Weed Control and Weather Predicted to Impact Future Soybean Yields
We all know that U.S. farmers produce a major portion of the global soybean supply each year. What we don’t know is just how long this can continue given the current state of increasing weed resistance to herbicides, and the predicted “climate change” that likely will involve an increase in global temperature and erratic rainfall patterns.
In an article titled “Deterioating weed control and variable weather portends greater soybean yield losses in the future” by Landau, Hager, and Williams [Science of the Total Environment 830 (2022) 154764], the authors use machine learning techniques to estimate the effects of these two occurrences on soybean yield loss in the midwestern U.S. They used a database of >1500 herbicide evaluation trials that spanned 28 years (1992-2019) and 106 weather environments to model the most important relationships among weed control, weather variability, and crop management to develop their yield loss estimates. Each trial consisted of herbicide, spray adjuvant, and weedy control treatments, and data collected from each trial often included percent control of dominant weed species and percent soybean injury that resulted from herbicide treatments. Pertinent points from the article follow.
• The objective of the analysis reported in the article was to identify the major linkages among weather variability, weed control, and crop management as related to soybean yield. The authors hypothesized that soybean yield loss due to weeds results primarily from incomplete weed control paired with drought and/or heat stress during soybean reproductive development.
• In the introduction of the article, the authors present several statements of known facts. Major ones are:
►Even under the best of growing conditions, soybean production is adversely affected by interference from uncontrolled weeds.
►Weeds can reduce soybean yield by competing for resources such as light, water, and nutrients during the growing season, and by adversely affecting harvest and quality of harvested seed.
►Weather conditions at the time of herbicide application can affect herbicide efficacy against targeted weeds. For example, with PRE herbicides, inadequate rainfall following their application can result in unsuccessful weed control, and excessive rainfall following their application can leach certain PRE herbicides away from the weed seedling emergence zone.
►Warmer temperatures will increase the growth rate of numerous weed species, which will result in a reduced window of opportunity for effective weed control by POST herbicides that are usually applied based on weed size.
►The increasing prevalence of herbicide resistance (HR) in an increasing number of weed species is reducing the effectiveness of chemical weed control in many cropping systems.
• Classification and regression tree (CART) analysis was used to model the relationship among weed control, weather, and soybean management practices on both soybean yield and soybean yield loss due to weeds.
• Random forest analysis was used to determine the importance of each independent variable for predicting soybean yield and yield loss.
• According to the CART model, the greatest soybean yield losses due to weeds occurred when late-season average control of all weed species was <51%, and the least yield loss occurred when late-season average control of all weed species was ≥94%.
• According to the random forest analysis, average late-season weed control and season-long control of all weed species were identified as the two most important weed control variables for predicting soybean yield loss due to weeds.
• Similar to soybean yield loss to weeds, average late-season control of all weed species was determined to be the most important variable for predicting soybean yield.
• The two most important weather variables for predicting soybean yield were vapor pressure deficit (VPD) during early vegetative growth of soybean and total rainfall during seedfill.
• Both models determined that late-season control of all weed species was the single most important determinant of soybean yield loss to weeds. Treatments that resulted in poor late-season control (<51%) of all weed species had average yield loss of 48%.
• Later-maturing soybean varieties alleviated some of the yield losses caused by poor late-season weed control in these tests that were conducted mostly at Urbana, Ill., and this was attributed to their increased competitiveness for light. The authors suggest these results lend credence to using later soybean maturity groups as part of a weed control strategy that may reduce the risk associated with poor weed control as climate becomes more variable and/or erratic.
• When the highest level of weed control was not accomplished, treatments with lower rainfall and hotter temperatures during seedfill had greater yield loss. This was attributed to several weed species in the U.S. having increased tolerance to these stresses compared to soybean.
• Since these results indicated that soybean yield loss due to weeds is driven by incomplete weed control coupled with heat and drought stress, the authors concluded that the combination of these stresses coupled with more competitive weeds under increasingly stressful conditions that will be associated with predicted climate change was likely the cause of the increased yield loss when weeds were not controlled.
Takehome Message. Results from this analysis suggest the following should be considered in the face of predicted climate change. 1) Soybean management strategies must continually be improved in order to lower the risk of yield loss from predicted more variable and erratic weather. 2) Uncontrolled weeds in soybean fields are likely to be more competitive when drought occurs, and will thus pose a greater threat to soybean yield. 3) Much of the U.S. soybean growing region is likely to have increasingly warmer temperatures and more variable rainfall patterns (longer periods of drought, more intense rainfall events), and these weather changes will likely increase the growth rate and competitiveness of many of the current problematic weeds that plague soybeans. 4) As predicted unfavorable weather changes occur, complete weed control throughout the soybean growing season, and especially in the late-season, will become more essential. 5) In the face of changing climate that is predicted to be more adverse for U.S. crop production, drought and heat stress will exacerbate soybean yield loss, and this will be especially prevalent if uncontrolled weeds are present. 6) Weed control systems for soybeans will have to employ all available measures that will include cultural, mechanical, biological, and chemical methods in order to avoid soybean yield losses that will be associated with competition from uncontrolled weeds. 7) The increasing occurrence of HR weeds likely will require the development and adoption of new efficacious weed management strategies that are not yet being researched or are not yet considered viable alternatives to chemical weed control.
Composed by Larry G. Heatherly, Apr. 2022, larryh91746@gmail.com