|Iowa State University|
Weed Population Dynamics
by Bob Hartzler
November 6, 2000 - This article was prepared for the proceedings of the 2000 Integrated Crop Management Conference, Nov. 29-30, Ames, IA.
The weed infestation in a field is defined by three parameters: 1) the number of species present, 2) the density of each species, and 3) the distribution of the species across the field. While the number of species in a field remains relatively constant from year to year, the latter two factors fluctuate widely in response to environment, cultural practices, and weed management tactics. It is the continual changes in weed infestations that make successful weed control such a difficult task to achieve consistently.
The weed seed bank in agricultural fields is made up of many species, but in any given year the infestation typically is dominated by a few species. An Illinois field maintained in a corn-soybean rotation was found to have 25 weed species in the seed bank, yet four species accounted for 85% of the weed population. The species that dominate the infestation are those best adapted to current management practices. As farmers adjust their management program to improve control of the species currently dominating the infestation, they typically create an opportunity for other species in the seed bank to escape control and become part of the current problem.
This cyclical nature of weed problems is illustrated by a survey conducted by the Southern Weed Science Society. Weed scientists from across the region were asked to rank the most troublesome weeds in their states (Table 1). Only one species that made the top 10 list in the 1974 survey failed to make the top 10 list in the 1995 survey. Thus, despite the major changes in tillage and herbicide use that occurred between the two surveys the same species caused the majority of problems. The rankings of individual weeds did change over time, probably due to introductions on new herbicides that provided more effective control of specific weeds. For example, introduction of ALS herbicides probably played a big role in cocklebur moving from Number 1 in 1974 to Number 4 in 1995.
Table 1. Weed Survey results - Southern Weed
(Source: Webster and Coble. 1997. Weed Technology 11:308-317).
|Most Troublesome Weeds in Corn|
|Florida beggarweed||6||9 (tie)|
This does not mean that new species can not become major weed problems. Woolly cupgrass was found in only a limited number of Iowa fields in the 1970's, but now is fairly common throughout the state. A look at the Iowa Noxious Weed Law also illustrates weed problems can change over time. A large number of species found on the Noxious Weed List are relics of the agricultural system utilized prior to the introduction of synthetic fertilizers and herbicides, where a diverse rotation was required. Examples of weeds that were considered serios problems in those systems but are rarely found in Iowa fields today include red sorrel, curly dock, buckhorn plantain and horsenettle. In contrast, the two SWSS surveys (Table 1) were both conducted during a time period where crop rotations were similar and herbicides were a principal component of weed management systems.
Herbicide resistance is another type of shift that can occur within a weed population. This shift involves a change within a weed species rather than among species. Just as the seed bank is composed of numerous weed species, a population of a single species is comprised of many biotypes. Certain biotypes may contain a genetic trait that allows them to survive a herbicide toxic to other biotypes of the same species. Repeated use of the same class of herbicide may result in the resistant biotype becoming a dominant component of the population. Numerous herbicide resistant biotypes have been selected throughout Illinois and the Corn Belt following repeated use of herbicides.
Since most weed management systems are heavily reliant on herbicides, the relative susceptibility of weeds to herbicides has a major influence on weed shifts. However, other crop production practices influence weed shifts observed in the field. Understanding how management practices influence weed shifts can be as important in developing efficient weed management programs as studying herbicide effectiveness charts.
Second to herbicide use patterns, tillage is the management practice that probably has the greatest impact on weed populations. Tillage can affect weeds directly, as in the destruction of winter annual weeds during seedbed preparation, or the effect may be more subtle, as in the shift from large-seeded broadleaf weeds to small-seeded weeds in reduced tillage systems. Research has found that fields maintained under no-till production have more diverse weed seed banks than fields managed with intensive tillage. The more diverse seed bank does not necessarily mean that weed control will be more difficult in no-till; however, this diversity increases the potential for shifts and will require greater vigilance in order to adapt management practices to prevent rapid swings in weed populations.
The date of planting and other management practices also influence weed population dynamics. The influence of these practices on weed shifts often is mitigated by differences in growth and behavior of different weed species. No other event in the life cycle of weeds affects management as greatly as weed emergence. The timing and intensity of weed emergence affect everything from the effectiveness of burndown herbicides and preplant tillage, to the timing of postplant tillage and herbicide application, to the competitiveness of weeds that escape control, to seed production by surviving plants, and eventually population shifts. Given the importance of weed emergence to all forms of weed management, it seems logical that we should give greater attention to understanding and predicting weed emergence as affected by environmental factors, weed species, and management practices.
Each weed species has a unique emergence profile. The emergence profile is defined by the initial emergence date, the length of time over which emergence occurs, and the distribution of emergence within this time period. While emergence profiles vary from year to year depending upon environmental conditions, the emergence patterns of different species remain relatively consistent in relation to each other. For example, in central Iowa the initial emergence date for giant foxtail ranged from April 29 to May 15 between 1996 and 1998. In each of these years the initial emergence of velvetleaf occurred within four days of the initial emergence of giant foxtail.
The emergence profiles (initial emergence date and emergence patterns) of 23 annual species common to Iowa are summarized in Figure 1. The size of oval provides information on both the length of emergence and distribution of emergence. Weed species with a small oval typically have most of their emergence close to their initial time of emergence, whereas species with a large oval have either an extended period of emergence or most of their emergence occurs further from their initial date of emergence. At any given time between early-April and mid-June there are several weed species at their peak emergence. Control tactics selected to be most efficacious on early-emerging weeds are unlikely to provide consistent control of later-emerging weeds.
The emergence profile of a species significantly affects the performance of weed management programs. Giant ragweed has become a greater problem for many farmers in recent years part of this increase may be due to the movement to earlier planting dates for corn and soybean than used in the past. Giant ragweed is typically the first summer annual weed of corn and soybean fields to emerge in the spring (Figure 1). Since most giant ragweed seedlings emerge shortly after its emergence begins, the majority of the population would be killed by seedbed preparation for late-April or early-May planting dates. Earlier planting dates result in the majority of the giant ragweed population emerging after planting, therefore the ragweed must be managed by the herbicide program. The large seed size of this species reduces the effectiveness of many preemergence herbicides, and the rapid growth rate of giant ragweed results in the plant reaching sizes difficult for most postemergence herbicides to control consistently. Fortunately, the relatively small seed production capacity of giant ragweed reduces the rate at which it spreads.
Waterhemp provides another example of how interactions between emergence timing and production practices influence weed population dynamics. Waterhemp is native to the corn belt but was considered a minor weed throughout the region until the mid-80s. At that time several factors fell into place that created an ideal environment for the survival of the species in corn and soybean fields. The most obvious factor favoring waterhemp was the widespread use of ALS-inhibiting herbicides and the subsequent selection of ALS-resistant biotypes. However, waterhemp remains one of the most problematic weeds for growers even after they have switched to alternative herbicide programs. Other factors that have contributed to waterhemp problems include: adoption of reduced tillage, decreased use of herbicides providing long residual control, and the decreased use of inter-row cultivation. Reductions in tillage favor small-seeded species, whereas the latter two factors favor a weed with a prolonged emergence pattern.
The prolonged emergence pattern of waterhemp allows late emerging weeds to escape control. These late-emerging weeds frequently do not impact yields because of the competitive advantage held by the earlier-emerging crop, but they can replinish the seed bank. Over time the increase in the seed bank enhances the risk of control failures that will impact yields. A simple model was created that predicts changes in waterhemp populations and soybean yield losses in a total post weed control system. A scenario in which a non-residual herbicide is applied at the V6 soybean stage and controls all weeds emerged at that point of time is represented by Figures 1A and 1B. Waterhemp that emerges after the post application produced seed that resulted in a continuous increase in the waterhemp seed bank and populations during the five year period. However, soybean yields were not affected because the late-emerging weeds did not accumulate sufficient biomass to impact soybean growth.
Figures 2A and 2B represent the same scenario except that in years 2 and 5 the effectiveness of the postemergence herbicide is reduced to 95% control from 100%, thus allowing some of the earlier emerging weeds to escape control. In year 2 soybean yields were not affected by the reduced control level due to the relatively small seed bank, but the waterhemp seed bank at the end of year 2 ( 583 seeds / m2 ) was increased significantly compared to where 100% control was obtained ( 298 seeds/ m2 ). In year 5 the seed bank had increased to a point where a reduction in herbicide performance resulted in a 14% yield loss.
The prolific seed production capability and ability of late-emerging waterhemp to contribute to the seed bank are major factors contributing to the weediness of this species. Waterhemp plants emerging after the V6 stage of soybean are able to produce 2000 or more seeds per plant. Most farmers would not know these plants are present since they fail to reach a size that extends above the soybean canopy. The late-emerging waterhemp that develop following postemergence applications are capable of increasing the seed bank even when high levels of control are provided by the herbicide program. This maintenance of the seed bank creates a train wreck waiting to happen whenever the herbicide program fails to provide complete control.
Weed shifts occur because no control tactic is equally effective against all weed species (or biotypes). The species that is controlled ineffectively will increase in density following repeated use. How does one effectively manage weed shifts? The key is to use a diverse management system - for most farmers today, that means using a variety of herbicides. Rotating herbicides is effective at reducing the rate that weed shifts arise, assuming that the rotational herbicide is highly effective against the weed species or biotype that is controlled ineffectively by the primary herbicide. The worst case scenario is illustrated in Figure 3A in which continuous selection pressure is placed on a weed population by a single herbicide. During the first few years the rate of growth is relatively slow but by year six the weed enters the exponential rate of growth and populations explode. Most farmers probably would not notice the gradual increase in the weed during the first 4-5 years of the selection pressure.
The impact of rotating herbicides is illustrated in Figure 3B. Under this scenario, the herbicide responsible for the shift is used only during the odd years, whereas an alternative chemical is used during the even years. Thus, the population increases during years 1, 3 and 5, but during the even years the alternative herbicide provides a high level of control and therefore reduces the population. In this scenario, the persistence of the seedbank allows the population to continue to increase over time and eventually the population would reach the level where it enters exponential growth and populations would get out of control. This type of scenario occurred with ALS resistant waterhemp - annual rotations of ALS herbicides were ineffective at preventing resistance, whereas in other situations (different herbicides or different species) annual rotations of herbicides have prevented widespread problems.
Weed infestations are dynamic, therefore requiring dynamic management programs. Although changes in weed infestations sometimes are caused by the introduction of new species, the majority of changes are due to weeds that already were present in the field but were maintained at non-economic levels by previous management tactics. As weed management systems become increasingly reliant on herbicides, it is likely that weed shifts will occur more rapidly than previously encountered. Thus, monitoring fields and adjusting management programs quickly before the increasing weed population reaches troublesome levels will be more important than in the past.
Prepared by Bob Hartzler, extension weed management specialist, Department of Agronomy, Iowa State University
more information contact:
ISU Extension Agronomy
2104 Agronomy Hall
Ames, Iowa 50011-1010
Voice: (515) 294-1923
Fax: (515) 294-9985
site designed and managed by Brent Pringnitz.
Submit questions or comments here.
Copyright © 1996-2003, Iowa State University, all rights reserved