Structure and Function in Agroecosystem Design and Management - Chapter 5 pdf

95 From Agriculture Based on Fossil Energy to Agriculture Based on the Use of Complex Biological Interactions.. which the interactions between organisms and environment and matter cyclin

CHAPTER 5

Utilization of Biological Interactions and Matter Cycling in Agriculture Masae Shiyomi

CONTENTS

Introduction 95

From Agriculture Based on Fossil Energy to Agriculture Based on the Use of Complex Biological Interactions 97

Plant-Grasshopper-Mantis-Bird Model 98

Grasshopper-Mantis Model 99

The Importance of Matter Cycling in the New Agriculture 101

Grassland Ecosystems 102

Upland Crop Field Ecosystems 106

Paddy Field Ecosystems 107

Conclusions 108

References 110

INTRODUCTION

For the 50 years following the Second World War, agricultural produc-tion markedly increased Examples are shown in Figure 5.1 for corn in the U.S and rice in Japan (Uchijima, 1990) In the U.S., the use of F1hybrid corn

in the 1960s led to a rapid increase in production per hectare Although the production of rice in Japan has not made such rapid strides as that of corn in the U.S., the production per unit area has gradually increased from 1900 to the very high present level, especially in the last 50 years

Modern agriculture, which depends on the consumption of large quanti-ties of fossil fuel, is now being forced to change to an alternative system in

95 0-8493-0904-2/01/$0.00+$.50

Corn, United States

Rice, Japan

7

6

5

4

3

2 5

4

3

2

Figure 5.1 Corn and rice yields in the U.S and Japan in the last 100 years (Uchijima,

1990).

which the interactions between organisms and environment and matter cycling in agricultural ecosystems are properly utilized (Edwards et al., 1990; Shiyomi, 1993) First we discuss the problems everyone is presently facing There are three reasons for making such a change One reason is the depletion of readily available fossil fuel resources According to the Tokyo

newspaper Asahi-shinbun (December 25, 1994),

Energy problem is serious The Central Institute of Electric Power Industry, Japan, predicts that the annual energy demand in the world in 2050 will reach an equivalent of 13 to 24 billion tons of petroleum If the present rate

of consumption of fossil fuel continues, all presently known oil deposits will have been mined by 2040, and all deposits to be found in the future will be mined by 2080, too Natural gas will be exhausted by 2080.

An American entomologist, D Pimentel, stated (Pimentel 1992),

“Unfortunately throughout the world more fossil energy is being used in order to increase food production for the ever expanding world population While the world population grows, the known supplies of fossil energy are being rapidly drawn down For example, most world oil and natural gas reserves will be consumed during the next 35 years.” Although the time

when the fossil deposits would have been exhausted differs among the reports, someday they will disappear

A second reason for change is that as the amount of fertilizers and agro-chemicals used increases, increase in the growth and yield of crops decreases exponentially, and eventually the growth and yield level off Furthermore, to these reduced marginal rates of return from input use, it is unlikely that new strains or varieties can be developed that will respond more effectively to an increase in input

Another reason for change is that the consumption of fossil fuel energy has led to the degradation of the environment According to Pimentel (1992),

“In addition, the heavy use of pesticides, especially in developed countries,

is having widespread impact on aquatic and terrestrial ecosystems Worldwide an estimated 2.5 billion kg of pesticide is applied to agriculture Yet, less than 0.1% of this pesticide reaches the target pests, with the remain-der negatively affecting humans, livestock, and natural biota Just in the U.S.,

it is estimated that pesticides cause $8 billion in damage to the environment and public health each year Million of wild birds, mammals, fishes, and ben-eficial natural enemies are destroyed each year because of the recommended use of pesticides in the U.S.”

It is clear that modern agricultural practices, which depend on inputs of fossil energy, have exerted a variety of harmful effects on both the local ecosystems and the global biosphere

This chapter discusses two topics The first concerns the importance of the use of complex biological interactions as an alternative to the heavy use

of fossil energy in modern agriculture The second discusses the impor-tance of matter cycling in agricultural ecosystems and uses examples of car-bon and nitrogen budgets in ecosystems of grassland, upland field and paddy field

FROM AGRICULTURE BASED ON FOSSIL ENERGY TO AGRICULTURE BASED ON THE USE OF COMPLEX

BIOLOGICAL INTERACTIONS

As mentioned above, the increases in agricultural production in advanced countries from the 1950s to the 1970s were largely due to large increases in the use of fossil fuel energy Specifically, the increases have been due to the increased use of fertilizers, agricultural chemicals, and big machines that are produced and operated with fossil energy sources, and to the breeding of new varieties of crops that are responsive to and compatible with such chemical inputs and cultural practices (Pimentel et al., 1973) Researchers have also promoted this agricultural system by focusing on research on improving crop yield through the direct use of these fertilizers, agrochemical inputs, and machinery Indeed, these research programs have been very efficient and have led to the increase of both crop and livestock production The use of intra- and interspecific interactions and interactions

between organisms and the environment, such as climatic factors and soils, have no particular place in the current agricultural system In modern agri-culture, these interactions are viewed as production constraints that must be overcome to make high production possible Since the direct effects of the use

of fossil fuel energy and products on agricultural production have been so powerful, reliable, and dramatic, little attention has been paid to the complex networks of interactions operating in agricultural ecosystems For example, competition between phytophagous insects, the effects of insect pathogens and other natural enemies on these phytophagous species, and antagonisms between them have intentionally been ignored Because of the clear, direct effectiveness of agrochemicals, it seemed that insect pests, plant pathogens, and weeds could be controlled at sufficiently low levels without considering the biological functions and interactions in agricultural ecosystems And because of the clear, direct effectiveness of fertilizers, it seemed that high crop yields could be guaranteed without the help of the subtle actions of soil-borne microorganisms Complex intercroppings have been excluded so that machinery can be operated more efficiently However, this modern agricul-ture has led to the three problems stated above In the alternative type of agri-culture, instead of modern agriagri-culture, analyses of indirect effects operating among the complex networks of biological interactions and between organ-isms and the environments in place of the direct effects must be considered

Plant-Grasshopper-Mantis-Bird Model

Because of the complexity of biological interactions, such interactions are most effectively understood by the use of system analysis (Edwards, 1990)

To demonstrate this concept, I will use a 4-component system composed of pasture plants, grasshoppers, mantes and birds (Figure 5.2) (Levins and Vandermeer, 1990) Grasshoppers eat plants, mantes eat grasshoppers, and birds eat both grasshoppers and mantes The first system (Figure 5.2a) is composed of only the three components, in which the population of grasshoppers increases as the biomass of pasture plants increases If the pop-ulation of grasshoppers increases, the poppop-ulation of mantes increases, and the biomass of plants decreases Then, when the biomass of plants increases, the populations of grasshoppers and mantes increase When the population

of mantes increases, the population of grasshoppers decreases, and then the plant biomass increases

If we add birds as the fourth component in the system (Figure 5.2b), the interactions operating among these components become much more compli-cated because the birds kill both grasshoppers and mantes As can be seen in Figure 5.2c, the bird population increases as the grasshopper population

increases In Figure 5.2d, I is an agrochemical Farmers do not ordinarily use

agrochemicals if many mantes, which can kill most of the grasshoppers, live there It becomes increasingly difficult to understand intuitively the interac-tions operating in such systems even in such a 4-component system like this

P P

H

I

I

I

C C

H

C

a

b

c

d

Figure 5.2 Plant-grasshopper-mantis-bird model P, H, C and I indicate the numbers

of plants, grasshoppers, mantes and birds, respectively In (d) I stands for

pesticide Arrows and circles indicate positive and negative feedbacks, respectively (From Levins and Vandermeer, 1990.)

example Indeed, even such a simple system may be too complicated for the human brain to understand

Grasshopper-Mantis Model

As another example for conceptualizing such simple systems, a 3-com-ponent system, is shown in Figure 5.3 In this system, there are two kinds of grasshoppers and one kind of mantis, where mantes eat both kinds of grasshoppers The two kinds of grasshoppers compete with each other for resources The time-dependent changes in these three components are expressed by the following equations (Levins and Vandermeer, 1990):

dH1/dt  H1(r1 a11H1 a12H2 a13C) (5.1)

dH2/dt  H2(r2 a22H2 a21H1 a23C) (5.2)

600

400

200

1000

500

0

200

100

100 200

200

300 400

400

500

600 800

800 600 400 200 0

1000

0

0

Time Time

(b) (a)

C

Figure 5.3 Grasshopper-mantis model (Levins and Vandermeer, 1990) There are

two kinds of competitive grasshoppers and one kind of mantis H1, H2,

and C indicate the numbers of the two kinds of grasshoppers and

mantes, respectively The simulated results on the left and right sides

depict the cases for r3 1.25 and 1.0, respectively A large negative

r3indicates large cannibalism by mantes.

Here, H1, H2, and C denote the densities of the two kinds of grasshoppers and mantes, respectively, and a and r are positive constants, except for r3which denotes cannibalism by the mantes and has a negative value Time is

expressed by t Equation 5.1 indicates that the population growth rate of

grasshopper 1 is proportional to the quantity indicated in parentheses, where

r1is a growth coefficient assuming the absence of interspecific competition

and predation The negative terms are corrections to r1due to interactions with each of the three organisms Equation 5.2 for grasshopper 2 is very sim-ilar to the first equation Equation 5.3 applies to the mantis, whose popula-tion increases in proporpopula-tion to the quantities of the two kinds of grasshoppers, and decreases with their own cannibalism

In the first simulation, r3was set at 1.25 The results are shown on the

left side of Figure 5.3 What changes will occur if r3increases to 1 (i.e., can-nibalism decreases)? Intuitively, one would expect an increase in the popula-tion of mantes and a decrease in the populapopula-tion of grasshoppers due to increased predation However, as shown in Figure 5.3 (right panel), the pop-ulation of mantes did not increase, and the poppop-ulation dynamics of grasshop-pers were very different from our expectation This phenomenon is known as

an example of a chaotic event The above two examples, the 4-component and 3-component systems, indicate that even in such simple systems it is not easy to predict how the individual components interact with each other Predicting the behavior of and properly managing an actual agricultural ecosystem may be too difficult without appropriate methods such as system simulations (Edwards, 1990)

THE IMPORTANCE OF MATTER CYCLING IN THE NEW

AGRICULTURE

To grow crops with reduced amounts of fertilizers in agricultural ecosys-tems in the next generation, it is important to develop methods to accelerate nutrient cycling, and there are two approaches: activation of inactive ele-ments that are stored in the ecosystem, such as inactive nitrogen and phos-phorus in the soil; and acceleration of the turnover rate Examples of the former are utilization of phosphorus by plants after solubilization by phos-phate-solubilizing soil microorganisms (Kimura et al., 1991) and utilization

of mineralized nitrogen from microbial biomass and organic matter by dry-ing and heatdry-ing of soil (Okano, 1990), although they have not been developed

as a technology yet

Iwama et al (1992) reported an example of improvement of nutrient turnover rate through the introduction of intermittent grazing At the National Grassland Research Institute, Nagano, Japan, a pasture was seeded

in 1966 with tall fescue, orchard grass, timothy, red clover, and white clover The grass was then cut three times a year Starting in 1973, grazing was allowed in one part of the pasture after the second cutting each year Dry

matter plant yield was found to be dramatically higher in the grazed pasture than in the ungrazed pasture Although no direct, numerical data were provided, the nutrient turnover rate in the pasture where grazing was intro-duced was clearly accelerated through the animal-excreta-soil microbial-plant interactions In this section, we present examples of carbon and nitrogen flow in agroecosystems

Grassland Ecosystems

Energy flow and nutrient cycling have been analyzed in various ecosys-tems for the past twenty years These analyses are essential to obtain a more detailed description of a system’s productivity and nutrient cycling In agri-cultural ecosystems, solar energy is converted into chemical energy by pho-tosynthesis in crops Some of the energy is used by the plant for respiration, and the remainder is fixed as net primary production The energy of net pri-mary production is passed on to the other compartments, and finally it flows out from the system to the inorganic environment in various ways Understanding the balance between the energy or carbon inflow and outflow and also the transfer functions is essential for the study of the dynamic behavior of an ecosystem The energy or carbon budget in an agricultural ecosystem indicates the degree of stability of the soil fertility or the sustain-ability of the agricultural ecosystem To explore these ideas, we discuss the carbon and nitrogen budgets in grasslands and then compare them with the corresponding budgets in upland and paddy fields

Surveys of energy and matter budgets in a grassland have been carried out at the National Grassland Research Institute, located in central Japan, a region where the livestock industry has predominated on the main island of Japan, since 1974 These budgets have been measured at the plant, animal, and ecosystem levels on a yearly basis (Akiyama et al., 1984; Koyama et al., 1986; Takahashi et al., 1989) Based on these measurements, an energy, or car-bon, and nitrogen flow model was constructed (Shiyomi et al., 1988; Shiyomi

et al., 2000) The outline of the model is as follows: we assume that the amounts of energy and nitrogen and their time-dependent variations in each compartment are determined by their fluxes into and out of each of these compartments Thus, the time-dependent variation in the amounts of energy

and nitrogen at time t, x(t)’s, can be described by dx(t)/dt’s although the

equations are omitted here The concept of the model is illustrated in Figures 5.4a and 5.4b

Key parameters in the model are as follow:

1 Global solar radiation, Q, which changes over the course of a year

according to a sine curve (kJ m2day1)

2 Conversion efficiency of global solar radiation to photosynthesis

f  [1  (2.4L  1)  1]a(aQ  1)1, where L is the leaf area index and a is a constant.

A Light intensity

Leaf area index

Sun

Grazing intensity Digestibility

Amount of

standing dead

material

Amount of available herbage

Amount of unavailable herbage

Amount of herbage intake

by cattle

Body weight of cattle

Air temperature

Amount of feces Respiration

Amount of belowground portions

Soil organisms

Soil organic matter

Turnover rate of soil organisms Water content

Amount of litter

Figure 5.4a Energy flow compartment model for grazing grassland (Shiyomi et al.,

1988) “A” indicates the link between energy and nitrogen models.

3 Respiration-loss energy by plants is expressed by a linear relation

of daily air temperature, and heat-loss energy from cattle is a func-tion of body weight, digestibility, etc (kJ m2day1)

4 The herd ingests each day an amount of herbage (dry weight) equivalent to 2.5% of live cattle body weight (kJ m2day1)

5 The energy accumulation in a cattle body is given by (herbage intake, kgDM) (digestibility)  0.414, where 41.4% of digested energy is accumulated in the cattle body Digestibility is given by the equa-tion 619.6/(herbage biomass, kJ m2) 0.398 (Koyama et al., 1986)

6 The total amount of nitrogen lost from the soil, which includes the amounts absorbed by plants and runoff/leaching, is expressed by linear functions of the number of days counted from March 1

7 A 100 kg heifer excretes 58.0 gN as dung and 26.8 gN as urine each day

Mortality rate

Amount in above-ground portion

A

Ingestion rate

Amount ingested

by cattle

Cattle body

Amount in standing dead material T/R

balance

Amount in litter

Decomposition

rate

Soil organic matter

Soil organisms

Rate from standing

dead material

to litter

Amount in below-ground portion

Amount in soil

Turnover rate

Fixation

Legume biomass

Volatilization, leaching etc Application

Amount in excreta

Volatilization rate etc.

Crop growth rate

Figure 5.4b Nitrogen flow compartment model for grazing grassland (Shiyomi et al.,

1988) “A” indicates the link between energy and nitrogen models.

8 Legumes fix 0.011 to 0.012 gN m2day1

9 The nitrogen concentration in plant leaves affects the leaf area index, which is expressed by a logistic function of nitrogen concen-tration

An annual gain of 1 ton cattle body weight ha1was attained in an inten-sively managed pasture (IMP) at the National Grassland Research Institute, Tochigi, in 1986 (Kobayashi et al., 1989) The carbon and nitrogen budgets estimated using the systems model for the ecosystem in this pasture were compared with those estimated in an extensively managed pasture (EMP)

In a computer simulation of the IMP, seven young Holstein oxen were grazed on a 1-ha orchard grass-white clover pasture, where 160 kgN ha1yr1 was applied, for a period of 200 days from April onward Likewise, in a com-puter simulation of the EMP, three young Holstein oxen were grazed on a

1-ha orc1-hard grass-tall fescue-red top-white clover pasture, where 50 kgN 1-ha1

yr1was applied, for the same grazing period The results are shown in Table 5.1 If we suppose that the amounts of carbon in plant bodies in both the EMP and IMP do not change between the successive two years in the simulations,