Interactions Among Species in Agroecosystems

This part of the course considers some of the other organisms, in addition to crops and soil organisms, that occur in agroecosystems, particular herbivores (mostly insects) and their predators, and competitors (weeds). Pathogens are discussed only briefly.

Herbivores

Why don't insects (and other herbivores) consume all available plant biomass? That is, Why is the world green?-most likely answers are plant defenses that limit which herbivores can feed on which plants, and predators that keep herbivore populations in check.

Groups of herbivores:

• Vertebrates-birds, mammals
• Invertebrates-insects, arachnids (mites), mollusks (snails, slugs). Of these groups insects cause the greatest crop losses in most agroecosystems.

Plant Strategies to cope with herbivory:

• Escape-short life cycle
• Tolerance--Compensation for tissue loss
• Defense--protection of tissues

Plant Defense Mechanisms:

Chemical defenses:

• Plant Secondary Compounds --Function not understood at first. [Contrast: primary vs. secondary metabolites]
• Plant compounds now known to affect all aspects of insect biology, acting as: toxins, feeding deterrents, digestion inhibitors, growth-regulators (hormones), etc.
• Examples include: alkaloids (heterocyclic N compounds); glucosinolates (contain sulfur); coumarins (heterocyclic ketones); terpenoids (e.g., gossypol); tannins (polyphenolics)
• Complex interactions: Plant released volatile terpenes which may serve as signals to predators or parasites.

Morphological defenses--Spines, thorns, pubescence, fibers, silica, etc. Various types of pubescence: simple, stellate, barbed, dendritic, glandular, etc., present different problems to different types of herbivores.

Quantitative vs. Qualitative defenses:

• Quantitative: Complex polymers (e.g., tannins), present in permanent tissues, found in late-successional or long-lived species. E.g., tannins, lignin, silica. Tannins are digestion inhibitors rather than toxins. They are effective against a broad spectrum of insect pests, but involve substantial investment by the plant. [Review succession.]
• Qualitative: Smaller, toxic molecules, generally <2% of dry weight; more common in short-lived leaves, flower buds, ripening fruits; found most often in short-lived or early successional species.

Constitutive vs. Inducible defenses:

• Constitutive: Always present, include morphological defenses and some chemical defenses
• Inducible: Produced in response to insect feeding. E.g., many low-molecular weight toxins (alkaloids, coumarins, monoterpenes) and proteinase inhibitors.

Quantitative defenses are always constitutive; qualitative defenses may be constitutive or inducible.

Herbivore [mostly insect] Responses to plant defenses:

Feeding strategies:

• Monophagous: Specialists, ca. 90% of plant-eating insects
• Oligophagous
• Polyphagous

Mechanisms for coping with plant toxins.

• Detoxification--mostly found in monophagous species.
• Mixed function oxidases--mostly found in polyphagous species.
• Sequestration--mostly in polyphagous species.

Theories of Plant-Herbivore Interactions. [Mostly developed by entomologists]:

Coevolutionary "arms race". Predicts a correspondence between the phylogeny of plants and the phylogeny of their specialist herbivores.

Plant Apparency. Types of defenses should depend on how "apparent" a plant is-i.e., on how easy or difficult is it for herbivores to locate their hosts.

• Early successional species are more difficult to locate and should have inducible defenses.
• Late-successional species that are easy to locate should have constitutive defenses.
• Suggests that crops may be particularly vulnerable since they are basically early-successional species that are easy to locate.

Resource Availability. Chemical defenses related to availability of limiting resources.

• Plants have a greater need to protect tissue in low-resource habitats where it is more difficult to replace.
• Constitutive defenses should be favored in low-resource habitats
• N-based toxins should be common in high N-habitats, C-based toxins should be common in low-N habitats.

Integrative Approach: Insect feeding determined by food quality, climate (particularly temperature), and avoidance of enemies.

Loss of Herbivore Defenses in Crop Plants. [Crops the least-well protected plants on the planet?]

During domestication--to improve palatability

During modern breeding--to increase yields. Examples:

• Reduced hairiness ----> Higher photosynthesis through reduced reflection of light
• Reduced allocation of photosynthate to toxins ----> higher harvest index

Regulation of Population Size:

Population growth: N(t) = N(0)(1+r)^t; r = b (birth rate) - d (death rate); when death rate is very small (d ~ 0), then r ~ b.

Example 1, hypothetical organism with low fecundity, short life span:

• b = 1 offspring per female per year = 0.5 y^-1 if the sex ratio is 50:50.
• d = 0.2 y^-1 (average lifespan of 5 years)
• If N(0) = 100, What is N(50)?
• r = b - d = 0.5 - 0.2 = 0.3 y^-1
• N(50) = 100 (1.3)⁵⁰ = 5 x 10⁷

Example 2, insect with higher fecundity, several generations per year (e.g., aphid):

• b = 20 offspring per generation, generation time = 3 weeks; for a 6-mo growing season there are ca. 9 generations per season;
• d = 0.5 (generation)^-1, i.e individuals live for a single year
• If N(0) = 10 in a one ha field, what is the population at the end of the growing season?
• r = 20 - 0.5 = 19.5 (generation)^-1; N(9) = 10 (1+19.5)⁹ = 6.4 x 10¹²
• What is the density of these insects (m^-2) over the of the field?
• 1 ha = 10⁴ m², density = 6.4 x 10⁸ m^-2. If each aphid occupies a volume of 1 mm³, the entire field would be covered by a layer of aphids 64 cm deep.

LESSON: Since populations in nature do not tend to increase at exponential rates, death rates (d) must be approximately equal to birth rates (b). Birth rates are primarily controlled by fecundity when food resources are not limiting. Populations are regulated mostly through their death rates, not birth rates.

What processes control death rates (mortality)?, or, How are populations regulated in nature?

Density-dependent mechanisms: d(t) = f(N(t)): death rate is not a constant, but increases as the population size increases. Density dependent mechanisms include:

• Predation
• Parasitism
• Disease.
• Food supply? In the case of agroecosystems, food supplies of herbivores are essentially unlimited. In practice, is it likely that an insect herbivore-particularly a generalist herbivore-would be limited by food supply? In general, food supply probably limits birth rates as well as death rates.

Density-independent mechanisms: d(t) = K: death rate is a constant (K), the mortality-causing event or process takes a constant proportion of the population. Included would be:

• Climate.
• Insecticides.

Density-dependent mechanisms are more effective and regulating population numbers. [Why?]

Ecological problems associated with insecticide use:

1. Insecticide resistance:

A response to selection requires that there be genetic variation for the trait and a difference in survival (actually fitness) among different genotypes.

The response to selection is:

R = ih²s_p

where:

• h² = heritability for the trait; h² for insecticide resistance is generally very low but not = 0
• i = intensity of selection pressure, in units of number of standard deviation of surviving or selected portion of the population
• s_p = population standard deviation

Very high selection pressure, even when combined with low heritability, can produce a rapid response to selection, i.e., spread of resistant genes throughout the population

Resistance develops more rapidly in herbivorous insects than in predators. [Why?]

Resistance mechanisms:

• Target-site modification (receptor is modified);
• Detoxification (toxin is modified);
• Increased excretion (ie., sequestration);
• Behavioral avoidance.

2. Pest Resurgence:

If herbivore (pest) insect populations recover more rapidly than predator populations, pest populations can "resurge" to a higher level.

Can happen because predators experience higher mortality or if predator growth rates are lower.

3. Secondary Pest Outbreaks:

Secondary pest is an insect that normally does not reach injurious population levels.

If such an insect is not affected by the insecticide, or if it rapidly develops resistance and the insecticide application eliminates the predators and parasites of the secondary pest, then the pest can reach injurious populations in the absence of competitors and predators.

In California 24/25 (96%) of major crop pests are secondary pests.

Summary of the ecological problem:

• Insecticides kill herbivore pests but also kill their predators and parasites; thus:
• Insecticide use substitutes density-independent control for density-dependent control;
• Development of resistance leads to less effective control, and (at least in the past) increased insecticide applications.

Integrated Pest Management (IPM).

IPM, in theory, integrates ecological principles into herbivore-management.

IPM has relativelly recently been supported by the insecticide industry:

• Decreased rate of discovery of new insecticides.
• Cost of developing new insecticides ($15 - $30 M).

Important IPM Concepts:

1.Injury vs. Damage

Injury = loss of tissue to herbivores or the rate of plant feeding

Damage = yield loss

The relationship between injury and damage generally is not linear. Low levels of injury may not result in any damage; if so any application of insecticides to an insect population not causing damage is economically ineffective.

Damage Curve--relates Damage = f(Injury)

• Point at which a control action is initiated.
• Is necessarily lower than the EIL.

4. Monitoring (population censusing) is critical.

IPM Tactics

Preventative management vs Therapeutic control:

• Preventative tactics are applied before the pest problem develops; therapeutic tactics are applied after a pest problem develops.
• Some tactics can be applied either preventatively or therapeutically.
• Selection pressures are higher for preventative tactics.

Cultural tactics: Crop rotation, sanitation, planting and harvest dates,--almost always preventative.

Biological control:

• Conservation of natural enemies (preventative).
• Augmentation of natural enemies (therapeutic).
• Herbivore pathogens (e.g., BT) (therapeutic)

Chemical control:-IPM objective is to use chemicals (insecticides) only as a therapeutic control.

• Insecticides. IPM is basically a management system to prolong the effective (economic) life of insecticides-not remove insecticides from the system
• Pheromones (disrupt mating behavior; trap bait);
• Anti-feedant compounds

Genetically modified organisms (GMOs) (To date, mostly incorporation of Bt toxins):

• Bacillus thuringiensis, produces crystaline protein endotoxins
• GMOs use Bt preventatively rather than therapeutically.
• Development of resistance is likely to be rapid.
• UA researchers found in 1997 that a single gene mutation in diamondback moth conferred resistance to 4 Bt insecticides = Cross resistance
• There have been reports of failure of Bt cotton to control several lepitopteran pests.

To avoid resistance, need to plant refuges-fields of non-Bt crops. Federal regulations provide three althernatives for Bt cotton:

• Plant up to 80% (acreage) Bt cotton and at least 20% non Bt cotton; grower may spray non-Bt cotton with any insecticide (other than foliar Bt applications!).
• Plant up to 95% (acreage) Bt cotton and at least 5% non Bt cotton; no insecticide applications allowed on non Bt cotton
• Plant a mixture of up to 95% Bt cotton and at least 5% non Bt cotton (i.e., in the same field); foliar insecticides permited.

Best strategy to delay development of resistance to transgenic Bt varieties (Hokkanen and Wearing, 1995):

• Provide refugia within the crop
• Eliminate pesticides to conserve natural enemies
• Enhance natural enemies through agroecosystem diversification
• Rotate transgenic and susceptible cultivars when necessary.

High dose/refuge strategy (SS, Ss are susceptible to Bt; ss is resistant to Bt):

• plant tissue should be very toxic so that heterozygous individuals are killed
• resistance alleles should be recessive, rare
• susceptible insects must mate with resistant insects

Huang et al (1999): Resistance in European cornborer is incompletely dominant; most other Bt resistance is recessive. High dose/refuge strategy based on recessive resistance (heterozygotes susceptible).

Liu et al. (1999) found that resistant pink bollworm larvae develop more slowly than susceptible individuals, which could lead to assortative mating which would accelerate the development of resistance.

Hoy (1998) argues that the entire concept of "resistance management" is problematic:

• it may be possible to mitigate resistance, but not "manage" it in any meaningful sense
• simulation models of resistance are too simplistic to capture the important ecological interactions

Hoy (1998) argues that Bt transgenic crops are not likely to be sustainable because of development of resistance. This could have negative effects in agroecosystems which have relied on traditional Bt (therapeutic) sprays, for instance in minor crops or organic systems. Hoy further argues that IPM programs need to be based on holistic, multitactic strategies that seek to enhance the compatibility of insecticides and biological control agents.

Competitors (Weeds)

Characteristics of Weeds

• High seed production, competitiveness, low palatability, seed longevity, seed dormancy, rapid emergence.
• Most weeds evolved from early successional species; many are crop relatives
• Most weeds in the United States are exotics--WHY? (Habitats with high levels of resources and frequent disturbance represent a comparatively new niche in North America).

Competition/Niche Theory

Two species can occupy the same habitat and not compete if:

• The species use different resources. This is often true for animals, but seldom true for plants.
• Resources are sufficient for both. For example, plants in the desert seldom compete for light.
• The species obtain their resources from different parts of the habitat. I.e., the species have a somewhat different niche with respect to resource acquisition.
• Many plant ecologists (e.g., David Tilman) maintain that plant species specialize with respect to their ability to capture different resources. This is probably not true, however, for crops and weeds.

Weeds reduce crop yield by reducing the supply of resources through competition.

• Plants use common resources--Light, C0₂, Water, Nutrients.
• Plants obtain resources from resource depletion zones, which depend on root and shoot architecture, and on resource mobility.
• Intensity of competition depends on the degree of overlap of resource depletion zones.

Resource depletion theory:

• R*_{i, k} = concentration of resource i at which uptake = loss for species k. That is, it is the minimum supply or level of the resource at which the plant achieves a positive net uptake.
• Given 2 species competing for the same resource i, the species with the lowest R*_i is the superior competitor. Mycorrhizal associations probably function to lower R*_i

There are often tradeoffs in the acquisition of different resources. For example, allocation of carbon to shoots helps the plant compete for light while allocation of carbon to roots improves competition for water and nutrients.

Plants generally compete by preemption. The first plant to use a resource depletes the supply for plants developing later or more slowly. Early germination, rapid height growth, and rapid development of LAI contribute to competitive ability by preemption.

Resource augmentation [e.g., fertilization and irrigation] generally results in increased growth and therefore increased competition among neighbors. Rate of thinning is faster under high resources. This could favor either the crop or the weed, but probably favors the crop under most circumstances?

Intensity of competition:

• Ecological theory suggests that the more similar are two species niches, the more intense will be the competition between them.
• Thus intraspecific competition should be more intense than interspecific competition. We would also expect that competition between closely related species would be more intense than competition between unrelated species. This is significant in agriculture because many weeds are genetically related to crops. Examples: wheat, barley, sorghum, rice, oats, millet, maize, potato, tomato, pepper, sunflower, carrots, radish, lettuce, etc. all have closely related weed species.
• Intraspecific competition in crops is reduced by selecting for compact canopies (high leaf angles) and short, branching root systems. This also reduces the crop's ability to compete with weeds, except when grown at very high densities.

Relationship of weed density to crop yield ("Threshold theory" vs. "weed-free period" theory).

• If the crop tolerates a low level of weeds before experiencing yield loss, the relationship between yield and weed density will be sigmoidal. A sigmoidal relationship may be due to "apparent competition"-facilitation by weeds at low density by some mechanism.
• If yield losses are greatest at low weed density, the relationship between yield and weed density will be a negative exponential. This relationship is the one most commonly observed.

[What are the management implications for these two theories?]

Allelopathy

Allelopathy is the chemical inhibition of one species by another. A variety of compounds may be involved including phenolics, terpenoids, and quinones. Examples:

• Juglone (a quinone) below walnut trees.
• Bare ground below eucalyptus trees. Accumulation of terpenes.

The source of allelochemicals could be crop plants, crop plant residues, weeds, or microorganisms; affected species could be crops, weeds, or microbes. Mycorrhizal fungi are believed to be sensitive to many allelochemicals (Einhellig, 1996).

Inhibitory compound(s) may be released by:

• Volatilization from living tissue
• Root exudates
• Decomposition of litter
• Leaching of compounds from the canopy.

In the laboratory it is easy to demonstrate that species produce compounds that inhibit seed germination and/or root growth of other species. In the field it has been very difficult to demonstrate that allelopathy is actually effective. Nevertheless, many plant ecologists and agroecologists have argued that allelopathy is an important process.

Plant-plant interference may be through competition or allelopathy; these are difficult to separate and likely to interact (Inderjit & del Moral, 1997). Smother Crops combine good competitive ability and allelopathy.

Crops known or suspected to be highly allelopathic include: rye, barley, and cucurbits (squash, cucumbers, melons, etc. Many crops can be inhibited by their own residues.

Plants are most sensitive to allelochemicals when they are growing under suboptimal conditions of moisture, temperature, or nutrients (Einhellig 1996)

Wu et al. (1999) recommend identifying genes in wild species, weeds, or crops that control the production of particularly effective allelochemicals and transferring them to other selected crop plants.

Crop Mimicry = Associative weeds.

Seed mimics. Resemble crops in seed size, shape, and phenology. Seed harvested with that of the crop is difficult to separate by seed cleaning. E.g., Camelina sativa in flax, Lolium in barley.

Vegetative mimics. Avoid weeding. E.g., teosinte in maize.

Some weeds show both: E.g., barnyard grass in rice.

Herbicide Resistance

As with insecticides, concerns have been expressed about excessive applications of herbicides. Some weeds have developed resistance to herbicides, and herbicides can have unwanted effects on non-target organisms.

Herbicide resistance has been slower to evolve than insecticide resistance-why?

• Fewer generations
• Lower fitness of herbicide-resistant weeds

Observations on herbicide resistance:

• About 100 weed species known to be resistant to one or more herbicides.
• Acetolactate synthase (ALS) inhibitors [e.g., chlorosulfuron herbicide]-natural mutations conferring resistance appear at a frequency of one in a million.
• Resistance to glyphosate (RoundUp) has been slow to evolve, but is considered inevitable.
• Metabolic resistance to diclofop-methyl and chloroluron/isoproturon herbicides in wheat has resulted in cross-resistance to all other wheat herbicides (Gressel et al., 1996). An extensive problem in Australia.

Herbicide-resistant (GMO) crops have been a major effort of biotechnology companies:

• Herbicides are the major strategy for controlling weeds and herbicide-producing companies are seeking ways to expand their markets. Agrochemical companies have purchased seed companies and combined them with biotechnology projects to produce crops resistant to their own herbicide products.
• Need exists for herbicide-tolerant crops in the developing world (Gressel et al., 1996), especially for parasitic weeds like Striga spp. (witchweeds). However, owners of resistant germplasm have often refused to release such germplasm to plant breeders in the developing world.

Are there ecological risks associated with transgenic herbicide-resistant crops?

• Many crops have weedy relatives; genetically-engineered tolerance is likely to be transferred to these weedy relatives by natural hybridization. For example, hybrids (2%) between grain sorghum and Johnson grass occur up to 100 m from crop, despite the facts that these species are partially incompatible and have different chromosome numbers (Arriola & Ellstrand, 1996).
• Excessive use of a single herbicide could lead to build-up in residues, limiting crop rotation options.
• Herbicide effects on soil organisms are not well-known, but thus far do not appear to be detrimental. Some herbicides are toxic to Rhizobium (Clark et al., 1996). Herbicide use may interfere with VA mycorrhizae by decreasing spore viability or eliminating weeds that serve as alternative hosts.

Integrated Weed Management.

• Less-well developed than IPM.
• Goal is to shift the competitive balance between crop and weed.
• Tactics include: mechanical cultivation, early emergence, crop rotation, weed-suppressive varieties, smother crops, intercropping, allelopathic cover crops.

Population dynamics of weeds are very difficult to model because fields generally contain many species of weeds with patchy distributions.

Liebman and Davis (2000) suggest developing IWM strategies based on ecological differences between crops and weeds, particularly seed size:

• Crop seed weights are typically 10-1000 times those of weed seed weights.
• Weed seedlings have rapid RGRs but are susceptible to low concentrations of soil resources early in their lifespan.
• Seedlings produced by the larger seeds of crops are more resistant to low nutrient stress, allelopathy, disease, and herbivory. Small-seeded weeds are most susceptible to allelopathic crops such as red clover.
• Delayed fertilizer nutrient applications, or use of residues instead of inorganic fertilizers, should shift the competitive balance to crops.

Jordan et al. (2000) no-till and cover cropping should promote mycorrhizal fungi, which then should inhibit germination and seedling of non-host plants, which include families with many weeds: Amaranthaceae, Brassicaceae, Chenopodiaceae, Polygonaceae, and some Cyperaceae and Poaceae.

Pathogens

Plant populations in nature are seldom decimated by disease. Primary "defense" is genetic diversity--plants and their pathogens coexist in a state of "balanced polymorphism".

Diseases reduce ecological resource use efficiency by reducing resource uptake by various mechanisms: obstructing vascular tissues, damaging roots, restricting root growth, or removing leaf area.

Plants possess morphological and chemical defenses against pathogens:

• Morphological-- cuticle
• Chemical-- both constitutive and inducible (inducible defenses against pathogens are called phytoalexins)
• These defenses most effective for aboveground pathogens.

Soil Ecology

• Disease organisms are endemic in most soils; their populations seldom buildup to damaging levels due to competition with other soil organisms.
• Examples; bacterial antibiotics and mycorrhizae inhibit disease organisms.
• Soils with high levels of organic matter and soil microbes are disease suppressive.

Soil-borne fungal diseases such as Fusarium, Rhizoctonia, and Verticillium are long lived (4-6 years) and therefore require long rotations to prevent build-up of their inocula.