Irrigation scheduling based on reference evapotranspiration (ETo) multiplied by a crop coefficient (Kc) is an accepted approach for managing and conserving water applied to turfgrass. However, increasing use of recycled water that is often high in salinity warrants further examination of irrigation practices for turfgrass health and salinity management. A 2-yr study was conducted in Riverside, CA to evaluate the response of perennial ryegrass (Lolium perenne L. `SR 4550') turf to varying quality and quantity of irrigation water. A modified line-source gradient experiment was designed to alternate between distribution of potable and saline water to establish an irrigation salinity gradient (EC ~ 0.6 to 4.2 dS m-1) in between lines. Irrigation was scheduled in four separate irrigation zones perpendicular to the irrigation lines according to 80, 100, 120, and 140% ETo. Changes in turf quality (R2 = 0.30***), cover (R2 = 0.26***), and clipping yield (R2 = 0.08***) were primarily driven by the number of days that the area had been irrigated with saline water. When data were separated by irrigation amount, both time and water quality accounted for 54% and 46% of the variability (P < 0.001) in quality and cover, respectively at 80% ETo. Soil salinity (ECe), and sodium absorption ratio (SAR) were highly correlated with irrigation water quality, but not irrigation amount. Stepwise linear regression revealed that soil ECe at 20-30 cm (P < 0.05), and SAR at 10-20 cm (P < 0.001) and 20-30 cm (P < 0.01) accounted for 43% of the variability in quality and cover in August 2012. In September 2012, soil ECe at 10-20 cm (r = -0.62) and 0-30 cm (r = -0.60) had the highest correlation with turfgrass quality and cover. Regression results revealed that soil ECe at 10-20 cm (P < 0.001) and SAR at 10-20 cm (P < 0.05) accounted for 41% of the variability in quality and cover in September 2012. Our results suggest that perennial ryegrass requires irrigation scheduling above 120% ETo, irrigation water quality below ECw ~ 1.7 dS m-1, and soil salinity (ECe) below 3.8 dS m-1 to maintain acceptable quality and cover for over one year in Riverside, CA.

Kikuyugrass is an invasive, warm-season turfgrass (C4) that was introduced into California in the 1920s. Although it is used in many sport, and residential turfgrass areas, very little focus has been on the improvement of the species. A comprehensive project was designed and implemented to address kikuyugrass management, improvement through anther culture, and genetic variation within California and other collections.

To study kikuyugrass management a field trial took place in 2012 and 2013 to identify cultural and chemical practices that are most important for producing quality turf and optimal playing conditions on golf course fairways. A 25 factorial design was used to evaluate mowing frequency, cultivation, trinexapac-ethyl, nitrogen, and fungicide treatments. Turf quality was assessed visually and by an instrument that measures “greenness” or Normalized Difference Vegetation Index (NDVI). Turf firmness and ball roll were measured with a Clegg Soil Impact Tester (2.5 kg hammer Gmax) and Pelz meter, respectively. Treatments of trinexapac-ethyl, heavy verticutting, and more frequent mowing practices produced the highest quality turfgrass by increasing firmness, improving turf quality and color, and reducing scalping. Androgenesis through anther culture was attempted on kikuyugrass in order to develop a less vigorous variety. Androgenesis has been successful on the closely related pearl millet but never attempted on kikuyugrass. Kikuyugrass and later pearl millet were tested with two types of media (C17 and Yu-Pei) combined with two different incubation regimes. No signs of mitosis were seen on any of the observed anthers and no embryos were formed in either species. The results were inconclusive and the possibility of successful androgenesis in kikuyugrass remains an open question.

To assess the presence of genetic variation as an indispensable condition for eventual breeding efforts, genotypes were sampled from collections in California (from around the state and at the University of California, Riverside), Hawaii, and Australia. A total of 20,000 single nucleotide polymorphism (SNP) makers were discovered using the Diversity Array Technology sequencing (DArTseq) platform. The hierarchical plot, gap statistics, and principle coordinate analysis all show that the 336 accessions in the study separated into three main clusters. The analysis shows that the amount of genetic variation is low, with 77% of the total genetic variation due to within population variation, while 23% is due to among population variation. The accessions from California show the least genetic variation with accessions from Australia and Hawaii showing a much broader degree of genetic diversity. Although California showed low genetic diversity, there may be enough variation between the accessions at the University of California, Riverside and those collected in the state to begin breeding a turf-type cultivar in California.

ABSTRACT

Recurrent selection for drought and heat tolerance among hybrids of perennial ryegrass (Lolium perenne L.) with meadow fescue (Festuca pratensis Huds.), was used to develop turf-type populations with a marked increase in stress tolerance. Increased tolerance was associated with the presence of an introgression of F. pratensis chromatin on chromosome 3 of L. perenne. To determine if root characteristics were responsible for the improved stress tolerance, a greenhouse study was conducted to compare sister lines of Festulolium both with or without the introgression; the recurrent backcross parent; a representative F. pratensis; and turf-type tall fescue (Festuca arundinacea Schreb.) grown under well-watered conditions in 5.1-cm diam. x 160-cm long tubes containing sand. Two separate 120-day experiments revealed that F. pratensis produced deeper roots, more root biomass, and a higher root:shoot, while the industry standard, F. arundinacea, ranked at or near the lowest value. For the hybrids and ryegrass, the root parameters were intermediate between the 2 fescues, with no statistically significant difference among the 3 lines tested. The results indicate that in the tested Festulolium turf, drought and heat tolerance were not a consequence of increased root depth or root biomass as reported in previous reports for forage-type intergeneric hybrids of ryegrass and fescue.

Turf is an essential part of urban landscapes and an effective sink for soil organic carbon, yet little is known about the relative carbon fixation (CF) capabilities among various turfgrass species and cultivars and how CF is influenced by irrigation practices. Ten commonly used cool-season and warm-season turfgrasses were evaluated for CF and water use efficiency (WUE) rates under optimal and deficit irrigation practices. Cool- season species consisted of: Kentucky bluegrass and perennial ryegrass (KB/PR); fineleaf fescue (FF); tall fescue and Kentucky bluegrass (TF/KB); and tall fescue (TF). Warm- season species included: zoysiagrass (ZOY); st. Augustinegrass (SA); seashore paspalum (SP); bermudagrass (328B and 419B); and buffalograss (BUF). CF was measured and calculated as gross ecosystem productivity (GEP) along with WUE using an open-path infrared gas analyzer. KB/PR, FF, TF/KB, TF, and SP had the highest GEP under deficit irrigation. When grouped, GEP was significantly higher during recovery under deficit irrigation (p<0.0001) for cool-season compared to warm-season turfgrasses, indicating that C3 photosynthesis is labile and C4 photosynthesis is stabile during recovery. Overall, WUE for warm-season grasses was higher than cool-season grasses (p<0.0001). However, species did not differ in WUE between optimal and deficit irrigation, suggesting that WUE is conserved when water is limited. Our results confirmed that warm-season turfgrasses are the most appropriate species for water conservation in regions where they are adapted.

Methiozolin is a new herbicide from South Korea currently under development for pre- (PRE) and postemergence (POST) control of annual bluegrass in bentgrass and most other cool- and warm-season turfgrasses. Greenhouse studies were conducted in 2012 at the University of California, Riverside, CA, and Auburn University, Auburn, AL, to evaluate the relative tolerance of nine creeping bentgrass (CRBG) cultivars, velvet bentgrass (VBG) and colonial bentgrass (COBG) to methiozolin at 0, 0.6, 1.1, 2.2, 4.5 and 9.0 kg ai ha^-1. Methiozolin was applied 7 weeks after seeding followed by a second application at 12 weeks. Methiozolin rates that produced 25% injury (TI₂₅) and 50% clipping dry weight reduction relative to an untreated control for each species or cultivar (GR₅₀) were calculated using four parameter logistic regression. Results indicated that TI₂₅ rates at 56 DAIT were the most accurate in describing relative tolerance among BG species; as a result CRBG was determined to be more tolerant to methiozolin than VBG or COBG. Respective TI₂₅ rates were 1.1, 0.2, and 0.3 kg ai ha^-1 for CRBG, VBG and COBG. Furthermore, VBG and COBG are unlikely to tolerate sequential applications necessary to control annual bluegrass with methiozolin. Evidence of soil residual activity was documented for methiozolin; therefore, herbicide rates that caused TI₂₅ and GR₅₀ decreased with increasing number of applications. Creeping bentgrass `Penn A-4' was the most tolerant of methiozolin application with TI₂₅ of 4.5 and 2.5 kg ai ha^-1 at 28 and 56 DAIT, respectively. All CRBG cultivars tested tolerated methiozolin at the recommended sequential application rate (0.5 kg ai ha^-1).

Rising global temperatures and the decreasing amount of water available for irrigating crops has become a major concern, especially in the turfgrass industry. These trends highlight the need adopt turfgrasses that are better adapted to both current and future climates. Bermudagrasses (Cynodon (L.) Rich) are warm-season grasses that are well-adapted to warm and dry climates, making them an ideal turfgrass for regions where water may be scarce. The genus also boasts a large range of genetic diversity, which can be leveraged to improve consumer- and agronomic-related traits, such as winter color retention. However, genomic resources are lacking for bermudagrass and distinguishing between species can be difficult given their phenotypic plasticity and morphological variation. This makes it difficult to effectively utilize the genetic diversity available in germplasm collections for breeding. In this dissertation, I first examined the systematics of the Cynodon genus and proposed the reclassification of several accessions collected from publicly available germplasm repositories. Exploratory work was done to identify and develop genetic markers that may aid breeders and taxonomists in identifying some of the bermudagrass species. Next, I evaluated the performance of inter- and intraspecific bermudagrass hybrid accessions under prolonged and repeated drought conditions using a novel method. Results from this study suggest the presence of different types of stress memory among the accessions, suggesting that this method may be useful in both selecting for improved drought tolerance as well as enhancing the drought tolerance of established stands. Lastly, I conducted a genome-wide association study (GWAS) to identify markers associated with winter color retention in a population of bermudagrasses. Despite several limiting factors in the study, such as admixture and differences in ploidy throughout the population, results from the GWAS identified two significant DNA polymorphisms, with one aligning to a gene known to be involved in cold stress response in maize. This dissertation emphasizes the need to develop more genomic resources for bermudagrass in order to select for accessions with improved consumer- and agronomic-related traits.