Performance of Turf Bermudagrass Hybrids with Deficit Irrigation in the Desert Southwest USA

Abstract

Water scarcity poses a substantial challenge for turfgrass irrigation in the drought- and heat-stressed Desert Southwest region of the United States. Bermudagrass (Cynodon spp.), renowned for its exceptional drought resistance, is the predominant warm-season turfgrass in the region. Selecting and using drought-resistant bermudagrass cultivars remains a primary strategy for sustaining the turfgrass industry in the region. This study evaluated 48 hybrid bermudagrasses (Cynodon dactylon × C. transvaalensis Burtt-Davy), including two commercial cultivars (‘TifTuf’ and ‘Tifway’, as controls), under 80% × ETo (0.8ET), 60% × ETo (0.6ET) and 40% × ETo (0.4ET) reference evapotranspiration (ETo) replacement irrigation systems at Maricopa, AZ. The experiment was laid out in a split-plot design with two replications, where the 3 irrigation treatments were assigned to main plots and 48 genotypes were in sub-plots. Analysis of data from two years (2022 and 2023) revealed significant differences among bermudagrass hybrids, irrigation treatments, and their interaction effects. The hybrids exhibited substantial variation for spring green-up, density, turf color, and quality. With the largest deficit irrigation treatment 40% × ETo (0.4ET), OSU2104, OSU2106, and OSU2105 showed greater mean greenness and aesthetic quality scores than recorded for ‘TifTuf’ (6.5), a popular drought-tolerant cultivar. The results highlight the prevalence of genetic variation in germplasm with potential for development of improved varieties for drought tolerance.

1. Introduction

Bermudagrass (Cynodon spp.) is a warm-season turfgrass widely used in golf courses, sports fields, parks, and landscapes in warm climatic regions of the world. Among warm-season grass species commonly used as turfgrass throughout the world, bermudagrass is the most important and widely adapted [1]. Its relative drought and heat tolerance, high traffic tolerance, and recuperative ability makes it an ideal turfgrass for arid environments with significant ecological, environmental, recreational, and economic importance. Interspecific hybrid bermudagrass (C. dactylon × C. transvaalensis Burtt-Davy) forms higher density and thicker turf with aesthetic beauty and delightful uniformity [2]. Its ability to withstand stresses better than common bermudagrass (C. dactylon) [3] makes it the most economically important and widely used for home lawns, sports fields, and golf courses [4].

Water scarcity in arid regions imposes considerable challenges for maintaining turfgrass, which requires substantial amounts of water for its survival and acceptable aesthetics [5]. High temperatures and low humidity in desert environments cause rapid water evaporation, requiring more water to maintain adequate moisture levels in turfgrass, thereby rendering reduced irrigation efficiency. In the Desert Southwest U.S., water resources are already over-stressed due to agriculture, urbanization, and rapid population growth [6,7]. Climate change has significantly exacerbated water shortages in the region due to increased frequency and intensity of droughts, shrinking snowpacks, increased evaporation rates, reduced flow of perennial streams, and modified hydrological systems [8]. Therefore, due to the heightened water demand, the available water must often be diverted for essential human need, leaving limited supplies for landscaping and turfgrass.

Turfgrass water use is substantially affected by a complex interplay of factors including grass species and cultivar, soil characteristics, atmospheric conditions, water chemistry, and turfgrass management practices such as mowing height and fertilization regimens [9,10,11]. The extent of turfgrass water requirements vary between species, with substantial genotypic variations observed within species [12]. Among warm-season turfgrasses, bermudagrass has gained widespread adoption in arid environments because of its broad genotypic variations for stress response [13,14].

Anthropogenic climate variability will likely have further impact on the availability of water resources for turfgrass irrigation. In the Desert Southwest region of the United States, the combination of frequent drought and heat stresses and limited surface water availability are serious environmental challenges that limit turfgrass establishment and maintenance. This difficult environment has created a persistent demand for improved stress-tolerant turfgrass varieties that perform well with limited management inputs [15]. Consequently, it is important to enhance turfgrass water use efficiency to keep it alive and at acceptable quality levels under deficit irrigation. Such an approach is essential to preserve the long-term recreational function and ecosystem services provided by turfgrass [16,17].

The prevalence of both inter- and intraspecies genetic variations for environmental stress tolerance has been instrumental in the development of stress-tolerant varieties in different crops [18]. This genetic variability may be attributed to several complex interacting factors, including efficient water absorption, internal water retention, and accumulation of thermo- and drought-protective compounds [19]. Several drought tolerance studies conducted on different turfgrass species found that certain genotypes maintain acceptable turf quality and greenness during drought stress due to their relatively low water demand [20], specific shoot or root structures [12,21], or dormancy mechanisms [22]. Understanding the natural genetic variation for drought tolerance among available germplasm will provide novel insights to advance bermudagrass breeding programs. Therefore, evaluation of available germplasm is expected to yield relatively stress-tolerant genotypes with acceptable aesthetic value in challenging and stressful environments. The objectives of this study were to (1) investigate the relative performance of 48 hybrid turf bermudagrass genotypes under deficit irrigation; (2) better understand the interaction of genotypes with deficit irrigation input; and (3) generate locally relevant research-based information on the water conservation potential of drought-tolerant bermudagrass hybrids.

2. Materials and Methods

2.1. Experimental Description and Plant Materials

A field experiment of 46 interspecific hybrid bermudagrass hybrids and 2 commercial cultivars (‘TifTuf’ and ‘Tifway’, as checks) (C. dactylon × C. transvaalensis) was planted 30 August 2021, where four 100 cm² plugs of each genotype were transplanted 50 cm apart in the center of each plot with 0.50 m alleys between plots. These hybrids were developed by Oklahoma State University and selected based on their performance in previous small-plot progeny testing in Stillwater, Oklahoma. The checks (‘TifTuf’ and ‘Tifway’) were selected due to industry prevalence and adaptability [23,24]. Vegetative clones from stolons of the genotypes were initially propagated in a greenhouse environment. These were grown in 4.0 cm diameter 10 cm deep containers filled with a potting mix of sand–soil–peat–perlite in 2:1:3:3 ratios (v:v) in the greenhouse. Four plugs were planted in a square of 0.5 m × 0.5 m at the center of 2.25 m² plots. All 48 genotypes were established on Casa Grande sandy loam soil (92% sand, 4% silt, and 4% clay containing 0.84% organic matter) at the Maricopa Agricultural Center in Maricopa (MAC), Arizona (33°3′24″ N 112°2′48″ W).

The trial was laid out in split-plot design, arranged as a randomized complete block design with two replications, where irrigation treatments were assigned to main plots and genotypes were sub-plots. Three irrigation levels, 80% × ET_o (0.8ET), 60% × ET_o (0.6ET), and 40% × ET_o (0.4ET), were assigned to the main plots, with 48 genotypes randomized across the subplots. Each subplot was established with a plot area of 1.5 m × 1.5 m.

All plots were planted 31 August 2021 and were established using 100% × ET_o replacement irrigation using an automatic sprinkler system. Then, the plots were maintained on a minimum irrigation in the off-season until April 30, both in 2022 and 2023. The deficit irrigation treatments were administered from 1 May to 31 October in both 2022 and 2023.

Plots were watered four times weekly based on a previous weekly ET_o multiplied by their respective irrigation level. Plots were mowed once weekly using a rotary mower at a height of 5.0 cm from April through October during treatment years and as needed November through March. Regular mowing and fertility management was performed as recommended for the area [25]. Plots were fertilized annually with 193 kg N ha⁻¹.

2.2. Data Collection and Statistical Analysis

Visual ratings (1–9 scale) for green-up, density, turf color, and quality were collected from the planting year through 2023. The plots’ color and quality performance under the three irrigation levels were scored using a scale from 1 (brown plot) to 9 (ideal turfgrass), as described by the National Turfgrass Evaluation Program (NTEP) standards [26], where 1 represents a poor performing plot, 6 is considered acceptable for the rated trait, and 9 represents outstanding performance. Green-up of the plots was assessed early in the spring. Turfgrass color and quality were rated every two weeks starting in the month of March to the end of the trial period in 2023. Percentage ground cover, which represents establishment of the genotypes, was visually rated on a 0-to-100 scale. A density rating was made at the middle of the summer using a 1-to-9 scale. Fall color was evaluated in the fall of 2022 and 2023, with 1 representing straw brown and 9 representing dark green.

A combined analysis of variance for spring green-up and turf density for the 48 hybrid bermudagrass cultivars and irrigation treatments was conducted using a general linear model in the SAS Software, version 9.4 [27]. A repeated measures ANOVA was conducted to examine the effect of year, irrigation level, genotype, and their interactions on turfgrass quality and color performance in 48 hybrid turf bermudagrass. Repeated measures ANOVA was performed using the general linear model (GLM) with the polynomial function to examine the effects of subject factors (irrigation level, genotype, and their interaction) and the effect of time within subject factors on turfgrass quality and color performance. To distinguish the irrigation treatments, the genotypes, and the interaction effects, combined analysis of variance was conducted following the general linear model:

y_ijk = µ + αi + δj + (αδ)_ij + γ_k + (αγ)_ik + (δγ)_jk + (αδγ)_ijk + e_ijk;

where y_ijk is the empirical mean response of the ith genotype (i = 1, 2,…, g) in the jth irrigation (j = 1, 2,…, e) with r replications in each of the g×e cells, μ is the grand mean across all genotypes (across irrigations and years), α_i is the effect of the ith genotype, δ_j is the effect of the jth irrigation, (αδ)_ij is the interaction of the ith genotype in the jth irrigation, γ_k is the kth year effect, (αγ)_ik is the interaction effect of the ith genotype and kth year, (δγ)_jk is the interaction effect of the jth irrigation and kth year, (αδγ)_ijk is the interaction effect of the ith genotype, jth irrigation, and kth year, and eijk is the random error effect assumed to be normally and independently distributed.

The means for different levels of treatment were compared using Fisher’s protected least significant difference (LSD) test (p < 0.05). Then, the LSD_0.05 values for the comparisons were presented with a comparison table. The rank summation index (RSI) was also computed for the means of the four traits following the method outlined by Mulamba and Mock [28].

3. Results

3.1. Environmental and Experimental Conditions

Main plots were irrigated at levels of 0.8ET, 0.6ET, and 0.4ET as determined weekly based on reference evapotranspiration (ET_o) data from a weather station located ~100 m from the trial plots. The weather station is linked to the Arizona Meteorological Network (AZMET) (cales.arizona.edu/AZMAT/index.html) and provides daily ET_o values based on the FAO Penman–Monteith equation [29]. Monthly average air and soil temperatures (°C), relative humidity (RH%), total rainfall (mm), and reference ET_o (mm) variations at Maricopa, Arizona during 2022 and 2023 are presented in Figure 1. The highest irrigation level of 0.8ET was chosen to be a representative of the average summer crop coefficient (Kc) of ‘Tifway’ hybrid bermudagrass determined by previous research [30].

One degree of freedom for the replication error was decided based on the availability of resources for the experiment such as planting material and land area available. The minimum was used from the estimate based on the expected standard deviation, mean difference, and significance level.

3.2. Spring Green-Up and Summer Turf Density

The spring green-up and mid-summer turfgrass density of the genotypes varied across years and irrigation levels (Figure 2A,B). The spring green-up data were collected mid-March in both years 2022 and 2023. In 2022, all the genotypes greened-up uniformly irrespective of the irrigation levels (Figure 2A). However, in 2023, the median green-up data of the genotypes was greater for the largest deficit irrigation (0.4ET) as compared to the full (0.8ET) and moderate (0.6ET) ET_o replacement irrigation. This result was primarily attributed to lower thatch accumulation from the previous year on the largest deficit irrigated pots than the moderate deficit and the full. In 2023, most of the plots irrigated with 0.6 ET were 4 to 5 in green-up score. The median turfgrass density scores were higher in the year 2023 than 2022 across all three irrigation levels (Figure 2B), indicating that irrigation level affected the density scores of the genotypes.

Statistical analysis of the two years of green-up data revealed that there were statistically significant differences among years, irrigation treatments, genotypes, and year × irrigation and year × genotype interactions (

Table 1. Analysis of variance of spring green-up, summer turf density, and seasonal average quality and color ratings for 48 hybrid bermudagrass during 2022 and 2023 at Maricopa, Arizona.

Source of Variation	DF	Mean Squares
		Green-Up	Density	Quality	Color
Year	1	1326.2 ***	59.4 ***	1.1 **	0.6 *
Irrigation	2	7.4 ***	114.4 ***	25.7 ***	5.4 ***
Replication	1	0.1	0.6	0.1	0.7
Genotype	47	10.1 ***	6.6 ***	3.4 ***	3.2 ***
Year × Irrigation	2	7.4 ***	2.4	0.1	0.3
Year × Genotype	47	2.4 ***	0.2	0.3 **	0.3 ***
Irrigation × Genotype	94	0.9	1.6 ***	0.3 ***	0.2 **
Year × Irrigation × Genotype	94	0.8	0.19	0.1	0.1
Error	287	0.09	0.80	0.15	0.09
Total	575
R2		0.91	0.87	0.86	0.88
COV (%)		13.3	16.3	6.6	5.2
Mean		6.5	5.5	5.8	5.6

*, **, *** Significant at 0.05, 0.01, 0.001 probability levels.

). The overall mean green-up scores across years and irrigation levels for most of the hybrids were higher than 6.0. Ten hybrids, such as OSU2107(7.9), OSU2117(7.7), OSU2110 (7.4), OSU2073 (7.3), OSU2015 (7.4), OSU2104 (7.3), OKC1873 (7.3), OSU2035 (7.2), OSU2106 (7.1), and OSU 2101 (7.1), had green-up scores above 7.0. Similarly, the analysis of two years of turfgrass density data showed significant differences among years, irrigation levels, genotypes, and year × irrigation and year × genotype interactions. A total of 14 hybrids had an overall mean density score of 6.0 and above. OSU2104 had an overall mean density score of 7.1, followed by OSU2106 (6.8) and OSU2105 (6.5). The analysis of variance of color and quality data for each time point is provided in Supplementary Tables S1 and S2.

3.3. Turf Color and Quality

Turfgrass visual color and quality data for the bermudagrass hybrids were collected 13 times at intervals of approximately every 15 days during the 2022 and 2023 growing seasons. Figures S1 and S2 depict the distribution of visual color and quality assessment data, respectively, collected from early June to mid-November for three different irrigation regimes across two years. The violin plots illustrate that the distribution of visual greenness and turf quality scores (assessed on a 1–9 scale) varied throughout the season and among the irrigation levels.

The variations among the hybrids in greenness were highly significant (p < 0.001). Significant differences were also observed between years (p < 0.01), irrigation treatments (p < 0.001), year × genotype (p < 0.01), and irrigation × genotype (p < 0.01) interaction effects (Table 1). Visual quality scores differed significantly between the two years (

Table 2. Overall mean year and irrigation least significant differences (LSD_0.05) of turf bermudagrass hybrids evaluated for three irrigation levels during 2022 and 2023 at Maricopa, Arizona.

Variation	Green-Up	Density	Color	Quality
Year
2022	8.00	5.15	5.88	5.78
2023	4.97	5.80	5.81	5.87
LSD (0.05)	0.14	0.15	0.05	0.06
Irrigation
0.4ET	6.69	4.66	5.71	5.45
0.6ET	6.30	5.56	5.79	5.87
0.8ET	6.46	6.20	6.03	6.17
LSD (0.05)	0.17	0.18	0.06	0.08

). An average quality rating of 5.8 and 5.9 were scored in 2022 and 2023, respectively (LSD_0.05 = 0.06), while the average greenness rating was 5.8 both years (LSD_0.05 = 0.05).

The three irrigation levels had significant effects on both greenness and visual quality ratings (Table 2). Mean greenness ratings were 5.7, 5.8, and 6.0 for the 0.4ET_o, 0.6ET_o, and 0.8ET_o irrigation levels (LSD_0.05 = 0.06), respectively. Similarly, mean quality ratings of 5.5, 5.9., and 6.2 were recorded for the 0.4ET_o, 0.6ET_o, and 0.8ET_o irrigation levels (LSD_0.05 = 0.08), respectively.

Of the 46 hybrids evaluated, 21 had overall mean greenness scores of at least 6.0 or higher, with 14 of these also attaining quality scores of at least 6.0 (

Table 3. Two-year mean performance and rank summation index (RSI) for turf bermudagrass hybrids evaluated for three irrigation levels during 2022 and 2023 at Maricopa, Arizona.

Genotype	Green-Up	Rank	Density	Rank	Color	Rank	Quality	Rank	RSI
OKC1873	7.3	5	5.8	25	6.0	20	5.4	22	72
OKC1876	6.8	18	5.5	35	6.2	8	5.3	31	92
OSU1101	6.4	30	5.5	35	5.6	34	5.1	36	135
OSU1156	6.7	26	5.2	42	5.5	37	4.4	43	148
OSU1617	6.8	18	5.0	46	5.3	41	4.3	46	151
OSU2015	7.4	3	6.1	16	5.7	31	5.3	31	81
OSU2018	6.8	18	6.4	5	6.1	16	6.0	12	51
OSU2021	5.8	41	6.0	18	5.8	28	6.2	8	95
OSU2022	6.4	30	5.5	35	5.1	44	5.1	36	145
OSU2026	6.3	33	6.2	11	6.0	20	5.4	22	86
OSU2034	6.9	11	6.2	11	5.9	24	5.9	16	62
OSU2035	7.2	8	6.2	11	6.1	16	5.8	17	52
OSU2037	5.7	42	5.8	25	5.6	34	5.4	22	123
OSU2039	6.3	33	5.2	42	5.3	41	4.3	46	162
OSU2043	6.8	18	5.8	25	5.7	31	5.4	22	96
OSU2053	6.9	11	6.0	18	6.1	16	6.0	12	57
OSU2066	4.9	45	4.0	48	4.4	48	3.4	48	189
OSU2073	7.3	5	6.5	4	6.5	4	6.0	12	25
OSU2074	2.4	48	5.2	42	5.1	44	5.3	31	165
OSU2075	6.9	11	5.8	25	6.0	20	5.7	18	74
OSU2081	4.8	46	6.2	11	6.2	8	5.5	20	85
OSU2082	6.8	18	6.2	11	6.0	20	6.2	8	57
OSU2088	6.3	33	5.9	23	5.9	24	5.4	22	102
OSU2094	6.3	33	6.0	18	6.2	8	6.2	8	67
OSU2101	7.1	9	5.3	40	5.6	34	4.6	41	124
OSU2102	6.7	26	5.8	25	5.8	28	5.4	22	101
OSU2104	7.3	5	6.6	3	6.7	2	7.1	1	11
OSU2105	6.3	33	6.4	5	6.5	4	6.5	4	46
OSU2106	7.1	9	6.4	5	6.6	3	6.8	2	19
OSU2107	7.9	1	6.7	2	6.5	4	6.3	5	12
OSU2108	6.6	28	6.0	18	5.9	24	6.2	8	78
OSU2109	6.9	11	6.3	8	6.3	7	6.3	5	31
OSU2110	7.4	3	6.1	16	6.2	8	6.0	12	39
OSU2111	5.9	40	5.4	38	5.4	38	5.0	38	154
OSU2112	5.4	44	5.6	32	5.4	38	4.9	39	153
OSU2113	6.9	11	5.8	25	5.7	31	5.4	22	89
OSU2114	4.5	47	5.2	42	4.7	47	4.4	43	179
OSU2116	6.9	11	6.3	8	6.2	8	6.3	5	32
OSU2117	7.7	2	6.3	8	6.1	16	5.5	20	46
OSU2118	6.8	18	6.0	18	6.2	8	5.4	22	66
OSU2119	5.7	42	5.6	32	5.8	28	5.4	22	124
OSU2120	6.6	28	5.8	25	6.2	8	5.2	34	95
OSU2121	6.2	38	5.0	46	5.0	46	4.4	43	173
OSU2122	6.4	30	5.4	38	5.3	41	4.8	40	149
OSU2123	6.8	18	5.3	40	5.4	38	4.6	41	137
OSU2124	6.2	38	5.6	32	5.9	24	5.6	19	113
TifTuf (Check)	6.8	18	6.8	1	7.0	1	6.6	3	23
Tifway (Check)	6.9	11	5.9	23	6.2	8	5.2	34	76
LSD (0.05)	0.7		0.72		0.25		0.31		93

). The highest greenness score of 6.7 was scored for OSU2104, followed by OSU2106 (6.6), OSU2073 (6.5), OSU2105 (6.5), and OSU2107 (6.5), as compared to 7.0 and 6.2 scored for TifTuf and Tifway, respectively. An overall turf quality score of 7.1 was also recorded for OSU2104, followed by OSU2106 (6.8) and OSU2105 (6.5), as compared to the 6.6 scored for TifTuf and 5.2 for Tifway. Therefore, these experimental hybrids outperformed Tifway (5.2) and were comparable to or exceeded TifTuf.

The rank summation index of all the 48 hybrids was also computed for the means of the four traits (Table 3). Four genotypes, such as OSU2104, OSU2107, and OSU2106, had an RSI of 11, 12, and 19, respectively. Four other genotypes, namely OSU2073, OSU2109, OSU2116, and OSU2110, respectively, had an RSI of 25, 31, 32, and 39. The two commercial checks TifTuf and Tifway had an RSI of 23 and 76, respectively.

3.4. Repeated Measure Analysis of Turf Color and Quality

Visual turf color and quality scores were collected 13 times throughout the growing seasons in 2022 and 2023. Repeated measures analysis was conducted to examine temporal change in turf color and quality scores among irrigation treatments in the same hybrid and to isolate variability between hybrids. A repeated measures analysis of color data between subjects revealed statistically significant differences between irrigation, genotype, year × irrigation, year × genotype, and irrigation × genotype effects (

Table 4. Repeated measures analysis of variance for quality and color between-subject effect.

Source of Variation	DF	Mean Square
		Quality	Color
Year	1	13.3 **	4.4
Irrigation	2	331.3 ***	83.5 ***
Replication	1	1.2	5.9 *
Genotype	47	44.6 ***	42.2 ***
Year × Irrigation	2	1.0	5.3 *
Year × Genotype	47	3.7 **	4.1 **
Irrigation × Genotype	94	3.9 ***	2.5 ***
Year × Irrigation × Genotype	94	1.0	0.8
Error	287	1.9	1.3

F-test was adjusted by G-G ε and H-F-L ε for quality (0.73, 0.75) and (0.66, 0.68) for color; *, **, *** significant at 0.05, 0.01, 0.001 probability levels based on adjusted F-test.

). The main effect of year on turfgrass color was not significant. There were significant differences in quality between subjects such as years, irrigation levels, and genotypes. The interaction effects of year × genotype and irrigation × genotype were also significant. The three-way interaction of year × irrigation × genotype did not significantly affect turf quality.

Analysis of within-subject effects over time was also assessed for both turf color and quality. The results showed that there were highly significant differences among data collection time points for both color and quality traits (

Table 5. Repeated measures analysis of variance for univariate tests of hypotheses for within-subject effects over time for turf color and visual quality.

Source of Variation	DF	Mean Square
		Quality	Color
Time	12	34.4 ***	198.9 ***
Time × Year	12	60.5 ***	72.61 ***
Time × Irrigation	24	18.9 ***	12.1 ***
Time × Replication	12	0.9 *	1.6 **
Time × Genotype	564	1.6 ***	2.0 ***
Time × Year × Irrigation	24	2.8 ***	6.2 ***
Time × Year × Genotype	564	0.8 ***	0.8 ***
Time × Irrigation × Genotype	1128	0.6 ***	0.7 ***
Time × Year × Irrigation × Genotype	1128	0.4	0.4
Error (Time)	3444	0.4	0.5

F-test was adjusted by G-G ε and H-F-L ε for quality (0.73, 0.75) and (0.66, 0.68) for color; *, **, *** significant at 0.05, 0.01, 0.001 probability levels based on adjusted F-test.

). The time interaction effect with year, irrigation, genotype, and the two-way interaction effects were also significant for both traits. However, the three-way interaction of time × year × irrigation × genotype was not significant for either of the traits.

3.5. Performance of Hybrids Under Deficit Irrigation Based on Rank Summation Index

The results of the analysis of variance and mean separation procedures revealed that deficit irrigation significantly affected the overall mean performance of the hybrids. Consequently, it is paramount to evaluate the performance of these hybrids relative to the check cultivars under deficit irrigation conditions. This evaluation aimed to identify hybrids with more efficient use of water, as reflected in their spring green-up, turf density, greenness, and overall quality.

Mean green-up scores with the largest deficit irrigation (0.4ET_o) were greater for seven hybrids (OSU2107, OSU2104, OKC1873, OSU2035, OSU2073, OSU2101, OSU2102) than with moderate deficit (0.6ET_o) and full (0.8ET_o) irrigation levels, probably attributed to lesser thatch accumulation from the previous years in this irrigation treatment. These seven hybrids outperformed TifTuf (7.5) in green-up scores. Six hybrids, including OSU2018, OSU2021, OSU2105, OSU2106, OSU2109, and OSU2116, demonstrated superior mean density scores compared to TifTuf (5.3) under the largest deficit irrigation. The mean greenness scores of OSU2105, OSU2109, OSU2073, OSU2106, OSU2110, and OSU2104 were equal to or exceeded that of TifTuf (6.5) under the largest water deficit irrigation. Most of the hybrids showed little variation in their greenness scores across the three irrigation levels (LSD_0.05 = 0.06). With the largest deficit (0.4ET_o), OSU2105, OSU2073, OSU2109, OSU2018, OSU2106, OSU2116, OSU2104, OSU2081, and OSU2082 achieved mean turf quality scores surpassing TifTuf. Six of these hybrids, OSU2073, OSU2109, OSU2018, OSU2116, OSU2081, and OSU2082, maintained similar mean quality scores with moderate deficit (0.6ET_o) irrigation.

The average greenness score of the hybrids under different irrigation levels indicated a general decline in color score starting in mid-September (MSepC) irrespective of the irrigation treatment (Figure 3a). The late June score (LJunQ) showed no variation among irrigation levels, which may be attributed to favorable rainfall in 2022 (Figure 1). Average turf quality scores under different irrigation levels at specific time intervals over the two years showed that the impact of largest deficit irrigation (0.4ET) became noticeable towards the end of the summer season (Figure 3b). The late June score (LJunQ) showed no variation between the irrigation levels and may be attributed to adequate rainfall in 2022 (Figure 1 above). Both turf color and quality vary over the seasons for all the three irrigation levels. Overall, the 0.8ET irrigation showed a better visual quality, followed by 0.6ET, at critical evapotranspiration rates in the middle of the summer. In most of the evaluation time points, low color and quality scores were observed for the 0.4ET irrigation.

4. Discussion

4.1. Turfgrass Water Use

In the current study, the performance of 46 turf bermudagrass hybrids were assessed under three levels of irrigation treatments managed to replace 40%, 60%, and 80% ET_o. Two levels of deficit irrigation treatments were implemented to investigate the level of drought tolerance among bermudagrass hybrids that optimize water use efficiency without significantly compromising turf quality. The results of green-up, density, color, and quality in this study revealed significant differences among hybrids and irrigation-by-genotype interaction effects for all the traits. The significance of the irrigation x genotype interaction effect implies the prevalence of substantial genetic variation for drought resistance among the hybrids and their efficiency in water resources utilization [5,31]. This considerable genetic variability, partly generated through directional breeding [32], represents a valuable genetic resource that can be harnessed for improving drought tolerance in turf bermudagrass. Although irrigation water quality influences turfgrass growth, considerable variability was observed among bermudagrass hybrids for evapotranspiration rates and turfgrass visual quality [33]. The results demonstrate that careful turfgrass selection for drought tolerance and improved irrigation management can contribute to water conservation in arid environments.

An investigation of the same set of hybrids using active spectral reflectance captured at 670 nm, 730 nm, and 780 nm bands under controlled environment conditions showed significant differences for spectral vegetative indices related to photosynthetic efficiency and chlorophyll content [34]. The six reflectance-based high-throughput plant phenotyping indices are relevant to key physiological properties such as chlorophyll content and photosynthetic area at both the leaf and canopy scales [35]. These physiological differences are the basis for the variation in their field performance under different irrigation regimes. This genetic diversity is a foundation for the development of stress-tolerant bermudagrass cultivars that conserve water without compromising the turf quality of bermudagrass turf system in the desert environment. We observed across the years consistent performance differences among the hybrids tested for deficit irrigation. This is supported by the statistically significant effect of the irrigation × genotype interaction. Notably, the year × irrigation × genotype interaction was not statistically significant, indicating consistent results over years.

4.2. Effect of Deficit Irrigation

A range of performance was observed among the hybrids under different levels of deficit irrigation. With moderate (0.6ET) deficit irrigation, most of the hybrids evaluated maintained acceptable visual quality. However, under the most severe deficit level (0.4ET), most hybrids exhibited reduced turf quality, including discoloration and reduced density. Several hybrids, such as OSU2105, OSU2073, OSU2109, OSU2018, OSU2106, OSU2116, OSU2104, OSU2081, and OSU2082, demonstrated mean turf quality scores surpassing TifTuf, which is currently recognized as one of the most drought-tolerant hybrid bermudagrasses [3,36,37]. These findings align with previous studies indicating that bermudagrass can maintain greenness and functionality while receiving 60–80% of its full water requirements [22]. A study conducted in Tucson, Arizona using a linear gradient irrigation system showed an acceptable quality of bermudagrass under 66–75% × ET_o under accumulated water stress conditions [38]. A two-year deficit irrigation study of bermudagrass in southern California revealed that acceptable-to-minimally acceptable turf quality can be maintained at 63% to 41% ET_o [39]. Consequently, the severe deficit irrigation treatment that replaced only 40% × ET_o was ideal for screening hybrids to assess their tolerance to limited irrigation in the desert environment.

Low-water-use grass species often exhibit characteristics such as slow vertical growth, a prostrate growth pattern, and a dense canopy [40]. In 2022, the hybrids fully recovered by late March across all the three irrigation treatments. With the largest deficit irrigation (0.4 ET), most of the hybrids showed a median regrowth of 6 (1–9 scale), potentially due to less thatch accumulation from limited biomass production during the previous year. There is also variation between the two years that may be attributed to differing weather conditions at the location. Under conditions of plant water stress, factors such as stomatal closure and cuticle formation play crucial roles in controlling water use [1]. A linear relationship between irradiance and water use rate has been documented in turfgrasses [21,41,42,43]. A study on turf bermudagrass evapotranspiration rates under non-limiting soil moisture conditions reported a range of variations among genotypes in water use rates, highlighting the importance of selecting for lower ET rates in breeding for bermudagrass cultivars [12]. They reported that TifTuf exhibited higher ET but a deep root mass that contributed to its high drought tolerance.

Research has demonstrated the efficacy of deficit irrigation strategies for maintaining acceptable turfgrass quality in bermudagrass while conserving water. In south-central Texas, 30% × ET_o was sufficient in maintaining acceptable turfgrass quality of Tifway bermudagrass fairway plots during summer months, in which water was applied three times per week on a fine sandy loam soil [44]. Similarly, in inland southern California, a 75% ET_o replacement irrigation strategy successfully maintained hybrid bermudagrass quality in the dry season [45]. The implementation of deficit irrigation for water conservation in turf bermudagrass necessitates consideration of various management practices. Hejl et al. [11] highlighted that mowing height can significantly influence water use under deficit irrigation conditions. This finding underscores the importance of integrating multiple management strategies to optimize irrigation scheduling while maintaining acceptable turf quality.

These studies collectively demonstrate the potential for significant water savings in bermudagrass management through carefully designed deficit irrigation strategies. However, the specific irrigation requirements may vary depending on local climate conditions, soil type and characteristics, and management practices, emphasizing the need for site-specific irrigation planning and optimization.

4.3. Repeated Measure Analysis

Repeated measure analysis of variance was conducted to assess differences in mean turf quality and color among bermudagrass hybrids at each time interval of visual assessment over two years. Repeated measures analysis of variance between-subject effects for turf color and visual quality data over two years showed highly significant differences (p < 0.001) among irrigation treatments, genotypes, and irrigation-by-genotype interaction effects. These results suggest that genotypes responded differently to irrigation levels within each year. The repeated measures condition occurs when experimental units, treated as the primary independent variable across two or more experimental groups, receive varying levels of experimental conditions and have multiple dependent variable observations collected at several time points [46]. In this study, genotypes served as the primary experimental units and were subjected to three different irrigation levels. Then, the repeated measure analysis assessed the fixed time effect that captured the random effect of the genotypic variation among the hybrids for mean visual quality and greenness differences across seasons.

This analytical approach allowed for a comprehensive evaluation of bermudagrass hybrid performance under different irrigation regimes over time. By accounting for both fixed and random effects, the analysis provided insights into the complex interactions between genotype, irrigation level, and temporal factors influencing turf quality and color. The analysis revealed complex interactions between genotype, deficit irrigation level, and temporal factors affecting turf quality and color. The significant genotype-by-irrigation interaction effect indicated that the performance of the different bermudagrass hybrids is not consistent across all irrigation levels. We observed that most of the hybrids performed well under full irrigation while few showed greater resilience to drought stress in the deficit irrigation. The significant irrigation-by-genotype interactions observed underscore the importance of selecting appropriate bermudagrass cultivars for specific irrigation regimes. This finding has practical implications for turfgrass managers and breeders, as it suggests that optimal water use efficiency and turf quality may be achieved through careful matching of cultivars to local water availability and irrigation practices.

5. Conclusions

This study has revealed the prevalence of extensive genetic variation in turf bermudagrass hybrids for drought tolerance. This rich genetic diversity provides a robust foundation for further improvement and development of new bermudagrass cultivars with enhanced adaptive capabilities. The deficit irrigation approach proved important in identifying drought-tolerant bermudagrass cultivars within a germplasm panel of diverse genetic background. The superior performance of several hybrids under water-limited conditions suggests the involvement of multiple stress tolerance mechanisms. These may include, but are not limited to, enhanced root systems, efficient stomatal regulation, and improved osmotic adjustment capabilities. Further exploration of these mechanisms responsible for drought tolerance will be crucial in developing improved cultivars with high levels of stress tolerance. This research provides valuable insights for breeding programs aimed at developing water-efficient bermudagrass cultivars for water-scarce regions.

The findings of this study have significant implications for both turfgrass breeding programs and water conservation efforts in arid and semi-arid regions. By leveraging the identified genetic variation and understanding the underlying physiological mechanisms, breeders can develop bermudagrass cultivars that maintain high turf quality under reduced irrigation regimes. This approach aligns with the growing need for water-efficient landscaping solutions in regions facing increasing water scarcity.

Future research should focus on elucidating the specific genetic and physiological bases of drought tolerance in the top-performing hybrids. Additionally, long-term field trials across diverse environmental conditions will be essential to validate the stability and consistency of drought tolerance traits. These efforts will contribute to the development of bermudagrass cultivars with superior drought tolerance and other desirable turfgrass characteristics, ultimately leading to more sustainable and resilient turfgrass systems in the Desert Southwest region of the United States.