NE1047: Ecological Bases for Weed Management in Sustainable Cropping Systems
(Multistate Research Project)
NE1047: Ecological Bases for Weed Management in Sustainable Cropping Systems
Duration: 10/01/2011 to 09/30/2016
Statement of Issues and Justification
The need as indicated by stakeholders
Recent surveys, farmer focus groups, discussions at grower meeting, and individual interactions with farmers indicate that weeds are the most critical management problem facing organic or sustainable farmers who limit their use of herbicides. Weeds reduce crop yields and quality, increase harvest difficulty, and add to the seed bank contributing to future management problems. Our previous research developed cover crop systems, cultivation methods, and natural product herbicides to manage weeds early during the growing season. Sustainable weed management, however, is based on the integration of multiple tactics that attack weed populations throughout the weed life cycle. This requires an understanding of how ecological processes interact with control measures at various life stages. First, cultivation is a major method of weed management in organic and reduced herbicide systems, but little is known about how soil texture, tilth, and moisture affect the efficacy and selectivity of cultivation implements. Substantial gains in weed control are possible through better matching of the type and timing of implement use to soil conditions. Second, some weeds are missed by cultivation, or are resistant to herbicides or emerge through gaps in the crop canopy. Controlling these weeds by rogueing or post-harvest operations can greatly affect seed input to the weed seed bank. Weed seeds sometimes continue to develop on plants that have been cut, pulled or treated with herbicide, but information is sorely lacking on how the method of termination and the developmental stage at death affect the number of viable seeds that various weed species ultimately produce. We regularly get questions from growers about the potential for seed production on weeds that have been killed but not removed from the field. Third, once weed seeds have entered the seed bank, they may die, produce emerged seedlings, or persist in the soil and germinate in subsequent years. Information is available on how tillage and seed position in the soil affect the relative probabilities of these three outcomes. Much less is known, however, about how organic matter sources like cover crops and compost affect seed survival and seedling emergence. Some types of organic matter inputs may reduce seed persistence by promoting germination in circumstances where emergence is unlikely or by increasing microbial populations that cause seed decay. Reducing the persistence of weed seeds would reduce weed infestations in subsequent crops. Organic farmers in particular are interested in multi-tactic weed management strategies that involve attack on several stages of the weed life cycle. With the growing prevalence of herbicide resistant weeds, conventional growers too are realizing the usefulness of integrating a variety of practices into their weed management programs. The applied research proposed here will provide a knowledge base for improving several important management tactics.
The importance of the work, and what the consequences are if it is not done
Multiple weed management strategies directed at seeds and weed seedlings improve weed control in organic and sustainable systems as compared to a single tactic, thereby improving food production, quality and net return to farmers.
Objective 1. Determine how soil conditions affect efficacy and selectivity of cultivation implements for the control of various weed species.
Physical weed control implements kill only a proportion of weed seedlings. Improving the mean efficacy and reducing within field variability in efficacy should be possible by choosing implements and making adjustments that are most appropriate for soil conditions in a field. This requires an understanding of how site conditions affect the efficacy of particular implements against various weed species. Highly experienced operators are often able to achieve reasonable performance of their cultivators. But in the absence of the information provided by the proposed work, they may have no way of knowing that a delay in cultivation to reduce soil moisture, a change in setting based on moisture conditions, or a different cultivator may improve weed control. Moreover, the recent increase in the number of organic growers and the growing need for cultivation to control herbicide resistant weeds means that many growers are relatively new to cultivation. They could be helped by the availability of information on how to choose implements and settings and time cultivation based on their soil conditions.
Objective 2. Determine the reproductive growth stage at which summer annual weeds can be terminated and still produce viable seeds and quantify the effect of method of life-termination on seed production.
At present, the point in plant development at which seeds become physiologically independent of the parent plant is known for only a few weed species. Consequently, growers cannot assess whether the weeds are about to produce viable seeds, and hence cannot assess the urgency of control measures. Moreover, knowledge about how the method of killing a weed influences seed production is meager. Can a plant that is merely in flower produce seeds if it is cut or uprooted or sprayed with a slow acting herbicide like glyphosate? Understanding how production of viable seeds relates to the point in the life-cycle at which a weed is killed and the method by which it is killed will improve preventive weed management. The findings of this effort will directly benefit agricultural producers who rely on hand labor, mowing, or glyphosate applications for late-season removal of summer annual weeds. If weeds continue to produce viable weed seeds after being mowed, pulled, or following a herbicide application, growers will need to know how early in the life history escaped weeds must be removed to prevent viable weed seed from forming and entering the soil seed bank. If this work is not done, growers will continue to allow weed seed set, even when they are trying to avoid this.
Objective 3. Determine the extent to which soil amendments such as green manures and compost affect seed mortality of various weed species.
At present, the principal method growers have for decreasing the weed seed bank is tilled fallow in which the soil is disturbed and left bare. Weeds are stimulated to germinate by the soil disturbance and then killed by further tillage or by herbicides. Improved understanding of how organic matter amendments, including green manures and compost, affect seed survival following initial seed rain will allow growers to implement cropping systems that reduce weed seed banks and thus weed pressure in subsequent crops. Without this work, growers will either have to suffer the consequences of high density seed banks or risk the erosion potential of bare fallows. The technical feasibility of the research
Objective 1. Pilot studies at Maine and New York have examined effects of various measures of soil moisture and soil tilth on the performance of cultivators. The development of hand held soil moisture meters allows rapid objective measurements of soil moisture that will be correlated with visual and tactile information that could be obtained by a farmer. Soil texture can be read from county soil surveys and, for research, measured more precisely by the hydrometer method. Members of the research group are experimenting with photographic and dry screening methods of measuring soil roughness. Cultivation efficacy will be assessed by weed counts before and after cultivation.
Objective 2. A preliminary study in Michigan examined the reproductive growth stage at which three summer annual weeds produced viable seeds. Termination methods that farmers use include pulling, cutting, chopping, and applying herbicides. Grass and broadleaf weeds will be terminated by these methods during flowering and seed development and the plants confined in fine mesh to retain any seeds that subsequently mature. In late autumn, seeds will be removed from the plants, counted and tested for viability to determine how mode of death and phenology at termination affect seed production. Some seeds will be returned to the soil and tested again in the spring to determine whether these factors affect the short term persistence of seeds in the soil.
Objective 3. Several of the investigators have previous experience with buried seed experiments. Seeds will be buried in soil in mesh packets in the fall and exhumed annually. At each exhumation, the soil will be stirred and returned to the packet, or an organic matter amendment (e.g., chopped clover, compost) mixed into the soil and then returned to the packet. Packets will be buried in plots receiving the same type and rate of the soil amendment. A subsample of the packets will be assayed for surviving seeds at each exhumation and the seed mortality trajectories of various treatments compared.
The advantages for doing the work as a multistate effort
Objectives 1-3 are all better accomplished through a Multi-State project as opposed to an intensive effort at one or two sites.
For Objective 1, a Multi-State initiative will allow assessment of many more types of machines in a much broader range of soil conditions than could be obtained by an investigator in one state. The utility of the study is directly proportional to the number of observations, and since most cultivation is concentrated in early summer, the number of observations that can be made by any one research team is limited. Cross comparison between states will also help in assessing how the optimal equipment and settings may shift with climate change.
For Objective 2, if each participating lab investigates three weed species at three phenological stages of flower development and includes three termination methods there will be over 150 samples to assess. A Multi-State effort, will allow (i) the assessment of many more species, which will reveal taxonomic patterns in the response of seed production to time and mode of death, (ii) replication in different environments to assess the robustness of results, (iii) processing of seed samples in a timely manner, and (iv) allow a comparison across latitudes to provide an indication of how climatic shifts may affect seed production on terminated plants of various species.
For Objective 3, a Multi-State project will allow assessment of the effects of more types of amendments on more species of weeds than could be accomplished in any one state. We believe that legume green manures may be particularly useful for promoting seed mortality, but the types of legume cover crops used across the U.S. varies widely. A Multi-State initiative will allow us to determine whether effects are general to broad classes of organic amendments or highly specific to particular inputs and whether they relate to measurable characteristics such as carbon/nitrogen ratio. What the likely impacts will be from successfully completing the work This research will be coupled with extension efforts in each state that will disseminate results to growers through field days, presentations, and publications. Results will be widely disseminated through synthesis articles in the e-Organic/eXtension web site, New Ag Network, Organic Broadcaster, and MOSES. We expect this work to reduce weed problems in organic and low herbicide cropping systems with consequent benefits to growers.
Related, Current and Previous Work
Ecology provides the theoretical basis for weed science, much as physics provides a theoretical basis for engineering and biology the theoretical basis for medicine (Liebman et al. 2001). Although much research has focused on the ecological relationships of weeds in agroecosystems in recent years, substantial gaps in knowledge relevant to weed management still exist. Weed management strategies that integrate diverse tactics which attack weeds at multiple points in their life cycles (Liebman and Gallandt 1997, Bond and Grundy 2001) are increasingly accepted as the key to sustainable weed management. The research proposed here will fill critical knowledge gaps pertaining to three stages in the life cycle of annual weeds, and each of the three areas of research has direct application in weed management.
Objective 1. Determine how soil conditions affect efficacy and selectivity of cultivation implements for the control of various weed species.
Very few studies have addressed the effects of soil conditions on the efficacy of cultivation implements. Lovely et al. (1958) found that three passes with a rotary hoe provided 72% weed control but that when the soil was wetted either before or immediately after rotary hoeing weed control dropped to 33%. Mohler et al. (1997) noted that a coarse seed bed decreases the effectiveness of in-row weeding tools like rotary hoes and tine weeders since (i) weeds rooted in clods may roll around but not be damaged, (ii) rolling clods about uncovers weed seedlings that would not have been able to emerge, and (iii) light penetrates several cm into the cloddy soil so that seedlings establish too deeply to be uprooted by these implements. Cussans et al (1996) showed that seedlings of some species emerge from greater depths in cloddy soil. A coarse soil surface during inter-row cultivation may also prevent effective burial of weed seedling when soil is thrown into the crop row. First, if the soil is cloddy or slabbing off of the cultivator sweeps, the operator may have to decrease speed or depth settings to throw less soil so as to preserve the crop. Second, gaps between clods may allow light to reach weed seedlings, thereby allowing survival. Mohler (unpublished data) found that soil thrown into the crop row in organically managed plots had a finer aggregate size distribution than soil in conventionally managed plots, but post-cultivation weed density did not differ in that study.
Gallandt's research group recently completed on-farm evaluations of a wide range of hand and wheeled (pushed) tools, including documentation of efficacy and working rates. While mean efficacy values were generally similar among a large group of tools, and lower than hand pulling, there were large differences in the variability in efficacy (Gallandt and Costanzi, unpublished). Working rates of the tools varied considerably, but predictably, from hand pulling < short-handled tools < long-handled tools < wheeled-tools. Measurements of volumetric soil moisture in subplots surveyed to measure efficacy indicated that each 1% increase in soil moisture resulted in an 8% decrease in cultivation efficacy. Understanding how this, and other site characteristics (e.g., soil texture, surface roughness, crop residues) affect the variability in cultivation efficacy is the central aim of this objective.
A search of the CRIS data base using the keywords: cultivation or physical weed control X efficacy or weed control revealed 273 records, indicating growing interest in cultivation for weed management. Few studies, however, are related to the work proposed here. Cultivation is being evaluated in high-residue systems (0418352; Mirsky and Moyer, "Reduced-tillage weed management for organic farming"), and in reduced-tillage vegetable systems (0215417; Rangarajan et al., "Optimizing reduced tillage systems for conventional and organic vegetables grown in the upper Northeast"), but without an explicit goal of documenting and understanding causes of variation in efficacy. Fennimore (0208114; "Evaluation of low-rate herbicides and precision cultivation for weed management in vegetable crops") evaluated use of a precision-guidance system in California vegetable production. They concluded that the RoboCrop offered good weed control and perhaps reduced hand weeding labor. Efficacy values, however, were not reported, and a more mechanistic understanding of this cultivation system appears to have been beyond the scope of the study.
The proposed research is closely aligned with a new Hatch Project of one of the Co-PIs (0222914; Gallandt, "Integrating physical and cultural weed management strategies in organic farming systems"). A component of this research aims to evaluate efficacy, working rates, and sources of variation in efficacy of a wide range of cultivation tools, with a particular emphasis on hand tools used in small-scale organic systems.
Objective 2. Determine the reproductive growth stage at which summer annual weeds can be terminated and still produce viable seeds and quantify the effect of method of life-termination on seed production.
Seed production of escaped weeds is a major concern in both conventional and organic vegetable and row crop farming systems, and the continued evolution of weed resistance to herbicides in herbicide tolerant crops has exacerbated these concerns. Each weed that is not controlled produces hundreds to thousands of seeds that will emerge over several growing seasons. When late-season weed control measures like hand hoeing, mowing, or herbicide applications are used it is unclear what happens to seeds that were forming on plants at the time of control. Will pollinated flowers develop viable seeds during the process of desiccation? Will green weed seeds after-ripen and be viable if the weeds are left lying in the field? If weeds are found to produce seed post-termination, tactics for weed removal will need to be considered as well the economics of late-season weed control. Research into seed maturation on terminated weeds is scarce. In 1918, Longyear found that dandelion seeds become viable seven days after the flowers opened. Gill (1938) found that weeds such as annual sowthistle and common groundsel were capable of producing viable seed when terminated at the flowering stage. Longspine sandbur had 20% seed viability at heading and 40% at anthesis (Anderson 1997), while other weeds, such as common chickweed, curly dock, and jimsonweed produced viable seed when killed in the milk/fruit stage (Gill 1938). A study somewhat similar to the proposed work was conducted in Australia on variegated thistle. Dodd (1989) examined the potential for seed production at various capitulum (seed head) maturities. Two termination treatments were examined, very similar to our proposed whole plant and cut/chop termination treatments. Mature seed was produced at all capitulum maturities except the earliest stage which was pre-flowering, and seed production increased with capitulum maturity. At each maturity level, the whole plant termination method resulted in significantly higher seed production than when reproductive structures were removed from the plant. Although these studies provide information for a few weeds, they have not compared weed termination methods, nor have the most prevalent weeds species in the Northeast and North Central regions been studied. In Michigan, the phenology of three summer annual weeds, including common lambsquarters, velvetleaf, and giant foxtail, was recorded the past three years. The time of initial flowering, immature seed formation, and seed dispersal for two cohorts is presented in Figure 1 (see attachment). The early cohort of each weed species was made up of the first flush of plants emerging in the spring. The late cohort consisted of weeds germinating 1500 growing degree days (GDD, base 32 F) after initial emergence of the species began in the spring. The late cohort was designed to look at reproduction of weeds that escaped management. In organic fields where hand hoeing is used for late-season weed control beginning in early August, plants being terminated range in phenology from pre-flowering (e.g. late emerging common lambsquarters) to mature seed formation (e.g. early emerging common lambsquarters and giant foxtail). Information on the potential seed production of these weeds at these stages is currently unavailable and will be important in evaluating the economics of late-season weed control and designing tactics to ensure weed seeds are not added to the weed seed bank.
A search of the CRIS data base showed a few somewhat related projects. Ball (0194956: "Integrated Weed Management Systems for Eastern Oregon Dryland Crops" studied climatic factors affecting phenology of maturation in jointed goatgrass, and feral rye, important weeds of dryland wheat cropping systems. Knapp et al. (0213350) "Seed Biology in Agroecology" are going to harvest seeds from common vetch (Vicia sativa) at five developmental stages to determine how phenology affects seed dormancy and other seed properties. The predecessor Multi-State project to the one proposed here ("Weed Management Strategies for Sustainable Cropping Systems") had as one objective "Determine the response of various weed species in several developmental stages to burial, uprooting and breakage to optimize timing and type of cultivation for weed control" (e.g., Haar: 0208850), Bellinder: 0208621). The focus of this work, however, was on effects of simulated cultivator damage on young weeds, well before flowering.
Objective 3. Determine the extent to which soil amendments such as green manures affect seed mortality of various weed species.
Managing seed banks is a critical but difficult component of integrated weed management. One potential way of decreasing seed survival is through incorporation of green manures and other organic amendments. Several studies have observed seed bank density in cropping systems with contrasting organic matter inputs and some have found relations between weed seed density and microbial activity. Fennimore and Jackson (2003) showed both above-ground weed density and weed seed banks were lower in soil amended with cover crops and compost. De Cauwer et al. (2011) showed seed bank density after 4 years of low C:N ratio compost was lower than with manure slurry, though neither differed from the control. Seed density of species with hard seed coats like Chenopodium spp. was lower in plots with high microbial activity in their study. A problem with these studies is that weed seed rain was not controlled across treatments so that one cannot be certain that differences among treatments were due to affects on persistence of the seed bank. Controlled studies on seed survival in contrasting organic matter input treatments using seed packet and artificial seed banks have produced variable results. Ullrich et al. (2011) buried seeds of smooth pigweed and common lambsquarters in packets in cropping systems treatments with high and low organic matter inputs and followed survival over two years. Overall, they found relatively small differences in survival among treatments, though some differences were significant. Seed survival of smooth pigweed was lower in the high organic matter input systems in two of four experiments but survival of common lambsquarters was inconsistent and sometimes better with high organic matter. Comparison of seed survival in packets buried in soils across the Midwest showed that common lambsquarters survival was positively correlated with soil organic matter (Davis et al. 2005), and incubation of velvetleaf and giant foxtail in soils from different long term cropping systems in Michigan indicated that seeds of these species tended to survive better in soil from systems with high organic matter inputs (Davis et al. 2006). Using artificial seed banks in open-bottomed containers in the field, Menalled et al. (2005) showed that high rates of composted swine manure had no affect on seed survival of any of the weeds tested. Mohler et al. (2012) showed that weed emergence was reduced after incorporation of cover crops and that this was associated with attack on seeds by micro-organisms. Organic amendments incorporated into the soil might be expected to reduce survival of weed seeds in several ways. First, incorporation of fresh green organic matter into the soil stimulates a short lived population explosion of pathogenic fungi and oomycetes (Pythium, Rhizoctonia, Fusarium and others), which can be expected to attack seeds along with the newly incorporated residue (Chung et al. 1988, Manici et al. 2004, Mohler et al. 2012). Second, organic amendments will support macroinvertebrate populations, particularly earthworms, and these are known to consume weed seeds (McRill and Sagar 1973, Grant 1983) or move them too deep in the soil for successful emergence (Regnier et al. 2008). Finally, if the amendment has a low C:N ratio, as is the case for most legume green manures, then N release during decomposition will produce nitrate in the soil solution, and nitrate is a known germination cue for many weed species (e.g., lambsquarters Hanson 1970, common chickweed Roberts and Lockett 1975, Powell amaranth Brainard et al. 2006). Germination deep in the plow layer usually results in death since few weed species are able to emerge from more than 1 to 2 inches deep. These mechanistic considerations indicate possible reasons why results from cropping systems experiments appear to be contradictory. All of the mechanisms just listed are relatively short lived. Populations of pathogens promoted by fresh green crop residues quickly die off as the residue decomposes and are frequently replaced by microorgnanisms antagonistic to pathogens (Chung et al. 1988, Manici et al. 2004). Similarly, large earthworms that consume weed seeds are favored by partially decomposed organic debris and cannot subsist on highly decomposed humic materials. Finally, only relatively low C:N organic inputs can be expected to stimulate seed germination by release of nitrates through decomposition in the short term, and soils may release nitrate following disturbance even if no organic matter is incorporated. Consequently, comparison of systems with long term inputs of diverse organic matter sources versus systems without additional inputs may have little power to reveal whether a specific organic amendment affects survival of a particular weed species. The considerations above indicate that different types of organic amendments should have differing effects on weed seed persistence and that various species are likely to respond differently to a given amendment, depending on seed coat properties (Davis et al. 2008), germination response to nitrate, and other factors. The individualistic effect of amendments on different weed species is supported by Davis (2007). He incubated eight species of annual weeds in field soil with different rates of N fertilizer either with or without ground corn stover. Five of the eight species showed no significant response to the treatments, velvetleaf seed mortality was reduced by corn stover at all N rates, giant ragweed seed mortality was reduced by stover only at the high N rate, and wooly cupgrass seed mortality was increased by stover only at the high N rate. A search of the CRIS data base using the key words 'cover crop' or 'compost' X 'seed bank' revealed many additional cropping systems studies in which seed banks are being monitored in systems with contrasting organic matter amendments: 0196015, Weller et al., "Seed Bank Dynamics in Alternative Vegetable Cropping Systems"; 0204932, Gallandt and Hutton, "Soil Improving Practices for Ecological Weed Management" (see Mirsky et al. 2010); 0210970,Haar et al., "Weed Seedbank Dynamics for Weed Management"; 0408813, Teasdale and Mortensen, "Ecological Mechanisms Underlying Biologically Based Weed Management"; and several others with a similar approach. As with some of the published studies cited above, all of these have the problem that seed input is not controlled, and therefore the role of the organic amendment on persistence of seeds cannot be deduced. A search using the key words 'seed bank' or 'weed seed' X 'microbial' produced only one slightly related project. Chef et al. (0409450) "Biologically and Ecologically Based Knowledge for Integrated Weed Management Systems" studied microbial attack on seeds and cover cropping, but the connection between the two is unclear from the report. Wolf et al. (0208745),"Winter Soil Processes and New Tools for Nitrogen, Weed and Nematode Management" found that a living rye cover crop sometimes affected over-winter survival of weed seeds. A search using the key words 'seed survival' X 'weed' found two relevant studies, but they have been published and are discussed above (Menalled et al. 2005, Regnier et al. 2008).
Determine how soil conditions affect efficacy and selectivity of cultivation implements for the control of various weed species. Specifically, we hypothesize that (i) efficacy of cultivation decreases as soil moisture increases and (ii) efficacy decreases as soil roughness (cloddiness) increases. In addition, this objective seeks to obtain information on how different implements are affected by soil factors.
Determine the reproductive growth stage at which summer annual weeds can be terminated and still produce viable seeds and quantify the effect of method of life-termination on seed production. The goal of this objective is to provide preliminary insight into an area of weed seed management that has been previously under studied. This research will enable us to develop recommendations for farmers on managing late-season weed escapes.
Determine the extent to which soil amendments such as green manures affect seed mortality of various weed species. In particular, we hypothesize (i) that seed persistence will be lower in soil with incorporated green manure, (ii) that species which are cued to germinate by the presence of nitrate will have lower persistence when legume cover crops are incorporated than when no cover crop or a small grain cover crop is incorporated, and (iii) that species with shorter seed half-lives in the soil will be more affected by incorporated green manure than species with relatively longer lived seeds.
MethodsObjective 1 The states that will participate in this Objective include Maine (Gallandt), New York (Mohler, DiTommaso), Pennsylvania (Curran), Michigan (Brainard) and Deleware (VanGessel). Objective 1 will be pursued through a variety of different cultivation experiments at several sites that target various aspects of the problem. The studies will be linked by collection of data through a standard format that will allow meta-analysis of results across multiple experiments, cultivation events, sites and years. A standard data sheet and computer spreadsheet template will be developed for standardized collection and entry of data. Data collected for each cultivation operation in each experiment will include the following. (1) A description of the machine sufficient to duplicate the implement: make and model, size and pitch of sweeps, type of shank, diameter of disk-hillers, and other relevant factors. (2) Information on machine settings: distance of front sweeps from the crop row, operating depth, operating speed, and analogous relevant records for tine weeders, rotary hoes and hand tools. (3) Information on the weeds: size of the three most abundant species, counts in quadrats before and after cultivation of the three most abundant species, and the total of all species. The quadrats will be the width of the row spacing. Sufficient quadrats will be measured to obtain a good estimate of the abundance of the major weed species. (4) The size and growth stage of the crop at the time of cultivation will be recorded, and the stand density measured before and after cultivation. (5) Measurements of edaphic factors will include soil texture (taken from soil survey or by hydrometer measurement), soil moisture measured with a hand held electronic probe, soil roughness (cloddiness), residue cover and, if significant residue is present, residue biomass. In addition, rainfall between cultivation and post-cultivation weed counts will be recorded. The cultivation machinery available differs among states, and purchase of new machinery that would enable standardization of experiments across states is beyond the means of a Multi-State project. Moreover, the crops grown differ with climate and whether the academic appointment of the participating scientist is primarily in the area of field crops or horticulture, and the types of cultivation equipment used necessarily varies with crop and cropping system. In addition, the most problematic weed species vary between states. Ensuring equivalent weed floras at the various locations would be difficult and could potentially crate lasting problems. Finally, cultivation is something of an art, and efficacy depends on the user's familiarity with the machine. Having each lab use their standard equipment will help ensure performance that is similar to that achieved by growers using the same machinery. Consequently, our approach with this Objective is not to duplicate experiments at multiple sites, but to focus on standardized data collection that will allow comparison of data across experiments. Within a state we can test the utility of particular cultivation regimens and compare effects of soil conditions on performance of different implements. By standardizing data collection, however, we can achieve a Mult-State goal of determining whether general trends relate performance to soil conditions. For example, we hypothesize that sweep cultivators will have decreasing efficacy with increasing soil moisture, that tine weeders will have decreasing efficacy with increasing soil roughness, and that these trends will hold regardless of specific implement or weed species. Maine (Gallandt) will test a range of different hand tools and cultivators in several large, edaphically heterogeneous fields to determine how edaphic factors affect both mean performance and variation in performance across variable conditions. Regression procedures will be used to reveal how soil moisture, surface roughness and soil texture affect implement performance within a field. The hypothesis is that some implements may have greater variability in efficacy than others, even though mean efficacy may differ little. Such a lack of consistency would be considered undesirable by many growers. New York (Mohler and DiTommaso) will create treatments with variation in cloddiness by plowing in different moisture conditions, and then doing secondary tillage shortly before planting. Treatments will be arranged in a replicated block design and planted with soybeans. Plots will be tine weeded twice and cultivated with a row crop cultivator twice, with crop rows hilled up at the second cultivation. New York and Maine are currently developing a rapid method for assessment of soil roughness using a fine beaded chain: when the chain is laid onto the soil such that it follows the soil micro-topography, the difference between the observed length of the chain and its actual length is a measure of soil roughness. This measure will be calibrated by regression against parameters of the size distribution of aggregates obtained by dry screening of samples of the surface soil. Pennsylvania (Curran) and Deleware (VanGessel) will explore the use of high-residue cultivators as a means for supplementing the weed suppression achieved by cover crop residues flattened with a roller-crimper. Rye and legume cover crop (hairy vetch or winter peas) will be used as cover crops and the experiment planted with soybeans in Pennsylvania and lima beans or sweet corn in Delaware. Since the residue is expected to suppress weed emergence for the first month after planting, cultivation will begin relatively late (five to six weeks after planting) in these experiments. Delaying cultivation will also help crops survive clumps of soil moved by the cultivator. Treatments will consist of cover crops and number of cultivations (once versus twice) arranged in a randomized block design. Since the cover crop affects soil moisture by depletion before roller-crimping and by shielding the soil surface from evaporation by sunlight after flattening, the effect of variation in soil moisture on cultivation efficacy between locations, years and cultivation events is expected to be substantial. In Michigan (Brainard), the efficacy of a tine weeder and an S-tine row-crop cultivator on weeds and a snap bean crop will be evaluated under various soil conditions. Cultivator efficacy will be examined within each of 8 treatments varying in their previous cover crops (none, oats, buckwheat, and yellow mustard grown the previous summer), and the timing of primary tillage (fall versus spring chisel plowing). We hypothesize that both cover crop and tillage timing will influence efficacy of cultivation due to differences in surface roughness (cloddiness) at the time of cultivation. Effects of soil moisture will be explored by watering micro-plots at various times before cultivation. The experiment will be arranged in a split-split-plot design with tillage timing as the main plot factor, cover crop as the sub-plot factor, and watering as the sub-sub-plot factor. Resident weed species (primarily common lambsquarters, Powell amaranth, and giant foxtail) will be supplemented with a surrogate-weed (Idagold yellow mustard) sown immediately following secondary tillage in the spring. Flex tine weeding will occur approximately 10 days after planting beans, with s-tine cultivation occurring at approximately 20 and 30 days after planting. At each participating location, additional experiments will be devised and run based on the results from the experiments described above. In addition to analyzing the experiments at various locations independently, data from various studies will be subjected to a meta-analysis. Multiple regression and multivariate analysis procedures will be used to reveal overall patterns of how percentage weed control is affected by weed species, weed size and, most especially, the several edaphic factors studied. Objective 2 The states that will participate in this objective include Michigan (Renner), New York (Mohler, Bellinder, DiTommaso, Taylor), Illinois (Masiunas), Pennsylvania (Curran), and Delaware (VanGessel). Maine (Gallandt) will participate if competitive funds can be obtained to support the work. For the purposes of this study we have chosen to group common weeds based on floral structure. Group A includes broadleaves that form seeds on individual floral stalks. Group B consists of broadleaf weeds that form seeds within capsules or berries. Members of the Asteraceae family form Group C as the develop seeds on compound flower heads. Finally, Group D consists of grassy weeds which have seed heads. Weeds common in the Northeast and North Central regions that fall into these groups are as follows: Group A " Pigweed species (Amaranthus sp.) " Common lambsquarters (Chenopodium album) " Common ragweed (Ambrosia artemisiifolia) Group B " Velvetleaf (Abutilon theophrastii) " Jimsonweed (Datura stramonium) " Nightshade species (Solanum sp.) Group C " Horseweed (Conyza canadensis) " Canada thistle (Cirsium arvense) " Perennial sowthistle (Sonchus arvensis) Group D " Foxtail species (Setaria sp.) " Barnyardgrass (Echinochloa crus-galli) " Crabgrass species (Digitaria sp.) Laboratories participating in Objective 2 will select one weed species from Groups A, B, and D based on naturally occurring populations at research and on-farm sites; Group C weeds will be optional as a fourth weed choice. Plants from each species will be terminated at the early floral stage for Groups A, B, and C or at the first appearance of the green seed head for Group D. Additional plants will be terminated when immature seed is present and again at 50% seed maturity. Multiple stages of development may be present at the same time on a plant. Collection times will be based on the most advanced stage on a given plant. Flowering weeds will be collected when only open flowers or floral buds are present. Weeds with immature seed will be determined visually or by dissection of reproductive structures from multiple locations on the same plant. The same examination of seed will be used to determine when seed is mature for the final collection. Extra plants not destined for analysis of seed production will be grown so that reproductive structures can be sacrificed to determine developmental stage without affecting target plants. Notes will be kept so that developmental stage of each species can be related to thermal time (growing degree days) obtained from nearby weather station data. Three methods of termination will allow us to determine how method and timing of termination influence viable seed production. The termination methods are: " Pulling the plant (simulating hand pulling or hoeing) " Clipping/chopping (simulating mowing) " Applying the herbicide glyphosate Six replicates will be used for each species, termination method, and termination date. Each plant or collection of reproductive structures from a plant will be stored separately in a nylon-mesh residue bags or baskets. The bags/baskets will allow for some soil contact and exposure to the elements. Baskets will be secured to the soil surface between rows of soybeans or a vegetable crop. Plants sprayed with glyphosate will be allowed to remain standing in their original field and will be covered with a residue bag to avoid seed loss and predation, while still allowing plants/structures to be exposed to fluctuations in temperature and precipitation. The bottoms of these vertical bags will be sealed by folding the open end over several times and stapling through the folds; the juncture of the cloth with the stem will be held tight to the stem with electrician's waterproof rubber tape to prevent loss of seeds. Temperature will be recorded throughout storage using data loggers and precipitation will be recorded via the local automated weather stations. In early November of each year, the baskets and residue bags will be retrieved from the field, and seed production determined for each sample. If seeds are found they will be divided into three groups. Group 1 will be immediately tested for viability by subjecting seeds to a dormancy breaking treatment appropriate to the species (e.g., piercing the seed coat of velvetleaf, thiourea treatment of common lambsquarters, Buhler and Hoffman 1999) and then germinating seeds in deionized water in petri dishes lined with filter paper. Seeds that do not germinate promptly will be sectioned and stained with a 0.5% solution of tetrazolium chloride (Peters 2000). After 24h all viable embryos will turn pink in color. Viability will be scored as the percentage germinating plus the percentage turning pink in tetrazolium chloride. Groups 2 and 3 will be placed in new, smaller nylon bags and returned to the field surface. In the spring, groups 2 and 3 will be retrieved from the field. Group 2 will be tested for viability as described above and group 3 will be tested for germination by placing seeds in Petri dishes on filter paper soaked in deionized water. Dishes will be sealed using laboratory film, placed in the dark at 25 C, and examined weekly for three weeks to count the number of germinated seeds. For each weed species, viability and germination will be analyzed using ANOVA in SAS to determine if termination methods and timings differ. Results will be analyzed over initiation years and sites using generalized linear mixed models (Stroup 2011). Objective 3 The states that will participate in this Objective include New York-Cornell (Mohler, Bellinder, DiTommaso, Hahn), New York-Geneva (Bjorkman, Taylor), Michigan (Brainard, Renner), Indiana (Gibson), Illinois (Masiunas), New Hampshire (Smith), and Florida (Chase). Study A: medium term survival of seeds in amended soil. At each participating location, the experiment will be done in small plots in a replicated block design with three or more treatments differing in amendment inputs and five replications. Plot size will vary some with the equipment available for mowing and incorporating the cover crops but it will be sufficiently large for tractor operations and minimizing edge effects. Typecal plot size will in the range 6' by 15' to 8' by 20'. The type of amendments investigated and rate of application will be consistent from year to year at a location, but vary between participating laboratories. Amendment rates will be at the high end of rates commonly used by farmers in the local area. All laboratories will include a control treatment that receives no amendment, at least one legume cover crop, and at least one small grain cover crop. Cover crops will be planted at the appropriate time for each location. For cover crops, the target rate of top growth will be set, and adjusted based on sampling prior to incorporation in the spring by removing or (more usually) adding material from adjacent areas not in the plot proper. Compost applications will be made on a consistent dry weight basis from year to year. At each participating location seeds of two or more weed species will be buried in fine mesh packets in each plot late in the fall of year 0 while cover crops are growing, but before any cover crop residue or other amendment has been incorporated. Seeds will be collected from local populations when ripe and air dried in an unheated shed at ambient air temperatures, cleaned and then again held at ambient air temperatures until burial. Each packet will contain several hundred seeds mixed with 100 g of fine sand and initially no amendment. Since we expect that species with relatively short lived seeds (e.g., hairy galinsoga) will require more seeds per packet than species with persistent seeds (e.g., velvetleaf, common lambsquarters), the number of seeds per packet will need to vary with the weed species investigated. Packets will be buried at 15 cm which is too deep to allow emergence of any of the species that will be investigated. The intent is to study effects of amendments on seed persistence, not on emergence. Any germination that occurs below the depth of possible emergence is fatal, and removes seeds from the soil seed bank. Enough packets of each species will be buried to allow removal of one packet from each replication in each of the following two or more springs. Strings will be attached to the packets, and the distal ends of the strings attached to labeled flags or plot markers to facilitate relocating the packets the next year. Each spring, just before incorporation of cover crops and other amendments, all packets will be removed from the soil. One set will be set aside for analysis of surviving seeds. The other packets will be weighed, opened, the sand and seeds dumped out and mixed with amendment at a rate corresponding to the rate incorporated in the field plots. For cover crop residue, the material will be finely chopped fresh (not dry) residue to mimic what is happening in the plots. While the packets are being processed, cover crops and other amendments will be incorporated in the field plots. After the amendment has been added, the sand and seeds will then be returned to a packet, and reburied in the plot from which they came. The same crop or summer cover crop will be grown on all plots during the summer. If a cover crop is grown, it will be removed prior to planting fall cover crops. To assess the number of seeds remaining in a packet, the sand-seed-residue mix will be spread shallowly in a dish, dried at 40 C, screened to remove the fine sand and decomposed organic matter, and screened again with a larger mesh size to remove coarse organic fragments. Partially decomposed seeds and empty seed coats will be destroyed when screening out the sand. The seeds will be counted and viability assessed by light pressure. We have found from past work that a very high percentage of firm seeds recovered from the soil are viable when tested by germination and tetrazolium. Species that will be investigated include common lambsquarters, smooth pigweed, Powell amaranth, common ragweed, giant foxtail, hairy galinsoga, and velvetleaf. The various labs will coordinate species choice so that each species is studied at two or more locations. Since the first three species above are known to use the presence of nitrate as a germination cue whereas the latter three do not, we will be able to see whether this aspect of weed biology interacts with the type of residue (legume, small grain) in affecting seed persistence in the soil. Differences in seed survival between amendment treatments, burial periods and species will be assessed initially at the state level by time-series ANOVA of arcsine square root transformed seed survival proportions. The experiments will be initiated again in the second project year to provide replication across years with different soil moisture and temperature conditions. Results will be analyzed over initiation years and sites using generalized linear mixed models (Stroup 2011). Study B: Short Term survival of seeds in amended soil. Laboratories participating in this experiment will bury additional packets of species to be tested during the fall of year 0. The following spring, amendments will be added and the packets reburied as for the Medium Term experiment. Packets will be recovered the spring after burial and every three months thereafter for the next year. Based on the literature discussed in Related, Current and Previous Work we expect that effects of amendments on weed seed survival will depend on the type of amendment and the weed species. The Multi-State approach will allow us to screen many weed-amendment combinations with some replication across sites. Identification of amendments that impact seed survival of particular weed species will provide a foundation for design of experiments to elucidate the mechanisms whereby amendments reduce seed survival.
Measurement of Progress and Results
- A refereed journal article on effects of variation in soil moisture and other edaphic factors on efficacy of cultivators and hand tools.
- Article on effects of variation in soil moisture and other edaphic factors on efficacy of cultivators and hand tools published in several growers journals, newsletters and proceedings of grower meetings.
- Refereed journal article on effects of soil cloddiness on efficacy of inter-row cultivators and tine weeders.
- Articles on effects of soil cloddiness on efficacy of inter-row cultivators and tine weeders in several growers journals, newsletters and proceedings of grower meetings.
- One or more refereed journal article on augmentation of weed control with inter-row cultivation in crops planted into roller-crimped cover crops.
- Extension publications on cultivation in roller-crimped cover crops in growers magazines, newsletters and proceedings of grower meetings.
- A refereed journal article on the effects of cover crops, tillage method and simulated rainfall on cultivation efficacy.
- A synthesis article on effects of edaphic factors on weed control by cultivation on the eXtension web site.
- A refereed journal article describing how the timing of life termination and the method of termination affects seed production of a wide range of species.
- Extension articles on effects of mode and timing of weed death on seed production published on web sites and in grower publications.
- A refereed journal article on the effects of cover crop residue on weed seed persistence across multiple states.
- Articles on the effects of cover crop residue and other organic amendments on weed seed persistence in growers journals, newsletters and proceedings of grower meetings.
- Article on effects of cover crop residue and other organic amendments on weed seed persistence on the eXtension web site.
Outcomes or Projected Impacts
- Improved effectiveness of cultivation for weed management resulting in improved yields, greater harvest efficiency and higher net profits for farmers.
- Improved grower knowledge of how and when to remove weeds before they set seeds with consequent reductions in weed seed banks leading to better yields, reduced costs for weed management and higher net profits for farmers.
- A new method for reducing the density of weed seeds in the seed bank leading to better yields, reduced costs for weed management and higher net profits for farmers.
Milestones(2012): Objective 1. Plant cover crops in fall. Till ground or roller-crimp cover crops and plant crops. Measure soil properties, count weeds in quadrats, assess crop stand; cultivate; count weeds in quadrats and assess crop stand after cultivation. Repeat for additional cultivations. Objective 2. Determine which labs will study which species. Plant crops and weeds in spring. Terminate sample plants at various phenological stages and confine plants in mesh to capture seeds as they mature. Objective 3. Plant cover crops, gather seeds, and bury seed packets in the fall. Run tests of initial seed viability. Recover seed packets in the spring, till plots, mix amendments into packets, rebury most packets. Assess seed viability in spring sample packets. Plant and maintain crops or cover crops over the summer and remove crops in the fall in time for cover crop planting. Collect and analyze Study B rapid assessment packet at 3 month intervals.
(2013): Objective 1. Plant cover crops in fall. Analyze data on cultivation efficacy from year 1. Repeat experiments from Year 1. Objective 2. Collect plants in Nov., clean seed and replace some seed on soil in mesh packets. Assess seed viability. In spring, collect over-wintered seeds and assess germination and viability. Plant crops and weeds in spring. Terminate sample plants at various phenological stages and confine plants in mesh. Objective 3. Plant cover crops, gather seeds, and bury seed packets for the second entry point. Run tests of initial seed viability these seeds. Recover seed packets in the spring, till plots, mix amendments into packets and rebury. Assess seed viability in spring sample packets from both entry points. Plant and maintain crops or cover crops over the summer and remove crops in the fall in time for cover crop planting.
(2014): Objective 1. Plant cover crops in fall. Analyze data and assess results from work in Years 1 and 2. Prepare extension publications and present results to grower audiences. Repeat experiments from Years 1 and 2 or conduct new phase 2 experiments based on results so far. Objective 2. Collect Year 2 plants in Nov., clean seed and replace some seed on soil in mesh packets. Assess seed viability. Present preliminary results to grower audiences. In spring, collect over-wintered seeds and assess germination and viability. Plant crops and weeds in spring. Terminate sample plants at various phenological stages and confine plants in mesh. Objective 3. In fall, plant cover crops, gather seeds, and bury seed packets for second phase experiments examining mechanisms affecting seed persistence for species-amendment combinations that show promising responses. Present results to grower audiences. Recover seed packets in the spring, till plots, mix in amendments, rebury packets, and initiate second phase treatments during mixing and reburial. Assess seed viability in spring sample packets. Plant and maintain crops or cover crops over the summer and remove crops in the fall in time for cover crop planting.
(2015): Objective 1. Plant cover crops in fall. Analyze data. Present results to grower audiences. Prepare articles for peer reviewed journals. Repeat phase 2 experiments initiated in Year 3. Objective 2. Collect Year 3 plants in Nov., clean seed and replace some seed on soil in mesh packets. Assess seed viability. Present results to grower audiences. In spring, collect over-wintered seeds and assess germination and viability. Objective 3. In fall, plant cover crops, gather seeds, and bury seed packets for repeat of second phase experiments examining mechanisms. Present results to grower audiences. Recover seed packets in the spring, till plots, mix amendments into packets. Assess seed viability in spring sample packets. Plant and maintain crops or cover crops over the summer and remove crops in the fall in time for cover crop planting.
(2016): Objective 1. Plant cover
Projected ParticipationView Appendix E: Participation
The findings of our research will be communicated to two audiences; growers in the eastern United States and researchers throughout the United States, Canada and elsewhere.
To reach our primary audience of growers, we will present information regarding the various topics addressed here at extension and grower meetings that already occur throughout the winter months in participating states. Many of these meeting have over 1,000 attendees annually. These meetings will also serve in our evaluation plan (see below). In addition to the winter meetings, articles will be published in extension newsletters, grower magazines and on web sites. Web sites that will make this information available to growers include eXtension, The New Agriculture Network (Michigan State University, Purdue University, and University of Illinois), and web sites operated by various participating laboratories.
The short-term goal for our primary audience is to provide much needed information (1) on the effects of soil conditions on the efficacy of cultivation, (2) on the risks associated with leaving terminated weeds in the field that have reached a certain level of maturity, and (3) on the effect of incorporated cover crop residue on persistence of weed seeds in the soil. The evaluation indicators for this goal will be measured through surveys of management tactics at grower meetings. These surveys will be administered after the presentation and ask a variety of questions relevant to the talk. For example, if the talk is on late-season weed seed management, we will ask them what late-season weed management tactics they use and their knowledge of weed seed production and seed banks before and after the presentations. We will also ask whether the presented information will impact future weed management decisions and in what way. The surveys will be evaluated to determine whether the information and the format in which it was provided had a positive impact on growers understanding of ecologically based weed management and intentions to change management strategies. Our longer term goal is to change grower behavior. Our intent is to help growers use cultivation tools more effectively, reduce inputs to the seed bank by removing weeds that are likely to produce seeds, and manage seed banks more effectively
To reach our audience of fellow researchers we will publish several scientific articles covering various aspects of the research (see Measurement of Progress and Results) in refereed journals such as Weed Technology and Weed Science. We will also present our findings at the annual meetings of the Weed Science Society of America and the regional weed science societies.
The project will be executed by a Technical Committee composed of the senior members of each participating laboratory. The officers consist of a chair and secretary who are elected by the committee annually. The secretary normally succeeds to chair the following year. The committee will meet yearly in conjunction with the annual meeting of the Weed Science Society of America. Sub-committees focused on particular objectives will confer as needed by teleconference to plan and coordinate details of the research. Eric Gallandt (Maine) will coordinate the sub-committee for Objective 1: Karen Renner (Michigan State) will coordinate the sub-committee for Objective 2; and Charles Mohler (New York, Cornell) will coordinate the the sub-committee for Objective 3.
Anderson, R. L. 1997. Longspine sandbur (Cenchrus longispinus) ecology and interference in irrigated corn (Zea mays). Weed Technology 11:667-671.
Bond, W. and A. C. Grundy. 2002. Non-chemical weed management in organic farming systems. Weed Res. 41:383-405.
Brainard, D. C., A. DiTommaso, and C. L. Mohler. 2006. Interspecific variation in germination response to ammonium nitrate in powell amaranth (Amaranthus powellii) seeds originating from organic vs. conventional vegetable farms. Weed Sci 54:435-442.
Buhler, D. D. and M. Hoffman. 1999. Anderson's Guide to Practical Methods of Propagating Weeds and Other Plants. Lawrence, KS: Weed Science Society of America.
Chung, Y. R., H. A. H. Hoitink, and P. E. Lipps. 1988. Interactions between organic matter decomposition level and soilborne disease. Agric. Ecosys. Env.24:193-193.
Cussans , G. W. S. Raudonius, P. Brain and S. Cumberwrth. 1996. Effects of depth of seed burial and soil aggregate size on seedling emergence of Alopecurus myosuroides, Glium aparine, Stellaria media and wheat. Weed Research 36:133-141
Davis, A. S. 2007. Nitrogen fertilizer and crop residue effects on seed mortality and germination of eight annual weed species. Weed Sci. 55:123128.
Davis, A. S., K. I. Anderson, S. G. Hallett, and K. A. Renner. 2006. Weed seed
mortality in soils with contrasting agricultural management histories. Weed Sci. 54:291297.
Davis, A. S., J. Cardina, F. Forcella, G. A. Johnson, G. Kegode, J. L. Lindquist, E. C. Luschei, K. A. Renner, C. L. Sprague, and M. M. Williams. 2005. Environmental factors affecting seed persistence of annual weeds across the U.S. Corn Belt. Weed Sci. 53:860868.
Davis, A. S., B. J. Schutte, J. Iannuzzi, and K. A. Renner. 2008. Chemical and physical defense of weed seeds in relation to soil seedbank persistence. Weed Sci. 56:676684.
De Cauwer, B, T. DHose, M. Cougnon, B. Leroy, R. Bucke, and D. Reheul. 2011. Impact of the quality of organic amendments on size and composition of the weed seed bank. Weed Research.(Just published on their web site DOI: 10.1111/j.1365-3180.2010.00840.x)
Dodd, J. 1986. Phenology and seed production of variegated thistle, Silybum marianum (L.) Gaertn., in Australia in relation to mechanical and biological control. Weed Res. 29: 255-263.
Fennimore, S. A. and L. E. Jackson. 2003. Organic amendment and tillage effects on vegetable field weed emergence and seedbanks. Weed Technology 17:4250.
Gills, N. T. 1938. The viability of weed seeds at various stages of maturity. Ann. Appl. Biol. 25:447-456.
Grant, J. D. 1983. The activities of earthworms and the fates of seeds. In, Earthworm Ecology, J. E. Satchell, ed., pp. 107-122. London: Chapman Hall.
Hanson, I. E. 1970. The effects of light, potassium nitrate and temperature on the germination of Chenopodium album L. Weed Res. 10:27-39.
Liebman, M., C. L. Mohler and C. P. Staver. 2001. Preface. In, Ecological Management of Agricultural Weeds, pp. ix-xi. New York: Cambridge University press.
Liebman, M. and E. Gallandt. 1997. Many little hammers: ecological management of cropweed interactions. Pages 291343, in L. Jackson, ed. Ecology in Agriculture. San Diego, CA: Academic Press.
Longyear (1918), cited in Stewart-Wade, S. M., S. Neumann, L. L. Collins, and G. J. Boland. 2002. The biology of Canadian weeds. 117. Taraxacum officinale G. H. Weber ex Wiggers. Canadian Journal of Plant Science 82:825-853.
Lovely, W. G., C. R. Weber, and D. W. Stniforth. 1958. Effectiveness of the rotary hoe for weed control in soybeans. Agron, J. 50:621-625.
McRill, M. and G. R. Sagar. 1973. Earthworms and seeds. Nature 243:482.
Menalled, F. D., A. K. Kohler, D. D. Buhler, and M. Liebman. 2005. Effects of composted swine manure on weed seedbank. Agric. Ecosys. Env. 111:163-169.
Mirsky, S. B., E. R. Gallandt, D. A. Mortensen, W. S. Curran, and D. L. Shumway (2010). Reducing the germinable weed seedbank with soil disturbance and cover crops. Weed Research 50:341-352
Mohler, C. L., J. C. Frisch, and J. Mt. Pleasant. 1997. Evaluation of mechanical weed management programs for corn (Zea mays). Weed Tech. 11:123-131.
Mohler, C. L., C. Dykeman, E. Nelson, and A. DiTommaso. Reduction of weed seedling emergence by pathogens following incorporation of green crop residue. Weed Res., In prep.
Peters, J., Ed. 2000. Tetrazolium testing handbook. Contrib. No. 29 to the handbook on seed testing. Lincoln, NE, Association of Official Seed Analysts. Pp 151-154.
Regnier, E., S. K. Harrison, J. Liu, J. T. Schmoll, C. A. Edwards, N. Arancon, and C. Holloman. 2008. Impact of an exotic earthworm on seed dispersal of an indigenous US weed. J. Appl. Ecol. 45:1621-1629.
Roberts, H. A. and P. M. Lochkett. 1975. Germination of buried and dry-stored seeds of Stellaria media. Weed Res. 15:199-204.
Stroup, W. W. 2011. Living with generalized linear mixed models. SAS Global Forum, Paper 349-2011. http://support.sas.com/resources/papers/proceedings11/349-2011.pdf
Ullrich, S. D., J. S. Buyer, M. A. Cavigelli, R. Seidel, and J. R Teasdale. 2011. Weed seed persistence and microbial abundance in long-term organic and conventional cropping systems. 59:202-209.
Wagner, M. and N. Mitschunas. 2008. Fungal effects on seed bank persistence and potential applications in weed biocontrol: a review. Basic Appl. Ecol. 9:191203.