheless, despite the lack of clear evidence for any of the hypothesised curves in Figure 11.2, it is possible to detect some patterns of change in specific examples.

11.2.4 Patterns of change

Impact on the above-ground fauna of intensification in the Central American coffee ecosystem The dynamics of transformation of the coffee agroeco-system in Costa Rica provides an instructive example of the impact of management intensity on biodiversity. As with other major ecosystem transformations in tropical latitudes, the transformation of coffee (Coffea arabica) production involves spectacular landscape changes. At the two extremes of this transformation lie the traditional system and the modern intensive coffee monoculture. The former follows the common pattern of traditional agroforestry, with a variety of shade-tree species, frequently interspersed with fruit trees, sometimes with relatively dense plantings of bananas (Musa sp.) in a forest "canopy" above the coffee bushes. The coffee itself tends to be managed at the level of individual coffee plants, such that pruning creates small light gaps into which cassava (Manihot esculenta), yam (Dioscorea alata) or other annual crops are planted. When a whole group of coffee bushes are to be "renovated" (removed and replanted with new bushes), a larger "light gap" is created and may receive a planting of corn, beans or other light-demanding crops. Thus, traditional coffee farms share many of the structural attributes normally associated with forests.

The new monocultura! system that is being promoted all over Central America (Reynolds 1991; Babbar 1993) could not be more different. All shade trees are eliminated and the traditional coffee varieties are replaced by new sun-tolerant and shorter varieties which are genetically homogeneous. The coffee is pruned either by row or by plot and is heavily dependant on agrochemicals, especially herbicides and fertilizers (ICAFF-MAG 1989).

These two systems represent the two extremes in a continuum of coffee management systems with varying degrees of complexity. The vegctational changes associated with intensification are obvious at the landscape ievel, but the more subtle changes in biodiversity are even more spectacular. Nestel and Dickschen (1990) reported a high diversity of arthropods in the traditional system compared with the unshaded modern one, and Perfecto and co-workers (Perfecto and Snelling 1994; Perfecto and Vandermeer 1994) reported a significant reduction in species diversity of ground-foraging ants as the transformation proceeds. Preliminary samples of the arboreal entomo-fauna, using canopy insecticidal fogging, illustrate the high diversity of insects that can be found in this agroecosystem (Table 11.1). The diversity of Coleoptera and ants (Hymenoptera: Formicidae) in shade trees in coffee plantations is within the same order of magnitude as that reported by others (Erwin and Scott 1980; Adis et al. 1984; Wilson 1987, 1988) for beetles and ants in rainforest trees (Table 11.1). Note that the traditional coffee farms sampled were hundreds of kilometres from anything resembling a tropical rainforest, and were effectively islands in a sea of modern unshaded coffee. While the beetle and ant diversity of the traditional coffee agroecosystem are surprisingly similar to the figures for a natural rainforest, they decline very rapidly once the system is modernized, suggesting an approximation to the Type II or Type IV curve.

impact of land-use change on the soil biota The conversion of natural systems to intensively cultivated monocultures results in a loss of diversity in soil invertebrates and microorganisms. With increasing inputs of energy, water and agrochemicals to maintain production, the functional groups which regulate soil biological processes in natural systems are replaced by mechanical and chemical controls of soil fertility.

The key factors determining total biodiversity in the different agroecosys-tems shown in Figure 11.2 are microclimate, habitat structure and food resources. Hence forest plantations and home gardens retain many of the community characteristics of natural forest, and pasture communities are functionally similar to savannas. The pattern of species losses in the soil biota along the gradient of agricultural development approximates that of Curve IV in Figure 11.2. The point of inflection from more to less diverse communities varies, however, for different sizes and functional groups according to the type of farming system and agricultural practice. In the extreme situation of direct conversion of forest to intensive cultivation of short-rotation crops, there is rapid disappearance of surface-active macro-invertebrates (millipedes, earthworms, beetles) which use leaf litter as both a habitat and a food resource (Lavelle et al. 1994). As the mass of soil and litter organic matter pools is further rcduced, there is a progressive shift in the structure of the community towards small organisms (microarthropods, nematodes, protozoa, bacteria, fungi) occupying soil pores within the buffered soil environment, and larger organisms such as earthworms and termites which can modify soil structure, whilst the diversity of soil mesofauna (mites, collembola) declines with the increasing intensity of tillage from hand-cultivation to ploughing because of the disruption of macrohabitats.

Table 11.1 Number of species of ants (Hymenoptera: Formicidae) and beetles (Coleoptera) collected by insecticidal fogging from the canopy of trees in coffee plantations in the Central Valley of Costa Rica, and tropical forests in Panama, Peru and Brazil


Tree species



No. of species



Luehea seemannii

Moist seasonal forest


Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook

Post a comment