Preemergence herbicide effects on St. Augustinegrass

Field research was conducted at the Mississippi State University Turfgrass Research Center on the R.R. Foil Plant Science Research Farm in Starkville, Miss., from June to November 2019 and from May to November 2020. Ten plugs (5 inches by 5 inches, 25 square inches apiece) were planted in each plot. Photo by Amy Wilber

St. Augustinegrass [Stenotaphrum secundatum (Walt.) Kuntze] is a widely grown lawn grass in the southeastern and Gulf-coastal United States. Unlike other warm-season turfgrasses that spread through rhizomes, St. Augustinegrass relies solely on stoloniferous growth. While the total acreage of St. Augustinegrass is unknown, it is the most used lawn grass in Florida (11) and is widely grown in Texas (10). Typically, St. Augustinegrass is established from plugs or sod, and weed competition can slow this process. Weed-infested areas are undesirable and require postemergence weed control, often achieved with herbicides. Although preemergence herbicides can prevent annual weed growth, they might hinder turfgrass establishment.

During St. Augustinegrass establishment, common grassy weeds like goosegrass [Eleusine indica (L.) Gaertn], crabgrass (Digitaria spp.) and annual bluegrass (Poa annua L.) are often present. Preemergence herbicides such as atrazine (AAtrex, Syngenta), oxadiazon (Ronstar, Envu), S-metolachlor (Pennant Magnum, Syngenta), prodiamine (Barricade, Syngenta), pendimethalin (Pendulum, BASF), and dithiopyr (Dimension, Corteva Agriscience) are commonly used for controlling these weeds in St. Augustinegrass sod production (4, 8). However, not all of these herbicides are approved for use on St. Augustinegrass during establishment. These herbicides each have a unique spectrum of weeds they control, but there is limited research on their effects on St. Augustinegrass sod production. Moreover, some studies do not align with standard production practices and label guidelines.

The traditional way to assess turfgrass research plots involves a trained evaluator making subjective judgments about factors like phytotoxicity and percentage cover. Multispectral imaging measures the reflectance from different wavelengths in the electromagnetic spectrum. Certain wavelengths of reflectance are linked to turfgrass growth, coverage and density (2, 7). Indices like the normalized difference vegetation index (NDVI) have been found to align with visual quality ratings (7). Red reflectance is influenced by chlorophyll absorption and content, while near-infrared (NIR) reflectance is affected by light scattering within leaf cells (3). Red reflectance is closely linked to visual assessments of turfgrass quality, and NIR reflectance indicates plant water status (2, 5). Reflectance from the red-edge region is used to estimate canopy chlorophyll, leaf area index and nitrogen content (3). Multispectral data can identify differences in turfgrass quality in agreement with visual observations (7). However, studies have shown that while visual quality correlated with NDVI, the values are often indistinguishable statistically (6). This lack of differentiation might be due to limitations in detection and errors introduced by remote methods.

There is limited research on the impact of preemergence herbicides on establishing St. Augustinegrass using plugs, as previous studies used sprigs or ribbons. Establishing from ribbons left in the field involves regrowth from plants with intact root systems, while plugging uses small pieces of turfgrass with minimal roots. Preemergence herbicides can affect root growth, potentially leading to variations between establishing with shallow root systems versus deep, intact roots. Research was conducted to assess the effects of contemporary preemergence herbicides on plugged St. Augustinegrass establishment, employing both multispectral and traditional subjective evaluation methods.

Table 1. Preemergence herbicide treatments applied to MSA 2-3-98 St. Augustinegrass plugs to determine effects on establishment.

Materials and methods

Field research was conducted at the Mississippi State University Turfgrass Research Center on the R.R. Foil Plant Science Research Farm in Starkville, Miss., from June to November 2019 and from May to November 2020. Research was conducted as a randomized complete block with four replications repeated in 2019 and 2020. Soil was a Marietta fine sandy loam with pH 5.5 (2019) and 6.5 (2020).

Sites were prepared by eliminating vegetation with glyphosate and then cultivated. Plots (5 feet by 5 feet/1.5 meters by 1.5 meters; 25 square feet/2.25 square meters) were plugged with MSA 2-3-98 St. Augustinegrass on June 12, 2019, and May 14, 2020. Ten plugs (5 inches by 5 inches/13 centimeters by 13 centimeters; 25 square inches/161 square centimeters apiece) were planted in each plot (Image 1). One day after planting, preemergence herbicide treatments (Table 1) were applied with a backpack sprayer calibrated to 40 gallons per acre (374 liters per hectare), and granular products were applied evenly with a shaker jar. Plots were irrigated with 0.2 inches (0.5 centimeters) of water four hours after treatment and received irrigation as needed during establishment. Plots were rolled one week after initiation and were maintained at a 4-inch (10-centimeter) height of cut using a rotary mower during periods of active growth. Lime and fertilizer were applied to the study areas according to soil test recommendations. In select instances, weeds were hand-pulled to limit herbicide application and assessment bias.

Aerial image of the 2019 field site in late October 2019. Indaziflam plots are easy to distinguish as there is no growth. Photo by Addison Meeks and Amy Wilber

Cover assessment

Treatment effects on St. Augustinegrass establishment were assessed weekly during active growth for the duration of each site-year. For the 2019-planted site, data were collected from June until dormancy in November. Evaluations resumed from May to October 2020, when all plots reached full cover. Data were collected on the 2020-planted site from May until dormancy in November. Cover was visually estimated on a 0-to-100% scale (0 = no turfgrass cover, 100 = complete turfgrass cover). A Matrice 100 quadcopter (DJI Ltd., Shenzhen, China) equipped with a RedEdge-MX sensor (MicaSense Inc., Seattle) collected reflectance data in narrow bands centered at 475 nanometers (blue), 560 nm (green), 668 nm (red), 717 nm (red edge) and 840 nm (near infrared). Images collected using the aerial sensor were mosaiced in Pix4D Mapper (Pix4D S.A., Prilly, Switzerland) (Image 2), and plot-level NDVI values were extracted from mosaics in ArcGIS Pro.

Statistical analysis

Data were regressed in GraphPad Prism 9.3.1 (San Diego) using a four-parameter variable slope curve like the methods of Begitschke et al. (1) and Schiavon et al. (9) to determine days to reach 50% of the maximum non-treated response. Predicted 95% confidence intervals for time to reach 50% maximum response were used to determine differences between treatments. Treatments with nonoverlapping confidence intervals were considered significantly different.

Table 2. Regression estimates of 95% confidence intervals (CI) for the number of days to reach 50% visual St. Augustinegrass cover following the application of preemergence herbicides. Treatments with non-overlapping CI are considered significantly different.

Results

Percentage St. Augustinegrass cover and NDVI estimates to reach 50% of the maximum recorded response differed due to herbicide treatment and year. The variation in results is likely due to earlier initiation in Year 2, decreased disease pressure and increased irrigation efficiency in 2020. Results also may have differed because a relatively low pH in 2019 was corrected mid-season and prior to 2020 study initiation.

In 2019, all treatments except liquid-applied oxadiazon and atrazine increased days to reach 50% cover compared to the non-treated control (Table 2). Prodiamine, S-metolachlor, dithiopyr and indaziflam (Specticle FLO, Envu) treatments increased days to reach 50% cover in 2019 and 2020 (Table 2). Liquid-applied oxadiazon and atrazine were the only treatments that did not increase days to reach 50% cover in either year.

Dithiopyr, pendimethalin, prodiamine and indaziflam increased days to reach 50% maximum NDVI relative to the non-treated in 2019 (Table 3). S-metolachlor and indaziflam were the only treatments to increase days to reach 50% maximum NDVI in 2020 (Table 3). Granular and liquid-applied oxadiazon, atrazine and atrazine + S-metolachlor treatments did not increase days to reach 50% maximum NDVI in 2019 or 2020.

Discussion

Prodiamine delayed the time for St. Augustinegrass to reach 50% cover and 50% of the maximum NDVI in 2019. These findings are supported by other studies (4) regarding hindered growth after prodiamine application. Pendimethalin also prolonged the time to reach 50% cover and maximum NDVI in 2019, aligning with previous findings of hindered stolon growth due to pendimethalin application (8). Dithiopyr delayed time to 50% cover in both 2019 and 2020 and maximum NDVI in 2019, consistent with previous studies reporting hindered stolon growth after application (8).

Oxadiazon is a common active ingredient used in St. Augustinegrass sod production for preemergence control of annual weeds (4). Granular oxadiazon delayed time to 50% cover in 2019. Liquid oxadiazon did not affect cover or NDVI. This research suggests that liquid-applied oxadiazon is safe to use for weed control although it is not labeled for application to non-dormant, non-established St. Augustinegrass.

Atrazine did not significantly delay time to 50% cover or NDVI, supporting its use for weed control in St. Augustinegrass. Atrazine is a widely used preemergence herbicide in turfgrass production because of its known safety for desired turfgrasses. Due to its repeated use over time, resistance to atrazine has evolved in annual bluegrass (Poa annua), often requiring a tank mix with herbicides of other modes of action, such as S-metolachlor, for preemergence control of annual grassy and broadleaf weeds. Atrazine + S-metolachlor cover results were not consistent across years, delaying growth in one year. S-metolachlor alone delayed time to 50% cover in both years, indicating that it may be safer when tank-mixed with atrazine.

Table 3. Regression estimates of 95% confidence intervals (CI) for the number of days to reach 50% maximum Normalized Difference Vegetation Index collected from an aerial sensor following the application of preemergence herbicides. Treatments with non-overlapping CI are considered significantly different.

Indaziflam was included as a treated check, as it is not labeled for use during establishment in St. Augustinegrass sod production. It is labeled for use in bermudagrass and zoysiagrass production when coverage reaches at least 80% and up to four months before harvest. Indaziflam was injurious in both years, indicating long-term effects on establishment.

In summary, the study revealed that prodiamine, pendimethalin and dithiopyr negatively impacted St. Augustinegrass establishment, even when used at recommended rates. Atrazine showed safety for desired turfgrasses, while S-metolachlor's effects were inconsistent.

Oxadiazon safety was dependent on application method, as liquid-applied oxadiazon was safe, and granular-applied had mixed results. Indaziflam, not recommended for St. Augustinegrass, exhibited long-term inhibitory effects on growth. These findings provide insight for turf managers regarding preemergence herbicide choices during St. Augustinegrass establishment.

NDVI detected similar trends as visual cover but failed to directly estimate time to reach 50% cover. NDVI overestimated the time to reach 50% visual cover in 2019 and underestimated time to reach 50% visual cover in 2020. As spectral reflectance is affected by plant water status and nutrient content (2, 3, 5), turfgrass experiencing drought or nutrient deficiency may impact reflectance values without necessarily affecting ground cover.

Withstanding economic decisions regarding price and availability of weed control options, preemergence herbicide applications should be selected based upon two principal agronomic criteria: 1) range of weeds controlled, and 2) effects upon turfgrass growth. Weed control was not evaluated in the present study — turf was maintained weed-free to eliminate bias of establishment. Although some herbicide treatments increased time to reach 50% cover or time to 50% maximum NDVI, time to 100% cover ultimately may be the true concern for turfgrass managers — not measured in this study. Confidence intervals for time to reach 50% cover or time to 50% maximum NDVI were relatively narrow, so while some treatments were significantly different, those differences may be of little consequence given the benefits of weeds controlled. Future research may explore the cost-benefit analysis of applying less-injurious herbicides that might have a narrower spectrum of control compared to applying a more-injurious herbicide with a broader spectrum of control that may not require the use of postemergent herbicides for weed-free turfgrass. Results suggest that turfgrass managers can make spray applications of atrazine and oxadiazon without affecting St. Augustinegrass establishment, though this was not tested across a broad range of soils and sites. Regarding oxadiazon, there is a large cost savings when applied as a liquid-broadcast versus a granular product. Pendimethalin, granular oxadiazon and atrazine + S-metolachlor, all historically common treatments in St. Augustinegrass sod production, may also delay establishment. This and other factors (e.g., soil pH, soil type, planting time) should be tested in future research.

The research says

• The study revealed that prodiamine, pendimethalin and dithiopyr negatively impacted St. Augustinegrass establishment, even when used at recommended rates.
• Results suggest that turfgrass managers can make spray applications of atrazine and oxadiazon without affecting St. Augustinegrass establishment.
• Regarding oxadiazon, there is a large cost savings when applied as a liquid-broadcast versus a granular product.
• Withstanding economic decisions regarding price and availability of weed control options, preemergence herbicide applications should be selected based upon two principal agronomic criteria: 1) range of weeds controlled, and 2) effects upon turfgrass growth.

Acknowledgements

The authors thank the Geosystems Research Institute staff for the collection of aerial imagery and the turfgrass research technicians at the Mississippi State University Turfgrass Research Center for maintenance of the trial sites. The full manuscript of this experiment is open access, available in Agronomy Journal (https://doi.org/10.1002/agj2.21304).

Literature cited

1. Begitschke, E.G., J.D. McCurdy, T. Tseng, T.C. Barickman, B.R. Stewart, C.M. Baldwin, M.P. Richard and J.K. Ward. 2018. Preemergence herbicide effects on establishment and tensile strength of sprigged hybrid bermudagrass. Agronomy Journal 110(6):2243-2249 (https://doi.org/10.2134/agronj2017.12.0720).
2. Bremer, D.J., H. Lee, K. Su and S.J. Keeley. 2011. Relationships between normalized difference vegetation index and visual quality in cool-season turfgrass: II. Factors affecting NDVI and its component reflectances. Crop Science 51(5):2219-2227 (https://doi.org/10.2135/cropsci2010.12.0729).
3. Clevers, J.G.P.W., and A.A. Gitelson. 2013. Remote estimation of crop and grass chlorophyll and nitrogen content using red-edge bands on Sentinel-2 and -3. International Journal of Applied Earth Observation and Geoinformation 23:344-351 (https://doi.org/10.1016/j.jag.2012.10.008).
4. Grichar, W J., and R.D. Havlak. 2009. Evaluating herbicidal injury to St. Augustinegrass in sod production. Texas Journal of Agriculture and Natural Resources 22:61-68.
5. Guyot, G. 1990. Optical properties of vegetation canopies. Pages 19-43. In: M.D. Steven and J.A. Clark, eds. Applications of remote sensing in agriculture. Elsevier Science, Amsterdam.
6. Lee, H., D. Bremer and K. Su. 2009. Comparing estimates of turfgrass quality obtained using normalized difference vegetation index, digital imagery and the human eye. K-State Turfgrass Research Report of Progress. 1015.
7. Leinauer, B., D.M. VanLeeuwen, M. Serena, M. Schiavon and E. Sevostianova. 2014. Digital image analysis and spectral reflectance to determine turfgrass quality. Agronomy Journal 106(5):1787-1794 (https://doi.org/10.2134/agronj14.0088).
8. McCarty, L.B., D.W. Porter and D.L. Colvin. 1995. Sod regrowth of St. Augustinegrass after preemergence herbicide application. Agronomy Journal 87(3):503-507 (https://doi.org/10.2134/agronj1995.00021962008700030017x).
9. Schiavon, M., M. Serena, B. Leinauer, R. Sallenave and J.H. Baird. 2015. Seeding date and irrigation system effects on establishment of warm-season turfgrasses. Agronomy Journal 107(3):880-886 (https://doi.org/10.2134/agronj14.0322).
10. Segars, C.A., and B. Bowling. 2020. St. Augustinegrass lawn management (Fact Sheet EHT-144). Texas A&M AgriLife Extension.
11. Trenholm, L.E., J.B. Unruh and T.W. Shaddox. 2017. St. Augustinegrass for Florida lawns (EDIS publication ENH5). Department of Environmental Horticulture, UF/IFAS Extension.

Amy Wilber is an Extension Associate I; Jay McCurdy, Ph.D., is an associate professor; Joby Czarnecki, Ph.D., is an associate research professor; and Barry Stewart, Ph.D., is a professor, all in the Department of Plant and Soil Sciences at Mississippi State University, and Hongxu Dong, Ph.D., was an assistant professor until August 2022 at Mississippi State University.

You may also like: How much nitrogen fertilization does seeded bermudagrass require?