Frogeye leaf spot (FLS), caused by
, is an economically important disease of soybean in the United States. Data from 66 uniform fungicide trials (UFTs) conducted from 2012 to 2021 across eight states ...(Alabama, Arkansas, Illinois, Iowa, Kentucky, Louisiana, Mississippi, and Tennessee) were gathered and analyzed to determine the efficacy and profitability of the following fungicides applied at the beginning pod developmental stage (R3): azoxystrobin + difenoconazole (AZOX + DIFE), difenoconazole + pydiflumetofen (DIFE + PYDI), pyraclostrobin (PYRA), pyraclostrobin + fluxapyroxad + propiconazole (PYRA + FLUX + PROP), tetraconazole (TTRA), thiophanate-methyl (TMET), thiophanate-methyl + tebuconazole (TMET + TEBU), and trifloxystrobin + prothioconazole (TFLX + PROT). A network meta-analytic model was fitted to the log of the means of FLS severity data and to the nontransformed mean yield for each treatment, including the nontreated. The percent reduction in disease severity (%) and the yield response (kg/ha) relative to the nontreated was the lowest for PYRA (11%; 136 kg/ha) and the greatest for DIFE + PYDI (57%; 441 kg/ha). A significant decline in efficacy over time was detected for PYRA (18 percentage points p.p.), TTRA (27 p.p.), AZOX + DIFE (18 p.p.), and TMET + TEBU (19 p.p.) by using year as a continuous covariate in the model. Finally, probabilities of breaking even were the greatest (>65%) for the most effective fungicide DIFE + PYDI and the lowest (<55%) for PYRA. Results of this meta-analysis may be useful to support decisions when planning fungicide programs.
Soybean yield response variability to foliar fungicide applications was evaluated in on-farm replicated strip trials (OFTs) and small-plot trials (SPTs) from 2008 through 2015 in Iowa. A total of 230 ...OFTs and 49 SPTs were compared for yield response to pyraclostrobin, pyraclostrobin + fluxapyroxad, or trifloxystrobin + prothioconazole fungicides. OFTs (18 to 55 m wide and 200 to 800 m long strips) were harvested with farmers' combines equipped with yield monitors and GPS, while SPTs (3.0 to 4.6 m wide and 10.7 to 15.3 m long plots) were harvested by small research plot combines. Variance component and power analyses were conducted with a subset of data consisting of 12 OFTs and SPTs, each with pyraclostrobin and evaluated in 2008 and 2009. While average yield responses were similar, the residual random yield variation was smaller in OFTs than SPTs. Power analysis showed that SPTs need more replications than OFTs to detect the same overall treatment differences. To detect a yield response of 134 kg/ha, it would require at least three treatment replications with 12 locations in OFTs and seven replications with 12 locations in SPTs. Researchers need to acknowledge the differences in statistical power of detecting yield responses to foliar fungicide on soybean in different types of field experiments, especially with smaller plot sizes in situations with less foliar disease.
As complete host resistance in soybean has not been achieved, Sclerotinia stem rot (SSR) caused by
continues to be of major economic concern for farmers. Thus, chemical control remains a prevalent ...disease management strategy. Pesticide evaluations were conducted in Illinois, Iowa, Michigan, Minnesota, New Jersey, and Wisconsin from 2009 to 2016, for a total of 25 site-years (
= 2,057 plot-level data points). These studies were used in network meta-analyses to evaluate the impact of 10 popular pesticide active ingredients, and seven common application timings on SSR control and yield benefit, compared with not treating with a pesticide. Boscalid and picoxystrobin frequently offered the best reductions in disease severity and best yield benefit (
< 0.0001). Pesticide applications (one- or two-spray programs) made during the bloom period provided significant reductions in disease severity index (DIX) (
< 0.0001) and led to significant yield benefits (
= 0.0009). Data from these studies were also used in nonlinear regression analyses to determine the effect of DIX on soybean yield. A three-parameter logistic model was found to best describe soybean yield loss (pseudo-
= 0.309). In modern soybean cultivars, yield loss due to SSR does not occur until 20 to 25% DIX, and considerable yield loss (-697 kg ha
or -10 bu acre
) is observed at 68% DIX. Further analyses identified several pesticides and programs that resulted in greater than 60% probability for return on investment under high disease levels.
Three soybean field trials were conducted in Indiana to evaluate the integration of seed treatment, cultivar selection, and seeding rate on sudden death syndrome (SDS) root rot, pathogen load in the ...root, foliar symptoms, yield, and net return. Two soybean cultivars, one moderately resistant and one susceptible to SDS, were planted at three seeding rates (272,277 seeds/ha, 346,535 seeds/ha, and 420,792 seeds/ha). Fluopyram and pydiflumetofen seed treatments were applied to both cultivars, and the cultivars were then compared with a control. Low foliar SDS disease pressure was observed in our study. Seed treatment with either fluopyram or pydiflumetofen and the use of a moderately resistant cultivar decreased
DNA concentration in the root relative to the control and the use of a susceptible cultivar. Fluopyram significantly reduced visual root rot severity by 8.8% and increased yield by 105 kg/ha relative to the control but was not different from pydiflumetofen. However, pydiflumetofen performed the same as the control with respect to root rot severity and yield. Findings from this study support the use of a seed treatment to protect roots from infection and the use of a moderately resistant cultivar planted at a seeding rate of 346,535 seeds/ha to protect yield and maximize net returns when a field has low foliar SDS pressure.
One hundred fifty-two Diaporthe isolates were recovered from symptomatic soybean (Glycine max) stems sampled from the U.S. states of Iowa, Indiana, Kentucky, Michigan, and South Dakota. Using ...morphology and DNA sequencing, isolates were identified as D. aspalathi (8.6%), D. caulivora (24.3%), and D. longicolla (67.1%). Aggressiveness of five isolates each of the three pathogens was studied on cultivars Hawkeye (D. caulivora and D. longicolla) and Bragg (D. aspalathi) using toothpick, stem-wound, mycelium contact, and spore injection inoculation methods in the greenhouse. For D. aspalathi, methods significantly affected disease severity (P < 0.001) and pathogen recovery (P < 0.001). The relative treatment effects (RTE) of stem-wound and toothpick methods were significantly greater than for the other methods. For D. caulivora and D. longicolla, a significant isolate × method interaction affected disease severity (P < 0.05) and pathogen recovery (P < 0.001). Significant differences in RTEs were observed among D. caulivora and D. longicolla isolates only when the stem-wound and toothpick methods were used. Our study has determined that the stem-wound and toothpick methods are reliable to evaluate the three pathogens; however, the significant isolate × method interactions for D. caulivora and D. longicolla indicate that multiple isolates should also be considered for future pathogenicity studies.
A three-year study was conducted in Illinois, Indiana, Iowa, Michigan, and Ontario, Canada, from 2013 through 2015 to determine the effect of soybean (Glycine max) cultivars' source of soybean cyst ...nematode (SCN; Heterodera glycines) resistance on SCN population densities, sudden death syndrome (SDS; caused by Fusarium virguliforme), and yield of soybean. Five cultivars were evaluated with and without fluopyram seed treatment at each location. Cultivars with no SCN resistance had greater SDS severity, greater postharvest SCN egg counts (Pf), and lower yield than cultivars with plant introduction (PI) 548402 (Peking) and PI 88788-type of SCN resistance (P < 0.05). Cultivars with Peking-type resistance had lower Pf than those with PI 888788-type and no SCN resistance. In two locations with HG type 1.2-, cultivars with Peking-type resistance had greater foliar disease index (FDX) than cultivars with PI 88788-type. Fluopyram seed treatment reduced SDS and improved yield compared with a base seed treatment but did not affect SCN reproduction and Pf (P > 0.05). FDX and Pf were positively correlated in all three years (P < 0.01). Our results indicate that SDS severity may be influenced by SCN population density and HG type, which are important to consider when selecting cultivars for SCN management.
Insect pests cause significant damage to food production, so early detection and efficient mitigation strategies are crucial. There is a continual shift toward machine learning (ML)‐based approaches ...for automating agricultural pest detection. Although supervised learning has achieved remarkable progress in this regard, it is impeded by the need for significant expert involvement in labeling the data used for model training. This makes real‐world applications tedious and oftentimes infeasible. Recently, self‐supervised learning (SSL) approaches have provided a viable alternative to training ML models with minimal annotations. Here, we present an SSL approach to classify 22 insect pests. The framework was assessed on raw and segmented field‐captured images using three different SSL methods, Nearest Neighbor Contrastive Learning of Visual Representations (NNCLR), Bootstrap Your Own Latent, and Barlow Twins. SSL pre‐training was done on ResNet‐18 and ResNet‐50 models using all three SSL methods on the original RGB images and foreground segmented images. The performance of SSL pre‐training methods was evaluated using linear probing of SSL representations and end‐to‐end fine‐tuning approaches. The SSL‐pre‐trained convolutional neural network models were able to perform annotation‐efficient classification. NNCLR was the best performing SSL method for both linear and full model fine‐tuning. With just 5% annotated images, transfer learning with ImageNet initialization obtained 74% accuracy, whereas NNCLR achieved an improved classification accuracy of 79% for end‐to‐end fine‐tuning. Models created using SSL pre‐training consistently performed better, especially under very low annotation, and were robust to object class imbalances. These approaches help overcome annotation bottlenecks and are resource efficient.
Core Ideas
Insect pests cause significant damage to food production.
Early detection and mitigation of insect pests are crucial in managing economic threshold level.
We developed a self‐supervised learning (SSL) model to identify insect pests with minimal annotations.
SSL models greatly improve the identification and classification tasks.
Entropy‐masking‐based segmentation aids SSL effectiveness.
Abstract
Soybean (Glycine max L.) is a major crop grown in the United States but is susceptible to many diseases that cause significant yield losses each year. Consistent threats exist across both ...northern and southern production regions and include the soybean cyst nematode, charcoal rot, and seedling diseases. In contrast, significant soybean diseases like Phytophthora stem and root rot, sudden death syndrome, and Sclerotinia stem rot (white mold) are intermittent threats that can be heavily influenced by environmental factors. Additional threats to soybean production that have emerged in recent years as more common problems in soybean production include root-knot and reniform nematodes, frogeye leaf spot, and Diaporthe diseases. Disease in any crop will only occur when the three components of the disease triangle are present: a susceptible host, a virulent pathogen, and a conducive environment. If an environment is becoming more conducive for a particular disease, it is important that farmers and practitioners are prepared to manage the problem. The information in this review was compiled to help assist agriculturalists in being proactive in managing new soybean diseases that may be emerging in new areas. To do this, we provide: 1) an overview of the impact and disease cycle for major soybean diseases currently causing significant yield losses in the United States, 2) a comprehensive review of the current management strategies for each soybean disease, and 3) insights into the epidemiology of each pathogen, including the likelihood of outbreaks and expansion to additional geographic regions based on current trends in climate change.
Radiation use efficiency (RUE) is difficult to estimate and unreasonable to perform on a small plot scale using traditional techniques. However, the increased availability of Unmanned Aerial Vehicles ...(UAVs) provides the ability to collect spatial and temporal data at high resolution and frequency, which has made a potential workaround. An experiment was completed in Iowa to (i) demonstrate RUE estimation of soybean Glycine max (L.) Merr. from reflectance data derived from consumer-grade UAV imagery and (ii) investigate the impact of foliar fungicides on RUE in Iowa. Some fungicides are promoted to have plant health benefits beyond disease protection, and changes in RUE may capture their effect. Frogeye leaf spot severity did not exceed 2%. RUE values ranged from 0.98 to 1.07 and 0.96 to 1.12 across the entire season and the period post-fungicide application, respectively, and fell within the range of previously published soybean RUE values. Plots treated with fluxapyroxad + pyraclostrobin had more canopy cover (p = 0.078) compared to the non-treated control 133 days after planting (DAP), but yields did not differ. A “greening effect” was detected at the end of the sample collection. RUE estimation using UAV imagery can be considered a viable option for the evaluation of management techniques on a small plot scale. Since it is directly related to yield, RUE could be an appropriate parameter to elucidate the impact of plant diseases and other stresses on yield.
In soybean, Sclerotinia sclerotiorum apothecia are the sources of primary inoculum (ascospores) critical for Sclerotinia stem rot (SSR) development. We recently developed logistic regression models ...to predict the presence of apothecia in irrigated and nonirrigated soybean fields. In 2017, small-plot trials were established to validate two weather-based models (one for irrigated fields and one for nonirrigated fields) to predict SSR development. Additionally, apothecial scouting and disease monitoring were conducted in 60 commercial fields in three states between 2016 and 2017 to evaluate model accuracy across the growing region. Site-specific air temperature, relative humidity, and wind speed data were obtained through the Integrated Pest Information Platform for Extension and Education (iPiPE) and Dark Sky weather networks. Across all locations, iPiPE-driven model predictions during the soybean flowering period (R1 to R4 growth stages) explained end-of-season disease observations with an accuracy of 81.8% using a probability action threshold of 35%. Dark Sky data, incorporating bias corrections for weather variables, explained end-of-season disease observations with 87.9% accuracy (in 2017 commercial locations in Wisconsin) using a 40% probability threshold. Overall, these validations indicate that these two weather-based apothecial models, using either weather data source, provide disease risk predictions that both reduce unnecessary chemical application and accurately advise applications at critical times.