Materials and methods

Field trials were conducted at the University of Florida Gulf Coast Research and Education Center in Wimauma (lat. 27.76°N, long. 82.23°W) during the 2016–17, 2017–18, and 2018–19 strawberry production seasons. The soil type at the experimental site is a Myakka fine sand (Siliceous Hypothermic Oxyaquic Alorthod) slightly acidic (pH 6.5–6.8), and low in organic matter (0.9% to 1.2%).

Field preparation included disc cultivation and leveling, followed by bed formation using a bed press. Bed dimensions were 4-ft center-to-center spacing, 28 inches wide at the top, and 12 inches tall. The plot length was 50-ft linear bed length. All plots were treated with chloropicrin 35% + 1,3-dichloropropene 65% as a primary bed fumigant at 300 lb/acre using a two-shank fumigation rig at 8-inch depth in the bed. Primary bed fumigant was the fumigant used at the time of bed formation. Chloropicrin 35% + 1,3-dichloropropene 65% was used to suppress the soilborne pathogen and nematodes with the limited control of purple nutsedge (Minuto et al., 2006). The bed formation and primary fumigation occurred on 10, 20, and 17 Sept. for the 2016–17, 2017–18, and 2018–19 production seasons, respectively.

The field was subdivided into four blocks based on the anticipated soil moisture level during the production season. Treatments were then arbitrarily assigned to each block. In 2016–17, 60 gal/acre metam-K [potassium N-methyldithiocarbamate 54% (K-PAM HL; AMVAC Chemical Corp., Newport Beach, CA)] was applied 1) using a minicoulter at the time of bed formation, 2) through the drip tape within 24 h after bed formation and plastic application, and 3) through the drip tape 2 weeks after bed formation (WAB). In the 2017–18 and 2018–19 production season, two additional herbicide treatments were added including 4) 4 fl oz/acre sulfentrazone [sulfentrazone 40% (Spartan 4F; FMC Corp., Philadelphia, PA)] and 5) 32 fl oz/acre EPTC [S-ethyl dipropylthiocarbamate 88% (Eptam 7E; Gowan Co., Yuma, AZ) over the bed top using carbon dioxide pressurized backpack sprayer immediately before laying plastic. The backpack sprayer was calibrated to deliver 0.78 L·min⁻¹ and was equipped with a single flat-fan nozzle (8002VS; Teejet Spraying Systems Co., Glendale Heights, IL) at 35 psi. Beds were covered with virtually impermeable film (Blockade; Berry Plastics, Evansville, IN) with a single drip tape (Chapin drip tape-BTF; Jain Irrigation, Haines City, FL) buried 1 inch deep at the center of the bed. All experiments included a control that was only fumigated with chloropicrin 35% + 1,3-dichloropropene 65% at the time of bed formation that was expected to kill soilborne pathogens and nematodes but not the purple nutsedge. Soil moisture content at the time of treatment application ranged from 15% to 20% water by volume.

‘Florida Radiance’ strawberry transplants were planted in the middle 15-ft area of 50-ft-long plot. The plants were placed in two rows per bed with 14 inches between rows and plant-to-plant spacing of 16 inches at 3 to 4 WAB. Planting occurred on 5, 18, and 15 Oct. for the 2016–17, 2017–18, 2018–19 strawberry seasons, respectively. Purple nutsedge shoots that penetrated the plastic mulch were counted in the transplanted area 5, 10, and 20 WAB using a 0.3 × 0.3-m quadrat. Marketable fruit was harvested two times per week starting from late December to mid-February.

The PROC GLIMMIX procedure was used to determine the effect of treatments on purple nutsedge population density and strawberry yield in SAS (version 9.2; SAS Institute, Cary, NC). Block was treated as a random effect and the treatment as a fixed effect. Repeated-measure analysis was used for the purple nutsedge population density and fruit yield, which were collected at multiple times. Purple nutsedge population density data were transformed using the lognormal distribution function in the model statement; however, means presented in Table 1 are the back-transformed means. Tukey’s procedure was used to separate means whenever appropriate.

Table 1.

Effect of different supplemental metam potassium (metam-K) fumigation application techniques and herbicides in purple nutsedge population density in the strawberry field trials conducted at Wimauma, FL, in 2016–17, 2017–18, and 2018–19 production seasons.

View Table

Results and discussion

The treatment by date interaction for purple nutsedge was not significant in 2015–16 (P = 0.99) and 2016–17 (P = 0.56) but was significant in 2017–18 (P = 0.01). Therefore, purple nutsedge population densities are presented separately for each weed assessment date for all years. Supplemental fumigants and herbicides did not affect (P > 0.05) purple nutsedge population density in any of the three production seasons compared with the control except at the final weed assessment at 20 WAB (P = 0.005) in the 2018–19 production season. Supplemental herbicide applications had the lowest purple nutsedge population density (<80/m²) at 20 WAB in 2017–18, but the difference was not significant (Table 1).

In 2018–19, herbicides EPTC and sulfentrazone consistently had a significantly lower number of purple nutsedge than other treatments at 20 WAB (Table 1). This indicates that preemergence herbicides in conjunction with fumigants may be a more effective option than the Florida 3-way. Previous research found similar trends. For example, incorporation of herbicides such as pebulate, S-metolachlor, and napropamide in addition to chloropicrin 17% + 1,3-dichloropropene 83% (Telone C-17, Corteva Agriscience) showed significant purple nutsedge control in a field trial (Gilreath et al., 2004) compared with chloropicrin 17% + 1,3-dichloropropene 83% alone. In our study, supplemental metam-K was beneficial in only one of three production seasons. We are uncertain why seasons varied because all three seasons were comparatively similar in terms of soil moisture, rainfall, and temperature. The results are similar to those of Boyd et al. (2017) where supplemental metam-K did not provide additional benefit in terms of purple nutsedge control and tomato yield when 1,3-dichloropropene + chloropicrin was used as the primary bed fumigant. Furthermore, sequential application of metam fumigant and chloropicrin 35% + 1,3-dichloropropene 65% did not result in a synergistic effect for weed control or higher berry yield in California strawberry fields (Fennimore et al., 2003; Trout and Ajwa, 1999).

Lack of efficacy of our supplemental treatments observed in these experiments may be due to the prolific reproductive capacity of purple nutsedge and high nutsedge density (>100/m²) in the experimental sites. Additionally, one tuber can produce up to 36 plants and 332 tubers over 16 weeks and is capable of forming 6-m-diameter patch containing 19,000 plants and 7000 tubers within 1 year (Anderson 1999). In this context, nontreated regions in the soil bed could result in inconsistent control and high purple nutsedge population densities over time.

Metam application timing had no effect on control levels achieved. It was hypothesized that delaying injections for 1 week would give the tubers the opportunity to sprout. Reed at al. (2016) reported that herbicides such as EPTC and fomesafen are more effective against purple nutsedge when they are applied at the tuber sprout stage. There was no significant difference between supplemental metam-K at the time of bed formation and after 2 weeks of bed formation in their effect on purple nutsedge control; however, tubers were not extracted to observe if they had sprouted at the time of application.

Effectiveness of treatments on purple nutsedge control in the experiment were based on the aboveground number of purple nutsedge shoots piercing through the plastic mulch; however, these numbers are affected by not only fumigation treatments but also the natural underground nutsedge tuber population existing in the experimental plot. The underground tuber population was not determined, which may be different across the experimental plots. This is one limitation when weed control is determined based on natural weed population, which was addressed by selecting the experimental sites with a history of a homogenous weed population.

None of the treatments affected (P > 0.05) total strawberry yield in any of the three production seasons. There was no apparent adverse effect on strawberry growth and no visual strawberry injury (data not shown).

The supplemental applications of metam-K at 390 kg·ha⁻¹ after chloropicrin 35% + 1,3-dichloropropene 65% did not improve purple nutsedge control compared with chloropicrin 35% + 1,3-dichloropropene 65% alone. Herbicides were also ineffective in 2 of 3 years with rates of control for sulfentrazone and EPTC at 51% and 80%, respectively, in 2018 and 2019. We conclude that the Florida 3-way and modifications of this management system did not effectively control purple nutsedge or improve strawberry yield.

Units

View Table