DROUGHT AVOIDANCE CHARACTERISTICS OF HYBRID BLUEGRASS AND TALL FESCUE CULTIVARS Leonard Githinji, Jacob Dane, Robert Walker, and Edzard van Santen Auburn University, Agronomy and Soils Department, 285 Funchess Hall, Auburn, AL 36849-5412 Table 1. Results from mixed models ANOVA for root mass and root length density (RLD). Four- and three-way interactions were dropped from the model because of P-values ≥ 0.50. RESULTS AND DISCUSSIONS Root Length Density (RLD) and Root Dry Mass (RDM) The RLD over the entire 2-year period was significantly (P = 0.001) affected by cultivar (C), year (Y), and month (M), but not by irrigation (Table 1). The only important (P ≤ 0.15) two-way interactions involving cultivars were the C x M and the C x Y interaction. The results for pair-wise comparison showed that in the 2-yr period the RLD for HB 130 (average RLD = 8.9 cm cm-3) was significantly (P ≤ 0.001) higher than that of any other cultivar except HB 129 (Table 2). In every case the RLD was significantly larger in July than in September (Fig. 1). The RLD was higher during the close-to-normal year (2005) compared to the extreme dry and hot year 2006 (Fig. 2). The RLD ranking using the combined means for the 2-yr period was HB 130 (best) > HB 129 > HB 329 > HB 328 > Green Keeper > KY 31. Drought resistance has been correlated with root length, extensive root system or root length density in field crops and turfgrass. Turf Color Quality For the entire 2-yr period turf color quality was significantly (P = 0.05) different for cultivar (C), year (Y), and week (W). Among the 2-way interactions involving irrigation, only I × C interaction is of interest in the context of evaluating the color response of cultivar. The C × Y and C × W interactions were both significant (P  0.001) indicating that the turf color trait differences among cultivars were dependent on the year and time of year (Table 3). The ranking for turf quality was based on hue, saturation and brightness was HB 329 (best) > HB 328 > HB 130 > HB 129> KY31 > Green Keeper. Combined RLD and turf color quality The ranking based on RLD and turf color quality was: HB 329 (best) > HB 130 > HB 328 > HB 129 > KY 31 > Green Keeper. INTRODUCTION Tall fescue (Festuca arundinacea Schreb.) has been reported to have superior drought avoidance and is able to maintain growth and green color for longer periods compared to many other cool-season species (Sleper and West, 1996). Kentucky bluegrass (Poa pratensis L, KBG) has good turf quality and forms attractive turf when supplied with adequate water but has moderate to low drought resistance (Beard, 1989). The grass usually goes dormant and loses color during periods of high temperature and drought. While tall fescue has good drought resistance, it is not as fine textured as KBG. Furthermore, some turfgrass managers and home owners prefer the finer texture and recuperative capacity that KBG offers (Bremer et al., 2006). Hybrid bluegrasses (HBG) are genetic crosses between KBG and Texas bluegrass (Poa arachnifera Torr., TBG). They have the desirable appearance of KBG and like TBG, may withstand higher temperatures and extended drought without going dormant. Although HBGs have potential for lawns and golf courses, there is little scientific information available about their performance relative to KBG and tall fescue under different climatic stresses or cultural practices (Bremer et al., 2006). Objectives Determining the turf color quality, root length density and root dry mass under different stress regimes. Comparing the performance of the hybrid bluegrasses (HB 129, HB 130, HB 328 and HB 329) with tall fescue cultivars (Green Keeper and KY 31). Source of Variation Degrees of freedom P > F Numerator Denominator Root mass RLD Irrigation (I) 2 16 0.144 0.059 Cultivar (C) 5 160 0.382 <0.001 I × C 10 0.937 0.170 Year (Y) 1 I × Y 0.683 0.001 C × Y 0.373 0.060 Month (M) 207 I × M 0.120 0.437 C × M 0.295 0.027 M × Y 0.056 Table 2. Pair-wise differences for RLD among the hybrid bluegrass and tall fescue cultivars. Differences are given below the diagonal and P-values above the diagonal. Cultivar HB 129 HB 130 HB 328 HB 329 G. Keeper KY 31 Avg. RLD (cmcm-3) 0.083 0.307 0.618 0.001 < 0.001 7.7 1.19 8.9 -0.91 2.10 0.996 0.215 0.093 6.8 -0.69 -1.88 0.22 0.071 0.024 7.1 -1.91 -3.10 -1.00 1.22 0.999 5.8 -2.08 3.27 1.17 1.39 0.17 5.7 METHODS AND MATERIALS Location The location of the experiment was the Auburn University Turfgrass Research Facility, Auburn, AL. Experimental Set-up The experimental design was a 6 by 6 Latin square with six replicates of each treatment combination within an irrigation block Three ET-based irrigation replacements were applied to meet 100%, 80% and 60% of potential evapotranspiration and irrigation was treated as an external variable. Turf Color Quality Determination Turf images were taken with the aid of a digital camera once a week from June through September. The average RGB levels of the digital images were calculated using SigmaScan Pro version 5.0 software (SPSS, 1999). The digital values of RGB were converted to HSB values. Determination of Root-Length Density and Mass Root samples were collected from each plot in July and September. The sampling depth was 0-10 cm below the surface. The roots were stained with Congo red dye. Root-length measurement was done using a Comair root length scanner, calibrated to known lengths and widths of thread. Root-length density (RLD) was eventually calculated by dividing total root length in centimeters by the volume of the sampler in cubic centimeters. Root dry mass (RDM) was recorded after oven drying for 72 h at 80° C. Statistical Analysis Mixed models as implemented in PROC MIXED of PC SAS Version 9.1.3 (SAS Institute, 2006) was used to analyze the response data. For RLD and RDM this involved analysis as a replicated Latin square (cultivars=6, rows=6, columns=6), where irrigation was treated as an environmental factor with three levels. Table 3. Results from mixed models analysis of variance for turf color response variable hue, saturation, and brightness. Source of Variation Degrees of freedom P > F Numerator Denominator Hue Saturation Brightness Irrigation (I) 2 3 0.686 0.129 0.495 Cultivar (C) 5 15 <0.001 I × C 10 16 0.551 0.872 0.461 Year (Y) 1 0.009 0.001 0.122 I × Y 0.126 0.090 0.848 C × Y 585 I × C × Y 0.019 Week (W) 13 I × W 26 0.008 0.005 C × W 65 I × C × W 130 1.000 0.829 W × Y 12 I × W × Y 24 0.007 0.221 C × W × Y 60 CONCLUSIONS The results of this study showed that hybrid bluegrasses performed better in terms of root length density and turf color quality, and would likely withstand drought better than tall fescue cultivars. Furthermore, the authors suggest that hybrid bluegrass cultivars were able to withstand summer heat better than tall fescue cultivars. This is collaborated by visual observations and digital pictures which showed considerable browning for the tall fescue plots, but not for hybrid bluegrasses. REFERENCES Beard, J.B. 1989. Turfgrass water stress: Drought resistance components, physiological mechanisms, and species-genotype diversity. p. 23–28. In Takatoh H. (ed.) Proc. Japanese Soc. Turfgrass Sci., Tokyo, Japan Bremer, D. J., K. Su, S. J. Keeley, and J. D. Fry. 2006. Performance in the transition zone of two hybrid bluegrasses compared with Kentucky bluegrass and tall fescue. Applied Turfgrass Science.10:1094. Sleper, D.A., and C.P. West. 1996. Tall fescue. p. 471–502. In Moser L.E. (ed.) Cool-season forage grasses. Agron. Mon. 34. ASA, CSSA and SSSA, Madison, WI. Fig. 1. Cultivar x month interaction means for root length density. Fig. 2. Cultivar x year interaction means for root length density.