Intensified cover-cropping practices are increasingly viewed as a herbicide-resistance management tool but clear distinction between reactive and proactive resistance management performance targets is needed. We evaluated two proactive performance targets for integrating cover-cropping tactics, including (1) facilitation of reduced herbicide inputs and (2) reduced herbicide selection pressure. We conducted corn (Zea mays L.) and soybean [Glycine max (L.) Merr.] field experiments in Pennsylvania and Delaware using synthetic weed seedbanks of horseweed [Conyza canadensis (L.) Cronquist] and smooth pigweed (Amaranthus hybridus L.) to assess winter and summer annual population dynamics, respectively. The effect of alternative cover crops was evaluated across a range of herbicide inputs. Cover crop biomass production ranged from 2,000 to 8,500 kg ha−1 in corn and 3,000 to 5,500 kg ha−1 in soybean. Experimental results demonstrated that herbicide-based tactics were the primary drivers of total weed biomass production, with cover-cropping tactics providing an additive weed-suppression benefit. Substitution of cover crops for PRE or POST herbicide programs did not reduce total weed control levels or cash crop yields but did result in lower net returns due to higher input costs. Cover-cropping tactics significantly reduced C. canadensis populations in three of four cover crop treatments and decreased the number of large rosettes (>7.6-cm diameter) at the time of preplant herbicide exposure. Substitution of cover crops for PRE herbicides resulted in increased selection pressure on POST herbicides, but reduced the number of large individuals (>10 cm) at POST applications. Collectively, our findings suggest that cover crops can reduce the intensity of selection pressure on POST herbicides, but the magnitude of the effect varies based on weed life-history traits. Additional work is needed to describe proactive resistance management concepts and performance targets for integrating cover crops so producers can apply these concepts in site-specific, within-field management practices.

The long-term success of weed management programs requires that all crops in a rotation receive satisfactory weed control. Band sowing with inter-band hoeing has been proposed as an innovative weed management strategy for grain crops. In the band-sowing system, crops are sown in a broadcast pattern within a band of some chosen width (here we selected 12.7 cm); weeds between bands are controlled with inter-band hoeing, with or without so-called “blind cultivation,” for example, tine harrowing. Alteration of the crop spatial arrangement from typical single-line rows to a more evenly distributed pattern aims to enhance interspecific competition while reducing intraspecific competition. Field experiments, conducted in Maine in 2016 and 2017, compared band sowing with inter-band hoeing to the region’s standard practice of planting in 16.5-cm rows and tine harrowing in four test crops: spring wheat (Triticum aestivum L. ‘Glenn’), oat (Avena sativa L. ‘Colt’), field pea (Pisum sativum L. ‘Jetset’), and flax (Linum usitatissimum L. ‘Prairie Thunder’). Band sowing improved weed control relative to the standard practice, especially in crops with greater competitive ability (wheat and oat). Despite improved weed control, in most cases, yields were unaffected by treatment. While band sowing with hoeing provided improved weed control in multiple crops, further study is warranted to optimize seeding rate, band width, and inter-band width to improve crop yields.

Spot spraying POST herbicides is an effective approach to reduce herbicide input and weed control cost. Machine vision detection of grass or grass-like weeds in turfgrass systems is a challenging task due to the similarity in plant morphology. In this work, we explored the feasibility of using image classification with deep convolutional neural networks (DCNN), including AlexNet, GoogLeNet, and VGGNet, for detection of crabgrass species (Digitaria spp.), doveweed [Murdannia nudiflora (L.) Brenan], dallisgrass (Paspalum dilatatum Poir.), and tropical signalgrass [Urochloa distachya (L.) T.Q. Nguyen] in bermudagrass [Cynodon dactylon (L.) Pers.]. VGGNet generally outperformed AlexNet and GoogLeNet in detecting selected grassy weeds. For detection of P. dilatatum, VGGNet achieved high F1 scores (≥0.97) and recall values (≥0.99). A single VGGNet model exhibited high F1 scores (≥0.93) and recall values (1.00) that reliably detected Digitaria spp., M. nudiflora, P. dilatatum, and U. distachya. Low weed density reduced the recall values of AlexNet at detecting all weed species and GoogLeNet at detecting Digitaria spp. In comparison, VGGNet achieved excellent performances (overall accuracy = 1.00) at detecting all weed species in both high and low weed-density scenarios. These results demonstrate the feasibility of using DCNN for detection of grass or grass-like weeds in turfgrass systems.

Cover crop residue can act as a mulch that will suppress weeds, but as the residue degrades, weed suppression diminishes. Biomass of cover crop residue is positively correlated to weed suppression, but little research is available regarding the composition of cover crop residue and its effect on weed suppression. Field experiments were conducted to determine the impact of cover crop residue properties (i.e., total carbon, total nitrogen, lignin, cellulose, and hemicellulose) on summer annual weed suppression and cash crop yield. Cover crop monocultures and mixtures were planted in the fall and designed to provide a range of biomass and residue properties. Cover crops were followed by corn (Zea mays L.) or soybean [Glycine max (L.) Merr.]. At termination, cover crop biomass and residue components were determined. Biomass ranged from 3,640 to 8,750 kg ha−1, and the carbon-to-nitrogen (C:N) ratio ranged from 12:1 to 36:1. As both cover crop biomass and C:N ratio increased, weed suppression and duration of suppression increased. For example, a C:N ratio of 9:1 is needed to suppress redroot pigweed (Amaranthus retroflexus L.) 50% at 4 wk after termination (WAT), and that increases to 16:1 and 20:1 to have 50% suppression at 6 and 8 WAT, respectively. Similarly, with biomass, 2,800 kg ha−1 is needed for 50% A. retroflexus suppression at 4 WAT, which increases to 5,280 kg ha−1 and 6,610 kg ha−1 needed for 50% suppression at 6 and 8 WAT, respectively. In general, similar trends were observed for pitted morningglory (Ipomoea lacunosa L.) and large crabgrass [Digitaria sanguinalis (L.) Scop.]. Corn and soybean yield increased as both cover crop biomass and C:N ratio increased where no weed control measures were implemented beyond cover crop. The same trend was observed with cash crop yield in the weed-free subblocks, with one exception. This research indicates that cover crop residue composition is important for weed control in addition to biomass.