Sorghum cover crop and repeated soil fumigation for purple nutsedge management in tomato production.

BACKGROUND: Purple nutsedge (Cyperus rotundus L.) is one of the most common and troublesome weeds. Field research trials were conducted in Florida to evaluate the effects of repeated fumigation and a sorghum sudangrass [Sorghum bicolor S. bicolor var. sudanense (Piper) Stapf.] cover crop on purple nutsedge (Cyperus rotundus L.) populations over time in tomato (Solanum lycopersicum L.) production. RESULTS: Among the soil fumigants, DMDS + metam potassium was consistently the most effective treatment in terms of in-crop purple nutsedge control. Plots with a sorghum cover crop during the fallow period exhibited higher purple nutsedge density during the tomato growing season as well as higher purple nutsedge shoot and tuber densities during the fallow period compared to the chemical fallow. CONCLUSION: DMDS + metam potassium was the most effective fumigant for purple nutsedge control. Unexpectedly, a sorghum cover crop during the fallow period was less effective than chemical fallow for purple nutsedge management, and therefore we do not recommend the use of sorghum cover crops for weed management in fields where purple nutsedge is the major weed species.

Fallow management input strategies for weed management in Florida cabbage production.

BACKGROUND: Fallow period weed management is an important Florida production consideration due to its duration and impact on the cash crop. Cover cropping is a valuable summer fallow period option for weed suppression. Sorghum-sudangrass is a commonly used, competitive, and allelopathic Florida summer cover crop. The effect of increased seeding rate, a herbicide application, and added fertilizer inputs during the fallow period on the cover crop, weed populations, and cabbage yield was explored and compared to nontreated and chemical fallow controls. RESULTS: Increased sorghum-sudangrass seeding rates had no effect on the resultant stand density or biomass compared to the standard seeding rate. Cover cropping did not consistently suppress purple nutsedge, Florida pusley, or wild radish and added fertilizer inputs produced variable results. S-metolachlor enhanced purple nutsedge suppression at low densities but did not improve grass, wild radish, and Florida pusley suppression. CONCLUSIONS: Increased fallow management inputs did not consistently enhance weed suppression or provide benefit to the cash crop. Sorghum-sudangrass suppressed Poaceae densities during the fallow period but did not adequately suppress nutsedge, wild radish, or Florida pusley densities over time. We conclude that weed management inputs should be focused on the cash crop and that enhanced management during the fallow period has limited benefit. © 2021 Society of Chemical Industry.

Long-term effect of fumigation and a sorghum cover crop on broadleaf and grass weeds in plastic-mulched tomato.

BACKGROUND: Broadleaf and grass weeds can adversely affect growth and productivity of plastic-mulched tomato (Solanum lycopersicum L.). Two, four-year research trials were conducted in Florida to evaluate the effect of repeated fumigation and chemical fallow versus a sorghum [Sorghum bicolor S. bicolor var. sudanense (Piper) Stapf.] cover crop on broadleaf and grass weeds in tomato plasticulture. RESULTS: 1,3-Dichloropropene (1,3-D) + chloropicrin (Pic), dimethyl disulfide (DMDS) + Pic, and DMDS + metam potassium effectively controlled broadleaf weeds in-crop and reduced densities by 79-98% compared to the non-fumigated control but provided inconsistent control of grass weeds. DMDS + metam potassium was generally the most effective fumigant. During the fallow period, a sorghum cover crop effectively reduced broadleaf weed density than the chemical fallow, while chemical fallow effectively reduced grass weed density than the cover crop. The fallow program did not affect in-crop densities of broadleaf and grass weeds. In some measurements, the evaluated fumigants resulted in taller tomato plants and higher yield compared to the non-fumigated control. CONCLUSION: We conclude that the evaluated soil fumigants effectively control broadleaf and grass weeds. Planting a sorghum cover crop effectively suppresses broadleaf weeds but not grasses during the fallow period. However, this suppression does not result in reduced weed density in-crop despite the fact that similar weed species were observed in both time periods. © 2020 Society of Chemical Industry.

Goosegrass Detection in Strawberry and Tomato Using a Convolutional Neural Network.

Goosegrass is a problematic weed species in Florida vegetable plasticulture production. To reduce costs associated with goosegrass control, a post-emergence precision applicator is under development for use atop the planting beds. To facilitate in situ goosegrass detection and spraying, tiny- You Only Look Once 3 (YOLOv3-tiny) was evaluated as a potential detector. Two annotation techniques were evaluated: (1) annotation of the entire plant (EP) and (2) annotation of partial sections of the leaf blade (LB). For goosegrass detection in strawberry, the F-score was 0.75 and 0.85 for the EP and LB derived networks, respectively. For goosegrass detection in tomato, the F-score was 0.56 and 0.65 for the EP and LB derived networks, respectively. The LB derived networks increased recall at the cost of precision, compared to the EP derived networks. The LB annotation method demonstrated superior results within the context of production and precision spraying, ensuring more targets were sprayed with some over-spraying on false targets. The developed network provides online, real-time, and in situ detection capability for weed management field applications such as precision spraying and autonomous scouts.

Pest control with drip-applied dimethyl disulfide and chloropicrin in plastic-mulched tomato (Solanum lycopersicum L.).

BACKGROUND: Dimethyl disulfide (DMDS) is used as a preplant soil fumigant for weed and soilborne pathogen control in plasticulture vegetable crops. The objective of this research was to determine the control efficacy of emulsifiable concentrate (EC) formulation of DMDS or DMDS + chloropicrin (Pic) on weed and Fusarium wilt in tomato (Solanum lycopersicum L.) plasticulture. RESULTS: The effective DMDS rates required to provide 50% (ER50 ) control of purple nutsedge (Cyperus rotundus L.) were 210 and 340 kg ha-1 at 4 weeks after fumigation (WAF) in fall 2017 and fall 2018, respectively, while these values increased to 348 and >467 kg ha-1 , respectively, at 12 WAF. The ER50 values of DMDS + Pic were 150 and 240 kg ha-1 at 4 WAF in fall 2017 and fall 2018, respectively, while these values increased to 255 and 450 kg ha-1 , respectively, at 12 WAF. DMDS + Pic was generally more effective than DMDS for C. rotundus control. The high rates of DMDS or DMDS + Pic provided adequate C. rotundus control in early season but failed to provide effective control by season end. In addition, DMDS + Pic injections through drip tape effectively reduced Fusarium oxysporum f. sp. lycopersici (FOL) inoculum while DMDS alone was generally ineffective. CONCLUSION: Injection of the EC formulation of DMDS or DMDS + Pic through drip tape should have been provided a viable option for C. rotundus and Fusarium wilt control in plastic-mulched tomato. However, supplemental weed management actions, such as herbicide applications, may be required to achieve season-long control. © 2019 Society of Chemical Industry.

Weed Detection in Perennial Ryegrass With Deep Learning Convolutional Neural Network.

Precision herbicide application can substantially reduce herbicide input and weed control cost in turfgrass management systems. Intelligent spot-spraying system predominantly relies on machine vision-based detectors for autonomous weed control. In this work, several deep convolutional neural networks (DCNN) were constructed for detection of dandelion (Taraxacum officinale Web.), ground ivy (Glechoma hederacea L.), and spotted spurge (Euphorbia maculata L.) growing in perennial ryegrass. When the networks were trained using a dataset containing a total of 15,486 negative (images contained perennial ryegrass with no target weeds) and 17,600 positive images (images contained target weeds), VGGNet achieved high F1 scores (≥0.9278), with high recall values (≥0.9952) for detection of E. maculata, G. hederacea, and T. officinale growing in perennial ryegrass. The F1 scores of AlexNet ranged from 0.8437 to 0.9418 and were generally lower than VGGNet at detecting E. maculata, G. hederacea, and T. officinale. GoogleNet is not an effective DCNN at detecting these weed species mainly due to the low precision values. DetectNet is an effective DCNN and achieved high F1 scores (≥0.9843) in the testing datasets for detection of T. officinale growing in perennial ryegrass. Moreover, VGGNet had the highest Matthews correlation coefficient (MCC) values, while GoogleNet had the lowest MCC values. Overall, the approach of training DCNN, particularly VGGNet and DetectNet, presents a clear path toward developing a machine vision-based decision system in smart sprayers for precision weed control in perennial ryegrass.

Detection of broadleaf weeds growing in turfgrass with convolutional neural networks.

BACKGROUND: Weed infestations reduce turfgrass aesthetics and uniformity. Postemergence (POST) herbicides are applied uniformly on turfgrass, hence areas without weeds are also sprayed. Deep learning, particularly the architecture of convolutional neural network (CNN), is a state-of-art approach to recognition of images and objects. In this paper, we report deep learning CNN (DL-CNN) models that are remarkably accurate at detection of broadleaf weeds in turfgrasses. RESULTS: VGGNet was the best model for detection of various broadleaf weeds growing in dormant bermudagrass [Cynodon dactylon (L.)] and DetectNet was the best model for detection of cutleaf evening-primrose (Oenothera laciniata Hill) in bahiagrass (Paspalum notatum Flugge) when the learning rate policy was exponential decay. These models achieved high F1 scores (>0.99) and overall accuracy (>0.99), with recall values of 1.00 in the testing datasets. CONCLUSION: The results of the present research demonstrate the potential for detection of broadleaf weed using DL-CNN models for detection of broadleaf weeds in turfgrass systems. Further research is required to evaluate weed control in field conditions using these models for in situ video input in conjunction with a smart sprayer. © 2019 Society of Chemical Industry.