Materials and methods

The study was conducted at the Experimental Agricultural Farm of the University of Padova, which is located in Legnaro in northeastern Italy (lat. 45°20′N, long. 11°57′E, 8 m above sea level). The soil at the site consisted of a coarse-silty, mixed, mesic, Oxyaquic Eutrupedt (Morari 2006) containing 14% clay, 69% silt, and 17% sand (pH, 8.1), 2.4% organic matter (loss on ignition method), a carbon (C):nitrogen (N) ratio of 12.2:1, an Olsen phosphorus (P) content of 29 mg·kg⁻¹, and an exchangeable potassium (K) content of 140 mg·kg⁻¹. The area’s climate is classified as sub-humid (Köppen 1936), with a mean annual temperature of 12.3 °C and annual rainfall of 837 mm mostly distributed from April to November (ARPAV 2023). The average daily air temperature at 2 m aboveground, and the daily precipitation is reported in Fig. 1; the monthly air temperature, monthly precipitation, and daily reference evapotranspiration (ET₀) (Berti et al. 2014) are reported in Table 1.

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Monthly mean air temperature, precipitation, and reference evapotranspiration during the study period and long-term averages of monthly mean temperature and precipitation (1994–2020) at the Experimental Agricultural Farm of the University of Padova, located in Legnaro, northeastern Italy (lat. 45°20′N, long. 11°57′E, 8 m above sea level).

View Fullscreen

The study was conducted from Mar 2021 to Mar 2023. The experimental design was a complete randomized block with three replications, and the individual plot size was 4 m² (2 m × 2 m). Seedbed preparation consisted of removing existing vegetation, cultivating the site with a tractor-mounted rototiller (Lamborghini R70; Cento, Ferrara, Italy) to a depth of 30 cm, and grading to provide a smooth planting surface of desired contour. Individual plots were seeded carefully by hand to avoid contamination into adjacent plots in Sep 2020 with 20 commercially available tall fescue cultivars at a rate of 40 g seed/m² (Caturegli et al. 2015; Schiavon et al. 2021a). The cultivars were selected based on their availability in the European turf market or their potential to be introduced based on turf industry requirements. The germination rate listed on the seed labels were in the range of 90% to 95%. The 20 tall fescue cultivars evaluated are reported in Table 2.

Table 2. Sources of the 20 tall fescue cultivars evaluated at the Experimental Agricultural Farm of the University of Padova, located in Legnaro, northeastern Italy (lat. 45°20′N, long. 11°57′E, 8 m above sea level).

View Fullscreen

Before seeding, plots were fertilized with 50 kg N/ha, 150 kg P₂O₅/ha, and 150 kg K₂O/ha from an 8N–24P₂O₅–24K₂O granular fertilizer. Immediately after seeding, the study site was irrigated daily to maintain adequate soil moisture by means of an overhead sprinkler irrigation system during Sep and Oct 2020. At no time during this establishment period was seed observed being moved into adjacent plots. By Nov 2020, tall fescue had emerged and the visual turf cover was observed in all plots; therefore, successful establishment was considered and irrigation was suspended. A single irrigation intervention (40 mm), however, was performed on 14 Jul 2022 to avoid irreversible damage caused by the prolonged drought period at the beginning of Summer 2022 (Fig. 1, Table 1).

Fertilization was applied in equal amounts in Mar, May, and Sep in 2021 and 2022 using controlled-release fertilizer (22N–5P₂O₅–8K₂O + Mg + Fe + S) for a total of 120 kg N/ha/year. Plots were mowed once per week at a height of 52 mm with a rotary mower (Honda HRX 476 SC E; Honda Motor Co. Ltd., Tokyo, Japan) and clippings were removed. For both 2021 and 2022, mowing began in March and concluded in October. Four soil moisture (WaterScout SM100 Soil Moisture; Spectrum Technologies Inc., Plainfield, IL, USA) sensors connected to a data logger (Spectrum Technologies Inc.) were installed at a depth of 5 cm at random at the test site. Soil moisture was recorded hourly as the percent volumetric water content; thereafter, the average soil availability water capacity (Berti et al. 2014) was calculated (Fig. 1).

Grass weeds were manually removed, whereas broadleaf weeds were controlled as needed using Vithal Turfene plus selective herbicide (active ingredients = dicamba/MCPP-P; Fitogarden, Scanzano Jonico, Italy). The disease control strategy was limited to fungicide applications against Pythium spp. (active ingredients = propamocarb plus fosetyl-aluminum; Previcur Energy; Bayer CropScience, St. Louis, MO, USA) and consisted of two foliar applications in Summer 2022. No tall fescue foliar injury or phytotoxicity was observed from the applications of those herbicides or fungicides.

The tall fescue stand in each plot was evaluated every 14 d from Mar 2021 to Mar 2023 to determine texture, uniformity, density, color, and quality based on a visual scale (1 = worst; 6 = minimally acceptable; 9 = best) (Krans and Morris 2007). The green turf cover (percent plot area with green turf cover) was evaluated using a digital image analysis (Richardson et al. 2001) and the normalized difference vegetation index (NDVI) using RapidScan CS-45 (Holland Scientific, Lincoln, NE, USA). The biweekly data were averaged for each month.

All data were subjected to an analysis of variance (ANOVA), which was performed using a linear mixed effects model to test the effects of cultivar, sampling date, and their interactions on the measured parameters (i.e., texture, uniformity, density, color, quality, green turf cover, and NDVI). The blocks within the sampling date and main plots within blocks were included as random effects to account for the clustering of observations. An additional ANOVA was performed using a linear mixed effects model to test the effects of cultivars on soil moisture. To exclude spatial correlation, the longitudinal and latitudinal positions of each plot were included as covariates (X and Y positions) in each model. The statistical difference was determined by likelihood-ratio tests of models including and not including the position as a covariate. For soil moisture, the position was not significant based on Akaike’s information criterion, excluding any inhomogeneous soil moisture distribution. A least significant difference test with the Bonferroni correction at P = 0.05 was used to identify significant differences among means. All statistical analyses were performed using R version 4.3.1 (R Core Team 2023).

Results

The only two parameters with a significant interaction between cultivar × assessment date main effects were texture and color (Table 3). For the remaining parameters (i.e., turf quality, density, uniformity, NDVI, and green turf cover), the ANOVA revealed significant main effects (Table 3). Seasonal soil moisture levels in spring, summer, and fall were not significantly affected by cultivars in either year of the study period (data not shown).

Table 3. Results of an analysis of variance for the effects of sampling date, cultivar, and their interactions on turf quality, density, texture, color, uniformity, normalized difference vegetation index (NDVI), and percent green turf cover (cover) of 20 tall fescue cultivars, 2021–23.

View Fullscreen

Turf quality, color, and uniformity for all tall fescue cultivars displayed a similar performance trend during the study period (Fig. 2), with low ratings (<6) in winter and summer, thus reflecting the bimodal seasonal growth of tall fescue (Jesperson et al. 2023). Density was maintained above the minimal acceptable level during the entire study period (Fig. 2), although with a noticeable slight decrease during Summer 2021, Summer 2022, and Winter 2022–23. Also, in Summer 2022, most density ratings were lower than those of Summer 2021 (Fig. 2). Texture ratings were always above the minimal acceptable rate (6), with some decline observed during Summer 2021 and after Apr 2022 (Fig. 2). Despite the considerable change in turf quality, color, and uniformity over seasons, texture did not exhibit relevant or significant variations during the entire study period and, therefore, environmental conditions had a minimal influence on this important trait (Fig. 2). Because of the consistency of texture ratings over the study period, the interaction between cultivar × sampling date did not provide any additional information beyond the main effects (Table 3).

View Figure

• Download as Powerpoint
• Download Figure

On average, across the 2-year study period, the cultivars Triad, ZRC-1, PPG-TF336, and PPG-TF238 achieved significantly higher quality ratings compared with those of Bullseye, Lexington, Forlaine, and RGT Nuance, which had lower ratings than those of Bullseye (Fig. 3). All cultivars produced an overall average quality rating >5 (Fig. 3); therefore, a quality rating between 5 and 6 could be considered “marginally acceptable” (Morris et al. 2023). All cultivars produced an acceptable density rating (i.e., rating >6) averaged across the entire study period (Fig. 3). The cultivars ZRC-1, Triad, PPG-TF254, and PPG-TF336 displayed significantly higher density ratings compared with those of Lexington, Forlaine, Grande II, RGT Nuance, and Melyane (Fig. 3). With uniformity, all cultivars produced an overall average rating >5 (Fig. 3); therefore, a uniformity rating between 5 and 6 could be considered marginally acceptable (Morris et al. 2023). The cultivars ZRC-1 and Triad exhibited higher uniformity ratings than those of Grande II, Forlaine, Lexington, RGT Nuance, and Melyane (Fig. 3). Thus, within the environmental parameters of this 2-year study, the cultivars with consistently best quality, density, and uniformity were ZRC-1, Triad, and PPG-TF336.

View Figure

• Download as Powerpoint
• Download Figure

For most of the cultivars evaluated, color ratings were higher in spring and fall than they were in summer and winter, with the exceptions of Forlaine, RGT Nuance, and Melyane, which never reached the minimum acceptable rating throughout the entire study period (Fig. 4). In Aug 2021, no significant color differences among cultivars were observed; however, during Fall 2021, the cultivars Darlington II, Turfway, and Triad displayed significantly higher ratings than those of Forlaine, Rhambler SRP, and Melyane (Fig. 4). Significant color differences were observed in May 2022, with Thor, Darlington II, Tough, Triad, and Thunderstruck producing higher color values than those of Forlaine, RGT Nuance, and Melyane (Fig. 4). Similarly, in Aug 2022, significantly higher color ratings were observed for cultivars Darlington II, Lexington, Thor, PPG-TF336, Grande II, Tough, Rhambler SRP, and Thunderstruck compared with those of Forlaine, Bullseyes, RGT Nuance, Detonate LS, and Melyane (Fig. 4). In Oct 2022, in the second year of this study, color ratings of cultivars Darlington II, ZRC-1, Tough, and Turfway were significantly higher than those of Forlaine, RGT Nuance, and Melyane (Fig. 4). Overall, for all cultivars from Jun 2022 onward, the color ratings were lower compared than those during the same period in 2021 (Fig. 4).

View Figure

• Download as Powerpoint
• Download Figure

A similar trend for all tall fescue cultivars was observed for green turf cover and NDVI, with lower values measured during the Summers and Winters of 2021 and 2022 (Fig. 5). Of note, in Summer 2022, green turf cover reached 40% and NDVI measured 0.4, which were lower than 70% for green turf cover and 0.7 for NDVI recorded in Summer 2021 (Fig. 5). After Summer 2022, however, tall fescue recovered, as reflected in green turf cover and NDVI values similar to those during Fall 2021 (Fig. 5).

View Figure

• Download as Powerpoint
• Download Figure

Discussion

During the study period of Mar 2021 through Mar 2023, the regional weather was characterized by prolonged drought periods and high summer temperatures (Table 1, Fig. 1). In late Spring (May) and Summer 2022, air temperatures were higher than those during the same period in 2021; however, air temperatures were lower than the long-term averages during the spring of both years (Table 1), and air temperatures were lower in Apr 2022 than in Apr 2021 (Fig. 1). Regarding precipitation, the total amount from Jan to Aug 2022 was 276 mm; however, during the same period in 2021, the total amount measured 467 mm, and both years had lower precipitation values than the long-term average of 522 mm (Table 1). Data of the soil available water capacity revealed a long period of drought condition from Mar to Nov 2022, with values lower than the wilting point calculated at 5 cm depth. In addition, during this field study, natural precipitation was concentrated in a few high-intensity events (Fig. 1). A single emergency irrigation was needed during the study period because of unusual climatic conditions. This occurred in Jul 2022, after 6 months with a total rainfall of 128 mm and precipitation event of no more than 20 mm of water each, and 1 month (June) with a mean ET₀ value of 4.55 mm·d⁻¹ (Table 1, Fig. 1).

A significant interaction between the cultivar × assessment date was detected only for texture and color (Table 3). Of note, cultivars that expressed a darker green genetic color during optimum growth conditions in the spring and fall were the poorest-performing cultivars in summer and winter when compared with cultivars with a lighter green color (Fig. 4). The lack of interaction between the assessment date × cultivar suggested that the cultivars tested responded similarly to environmental variations over the 2 years. However, the selection of cultivars for drought resistance could be based on their ability to maintain green color and uniformity during the drought period.

The visual characteristics and traits of tall fescue during summer and winter have been observed in previous studies (Pornaro et al. 2020; Schiavon et al. 2021b). In this field study, leaf texture did not vary significantly among seasons. However, lower texture ratings were observed during drought stress periods as a potential consequence of a tall fescue stress-coping mechanism of rolling-up or curling leaves to reduce leaf surface area, thus conserving physiological water (Fig. 2) (Lang et al. 2004; Wang et al. 2008; Yuan et al. 2015). This behavior most likely affected texture ratings beginning from Spring 2022, with increasing ET₀ when the cumulative rainfall was low (Fig. 1, Table 1), which contributed to reducing any noticeable or detectable texture differences among the 20 cultivars.

The weather and subsequent environmental conditions may justify the different behavior of tall fescue cultivars during the first and the second years of the study (Table 1), especially with NDVI and green turf cover starting in May 2022. In Aug 2021, green turf cover was 65% and the NDVI was 0.7 among all tall fescue cultivars; however, in Aug 2022, it was reduced to 50% and 0.4, respectively (Fig. 5). The differences could be attributed to the different weather conditions (Table 1, Fig. 1). This trend also affected other turf parameters (Fig. 2). Of note, it would be difficult to compare findings from this field study to those of other studies conducted under controlled environmental conditions. However, the results of this field study under drought stress conditions are consistent with those of Hong et al. (2021) and Karcher et al. (2008) regarding the aptitude of tall fescue to function and survive under water deficit conditions.

Throughout the 2-year study period, ZRC-1, Triad, and PPG-TF336 emerged as the better-performing cultivars for turf quality, NDVI, and green turf cover. These three tall fescue cultivars were also evaluated by the National Turfgrass Evaluation Program (NTEP; Beltsville, MD, USA) and designated as having the best for turf quality in the transition zone locations of Kansas, Maryland, North Carolina, Oklahoma, Texas, and Virginia (in the United States), with mean ratings of 6.33 for ZRC-1, 6.29 for Triad, and 6.19 for PPG-TF336 (NTEP; Beltsville, MD, USA). The NTEP results also indicated high quality ratings as averaged over drought-stressed locations for ‘ZRC-1’ and ‘PPG-TF336’, at 7.0 and 6.7, respectively; however, ‘Triad’ displayed a lower rating of 6.0. ‘Talladega’, ‘PPG-TF238’, and ‘PPG-TF315’ were also drought-tolerant, with ratings of 7, 7.3, and 6.7, respectively (NTEP; Beltsville, MD, USA). The results of this field study are consistent with the NTEP results and displayed no significant differences when compared with those of the best-performing cultivars (Fig. 3).

This field study site was considered spatially homogeneous in terms of soil moisture. The genetic potential of each cultivar to create and maintain a high canopy density is considered beneficial to reducing water loss through soil evaporation (Huang 2008), and this could be related to a decreased need for water by the cultivar, as associated with the covering effect of the turf canopy. Drought-resistant turf cultivars can decrease irrigation needs and inputs and conserve valuable water resources, which are crucial concerns in regions of northern Italy that are increasingly prone to prolonged drought because of climate change. However, it is evident that this species has the ability to endure long periods of drought by using water stored in the deeper soil layers. Therefore, these results further confirm the excellent environmental adaptability of this species to northern Italy’s conditions, thus demonstrating the possibility of maintaining healthy turfgrass even with minimal irrigation.