Dose–response analysis is widely used in biological sciences and has application to a variety of risk assessment, bioassay, and calibration problems. In weed science, dose–response methodologies have typically relied on least squares estimation under the assumptions of normal, homoscedastic, and independent errors. Advances in computational abilities and available software, however, have given researchers more flexibility and choices for data analysis when these assumptions are not appropriate. This article will explore these techniques and demonstrate their use to provide researchers with an up-to-date set of tools necessary for analysis of dose–response problems. Demonstrations of the techniques are provided using a variety of data examples from weed science.

Weeds that emerge between rows in fresh market tomatoes after the critical period of competition are not suppressed by the crop and can produce large quantities of seed. A living mulch planted between rows might limit weed seed production. Buckwheat was seeded between tomato rows after the critical period in 2007 and 2008 in field studies near Lafayette, IN. Weeds were allowed to emerge after the critical period (CP), controlled throughout the growing season (no seed threshold [NST]), or mowed to limit seed production (MOW). Buckwheat and MOW plots were mowed twice after the critical period in 2007 and once in 2008. Seed banks were sampled after the critical period and in the following spring. Tomato yields were not reduced by growing buckwheat between rows. Seed bank densities for common purslane and carpetweed, which escaped mowing due to their prostrate habits, increased in all treatments. Germinable seed bank densities were 306 seeds m−2 or less in the NST and buckwheat treatments but 755 seeds m−2 or more in the CP treatments for species with erect habits in both years. Seed bank densities were lower in the MOW treatment than in the CP treatments in 2007 but not in 2008. In a parallel experiment conducted in adjacent plots, buckwheat was seeded at five rates (0, 56, 112, 168, and 224 kg seed ha−1). Plots were mowed and emergent weeds sampled as described for the intercrop experiment. Weed densities before mowing decreased linearly with buckwheat seed rate. After mowing, no relationship was detected between seed rate and weed densities. This study supports the hypothesis that a living mulch planted after the critical period can be used to limit seed bank growth without reducing tomato yields, but additional research is needed to better understand the effect of mowing on living mulch growth and weed suppression.

Research was conducted at experimental research stations near Keiser and Marianna (Marianna-A), AR, in 2007, and in a grower's field near Marianna (Marianna-B), AR, in 2008, to compare herbicide programs, including POST application(s) of glyphosate/glufosinate alone or in combination with residual herbicides applied as PRE, mid-POST (MPOST), or layby POST-directed (PD) in enhanced glyphosate- and glufosinate-resistant cotton. Weed species evaluated included Palmer amaranth, pitted morningglory, hemp sesbania, barnyardgrass, and a mixture of large crabgrass and goosegrass. At Marianna-B, AR, the Palmer amaranth population was a mixture of glyphosate-resistant and -susceptible plants. For both cotton cultivars and at all locations, inclusion of S-metolachlor plus fluometuron PRE increased weed control and/or decreased the number of glufosinate or glyphosate applications needed in-season. At Marianna-B, AR, PRE residual herbicides and/or glufosinate were required to control glyphosate-resistant Palmer amaranth. Addition of pyrithiobac to glufosinate or glyphosate did not increase weed control. A layby PD application of flumioxazin plus MSMA was required to increase late-season control of all weed species in POST glufosinate-only programs, but not in POST glyphosate-only programs. None of the programs caused > 5% injury to either cotton cultivar. Seed-cotton yield was similar in all herbicide programs at Keiser, AR, and Marianna-A, AR, except for the POST glyphosate-only program, which yielded less than the PRE followed by POST programs in glyphosate-resistant cotton at Keiser, AR. In general, PRE herbicides did not increase cotton yield but did improve early and late-season control of glyphosate-susceptible and -resistant weeds in both cotton cultivars.

A four-year experiment was conducted near Scottsbluff, NE to evaluate the effect of different methods of preplant tillage, post-plant cultivation, and herbicides on weed density and grain yield. Ridge-till enhanced the development of kochia and reduced the density of wild proso millet and common lambsquarters. Tandem disking increased longspine sandbur and redroot pigweed density whereas moldboard plowing increased common sunflower density. Cultivation reduced weed density 86% but weeds that remained after cultivation reduced corn grain yield by 40% as compared with a hand-weeded control. Weed control treatments that integrated cultivation and herbicides controlled a broader spectrum of weed species than cultivation or herbicide treatments alone.

Field experiments were conducted to compare non-phenoxy herbicides with 2,4-D for broadleaf plantain, dandelion, and white clover control in cool-season turf. Herbicides tested were BAS 514, chlorflurenol, clopyralid, dicamba, triclopyr, and 2,4-D, alone and in combinations. Broadleaf plantain was controlled by 2,4-D and by clopyralid plus triclopyr at 0.21 plus 0.63 kg ae ha-1. Dandelion was controlled by 2,4-D, BAS 514 at 0.84 kg ha-1, and by the combination of clopyralid plus triclopyr at 0.16 plus 0.47 kg ha-1. Chlorflurenol, clopyralid, dicamba, or triclopyr did not control dandelion. White clover was controlled by all herbicides except 2,4-D. The combination of clopyralid plus triclopyr was the only non-phenoxy treatment which controlled broadleaf plantain, dandelion, and white clover.

A morningglory at my window satisfies me more than the metaphysics of books.

Song of Myself, Walt Whitman, 1855 (Mitchell 1998)

Yellow starthistle is an aggressive annual forb that has invaded millions of hectares of California's annual range. Control efforts such as burning and herbicides have been effective for short-term management. However, recruitment from the seedbank or reinvasion of the annual grassland system results in a rapid return to yellow starthistle dominance. Establishing perennial grasses would be ideal for suppression of yellow starthistle. However, a lack of effective weed control options in California during a seeding program has limited perennial grass establishment. Clopyralid was used to control yellow starthistle annually for 1, 2, or 3 yr to provide a window of reduced competition for pubescent wheatgrass establishment. Total plant cover, yellow starthistle density, biomass, and seedhead number were quantified for 6 yr. Clopyralid treatment significantly reduced yellow starthistle and allowed pubescent wheatgrass establishment with a single treatment. Both clopyralid treatment and pubescent wheatgrass establishment significantly affected the range plant community composition. Annual grasses and forbs increased in plots only treated with clopyralid for 2 or 3 yr, whereas clopyralid-treated pubescent wheatgrass plots maintained lower annual grass and forb cover. Integrating pubescent wheatgrass seeding with clopyralid treatment provided long-term yellow starthistle suppression, whereas clopyralid treatment alone resulted in a plant community susceptible to repeated invasion. These findings support the establishment of competitive perennial grasses in annual grasslands as an important component of long-term yellow starthistle management.

Field experiments were conducted in 1996 and 1997 to evaluate barnyardgrass (Echinochloa crus-galli) control, rice (Oryza sativa) yield, and grain quality with cyhalofop-butyl applied early postemergence (EPOST) or late postemergence (LPOST) in drained or flooded plots at rates of 0.15 to 0.4 kg ai/ha and standard treatments of molinate or propanil. Control with cyhalofop-butyl at 0.15 and 0.2 kg/ha applied in drained plots was greater compared to propanil and generally increased when treatments were applied EPOST to plots reflooded 2 d after treatment. Barnyardgrass control with cyhalofop-butyl at 0.15 and 0.2 kg/ha applied in flooded plots was similar to or better than the control obtained with molinate. Rice yields generally reflected differences in barnyardgrass control and were greatest with cyhalofop-butyl applied EPOST in drained plots. The 0.4 kg/ha rate of cyhalofop-butyl resulted in rice foliar injury and significant reductions in grain yield. Grain qualitative characteristics (total milling yield, grain vitreosity, grain length, and grain length to width ratio) were better with cyhalofop-butyl applied EPOST in drained plots compared with propanil or molinate.

Herbicides and herbicide prepackaged mixtures registered for use on established bermudagrass turf may cause significant injury to recently seeded bermudagrass cultivars, delaying full establishment. Research was conducted to evaluate the use of 12 herbicide treatments applied at onset of uniform stolon development (4 to 8 wk after seeding) to recently seeded ‘Princess 77,’ ‘Riviera,’ ‘Savannah,’ and ‘Yukon’ bermudagrass cultivars. In general, Yukon was more susceptible to herbicide injury than other cultivars. Atrazine at 1.1 kg ai/ha injured all cultivars 55 to 59% 14 d after initial treatment (DAIT), which lead to reduced bermudagrass cover 21 DAIT. Triclopyr + clopyralid at 0.63 + 0.21 kg ae/ha, respectively, injured Savannah and Yukon greater than other broadleaf weed herbicides (2,4-D + mecoprop + dicamba or 2,4-D + clopyralid + dicamba). Foramsulfuron did not injure or reduce bermudagrass cover of any cultivar evaluated at any rating date. The only adverse effect of trifloxysulfuron was a reduction in Riviera and Yukon ground cover at 21 DAIT. MSMA applied sequentially and quinclorac treatments did not injure or reduce ground cover of Savannah or Princess 77. MSMA applied sequentially and/or quinclorac injured and reduced ground cover of Riviera and Yukon; however, both cultivars completely recovered from MSMA or quinclorac injury by 42 DAIT.

Glyphosate-resistant Palmer amaranth has become a major problem for cotton producers throughout much of the southern United States. With cotton producers relying heavily on glyphosate-resistant cotton, an alternative solution to controlling resistant Palmer amaranth is needed. A field experiment was conducted during 2009 and 2010 at Marianna, AR, in which a rye cover crop and no cover crop were tested in combination with deep tillage with the use of a moldboard plow and no tillage to determine the impact on Palmer amaranth emergence in cotton. To establish a baseline population, 500,000 glyphosate-resistant Palmer amaranth seeds were placed in a 2-m2 area in the middle of each plot and incorporated into the soil, and emergence was evaluated five times during the season. In 2009, both tillage and the cover crop reduced Palmer amaranth emergence in cotton, but the combination of the two reduced emergence 85%. In the second year, only the cover crop reduced Palmer amaranth emergence in cotton, a 68% reduction. Cover crops and deep tillage will not eliminate glyphosate-resistant Palmer amaranth; however, use of these tools will likely reduce the risks of failures associated with residual herbicides along with selection pressure placed on both PRE- and POST-applied herbicides. Additional efforts should focus on the integration of the best cultural practices identified in this research with use of residual herbicides and greater focus on limiting Palmer amaranth seed production and reducing the soil seedbank.

Dose-response experiments were conducted on a biotype of prickly lettuce collected from Whitman County, WA, to determine the level of resistance to 2,4-D. Initially, progeny of prickly lettuce that survived two applications of glyphosate and 2,4-D in mixture were collected to determine if antagonism of the 2,4-D or glyphosate was occurring. Prickly lettuce survival was determined to not be due to antagonism of 2,4-D or glyphosate when the two herbicides were applied in mixture. The doses required to reduce growth 50% (GR50 ) for resistant and susceptible field-collected prickly lettuce were 150 and 6 g ae/ha 2,4-D, respectively, indicating the resistant biotype was 25 times more resistant to 2,4-D than the susceptible biotype. The resistant biotype expressed injury but produced regrowth following application. A dose of 2,4-D at 220 g/ha was required to reduce regrowth frequency 50% (FR50 ) for resistant field-collected prickly lettuce. Regrowth was also observed with the susceptible biotype, although the FR50 was much lower (10 g/ha), resulting in an R/S ratio of 22 based on the respective FR50 values. A rate of 4,300 g/ha 2,4-D (10 times the maximum labeled rate in wheat) was required to reduce the regrowth frequency in the resistant biotype to zero.

Two field experiments were conducted between 1990 and 1992 under sprinkler irrigation at the Research and Extension Center, Torrington, WY to determine the influence of mixed densities and durations of kochia and green foxtail interference in sugarbeet. Sugarbeet root yield and top weight generally decreased as densities of green foxtail and kochia increased whether alone or in combination. Reductions in sugarbeet root yield and above ground biomass from mixed densities of kochia and green foxtail were additive at the low and intermediate density but less than additive at the high density of either species. Sugarbeet root yield decreased as the duration of interference after sugarbeet emergence from a mixed density of kochia and green foxtail increased. Since sugarbeet plants were irrigated to avoid water stress and adequate nutrients were applied, it appears that kochia and green foxtail interfered with sugarbeet primarily for light, based on light and height measurements. Season-long as well as duration of interference did not show any significant effect on sucrose content. Based on regression analysis the lowest densities of kochia and green foxtail required to reduce root yield were approximately 0.3 and 0.06 plants/m of row, respectively. The minimum duration of time that 0.5 kochia and 3.0 green foxtail plants/m of row can interfere with sugarbeets before root yield is economically reduced was approximately 3.5 wk after sugarbeet emergence.

Twelve winter cover crops were planted in Citra and Live Oak, FL, in 2004, to evaluate their potential for use as living mulches in organic vegetable production: black oat, rye, annual ryegrass, hard fescue, two cultivars of white clover, berseem clover, crimson clover, subterranean clover, arrowleaf clover, barrel medic, and a hybrid disc × strand medic cultivar. The best canopy development and weed suppression occurred with black oat, rye, and annual ryegrass. In 2005, black oat, two rye cultivars, and annual ryegrass were evaluated as living mulches in broccoli at Citra and Live Oak, using organic production methods. ‘Florida 401’ (FL 401) rye was tallest, black oat was intermediate, and ‘Wrens Abruzzi’ (WA) rye and ‘Gulf’ ryegrass were of similar height and were the shortest living mulches. Biomass harvested at 12 and 13 wk after planting at Citra and Live Oak, respectively, was greatest with FL 401 rye. At Live Oak, the three other mulches had similar amounts of biomass; however, at Citra, black oat biomass was greater than that of WA rye, and biomass of ryegrass was lowest. The greatest weed infestation occurred with the weedy control. Weed biomass was highest with the weedy control, intermediate with ryegrass, and lowest with rye and black oat. However, the biomass of the weedy control was lower than that of the living mulches plus any associated weeds. Marketable broccoli yield was highest with the weed-free control. Yields with black oat, WA rye, and ryegrass were similar to that of the weedy control, whereas yield with the FL 401 rye was lower than with the weedy control. Suppression of living mulches by mowing at 3 and 7 wk after planting had no effect on broccoli growth or yield.

Two separate field experiments were conducted over a 2-yr period in Fayetteville, AR, during 2012 and 2013 to (1) evaluate POST herbicide programs utilizing a premixture of dimethylamine (DMA) salt of glyphosate + choline salt of 2,4-D in a soybean line resistant to 2,4-D, glyphosate, and glufosinate and (2) determine efficacy of herbicide programs that begin with PRE residual herbicides followed by POST applications of 2,4-D choline + glyphosate DMA on glyphosate-resistant Palmer amaranth. In the first experiment, POST applications alone that incorporated the use of residual herbicides with the glyphosate + 2,4-D premixture provided 93 to 99% control of Palmer amaranth at the end of the season. In the second experiment, the use of flumioxazin, flumioxazin + chlorimuron methyl, S-metolachlor + fomesafen, or sulfentrazone + chloransulam applied PRE provided 94 to 98% early-season Palmer amaranth control. Early-season control helped maintain a high level of Palmer amaranth control throughout the growing season, in turn resulting in fewer reproductive Palmer amaranth plants present at soybean harvest compared to most other treatments. Although no differences in soybean yield were observed among treated plots, it was evident that herbicide programs should begin with PRE residual herbicides followed by POST applications of glyphosate + 2,4-D mixed with residual herbicides to minimize late-season escapes and reduce the likelihood of contributions to the soil seedbank. Dependent upon management decisions, the best stewardship of this technology will likely rely on the use multiple effective mechanisms of action incorporated into a fully integrated weed management system.

Pesticide bans in Canada have resulted in a requirement for municipal turfgrass managers to use cultural methods of weed control to provide a safe playing surface for athletes. A field study was conducted to determine if overseeding provides enough competition to decrease weed populations in Kentucky bluegrass athletic turf typically used in municipal parks for recreation. Perennial ryegrass was overseeded at 2, 4, and 8 kg/100 m2 in May, July, or September, and all permutations of these timings in nonirrigated and irrigated trials at the Guelph Turfgrass Institute (GTI) field station in Guelph, and on in-use soccer fields at the University of Guelph campus and in the town of Oakville, Ontario, Canada over 2 yr. Plant cover by species was recorded every other month using a randomized point quadrat method throughout the growing seasons of 2005 and 2006. Weed populations were not affected by overseeding in 2005, a dry growing season. However, when weed populations were high and normal growing conditions existed in 2006, overseeding applications in May/July/September at 4 and 8 kg/100 m2 decreased perennial weed cover, specifically white clover in the irrigated trial and dandelion in the nonirrigated trial at the GTI. An increase in perennial ryegrass was observed in all plots that received an overseeding treatment. Treatments applied on the in-use soccer fields in Oakville and Guelph, which included May/September and May only overseedings, had no effect on weed populations or perennial ryegrass populations compared to the weedy control. Over the short term, high-rate and frequent overseeding with perennial ryegrass appears to provide competition against perennial weeds when weed cover is high and should be considered an important part of a weed management program for municipal turfgrass managers.

Laboratory and field experiments were conducted to determine whether manipulation of radiation environment during the period of winter wheat establishment can be used to improve wheat yields in Italian ryegrass-infested fields. Percentage of total irradiation and ratio of red (R) light (around 660 ηm) to far-red (FR) light (around 730 ηm) reaching the soil surface were important factors in regulating Italian ryegrass germination, growth, and competitive interactions with wheat. Reducing total irradiation to about 10% of full sunlight while maintaining the normal R/FR ratio of about 1.0 reduced wheat grain production in the presence of Italian ryegrass by about 40% compared with weed-free wheat in full sunlight. Further, reducing total irradiation to 3% of full sunlight plus reducing the R/FR ratio to about 0.2 reduced wheat grain production competing with Italian ryegrass by about 35% compared with production in the control conditions. Wheat production of dry matter, spikes, and seeds in the presence of Italian ryegrass increased more than threefold under both shading treatments compared with production in full sunlight.

Glyphosate at simulated drift rates representing 12.5, 6.3, and 1.6% of the usage rate of 1,120 g ai/ha (140, 70, and 18 g/ha, respectively) was applied to wheat at first node, boot stage, or at early flowering. At 14 d after treatment (DAT) wheat injury, expressed as bleaching of leaf foliage and growth inhibition, was 40 to 55% for 70 g/ha applied at first node and for 140 g/ha applied at all growth stages. Wheat height 28 DAT was reduced 47% with glyphosate applied at 140 g/ha at first node and was reduced around 26% for 70 g/ha applied at first node and 140 g/ha applied at boot stage. Wheat height was not reduced with glyphosate at 18 g/ha applied at first node or boot stage and with all rates applied at early flowering. Wheat yield was reduced 72% when glyphosate was applied at 140 g/ha at first node, 45% when applied at boot stage, and 54% when applied at early flowering. For 70 g/ha, wheat yield was reduced 25 to 30% for the three application timings. Wheat yield was not reduced for 18 g/ha glyphosate. In another study, six wheat varieties responded the same to glyphosate applied at 140 and 70 g/ha. Wheat height 28 DAT was reduced an average of 34% for 140 g/ha glyphosate and 17% for 70 g/ha applied at first node, but height was not reduced when applied at early flowering. Yield was reduced an average of 58 and 43% for 140 and 70 g/ha applied at first node and 38 and 19% for 140 and 70 g/ha applied at early flowering. In both studies yield reductions in most cases were reflected in reduced spike density, spikelet number per spike, and seed weight.

PRE herbicides are less effective in the zero-tillage system because of increased residual crop stubble and reduced soil incorporation. However, since weeds are not physically controlled in the zero-tillage system, reliance on efficacy of PRE herbicides is increased. This research investigated the impact of carrier volume and droplet size on the performance of PRE herbicides (in wheat crops at four sites in 2010) to improve herbicide efficacy in conditions of high stubble biomass in zero-tillage systems. Increasing carrier volume from 30 to 150 L ha−1 increased spray coverage on water-sensitive paper from an average of 5 to 32%. Average control of rigid ryegrass by trifluralin (at Cunderdin and Merredin sites) and trifluralin or pyroxasulfone (at Wickepin and Esperance sites) improved from 53 to 78% with increasing carrier volume. Use of ASABE Medium droplet size improved spray coverage compared with ASABE Extremely Coarse droplet size, but did not affect herbicide performance. It is clear that increased carrier volume improves rigid ryegrass weed control for nonwater-soluble (trifluralin) and water-soluble (pyroxasulfone) PRE herbicides. Western Australian growers often use low carrier volumes to reduce time of spray application or because sufficient high-quality water is not available, but the advantages of improved weed control justifies the use of a high carrier volume in areas of high weed density.

Olive jift is a solid by-product of olive (Olea europea) oil processing. Greenhouse experiments were conducted to evaluate olive jift effect on broomrape (Orobanche spp.) infections on three crops. Soil—jift mixtures were used as potting medium in ratios of 1:0, 1:1, and 3:1 soil to jift. Broomrape seeds were evenly incorporated in the potting media at 0, 150, 300, 450, 600, 750, and 900 seeds/kg. Pots planted with faba bean or pea were inoculated with seeds of O. crenata Forsk., whereas pots planted with tomato were inoculated with seeds of O. lavandulacea Reichenb. Jift in soil reduced broomrape germination and infection on all three hosts regardless of inoculation densities. Pea was not infected with broomrape in jift-containing media at any inoculation density, whereas only sporadic broomrape infections were observed in faba bean and tomato grown in jift-mixed soils. Fresh and dry weights of all crops were not adversely affected by mixing jift with soil. These results suggest a possible use of jift as an inexpensive organic material for broomrape control.

This study was designed to develop a protocol for using a biologically-based system to detect and tract airborne herbicides. Common bean, lentil, and pea were selected for their quasi-diagnostic sensitivity to chlorsulfuron, thifensulfuron, metsulfuron, tribenuron, paraquat, glyphosate, bromoxynil, 2,4-D, and dicamba. Plants were grown in the greenhouse at Prosser, WA, and placed at 25 exposure sites at weekly intervals between Apr. 2 and Oct. 15, 1991. After 1 wk of field exposure plants were brought back and observed for herbicide symptoms over a 28-d period. Symptoms that developed were compared with symptoms caused by disease, insects, adverse weather conditions, and herbicides applied at different rates under controlled conditions on these species. In addition, if herbicide symptoms were observed, herbicide spray records and weather data in the area were used in a computer model to determine the source of potential herbicide drift. This study demonstrates that indicator plant species selected for high sensitivity to herbicides can be used to monitor the occurrence of herbicide movement.