Structure and Function in Agroecosystem Design and Management - Chapter 4 pot

CHAPTER 4
Ecological Management of
Crop-Weed Interactions
Chris Doyle, Neil McRoberts, Ralph Kirkwood, and George Marshall
CONTENTS
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Ecological Consequences of Modern Weed Control Systems . . . . . . . . . . . . 63
Weeds in the Ecosystem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Weed Adaptation to Management Practices . . . . . . . . . . . . . . . . . . . . 64
In Search of New Approaches to Weed Management. . . . . . . . . . . . . 64
The Role of Mathematical Models in Predicting Weed
Population Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Spatial and Temporal Dynamics of Weed Populations. . . . . . . . . . . . . . . . . . 66
The Dynamics of Weed Invasion and Spread. . . . . . . . . . . . . . . . . . . . 66
Predicting Weed Invasion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Seed Dispersal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
The Dynamics of Weed Population Density. . . . . . . . . . . . . . . . . . . . . 69
Optimum Weed Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Extrinsic Factors Affecting Weed Populations . . . . . . . . . . . . . . . . . . . 73
Weed Control Decision Thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
Timing of Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Optimal Weed Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Integrated Weed Management. . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Required Advances in Modeling Weed-Crop Interactions . . . . 78
Biological Control of Weeds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
Weed Adaptation to Management Practices . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Adaptation to a Single Control Measure. . . . . . . . . . . . . . . . . . . . . . . . 81
Adaptation to Integrated Weed Management Systems . . . . . . . . . . . 83
61
0-8493-0904-2/01/$0.00+$.50
© 2001 by CRC Press LLC

920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 61
62 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
INTRODUCTION
In recent years, two very different approaches to controlling weeds have
developed. On the one hand, there has been the introduction of herbicide-
tolerant crops in North America with their specific reliance upon herbicides.
Clearly, however, the widespread application of such techniques will alter
the dynamic equilibrium which normally exists in vegetation. Thus, a key
research issue must be the long-term ecological consequences of the regular
use of nonselective herbicides on the community structure of seminatural
vegetation (Willis, 1990). In direct contrast, in response to both public and
industry concerns, there has been the development of sustainable systems of
crop production, in which the emphasis has been on minimizing herbicide
use. Instead, a mixture of biological, chemical, and mechanical methods are
combined to control weeds, pests, and diseases to provide stable long-term
protection to the crop (Lockhart et al., 1990; Swanton and Weise, 1991;
Gressel, 1992; Wyse, 1994; Holt, 1994; Viaux and Rieu, 1995). Fundamental to
this latter approach is a sound understanding of weed demography and of
the efficacy and impact of different control methods. Although the two
approaches represent very different strategies to weed control, both require
an understanding of the population biology of weeds, including evolution-
ary aspects (Jordan and Jannink, 1997), and the dynamics of weed popula-
tions. Accordingly, this chapter summarizes current understanding on these
matters, including the effects of crop rotation, tillage systems, and herbicide
use on weed communities.
However, one of the most striking developments in regard to research
into improved management systems, with reduced dependency on herbi-
cides, has been a move towards systems type investigations. Thus, Kropff et

al. (1996) have stressed that the complexity of the population dynamics of
weeds and of the crop-weed interactions necessitates the use of mathemati-
cal models. Certainly, models of weed infestation, population growth, and
control have served as a valuable framework for organizing biological infor-
mation on weeds and for developing weed control strategies (Mortimer et al.,
1980; Doyle, 1991; Colbach and Debaeke, 1998). In particular, they have
helped to identify information gaps, set research priorities, and suggest con-
trol strategies (Maxwell et al., 1988). Furthermore, their value has arguably
extended beyond being simply useful research tools. Several key questions in
weed control cannot be answered using conventional field trials because of
the constraints of cost, time, or complexity (Doyle 1989; 1997). As such, mod-
els have come to serve as experimental test beds. Accordingly, this chapter
will deliberately treat the ecological management of crop-weed interactions
from a modeling and systems perspective.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 62
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 63
ECOLOGICAL CONSEQUENCES OF MODERN WEED
CONTROL SYSTEMS
Weeds in the Ecosystem
Any ecosystem, made up as it is of an integrated community of the
organisms present and their controlling environment, evolves over time into
a relatively stable community. Interactions at the physical, chemical, and bio-
logical levels lead to the establishment of dynamic interrelationships among
the species within the community and a degree of stability (Willis, 1990).
However, most crop production systems directly aim to produce monocul-
tures, as in arable crops, or simple mixtures of species, as in grass leys, in
order to maximize crop yield or economic profitability. This means disturb-
ing the “natural” vegetation of an area, either by introducing new species or
selecting out specific species at the expense of others. Weed control strategies
are concerned with controlling the unwanted species—a weed being defined

as “a plant growing where it is not wanted” (Buchholtz, 1967; Roberts et al.,
1982). Thus, the ingress of weeds into an area used for cropping is intrinsi-
cally an adjustment towards a more natural plant community.
Historically, weed control measures have been pursued to minimize the
damage done by weeds to crop yields and quality. Weed control practices have
typically involved a combination of periodic habitat disturbance through cul-
tivation and crop rotation and more recently the widespread use of herbicides.
On an ecological level, these practices have acted as a very powerful force
in the interspecific selection of weed flora through the mechanisms of pre-
adaptation, evolution, and alien immigration (Mortimer, 1990). Plant species
may be pre-adapted in the sense that they are resident in a natural plant com-
munity within dispersal distance of a crop and come to predominate within the
crop as a consequence of a change in management practices. The successful
invasion of a crop by a species from the natural habitat, therefore, depends on
a match of the life history characteristics of the weed to the habitat provided by
the cropping system. As such, the combination of management practices and
the pattern of crop development through time results in interspecific selection,
leading to particular species becoming “weeds” (Cousens and Mortimer, 1995).
However, management practices may give rise to interspecific selection as a
result of evolutionary processes. Where agricultural practices are continued
for a sufficient length of time and sufficient genetic variation occurs within a
species, locally adapted races of weeds are likely to arise (Mortimer, 1990).
Finally, where intensive agriculture is practiced, it is common for species not
endemic to the area to be present as weeds. While additional species are con-
tinually being introduced into agricultural environments, both inadvertently
by industry and consciously by seed firms, few alien species succeed in estab-
lishing themselves as damaging weeds. However, as with pre-adaptation,
changes in land management practices are often a critical ingredient, as wit-
nessed by the spread of Rhododendron ponticum in the U.K. (Mortimer, 1990).
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 63

64 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Weed Adaptation to Management Practices
The ability of weeds to adapt to changes in management practices is cer-
tainly one explanation for the persistent nature of crop yield losses to weeds,
despite technological advances (Ghersa et al., 1994; Cousens and Mortimer,
1995). Thus, observations by Fryer and Chancellor (1970) suggested that the
continued and widespread use of herbicides had markedly altered the com-
position of grassland weeds, but it was doubtful that it had led to the eradi-
cation of any weed species. In a specific experiment to examine the effects of
several herbicides on species composition over a 5-year period, Mahn and
Helmecke (1979) noted that, while the different herbicide treatments changed
the density and dominance of individual weeds, there was no change in the
species present in the community. Likewise, in a much longer trial involving
herbicides on wheat, run over more than thirty years, Hume (1987) observed
that no weeds were eliminated and no new species were able to invade the
community. The only changes in community structure were changes in the
relative abundance of species. Thus, fundamental to successful control of
weeds is an ability to predict the evolutionary dynamics of weed popula-
tions, as shaped by human and natural factors (Jordan and Jannink, 1997).
However, to make such predictions, a better understanding of the traits,
and especially the variation of those traits, that confer adaptation to weed
management practices is needed (Hartl and Clark, 1989). Focusing on the
evolutionary dynamics and mechanisms will allow questions of practical sig-
nificance in regard to ongoing weed adaptation to be addressed (Jordan and
Jannink, 1997). These include the speed with which weed adaptation can
erode the efficacy of non-chemical control methods. Insofar as weed adapta-
tion proceeds at a pace that negates technological advances in control, then
future research may need to concentrate on ways of impeding adaptation,
raising the issue of whether it is possible to design management systems that
inhibit weed evolution.

In Search of New Approaches to Weed Management
It is clear from the preceding discussion that, despite the high level of
crop management and the array of options at the disposal of farmers, weeds
continue to be a major problem. As Cousens and Mortimer (1995) noted,
some grass weeds have become increasing problems in cereal crops, requir-
ing new herbicides or major changes in cropping to ensure continued pro-
ductivity. Herbicide resistance is also on the increase. As a result, it is widely
accepted that programs in which weed control is almost exclusively achieved
by herbicides can be very unstable (Swanton and Weise, 1991; Gressel, 1992;
Zimdahl, 1993; Wyse, 1994; Shaw 1996). This acknowledgment, coupled with
increasing public concern about the levels of chemicals being used and their
potential environmental effects, has led to a renewed emphasis on long-term
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 64
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 65
weed management and the integration of a range of environmentally safe
and socially acceptable control tactics (Thill et al., 1991). Consequently, the
focus of much recent weed research has become the study of how crop yields
and weed interference are affected by changes in cropping management,
including tillage methods, the timing and rates of herbicides, cover crops,
and planting patterns (Swanton and Murphy, 1996). However, the efficacy of
what has become termed integrated weed management (Thill et al., 1991; Elmore
1996) clearly depends on a thorough understanding of the population
dynamics of weed communities and their constituent populations. In partic-
ular, it requires an understanding of
• the factors that determine the rates at which weeds spread;
• the rates at which they increase when they reach a given location;
• the maximum extent to which they will increase; and
• the ways in which the spatial spread and abundance of weeds can
be minimized and reduced (Doyle, 1991; Cousens and Mortimer,
1995).

For this reason, it has become fashionable to talk of the need to employ a
systems approach to the study of weed control (Müller-Schärer and Frantzen,
1996; Swanton and Murphy, 1996). The problem, as a number of researchers
(Cousens and Mortimer, 1995; Swanton and Murphy, 1996; Jordan and
Jannink, 1997) have pointed out, is that research into integrated weed man-
agement (IWM) has not progressed beyond description. However, to be of
practical use, IWM must move from a descriptive to a predictive phase. As
Cousens and Mortimer (1995) have underlined, most studies of weed popu-
lation dynamics are capable only of providing information on the outcomes
of management changes, but not on the processes involved. Equally, few
studies on integrated weed management have as their specific aim finding a
solution to specific weed management problems. Finally, the emphasis of
much work on natural communities is the prediction of long-term changes.
However, Cousens and Mortimer (1995) argue that, not only is predicting
long-term behavior of natural systems difficult, but it is also not what the
farmers are interested in. They are concerned with the short- to medium-term
consequences of their management actions and with plant communities that
may be in a state of unstable equilibrium. Given this, it is interesting to ask
the quality of our current ability to predict changes in weed populations.
The Role of Mathematical Models in Predicting Weed Population
Dynamics
Linking management changes to models of crop-weed interactions,
which include such issues as weed population dynamics and the ecophysical
basis of competition, permits the prediction of future weed problems and
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 65
66 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
their solutions, together with the economic risks and benefits involved
(Doyle, 1991; Doyle, 1997). Accordingly, the following discussion focuses on
the ability of mathematical models to predict the changes in weed populations
and the consequences of changes in weed management. The first part con-

siders the contribution of quantitative models to the understanding of the
spatial and temporal dynamics of weed populations. Central to this is an
appreciation of the types of factors driving population change. At any given
point in time, the state of a given weed population can be defined in terms of
its spatial limits, its total size, its density, and its composition. From the
moment that environmental and management changes occur, alterations in
the state of the population will occur; it is the dynamics of these changes
which are of interest.
Nevertheless, comprehending the changes in the spatial distribution and
abundance of weeds is only one element of weed management. It is necessary
to understand how different management practices influence the size and
spread of weed populations. Accordingly, the second part of the chapter
looks at the various attempts to use biological and ecophysical models to
explore the efficacy of integrated weed management systems. However, inso-
far as weeds adapt to management conditions, there is also a need to predict
weed evolution (Jordan and Jannink, 1997). Thus, the third and final part of
the chapter considers our ability to predict the speed with which weeds can
adapt to control measures and whether management systems can be
designed which impede weed evolution.
SPATIAL AND TEMPORAL DYNAMICS OF WEED
POPULATIONS
The Dynamics of Weed Invasion and Spread
As in medicine, prevention rather than cure is likely to be the most cost-
effective strategy, so understanding how and why weeds invade a given area
and being able to predict the pattern of spread is fundamental to control
(Doyle, 1991). However, only very recently has any attention been paid to
predicting the process of weed invasion. As late as the middle of the 1980s,
Mack (1985) reported that there were no mathematical models simulating the
spread of weeds. Part of the reason for this lack of models was that spatial
processes were given very limited consideration in weed management mod-

els, which were almost exclusively concerned with the temporal dynamics of
weeds. However, in the last decade there has been an increased interest in
understanding the processes involved in the spread of weeds at both the
national and regional level and within fields. The former has been driven by
a concern to limit the geographic spread of unwanted plant species, while the
latter gained impetus from the pressure to reduce herbicide usage and
increase the efficacy of any chemical control.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 66
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 67
Predicting Weed Invasion
The simplest model to simulate the geographic spread of weeds is
obtained by assuming that a species spreads outwards along a front at a con-
stant rate in all directions. If the distance advanced each year is r, and pre-
suming that the spread starts from a single focus, the area A occupied after t
years is given by
A ϭ
␲
(rt)
2
(4.1)
while the rate of annual increase in area is given by
ᎏ
d
d
A
t
ᎏ
ϭ 2
␲
r

2
t (4.2)
and the instantaneous proportional rate of increase is then
΂
ᎏ
d
d
A
t
ᎏ
΃
/A ϭ
ᎏ
2
t
ᎏ
(4.3)
Auld extended this simple model first by simulating the spread of weeds
from several foci (Auld et al., 1979) and then by incorporating it within a
model for predicting the population density of weeds (Auld and Coote,
1980). For any given site, the level of weed infestation in year t(P
t
) was pre-
sumed to increase according to the exponential model:
P
t
ϭ P
0
(1 ϩ c)
t

(1 Ϫ s)
t
(4.4)
where P
0
is the initial weed population at the site, c is the proportionate rate
of growth, as given in Equation 4.3, and s is the proportion dispersed away
from the site. The model was subsequently used (1) to simulate the possible
spread of serrated tussock (Nasella trichotoma) in southeast Australia (Auld
and Coote, 1981); (2) to gauge the potential costs of an effective regional con-
trol policy (Auld, Vere and Coote, 1982); and (3) to compare the costs of dif-
ferent strategies for controlling the spread of a localised weed population
(Menz et al., 1980)
Implicit in such a model is the assumption that weed seed is distributed
equally in all directions, so the spread may be described by a series of con-
centric circles. However, likening the spread of weeds to the ripples from a
stone dropped in water involves considerable simplification of reality (Mack,
1985). In practice, environmental heterogeneity and spatial irregularity are
likely to result in an uneven spread (Plumber and Keever, 1963; Rapoport,
1982). Random processes may also influence the observed pattern of weed
diffusion, as Skellam (1951) noted in a seminal study, which modelled the
areal spread of a plant population using random-walk techniques. As a con-
sequence, more recent research has focussed on identifying areas potentially
suitable for the growth of particular weed species. The earliest of these
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 67
68 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
studies, by Medd and Smith (1978), involved the development of a simple
model to predict the growth, phenological development, and seed yield of
musk thistle (Carduus nutans) from climatic data. Using the model, they were
able to determine areas within Australia that were suitable for the growth

and development of the weed, including uninfested regions. More recently,
Panetta and Mitchell (1991) have used a computer program to analyze the
climatic factors at locations where particular weed species occur in Australia
in order to describe the climatic profiles of the species and to examine the
possibility of the invasion of New Zealand by these species. Others, such as
Patterson et al. (1979), Williams and Groves (1980), and Patterson (1990) have
used experiments under controlled environment conditions to infer the
limits to the spread of particular weed species. The problem with all these
models that use climatic data to predict spread from present occurrences is
that there is no guarantee that climate is the limiting factor (Cousens and
Mortimer, 1995).
However, the recent advent of geographic information systems (GIS) has
allowed the spatial distribution of weeds to be mapped against a wider range
of limiting factors, including soils, management techniques, competitor
species, and climatic variables. As a consequence, it is possible to derive a
more complex picture of the environmental and ecological determinants that
favor the growth of a particular species. Such techniques have been used by
Prather and Callihan (1993) to study the efficacy of eradication programs
and by Wilson et al. (1993) to predict the environmental consequences of
weed control. Nevertheless, even these models do not strictly predict
whether a particular area will be invaded by a given weed species but rather
if it is possible.
Seed Dispersal
Although the spatial diffusion models discussed may describe the spread
of weeds, they are essentially descriptive models, in that they do not really
explain the mechanism through which dispersal occurs. As Cousens and
Mortimer (1995) outline, the mechanisms are complex, including dispersal by
wind, animals, water, and tillage operations, as well as vegetative spread.
However, quantitative studies of weed dispersal have been few and most
modeling work has focussed on wind dispersal. Thus, Smith and Kok (1984)

studied the factors responsible for the direction and distance over which the
seed of Carduus nutans was spread from a single point source. They found
that local seed dispersal was a function of wind velocity and the degree of
turbulence. Specifically, the observed seed dispersal could be described by a
Gaussian plume model, in which the concentration of seeds (C) at a point
(x,y,z) in three-dimensional space at a relative time (T) is given by
C(x,y,z,T) ϭ ͵
T
0
Q(t)C
0
(x,y,z,t)dt (4.5)
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 68
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 69
where C
0
denotes the rate at which seeds pass through the point x,y,z at time
t, Q(t) is the rate at which seed is released and T is the cumulative time since
the initial release of seed.
However, the Gaussian dispersal model was originally constructed to
describe movements of molecules in a gas cloud and so implicitly assumes
that particles will continue to disperse indefinitely. With heavy particles,
such as seeds, this is evidently not true. Thus, Johnson et al. (1981) used a dif-
ferent approach to predicting the distance (d) over which weed seeds would
disperse, assuming a steady wind and no turbulence:
d ϭ HU/V
s
(4.6)
where H is the release height, U is the wind speed and V
s

is the terminal
velocity of the propagule. However, while the model describes in some detail
the mechanisms by which seed is spread, in the absence of a population com-
ponent, it is difficult to see how it can be extended to study problems of weed
invasion on a field or regional scale.
A model that does combine mechanistic modeling of seed dispersal with
the life-cycle dynamics of a weed population was developed by Ballaré et al.
(1987). In their work, they simulated the population dynamics and spread of
Datura ferox in a soybean crop. Apart from a series of simple mathematical
expressions describing the life cycle of the weed, the model also included a
specific weed-dispersal algorithm, in which the spatial dispersion of the
weed over time was a function of both the dispersal characteristics of the
species and the type and direction of the combine harvester. The result is a
dispersion pattern in which the seed is principally spread in the direction of
the combine moves.
One weakness of these models of seed dispersal is that they describe the
likelihood of weed invasion solely in terms of proximity to an existing area of
infestation. While this may explain most of the observed spatial heterogene-
ity in weed incidence in arable crops, for perennial crops, such as forages,
past management practices and weather conditions may be just as important
in influencing the spatial configuration. In other words, the likelihood of
invasion may be as much a function of the susceptibility of the area to inva-
sion as it is to the proximity of the weed source.
The Dynamics of Weed Population Density
Given the presence of an infestation, using knowledge of the temporal
dynamics of weed populations, it should be possible to predict how fast the
weed population will grow in the absence of controls. Because of the com-
plexity of the problem and the long-term character of weeds, as early as 1980
Mortimer et al. (1980) were advocating the use of simple mathematical mod-
els of the life cycle of weeds to predict population densities. The current state

920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 69
70 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Mature
flowering
plants
Seed shed
Seed
bank
Emergent seedlings
Viable
seedlings
Predation Mortality
Seed rate
Germination
rate
Survival rate
Flowering
rate
Figure 4.1 Diagrammatic representation of a typical weed life-cycle model.
(Reprinted from Crop Protection, 10, Doyle, C.J., Mathematical Models
in Weed Management, 432–446. Copyright 1991, with permission of
Elsevier Science.)
of the attempts to model life-cycle processes has been described in Doyle
(1991), Cousens and Mortimer (1995), and Kropff et al. (1996). In general,
comprehensive models based on physiological principles are only available
for parts of the life-cycle, such as plant growth, competition (Kropff and Van
Laar, 1993), germination, and emergence (Vleeshouwers and Bouwmeester,
1993). Instead, most models encompassing the whole life cycle have repre-
sented it in terms of a series of growth stages, as diagrammatically repre-
sented in Figure 4.1. The complex processes involved in the transition from

one stage to the next are then “blended into a few lumped parameters like a
germination rate, a reproduction rate and a mortality rate” (Kropff et al.,
1996, p. 7). Good examples of such models are Cousens et al. (1986), Doyle et
al. (1986), and Van der Weide and Van Groenendael (1990).
However, the detail in which the life-cycle processes in weeds are stud-
ied is only one issue. More critically, there are various ways to extract the
population dynamics from the life-cycle processes, and these different ways
may lead to different results (Durrett and Levin, 1994; Kropff et al., 1996). In
particular, three different approaches to modeling the integration of indi-
vidual weed plants into a population have been adopted. Kropff et al. (1996)
stylized these as (1) the density-based models, (2) the density-based models
incorporating spatial processes, and (3) the individual-based models
accounting for spatial processes.
Of these, the most frequent modeling approach has been to assume that
the key determinant of rates of population growth is the density of the weeds.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 70
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 71
From the current density, the rate of population growth is derived to give the
new density value. From the middle 1970s, researchers such as Hassell (1975),
Bellows (1981), and Law and Watkinson (1987) were modeling the dynamics
of single species over generations using nonlinear difference equations of the
following type:
N
tϩ1
ϭ RN
t
(1 ϩ N
t
)
Ϫb

(4.7)
where N
t
is the population size in period t, R is the asymptotic per capita
increase in a population of uncrowded individuals, and a and b parameters
describe the form and intensity of self-regulation. At high densities b reflects
the extent to which a population compensates for a change in density. Such a
model has been found to apply readily to plants with discrete generations
and no persistent seed bank. However, such models can be expanded to
incorporate a seed bank or to model species behaviour in mixtures (Mortimer
et al., 1989). Thus, the effect of introducing a second species into a monocul-
ture is presumed to be a reduction in yield per unit area and the per capita
rate of growth of the first species. For a multispecies assemblage, comprising
three species, N
0
, N
1
, and N
2
, Equation 4.7 specifically becomes
N
0,tϩ1
ϭ (4.8)
where
␣
,
␤
,
␥
, and

␦
are parameters and R is the per capita growth, where the
density of plants is low. However, an implicit assumption in this approach is
that each weed experiences a similar environment, so that it is impossible to
incorporate the spatial dispersal of weeds (Kropff et al., 1996).
A rather obvious way of including the dispersal of weeds is to include
space in the model and allow for spatial gradients in density. This has led to
the so-called reaction-diffusion models. Versions of this type of model have
been used to simulate the spread of weeds (Auld and Coote, 1980; Ballaré et
al., 1987; Maxwell and Ghersa, 1992). The key variable still remains weed
density, but it is now possible to look at spatial processes. Thus, a recent
model developed by González-Andujar and Perry (1995) has enabled the
examination of weed dynamics within patches over time, as well as permit-
ting the testing of hypotheses about patch persistence and the extent of seed
dispersal. However, as Kropff et al. (1996) have pointed out, over time the
spatial gradients in these models either move or flatten out. As a result, for
any particular site, this approach to modelling weed density and dispersal
rapidly reduces over time to modelling density alone.
One step further is to abandon weed density as the basic variable in the
model and proceed with the configuration of weeds over space. This is the
modelling approach adopted by Antonovics and Levin (1980), Weiner (1982),
Goldberg and Werner (1983), Barkham and Hance (1982), Pacala and Silander
(1985), Silvertown et al. (1992), and Wallinga (1995). Although a distinction
RN
0,t
ᎏᎏᎏᎏ
[1 ϩ
␣
(N
0,t

ϩ
␥
N
1,t
ϩ
␦
N
2,t
)]
␤
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 71
72 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
can be made between the individual-based models (e.g., Pacala and Silander,
1985) and cellular automaton models (e.g., Silvertown et al., 1992), the under-
lying principles can be understood by examining the work of Pacala and
Silander (1985, 1987). The basic idea is that the performance of an individual
plant or cell can be determined from the number, distance, and type of neigh-
bors. For each individual species, the population, dynamics is described in
terms of a series of sequential steps, comprising seed dispersal, germination,
seedling survival, and seed production. The novel aspect is that these
processes are formulated for single individuals, which germinate, grow, and
yield seed that is dispersed into a defined area from which new individuals
are established. Essentially, the objective is to estimate a neighborhood area
within which there is interference from neighbors on the target plant and out-
side which the effects are negligible. This estimation is achieved by deter-
mining statistically the relationship between various biological processes,
such as seed production, and the number of neighbors within a defined
radius from the target plant. By varying the radius, the appropriate neigh-
bourhood size can be determined. Knowing the radii together with the den-
sity of species, estimates can be made of the number of neighbors and the

consequent impact on a given biological process, such as seed production per
individual plant, assuming the species are randomly distributed. Allowing
for seed dispersal and germination, the level of infestation for the next year
can be projected.
However, even though these models can readily accommodate multiple
species, the application of models based on individuals and including spatial
aspects is likely to be restricted. As both van Groenendael (1988) and Kropff
et al. (1996) have observed, they are very difficult to parameterize and com-
putationally slow. For this reason, there has been a resurgence of interest in
the simple density-based models. Recently Mortimer et al. (1996) extended
the basic model given in Equation 4.8 to include spatial heterogeneity. This
involves treating weed populations as sets of sub-populations in a frag-
mented landscape interconnected by dispersing propagules. Accordingly,
they added to the basic growth function a probability distribution function
that describes the spread of propagules from each plant. Assuming a field
comprising n patches, each with a certain level and composition of weeds,
then the density of the weed population in patch x at time t is given by
N
x,t
ϭ
Α
n
y
ϭ 1
P
x,y
f [N
y,tϪ1
] (4.9)
where f[N

y,t
] is the population growth function at patch y and P
x,y
is the prob-
ability that seed will disperse from patch y to patch x. Although Mortimer et
al. (1996) confined their analysis to a linear, single dimension habitat, it is rel-
atively easy to generalize to a two-dimensional habitat.
To parameterize this model, it is necessary to decide two key issues. The
first is the form of the probability distribution function, P
x,y
. Except where
seed dispersal is affected by cultivation (Ballaré et al., 1987), it is probably not
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 72
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 73
unrealistic to assume a weed plant disperses propagules symmetrically
around each individual on a normal (Gaussian) distribution. Where wind or
cultivation leads to a skewed distribution, it can be represented by the use of
a generalized (or skewed normal) distribution. The second issue concerns the
choice of growth function. One possible functional form is Equation 4.8
Streibig et al. (1993) contended that this growth function was adequate for
predicting compositional change. Certainly, using this function, Mortimer et
al. (1996) were able to describe the spatial and temporal stability of weed
populations.
OPTIMUM WEED MANAGEMENT
Extrinsic Factors Affecting Weed Populations
So far, attention has focused on the dynamics of weed populations under
a constant environment, where population changes are driven solely by
intrinsic processes, such as intraspecific competition. However, the environ-
ment of a weed population is rarely constant, with factors such as manage-
ment, weather conditions and interactions with other organisms varying

both within and between generations. As Cousens and Mortimer (1995) have
observed, the relative importance of the different factors will vary with year,
geographic location, and habitat. However, insofar as weather and disease
factors are unpredictable and uncontrollable, most attention has focused on
how crop management can affect weed populations. By using this knowl-
edge, hopefully better weed control strategies can be developed. In particu-
lar, an understanding of the effects of management practices on the
composition and density of weed populations offers not only the prospect of
being able to predict the consequences of a particular management change,
but also the ability to determine the most effective and economic method of
controlling a particular weed. The development of management systems
with reduced dependency on herbicides has only shifted the emphasis still
further towards the management of weed populations through husbandry
practices (Kropff et al., 1996; Swanton and Murphy, 1996).
To achieve effective control of weeds requires the ability to answer three
questions:
1. What level of weed infestation justifies intervention?
2. At what stages during the weed life-cycle should intervention
occur?
3. How should the weeds be controlled?
As Doyle (1991; 1997) and Cousens and Mortimer (1995) have underlined, the
most powerful technique at our disposal for answering such questions is
mathematical modelling, coupled with experimental verification.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 73
74 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Accordingly, this section reviews how far it has been possible to incorporate
our knowledge of interactions between weeds and crop management into
models so as to provide quantitative insights into effective and environmen-
tally sustainable control techniques.
Weed Control Decision Thresholds

The prophylactic use of herbicides has come under increasing opposition
in the last decade and ways of reducing both the frequency and application
rates of chemicals have been investigated (Jordan and Hutcheon, 1993;
Turner, 1993; Elmore, 1996; Swanton and Murphy, 1996). Specifically, atten-
tion has focused on the question of what level of weed infestation justifies
intervention, which has been widely approached using economic threshold
modeling. In itself, the concept is easily understood and can be summarized
as follows: as the weed population per unit area increases, the gain in crop
yield from chemical control becomes greater than the cost of the control
measures. The threshold density is where the cost of the control is equal to the
net benefit from control. Provided that appropriate means for estimating
weed densities are available, then the theory is that the practical application
of the threshold concept will merely involve the farmer in judging whether
the actual level of infestation exceeds the critical threshold density. Examples
of such threshold models include Marra and Carlson (1983), Doyle et al.
(1984), Cousens et al. (1985), Cousens et al. (1986), Auld and Tisdell (1987),
Cousens (1987), Dent et al. (1989), Moore et al. (1989), Streibig (1989),
Mortensen et al. (1993), Swinton and King (1994), González-Andujar and
Perry (1995), Buhler et al. (1997), and Baziramahenga and Leroux (1998).
However, threshold models have come under attack in recent years on
four counts (Doyle, 1997). First, they are dependent on experimental evi-
dence regarding weed-crop competition. In many instances, the experiments
are conducted at weed densities that are of limited relevance to the determi-
nation of economic thresholds (Dent et al., 1989). Second, the vast majority of
threshold models developed have assumed that the weeds are uniformly dis-
tributed across the field. However, many weed species exhibit a marked ten-
dency to cluster, leaving large areas of a field relatively free of infestation.
Compared with a field in which the weeds are uniformly distributed, the
impact on crop yield will be less and the consequent threshold density will
tend to be higher (Dent et al., 1989; Brain and Cousens, 1990; Wiles et al., 1992;

Johnson et al., 1995; Mortensen et al., 1995; Wallinga, 1995; Lindquist et al.,
1998). To simulate the effect of patchy distributions of weeds, Brain and
Cousens (1990) developed a model incorporating a non-random distribution
of weeds. Essentially, it assumed that a field could be divided into a grid of
1 m
2
subplots. While within each plot the weeds were considered to be ran-
domly spread, the number of weeds per subplot were described by a negative
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 74
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 75
binomial distribution, which has been found to fit most weed seedling count
data (Johnson et al., 1995). The proportion of subplots containing weeds was
then a function of the mean weed density over the entire field (D) and the
degree of clumping (k). Assuming that for each subplot the effect of weed
density on crop yields could be represented by a hyperbolic function, then the
proportionate crop yield loss (Y
L
) is represented by
Y
L
ϭ
␣
͵
1
0
Dz
1/
␤
΂
ᎏ

k ϩ D
k
(1 Ϫ z)
ᎏ
΃
k ϩ 1
dz (4.10)
where α and β are estimated parameters, and z is a variable lying between 0
and 1. Lindquist et al. (1998) showed that where the mean weed density (D)
and the clumping factor (k) were known, an accurate estimate of field-scale
crop yield losses could be obtained.
The third criticism of threshold models is linked to the existence of
uncertainty (Auld and Tisdell, 1987). In weed control, there are three prin-
cipal sources of uncertainty that may modify the perceived optimal thresh-
old density for spraying: (1) the potential weed density; (2) the form of crop
loss function; and (3) the form of the herbicide dose-response function. A
major factor in deciding whether to use a herbicide is the size of the weed
population. Where a pre-emergent herbicide is to be used, then there must be
uncertainty about this. Second, although the general form of the crop
loss function may be known, its precise shape varies with location and agro-
nomic factors (Reader, 1985; Cousens et al., 1988). Thus, the economic thresh-
old for spraying will vary accordingly. Finally, the efficacy of a given
herbicide in controlling a weed infestation is sensitive to site and manage-
ment practices (Zimdahl, 1993). Not only do these factors mean that the eco-
nomic threshold density for a weed is subject to uncertainties, but the very
existence of uncertainty is known to modify grower behavior (Doyle, 1987;
Auld and Tisdell, 1987; Pannell, 1990). If farmers are risk averse, then they are
more likely to use herbicides in a prophylactic way and to apply them annu-
ally as a security against weed invasion (Cousens and Mortimer, 1995). The
consequence of all this is that specific weed threshold densities become less

relevant.
The final major conceptual problem with threshold models is that, in
practice, treating the damaging external effects of herbicides as a cost is not
really workable. Apart from the problem of whether environmental damage,
such as loss of plant and species diversity, can be measured in economic
terms, the resultant threshold densities may be unacceptable. Basically, the
effect of increasing the overall costs of applying chemical control is to
increase the threshold weed density at which significant crop losses occur
and which the grower would not be prepared to tolerate. Thus, in the absence
of alternative means of controlling weeds, the credibility of the predicted
thresholds is subject to attack.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 75
76 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
Timing of Control
More recently, interest in the threshold level at which weed control is jus-
tified has been replaced by consideration of when to apply the control meas-
ures during the life cycle of the weed. Although this has become integrally
bound up with moves towards non-chemical methods of control, the interest
predates the current focus on integrated weed management. In particular,
attention has focused on the relative efficacy of killing weed seeds rather than
controlling weed plants once they have emerged. In the early 1980s, Cussans
and Moss (1982) used an exponential multi-stage model of the annual grass
Alopecurus myosuroides to investigate the benefits of different cultivation tech-
niques to influence seed germination. The model was subsequently extended
to include density-dependent plant mortality and seed production (Cousens
and Moss, 1990). Medd and Ridings (1989) similarly investigated the relative
merits of seed versus plant kill using a three-cohort model of the life cycle of
wild oats Avena fatua. They were able to show that if relatively small improve-
ments in seed kill could be achieved, in conjunction with herbicides, signifi-
cant improvements in the rate of decline of weed populations could be

obtained. Finally, Pandey and Medd (1990) combined the technique of
dynamic programming with a population model of Avena species to examine
the efficacy of controlling weed seeds.
The conclusions about the importance of weed seed kill reflect more gen-
eral evidence from plant competition studies that the period between crop
and weed emergence is a critical factor which contributes to reductions in
crop yields. Accurate information on dates of weed emergence has been espe-
cially important in determining potential crop yield losses. However, the
practical difficulty is obtaining the daily information required (Kropff, 1988)
for models able to predict weed seedling emergence to be of practical benefit
(Forcella, 1993). Specifically, González-Andujar and Fernandez-Quintanilla
(1991) developed a population model of Avena sterilis, in which there were
two quite distinct periods of seedling emergence. Using the model, they were
able to show that two of the most critical factors influencing weed population
levels were the dispersal and mortality of seeds during the summer and the
fecundity of the first cohort of seedlings to emerge. Thus, they were able to
pinpoint the critical stages in the life cycle of A. sterilis as far as achieving
effective control was concerned. Elsewhere Grundy et al. (1996) and Prostko
et al. (1997) have focused on the influence of the distribution of weed seeds
within the soil profile on seedling emergence. While Prostko et al. (1997) used
Fermi-Dirac distribution functions to model weed emergence as influenced
by depth of weed seed burial, Grundy et al. (1996) used a simulation model
with several soil layers. The significance of these “seed burial” models is that,
by combining them with models that determine the effects of cultivation on
seed distribution, it should be possible to improve the predictions of seedling
emergence from the seed bank.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 76
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 77
Optimal Weed Management
Integrated Weed Management

A logical extension of the investigations into the most critical and effec-
tive stages in the life cycle of weeds for controlling them is the optimal
method of control. Thus, many of the studies concerned with examining the
relative efficacy of weed seed and plant killing investigated the merits of
chemical and non-chemical methods of control. In a review covering 1984 to
1996, Colbach and Debaeke (1998) found no less than twenty-six weed
demography models, which incorporated some cropping system effects. The
majority considered soil tillage and herbicide applications, but under 20%
considered other cultivation techniques, such as crop cultivar, sowing date,
sowing density, harvesting, or stubble burning. Few, if any, explicitly inte-
grated the effects of crop management. Thus, in most models, a constant
seedling mortality rate is associated with a set of weed control methods. For
chemical methods of control, the rate is typically determined by dosage and
active ingredients, and for mechanical control it is determined by timing of
tillage operations. However, as Debaeke and Sebillote (1988) observed, inter-
action between cultivation methods and weather conditions is frequently
critical in determining mortality rates. Likewise, the process of weed seed
dispersal is never considered, yet wind-borne seed from outside the field can
play a significant role in determining levels of weed infestation. Finally,
although some demographic models of weeds include consideration of the
patchy distribution of weed species, none of the models considering the
effects of cultivation practices, researched by Colbach and Debaeke (1998),
assumed anything but a uniform distribution of weeds.
The significance of this is that public concern about the environmental
costs of continued reliance on chemical methods of weed control has led to
the search for more sustainable practices that rely on a reduced use of all
inputs as a means of safeguarding natural resources and minimizing the neg-
ative impacts on the environment. This research has given birth to the con-
cept of integrated weed management, in which attention is focused on how
changes in crop management practices, such as tillage methods, planting pat-

terns, and the use of cover crops, can minimize the need for herbicides (Burn,
1987; Elmore, 1996; Swanton and Murphy, 1996). Certainly, through the use
of mathematical models, (Cussans and Moss, 1982; Wilson et al., 1984; Medd
and Ridings, 1989; Cousens and Moss, 1990; Pandey and Medd, 1996), it has
been shown that combining weed seed kill through cultivation practices with
a reduced herbicide application can be more cost effective than relying solely
on killing weed plants by chemical means. However, as the review by
Colbach and Debaeke (1998) revealed, weed population models must be
improved in three key areas if they are to make a tangible contribution to the
evaluation and management of cropping systems: (1) incorporation of
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 77
78 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
weed-crop interactions, (2) degree of detail in the description of crop man-
agement, and (3) the explicit recognition of spatial heterogeneity of the weed
population.
Required Advances in Modeling Weed-Crop Interactions
First, there is a need to incorporate the interaction between critical
processes, such as seed viability, seedling survival, and seed production, and
major cultural techniques. Thus, light (Ampong-Nyarko and de Datta, 1993;
Barbour and Bridges, 1995; Holt, 1995), water stress (Paterson, 1995), fertil-
ization strategies (Ampong-Nyarko and De Datta, 1993; di Tomaso, 1995),
and tillage methods (Rasmussen, 1993; Dyer, 1995) have all been shown to be
potential tools in managing weed levels. However, many of the current mod-
els used to explore the effects of cultivation techniques on weed infestation
levels do not separate direct weed-crop effects, such as shading and compe-
tition for water and nutrients, from indirect effects related to cultivation prac-
tices, such as soil tillage and date of sowing. Equally, the efficacy of
herbicides in killing weeds is usually assumed to be independent of the weed
emergence pattern and weather conditions, despite evidence to the contrary.
The exceptions are the models by Aarts (1986), Danuso and Zanin (1989), and

Debaeke (1988), which calculate weed control efficacy as a function of the
weed emergence pattern.
Second, there is a need for a much more complete description of crop
management, including the choice of cultivar, sowing dates, crop density,
and weeding and harvesting methods. The effect of crop cultivar on weed
dynamics has seldom been investigated, despite evidence from studies
involving wheat (Triticum aestivum) that there are noticeable cultivar effects
on weed population levels (Moss, 1985; Grundy et al., 1993). Only Melander
(1993) has modeled a cultivar effect on weed fecundity. Similarly, the sowing
date of the crop is widely reported to affect weed infestation levels
(Springensguth, 1960; Schneider et al., 1984). Only the models of Aarts (1986)
and Debaeke (1988) explicitly incorporate sowing date effects. Similarly, crop
density is important in determining seedling survival and fecundity for
many weed species. However, while most models include the effect of the
weed density on population weed dynamics, only a few of the most recent
models (eg., Wiles et al., 1996) specifically incorporate the effect of crop den-
sity on weed seedling survival.
The third area for improvement concerns the incorporation of within-
field variability of weeds. Specifically, variability within a field can take two
forms. First, weeds are not necessarily uniformly distributed in a field but
occur in patches. Second, a weed population is composed of different geno-
types and phenotypes. Both kinds of variability strongly interact with the
cropping system. The best example of this is given by herbicide efficacy. In a
field with a patchy weed distribution, herbicide rates can be excessive for
areas of low weed density and insufficient to destroy patches with high
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 78
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 79
densities. Knowledge of weed genotypes and phenotypes becomes increas-
ingly important because the development of herbicide resistance means that
only those plants with a sensitive phenotype will be killed by herbicide appli-

cations.
However, only a few studies directed at examining the interactions
between cultivation practices and weed populations have integrated
intrafield variability into the model. One such model was developed by
Jordan (1993) to examine the effects of tillage techniques on ridge-tilled corn.
In this model, a spatial distinction was made between the ridge and the fur-
row. A different approach was adopted by Ballaré et al. (1987) and González-
Andujar and Fernandez-Quintilla (1991). They treated a field as comprising
a series of subunits, for each of which the weed population dynamics were
modeled separately. Likewise, weed models that integrate both dynamic and
genetic aspects are uncommon. Examples include the models by Maxwell
et al. (1990) and Colbach and Meynard (1996), which are concerned with
predicting the evolution and dynamics of herbicide resistance. These are
discussed in more detail in the next section of this chapter.
Thus, to represent weed-crop interactions with sufficient realism, appre-
ciable changes are needed. Currently, the most complete descriptions of such
interactions are provided by the eco-physical models of Graf et al. (1990),
Wilkerson et al. (1990), Kiniry et al. (1992), Kropff and Spitters (1992), Kropff
et al. (1992), Weaver et al. (1992), Ball and Schaffer (1993), Dunan et al. (1994),
and Lindquist and Kropff (1996). These models simulate annual competition
for light, water, and nutrients between a crop and one or more weeds. As
such, they are suitable for exploring crop management effects, including
sowing date, crop density, nitrogen fertilization, and weeding, on weed bio-
mass and seed production. However, these models are not without their lim-
itations. In particular, they generally do not describe the dynamics of the
weed population in terms of evolution from seed to mature plant. Moreover,
they require the specification of a considerable number of parameters, for
many of which physiological data are lacking. The lack of ecophysiological
data on weeds is a true limitation to the wider use of these models. For this
reason, Colbach and Debaeke (1998) have argued that rather than integrating

ecophysiological models directly into weed population models, the former
should be used only to generate parameter values for the latter with respect
to seed mortality, seedling survival, and weed fecundity under different
crops, soil types, and weather conditions.
Biological Control of Weeds
Recently, there has been increased interest in and emphasis on biological
methods of weed control involving rhizobacteria (Kremer and Kennedy,
1996; Johnson et al., 1996), bioherbicides (Charudattan et al, 1996; Jackson et
al., 1996) and weed-feeding insects (Cofrancesco et al., 1984; Messersmith
and Adkins, 1995; Rees and Paynter, 1997). In general, biological control aims
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 79
80 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
to reduce the abundance of the weed to a level that can either be tolerated or
managed by other measures rather than be totally eradicated (Cousens and
Mortimer, 1995). However, expectations regarding the potential of biological
controls have arguably been inflated by much publicized successes with con-
trolling a few major weeds (Cousens and Mortimer, 1995; Auld and Moran,
1995). In practice, understanding the impact that biological control agents
will have on target weeds is complicated. First, the population biology of the
weed is just as important as that of the control agent (Rees and Paynter, 1997).
Second, controlling some weeds effectively may involve the release of several
control agents, possibly with different climatic preferences and genotypic
specificity (Cousens and Mortimer, 1995).
Accordingly, although uncommon, simulation models of biocontrol
(Frank et al., 1992; Lonsdale et al., 1995; Rees and Paynter, 1997) are likely to
be important in the successful application of such techniques to weed control.
Certainly, the power of modeling is exemplified by the spatial model of
Scotch broom Cytisus scoparius developed by Rees and Paynter (1997). Using
a population model of broom, which incorporates locally density-dependent
competition, seed dispersal, and an age-structured population of established

plants, Rees and Paynter were able to study the effects of introducing a seed-
eating insect on the growth, mortality, and seed production of broom. In par-
ticular, they projected that seed-feeding insects would reduce the longevity
of broom and impede its spread. However, what the model critically did not
predict was whether the elimination of Scotch broom would increase the
presence of desired plant species or whether other weed species would fill
the niche left. Thus, Burdon et al. (1981) reported on the successful control
of Chondrilla juncea using the rust Puccinia chondrillina. However, they also
noted that broad-leaved weeds, unaffected by the rust, replaced C. juncea.
Therefore, understanding adaptation of the weed community may be criti-
cal to effective biological control techniques, which is the focus of the next
section.
WEED ADAPTATION TO MANAGEMENT PRACTICES
At present the evolutionary potential of weed populations is a minor
consideration for weed managers (Bhowmik and Norris, 1996), although the
development of herbicide resistance is seen as a possible threat to chemical-
based weed management systems. However, there is growing evidence that
weeds are capable of adaptation to biological, mechanical, and cultural con-
trol practices and not just herbicides (Cavers, 1985; Barrett, 1988; Warwick,
1990; Gould 1991). This is likely to assume increasing significance as the
search for more sustainable weed management systems limits the range of
acceptable control methods. Taken together, the adaptive capability of weeds
and the likelihood of more stringent controls on weed management in future
cropping systems suggest that effective methods of crop protection against
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 80
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 81
weeds may need to be regarded as scarce resources (Jordan and Jannink,
1997). Thus, understanding weed evolutionary biology and how manage-
ment systems may actively impede weed evolution could well be the key to
future weed control. Accordingly, this section considers two issues:

• What do we know about the speed of weed evolution to a single
control measure?
• Will the adoption of integrated weed management systems impede
weed evolution?
Adaptation to a Single Control Measure
As Jordan and Jannink (1997) explain, the speed with which weeds adapt
to management practices is dependent on both the genetic variation in the
population and the selection process involved in adaptation. Most attention
has understandably focused on the development of herbicide resistance as it
reduces the efficacy of the key control method in modern farming systems
(Jasieniuk et al., 1996). As the biological factors that determine the evolution
of herbicide resistance are complex, a number of researchers have turned to
models to explore the issue. Thus, in the mid-1980s, May and Dobson (1986)
developed a general analysis of the evolution of pesticide resistance by con-
sidering the changes in allele frequency over generations with repeated
applications of pesticide. The resultant model is equally applicable to herbi-
cide resistance in weeds, at least where the resistant dominant genes are the
determinants of resistance. Assuming that the population of weeds contains
a susceptible allele, S, and a resistant dominant allele, R, then May and
Dobson predicted that the absolute time taken (T
R
) for 50% of the population
to be resistant to herbicide (r
f
ϭ 0.5) is given by
T
R
ϭ T
g
log

e
΂
ᎏ
0
r
.
0
5
ᎏ
΃
/log
e
΂
ᎏ
w
w
R
SS
S
ᎏ
΃
(4.11)
where T
g
is the time taken for a generation of population growth, r
0
is the ini-
tial frequency of the resistance allele, and w
RS
/w

SS
is the strength of selection.
As a general conclusion, May and Dobson noted that even if r
0
varies in the
range 10
Ϫ5
to 10
Ϫ6
, w
RS
/w
SS
in the range 10
Ϫ1
to 10
Ϫ4
, and T
g
is 1 year, then T
R
will lie in the range of 10 to 100 years, assuming recurrent selection.
Subsequent attention has focused on the selection “pressure” (w
RS
/w
SS
)
exerted by herbicides, which Maxwell and Mortimer (1994) observed will
depend on both the intensity and duration of selection. The former is a meas-
ure of the relative mortality exerted on a genotype and/or the relative reduc-

tion in seed production of survivors, and it will be related in some degree to
herbicide dose. The duration of selection is a measure of the period of time
over which phytotoxic effects occur. Both the intensity and duration will
interact to produce seasonal variation in the selection process, which in turn
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 81
82 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
will depend on the phenology and growth of the weed species. For instance,
with pre-emergent herbicide control of weeds that show germination over a
protracted period, the intensity of selection may be much higher on weed
seedlings recruited early in the life of a crop than seedlings emerging later.
The situation is further complicated by the presence of weed seed banks,
which in evolutionary terms represent a memory of past selection events and
which may buffer evolutionary processes and delay the onset of herbicide
resistance (Cousens and Mortimer, 1995). This situation was recognized by
May and Dobson (1986) who modified Equation 4.11 to incorporate a seed
bank. The resultant mathematical relationship was analogous to an alterna-
tive model, derived independently by Gressel and Segel (1978, 1990).
More recently, simulation modeling has been used to explore the evolu-
tion of herbicide resistance (Maxwell et al, 1990, Mortimer et al., 1992). In the
model by Maxwell et al. (1990), the flow of genes is seen as directly altering
the proportion of herbicide-resistant and nonresistant alleles in the weed
population. Herbicide-resistant genes are introduced into the population
both by immigration of pollen and seed and by genetic drift within the exist-
ing population. Attempts to manage the resistance then involve two distinct
strategies: the use of alternative herbicides to remove resistant plants and the
manipulation of the nonresistant type gene to increase its incidence in the
population. Maxwell et al. (1990) concluded from the modeling exercise that
the latter may be more cost effective.
However, despite the focus on herbicide resistance, weed adaptation may
become a greater issue with integrated weed management. Thus, Jordan and

Jannink (1997) note that many non-chemical systems of weed management
rest heavily on one particular control measure. For example, Ghersa et al.
(1994) suggested an innovative way to prolong the useful life of a herbicide by
manipulation of the patterns of selection imposed by management actions.
The basic step was to sow the crop earlier, leading to crop establishment
before the weed emerged and greater suppression of the weed by the crop.
Less herbicide use would then be needed and selection for early germinating
rather than later germinating herbicide-resistant weed genotypes would arise.
This selection was expected in this case to lead to greater frequencies of early
germinating, nonresistant weed genotypes. Subsequently, later sowing of the
crop and higher herbicide rates could be resumed to reverse patterns of selec-
tion. By continuing this cycle, the useful life of the herbicide would theoreti-
cally be extended. However, Jordan and Jannink (1997) observed that the
efficacy of the scheme rested on three assumptions about variation and selec-
tion in the weed population: the population must be genetically variable for
germination timing, selection against late-germinating resistant genotypes
must be highly effective, and late germination and herbicide resistance must
have some persistent association among genotypes. This situation illustrates
the need to develop a much more detailed understanding of the dynamics of
weed adaptation, which in turn requires more detailed case studies like those
of Putwain et al. (1982) on triazine-resistant Senecio vulgaris.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 82
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 83
Adaptation to Integrated Weed Management Systems
More critically, the well-attested ability of weeds to adapt to control
measures raises the question of what will be the durability of integrated
weed management (IWM) systems, in which multiple control measures are
deployed together. In theory, to adapt, the weeds would be required to evolve
successfully multiple traits, which is not very probable. However, as Jordan
and Jannink (1997) observed, it is not self evidently true that weeds cannot

effectively adapt to IWM systems as a whole. Thus, Jordan (1989) used mul-
tivariate selection analysis (Lande and Arnold, 1983) to estimate the rate at
which the annual dicotyledonous weed (Diodia teres Walt.) would adapt to a
soybean (Glycine max) cropping system by evolving a phenotype similar to
that of weed biotypes infesting soybean fields. The analysis indicated that the
D. teres population would evolve a multiple trait phenotype, similar to the
adapted weed population, within several decades. As such, the weed man-
agement system studied was clearly not sufficiently diversified to prevent
effective adaptation.
For effective adaptation to be prevented, Jordan and Jannink (1997)
emphasised that IWM must impose genostasis; it is structured in such a way
that it deprives the weed species of sufficient genetic variation to permit
adaptation to the controls. However, no examples of the imposition of geno-
stasis by diversified weed management have been documented, though sev-
eral lines of evidence suggest that it is possible. Thus, genostasis might be
imposed by exploiting negative cross-resistance to herbicides (Gressel, 1991;
Prado et al., 1992). Certainly, Jordan et al. (1997) have shown that net selec-
tion for herbicide-resistant genotypes can be avoided by herbicide rotation,
in which herbicides that select for resistant genotypes alternate with other
herbicides to which the resistant genotypes are especially susceptible.
Similarly, according to Jordan and Jannink (1997), sets of non-chemical weed
control measures might be identified, such that adaptation to one measure is
genetically associated with lack of adaptation to another. However, such sets
of control measures have not been specifically identified. Accordingly, Jordan
and Jannink (1997) stressed that further research is needed into
• how IWM affects the fitness of individual weeds;
• the genetic basis of variations in traits that affect weed fitness; and
• the prediction of trait evolution using models.
Nevertheless, the reality is even more complex than this discussion sug-
gests, as weed populations rarely consist of only a single species. Rather, they

exist as multi-species assemblages, and control of one particular weed
species may merely allow another species to increase. Certainly, there is
plenty of evidence that weed populations have adapted to management
practices. Thus, Thomas et al. (1996) reported that in Canada there had been
a change in the composition of weed communities in cereals with the shift
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 83
84 STRUCTURE AND FUNCTION IN AGROECOSYSTEMS DESIGN AND MANAGEMENT
from fallowing to continuous cropping. In an earlier study, Buhler and Daniel
(1988) observed that under continuous corn production giant foxtail (Setaria
faberi) had become more difficult to control with soil-applied herbicides as
tillage was reduced, while velvetleaf (Abutilon theophrasti) had become less of
a problem. Finally, in a particularly comprehensive study carried out by
Hallgren (1996) on annual dicotyledonous weeds in unsprayed cereal crops
in Sweden over the period 1951–1990, very evident shifts in the balance of
weed species were observed.
Accordingly, Buhler (1995) concluded that, with the reduced number of
herbicide options and the reduced effectiveness of other control practices,
knowledge of how weed populations shift is essential to weed management,
if herbicide use is not to increase to levels that are environmentally and eco-
nomically unacceptable. Understanding weed population shifts will identify
vulnerable stages in weed life cycles that can be exploited in management
systems. Understanding population shifts will also identify species that are
favored, as management systems are developed that target particular weed
species. While models, such as that developed by Mortimer et al. (1996) and
discussed earlier, are able to predict quite accurately weed population shifts,
they are basically descriptive and so cannot be used to determine effective
management strategies. Thus, to the research list by Jordan and Jannink
(1997) needs to be added the topic of increased understanding of the effects
of IWM systems on the species structure of weed populations.
CONCLUSIONS

Weed management is a means to the end of maintaining crop production
within a viable agricultural system (Swanton and Murphy, 1996). There has
been a tendency to develop weed management strategies to achieve per-
ceived economic goals without linking the strategies to biological factors
(Ghersa et al., 1994) and without investigating how these different factors
interact. The significance of the recent move towards a systems approach to
weed management means that weed control is considered part of the broader
economic and ecological objectives of society. If they are to be effective and
relevant to farmers and to the public, then weed management strategies can-
not be designed in isolation (Swanton and Murphy, 1996). This reality is
reflected in the recent public concerns in Europe regarding the introduction
of genetically modified crops that are tolerant of herbicides. Instead, the
demand is for the development of sustainable systems of crop production,
where the use of herbicides is minimized, and weed control is achieved
through a mixture of cultural, mechanical, and preventative techniques.
However, for ecologically acceptable methods of weed control to be
implemented, it is clear that increased understanding of several aspects of
weed biology, and ecology, is required including:
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 84
ECOLOGICAL MANAGEMENT OF CROP-WEED INTERACTIONS 85
• weed dispersal and spread;
• the determinants of the spatial and temporal variability of weeds;
• crop-weed interactions;
• the effects of cultivation practices on weed survival and fecundity;
and
• the ability of weeds to adapt to control measures.
Mathematical models may be a very useful tool in exploring these issues,
given the complexity of the interactions involved. However, while models
have partially addressed some of these issues, to be able to explore optimal
weed management in terms of economic efficiency and ecological accept-

ability, a new direction of enquiry is required. At the risk of some simplifica-
tion, mathematical models of weed control can be said to be directed towards
the scientific question of “what” rather than the practical question of “how”
(Doyle, 1997). Thus, weed management models have primarily addressed
three questions (Mortimer, 1987; Doyle 1991): (1) what is the relationship
between the level of weed infestation and the crop losses; (2) what is the level
of any control measure required to contain the infestation or totally eradicate
the weed; and (3) what is the level of weed infestation above which control
measures are justified. However, with respect to sustainable systems of weed
management, these questions are subordinate to the more central issues of (1)
how is it possible to promote the more selective use of herbicides, while
ensuring economically acceptable levels of weed control; (2) how is it possi-
ble to minimize the environmental impacts of herbicides through the use of
biological and physical control techniques; and (3) how are the economic
risks to farmers of switching to non-chemical controls to be minimized. This
is the challenge for future weed management research.
REFERENCES
Aarts, H.F.M., 1986. A computerised model for predicting changes in a population of
Galium aparine, in Proc. EWRS Symp. on Economic Weed Control, Stuttgart, European
Weed Research Society, Wageningen, the Netherlands, 277–284.
Ampong-Nyarko, K. and de Datta, S.K., 1993. Effects of light and nitrogen and their
interaction on the dynamics of rice-weed competition.
Weed Res., 33:1–8.
Antonovics, J. and Levin, D.A., 1980. The ecological and genetic consequences of
density-dependent regulation in plants.
A. Rev. Ecol. Systemat., 11:411–452.
Auld, B.A., Menz, K.M. and Monaghan, N.M. (1979). Dynamics of weed spread:
implications for policies of public control.
Protection Ecology, 1: 141–148.
Auld, B.A. and Coote, B.G., 1980. A model of a spreading plant population.

Oikos,
34:287–292.
Auld, B.A. and Coote, B.G., 1981. Prediction of pasture invasion by
Nasella trichotoma
(Graminae) in south east Australia. Prot. Ecol., 3:271–277.
Auld, B.A., Vere, D.T., and Coote, B.G., 1982. Evaluation of control policies for grass-
land weed
Nasella trichotoma in south east Australia. Prot. Ecol., 4:331–338.
920103_CRC20_0904_CH04 1/13/01 10:46 AM Page 85