Plant health evaluations of Belonolaimus longicaudatus and Meloidogyne incognita colonized bermudagrass using remote sensing

Microplot trials were conducted at the Plant Science Research Center (PSRC) at Auburn University, AL during the summers of 2018 and 2019. Two trials were conducted: one to evaluate four nematicides for their ability to reduce the population density of B. longicaudatus population density, and the other to reduce the population density of M. incognita. Trials were established in 26.5-liter 35.5 cm diameter by 35.5 cm depth plastic tree pots nested one on top of the other with a brick in between to limit root growth by air pruning and filled with 100% sand. The microplots were submerged 30 cm in the soil with 5.5 cm of the pot above the soil line. Experiments were arranged in a randomized complete block design (RCBD) with five replications for each treatment. In 2018, ‘Tifway’ hybrid bermudagrass was laid as sod in each microplot and given ten weeks for establishment.

At the end of the 10-week period, M. incognita eggs were inoculated on half of the microplots at a rate of 50,000 eggs per pot on a weekly basis for 4 weeks to build up the nematode population density. The remaining microplots received an inoculation rate of 100 B. longicaudatus nematodes per pot on a weekly basis for 4 weeks. After the 4-week inoculation period, a 100-cm³ sample was taken from each plot to confirm M. incognita or B. longicaudatus population density. Inocula of M. incognita race 3 were originally isolated from an infested field at Plant Breeding Unit (PBU) at E.V. Smith Research Center of Auburn University in 2017 (Groover et al., 2019). Nematode eggs were extracted following a modified version of the methodology of Hussey and Barker (1973). The root mass was placed in a 0.625% sodium hypochlorite (NaOCl) solution and shaken for 4 min at 1 g-force on a Barnstead Lab Line Max Q 5000 E Class shaker (Conquer Scientific, San Diego, CA) and the nematode eggs were extracted by centrifugal floatation method (Jenkins, 1964). The contents were centrifuged at 427 g-forces for 1 min in a 1.14 specific gravity sucrose solution based on Jenkins (1964) methodology.

For the sting microplot experiment, B. longicaudatus, originally isolated from a golf course in east Alabama, was used as inoculum. The nematode population was obtained from total soil collected on a 25-µm pore sieve and the modified sucrose centrifugal flotation technique (Jenkins, 1964) was used to extract nematodes.

Four nematicides were evaluated in the microplot setting during the study. Nematicides used were: (i) abamectin (Divanem; Syngenta, Greensboro, NC) at 0.89 L/ha; (ii) fluopyram (Indemnify; Bayer Environmental Science, Cary, NC) at 1.25 L/ha; (iii) fluensulfone (Nimitz Pro G; Control Solutions International, Pasadena, TX) at 134 kg/ha; and (iv) furfural (Multiguard Protect; Agriguard Company, LLC, Cranford, NJ) at 75 L/ha. An untreated treatment was included as the control. Abamectin, fluopyram, fluensulfone, and furfural were applied once at the start of the trial via a handheld spray bottle, and each treatment was diluted so that 10 mL of sprayed solution was the desired application rate per microplot. Fluensulfone was broadcast with a Scotts Easy Hand-Held Broadcast Spreader. All nematicides were watered with 0.64 cm of water after application.

Data collection occurred at 0, 30, and 60 days after treatment (DAT) for both 2018 and 2019 trials. Data collected included nematode population density, visual turf quality assessment, NDVI, and NDRE. At each collection date, a 100-cm³ soil sample was taken from each plot. For M. incognita, population density, the soil sample was placed in a modified Baermann funnel (Castillo et al., 2013), and after 48 hr, juveniles were collected on a 25-µm pore sieve. For B. longicaudatus, the 100-cm³ sample was collected and nematodes were extracted using the modified centrifugal flotation technique as previously described. Once extracted, both M. incognita and B. longicaudatus presence were confirmed and quantified via a Nikon TSX 100 inverted microscope at 40-x magnification.

Turfgrass vigor was calculated using the National Turfgrass Evaluation Program (NTEP) guidelines (Parsons et al., 2015). Visual ratings consisted of a 1-9 rating scale, where 1 was very poor quality turf, 6 was minimal acceptable turf quality, and 9 was exceptional turf quality (Morris, 2004). Multispectral data were collected the same day that visual ratings were made. Drone flights occurred within 2 hr of solar noon on clear sunny days or light overcast days when ambient light was not changing. Specific dates of flights in 2018 were July 5, August 4, and September 3. Flight dates in 2019 were July 19, August 18, and September 17. A DJI Phantom 4 Professional drone (SZ DJI Technology Co.; Shenzhen, China) equipped with a MicaSense RedEdge-M (MicaSense, Inc.; Seattle, WA) was used for flight and image acquisition. Flight speed was 10 meters per second at a height of 35 meters, and imagery was spaced with 80% front overlap and 88% side overlap. The sensor measured wavelengths of blue (475 nm), green (560 nm), red (668 nm), red edge (RE, 717 nm), and near infrared (NIR, 840 nm). Automated image processing was performed with Pix4Dmapper (Pix 4D; Prilly, Switzerland), and wavelength values were calculated using ArcMap (Esri; Redlands, California). Wavelengths red, red edge, and near infrared were used to calculate NDVI and NDRE values for each plot in ArcMap. NDVI and NDRE were calculated as previously described.

A field trial was conducted in the summer of 2019 at Montevallo Golf Club in Montevallo, AL to assess the ability of multispectral imagery to track nematicidal responses on a plant-parasitic nematode infested golf course. Four putting greens consisting of ‘TifEagle’ hybrid bermudagrass were selected for the experiment: three with a history of high plant-parasitic nematode population density, and one with low population density. Each green was divided into four quadrants of approximately 75 m², with three quadrants receiving a nematicide application, and one left as an untreated control. Treatments include: (i) untreated control; (ii) abamectin at 0.89 L/ha; (iii) fluopyram at 1.25 L/ha; and (iv) fluensulfone at 134 kg/ha.

The trial was initiated on July 22. Nematode samples were collected at 0 DAT on July 22, at 28 DAT on August 19, at 56 DAT on September 16, and 84 DAT on October 14. At each sample date, six soil cores (2.5 cm × 10 cm) were taken at roughly equal intervals in a zigzag pattern across each quadrant of a green. Collected soil samples were mixed and a 100-cm³ subsample was processed to determine the plant-parasitic nematode population density from each quadrant. Nematodes were extracted by gravity sieving followed by sucrose centrifugation following the methodology of Jenkins (1964) as previously described. Plant-parasitic nematodes were confirmed and enumerated by a Nikon TSX 100 inverted microscope at 40-x magnification, and morphologically identified to genus or species if possible based upon ‘Pictorial Key to Genera of Plant-Parasitic Nematodes, 4th Edition’ (Mai and Lyon, 1975) and ‘Identification Guides for Most Common Genera of Plant-Parasitic Nematodes’ (Eisenback, 2002).

Drone flights occurred during the trial on approximately 14-day intervals. Flights occurred under full sun as close to solar noon as possible to limit environmental impact. Specific dates were July 22, August 5, August 19, September 3, September 16, September 30, and October 14. Flights were performed similarly to microplot evaluations, capturing NDVI, and NDRE values for each treatment quadrant. Visual turfgrass quality ratings were also conducted on the 1 to 9 scale using the NTEP guidelines.

Data collected from microplot and field evaluations were statistically analyzed using SAS (version 9.4; SAS Institute, Cary, NC). Means separations were determined by Tukey’s multiple range test for each evaluation date (P ≤ 0.05). Rating methods (NTEP, NDVI, and NDRE) were compared statistically by the Pearson correlation coefficient at P ≤ 0.05, P ≤ 0.01, and P ≤ 0.001 to determine the strength of the linear relationship between nematode populations and foliar ratings.

M. incognita population density was significantly impacted by nematicide treatments in both 2018 and 2019 (Table 1). All nematicides significantly lowered M. incognita population density in August and September (30 and 60 days after treatment; DAT) for both years (P ≤ 0.05) (Fig. 1A, B). In 2018, only fluopyram significantly increased turf quality compared to the untreated control in August and September (P ≤ 0.05) (Fig. 1C). In 2019, both fluopyram and fluensulfone significantly increased turf quality compared to the untreated control in August, and fluopyram and abamectin significantly increased turf quality compared to the untreated control in September (P ≤ 0.05) (Fig. 1D). NDVI was significantly higher in both years for bermudagrass treated with fluopyram and abamectin in August, and all nematicides in September (P ≤ 0.05) (Fig. 1E, F). No significant differences were observed for NDRE in August 2018 and fluopyram, abamectin, and furfural had a significantly higher NDRE value compared to the untreated plots in September (P ≤ 0.05) (Fig. 1G). In 2019, all nematicides significantly increased NDRE in September compared to the untreated control (P ≤ 0.05) (Fig. 1H).

Figure 1:

In 2018, fluopyram and fluensulfone significantly reduced B. longicaudatus population density compared to the untreated control in August and September, and in 2019 fluopyram, fluensulfone, and furfural significantly reduced population density in August and September compared to the untreated control (P ≤ 0.05) (Fig. 2A, B). Turf quality was significantly improved by fluopyram and fluensulfone compared to the untreated control in August and abamectin, fluopyram, and fluensulfone in September in 2018 (P ≤ 0.05) (Fig. 2C). In 2019, fluopyram significantly increased turf quality compared to the untreated control in August, and abamectin, fluopyram, and fluensulfone increased turf quality compared to the untreated control in September (P ≤ 0.05) (Fig. 2D). In 2018, abamectin and fluopyram significantly increased NDVI compared to the untreated control in August, and fluopyram significantly increased NDVI compared to the untreated control in September (P ≤ 0.05) (Fig. 2E). For 2019, NDVI was significantly improved by abamectin, fluensulfone, and fluopyram in August, and abamectin and fluopyram in September (P ≤ 0.05) (Fig. 2F). Abamectin and fluopyram both significantly improved NDRE in August of 2018 compared to the untreated control, and fluopyram significantly improved NDRE in September compared to the untreated control (P ≤ 0.05) (Fig. 2G). In 2019, only fluopyram significantly improved NDRE in August compared to the untreated, and fluopyram and abamectin significantly improved NDRE compared to the untreated control in September (P ≤ 0.05) (Fig. 2H).

Figure 2:

Turfgrass visual ratings, NDVI, and NDRE were correlated with M. incognita and B. longicaudatus population density in both 2018 and 2019. In 2018, turf visual ratings significantly correlated with M. incognita population density in August (P ≤ 0.05), and all three sample dates in 2019 (P ≤ 0.05) (Table 1). NDVI was significantly correlated with M. incognita population density in August (P ≤ 0.05) and September (P ≤ 0.001) in 2018 and 2019. NDRE was significantly correlated with M. incognita population density in September in 2018 (P ≤ 0.01) and 2019 (P ≤ 0.001). B. longicaudatus population density was significantly correlated with turf visual ratings at July (P ≤ 0.001) and September dates (P ≤ 0.01) in 2018, and August and September dates (P ≤ 0.01) in 2019 (Table 2). Both NDVI and NDRE were significantly correlated with B. longicaudatus population density in September 2018 (P ≤ 0.001), and all sample dates in 2019 (P ≤ 0.01) (Table 2).

Turfgrass visual ratings, NDVI, and NDRE were highly correlated with each other throughout both the M. incognita and B. longicadatus microplot evaluations. In the M. incognita trial, turf visual quality was positively correlated with NDVI in July (P ≤ 0.01) of 2018, and September of 2019 (P ≤ 0.001) (Table 1). NDRE was positively correlated with turf visual quality in September of 2019 (P ≤ 0.001). NDVI and NDRE were positively correlated with each other at every evaluation date in 2018 and 2019 (P ≤ 0.001), except for July in 2018 (Table 1). For the B. longicaudatus trial, turf visual quality was positively correlated with NDVI and NDRE in September of 2018 (P ≤ 0.001), and both August and September of 2019 (P ≤ 0.05) (Table 2). NDVI and NDRE were positively correlated to each other at every flight date in 2018 and 2019 (P ≤ 0.001).

Five plant-parasitic nematode genera were identified throughout the duration of the 2019 field trial at Montevallo Golf Club. These include Hoplolaimus spp. (lance nematode, Fig. 3A), Helicotylenchus spp. (spiral nematode, Fig. 3B), Meloidogyne spp. (Fig. 3C), B. longicaudatus (Fig. 3D), and Criconemoides spp. (ring nematode, Fig. 3E). All nematicides reduced total plant-parasitic nematode population density on August 19, September 16, and October 14 compared to the untreated control (P ≤ 0.05) (Fig. 4A). Fluopyram significantly improved visual turf quality and NDVI on August 19, September 3, September 16, September 30, and October 14 (P ≤ 0.05) (Fig. 4B, C), and NDRE on August 5, August, 19, September 3, September 16, September 30, and October 14 (P ≤ 0.05) (Fig. 4D) compared to the untreated control. Abamectin and fluopyram also significantly improved visual turf quality at two evaluation dates apiece (P ≤ 0.05) (Fig. 4B). Abamectin significantly improved NDVI September 30 and October 14 compared to the untreated control, and significantly improved NDRE on September 3, September 30, and October 14 (P ≤ 0.05) (Fig. 4C, D). Fluensulfone significantly improved NDRE on September 30 (P ≤ 0.05) (Fig. 4D).

Figure 3:

Figure 4:

Total plant-parasitic nematode population density was highly correlated with turfgrass visual ratings, NDVI and NDRE. Correlation occurred with visual ratings on July 22 (P ≤ 0.01), August 19, September 16, and October 14 (P ≤ 0.001) (Table 3). Nematode population density and NDVI were correlated on July 22 (P ≤ 0.05), August 19, September 16, and October 14 (P ≤ 0.001) (Table 3). NDRE and nematode population density were correlated on August 19 (P ≤ 0.05), September 16 (P ≤ 0.01), and October 14 (P ≤ 0.001) (Table 3). NDVI and turf quality ratings were positively correlated at five of the seven evaluation dates, and NDRE and turf quality were positively correlated at four of the seven evaluation dates (P ≤ 0.01) (Table 3). NDVI and NDRE were positively correlated at five of the seven evaluation dates (P ≤ 0.05) (Table 3).