□ Intercropping without legum<

y//X With legumes legume

0.3-0.5 O.S-O.7 0.7-0.9 0.9-1 1 1.1-1 3 1.3-1 5 1.5—1 7

Land equivalent ratio (Land area of sole crop per land area ol intercropping at same yield)

Figure 11.4 The relative yield efficiency of intercrops (two-species mixtures) as compared with the two crops grown separately (from Trenbath 1976)

0.3-0.5 O.S-O.7 0.7-0.9 0.9-1 1 1.1-1 3 1.3-1 5 1.5—1 7

Land equivalent ratio (Land area of sole crop per land area ol intercropping at same yield)

Figure 11.4 The relative yield efficiency of intercrops (two-species mixtures) as compared with the two crops grown separately (from Trenbath 1976)

Principles of resource capture The yield gains obtained from intercropping systems have been extensively researched from the point of view of the biological production function. Agronomists engaged in intercropping research (usually with two species) started by employing well-established ecological techniques and analyses to examine the substantial yield advantages of intercropping over sole cropping (Willey 1981), The underlying assumption is that a single species is unable to fully occupy the ecological niches in a given environment and the addition of one or more species with appropriate traits can lead to a greater total productivity by capturing more light, water and nutrients than a single species. After a decade of systematic research on intercropping at ICRISAT and elsewhere, Willey and coworkers concluded that "probably the most common cause of higher yields from intercropping over sole cropping is the improved use of environmental resources. Put very simp!}', if component crops in an intercropping system use resource differently than when grown together, the crops complement each other and make better overall use of resources than when grown as separate sole crops" (Willey et at. 1986).

The same principle has since been extended to agroforestry research, which is still largely restricted to combinations of a single tree species with a single crop (Ong 1991). To what extent can these major underlying principles, based on two-species mixtures, be extended to examine the functional role of 10 or more species in complcx agroecosystems? If some species are lost as a result of agricultural intensification, are these the ones which are crucial for improvement in the use of environmental resources? Would the remaining species be able to provide the same overall productivity or resilience as compared with the initial natural ecosystems? In other words, is the impact of diversity on function a product of species richness per se or of the role of keystone species? Unfortunately there seem to be no investigations in which the only treatment is species richness.

The validity of the resource-capture hypothesis is most rigorously examined in experiments which target the most limiting resource. For instance, in cases where water is limiting, to understand the spatial and temporal relationships between the species concerned it is necessary to quantify the complete water balance of the tree/grass/crop system and their respective sole stands. Thus for each system, total precipitation, P can be expressed (in mm) as:

P= Tc +EC+ fVc + Rç P=T{ + E{ + lYt + Rx P = {EC + Tx) + ¿?ct + Wc{ + R,

•ct for crops alone for tress alone for mixtures where T is the transpiration, E the soil evaporation, W the water stored below the root zone and R the runoff from the hillslope (Figure 11.5), and

Figure 11.5 Major components of the water balance of an agroforestry system on hill slopes. Soil evaporation (£) and runoff (/?) are lost without passing through the vegetation. Transpiration by vegetation are shown for crops (Tc) and trees (7*,). Residual soil moisture is indicated as W

Figure 11.5 Major components of the water balance of an agroforestry system on hill slopes. Soil evaporation (£) and runoff (/?) are lost without passing through the vegetation. Transpiration by vegetation are shown for crops (Tc) and trees (7*,). Residual soil moisture is indicated as W

subscripts c and t denote crops and trees, respectively. Initial results indicate that (Tc+Tt) is twice Tc but only marginally greater than Tt. The greater value of (Tc+ Tx) is largely due to a reduction in £ct and Wct because Rcl is negligible (4% of P) during the period of study.

The same principles of resource capture have been used in temperate agriculture where the main emphasis has been on the capture and use of light (Montieth 1977), since water and nutrients are seldom limiting. Light interception has also been used in drought-prone environments to examine resource capture by intercropping systems (Marshall and Willey 1983), but light interception should be regarded as a "proxy" measurement for the effects of intercropping on growth rather than as the rate-limiting process. Whatever the limiting factor (light, water or nutrients), the basis of the resource capture principle is to quantify the amount of resource captured (Q) and the efficiency in which the resource is converted into dry matter (£>).

For example, Q of light is measured in megajoules of photosynthetically active radiation per square metre (MJ m"2) of ground area, and is expressed as grammes of dry matter per megajoule intercepted. Therefore total dry matter produced (TDM) = 2xi>. For both light and water the value of e for healthy canopies is relatively conservative for a given environment, and therefore it is possible to extrapolate across environments. Furthermore, there is much more quantitative information on how changes in moisture level, COi and temperature might influence e. The ICRISAT research indicates that improvements in both Q and e are possible with intercropping (Willey et al. 1986), but there is no reason to expect a significant change in resource capture or resource-use efficiency as a result of a loss in species richness alone unless the lost species has served an important function in the system.

Traditional management of plant diversity Farmers may deliberately maintain diversity among cultivars of a major crop species. Such practices can serve a variety of purposes - for diversity of product, to spread risk through the cropping season, or to suit different micro-environments. For instance, the Apatani tribe in northeastern India, who are traditionally involved with wet-rice cultivation, have selected different rice genotypes to suit sites of varied nutrient status within a landscape. Since waste recycling from the village is a key element in maintenance of soil fertility, fertility status is high closer to the village, declining gradually as one moves away. In flooded plots closer to the village, a long-duration rice cultivar is grown in combination with the maintenance of fish. This cultivatar has low nutrient-use efficiency but is compatible with pisciculture to capitalize upon the high nutrient levels. Farther away, a more nutrient-use efficient but shorter-duration rice cultivar is grown, but without pisciculture. Similar patterns of utilisation of rice cultivars have been observed by Richards (1985) among the farmers of west Africa.

Management of plant diversity is not confined to the crop species, but is also found in choices of associated plants such as mulch crops or agrofor-estry trees, and even among so-called "weeds". The adverse effect of a weed species may be altered depending upon the adaptive differentiation of ecotype populations to a given soil type. In a study of ccotypic differentiation in Cynodon dactylon, in the context of the calcicole - calcifuge problem in the Indo-Gangetic alluvial soils of western India, specially adapted ecotypes suited to a given soil type were shown to be more aggressive than others that were not so well suited (Ramakrishnan and Gupta 1972). In northwestern Indian wheat plots, Trigonella polycerata is a leguminous weed. At low densities of the weed, crop yield is enhanced due to improved nitrogen capture by this legume. The adverse effect of the presence of this weed is not significant until the density goes beyond 3200 plants m"2 (Kapoor and Ramakrishnan 1975).

Weed management can also be used to control productivity at a plot level, as is seen in the strategy of the jhum (shifting cultivation) farmer in northeast India, where loss of sediment and nutrients from plots by erosion is controlled by the weed-management practice. The farmer deals with it in. two different ways. In all cases of weed management, the jhum farmer leaves about 20% of the weed biomass standing in the plot. The remaining 80%

biomass is weeded and put directly back into the plot as mulch. Retention of this level of living weed biomass reduces the loss of sediment-labile elements such as potassium through run-off by about one-fifth when compared with total weeding. Although the losses are least in unweeded plots, this is to a large extent negated through weed-crop competition and the consequent reduced crop yield. Further study has shown that if it were not for the integrated weed management practices of the traditional jhum farmer, the highly shortened 4-5 year cycles now prevalent would have resulted in even more distorted ecosystem function than at present, through adverse effects on nutrient cycling. The 80% weed biomass put back into the plot, rather than being thrown outside, helps in nutrient cycling because of rapid decomposition (Ramakrishnan 1992).

The integrated weed management concept of the traditional jhum farmer indicates that the farmer knows precisely how intense the weeding should be so that the weed stops interfering with crop yield, and yet the beneficial effects are manifest. Such a subtle distinction between the "weed" and the "non-weed" status of the same species or set of species is widespread. For instance, in the Mayan region of southern Mexico and the Guatemalan highlands, farmers employ a system of mal monte and buen monte (bad weeds and good weeds). These terms are used to refer to the vegetation that grows in their fallow. Contrary to the modern agriculturalists point of view that all non-crop plants are undesirable weeds, the Mayan farmers regard the "other plants" in the system as a functioning part of the agroecosystem as a whole. Some are regarded as buen monte for obvious reasons (a Chenopodium that exudes a nematocide from its roots, several legume species that may fix nitrogen) and sometimes for no scientifically obvious reason. Frequently it is the collection of plant species that is regarded as buen or mal monte. Under traditional management conditions, mal monte is discouraged and buen monte encouraged in fallow plots. That some of the non-crop plants in the system are seen as functional is at least testament to the perception of local farmers that plant biodiversity is important (Chacon and Gleissman 1982).

11.3.3 Biodiversity and pest management

A major effect arising from the conversion of natural ecosystems to agroeco-systems is a destabilization of arthropod populations, often leading to pest outbreaks. Historically, the management of pest populations has followed two alternative routes: the first depends on the use of increasingly sophisticated chemical pesticides to assume the ecological function of population control; the second attempts to establish a biologically controlled equilibrium of pest populations in agroecosystems by the conservation, augmentation and importation of natural enemies. The impressive record of biological control, exemplifying this second route, provides clear evidence that popula-tion-level processes can be influenced by increasing biodiversity. On the other hand, many crop systems have proved not to be amenable to biological control, necessitating the continued intensive use of chemicals despite the environmental problems their use can generate. The realization that complete reliance on human inputs to manipulate pests can be extremely costly has led to the development of integrated pest management (IPM), combining the two approaches, in which inputs are minimized and geared to augmenting biological processes. Methods utilized include increasing the genetic diversity of the crop plants, intercropping, weedy cultivation and a variety of methods for conservation or addition of natural enemies. These practices are designed to increase the biodiversity of the system, and successes have been achieved, for example, in soybean (Kogan and Turnip-seed 1987) and corn (Luckman 1978). The particular practice of biological control and the more general practices of integrated pest management highlight the broad importance of maintaining some degree of biodiversity in agroecosystems in order to stabilize populations and decrease management costs.

The importance of the genotypic diversity of natural enemies to the biological control of insect pests and weeds is thus becoming increasingly appreciated. The importation of genetically distinct "biotypes" was instrumental in the spectacularly successful control of walnut and filbert aphid in California by the aphidiid parasitoid Trioxys pallidus Halliday (Unruh and Messing 1993). Similarly, the successful establishment of the weevil Rhino-cyllus cfínicas Foelich was responsible for the control of thistles, also in California (Goedon et al. 1985). Genetic diversity among populations of biological control agents can be critical to both their abilities to adapt to local climatic conditions and the range of host species which they are able to attack.

The role of diversity at the species-functional group level has been vigorously debated in the context of the biological control of insect pests (Ehler 1990), with many workers believing that increased parasitoid diversity generally results in better control. On the other hand, the historical biological control record indicates that diversity is not necessarily associated with the degree of pest control. Myers et al. (1989) analyzed 50 successful cases of biological control and found that in 68% of the cases control was afforded by a single natural enemy species. In the cases where multiple species were credited with control, the natural enemies complex contained an average of 2.8 species. Additional evidence that parasitoid community diversity may not be linked with the degree of pest population depression comes from quantitative comparative analyses of patterns in parasitoid species richness, parasitoid-induced host mortality, and biological control success rates achieved by Hawkins and collaborators. Initial analyses indicated that biological control success rates were indeed positively associated with parasitoid species richness (Hawkins 1993). More detailed studies, however, found that the relationship is not causally linked. Instead, the evidence suggests that host susceptibility drives parasitoid species richness and the impact of the parasitoid on host populations independently (Hawkins 1994). If so, parasitoid diversity per se is not critical to host dynamics, and relatively depauperate parasitoid communities are functionally identical to species-rich communities.

It is also possible to take advantage of genotypic and phenotypic differences in crop cultivars to reduce the impact of insect pests. For example, stink bugs, bean leaf beetle and Mexican bean beetle have been successfully trap-cropped using early maturing varieties of soybeans near maize plantings in "North and South America and Africa (Hokkanen 1991). In contrast, genetic uniformity in widely planted crop plants can exacerbate problems of adaptation of potential pests and pathogens, and can quickly frustrate attempts to utilize the properties of plants providing resistance. For example, the wheat variety Eureka, resistant to wheat-stem rust, was released into Australia in 1938, but a virulent strain of wheat-stem rust had appeared by 1942. The prevalence of virulent strains was thereafter positively associated with the proportion of the total wheat acreage sown to Eureka, causing a rapid decline in its use. Research elucidating the genetic basis of resistance in wheat varieties (Watson and Luig 1963) contributed to the now widely held realization that a broad genetic basis for resistance properties was essential for long-lasting protection against pathogens and insect pests.

The outcome of interactions between viral, bacterial and fungal pathogens is commonly determined by differences in a small number of genes in the microbe and/or plant. Much use of this gene-for-gene relationship has been made in modern agriculture (Van der Plank 1984). Early successes in developing resistant cultivars through the incorporation of resistant alleles at specific loci have often proved to be short-lived because of corresponding genetic shifts in host virulence. The risks of genetic uniformity have been recognised, and attempts have been made to re-introduce genetic variability through the use of multilines or variety mixtures (Browning and Frey 1969; Wolfe et al. 1981). The existence of genotype diversity within a crop population may have very significant effects on its productivity because of the reservoir of resistance to disease that it represents. These interactions can in some cases be quite complex. For instance, Jcnkyn and Dyke (1985) investigated the pattern of powdery mildew infection on three barley cultivars which have high, intermediate and low resistance. When pure stands of the cultivars were grown side by side, the outcome of infection in a given plot was dependent not only on the genotype of the cultivar in the plot, but also on that of cultivars in adjacent plots. For instance, the intermediate variety showed greater yield when adjacent to the susceptible variety than when next to the resistant one or to a mixture of all three. In the mixtures, the mean yield was significantly lower in plots next to the intermediate as compared with the resistant or susceptible varieties. Greater genetic diversity within the host population lowers the impact of pathogen attack not just as a direct function of the relative frequency of resistant genomes, but also as a consequence of interactions between genotypes. Chin and Wolfe (1984) have developed a model to describe the progress of a pathogen in a population of mixed host varieties.

A particularly spectacular example of the domino effects of loss of natural pest control as a result of diversity decline is that of cotton in Central America. The Pacific lowlands of Central America came to be dominated by cotton in the early 1950s, creating a situation that represents a variety of levels of interaction, from species to landscape. Cotton farmers began to use ever more applications of pesticides to control an increasing number of pests, and by the 1970s were spraying over 24 times per season for more than 20 species of insect pests. Costs of production consequently became so high that much of the cotton of Guatemala, Nicaragua and El Salvador had to be abandoned. Attempts were then made to convert to other crops, for example, soybeans and sunflowers in Nicaragua. Soybean production failed largely because of attacks from several species of armyworms (especially Spodoptera frugiperda and S. exigua: Savoie 1990). Armyworms can devastate a crop rapidly (Rosset et al. 1985), but can be held under control by natural enemies (Perfecto 1990). The former cotton fields of Nicaragua, however, are devoid of natural enemies, presumably because of the impact of the previous massive pesticide applications, thus precluding effective soybean production. With the failure of soybeans on the horizon, the Nicaraguan Ministry of Agriculture briefly began promoting sunflowers in the 1980s. Problems were immediately encountered with seed set, probably due to the absence of pollinators. It has also been suggested that Kieferia spp. and Lyriomiza spp. pests of tomatoes, some 100 km away from the cotton fields, may be a consequence of the pesticides previously applied in cotton (Rosset 1986).

The effects of increasing within-plot plant diversity on the densities of inscct pests are being evaluated in a number of experimental and theoretical studies (recently reviewed by Sheehan 1986; Russell 1989; Andow 1991). Results of field studies to date have been somewhat mixed, but increasing plant diversity through intercropping, trap-cropping and weedy cultures more often results in lowered pest densities than in pest increases. Unfortunately, elucidating which mechanism(s) might underline herbivore responses to vegetational diversity is difficult owing to potentially complex interactions between plants, herbivores and natural enemies. Nevertheless, there is a growing body of evidence that plant diversity can play an important functional role in the suppression of insect pests.

The importance of nectar sources, overwintering sites and alternate victims to natural enemies has been demonstrated in a large number of studies showing that plant diversity outside the crop provides reservoirs for enemies and leads to increased mortality of insect pests (reviewed by Van Emden 1990). Such effects can occur over relatively small spatial scales, such as enemies colonizing fields from wild plants growing along field margins, to much larger, landscape scales, such as enemy migration from woodlots or fallows (Van Emden 1990; Altieri et al. 1993).

Migration of organisms from one geographical region to another has been increasing dramatically in recent times (Drake et o.L 1989). These biological invasions can result in large-scale transformations at the landscape level. Invasions by weeding species such as Lantana camera. Eupatorium spp., Mikania micrantha, etc., and insects such as cassava mealy bug and citrus scale insects, are classical examples in the tropics. Such invasions have significant implications for biodiversity and agroecosystem function, including the conceptual basis of classical biological control which manipulates biodiversity at the largest geographical scale. Another obvious factor influencing the structure and function of agricultural landscapes that should be mentioned is that many of our major crops arc themselves exotic. Thus, intentional, as well as accidental, introductions play a fundamental role in agroecosystems.

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