. “We... ‘ 2. w... .14. $2.)...f. . a. J . ’7 t .341... 2.... flui- a7.,,. I 1......) 2.5.196... :1... I 3. . 28.. . £92.... 23.... .22.... z: . .4 .1: vat 4.0 .1... 5.... :3 '94:...) 5.1!... 5.3.... $320!; 1- unignjca‘f‘; .vlisrl. .11}. «I 5.49... {Luv . ”JED-:5: .l 1 vi 3 O. 9”a;...\~p¢.i.‘.l5...u~8. 52.3,“. J... I}. up“! [.1 3!. 1‘9... 3.. ilk-«WK ’3‘!“ $I.i~lfiuauu£flv.wfl .1... .IIX ! to. 11.3.2613) «)Vr. 9.).” . P! r I: 4..- This is to certify that the dissertation entitled GROWING DEGREE-DAYS AS A METHOD TO CHARACTERIZE GERMINATION, FLOWER PATTERN, AND CHEMICAL FLOWER SUPPRESSION OF A MATURE ANNUAL BLUEGRASS [Poa annua var reptans (Hauskins) 11mm] FAIRWAY IN MICHIGAN presented by RONALD NIGEL CALHOUN has been accepted towards fulfillment of the requirements for the Doctoral degree in Crop and Soil Sciences 7L 4/, 74. Major Pr'ofessor’s‘Signature January 14, 2010 Date MSU is an Affirmative Action/Equal Opportunity Employer LIBRARY Michigan State Ul liversity PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 K:/Prolecc&Pres/ClRC/Date0ue.indd GROWING DEGREE-DAYS AS A METHOD TO CHARACTERIZE GERMINATION. FLOWER PATTERN, AND CHEMICAL FLOWER SUPPRESSION OF A MATURE ANNUAL BLUEGRASS [Poa annua var reptans (Hauskins) Timm] FAIRWAY IN MICHIGAN By Ronald Nigel Calhoun A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Crop and Soil Sciences 2010 ABSTRACT GROWING DEGREE-DAYS AS A METHOD TO CHARACTERIZE GERMINATION, FLOWER PATTERN, AND CHEMICAL FLOWER SUPPRESSION OF A MATURE ANNUAL BLUEGRASS [Poa annua var reptans (Hauskins) Timm] FAIRWAY IN MICHIGAN By Ronald Nigel Calhoun Three turfgrass research projects characterized important seasonal events for two mature adjacent fairway populations of annual bluegrass [AB] as they relate to growing degree-days [GDD]. Experiments were conducted between 2001-2006 on a 9 to 15 yr old AB fairway maintained at 1.5 cm at the Hancock Turfgrass Research Center in East Lansing, MI. Environmental data was collected remotely by a Michigan Agricultural Weather Network [MAWN] weather station and verified with on-site measurements. AB seedling emergence [SEM] followed a bimodal pattern with peaks occurring in the spring and fall of each year. A simple growing degree- day [GDD] model accurately predicted mean soil temperature but did not adequately describe rate of SEM. However, logistic regression of mean soil temperature and SEM occurrence verified that SEM in this population fell within published soil temperature optima. Occurrence and rate of flowering of AB were observed for six years. Ordinal dates for onset, peak duration, and completion of flowering were recorded. GDD proved to be a reliable method to predict key flowering events in five of six years. A simple average-calculated GDD model, using a base temperature of -5 C was developed to describe seedhead production (R2=0.65) as compared to previously published model (R2=O.6S). Onset, peak period, and completion of flowering were 550, 800 to 1300, and 1550 GDD-5, respectively. Plant growth regulators [PGRs] may be used to suppress AB flowering. Five one-year studies examined the effects of application timing and PGRs, mefluidide [MF], ethephon [HP], or EP plus trinexapac-ethyl [TE] on suppressing AB flowering. AB seedhead suppression, but not the occurrence of turfgrass injury, could be explained by application timing for MF only. Application dates associated with maximum seedhead suppression from ME were used to identify optimum GDD range. Applications occurring between 350-550 GDD—5 resulted in 30 years) in a putting green will spread vegetatively and produce very few seedheads in a trade-off that favors competition in continuously vegetated environment. The C-S-R model of Grime (1977) attempts to describe the relative importance or tolerance of a species to competitors, stress-tolerators, and disturbance tolerant ruderals, represented by C, S, and R, respectively. Given the description of the AB community, we would place young populations in the lower right region of the triangle. As the population matures it would move toward the upper portion of the triangle (competitors). Selection pressure from microclimate, related to location, site history and cultural practices vary across a golf course property. Using the C-S-R model to represent the total population from a mature golf course would result in a large mass in the center of the triangle. Many researchers have proposed naming conventions to subdivide the AB taxa. Attempts to classify biotypes rely heavily on differences in life span and, to a lesser degree, seed production. Tutin (1957) described four biotypes differing in morphological features, germination, and rate of development. He distinguished between annual and perennial biotypes. Timm (1965) suggested three biotypes: 1) an erect annual biotype designated Poa annua ssp. annua, 2) a prostrate perennial biotype designated Poa annua spp. reptans, and 3) a very tall perennial biotype designated Poa annua ssp. aquatica, which occurs on terrestrial sites directly adjacent to water bodies. In turfgrass systems, only the annua and reptans biotype designations commonly appear. It is widely accepted that there are ranges of morphological and physiological characteristics in both biotypes creating intermediate forms or a continuum of biotypes. Gibeault and Goetze (1973) classified over 30 biotypes of P00 annua L. that could all broadly be grouped as annua or reptans. Biotypes are broadly segregated by climatic region or area of the country. However, microenvironments, management and cultural conditions such as shade, irrigation, mowing height, and compaction will allow biotypes to occur together within the same golf course. Sufficient diversity exists so that AB populations can easily adapt to any set of conditions from un-irrigated roughs to closely maintained putting greens. SEXUAL REPRODUCITON Annual bluegrass is typically characterized as an allotetraploid species (containing two copies each of two different genomes, presumably originating from very different parents) with 28 chromosomes (2n = 28) and is believed to have originated from a natural cross between two diploid species, Poa infirma, an annual species and P00 supina, a creeping perennial (Tutin 1952). However, researchers have not been able to produce annual bluegrass from a successful cross of these species (Nannfeldt 1937, Hovin 1958, Tutin 1952, Koshy 1968). Regardless of their identity, Huff (1999) suggested that the original cross between two diploids parents resulted in a sterile dihaploid hybrid that subsequently converted to the allotetraploid by spontaneous chromosome doubling. The enormous variation that exists within this species is probably a result of such hybridization and doubling events (Vargas and Turgeon 2004). Self-pollination of successive generation results in more stable, true breeding. and uniform strains with half the variability of previous generations. Ellis (1973) estimated out-crossing of annual bluegrass ranging from 0 to 7%. If annual bluegrass is primarily inbred (>95%), survivors in a location will produce more individuals ideally suited for that location. Allard et al. (1968) predicted that genetic variation in a population could be maintained when out crossing levels are as low as 5%. Therefore, he concluded that annual bluegrass would be able to maintain physiological plasticity to adapt to changing environmental conditions. Bradshaw (1972) proposed that environmental and cultural selection acts on seedling populations and results in less variable populations of adult plants. Warwick and Briggs (1978b) confirmed this conclusion finding that seed bank and progeny showed more phenotypic variability than existing surface population. It's easy to see why AB is so well adapted to so many sites. In warm-humid, semi-arid and warm-arid regions, annual bluegrass may behave as a winter annual; while in more northern portions of the cool-humid and cool-arid regions it may behave as a summer annual. In contrast, the perennial creeping types may persist as perennials in closely mowed, irrigated, fertilized turfs in the warmer portion of the cool-humid climate and cooler portions of the warm- humid climate. SEED PRODUCTION Seedhead production data for AB has been collected from multiple continents, under various climates, manmade and natural conditions. In general, AB is associated with a heavy pulse of seed production in the spring. Law et al. (1977) determined that each plant could produce 80 viable seeds per year in a low-density population in England. Lush (1988a) found that plants growing in Australian putting greens produce few viable seeds per year and produce seedheads with fewer florets per spikelet. However, high plant densities on putting greens could still result in production of 150 to 645 thousand seeds per m2. Other research has examined the effects of mowing height and plant density on seed head production. In a study of pasture weeds, Saruckhan (1974) noted that some plants reduce seed production with increased plant density but maintain seed weight. Renney (1964) estimated that a single annual bluegrass plant could produce 360 seeds during the four-month period from May to August in British Columbia. Gibeault (1970), in California, found that annual types (ssp. annua) produce more seed than perennial biotypes (ssp. reptans) and completed flowering in 50 d compared to 81 d, respectively. Danneberger and Vargas (1984) used growing degree-days to predict seedhead emergence. They found that maximum seedhead production occurred over a 14 d period in May in Michigan. They used growing degree-days (GDD13) calculated from 1 April. Peak seed head production occurred between 363 and 433 GDD 13. AB flowers more abundantly at cooler temperatures and shorter photoperiods (juhren et al., 1957). True annual types tend to be day-neutral and photoperiod-insensitive, while some perennial and intermediate biotypes respond to long-day conditions or require cold treatment to flower. Because true annual types are day-neutral, new plants that germinate in the spring will flower out-of- phase with the rest of the population in their first year of growth. If these plants overwinter, vernalization will force spring-biased flower synchronization in successive years. johnson and White (1997) found that vernalization encourages masting, where flowering is synchronized to the spring and promotes vegetative growth during the remainder of the season. Some perennials will continue to flower during the summer. johnson et al. (1993) termed plants that flower throughout the season continual types, and those that flower in the spring or fall seasonal types. However, as might be expected, several populations were considered intermediate. The ability of AB to produce viable seed at greens and fairway height is unique among grasses. New recruits can complete their lifecycle in as little as 55 to 80 d. Seeds can ripen on panicles excised from the mother plant on the same day that pollination occurs (Koshy, 1969). Plants growing at high densities and low mowing heights produce very compact inflorescence, which contributes to inbreeding and the success of future recruits. In young AB populations, vegetative reproduction is secondary to seed production. F ecundity and seed viability decrease over time as perennial biotypes increase. AB populations on established, closely mowed putting green may produce a few sterile seeds or stop producing seed altogether. This has limited researchers ability to breed annual bluegrass cultivars for commercial release (Huff, 1999). Anecdotal evidence indicates how AB invades golf courses. As golfers travel from course to course their equipment (e.g. shoes, clubs, carts) can act as dissemination agents for AB seed and ramets. Golf course superintendents first notice AB populations on the practice putting green and the first tee and green. Imported seed on dirty golf shoes could establish these populations. In addition, golf course maintenance crews routinely mow the practice green before mowing the greens on the course. In this way, the mowing equipment can spread infestations from the practice putting green to the rest of the course. In an attempt to limit course-to-course transfer, some new courses require that patrons have their spikes replaced and shoes cleaned by the golf course staff upon arrival at the course. Still other courses provide bleach bathes in the parking lot for the sterilization of shoes. Natural vegetative spread of well-established populations is very slow. These plants maintain a very low overall growth rate that helps them survive in compacted, shaded, high traffic environments. However, vegetative propagation is possible as coring equipment will spread some removed plugs during aerification. SEEDBANK DYNAMICS Seed from AB can form a major proportion of the weed seedbank in both arable and grassland soils. The seed can remain viable in soil for at least 4 years but losses are greater in cultivated soil. Seed mixed with soil and left undisturbed had declined by 76% after 6 years but in cultivated soil the loss was 92%. In soils sown with autumn crops and plowed annually, the time to 99% decline of AB seed was calculated at 4.3 years with an annual decline rate of 55%. In undisturbed soil the annual loss was 46%. Dry-stored seed was still 98% viable after 3 years. The annual decay rate of annual bluegrass has been reported between 37 and 50%. The relative abundance and shallow placement of AB seeds in turf systems would'seem to be a ready food source. Hutchison and Seymour (1982) report a wide variety of birds and invertebrates feed on AB seed. The seeds are known to survive animal digestion and are frequently found in worm casts. Renney (1964) calculated that a continuous stand of AB could produce 12 million seeds per hectare in the surface layer of soil. This estimate was based on a low-density population in Texas. It is likely that a high-density population could produce a great many more seeds. Branham and Gaussoin (1989) estimated the seed bank in the top 8 cm of soil under an AB turf in Michigan to range from 20,000 to 140,000 seeds per m2. Lush (1988b) characterized the seed bank of several annual bluegrass putting greens. She reported fluctuating levels of seed in the soil between 30,000 and 200,000 seeds per m2. Estimates of seed longevity in the soil range from 1 to 6 years. Most seeds (>80%) germinate or are devitalized within the first year. In undisturbed soils the seed bank declined 72% in 6 years but in cultivated soil the reduction was 92% (Roberts and Feast, 1973). The increased rate of decline in cultivated soils may be related to exposure to light or improve germination conditions. In their study 99 percent decline was calculated at 4.3 years with a mean decline per year of 55%. The seed bank must be considered in any management program. Overseeding putting greens with creeping bentgrass (Agrostis stolonifera) at 7 g m’z, equivalent to 60,000 seeds m‘z, in an attempt to diminish the impact of a robust AB seedbank is likely ineffective. Commercial bentgrass seed is non-dormant; therefore these seeds can only compete with the existing seed bank for a very short time. It is the opinion of this author that there is no long-term contribution to the seed bank from overseeding. SEED DORMANCY AND GERMINATION Seed dormancy is an important part of a weed's strategy for survival. AB seed dormancy seems to be related to the environment in which the seed was produced. Tutin ( 1957) collected seed from annual and perennial biotypes. He found that seeds from annual types had low initial germination (dormancy) while seeds from perennial types exhibited relatively high initial germination (no dormancy). These results are typical in the literature. Lush (1988) observed that seed produced in the fairway or rough were initially dormant and required chilling while nearly 100% of the seed produced on putting greens germinated immediately. In her study, subsequent generations exhibited the same characteristics as the parents. In nearly every case (Lush 1988, Wu et al. 1987, Bewely and Black 1994) either chilling or one year of after-ripening was necessary to produce consistent germination of seed from fairways or roughs. Germination of annual bluegrass is considered to occur in cool, moist conditions in late summer/early fall with a second germination period in the spring in some regions (Beard 1970, Harper 1965). Germination has been reported over an air temperature range of 4.5 to 21 C. Bogart (1972) reported substantial decrease in germination at temperatures above 27 C. Engel (1967) found that alternating day/night air temperatures (30 C/ 20 C) resulted in better germination than constant temperatures at either level. In nearly all studies, seedling establishment increased with increasing air temperatures and longer photoperiods (Johnson and White, 1997a). However, McElroy et al. (2004) found that germination could occur in complete darkness. Among the eight ecotypes studied, only one demonstrated a trend in germination rate associated with increasing photoperiod. In a four-year study, Kaminski and Dernoeden (2007) reported a majority of seedling emergence (SO-70%) in Maryland occurred between late September and mid-October. However, 25% of all seedlings emerged between November and May. They did not observe any major emergence flush in the spring of any year. AB germination occurred when mean daily air temperature was less than 20 C and was stimulated by natural precipitation. Branham (1991) reported high levels of germination in the 10 fall and spring in Michigan and submitted that germination could occur, to a lesser extent, during the entire growing season. 11 CHAPTER TWO CHARACTERIZING GERMINATION EMERGENCE PATTERN OF ANNUAL BLUEGRASS (Poa annua L.) IN AN IRRIGATED FAIRWAY IN MICHIGAN. ABSTRACT Accurate prediction of the germination time of annual grasses is critical to maximize performance of preemergence and postemergence herbicides. Soil temperature optima for germination of annual bluegrass (Poa annua L.) [AB] in controlled environments have been reported between 10 to 30 C. Three one-year observational studies were conducted to determine the effectiveness of growing degree-days (GDD) to estimate soil temperatures as a means to predict annual bluegrass seedling emergence [SEM]; or if GDD in combination with soil temperature, daily air temperature cycling (max-min>10 C), freezing events (air tempsO C), or day length could be used to accurately predict SEM pattern AB in Michigan. Germination of annual bluegrass was monitored on an Owosso sandy loam (fine-loamy, mixed, semiactive, mesic Typic Hapludalfs) from 2001 to 2003 at two locations in East Lansing, Michigan. The trial locations were previously maintained as AB fairways with automatic irrigation for more than ten years, therefore only the naturally occurring seedbank was used for observation. A bimodal SEM pattern was observed in all three years with a spring-timed and fall- timed peak. Growing degree-days accurately predicted soil temperatures from March to October (R2=0.96) with the formula Soil14 = -3.333083E-6*gdd2 + 12 1.840131E-2*gdd + 1.445839E-1, however, soil temperature and GDD did not predict rate of SEM or cumulative seasonal SEM equally in all three years. Linear regression was not appropriate due to the bimodal SEM pattern of this population. Logistic regression of SEM data confirmed iterative conclusions that major emergence periods occurred when 14 d moving average soil temperatures were above 10 C. Two peaks in emergence were observed in each year, however, low levels of emergence continued throughout the warmest periods of each season. AB emergence in this study did not strictly follow the pattern of a winter annual or a summer annual. AB seedling emergence in this, irrigated, 10 to 15 yr old population maintained was best described by considering separate life histories for the spring and fall seedling emergence pulse. The spring pulse, strongly influenced by seed from Poa annua var. reptans, and a fall pulse, strongly influenced by seed from Poa annua var. annua. Perennial plants producing non-dormant seed with high temperature enforced secondary dormancy explain the spring-timed seedling emergence pulse. The fall-timed seedling emergence pulse is explained by annual plants contributing seeds with primary dormancy and released by a high temperature after-ripening period. The spring pulse coincided with or just after the major flowering period of the surface population in each year. By inference, the major seed source, and therefore the seed in the seedbank, did not exhibit primary dormancy. Low levels of seedling emergence observed during the summer can be the attributed to overlapping ranges of dormancy release and enforcement in the two populations. 13 INTRODUCTION Annual bluegrass is historically classified as a winter annual (Warwick, 1979; Uva et al., 1997). Winter annual weeds produce dormant seed in the spring of the year that is released by after-ripening associated with higher summer temperatures for fall germination. Seedlings overwinter and resume vegetative growth in the spring, typically flowering in response to increasing day length and senescing as average daily temperatures continue to increase and the winter annual lifecycle repeats. Environmental conditions in the fall may not be conducive for germination. Even under ideal conditions not all seeds will germinate in the fall. Viable, non- germinated seed of winter annuals exposed to cold winter temperatures enter secondary dormancy where they will germinate only under an ever-narrovving set of parameters. In this way, primary and secondary dormancy work together to improve the chances that winter annual weed seedlings will establish under a favorable environment. Under non-irrigated conditions seed may continue to cycle in and out of secondary dormancy until conditions conducive for germination are experienced during a non-dormant state or the seed is devitalized. Typical response of winter annual weed seed to seasonal temperature cycling synchronizes germination to the fall of the year. As such, preemergence applications for suppressing annual bluegrass germination have typically been targeted in the late summer. Harsh conditions that facilitate spring senescence of winter annuals may be diminished by moderate-to-intense turfgrass management practices. Annual 14 bluegrass in the cool-season regions of North America and Mediterranean climates around the world may exhibit a perennial grth habit in response to increased use of fungicides and the ubiquitous nature of automatic irrigation systems. One of the byproducts of surviving in an environment where moisture is not a limiting factor is selection pressure that favors survival of weeds that produce non-dormant seed. Therefore, AB seed produced under the irrigated conditions of fairways and greens may germinate whenever temperature and other edaphic factors other than soil moisture are non-limiting (Lush, 1988b). This leads to a situation where germination may occur outside the traditional autumn period. Seedlings that emerge in the spring or summer are not controlled by traditional fall-timed preemergence herbicide applications. Predicting AB germination pulses is important for timing aerification. Turfgrass practitioners wishing to increase AB populations should time aerification procedures just prior to germination pulses. Conversely, aerification during the summer creates openings in the turf at a time of the year when germination rate is reduced. Accurately predicting germination and the resultant presence of seedlings is necessary to maximize postemergence herbicide applications targeted at selective removal of AB. One of the main factors modulating postemergence herbicide performance is age of the plant. Younger plants are easier to control. This principle is even more important when trying to control grasses in grasses. The margin of selectivity between the target weed(s) and the crop for active ingredients used for controlling grasses in grasses is often quite narrow. Because overall herbicide 15 activity is maximized on younger plants, it is often possible to use reduced rates of herbicide that increase the margin of safety on the desired turf. Younger plants have a greater percentage of biomass dedicated to meristematic activity, are typically rapidly growing, and lack well-developed underground asexual storage structures. All of these factors improve the performance of postemergence herbicides. Accurate prediction of germination would improve our ability to employ postemergence herbicides for the selective removal of AB. Annual bluegrass germination is most successful between 10 and 15 C (Beard 1980; Standifer 1988ab; Eggens and Ormond 1982; Naylor and Abdala 1982, Dernoeden and Kaminski 2005), but will germinate over a large temperature range. Most germination will occur between 2 and 40 C and decline above and below this range (Koch 1968). Bogart (1972) observed emergence between 4 and 30 C with low levels occurring above this range. Many anecdotal recommendations for timing preemergence and postemergence herbicide applications use calendar-based targets to identify the optimum application window. Calendar-based recommendations are inherently weak, as they did not translate well from year to year or location to location. Local environmental data should more accurately predict annual bluegrass germination patterns and improve herbicide application timing. We hypothesize that the use of public, electronically available climate data from nearby weather stations will improve prediction of AB germination over calendar date and function as a good approximation of soil temperature. This study was conducted to determine if GDD, as an indirect measure of soil temperature, 16 could be used alone or in combination with soil temperature, soil moisture, daily air temperature cycling (max-min>10 C), or freezing events (air tempsO C) to accurately predict AB germination time in Michigan. MATERIALS AN D METHODS Experimental Area Three field studies were conducted during 2001, 2002, and 2003 to determine the seasonal germination pattern of annual bluegrass from an existing seedbank. Plots were established on each of two sites in each year on two annual bluegrass fairways at the Hancock Turfgrass Research Center (HTRC) in East Lansing, Michigan. The two research sites had been maintained as predominantly annual bluegrass fairway since approximately 1983 and 1992 for the west and east location, respectively. Prior to the experiment soil cores from each location were pulverized and topdressed onto sterile media in the greenhouse and placed on a misting bench to facilitate germination to determine the robustness of the existing seed bank. It was determined that both locations contained an adequate reserve of annual bluegrass seed, such that no augmentation of the seed bank was necessary. Soil was an Owosso sandy loam (fine-loamy, mixed, semiactive, mesic Typic Hapludalfs) with a pH of 7.4. Phosphorus (37 ppm) and potassium (76 ppm) levels were supra optimum and low, respectively. Nitrogen was applied with 46-0-0 at 146 kg N ha‘1 in 2001, 2002 and 2003. Records indicate that both locations were core cultivated in the fall of 2000. Plots were mowed at 1.3 cm three days per week throughout the study. Irrigation was routinely applied throughout the observation 17 period to encourage reasonable soil moisture conducive for germination of annual bluegrass seed. Environmental conditions were monitored beginning 1 March because air and soil temperatures prior to this time were too low for annual bluegrass germination. This starting date is one month earlier than the one used for developing an annual bluegrass seedhead emergence model (Danneberger and Vargas, 1984). Soil temperature at each observation date was measured with a Brighton Electronics portable thermocouple probe that determined the average soil temperature from 0 to 5 cm. This data was compared to the soil temperature data collected by a Michigan Automated Weather Network (MAWN) weather station located approximately 5 km from the two study sites. It was determined that the MAWN weather station data approximated the conditions at the HTRC. The MAWN weather station recorded daily maximum and minimum air temperature, daily maximum and minimum soil temperature and rainfall. In addition, weather-based variables were generated from the same data. Weather-based generated variables included growing degree-days, freezing events, and diurnal air temperature cycling. Degree-day accumulation was calculated as follows: DD = 1210," — Tb) (1) where Tm is the mean daily temperature (averaged over all readings), and Tb is the base temperature, and n is the number of days elapsed since 28 February (Scott et al. 1984). 18 Beard et al. 1978, indicated that daily air temperature cycling of at least 10 C is important for annual bluegrass germination. In this study, diurnal events were tabulated as the number of days meeting the 10 C fluctuation criteria proposed by Beard and others (Buhler et al., 1999). Germination Observation Circles Evaluation of AB germination in a dense established turf stand is difficult if not totally impractical. Counting the number of seedlings with observations circles has previously been used to overcome this problem (Carlson 1994). In this study, we used observation circles to approximate AB germination and emergenceAnnual bluegrass emergence was monitored at the HTRC between 2001 and 2003. Two sets of four observation circles measuring 80 cm2 each were established at each location using pistol-triggered hand pump sprayer to apply a 30 ppm concentraion of glufosinate-ammonium [2-amino-4- (hydroxymethylphosyhinyl)butanoic acid] 7 d prior to the first observation date. The observation circles provided a means to simulate microclimate of the surrounding turf while providing a practical way to identify emerging seedlings. The total number of emerged seedlings in the observation circles was used to estimate germination rate. Evaluations and Data Collection Seedlings observations were recorded weekly from March to October in four of the eight observation circles in each year. Observed seedlings were chemically rogued with glufosinate-ammonium after tabulation. On the next evaluation date the remaining circles were evaluated and chemically rogued with glufosinate- 19 ammonium. Germination observations alternated between sets of circles at each site in each year of experiment for the entire season. Seedlings counts from each location for a particular date were totaled. After the final observations date within each year the total germination count from each observation was transformed and reported as percent germination for each observation date. New sets of observation circles were established in each year. The method described above was used for monitoring emerging seedlings in 2002 and 2003. Plant tissue killed from the non- selective herbicide was not removed and initially remained as ground cover, however, mostly bare soil was present after 6 to 8 weeks. Statistical Analysis Germination data did not differ by location and were pooled by each observation date within year prior to analysis. Data from both locations were analyzed using various methods in an effort to meet the objectives of this research. Seedling emergence data for each year was plotted to assess seasonal emergence patterns. Seedling emergence data for each year was also plotted against corresponding weather-based generated variables. Potential model parameters included air temperature, soil temperature, GDD, SDD, daily air temperature cycling, and freezing events. The complete list of candidate environmental data and generated weather-based variables is presented in Table 2.01. Because emerged seedlings were observed every 14 d in a particular set of observations circles it is unknown whether germination occurred equally over the 14 d period or occurred near the beginning or the end of the period. To account for this uncertainty, the 20 weather-based variable 801114 was created to represent the average soil temperature from the 14 d period immediately preceding the observation date. Two GDD base Table 2.01. Candidate environmental data1 and generated weather-based variables for predicting annual bluegrass emergence in and irrigated fairway in Michigan. Name Description air14 14 d average air temperature 301-11 4 14 d average soil temperature (0-2 cm depth) gdd-5 growing degree-day total since 1 March where base temperature equals -5 C gddo growing degree-day total since 1 March where base temperature equals 0 C diurnmet condition satisfied if daily max air - min air >10 C diurn14 14 d count of diummet satisfying >10 C air temperature cycling dI14 14 d average of day length in minutes frzmet condition satisfied if daily min air falls below 0 C fr214 14 d count of frzmet satisfying >10 C air temperature cycling 1Environmental data collected by MAWN weather station located approximately 5 km from study sites. temperatures were considered for inclusion in model development; 10 C, which represents a potentially conservative but widely published base temperature and -S C, a base temperature that may be more sensitive to accounting for biological activity occurring in the spring and fall of the year. This base temperature was selected by determining the smallest coefficient of variation from a series of candidate base temperatures using the method described by Arnold (1959). GDD-5 accurately predicted Soil14 (R2=0.96) with the quadratic formula 21 — Germination Rate -------- Soil_14 30 a 6i 14-d Average Soil Temperature (C) Percent Annual Germination 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 GDD Total from 1 March (base -SC) Figure 2.01 Aggregate annual bluegrass germination, from 2001-2003, 14 (I mean soil temperature, and predicted 14 d mean soil temperature for all observations plotted against GDD accumulation from 1 March. Hancock Turfgrass Research Center (HTRC), located in East Lansing, MI. Soil“ = -0.333083E-6"‘gdd2 + 1.840131E-2*gdd + 1.445839E-1. In Figure 2.01, real and predicted values for 801114 and percent annual germination rate for all years is plotted against GDD-5 totals. Raw and transformed data were unable to generate a response variable that was linear. Therefore, developing linear regression models describe this data was not appropriate. A binary response variable was created for seedling emergence where 0=no emergence observed and 1=emergence observed for each 14 d interval. The binary data set was subjected to PROC LOGISTIC in SAS 22 9.1.2 as a response to 801114 to confirm iteratively determined field-collected soil temperatures associated with AB seedling emergence. The output from this procedure, including the 95 percent confidence intervals, for all observation dates from three years is presented in Figure 2.02. bingerm 1.0 1 :1: III-1H an” wanna-H: mat *ngan...x .d" a." ’- f. " 0.9 i 1 J 0.8 g 0.7 3 0.6 0.5 g 0.4 0.3 .3 0.2 0.2 : 0.0 - “"411” Hz :1: minor-o: )1: no” aunt-bunk It 4* I' T fij 7 Y T 1' fi ‘ U I I U V V ‘ U U I U V Y 0 10 20 30 soil14 1* -I Figure 2.02. Logistic response of annual bluegrass germination and 95 percent confidence intervals where 0=no germination and 1=yes from three-years of observation, plotted against mean 14 (1 soil temperature. Hancock Turfgrass Research Center (HTRC), located in East Lansing, MI. 23 RESULTS AND DISCUSSION Seasonal Annual Bluegrass Emergence In 2001, annual bluegrass seedling emergence occurred between 8 May and 18 September. The total number of emerged seedlings was 384 seedlings 80 cm. A bimodal emergence pattern was observed in 2001. The first emergence peak occurred between 12 june and 24 july and accounted for 56% of the seedling emergence observed in 2001, while the second emergence peak was observed between 21 August and 18 September and accounted for 39% of the total annual emergence (Figure 2.03). Ninety-five percent of the total seedling emergence for 2001 occurred during these two periods representing 12 of the 30 observation weeks. The bimodal emergence pattern is similar to the one described for annual bluegrass in Michigan by Branham (1991). However, the major emergence period in the spring occurred later in the season than what had been previously reported (Carlson, 1994). The second pulse of germination occurred within the Branham's previously reported time frame for Michigan. In 2001, Soil14 ranged from 4.4 Cto 30 C, which is entirely within the range studied by Bogart (1972). N inety-nine percent of seasonal seedling emergence was recorded when Soil14 was between 19.5 C and 28.2 C and diminished greatly when Soil14 fell outside this range. No seedling emergence was observed when Soil“ was less than 14.6 C. In 2002, annual bluegrass seedling emergence occurred between 30 April and 1 October. The total number of emerged seedlings was 348 seedlings 80 cm'z. A bimodal emergence pattern was observed in 2002. However, the majority of the 24 emergence was observed in the spring. The first emergence peak occurred between 30 April and 18 june and accounted for 71% of the total emergence for 2002, while the second emergence peak was observed between 13 August and 21 September and accounted for 20% of the total annual emergence (Figure 2.04). Ninety-one percent ofthe total seedling emergence for 2002 occurred during these two periods representing 14 of the 30 observation weeks. Minimal emergence (<4%) was observed between 9 july and 6 August. In 2002, 501114 ranged from 2.5 C to 24.1 C. One hundred percent of seedling emergence was observed when Soil14 was between 10.6 C and 24.1 C. No seedling emergence was observed when Soil14 was less than 10.6 C. 25 2001 Mean annual bluegrass germination from an irrigated fairway. 24 20-j : _ .9 - E! 16-} <5 1 E : E 81 ‘1’ 1 CL - 0: f I I III 7 I — Air_14 9, 25.“ ........ Soil_14 """" e ——————————————— uh — — — ———————————— 3 S 15% ___________ _ _ _________ ._ _______ o. -- 0E9 . 1— 5! -5_: '15 d 1 1 1 l l 1 1 l 1 F 1 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 2.03. Mean annual bluegrass seedling emergence in a fairway-height turf, reported as percentage of annual total germination as recorded in observation circles, Hancock Turfgrass Research Center (HTRC), located in East Lansing, MI (2001) and associated moving 14 d average air and soil temperature collect by MAWN weather station (~5km). Shaded box represents spring flowering period of associated surface population. 26 2002 Mean annual bluegrass germination from an irrigated fairway. 20; c : .9 . E 16—: g 3 8 12—2 E : 8-. q, - CL 1 4: i i i ii 0: 15H. L.I.I I. —Air_14 = c: (D 5 E Q) Q. E Q) I..— -15i I I I I F I I I I I I Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 2.04. Mean annual bluegrass seedling emergence in a fairway-height turf, reported as percentage of annual total germination as recorded in observation circles, Hancock Turfgrass Research Center (HTRC), located in East Lansing, MI (2 002) and associated moving 14 (1 average air and soil temperature collect by MAWN weather station (~5km). Shaded box represents spring flowering period of associated surface population. 27 In 2003, annual bluegrass seedling emergence occurred between 27 May and 28 October. The total number of emerged seedlings was 137 seedlings 80 cm'z. A bimodal emergence pattern was observed in 2003 with a larger peak in the spring of the year. The spring emergence peak occurred between 27 May and 1 july and accounted for 62% of the total emergence for 2003, while the second emergence peak was observed between 2 September and 23 September and accounted for 20% of the total annual emergence (Figure 2.05). Eighty-two percent of the total seedling emergence for 2003 occurred during these two periods representing 11 of the 30 observation weeks. Seedling emergence observed during the summer period of 15 I july to 26 August and the fall period of 30 September to 28 October accounted for 11 and 8% of the annual emergence total, respectively. In 2003, 801114 ranged from 3.0 C to 26.1 C. Ninety-two percent of seedling emergence was observed when Soil14 was between 15.8 C and 25.7 C. No seedling emergence was observed when Soi114 was less than 9.4 C. A bimodal pattern of annual bluegrass seedling emergence was observed in each year with peaks occurring in the spring and fall. In each year a majority of seedling emergence was observed during the spring germination pulse. Peak germination rate occurred earliest in 2002 (30 April to 4 june) and latest in 2001 (12 june to 26 june). On average, spring annual bluegrass seedling emergence occurred during a 26 d period between 24 May and 19 june, with peak emergence 28 2003 Mean annual bluegrass germination in an irrigated fainlvay. 24_ 20; c I .9 — g 161 g : a) L (D 12: *c‘: : 8: a) : o. _ II I I O- = I In 1.. .I....| I — Air_14 A 25-2 g . a) I 5 15- E 2 3 2 E 5- m . l— 2‘ -5_E -15‘ 1 I I r I l l l I I I Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 2.05. Mean annual bluegrass seedling emergence in a fairway-height turf, reported as percentage of annual total germination as recorded In observation circles, Hancock Turfgrass Research Center (HTRC), located in East Lansing, MI (2 003) and associated moving 14 d average air and soil temperature collect by MAWN weather station (~5km). Shaded box represents spring flowering period of associated surface population. 29 observed on 5 june. The spring emergence peak period accounted for 56, 71, and 62% of total annual emergence for 2001, 2002, and 2003, respectively. The fall annual bluegrass seedling emergence pulse occurred, on average, during a 22 d period between 22 August and 13 September, with peak emergence observed on 2 September. The fall emergence peak period accounted for 39, 20, and 20% of total annual emergence for 2001, 2002, and 2003, respectively. On average, 10% of annual bluegrass SEM occurred outside of the two major emergence periods. The date of first SEM varied by 35 d between all years. Figure 2.06 displays the relative emergence by observation period and cumulative seedling emergence for each year. Germination observation data from each year is summarized in Table 2.02 and 2.03 for the spring-timed and fall-timed SEM periods, respectively. As hypothesized, the germination for annual bluegrass followed a bimodal pattern in each season with a spike in germination in the spring and fall of each year. However, the spring germination pulse was somewhat later than previously reported. Growing degree-days accurately predicted Soil14 throughout the growing season but did not accurately predict the onset of AB seedling emergence and did not describe the bimodal response. Fitting simple regression to the data was not appropriate as the response was not linear. Seedling emergence rate in this research was somewhat more variable that what has been reported in the literature. However, nearly all published work has been done in controlled environment studies where artificially seedbanks are incubated for 40 to 50 d at constant or diurnal temperatures. AB seed from the observed populations in this research may 30 —l— Percent Germ + Cum. Germ Percent Emergence Cumulative Emergence P60 -40 —20 0 Apr May Jun July Aug Sept Oct Figure 2.06. Mean and cumulative percentage annual bluegrass emergence recorded in observation circles in irrigated fairway-height turf, Hancock Turfgrass Research Center (HTRC) (2001, 2002, 2003), located in East Lansing, MI. 31 Table 2.02. Day of year, duration, and percent emergence occurring during spring-biased emergence period from two fairway-height locations, Hancock Turfgrass Research Center (HTRC), East Lansing, MI (2001-2003). Spring Emergence Pulse Duration Seedling Emergence Year _ . d 0/ Beglnnlng End I ays) I 0) 2001 12 jun 24 jul 42 56 2002 30 Apr 18 jun 49 71 2003 27 May 1 jul 35 62 Table 2.03. Day of year, duration, and percent emergence occurring during fall-biased emergence period from two fairway-height locations, Hancock Turfgrass Research Center (HTRC), East Lansing, M1 (2001-2003). Y Spring Emergence P “159 Duration Seedling Emergence ear 0 Beginning End (days) (/0) 2001 21 Aug 18 Sep 28 39 2002 13 Aug 21 Sept 39 20 2003 2 Sept 23 Sept 21 20 well germinate at lower temperatures than were reported if subjected to the same incubation treatments as other controlled environment studies. It is clear that AB germination and corresponding SEM can occur over a wide range of temperatures. Two major emergence periods were observed in the spring and fall of each season, accounting for 95, 91, and 82% of seasonal emergence for 2001, 2002, and 2003, respectively. Low levels of SEM occurred throughout the season. Two peaks in emergence were observed in each year, however, low levels of emergence continued throughout the warmest periods of each season. AB 32 emergence in this study did not strictly follow the pattern of a winter annual or a summer annual. Considering separate life histories for seed contributing to the spring-timed and fall-timed SEM pulse may explain the bimodal SEM pattern observed in this research. Winter annual ‘colonizer' types contribute disproportionately more seed to the seedbank in years closest to establishment (long-lived seed with cold-temperature-enforced dormancy). Regular mowing, frequent irrigation, and routing fungicide use provide intense selection pressure for perennial biotypes. Early-competitive types will contribute seed that is better adapted to irrigation, fertilizer, fungicides, and mowing and result in a temporal change in relative abundance between annual and perennial biotypes. Competitive perennial types may contribute non-dormant seed with high-temperature-enforced secondary dormancy. The temporal change in the surface population will affect the relative contribution of varying AB biotypes to the seedbank. The seedbank will be much more diverse than the surface population due to differences in dormancy, after-ripening, seed longevity, and relative fecundity of the contributing biotypes of the ever changing surface population. It seems plausible that seed contributed from competitor biotypes unequally influences the spring-timed SEM pulse while seed contributed to the seedbank by colonizers, either from early Site history or immigration sources, unequally influences the fall-timed SEM pulse. The SEM rate in the summer is intermediate. Intermediate seed and some seed produced by Spring germinated competitor types that is also non-dormant could very well contribute to summer germination. Low levels of SEM during warm temperatures could be partially explained as the intersection of several life histories. 33 The spring germination pulse is strongly influenced by seed from Poa annua var. reptans, whereas the fall germination pulse is strongly influenced by seed from Poa annua var. annua. Perennial plants producing non-dormant seed with high temperature enforced secondary dormancy explain the spring-timed seedling emergence pulse. The fall-timed seedling emergence pulse is explained by annual plants contributing seeds with primary dormancy and released by a high temperature after-ripening period. The spring pulse coincided with or just after the major flowering period of the surface population in each year. By inference, the major seed source at this site, and therefore the predominate seed in the seedbank, did not exhibit primary dormancy. Low levels of seedling emergence observed during the summer can be the attributed to overlapping ranges of dormancy release and enforcement in the two populations. Future research could attempt to model the spring and fall-timed emergence periods separately. There may be value in trying to describe the aggregate SEM pattern with non-linear regression. Preliminary examination of nonlinear regression with PROC NLIN in SAS 9.1.2 identified soil temperature, daily air temperature cycling, and freeze days through forward parameter selection using a grow curve model. The assumptions of the growth curve are unknown and may or may not be appropriate for describing AB seedling emergence. It may be possible to use non-linear regression to describe the bimodal pattern of AB seedling emergence. 34 CHAPTER THREE USING LOCAL WEATHER STATIONS TO GENERATE GROWING DEGREE-DAY DATA TO PREDICT THE FLOWERING PATTERN OF A PEREN N IAL ANNUAL BLUEGRASS (Poa annua L.) FAIRWAY IN MICHIGAN ABSTRACT Six one-year observational studies were conducted from 2001 to 2006 at the Hancock Turfgrass Research Center in East Lansing, Michigan to characterize the timing, duration, and amplitude of annual bluegrass [AB] (Poa annua L.) seedhead emergence in a 10 to 15 year old AB fairway. The objective of this research was to collect data that could be used in the development of a growing degree-day model to predict AB seedhead emergence at other locations using readily available weather station data. New GDD models were compared with previously published models. Plots were established on two adjacent perennial stands of AB maintained at 1.5 cm, receiving 0.5 to 0.6 cm automatic daily irrigation throughout the growing season and 120 kg N ha'1 yr'1 from 2001 to 2006. Soil type was a Mariette sandy loam (fine-loamy, mixed, mesic Glossoboric Hapludalfs). Line intersect and visual estimation of seedhead cover was evaluated multiple times throughout the spring seedhead emergence period. A base temperature of -5 C most accurately predicted onset, peak duration, and completion of the AB seedhead emergence period for all five years. The final model, flowering rate = -3.331599E-6"‘gdd2 + 6.968782E-3*gdd + -2.841894, accurately predicted (R2=0.64) flowering stages of an AB fairway turf over six years in Michigan. 35 INTRODUCTION Annual bluegrass [AB] is often maintained as a component of or as the primary species of intensely maintained turfgrass surface populations, particularly on golf courses, in many cool-humid climates. Profuse annual seed production contributes to AB's unique competitiveness in turfgrass. A given surface population of AB consists of a continuum of biotypes with divergent life strategies. Winter annual-type, colonizing, r-strategist biotypes termed Poa annua var. annua germinate in the fall, flower in the spring, and senesce in early summer. Perennial- type, competitive, K-strategist biotypes termed Poa annua var. reptans tend to dominate over time in situations that favor continual growth. Perennial biotypes produced non-dormant seed. Progeny flower at maturity in the first year of growth. In subsequent years, vernalization synchronizes the flowering pattern of the surface population to the spring of the year (johnson and White, 1997a). The result is that annual and perennial biotypes set seed in the spring. Understanding the onset, peak duration, and completion of AB seedhead emergence is important in order to properly time certain cultural practices like clipping collection, preemergence herbicide application, cultivation practices, and to evaluate performance of chemical seedhead suppression treatments. Seedhead production data for AB has been collected from multiple continents, under various climates, manmade and natural conditions. In general, AB is associated with a heavy pulse of seed production in the spring. Gibeault (1970), in California, found that annual types (ssp. annua) produce more seed than perennial biotypes (ssp. reptans) and completed flowering in 50 d compared to 81 d, 36 respectively. AB flowers more abundantly at cooler temperatures and shorter photoperiods (juhren et al., 1957). True annual types tend to be day-neutral and photoperiod-insensitive, while some perennial and intermediate biotypes respond to long-day conditions or require cold treatment to flower. Because true annual types are day-neutral, new plants that germinate in the spring will flower out-of- phase with the rest of the population in their first year of growth. If these plants overwinter, vernalization will force spring-biased flower synchronization in successive years. johnson and White (1997a) found that vernalization encourages masting, where flowering is synchronized to the spring and promotes vegetative grth during the remainder of the season. Some perennials will continue to flower during the summer. johnson et al. (1993) termed plants that flower throughout the season continual types, and those that flower in the spring or fall seasonal types. However, as might be expected, several populations were considered intermediate. The ability of AB to produce viable seed at greens and fairway height is unique among grasses. AB may complete its lifecycle in as little as 55 to 80 d. Seeds can ripen on panicles excised from the mother plant on the same day that pollination occurs (Koshy, 1969). Plants growing at high densities and low mowing heights produce very compact inflorescence, which contributes to inbreeding and the success of future recruits. It is no wonder that seed production plays such an important part in AB'S ability to colonize new sites. Turfgrass managers object to seedhead production of AB as it relates to interference with surface uniformity during prolific seed production. Much of the enjoyment and use of turfgrasses is related to their aesthetic value. Seedhead 37 production of AB diminishes the aesthetic value of the turf through visual interference. During seedhead production, plants divert a tremendous amount of energy from other plant processes, like rooting and production of defensive compounds, into flowering. In this way, heavy seed production during the spring may reduce overall plant vigor, predisposing annual bluegrass to impending summer stresses such as heat, drought, disease, and traffic. In situations where annual bluegrass is the primary component of the surface population, suppressing seedhead production may be key to improving overall plant vigor prior to summer stresses. There is a paradox related to AB seedhead suppression. Unregulated populations of AB will produce a tremendous amount of seed, building up the soil seed bank, well positioning AB for resource capture when future voids are created. These plants expend energy on seedhead production and are often predisposed to impending summer stress and are likely to die, creating voids in the turf that are readily filled by their progeny. Inguagiato et al. (2008) offers evidence of improved AB robustness when flowering is suppressed, Showing that AB develops less anthracnose in the summer when treated with PGRs in the spring. Gilmore and Rogers (1958) suggested using heat accumulation units, or growing degree-days, to model developmental stages of corn. There are numerous examples throughout the literature of GDD models developed for field and horticulture crops, insect development, and weed emergence and development. Arnold (1959) presents methods for determining the appropriateness of various base temperatures for use with simple average-calculated GDD models. The method was used by Fidanza et al. (1995) for determining the proper base temperature for a 38 model to predicit smooth crabgrass (Digitaria ischaemum) seedling emergence in Maryland. Danneberger and Vargas (1984) used growing degree-days to predict AB seedhead emergence in Michigan. They used a base temperature of 13 C to calculate GDD from 1 April. In their model, GDD were calculated based on a method proposed by Baskerville and Emin (1969). The Baskerville-Emin [BE] method uses a sine curve as an approximation of the diurnal temperature curve. The somewhat complicated math involved with this method has limited its used by turfgrass managers. Danneberger and Vargas (1984) reported that maximum seedhead production occurred over a 14 d period in May. Peak seed head production occurred between 363 and 433 GDD 13. Their proposed model was seedhead number = -6.249.4 + 41.099*gdd + 0.08573"‘,gdd2 + 5.75e-5"‘gdd3 with a coefficient of determination of 0.65. In 2001, this model was compared to real seedhead emergence data from one of the two locations used by Danneberger and Vargas in 1982 and 1983. The AB seedhead emergence period in 2001 occurred within the same calendar dates during May as previously reported, but the predicted occurrence was 35 d later. Danneberger and Vargas (1984) started GDD accumulation on 1 April as this date corresponded to snow-free turf being first present. In 2001, snow-free turf was present much earlier than 1 April. This research was conducted to 1) validate previously published GDD model for AB seedheads, 2) determine if simple average-calculated GDD model could be used to predict AB seedheads, and 3) compare simple average-calcuated GDD model with BE calculated model. 39 MATERIALS AND METHODS Site Description Seedhead emergence data ofAB fairways were collected from three 1.2 x 2 m plots at two Sites in six one-year studies from 2001 to 2006 at the Hancock Turfgrass Research Center in East Lansing, Michigan, USA. Study sites were 9 to 15 year-old AB fairways maintained at 1.5 cm, receiving 0.5 to 0.6 cm automatic irrigation throughout the growing season and 120 kg N ha‘1 yr'l. Soil type was a Mariette sandy loam (fine-loamy, mixed, mesic Glossoboric Hapludalfs). Percent seedhead cover was determined by visual estimation for each plot throughout the seedhead production period, which ranged from 30 April to 17 june. Visual estimation was compared with binomial coordinate mapping (grid method) in 2001 and 2002. It was determined that a trained rater could accurately determine percent cover by whole plot visual estimation. The accuracy and speed of this method facilitated multiple seedhead rating dates in each year. Daily maximum and minimum air temperatures were collected by a Michigan Agricultural Weather Network [MAWN] weather station located approximately 5 km from the study site. Two GDD methods were calculated from 1 March using the weather station data. Degree-day accumulation for the simple average method calculated as follows: GDD = 20... — Tb) (1) where Tm is the mean daily temperature (averaged over all readings), and Tb is 40 the base temperature, and n is the number of days elapsed since 28 February (Scott et al. 1984). The previously published method by Danneberger and Vargas (1984) was altered to begin calculations on 1 March instead of 1 April. GDD Seedhead emergence data did not differ by location and was pooled by observation date within year prior to analysis. Key seedhead emergence events (onset, peak duration, and completion) were identified for each year. Data for onset and peak flowering were used to select optimum base temperature with the method described by Arnold (1959). He warned that base temperatures set too low would overestimate biological development particularly as temperatures increase while base temperatures set too high would not be sensitive biological development during cooler periods. Arnold concluded that base temperatures that are too high are often reported. Selecting a base temperature with the lowest coefficient of variation in GDD should result in more reliable, repeatable accumulation of GDD from year to year. Fidanza et al. (1995) used the Arnold method to select an appropriate base temperature for developing a GDD model for smooth crabgrass germination in Maryland. Yang et al. (1995) proposed mathematical formulae to simplify the process of calculating the least standard deviation in days to determine appropriate base temperatures in GDD models. Their method was adapted here by substituting observation years for planting dates and is presented here: Standard deviation in days is defined as SDD 51),, = —8d—" (2) T — X 41 where SDday is standard deviation in days, Sngd is standard deviation in GDD, T is the overall mean temperature ofall [years], and x is the base temperature. By taking the derivative ofSDday, the equation / Ill dSDday 2:1[f,(x)-f(x)]2 f 7 = (n -1)(T - x)2 (3, \ J / is obtained, where Tis the overall mean of temperature of all [years] and n is the number of [years]. Let Eq. (3) be zero and the base temperature can be calculated with the equation (2:1tidi)z - 112:1 tizdiz glidizt‘ _ 2f=1tidizf=idi (4) where di is the number of days required to reach a developmental stage for the x=T- n ith [year] and ti is the difference of the overall mean of temperature in all [years] and the mean temperature of the ith [year]. Seasonal and combined data were plotted against GDD as a continuous variable and analyzed with PROC REG in SAS 9.1.2 for development a simple average-calculated model and for comparison with previously reported model. Model equations for each year were generated with PROC MIXED with GDD included in the RANDOM statement. 42 RESULTS AND DISCUSSION Date and ordinal day values for onset, peak duration, and completion of major flower production events for 2001—2006 are presented in Table 3.01. AB seedhead production occurred simultaneously at both locations in all years but onset, peak duration, and conclusion of flower production varied from year to year. On average, the seedhead production period occurred over a 40 d period between 2 May and 10 june with peak seedhead emergence occurring during a 14 d period between 13 May and 26 May. Table 3.01 Date and ordinal day values that characterize the occurrence and duration of flowering period from 2001-2006 of an annual bluegrass fairway in Michigan. Hancock Turfgrass Research Center, East Lansing, MI. Year Date Ordinal Dayi Onset Peak End Onset Peak End 2001 30 Apr 235M112; 12 jun 120 135' 163 2002 7 May 22 [my 19 jun 127 1145? 170 2003 2 May 1217:6137 17 jun 122 111’; 168 2004 4 May gm}; 8 jun 125 11136' 160 2005 10 May 1; my 10 jun 130 11351 161 2006 3 May [38’4”]; 23 May 123 1123?; 143 iOrdinaI day calculated from 1 ]anuary. Values for 2004 reflect leap year. 43 In all years, seedhead production was characterized by a rapid increase, a peak production period, and a rapid decline in seedhead production (Figure 3.01). 100 Year _ 2001 80 - ——— 2002 E) _ —--- 2003 a; 1 3 60 ................ 2004 o _ E _ __..__ 2005 fi _ $ ‘ —~— 2006 E 40 9.3 : o a 1 2C Qj"'l""l"''I“"I""I"r'I"'r 121 128 135 142 149 156 163 1"0 Ordinal Day Figure 3.01. Fairway height annual bluegrass (Poa annua var. reptans) seedhead production from 2001-2006 plotted as percent of peak flower production vs. ordinal day since 1 Ian. Hancock Turfgrass Research Center, East Lansing, MI. Data from all 6 yr were used to determine base temperature of the GDD model for predicting flower production rate of AB. On a year-by-year basis, growing degree- day totals calculated from average air temperatures collected approximately 5 km from the study site provided coefficients of determination ranging from R2=0.51 to 0.86. The base temperature used was -5 C because it coincided with the lowest CV for onset of seedhead production In GDD (Figure 3.02). The optimum base temperature determined by the Yang et al. (1995) method was -5.1 C and -5 C for prediction of onset and peak flowering, respectively. This base temperature (-5 C) is 44 much lower than what has previously been used for AB. Bogart (1972) reported that AB reinitiates grth at temperatures above 13 C. Therefore, Danneberger and Vargas (1984) used 13 C in developing a seedhead model for AB. Danneberger et al. (1987) used 13 C and Branham (1989) suggested that 10 C or 13 C could be used for timing application of plant grth regulators for suppressing flower production of AB. 4.5 I \ \\ , i yfi/ -16 -14 ~12 -10 -8 -6 -4 -2 0 Candidate Base Temperature (C) Coefficient of Variation in Degree-Days Figure 3.02. Coefficient of variation in growing degree-days for candidate base temperatures measured by remote weather station ~5 km from study site from 2001-2006. Seedhead numbers increased rapidly from onset to peak production in all years. On average, onset, peak period, and completion of flowering occurred at 550, 800 to 1300, and 1550 GDD-5, respectively (Figure 3.03). Flowering pattern of AB followed a similar pattern for calendar date and GDD. The model relating GDD to 45 percent peak flowering was: flowering rate = -3.331599E-6"‘gdd2 + 6.968782E-3*gdd + -2.841894 Generated model equations were plotted against GDD for each year and are presented in Figure 3.04. The model provides a reasonable fit all years, except 2006, when major flower production lasted only 22 d. 46 80 1.11 Percent Peak Flowering .p. o 20- O 400 years plotted as percent of peak flowering v. growing degree-day t 1 r . 1 . . . . 1 . . . . 1 . . 600 800 1000 . . I . . . 1200 GDD Total (base -SC) Figure3.03 Annual bluegrass (Poa annua L.) seedhead production from six 1600 accumulation (base -5 C) using simple average calculation method from 1 March. Hancock Turfgrass Research Center, East Lansing, Michigan. 100 80- O) O Percent Peak Flowering 8 20‘ Research Center, East Lansing, Michigan. - 1 ' ' ' I ' v ' I 1 600 800 1000 . I . 1 200 GDD Total (base 60) Figures 3.04 Predicted annual bluegrass (Poa annua L.) seedhead production from six years plotted against growing degree-day accumulation (base -5 C) using simple average calculation method from 1 March. Hancock Turfgrass 1600 Year 2001 2002 2003 2004 2005 2006 On average, onset, peak duration, and completion of flowering occurred at 50, 60 to 140, and 200 GDD 13, respectively (Figure 3.05). The plot of percent ofpeak flower production against GDD13 does not have the same shape as when plotted against ordinal day or GDD-5. With the higher base temperature, fewer GDD accumulate at or near the onset of flowering, however, warmer temperatures toward the end of May leads to more rapid GDD accumulation during and after the peak flowering period giving the appearance ofa less rapid decline in seedheads. The non- symmetrical shape of the GDD13 plot explains why Danneberger and Vargas (1984) used a third-order equation to describe seedhead number (R2=0.65). In this research, quadratic and third order models resulted in coefficients of determination of 0.45 and 0.65, respectively. However, their target range of 363 to 433 GDD13 did not coincide with onset, peak duration, or conclusion of flower production in the studies conducted between 2001 and 2006. Their observed peak seedhead period occurred between 13 May and 29 May. Peak flower production in 2001 to 2006 occurred during the same calendar dates but did not correspond with the target GDD range. In 2001 to 2006, 363 to 433 GDD13 occurred, on average, during a 9 d period between 1]ul and 9qu, nearly two months after peak flowering was observed. Generated quadratic model equations were plotted against GDD13 for each year and are presented in Figure 3.06. Using a quadratic equation with GDD13 accurately predicts onset and conclusion of the flowering period. However, because 48 100 Year < 2001 80 —— 2002 E --—2003 %60- ........... 2004 E . g _ ——2005 E I — 2006 E40 ‘D _ 8 o a. 20 0....,....,. ......,... 0 50 100 150 200 250 GDD Total (base 13C BE) Figure 3.05 Annual bluegrass (Poa annua L.) seedhead production from six years plotted as percent of peak flowering v. growing degree-day accumulation (base 13 C) using Baskerville Emin method from 1 March. Hancock Turfgrass Research Center, East Lansing, Michigan. 100 : Year ‘ . - 1 -— 2001 80 ,79\ m ‘ , / g \ -—‘ 2002 E ‘. - - - - 2003 as - ' //_\ 1 g 60 , ’,’ \. ............... 2004 Lg : — —- 2005 6‘3 I — --— 2006 E 40 g . G) l 20' ‘ \‘ ‘\ . \ 1 \\ O ' ' ' I ' ' ' ' I ' ‘ ' ' I ' ' ' ‘ I ' ' fir O 50 200 2510 100 150 GDD Total (base 13C BE) Figure 3.06 Predicted annual bluegrass (Poa annua L.) seedhead production from six years against growing degree-day accumulation (base 13 C) using Baskerville Emin method from 1 March. Hancock Turfgrass Research Center, East Lansing, Michigan. 49 the equation results in a symmetrical curve, the model underestimates occurrence of peak flower production in four of six years. The quadratic model would not be appropriate for predicting peak flower period for GDD13. The use of a third order model, like Danneberger and Vargas (1984), is necessary with GDD13 in order to predict peak flowering period. CONCLUSIONS Occurrence and rate of flowering of AB were observed for six years. GDD proved to be a reliable method to predict key flowering events in five of six years. Flowering pattern of AB followed a similar pattern for calendar date and GDD-S. A set of quadratic models to explain AB flower production for ordinal day, GDD-5, and GDD13 provided coefficients of determination of 0.54, 0.65, and 0.45, respectively, over Six years. Prediction with GDD 13 was improved by using a third order model (R2=0.65). The additional term did not improve fit for ordinal day or GDD-s. Onset, peak period, and completion of flowering were 550, 800 to 1300, and 1550 GDD-5, respectively. Peak flower production of this AB population had previously been reported between 363 to 433 GDD13. However, in this research, peak flower production occurred between 60 to 140 GDD 13 in all six years. 50 Determining least standard deviation in days for key developmental stages of AB flower production was useful in identifying a base temperature for use in simple average GDD calculations. Using the simple average GDD calculation method and a base temperature of -5 C provided accurate estimation of onset, peak duration, and conclusion of AB flower production. Calculating GDD from air temperature data collected from local weather stations could lead to the expanded use of GDD over geographical regions. Base temperatures, start dates, and target ranges developed in Michigan, may not be appropriate for other parts of the country, particularly where AB does not enter cold temperature induced dormancy. 51 CHAPTER FOUR MAXIMIZING SEEDHEAD SUPPRESSION 0F POA ANNUA L. IN MICHIGAN BY USING GROWING DEGREE-DAYS TO PREDICT OPTIMUM APPLICATION TIMING FOR MEFLUIDIDE, ETHEPHON. OR ETHEPHON PLUS TRINEXAPAC-ETHYL. ABSTRACT Five one-year studies were conducted in 2002 to 2006 at the Hancock Turfgrass Research Center in East Lansing, Michigan, to evaluate the chemical seedhead suppression of annual bluegrass [Poa annua L. var. reptans (Hauskins) Timm.] by single applications of the PGRS mefluidide [MF] (N-[2,4—dimethyl-5- [[(trifluoromethyl) sulfonyl]amino] phenyl]acetamide), ethephon [EP] (2- chloroethylphosphonic acid), or EP plus trinexapac-ethyl [TE] (4-(cyclopropyl- alpha-hydroxymethylene)-3, 5-dioxo-cyclohexanecarboxylic acid ethyl ester) made over a series of dates in the spring to determine if optimum application timings existed. The objectives of these field trials were to; 1) compare MF with EP or a tankmix of EP plus TE, 2) determine effects of application timing on EP or HP plus TE, 3) evaluate potential turfgrass injury associated with PGR application, 4) evaluate the accuracy of a simple average calculated GDD model (base -5 C) to predict best application timings and, 5) remotely predict best application dates using web-based technologies. Plots were established on a 10 to 15 year-old AB fairway maintained at 1.5 cm, receiving 0.5 to 0.6 cm daily automatic irrigation throughout the growing season and 120 kg N ha'1 yr'1 from 2001-2006. Soil type was a Mariette sandy loam (fine-loamy, mixed, mesic Glossoboric Hapludalfs). PGRS 52 were applied at 0.13, 3.74 and 0.08 kg ai ha'1 for ME, EP, and TE, respectively, between 24-Mar and 13-May. MF resulted in less than 10 percent seedhead cover in all years. EP interacted with application timing in 2002 but not 2003. EP plus TE interacted with application timing in 2004, but not 2005 or 2006. EP and EP plus TE did not provide less than 10 percent seedhead cover in 2003 or 2006, respectively. The proposed GDD model accurately predicted best application timing range in six ofsix years. MF treatments applied between 350 and 550 GDD-5, calculated from 1 March, resulted in less than 5 percent seedhead cover in all years. Severe, sporadic turfgrass injury was associated with MF applications in all years and was not associated with GDD accumulation. Applications of EP and EP plus TE resulted in 20 and 15 percent seedhead cover, respectively. No turfgrass injury was associated with applications of EP or EP plus TE. 53 INTRODUCTION Poa annua L., commonly known as annual bluegrass [AB], can be found anywhere in the world where human disturbance has occurred (Tutin, 1957). Automatic irrigation and routine fungicide provide selection pressure for perennial biotypes in these climates. The life history of AB is that of a winter annual. As such, populations of AB usually exhibit profuse seedhead production in the spring of the year as they complete their life cycle. When maintained as a perennial crop in an irrigated fairway environment, AB will germinate throughout the year and flower at maturity in the first year of growth. Vernalization of mature plants synchronizes the surface populations flowering into a spring-timed masting event (johnson and White, 1997). Reproduction from seed is less important for AB in a perennial crop system, however, the diversity of life history traits associated with the surface population and the ready occurrence of voids ensures that rapid resource capture from seed remains an important component of the AB competitive profile. Objections to AB Seedheads One of the major components to the competitive ability of AB is its ability to produce viable self-fertilized seed at any mowing height even when clipped at short intervals (Koshy, 1969). It is no wonder that seed production plays such an important part in AB'S ability to colonize new sites. The formation and maintenance of a robust soil seedbank positions AB to capture resources as they become available when factors like disease, divots, and traffic reduce the density of or eliminate portions of the surface population. It is therefore intuitive that reducing the contribution to the seedbank would eventually result in less AB in the surface 54 population. In mixed stands of AB and creeping bentgrass (CB), limiting new seed deposits to the seedbank will result in a reduction in AB over time. This was demonstrated by Gaussoin and Branham (1989) where diligent, season-long clipping collection in a mixed AB/ CB fairway was used to limit new deposits of AB seed to the seedbank resulting in an increase in CB. A similar benefit may be realized from effective chemical seedhead suppression. Turfgrass managers also object to seedhead production of AB as it relates to interference with surface uniformity during prolific seed production. Much of the enjoyment and use of turfgrasses is related to their aesthetic value. Seedhead production of AB diminishes the aesthetic value of the turf through visual interference. The presence of AB seedheads would interfere with surface uniformity and ball roll on a putting green. During seedhead production energy is diverted from other plant processes, like rooting and production of defensive compounds, into flowering. Therefore, heavy seed production during the spring may reduce overall plant vigor, predisposing annual bluegrass to impending summer stresses such as heat, drought, disease, and traffic. In situations where annual bluegrass is the primary component of the surface population, suppressing seedhead production may be key to improving overall plant vigor prior to summer stresses. There is a paradox related to AB seedhead suppression. Unregulated populations of AB will produce a tremendous amount of seed, building up the soil seed bank, well positioning AB for resource capture when future voids are created. Because plants expend energy on seedhead production, AB is predisposed to impending summer stress and likely to die, creating voids in the turf that are readily filled by their 55 progeny. Inguagiato et al. (2008) have shown that AB develops less anthracnose in the summer when treated with PGRS in the spring. On the other hand, suppressing seedheads results in healthier, more vigorous plants that are more likely to survive the summer, grow in stature, and become a larger proportion of the overall surface population. Therefore, when used alone, suppression of seedheads may not be a very effective long-term control strategy. Chemical Suppression of AB Seedheads Turf managers have few options for suppressing seedhead production. Mefluidide [MF] {N-[2,4-dimethyI-5-[[(trifluoromethyl)sulfonyl] amino]phenyl] acetamide} was first reported as a plant growth regulator in turf by Matteson et al. (1976). Mefluidide is primarily foliarly absorbed by turfgrasses and inhibits cell division. Activity in turfgrass plants is maximized 3 to 5 d after application. Affected plants exhibit stunted shoot and root growth for 2 to 5 wk followed by enhanced growth that exceeds that rate of untreated plants (Parups and Cordukes, 1977). Initially, research focused on maximizing clipping reduction and characterizing morphological responses of turfgrass to MF applications (Watschke et al., 1977, Schmidt and Bingham, 1977, Parups and Cordukes, 1977). Schott et al. (1980) made the first definitive mention of MP for seedhead suppression in turf. Since the early 1980's, MF has been the primary tool used by turfgrass managers for this purpose (Petrovic et al., 1985, Cooper et al., 1987, Kane, 2003). However, the effectiveness of MF applications varies from region to region and year to year (Petrovic et al., 1985, Danneberger 1987, Deroeden 1984, Branham, 1989). A summary of AB seedhead suppression provided by MF is Shown in Table 4.01 including rate, rating type, 56 Table 4.01. Reported annual bluegrass seedhead suppression from mefluidide by application timing method, location, author and year. kg Percent Year Author Location ai/ha Method Base Timing Control1 1984 Danneberger OH 0.035 BE 13 C 15 45% 1984 Danneberger OH 0.035 BE 13 C 25 43% 1984 Danneberger OH 0.035 BE 13 C 30 39% 1984 Danneberger OH 0.035 BE 13 C 45 5% 1984 Danneberger MI 0.035 BE 13 C 25 15% 1984 Danneberger MI 0.035 BE 13 C 30 -6% 1985 Branham MI 0.035 Average 10 C 40 37% 1985 Branham MI 0.035 Average 10 C 50 20% 1985 Danneberger OH 0.035 BE 13 C 15 28% 1985 Danneberger OH 0.035 BE 13 C 25 24% 1985 Danneberger OH 0.035 BE 13 c 30 19% 1985 Danneberger OH 0.035 BE 13 C 45 12% 1985 Danneberger MI 0.035 BE 13 C 25 37% 1985 Danneberger MI 0.035 BE 13 C 30 20% 1984 Danneberger OH 0.07 BE 13 C 15 60% 1984 Danneberger OH 0.07 BE 13 C 25 64% 1984 Danneberger OH 0.07 BE 13 C 30 82% 1984 Danneberger OH 0.07 BE 13 C 45 37% 1984 Danneberger MI 0.07 BE 13 C 25 61% 1984 Danneberger MI 0.07 BE 13 C 30 76% 1984 Cooper OH 0.07 Boot 62% 1985 Danneberger OH 0.07 BE 13 C 15 74% 1985 Danneberger OH 0.07 BE 13 C 25 82% 1985 Danneberger OH 0.07 BE 13 C 30 66% 1985 Danneberger OH 0.07 BE 13 C 45 38% 1985 Danneberger MI 0.07 BE 13 C 25 44% 1985 Danneberger MI 0.07 BE 13 C 30 54% 11]" multiple rating dates existed the percent control reported here is from seedhead production peak date. 57 Table 4.01. (cont’d) Reported annual bluegrass seedhead suppression from mefluidide by application timing method, location, author and year. kg Percent Year Author Location ai/ha Method Base Timing Control1 1999 Mahady CA 0.07 No Mention 70% 1999 Watschke PA 0.07 emergence 23% 2001 Stier WI 0.07 lst Mowing 74% 2001 Stier WI 0.07 3rd Mowing 86% 1985 Branham MI 0.09 Average 10 C 40 44% 1985 Branham MI 0.09 Average 10 C 50 54% 1982 Danneberger Ml 0.125 BE 10 C 25 76% 1982 Danneberger MI 0.125 BE 10 C 50 95% 1982 Danneberger MI 0.125 BE 10 C 75 46% 1983 Cooper OH 0.14 Boot 98% 1984 Danneberger OH 0.14 BE 13 C 15 75% 1984 Danneberger OH 0.14 BE 13 C 25 97% 1984 Danneberger OH 0.14 BE 13 C 30 91% 1984 Danneberger OH 0.14 BE 13 C 45 52% 1984 Danneberger Ml 0.14 BE 13 C 15 62% 1984 Danneberger MI 0.14 BE 13 C 25 39% 1984 Danneberger MI 0.14 BE 13 C 30 76% 1984 Danneberger MI 0.14 BE 13 C 45 77% 1984 Cooper OH 0.14 Boot 58% 1If multiple rating dates existed the percent control reported here is from seedhead production peak date. 58 Table 4.01. (co nt'd) Reported annual bluegrass seedhead suppression from mefluidide by application timing method, location, author and year. kg Percent Year Author Location ai/ha Method Base Timing Control1 1985 Branham MI 0.14 Average 10 C 25 77% 1985 Branham MI 0.14 Average 10 C 40 65% 1985 Branham MI 0.14 Average 10 C 50 79% 1985 Branham MI 0.14 Average 10 C 75 33% 1985 Danneberger OH 0.14 BE 13 C 15 82% 1985 Danneberger 0H 0.14 BE 13 C 25 86% 1985 Danneberger 0H 0.14 BE 13 C 30 79% 1985 Danneberger OH 0.14 BE 13 C 45 55% 1985 Danneberger MI 0.14 BE 13 C 15 77% 1985 Danneberger MI 0.14 BE 13 C 25 65% 1985 Danneberger Ml 0.14 BE 13 C 30 79% 1985 Danneberger MI 0.14 BE 13 C 45 32% lst 2001 Stier WI 0.14 Mowing 95% 3rd 2001 Stier WI 0.14 Mowing 98% 2001 Kane IL 0.14 mid-April 95% 2006 Bigelow IN 0.14 Spring 68% 2006 Bigelow IN 0.14 Spring 13% 1984 Cooper OH 0.21 Boot 59% 1If multiple rating dates existed the percent control reported here is from seedhead production peak date. 59 application date, location, and application determination method. Well-timed applications of MF can provide almost complete seedhead suppression (>95%) (Borger, 2008). Poorly-timed application results in inconsistent seedhead suppression and sporadic transient, but severe, turfgrass injury. It is not apparent from the literature if high levels of seedhead suppression are always associated with high levels of turfgrass injury. Turfgrass injury from MP is generally reported as yellowing or bronzing of shoots occurring 2 to 5 weeks after application and lasting for 7 to 14 d. There is some anecdotal evidence that associate increased injury symptoms with the occurrence of post-treatment frost events. The author has observed injury duration of 21 to 26 d when frost and cool temperatures follow MF application by 7 to 10 d (Figure 4.01). However, injured turf eventually recovers and exhibits excellent quality (Figure 4.02). Petrovic et al. (1985) reported reduced turfgrass quality that lasted for 10 wk after MF application. Cooper (1984) reported leaf tip yellowing that lasted for 3 to 4 wk after application followed by 4 to 6 wk of improved quality as compared to the control. Turfgrass injury associated with MF is reported in nearly every trial and is characterized as mild, severe, transient, persistent, and unacceptable. In the study by Cooper (1984), it was postulated that MF uptake, seedhead suppression, and resultant injury were reduced when applications were made under suboptimal environmental conditions associated with cooler air temperatures. Several authors have suggested using Fe tankmixed with MF to mask turfgrass injury after application. Using partially chelated Fe sources antagonizes MF (Watschke and Borger, 2005). Tankmixes of MF plus partially chelated Fe 60 Figure 4.01. Turfgrass injury from mefluidide can be exacerbated by post- treatment frost events. Unseasonably cool air temperatures after the onset of injury can delay recovery to 21 (1. Photo courtesy of RN. Calhoun, Indian Hills 66, Okemos, MI - 8 May 2003 ,- 1 “ ~-M;.-':P‘ ‘. a”. 19316....” 3‘." -' g . Figure 4. 02. Treated area recovered from initial injury, exhibiting excellent turfgrass quality and annual bluegrass seedhead control. Photo courtesy of RN. Calhoun, Indian Hills GC, Okemos, MI - 23 May 2003 61 products result in less severe turfgrass injury and reduced levels of seedhead control. Fully chelated Fe products can mask the transient turfgrass injury associated with MF applications without antagonizing performance (Branham, 1989). Borger (2008) has done extensive work examining possible tankmix combinations of MF with Fe products, seaweed extracts, wetting agents, and other biolstimulants in an attempt to minimize or mask turfgrass injury associated with MF. In general, turfgrass researchers seem more interested in maximum efficacy and less concerned with injury than turfgrass managers. Ethephon [EP] {2-chloroethylphosphonic acid} for use as a PGR in turfgrass was first reported in the 1960's (Heng and White, 1969). EP is foliar absorbed by turfgrass. EP translocation and activity in the plant is maximized in 7 to 14 (1. EP is metabolized to ethylene in the plant. Ambient air temperatures primarily regulate the rate of this process. In cooler spring conditions, plant response may be delayed by 14 d. Turfgrass response it typified by leaf senesce, internode elongation, and a 7 to 10 wk period of 30 to 35 percent clipping reduction. Buettner et al. (1976) found that EP could suppress seedheads on tall fescue (Festuca arundinacea), Kentucky bluegrass (Poa pratensis), and white clover (Trifolium repens). Petrovic et al. (1985) found no suppression of AB seedheads from EP. Christians and Nau (1984) measured shoot grth effects of EP on tall fescue, Kentucky bluegrass (Poa pratensis), and hard fescue (Festuca Iongifolia), but did not provide information on seedheads or turfgrass injury. Eggens et al. (1989) observed 4 wk of growth suppression on CB and improved turfgrass quality, while at the same time reporting severe injury in AB. EP was commercialized for use in the turfgrass industry in 62 1999. The introduction of EP renewed interest in the whole area of chemical suppression of AB seedheads (Kane and Miller, 2003; Gelernter and Stowell 2001; Borger, 2008; Askew, 2006; Kopec, 2004). The literature on EP since it's introduction to the turf market is primarily in the popular press. The large amount of somewhat conflicting data (when given), observations, conclusions, and recommendation has lead to some confusion in the industry regarding the use of this product. Cooper et al. (1987) evaluated EP for the selective removal of AB in mixed AB /CB stands. Eggens and Wright (1985) concluded that EP could be used to selective weaken AB compared to KBG or CB. EP has provided excellent suppression of AB seedheads in studies in northern California (Gertlener and Stowell, 2001) and Chicago (Kane and Miller, 2003). Variability of results in other regions (Bigelow, 2006; Street, 2004) could be related to the specific life history or surface populations and environmental conditions. There are varying reports of EP injury to turfgrass (Cooper et al. 1987; Diesburg, 2000). It is difficult to summarize injury findings because of various application methods and use rates. Most references prior to commercialization use rates two to four times the current label rate. Examples of application method include soil drench, Hoagland's solution, and immersion of imbibed seeds in solution of EP. Since its commercial introduction, injury associated with EP is characterized by mild chlorosis and internode elongation. The author has described the chlorosis Similar to that of a Granny Smith apple (Figure 4.03). The chlorosis is most noticeable on small plot research when treatments are adjacent to PGRS that darken the turf and is less noticeable on golf course green and fairways when the 63 Figure 4.03. Ethephon may cause temporary chlorosis similar to a Granny Smith apple color, however, in the absence of seedheads, the turf will appear darker green than the untreated area. Photo courtesy of RN. Calhoun, Hancock Turfgrass Research Center, East Lansing, MI - 21 May 2004 entire area has been treated. This chlorosis can be offset by tankmixing EP with trinexapac-ethyl [TE] {4-(cyclopropyl-a-hydroxy-methylene)-3,5-dioxocyclohex acid ethylester} (Kane and Miller, 2003). In recent years, researchers and turfgrass managers have evaluated tankmix combinations of PGRS for seedhead suppression and other uses (Stier 2001; Stowell 2000; Borger, 2008; Bigelow, 2004, 2006; Kane and Miller, 2003). However, only the combination of EP plus TE has come into wide use among golf course superintendents for seedhead suppression of AB (Borger, 2008), particularly on putting greens where there is less tolerance for injury and EP plus TE is seen as an easy way to ease into a TE program. The EP plus TE tankmix results in less turfgrass injury than EP alone. TE does not suppress seedheads, but will delay onset of seedheads for up to 14 d. TE is foliar absorbed and is maximized 64 in the plant in 3 to 4 d. TE inhibits gibberellic acid biosynthesis. Affected plants exhibit darker green color, improved density, and decreased clipping production for up to 28 d. Optimizing Application Timing AB seedhead suppression and severity of turfgrass injury from MF applications varies by year, application timing, and location. In an effort to explain this variability, several researchers have attempted to identify optimum application timing through the use of multiple calendar date application timings throughout the Spring (Jagschitz, 1984), growing degree-days (Danneberger and Vargas, 1982, Danneberger et al., 1987, Branham, 1989, Askew et al., 2006), phenological indicators (Askew et al. 2006), or by describing the morphological stage of AB (Cooper et al. 1987, Kane and Miller, 2003, Watschke and Borger, 2005). In the last example, researchers use vigorous scouting to identify the boot and pre-boot stage of AB. Boot stage was described as a swollen stem ready to give birth to a developed inflorescence. Danneberger et al. (1987) used a modified form of the Baskerville- Emins (1969) GDD accumulation method (Danneberger and Vargas, 1984) to identify the optimum application window in Michigan and Ohio. In those trials, GDD . were accumulated from 1 April with a base temperature of 13 C. Optimum application timing was determined to be between 15 and 30 GDD. End-user adoption of the Baskerville-Emins method has been limited either by the models use of advanced math or by the small number of agricultural weather stations that automatically calculate GDD in this way. In an attempt to encourage wider adoption of GDD use, Branham and Collins (1986), using the data set from Danneberger et al. 65 (1987), suggested that using the Simple average calculation method with a base temperature of 10 C would be equally appropriate. They concluded that applications between 45 to 90 GDD1o would provide commercially acceptable control. Later, Branham (1991) suggested that the simple average calculation method could be used with either GDD1o or GDD13 with target ranges of 45 to 90 and 25 to 50, respectively. The commercial introduction of EP sparked renewed interest in chemical suppression of AB and efforts to identify maximum application timing. This research is part of that effort. Application timing of EP has not been studied to the same degree as MF. Some attempt has been made to associate EP with GDD and growth stage of the turf. It is our hypothesis that EP will be affected by application timing and EP plus TE will not. This tankmix should allow for a wider application timing range. The plant rapidly takes up TE while cooler air temperatures delay EP metabolism. TE will typically delay AB seedheads for 10 to 14 d, which corresponds with the time it takes for EP to become active in the plant. The objectives of this study were to evaluate: 1) AB seedhead suppression from single applications of MF compared with EP, 2) AB seedhead suppression from Single applications of MP compared with EP plus TE, 3) influence of application timing and identify best application dates for MF, EP or EP plus TE in each year, and 4) Simple average-calculated GDD and Baskerville-Emin GDD for reliability to predict best application timing dates from year to year and, 5) if a simple, reliable GDD model can be constructed, ground-truth a web-based weather system whereby 66 golf course superintendents could use local weather data to predict best application timing to achieve maximum chemical seedhead suppression of AB. MATERIALS AND METHODS Site Description Five one-year studies were conducted from 2002 to 2006 at the Hancock Turfgrass Research Center in East Lansing, Michigan, USA. Plots were established on a 10 to 15 year-old AB fairway maintained at 1.5 cm, receiving 0.5 to 0.6 cm automatic irrigation throughout the growing season and 120 kg N ha'1 yr'1. Soil type was a Mariette sandy loam (fine-loamy, mixed, mesic Glossoboric Hapludalfs). Treatments The experimental design was a factorial with two factors. In 2002 and 2003 factor A included two PGR treatments: MF applied at 0.13 kg ha"1 or EP applied at 3.74 kg ha'l. In 2004, 2005, and 2006 factor A was changed to include: MF applied at 0.13 kg ha‘1 or EP plus TE applied at 3.74 plus 0.08 kg ha'1, respectively. The second factor (B), application timing, varied between 12 and 15 levels (timings) for each year of the study. Applications were made between 24 March and 19 May of each year. The experimental unit in this trial was plot (1.2 x 2 m). Throughout the application timing range both PGR treatments were applied on Monday and Thursday of each week. As climatic conditions vary from year to year, application date was converted to growing degree-days (GDD). A base temperature of -5 C (GDD—5) was used as it was previously shown to provide the best fit to the response 67 of seedhead formation in this AB population (Calhoun, 2009). In this way, GDD values for each application date could be compared from year to year and potentially extrapolated to other locations. Two untreated checks were included in each replication for determination of percent control but were not included in the data analysis. Application Equipment and Calibration PGR treatments were applied with a COz-pressurized four-nozzle backpack sprayer calibrated to deliver 518 L ha'1 at 276 kPa using Teejet 8002VS flat-fan nozzles. Application rates were 0.13, 3.74, and 0.08 kg ha'1 for MF, EP, and TE, respectively. Evaluations Percent seedhead cover was determined by visual estimation for each plot. Verification of visual estimation was previously done with binomial coordinate mapping (grid method) in 2001 and 2002. It was determined that a trained rater could accurately determine percent cover by whole plot visual estimation. The accuracy and speed of this method facilitated multiple seedhead rating dates in each year. Turfgrass injury was rated on a 1 to 9 scale where 1=none, 9=complete foliar blighting (phytotoxicity), and ratings greater than 3=commercially unacceptable. In 2002, PGR treatments were made on 14 dates over a 47 d period between 28-Mar and 14-May. Visual estimates of AB seedhead cover were made on 10 dates over a 38 d period between 7-May and 14-Jun. Turfgrass injury was rated six times over a 54 d period between 14-Apr and 28-May. In 2003, PGR treatments were 68 made on 15 dates over a 56 d period between 24-Mar and 19-May. Visual estimates of AB seedhead cover were made on 10 dates over a 46 d period between Z-May and 17-jun. Turfgrass injury was rated Six times over a 60 (1 period between 18-Apr and 17-jun. Figure 4.04 (A and B) summarizes mean daily air and soil temperatures, application dates, and GDD—5 accumulation for the period from 1-Mar to 30-june for 2002 and 2003. In 2004, PGR treatments were made on 12 dates over a 36 (1 period between 31-Mar and 6-May. Visual estimates of AB seedhead cover were made on four dates over a 27 d period between 12-May and 8-jun. Turfgrass injury was rated five times over a 47 d period between 22-Apr and 8-jun. In 2005, PGR treatments were made on 12 dates over a 39 d period between 29-Mar and 6-May. Visual estimates of AB seedhead cover were made on 10 dates over a 31 d period between 10-May and 10- jun. Turfgrass injury was rated 18 times over a 60 d period between 11-Apr and 10- jun. In 2006, PGR treatments were made on 11 dates over a 34 d period between 31-Mar and 4-May. Visual estimates of AB seedhead cover were made on seven dates over a 20 d period between 3-May and 23-May. Turfgrass injury was rated 10 times over a 35 (1 period between 22-Apr and 23-May. Figure 4.05 (A, B, and C) summarizes mean daily air and soil temperatures, application dates, and GDD-5 accumulation for the period from 1-Mar to 30-june for 2004, 2005, and 2006. The treatment design was a two Factor RCBD with repeated measures in three replications. Replications of the experiment were conducted in 2002, 2003, 2004, 2005, and 2006. The statistical model used is represented by: 69 y = p + Block + PGR + App + PGR*App + 81 + Rating + PGR*Rating + App*Rating + PGR*App*Rating + 192 Initial assessment of chemical seedhead suppression and turfgrass injury was done by plotting box and whisker diagrams of percent control and injury ratings from all rating dates for each application date for each year. This was helpful for identifying trends in seedhead suppression response and injury related to application timing. Turfgrass injury ratings were scored as acceptable or unacceptable. The averages of unacceptable turfgrass injury ratings (greater than or equal to 3) by application date were used to represent the severity of turfgrass injury associated with a particular PGR from a particular application date. The date range associated with consecutive unacceptable injury ratings is the duration of injury in days. Further evaluation of the data examined seedhead suppression data at or near maximum seedhead production [peak]. It was determined that seedhead evaluations at the beginning and end of the seedhead production period (when overall seedhead cover was low) were unequally benefiting poor performing application dates in comparison to well performing application dates. Finally, it was determined that due to differences in overall seedhead pressure by year that percent control, by its self, may not be the best measure of reporting results. For instance, while it could be that 99 percent control is always excellent, there are cases, in years with lower seedhead pressure, that significantly less control is necessary to achieve commercially acceptable results (say less than 5 or 10 percent cover). Therefore, percent seedhead cover was used for reporting purposes. 70 30 25 20 Temperature (degrees C) (.0 O -10 80 90 110 Ordinal Day 140 — Air_5 ------- Soil_5 GDD (-5C) — Air_5 -------- Soil_5 GDD (-5C) 160 2000 1800 1600 1400 1 200 1000 200 2000 1800 1 600 1400 1200 1000 200 Figure 4.04 Growing degree-day accumulation (GDD-5) from 1 March and corresponding 5 (1 moving mean air and soil temperature (A, 2002; B, 2003) as measured by MAWN weather station approximately 5 km from study location. 71 GDD Total (from 1 Mar) 30 2000 25. (A) . - E1800 w . 30’3- . - ' ’ I Application Range I . :2880 i (B) 625: $20: 0) 2 . 3,15- — A1r_5 a.) 2 {310—3 5.: ------- Soil_5 ‘5 1 Tu _ _ _ a 52 a 000(50) 5 o: '— o. - o E-Si D ,_ <9 25: (C) 20: 15: 10: 5g 7 05. ’“ ," -5‘ ‘ - 4.;— ‘101 '1‘: W 1 1 O eeeesseeeeess Ordinal Day Figure 4.05 Growing degree-day accumulation (GDD-5) from 1 March and corresponding 5 d moving mean air and soil temperature (A, 2004; B, 2005; C, 2006) as measured by MAWN weather station approximately 5 km from study location. 72 Statistical Analysis Data were analyzed as a factorial randomized complete block design with repeated measure using PROC MIXED procedure in SAS (SAS Institute, 2007). Normality and homogeneity of the variances were assessed using PROC UNIVARIATE. Normality of the residuals was assessed with stem-and-leaf, box and normal probability plots. When appropriate, data were transformed (1 /log) to correct for right skewness in the data set. Homogeneity of the variances was examined using side-by-side box plots, initial results suggested unequal variance. Unequal variance analysis using REPEATED/GROUP=trt statement (grouped data by application date). Treatments with similar variances were grouped and the unequal variances of these groups were accounted for with REPEATED/GROUP=app date statement. Results were analyzed and the Akaike Information Criterion (AIC) fit statistics were improved; therefore, data was analyzed using unequal variance grouped by application date. Several covariance structures were considered with compound symmetry with heterogeneous variance (csh) returning the lowest AIC value. However, data were analyzed with the compound symmetry (cs) structure as more complex structures did not converge or generated infinite likelihood or out of memory errors in SAS. LSMEANS was used to identify possible significance of main effects and interactions of application date and PGR treatment. It was determined that application timing interacted with MP in all years but not with HP or EP plus TE. Slicing results were used to make conclusions regarding significantly different application timings for MF and derive LSD values at a probability level of 0.05. Using 73 GDD-5 as a continuous variable also facilitated generating model equations to explain the response of PGR by application date, for each rating date, for each year. Model equations were generated with PROC MIXED with GDD-5 included in the RANDOM statement. The final model for percent seedhead cover by rating date by application timing for MP is: percent cover = ra te_date + rate_date*gdd + ra te_date*gdd*gdd The resulting curves were used to identify optimum application ranges within each year. Peak seedhead date from each year was used to compare optimum application timing ranges across all years. RESULTS Significant differences were observed among the PGRS. Both PGR treatments reduce seedhead production of annual bluegrass. MF reduced seed production of annual bluegrass more than EP or HP plus TE. Significant interactions occurred between MF and application timing in all years. Significant interactions also occurred for EP and application timing in 2002 and 2003, and for EP plus TE in 2004. PGR was always significant and interactions existed separately for PGR. Therefore, the data set for each PGR was analyzed by application timing. No turfgrass injury was observed from applications of HP or EP plus TE at any application timing in any year. Unacceptable turfgrass injury, greater than 30 percent, was observed from applications of MP in all years. 74 Annual Bluegrass Seedhead Suppression by Season 2002 The major flowering period in 2002 occurred over a 35 d period between 7- May and 11-Iun (Table 4.02). Peak seedhead production, ranging between 60 and 80 percent cover, occurred over a 15 d period between 20-May and 4-jun. On average, MF applications suppressed AB seedhead emergence by 77%. Seedhead emergence ranging from 26 to 97 percent control throughout the evaluation period corresponded to 1 to 40 percent cover (Table 4.03). Seedhead suppression observed on the peak seedhead date (28-May) was 69%, when averaged across application dates (Table 4.04). Maximum seedhead suppression occurred from MF applications made between 15-Apr and 2-May which corresponded with 353 to 607 GDD-5. Seedhead suppression from applications made within this range averaged 96 percent control over the entire flowering period and 99 percent control at peak. Treatment dates falling outside the optimum range averaged 58 percent control for the entire flowering period and 47 percent control at peak. Average percent cover at peak AB seedhead production from MF in 2002 was 1 and 39 for inside-range and outside-range application timings, respectively. Model parameters for ME by rating date by year were plotted against GDD-5 (Figure 4.06). The curves were analyzed to identify the range of GDD values associated with percent seedhead cover of less than 5%. In 2002, the optimum range was approximately 400 to 550 GDD-5. 75 Table 4.02. PGR application timing and various associated GDD accumulations, pre-treatment and post-treatment 5 d average maximum air temperature. 2002. Hancock Turfgrass Research Center, East Lansing, MI. Application Timing Mean 5 d Max Air (°C) Date GDD-5+ 81313i GDDso’r TD-4* TD+4* 28-Mar 137 1 1 3.9 8.3 1-Apr 170 1 1 8.3 4.3 4-Apr 188 1 1 5.8 6.7 8-Apr 220 1 1 6.7 17.7 11-Apr 264 5 6 15 22.9 15-Apr 353 24 58 22.9 27.9 16-Apr 382 35 84 24.3 24.4 18-Apr 435 52 138 27.4 15.7 22-Apr 488 57 142 15.7 12.3 25-Apr 526 59 142 11.3 11.3 29-Apr 570 59 142 11.3 12.2 2-May 607 59 142 12.3 16.6 6-May 673 68 161 16.6 16.2 14-May 793 73 178 13.5 15.4 TGDD accumulated from 1 March 2002 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Mean 5-d maximum air temperature. TD-4 represents treatment date and four preceding days, TD+4 represents treatment date and four following days. 76 Table 4.03. Annual bluegrass seedhead suppression provided by a single application of mefluidide at 0.13 kg ai ha‘1 as affected by application timing. 2002. East Lansing, MI. Application Timing Seedhead Suppression Date GDD-5’r [3513+ Percent Control Percent Coveri‘ 28-Mar 137 1 78 g§ 10 1-Apr 170 1 47 h 25 4-Apr 188 1 78 h 10 8-Apr 220 1 15 i 40 11-Apr 264 5 86 f 7 15-Apr 353 24 96 bd 2 16-Apr 382 35 96 bd 2 18-Apr 435 52 95 be 2 22-Apr 488 57 97 a 1 25-Apr 526 59 96 ab 2 29-Apr 570 59 96 acd 2 2-May 607 59 95 cde 2 6-May 673 68 78 h 10 14-May 793 73 26 i 35 TGDD accumulated from 1 March 2002 where GDD-5 and 60050 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 47 percentseedhead cover during flowering period. §Means followed by the same letter are not diflerent at P=0.05. Unequal variance was used in an alysis for means separation. 77 Table 4.04. Annual bluegrass seedhead suppression during peak seedhead production (28-May) provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2002. East Lansing, MI. Application Timing seedg‘::£::olzjlszéz:m at Date GDD-5i BE13’r Percent Control Percent Covert 28-Mar 137 1 66 b 2 5 1-Apr 170 1 66 b 2 5 4-Apr 188 1 66 b 2 5 8-Apr 220 1 -3 b 80 11-API' 264 5 89 b 8 15-Apr 353 24 99 a 1 16-Apr 382 35 99 a 1 18-Apr 435 52 99 a 1 22-Apr 488 57 99 a 1 25-Apr 526 59 99 a 1 29-Apr 570 59 99 a 1 2-May 607 59 93 1 6-May 673 68 5 5 33 14-May 793 73 -4 b 77 iGDD accumulated from 1 March 2002 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 B respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. =FUn treated plots averaged 74 percentseedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 78 Rating Dates Percent Seedhead Cover 5/1 3/2002 5/20/2002 5/23/2002 5/28/2002 5/31/2002 6/4/2002 err/2002 6/1 1/2002 400 500 600 700 800 900 Application Timing (G DD base -5C) 100 Figure 4.06. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for 2002 by rating date. Hancock Turfgrass Research Center, East Lansing, MI. On average, EP applications suppressed AB seedhead emergence by 42%, ranging from 16 to 66 percent control over all rating dates (Table 4.05). Maximum seedhead suppression observed on peak seedhead date was 64% (Table 4.06). Maximum seedhead suppression occurred from EP applications made between 22- Apr and 6-May and corresponded with 488 to 673 GDD—5. Seedhead suppression from applications made within this range averaged 60 percent control of seedheads over the entire flowering period and 39 percent control at peak. Treatment dates falling outside the optimum range averaged 33 percent control for the entire flowering period and 8 percent control at peak. Treatment dates falling outside the optimum range averaged 58 percent control for the entire flowering period and 47 percent control at peak. 79 Table 4.05. Annual bluegrass seedhead suppression provided by a single application of ethephon at 3.4 kg at ha'1 as affected by application timing. 2002. East Lansing, MI. Application Timing Seedhead Suppression Date 000-51 BE 131 Percent Control Percent Covert 28-Mar 137 1 41 d§ 28 1-Apr 170 1 23 f 36 4-Apr 188 1 23 f 36 8-Apr 220 1 16 h 39 11-Apr 264 5 27 e 34 15-Apr 353 24 47 bc 25 16-Apr 382 35 39 d 29 18-Apr 435 52 47 bc 25 22-Apr 488 57 66 a 16 ZS-Apr 526 59 54 ac 22 29-Apr 570 59 57 ab 20 2-May 607 59 61 b 18 6-May 673 68 62 b 18 14-May 793 73 38 d 29 TGDD accumulated from 1 March 2002 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 47 percent seedhead cover during flowering period. §Means followed by the same letter are not dijTerent at P=0.05. Unequal variance was used in analysis for means separation. 80 Table 4.06. Annual bluegrass seedhead suppression during peak seedhead production (28-May) provided by a single application of ethephon at 3.4 kg ai ha'1 as affected by application timing. 2002. East Lansing, MI. Application Timing seedgleesglffolzjlzfciizinon at Date GDD-5* 3513+ Percent Control Percent Covert 28-Mar 137 1 5 b 70 l-Apr 170 1 1 b 73 4-Apr 188 1 5 b 70 8-Apr 220 1 -4 b 77 11-Apr 264 5 2 b 72 15-Apr 353 24 17 b 51 16-Apr 382 35 19 b 50 18-Apr 435 52 28 b 53 22-Apr 488 57 64 a 25 25-Apr 526 59 26 b 55 29-Apr 570 59 33 b 50 2-May 607 59 36 b 43 6-May 673 68 34 b 50 14-May 793 73 0 b 74 iGDD accumulated fi'om 1 March 2002 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 74 percent seedhead cover during peak flowering period. §Means followed by the same letter are not difl’erent at P=0. 05. 81 Average percent cover at peak AB seedhead production from EP in 2002 was 46 and 76 for inside-range and outside-range application timings, respectively. Unacceptable turfgrass injury (greater than 30% blighted tissue) associated with MF occurred over the entire range of application timings. Table 4.07 summarizes the severity and duration of turfgrass injury observed in 2002. Unacceptable turfgrass injury ratings, ranging from 3.8 to 7.3 occurred for a period of 4 to 26 d. However, no injury occurred from the 11-Apr treatment corresponding with 264 GDD-5. This treatment timing fell outside the optimum range for 2002, but provided commercially acceptable levels of suppression averaging less than 10 percent seedhead cover over all rating dates. 82 Table 4.07. Annual bluegrass injury severity and duration as affected by mefluidide at 0.13 kg ha'1 and application timing. 2002. East Lansing, MI. Application Timing Turfgrass lnjury Rating Date GD [3-5+ 3513+ GDD50? Severityi Duration§ 28-Mar 137 1 1 5.4 bcd 16 1-Apr 170 1 1 4.8 abc 16 4-Apr 188 1 1 3.8 a 16 8-Apr 220 1 1 5.8 cd 16 11-Apr 264 5 6 -- -- 15-Apr 353 24 58 5.6 cd 16 16-Apr 382 35 84 4.3 ab 26 18-Apr 435 52 138 4.2 ab 21 22-Apr 488 57 142 5.2 bcd 21 25-Apr 526 59 142 5.3 bcd 21 29-Apr 570 59 142 7.2 e 8 2-May 607 59 142 7.3 e 8 6-May 673 68 161 5.3 bed 8 14-May 793 73 178 6.3 de 4 TGDD accumulated from 1 March 2002 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *lnjury rating where 1 =none, 9=complete bligh ting, and 3=unacceptable; only ratings greater than or equal to 3 reported for average. Means followed by the same letter are not different at p=0.05. §Duration of unacceptable injury in days. 83 2003 The major flowering period in 2003 occurred over a 46 d period between 2- May and 17-Jun (Table 4.08). Peak seedhead production, ranging between 25 and 36 percent cover, occurred over a 13 d period between 14-May and 27-May. On average, MF applications suppressed AB seedhead emergence by 69%. Seedhead emergence ranging from 31 to 95 percent control throughout the evaluation period corresponded to 1 to 13 percent cover (Table 4.09). Seedhead suppression observed on the peak seedhead date (19-May) was 60% (Table 4.10). Factor B, application timing, was significant. Maximum seedhead suppression occurred from ME applications made 10-Apr and between 18-Apr and 24-Apr which corresponded with 280 and between 402 to 492 GDD-S. Seedhead suppression from applications made within this range averaged 93 percent control of seedheads over the entire flowering period and 96 percent control at peak. Treatment dates falling outside the optimum range averaged 60 percent control for the entire flowering period and 47 percent control at peak. Average percent seedhead cover at peak AB seedhead production from ME in 2003 was 1 and 17 for inside-range and outside-range application timings, respectively. Model parameters for ME by rating date by year were plotted against GDD-5 (Figure 4.07). The curves were analyzed to identify the range of GDD values associated with percent seedhead cover of less than 5%. In 2003, the optimum range was approximately 350 to 620 GDD—5. 84 Table 4.08. PGR application timing and various associated GDD accumulations, pre-treatment and post-treatment 5 (1 average maximum air temperature. 2003. Hancock Turfgrass Research Center, East Lansing, MI. Application Timing Mean 5 d Max Air (°C) Date GDD-5i BE13? GDD50? TD-4* TD+4* 24-Mar 140 7 7 12.1 16.9 27-Mar 177 8 9 15.2 10.1 31-Mar 209 10 13 10.1 9.3 3-Apr 247 13 18 9.6 1.0 10-Apr 280 13 18 4.5 17.9 14-Apr 338 20 27 17.9 18.8 16-Apr 381 32 51 20.4 16.8 18-Apr 402 32 51 18.8 15.6 22-Apr 468 40 73 15.6 15.3 24-Apr 492 41 73 15.5 19.9 29-Apr 575 53 92 19.9 16 1-May 611 55 103 19.8 17.6 5-May 674 61 115 17.6 18.8 13-May 816 77 159 18.7 18.9 19-May 932 93 206 19.6 18.3 iGDD accumulated from 1 March 2003 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of13 C. *Mean S-d maximum air temperature. TD-4 represents treatment date and four preceding days, TD+4 represents treatment date and four following days. 85 Table 4.09. Annual bluegrass seedhead suppression provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2003. East Lansing, MI. Application Timing Seedhead Suppression Date GDD—5* BE13? Percent Control Percent Coveri 24-Mar 140 7 59 cfl1§ 8 27-Mar 177 8 35 chi 12 31-Mar 209 10 63 bce 7 3-Apr 247 13 31 dfij 13 10-Apr 280 13 93 a 1 14-Apr 338 20 85 bchj 3 16-Apr 381 32 65 bc 7 18-Apr 402 32 93 a 1 22-Apr 468 40 95 a 1 24-Apr 492 41 93 a 1 29-Apr 575 53 58 bcg 8 1-May 611 55 74 abcd 5 S-May 674 61 86 ab 3 13-May 816 77 65 bc 7 19-May 932 93 40 degh 11 TGDD accumulated from 1 March 2003 where GDD-5 and 60050 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un trea ted plots averaged 1 9 percent seedhead cover during flowering period. §Means followed by the same letter are not different at P=0.05. Unequal variance was used in analysis for means separation. 86 Table 4.10. Annual bluegrass seedhead suppression during peak seedhead production (19-May) provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2003. East Lansing, MI. Application Timing Seedhead Suppression at Peak Production Date GDD-5’r BE13+ Percent Control Percent Cover‘i 24-Mar 140 7 46 def§ 17 27-Mar 177 8 32 def 21 31-Mar 209 10 53 cde 15 3-Apr 247 13 14 ef 27 10-Apr 280 13 95 a 2 14-Apr 338 20 68 bed 10 16-Apr 381 32 51 cdef 15 18-Apr 402 32 96 a 1 22-Apr 468 . 40 97 a 1 24-Apr 492 41 97 a 1 29-Apr 575 53 51 def 15 1-May 611 55 85 b 5-May 674 61 80 be 13-May 816 77 49 def 16 19-May 932 93 -15 f 36 TGDD accumulated from 1 March 2003 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 31 percent seedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 87 Rating Dates Percent Seedhead Cover 5/3/2003 5/9/2003 5/1 4/2003 5/1 62003 5/1 9/2003 5/23/2003 5/27/2003 6/5/2003 6/1 0/2003 000000 geese Application Timing (GDD base -5C) 00 0 «sewage Figure 4.07. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for 2003 by rating date. Hancock Turfgrass Research Center, East Lansing, MI. On average, EP applications suppressed AB seedhead emergence by 53 percent ranging from 41 to 70 percent control over all rating dates (Table 4.11). Maximum seedhead suppression observed on peak seedhead date was 64% (Table 4.12). Maximum AB seedhead suppression occurred from EP applications made between 18-Apr and 22-Apr and corresponded with 402 to 468 GDD-5. Seedhead suppression from applications made within this range averaged 66 percent control of seedheads over the entire flowering period and 61 percent control at peak. Treatment dates falling outside the optimum range averaged 51 percent control for the entire flowering period and 17 percent control at peak. Average percent seedhead cover at peak AB seedhead production from EP in 2003 was 12 and 26 for inside-range and outside-range application timings, respectively. 88 Table 4.11. Annual bluegrass seedhead suppression provided by a single application of ethephon at 3.74 kg ai ha'1 as affected by application timing. 2003. East Lansing, MI. Application Timing Seedhead Suppression Date GDD—st BE13’r Percent Control Percent Coveri 24-Mar 140 7 52 af§ 9 27-Mar 177 8 43 cf 11 31-Mar 209 10 55 ae 8 3-Apr 247 13 32 c 13 10-Apr 280 13 74 ac 5 14-Apr 338 20 50 cefg 10 16-Apr 381 32 54 ag 18-Apr 402 32 62 ab 22-Apr 468 40 70 a 24-Apr 492 41 .61 ad 7 29-Apr 575 53 46 cdf 10 1-May 611 55 47 cdef 10 5-May 674 61 46 cdef 10 13-May 816 77 52 af 9 19-May 932 93 46 abg 10 iGDD accumulated from 1 March 2003 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 1 9 percentseedhead cover during flowering period. §Means followed by the same letter are not different at P=0.05. Unequal variance was used in analysis for means separation. 89 Table 4.12. Annual bluegrass seedhead suppression during peak seedhead production (19-May) provided by a single application of ethephon at 3.74 kg ai ha‘1 as affected by application timing. 2003. East Lansing, MI. Application Timing Seedhead Suppression at Peak Production Date GDD-5’r [3513+ Percent Control Percent Cover* 24-Mar 140 7 29 cd§ 22 27-Mar 177 8 17 d 26 31-Mar 209 10 30 cd 22 3-Apr 247 13 8 de 29 10-Apr 280 13 11 de 28 14-Apr 338 20 25 cd 23 16-Apr 381 32 -6 e 33 18-Apr 402 32 59 ab 13 22-Apr 468 40 64 a 11 24-Apr 492 41 45 bc 17 29-Apr 575 53 16 de 26 1-May 611 55 24 cde 24 5-May 674 61 6 de 29 13-May 816 77 de 29 19-May 932 93 3 de 30 iGDD accumulated from 1 March 2003 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 31 percentseedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 90 Unacceptable turfgrass injury (greater than 30% blighted tissue) associated with MF occurred over the entire range of application timings. Table 4.13 summarizes the severity and duration of turfgrass injury observed in 2003. Unacceptable turfgrass injury ratings, ranging from 3.0 to 6.7 occurred for a period of 14 to 26 d. However, no injury occurred from applications made on 3-Apr, 29- Apr, or 1-May, corresponding with 247 and 575 to 611 GDD-5. These treatment timings fell outside the optimum range for 2003. However, the 1-May treatment timing provided commercially acceptable levels of suppression averaging less than 5 percent seedhead cover over all rating dates. 2004 The major flowering period in 2004 occurred over a 35 (1 period between 3- May and 7-Jun (Table 4.14). Peak seedhead production, ranging between 37 and 72 percent cover, occurred over a 7 d period between 11-May and 18-May. On average, MF applications suppressed AB seedhead emergence by 87%. Seedhead emergence ranging from 79 to 95 percent control throughout the evaluation period corresponded to 1 to 12 percent seedhead cover (Table 4.15). Seedhead suppression observed on the peak seedhead date (12-May) was 65%, when averaged across application dates (Table 4.16). Maximum seedhead suppression occurred from ME applications made between 30-Mar and 12-Apr and on 22-Apr which corresponded with 282 to 402 and 580 GDD-5. Seedhead suppression from applications made within this range averaged 95 percent control of seedheads over the entire flowering period and 97 percent control at peak. 91 Table 4.13. Annual bluegrass injury severity and duration as affected by mefluidide at 0.13 kg ha"1 and application timing. 2003. East Lansing, MI. Application Timing Turfgrass Injury Rating Date 001151 3513+ GDDSOT Severity? Duration§ 24-Mar 140 7 7 6.0 c 14 27-Mar 177 8 9 3.5 a 14 31-Mar 209 10 13 4.2 ab 21 3-Apr 247 13 18 -- -- 10-Apr 280 13 18 5.6 bc 21 14-Apr 338 20 27 3.5 a 14 16-Apr 381 32 51 4.3 abc 14 18-Apr 402 32 51 4.2 ab 26 22-Apr 468 40 73 5.0 bc 14 24-Apr 492 41 73 4.9 be 14 29-Apr 575 53 92 -- -- 1-May 611 55 103 -- -- 5-May 674 61 115 4.8 abc 10 13-May 816 77 159 5.6 bc 14 19-May 932 93 206 6.0 c 10 ’rGDD accumulated from 1 March 2003 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *lnjury rating where 1 =none, 9=complete bligh ting, and 3=unacceptable; only ratings greater than or equal to 3 reported for average. Means followed by the same letter are not different at p=0. 05. §Duration of unacceptable injury in days. 92 Table 4.14. PGR application timing and various associated GDD accumulations, pre-treatment and post-treatment 5 d average maximum air temperature. 2004. Hancock Turfgrass Research Center, East Lansing, MI. Application Timing Mean 5 d Max Air (°C) Date GDD—5t BE13? GDD50? TD-4* TD+4* 30-Mar 282 7 21 13.9 10.6 2-Apr 306 7 21 12.6 11.2 S-Apr 330 7 21 10.3 11.7 8-Apr 367 8 21 10.6 8.7 12-Apr 402 8 21 8.7 16.1 15-Apr 441 15 23 12.0 24.4 19-Apr 533 33 81 24.4 18.3 22-Apr 580 39 91 20.7 17.2 26-Apr 642 43 100 17.2 18.9 29-Apr 693 59 122 18.9 16.6 3-May 752 60 138 16.6 17.6 5-May 800 68 149 15.9 24.2 ’rGDD accumulated from 29 February 2004 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Mean 5-d maximum air temperature. TD-4 represents treatment date and four preceding days, TD+4 represents treatment date and four following days. 93 Table 4.15. Annual bluegrass seedhead suppression provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2004. East Lansing, MI. Application Timing Seedhead Suppression Date GDD-5’r 3513+ Percent Control Percent Coveri‘ 30-Mar 282 7 92 abc§ 3 2-Apr 306 7 92 b 3 5-Apr 330 7 95 a 2 8-Apr 367 8 97 a 1 12-Apr 402 8 97 1 15-Apr 441 15 73 d 11 19-Apr 533 33 87 be 5 22-Apr 580 39 96 a 2 26-Apr 642 43 80 cd 8 29-Apr 693 59 91 b 4 3-May 752 60 72 de 11 5-May ' 800 68 69 e 12 iGDD accumulated from 29 February 2004 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 39 percentseedhead cover during flowering period. §Means followed by the same letter are not different at P=0.05. Unequal variance was used in analysis for means separation. 94 Table 4.16. Annual bluegrass seedhead suppression during peak seedhead production (12-May) provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2004. East Lansing, MI. Application Timing Seedhead Suppression at Peak Production Date GDD-5’r 3513+ Percent Control Percent Cover’F 30-Mar 282 7 93 bd§ 5 2-Apr 306 7 91 bc 6 5-Apr 330 7 97 ab 2 8-Apr 367 8 98 ad 1 12-Apr 402 8 96 ab 3 15-Apr 441 15 21 f 54 19-Apr 533 33 81 be 13 22-Apr 580 39 95 ab 3 26-Apr 642 43 32 ef 46 29-Apr 693 59 65 cf 24 3-May 752 60 7 f 63 5-May 800 68 2 f 67 'rGDD accumulated from 29 February 2004 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un trea ted plots averaged 68 percent seedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 95 Treatment dates falling outside the optimum range averaged 79 percent control for the entire flowering period and 49 percent control at peak. Average percent cover at peak AB seedhead production from ME in 2004 was 2 and 35 for inside-range and outside-range application timings, respectively. Model parameters for ME by rating date by year were plotted against GDD-5 (Figure 4.07). The curves were analyzed to identify the range of GDD values associated with percent seedhead cover of less than 5%. In 2004, the optimum range was approximately 300 to 440 GDD-5. 60 3 Rating Dates 50 — 5/4/2004 —— 5112/2004 .h 0 / \ K \ l/ / / -— 5/19/2004 \ / / / \ /’ // na/ Percent Seedhead Cover [0 CO 0 O / \ \ \ A O 111111 0 100 200 300 400 500 600 700 800 900 1000 Application Timing (G DD base ~5C) Figure 4.07. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for 2004 by rating date. Hancock Turfgrass Research Center, East Lansing, MI. On average, tankmix applications of EP+TE suppressed AB seedhead emergence by 66 percent ranging from 44 to 87 percent control over all rating dates (Table 4.17). Maximum seedhead suppression observed on the peak seedhead date was 38% (Table 4.18). Maximum AB seedhead suppression occurred from EP 96 applications made between 30-Mar and 12-Apr and 22-Apr, which corresponded with 402 to 468 and 580 GDD-5. Seedhead suppression from applications made within this range averaged 76 percent control of seedheads over the entire flowering period and 36 percent control at peak. Treatment dates falling outside the optimum range averaged 57 percent control for the entire flowering period and -8 percent control at peak. At peak AB seedhead production in 2004, the average percent cover from applications of EP+TE was 43 and 74 for inside-range and outside-range application timings, respectively. Unacceptable turfgrass injury (greater than 30% blighted tissue) associated with MF occurred at eight of the 12 application timings. Table 4.19 summarizes the severity and duration of turfgrass injury observed in 2004. Unacceptable turfgrass injury ratings, ranging from 3 to 7 occurred for a period of 10 to 21 (1. However, no unacceptable injury occurred from applications made on 15-Apr, 19-Apr, 26-Apr, or 5-May, corresponding with 441 to 533, 642, and 800 GDD-5. These treatment timings fell outside the optimum range for 2004. However, the 19-Apr and 26-Apr treatment timings provided commercially acceptable levels of suppression averaging 95 percent clean over all rating dates. 97 Table 4.17. Annual bluegrass seedhead suppression provided by a single application of ethephon plus trinexapac-ethyl at 3.74 plus 0.08 kg ai ha'1 as affected by application timing. 2004. East Lansing, MI. Application Timing Seedhead Suppression Date GDD-s’r 3513+ Percent Control Percent Coveri 30-Mar 282 7 82 ab§ 7 2-Apr 306 7 63 d 15 5-Apr 330 7 81 ab 8-Apr 367 8 87 a 12-Apr 402 8 72 cd 11 15-Apr 441 15 60 d 15 19-Apr 533 33 63 cd 15 22-Apr 580 39 73 bc 10 26-Apr 642 43 54 de 18 29-Apr 693 59 61 d 15 3-May 752 60 44 e 22 5-May 800 68 58 de 16 TGDD accumulated from 29 February 2004 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 39 percent seedhead cover during flowering period. §Means followed by the same letter are not different at P=0. 05. 98 Table 4.18. Annual bluegrass seedhead suppression during peak seedhead production (12-May) provided by a single application of ethephon plus trinexapac-ethyl at 3.74 plus 0.08 kg ai ha‘1 as affected by application timing. 2004. East Lansing, MI. Application Timing S€ed§§jsggpdzrfgzrn at Date GDD-5’r 3513+ Percent Control Percent Cover* 30-Mar 282 7 42 a§ 39 2-Apr 306 7 29 abc 48 S-Apr 330 7 39 a 42 8-Apr 367 8 44 a 38 12-Apr 402 8 35 ab 44 15-Apr 441 15 -1 bcd 39 19-Apr 533 33 -9 bcd 74 22-Apr 580 39 29 abc 48 26-Apr 642 43 -22 d 83 29-Apr 693 59 3 bed 66 3-May 752 60 -33 d 90 5-May 800 68 13 abcd 59 TGDD accumulated from 29 February 2004 where GDD-5 and GDDso use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 68 percentseedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 99 Table 4.19. Annual bluegrass injury severity and duration as affected by mefluidide at 0.13 kg ha'1 and application timing. 2004. East Lansing, MI. Application Timing Turfgrass Injury Rating Date GDD-5+ BE13’r GDD50i Severity* Duration§ 30-Mar 282 7 21 4.0 ab 10 2-Apr 306 7 21 4.0 ab 10 5-Apr 330 7 21 5.5 bc 14 8-Apr 367 8 21 6.0 c 14 12-Apr 402 8 21 4.7 ab 21 15-Apr 441 15 23 -- -- 19-Apr 533 33 81 -- -- 22-Apr 580 39 91 4.5 abc 10 26-Apr 642 43 100 -- -- 29-Apr 693 59 122 6.0 c 10 3-May 752 60 138 3.5 a 10 5-May 800 68 149 -- -- iGDD accumulated from 29 February 2004 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *lnjury rating where 1=none, 9=complete blighting, and 3=unacceptable; only ratings greater than or equal to 3 reported for average. Means followed by the same letter are not different at p=0.05. §Duration of unacceptable injury in days. 100 2005 The major flowering period in 2005 occurred over a 31 d period between 10- May and 10-]un (Table 4.20). Peak seedhead production, ranging between 18 and 45 percent cover, occurred over a 17 d period between 17-May and 3-Jun. On average, MF applications suppressed AB seedhead emergence by 78%. Seedhead emergence ranging from 47 to 95 percent control throughout the evaluation period corresponded to 1 to 11 percent cover (Table 4.21). Seedhead suppression observed on the peak seedhead date (17-May) was 80%, when average across application dates (Table 4.22). Maximum seedhead suppression occurred from ME applications between 26-Apr and 29-Apr which corresponded with 534 to 567 GDD-5. Seedhead suppression from applications made within this range averaged 99 percent control of seedheads over the entire flowering period and 99 percent control at peak. Treatment dates falling outside the optimum range averaged 74 percent control for the entire flowering period and 77 percent control at peak. Average percent cover at peak AB seedhead production from ME in 2005 was 1 and 8 for inside-range and outside-range application timings, respectively. Model parameters for ME by rating date by year were plotted against GDD-5 (Figure 4.08). The curves were analyzed to identify the range of GDD values associated with percent seedhead cover of less than 5%. In 2005, the optimum range was approximately 380 to 520 GDD—s. 101 Table 4.20. PGR application timing and various associated GDD accumulations, pre-treatment and post-treatment 5 d average maximum air temperature. 2005. Hancock Turfgrass Research Center, East Lansing, MI. Application Timing Mean 5 (1 Max Air (°C) Date 509-51 3513i GDD50? TD-4* TD+4¥ 29-Mar 110 1 0 10.6 15.8 1-Apr 155 5 7 16.3 16.1 5-Apr 215 10 18 16.1 20.9 8-Apr 266 17 32 20.5 18.7 11-Apr 315 24 41 19.4 16.7 14-Apr 355 25 41 17.6 21.4 18-Apr 427 39 67 21.4 19.5 21-Apr 483 48 90 22.2 9.7 26-Apr 534 48 90 9.1 11.5 29-Apr 567 48 90 12.0 10.1 3-May 608 48 90 10.1 15.6 6-May 649 52 93 13.5 23.4 iGDD accumulated from 1 March 2005 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Mean 5-d maximum air temperature. TD-4 represents treatment date and four preceding days, TD+4 represents treatment date and four following days. 102 Table 4.2 1. Annual bluegrass seedhead suppression provided by a single application of mefluidide at 0.13 kg ai ha‘1 as affected by application timing. 2005. East Lansing, MI. Application Timing Seedhead Suppression Date GDD-5’r [3513+ Percent Control Percent Cover* 29-Mar 110 1 80 bcd§ 4 1-Apr 155 5 84 be 3 5-Apr 215 10 87 b 3 8-Apr 266 17 86 b _ 3 11-Apr 315 24 78 bcd 4 14-Apr 355 25 82 be 4 18-Apr 427 39 71 bcde 6 2 1-Apr 483 48 63 de 7 26-Apr 534 48 95 a 1 29-Apr 567 48 93 a 1 3-May 608 48 66 de 7 6-May 649 52 47 e 11 iGDD accumulated from 1 March 2005 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 20 percent seedhead cover during flowering period. §Means followed by the same letter are not different at P=0.05. 103 Table 4.22. Annual bluegrass seedhead suppression during peak seedhead production (17-May) provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2005. East Lansing, MI. Application Timing Seedhead Suppression at Peak Production Date (3013-51 3513+ Percent Control Percent Cover1i 29-Mar 110 1 87 a5 5 1-Apr 155 5 91 a 3 5-Apr 215 10 89 a 4 8-Apr 266 17 91 a 3 11-Apr 315 24 82 a 6 14-Apr 355 25 86 a 5 18-Apr 427 39 89 a 4 21-Apr 483 48 66 a 12 26-Apr 534 48 97 b 1 29-Apr 567 48 97 b 3-May 608 48 60 c 14 6-May 649 52 29 c 25 ’rGDD accumulated from 1 March 2005 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 35 percent seedhead cover during peak flowering period. §Means followed by the same letter are not different at P=0.05. 104 Rating Dates — 5/10/2005 5’ — 5/13/2005 é —— 5/17/2005 (5 2 — 5/20/2005 g — 5/23/2005 2 5/26/2005 [515 — 5/31/2005 5‘ — 6/10/2005 :1 0 100 200 300 400 500 600 700 800 900 1000 Application Timing (G DD base -5C) Figure 4.08. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for 2005 by rating date. Hancock Turfgrass Research Center, East Lansing, MI. On average, tankmix applications of EP+TE suppressed AB seedhead emergence by 56% ranging from 44-68 percent control over all rating dates (Table 4.23). Average seedhead suppression observed on peak seedhead date was 49% (data not shown). Factor B, application timing. was not significant. Unacceptable turfgrass injury (greater than 30% blighted tissue) associated with MF occurred over the entire range of application timings. Table 4.24 summarizes the severity and duration of turfgrass injury observed in 2005. Unacceptable turfgrass injury ratings, ranging from 3 to 7 occurred for a period of 4 to 28 d. The lowest average injury (3.4) with the shortest duration (4 to 14 d) occurred when applications were made between 11-Apr and 21-Apr corresponding with 315 to 483 GDD.5. 105 Table 4.23. Annual bluegrass seedhead suppression provided by a single application of ethephon plus trinexapac-ethyl at 3.74 plus 0.08 kg ai ha'1 as affected by application timing. 2005. East Lansing, MI. Application Timing Seedhead Suppression Date 000-5? BE 13+ Percent Control Percent Cover* 29-Mar 110 1 43 a§ 11 1-Apr 155 5 49 a 10 5-Apr 215 10 65 a 7 8-Apr 266 17 58 a 8 11-Apr 315 24 56 a 9 14-Apr 355 25 59 a 8 18-Apr 427 39 61 a 8 21-Apr 483 48 51 a 10 26-Apr 534 48 68 a 29-Apr 567 48 61 a 3-May 608 48 44 a 11 6-May 649 52 52 a 10 TGDD accumulated from 1 March 2005 where GDD—5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F) respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 35 percentseedhead cover during peak flowering period. §Means followed by the same letter are not differen t at P=0.05. 106 Table 4.24. Annual bluegrass injury severity and duration as affected by mefluidide at 0.13 kg ha'1 and application timing. 2005. East Lansing, MI. Application Timing Turfgrass Injury Rating Date GDD-5+ BE13? GDD50? Severity* Duration§ 29-Mar 110 1 0 5.5 de 28 1-Apr 155 5 7 5.3 de 28 5-Apr 215 10 18 4.9 cd 28 8-Apr 266 17 32 4.2 bc 28 11-Apr 315 24 41 3.2 a 11 14-Apr 355 25 41 3.1 a 14 18-Apr 427 39 67 3.5 ab 7 21-Apr 483 48 I 90 3.7 ab 26-Apr 534 48 90 5.9 e 25 29-Apr 567 48 90 5.2 de 25 3-May 608 48 90 3.7 ab 18 6-May 649 52 93 3.7 ab 7 TGDD accumulated from 1 March 2005 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *lnjury rating where 1 =none, 9=complete bligh ting, and 3=unacceptable; only ratings greater than or equal to 3 reported for average. Means followed by the same letter are not different at p=0.05. §Duration of unacceptable injury in days. 107 2006 The major flowering period in 2006 occurred over a 20 d period between 3- May and 23-May (Table 4.25). Peak seedhead production, ranging between 20 and 45 percent cover, occurred over a 6 d period between 12-May and 18-May. On average, MF applications suppressed AB seedhead emergence by 56%. Seedhead emergence ranging from 34 to 82 percent control throughout the evaluation period corresponded to 4 to 16 percent cover (Table 4.26). Seedhead suppression observed on peak seedhead date (17-May) was 52%, when averaged over application dates (Table 4.27). Maximum seedhead suppression occurred from MF applications between 7-Apr and 18-Apr and on 3-May, corresponding with 300 to 474 and 728 GDD-5. Seedhead suppression from applications made within these ranges averaged 69 percent control of seedheads over the entire flowering period and 78 percent control at peak. Treatment dates falling outside the optimum range averaged 45 percent control for the entire flowering period and 31 percent control at peak. Average percent cover at peak AB seedhead production from ME in 2006 was 9 and 29 for inside-range and outside-range application timings, respectively. Model parameters for ME by rating date by year were plotted against GDD-5 (Figure 4.09). The curves were analyzed to identify the range of GDD values associated with percent seedhead cover of less than 5%. In 2006, the optimum range was approximately 400 to 540 GDD-5. 108 Tankmix applications of EP+TE suppressed AB seedhead emergence by 27% (16 percent seedhead cover) when averaged over all rating dates in 2006. Seedhead suppression observed on the peak seedhead date was 20% or Table 4.2 5. PGR application timing and various associated GDD accumulations, pre-treatment and post-treatment 5 d average maximum air temperature. 2006. Hancock Turfgrass Research Center, East Lansing, MI. Application Timing Mean 5 d Max Air (°C) Date GDD-5+ 3513+ GDD50? TD-4 TD+4 31-Mar 219 7 9 13.5 12.0 5-Apr 276 7 9 10.7 11.6 7-Apr 300 8 9 11.7 13.9 10-Apr 329 9 9 12.5 21.9 13-Apr 387 18 32 19.0 19.9 18-Apr 474 29 56 19.2 21.7 21-Apr 533 39 69 20.8 18.2 25-Apr 594 44 91 18.2 16.6 28-Apr 634 46 91 16.8 17.9 3-May 728 58 110 19.3 19.3 8-May 811 66 132 19.0 18.4 ’rGDD accumulated from 1 March 2006 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Mean S-d maximum air temperature. TD-4 represents treatment date and four preceding days, TD+4 represents treatment date and four following days. 109 Table 4.2 6. Annual bluegrass seedhead suppression provided by a single application of mefluidide at 0.13 kg ai ha‘1 as affected by application timing. 2006. East Lansing, MI. Application Timing Seedhead Suppression Date GDD-51‘ 35131 Percent Control Percent Coveri 31-Mar 219 7 34 de§ 14 5-Apr 276 7 53 cde 10 7-Apr 300 8 82 a 4 10-Apr 329 9 53 cde 10 13-Apr 387 18 73 ab 18-Apr 474 29 76 ab 21-Apr 533 39 29 e 16 25-Apr 594 44 63 bc 8 28-Apr 634 46 47 cde 12 3-May 728 58 62 bcd 8 8-May 811 66 42 cde 13 TGDD accumulated from 1 March 2006 where GDD-5 and 60050 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un trea ted plots averaged 22 percentseedhead cover during flowering period. §Means followed by the same letter are not different at P=0. 05. 110 Table 4.27. Annual bluegrass seedhead suppression during peak seedhead production (17 -May) provided by a single application of mefluidide at 0.13 kg ai ha'1 as affected by application timing. 2006. East Lansing, MI. Application Timing Seedhead Suppression at Peak Production Date GDD-51‘ 3513+ Percent Control Percent Coveri 31-Mar 219 7 -8 e§ 45 5-Apr 276 7 54 bcd 19 7-Apr 300 8 84 a 7 10-Apr 329 9 64 abc 15 13-Apr 387 18 74 ab 11 18-Apr 474 29 85 a 6 21-Apr 533 39 9 cde 38 25-Apr 594 44 77 ab 10 28-Apr 634 46 36 cd 37 3-May 728 58 82 ab 7 8-May 811 66 19 cde 34 TGDD accumulated from 1 March 2006 where 600.5 and 60050 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *Un treated plots averaged 58 percent clean during peak flowering period. §Means followed by the same letter are not different at P=0. 05. 111 Rating Dates — 5/3/2006 g. — 5/5/2006 0 g —— 5/8/2006 (6 g — 5/12/2006 3 — 5/16/2006 (I) 5 5/18/2006 5 — 5/23/2006 ll O 100 200 300 400 500 600 700 800 1000 Application Timing (GDD base -5C) Figure 4.09. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for 2006 by rating date. Hancock Turfgrass Research Center, East Lansing, MI. 32 percent cover (data not shown). Factor B, application timing, was not significant for EP+TE in 2006. Unacceptable turfgrass injury (greater than 30% blighted tissue) associated with MF occurred over the entire range of application timings. Table 4.28 summarizes the severity and duration of turfgrass injury observed in 2006. Unacceptable turfgrass injury ratings, ranging from 3 to 7 occurred for a period of 4 to 14 d. The lowest average injury (3.0) with the shortest duration (5 d) occurred when applications were made on 10-Apr corresponding with 329 GDD—5. No injury was observed from treatments made on 31-Mar, 18-Apr, and 8-May, corresponding with 219, 474, and 811 GDD-s, respectively. Table 4.2 8. Annual bluegrass injury severity and duration as affected by mefluidide at 0.13 kg ha'1 and application timing. 2006. East Lansing, MI. Application Timing Turfgrass Injury Rating Date GDD-s’r [3513+ GDD50? Severity¢ Duration§ 31-Mar 219 7 9 -- -- 5-Apr 276 7 9 3.7 ab 10 7-Apr 300 8 9 3.3 a 10 10-Apr 329 9 9 3.0 a 5 13-Apr 387 18 32 3.5 ab 14 18-Apr 474 29 56 -- -- 21-Apr 533 39 69 4.7 bc 25-Apr 594 44 91 3.3 a 28-Apr 634 46 91 5.8 c 14 3-May 728 58 110 3.0 a 7 8-May 811 66 132 -- -- iGDD accumulated from 1 March 2006 where GDD-5 and GDD50 use simple average calculations and base temperatures of -5 C and 50 F, respectively, and BE 13 uses the Baskerville-Emin calculation method with a base temperature of 13 C. *lnjury rating where 1 =none, 9=complete bligh ting, and 3=unacceptable; only ratings greater than or equal to 3 reported for average. Means followed by the same letter are not different at p=0.05. §Duration of unacceptable injury in days. BestApplication Timing As previously mentioned, percent control, by itself, may not accurately represent the seedhead cover of the surface population. Because AB seedhead cover is often below 50%, the percent control necessary to achieve commercially acceptable results (<10 percent seedhead cover) could be significantly different from the 'best’ treatment timing yet still provide acceptable results. Acceptable 113 application timing ranges were identified by using 90 and 95 percent clean as the performance thresholds. The resulting application timing and corresponding GDD-S ranges are summarized by PGR by year in Table 4.29. EP applications did not provide commercially acceptable AB seedhead suppression in 2002. In 2004, EP+TE applications made between 30-Mar and 12-Apr resulted in commercially acceptable seedhead suppression while applications made on 5-Apr or 8-Apr resulted in superior seedhead suppression (<5 percent seedhead cover). Performance of EP+TE was not affected by application timing in 2005 or 2006. Applications of EP+TE provided acceptable and unacceptable chemical seedhead suppression in 2005 and 2006, respectively. Date ranges existed for each year where MF provided superior control. The application widow for achieving commercially acceptable results ranged from 13 din 2006 to 43 din 2003. Target application timing ranges to achieve superior seedhead suppression (>95 percent clean) from MF varied from 5 d in 2006 to 35 d in 2005. Varying application timing to the early or late end of the application range did not consistently reduce observed turfgrass injury ratings. DISCUSSION Single applications of MF, EP or EP plus TE suppressed AB seedheads during the spring flowering pulse of a 10 to 15 yr old AB fairway in Michigan in a series of 1 yr studies from 2002 to 2006. MP reduced seedhead formation more than EP or EP plus TE in each year. MF treatments, averaged over application timings and years, resulted in 58 percent suppression of AB seedheads during the spring flowering 114 Table 4.29 Contiguous application timing ranges associated with maximum annual bluegrass seedhead suppression from single PGR application. 2002- 2006. East Lansing, MI. Contiguous Application Range* Year P GRT Percent Max Ordinal (300-5 Duration Cover Iniury§ MF 101-126 264-673 26 510 7 105-122 353-607 18 55 7 2002 EP 105-126 353-673 22 525 1 112-126 488-673 15 520 1 MP 110-124 209-816 44 510 6 110-124 280-492 15 55 6 2003 EP 83-139 140-932 57 510 1 110 280 1 55 1 MP 90-126 282-800 37 510 6 90-120 282-693 31 55 6 2004 90-103 282-402 34 510 1 EP+TE 96-99 330-367 4 55 1 MP 88-123 110-608 36 510 6 88-123 110-608 36 55 6 2005 87-126 110-649 40 510 1 EP+TE MF 95-108 276-474 14 510 4 103-108 387-474 6 55 2006 90-128 219-811 39 520 EP+TE TPGR treatment where mefluidide[MF], ethephon[EP], and trinexapac-ethyl[TE] applied at 0.13, 3. 74, and 0.08 kg ha'1, *Con tiguous application range represented as ordinal days from 1 fanuary, growing degree-days accumulated from ordinal day 60, and duration of optimum application range in days. §lnjury rating on 1-9 scale where, 1=none, 9=completefoliar bligh ting, and >3 is considered unacceptable. 115 period corresponding to 11 percent cover. A simple average growing degree-day model was developed to predict optimum application timing for PGRS. The base temperature for this model was -5 C, which is lower than previously reported base temperatures used for AB (Danneberger and Vargas, 1984). GDD calculated with -5 C base temperature most accurately predicted various phases of AB seedhead production in the same surface population in a simultaneous set of observational studies as compared to other base temperatures between -15 C and 15 C. Growing degree-days were useful for describing the contiguous subset of application dates that resulted in improved seedhead suppression. The most effective combination of PGR and application date suppressed AB seedheads by 75% corresponding to 4 percent cover over all years. The GDD model consistently indentified application date ranges that resulted in maximum seedhead suppression. GDD values corresponding to application dates that resulted in maximum seedhead suppression were selected from curves of modeled seedhead cover estimates plotted against GDD (Figure 4.10). Seedhead suppression from applications made between 350 to 550 GDD-5, provided 71 percent control corresponding to 5 percent cover. Application dates resulting in maximum AB seedhead suppression from ME occurred within this range in all six years. EP suppressed AB seedheads, and suppression varied by application date in 2002 and 2003. In 2002, later application timings improved seedhead suppression as compared to earlier treatments. However, EP did not provide commercially acceptable control, less than 10 percent cover, in 2002. In 2003, EP efficacy was maximized between 280 and 492 GDD—5. EP treatment, averaged over all application 116 Year — 2002 g — 2003 0 — 2004 8 2 —- 2005 8 a) — 2006 (D E a) e (D CL 100 200 300 400 500 600 700 800 900 1000 Application Date (GDD base -SC) Figure 4.10. The result of model equations for mefluidide suppression of annual bluegass seedhead production as affected by application timing for peak seedhead date from 2001-2006. Hancock Turfgrass Research Center, East Lansing, MI. timings in 2002 and 2003, resulted in 35 percent control of AB seedheads during the spring flowering period corresponding to 20 percent coverapplications made between 275 and 500 GDD—5 resulted in 43 percent control of AB seedheads, corresponding to 15 percent cover. Seedhead suppression in response to EP plus TE treatment varied dependent on application date in 2004, but not in 2005 or 2006. EP plus TE treatment, averaged over all application timings in all years, resulted in 36 percent control of AB seedheads during the spring flowering period corresponding to 14 percent cover. In 2004, earlier application timings resulted in improved seedhead suppression as compared to later treatments. EP plus TE treatments made between 250 and 400 GDD-5 resulted in 78 percent control of seedheads during the spring flowering period, corresponding to 8 percent cover. 117 However, this only resulted in improved seedhead suppression in 2004. EP plus TE treatments from 2004 to 2006 made between 250 and 400 GDD-5 resulted in 35 percent control of seedheads, corresponding with 15 percent cover. The tankmix of EP plus TE allows turfgrass managers to treat at anytime in the spring up to conventional MF timing. The results of this research indicate that reported variability in performance of EP or EP plus TE has more to do with the underlying level of seedhead production in a given year and less to do with application timing. This helps explain why golf course superintendents have such varying levels of success with this tankmix. Generally consistent control by EP plus TE on putting greens may be due to the fact that perennial growth habit, competitive-type AB, that dominates putting greens produces less seed than the adjacent fairway population. Single spring-timed applications of EP plus TE provide 40 to 60 percent control of AB seedheads. Therefore, in years when seedhead cover doesn't exceed 25%, the tankmix will likely provide commercially acceptable results. Sporadic and severe turfgrass injury, resulting from ME applications, occurred in all years. In some cases, unacceptable injury persisted for up to 26 d. MF applications resulting in low or no injury occurred intermittently across the range of application dates from year to year. Early treatment timings resulting in somewhat lower injury; however, the injury persisted longer than later application timings, which caused more severe injury that was more transient. It seems likely that the rate of recovery in the late application timings is related to more rapid turfgrass 118 growth associated with warmer temperatures, resulting in faster removal of blighted foliage by mowing. Danneberger et al. (1987) and Branham and Collins (1986) suggested GDD models for timing MF applications based, at least in part, on data collected at the same location as these current studies. The model proposed by Danneberger uses a base temperature of 13 C and the Baskerville-Emin [BE] method of calculating GDD accumulating from 1 April. In part, this method assumes that daily fluctuations in air temperature generally resemble a sine wave. This model attempts to account for biological activity that occurs in the spring when daily maximums are adequate for development but nightly lows are such that no GDD accumulation would occur with the simple average method. The complexity of the math involved with this method may limit its use by turfgrass managers. Branham and Collins (1986) suggested that a simple average method could be used with either GDD1o or GDD13 with target ranges of 45 to 90 or 25 to 50, respectively. The results from this research indicated that the BE method would accurately predict the optimum application timing range in four of six years. However, modifying the start date to from 1 April to 1 March allowed our model to accurately predict the optimum application range in all five years. The simple average GDD models suggested by Branham and Collins (1991) resulted in accurate prediction of application timing range in four of five and two of five years for the GDD1o and GDD 13 model, respectively. The GDD13 simple average model did not exhibit good fit with our data. The simple average model developed in 119 this research, 350 to 550 GDD—5, resulted in 92% overlap with most effective application dates in all five years. Data from this series of five one-year studies suggests that the GDD model suggested by Danneberger et al. (1987) provide appropriate estimates of best application timing when the start date is adjusted from 1 April to 1 March. The simple average GDD calculation method, with appropriate base temperature of -5 C, accurately predicted best application dates for ME in this perennial AB fairway population. The use of the simple average calculation method allows data from more weather stations to easily incorporated into weather-based online decision making tools. The appropriateness of the simple average model is currently being evaluated across Michigan and the Great Lakes Region through a network of weather stations in Michigan, Indiana, Illinois, and Wisconsin. A web site was developed (wwwgddtrackernet) that allows golf course superintendents from across the region to estimate growing degree-day accumulations on their course by linking their location (zip code) with nearby weather stations. Users may choose to receive automatic email alerts when GDD target range is close, at, over, or done. In 2009, there were 803 registered users and they received 2,488 automatic email alerts. Future expansion of the system includes incorporating soil temperature and relative humidity data into the data feed so that models for insects and disease can be added. 120 BIBLIOGRAPHY Allard, R.W., S.K. lain, and PL. Workman. 1968. The genetics of inbreeding populations. Advances in Genetics 14:55-131. Arnold, C.Y. 1959. The determination and significance of the base temperature in a linear heat unit system. journal of the American Society for Horticultural Science. Geneva. 74:430-445. Arnold, C.Y. 1960. Maximum-minimum temperatures as a basis for computing heat units. Proc. Am. Soc.Hort. Sci. 76:682-692. Askew, S. D., D. R. Spak, W. L. Barker, 1. B. Willis, and D. B. Ricker. 2006. Degree day for predicting annual bluegrass seedhead emergence. Proc. Annu. Meet Northeast Weed Sci. Soc. 60:p. 114. Bakersville, G.L. and P. Emin. 1969. Rapid estimation of heat accumulation from maximum and minimum temperatures. Ecology. 50:414-517. Beard, ].B. 1970. An ecological study of annual bluegrass. USGA Green Sec. Rec. 8(2):p. 13-18. Beard, ].B. 1980. Minimum temperatures for turfgrass seed germination. Grounds Maint 15(2):p. 32, 34, 36. Beard, ].B., P.E. Rieke, AJ. Turgeon, and ].M. Vargas jr. 1978. Annual bluegrass (Poa annua L.): Description, adaptation, culture and control. Research Report 352. Michigan Agricultural Experiment Station. East Lansing, MI. Bewley, ].D. and Black, M. 1994. Seeds. Physiology of Development and Germination. Second Edition. Plenum Press, New York, p. 232. Bigelow, CA, and GA. Hardebeck. 2004. Evaluation of ethephon (Proxy) and trinexapac-ethyl (Primo MAXX) combinations for suppression of annual bluegrass seedheads on a golf course putting green. Annu. Rep. Purdue Univ. T urfgrass Sci. Progr. p. 66-72. 121 Bigelow, C.A., and GA. Hardebeck. 2006. Annual bluegrass seedhead suppression in two contrasting golf turf areas as affected by Proxy, a Proxy + Primo Maxx tank-mix and Embark Lite applications, 2006; Purdue University. Ann. Rep. Purdue Univ. Turfgrass Sci. Prog. p. [1-11]. Bigelow, C.A., and GA. Hardebeck. 2006. Annual bluegrass seedhead suppression in two contrasting golf turf areas as affected by Proxy, a Proxy + Primo Maxx tank-mix and Embark Lite applications, 2006; Purdue University. Ann. Rep. Purdue Univ. Turfgrass Sci. Prog. p. [1-11]. Bogart, ].E. 1972. Factors Influencing Competition ofAnnuaI Bluegrass (Poa Annua L.) Within Established Turfgrass Communities. MS. Thesis: Michigan State University, East Lansing, MI. Bogart, ].E. and ].B. Beard. 1973. Cutting height effects on the competitive ability of Annual Bluegrass (Poa annua L.). Agron. 1., 65, 513. Borger, ].A. 2008. Managing Poa annua seedheads on putting greens: Successful seedhead inhibition can improve spring playability of Poa annua putting greens. USGA Green Sec. Rec. 46(4):p. 7-8. Bradshaw, A.D. 1972. Some of the evolutionary consequences of being a plant. Evol. Biol. 5:25-47. Branham, B. E., and M. Collins. 1986. 1985 research update: [1. Timing of plant growth regulators for seedhead control]. p. 31-35. In Proceedings of the 56th Annual Michigan Turfgrass Conference. East Lansing, MI: january 13-15, 1986. East Lansing, MI: Michigan State University. Branham, B.E. 1989. 1988 turf weed control, PGR, and management studies: [111. Iron sources and Embark masking on Poa annua]. p. 4, 13-14. In Proceedings of the 59th Annual Michigan Turfgrass Conference. East Lansing MI: January 16-18, 1989. East Lansing, MI: Michigan State University. Branham, B.E. 1991. Dealing with Poa annua: Understanding the strength and weaknesses of annual bluegrass is the first step in developing a successful management program. Golf Course Manage. 59(9):46-60. Buettner, M.R., R.D. Ensign, and AA. Boe. 1976. Plant growth regulator effects on flowering of Poa pratensis L. under field conditions. Agron. ]. 68(2):p. 410- 413. 122 Buhler, D.D., M.L. Hoffman, and RN. Andersen. 1999. Andersen's Guide to Practical Methods of Propagating Weeds and Other Plants. 2nd ed. Lawrence, KS: Weed Science Society of America. p. 97. Carlson, Thomas Mark 1994. Pre and postemergence control of Poa annua in turf with ethofiimesate. MS. Thesis: Michigan State University. Christians, NE, and]. Nau. 1984. Growth retardant effects on three turfgrass species. ]. Am. Soc. Hortic. Sci. 109(1):p. 45-47. Cooper, R.]., P.R. Henderlong, and ].R. Street. 1984. Annual bluegrass management: Getting to the root of the problem. Golf Course Manage. 52(3):p. 39, 41, 43. Cooper, R.]., P.R. Henderlong, LR. Street, and K.]. Kamok. 1987. Root growth, seedhead production, and quality of annual bluegrass as affected by mefluidide and a wetting agent. Agron. ]. 79(5):p. 929. Danneberger, T.K. and j.M. Vargas jr. 1983. Predicting the onset of annual bluegrass seedhead production from degree-days. p. 78-81. In Proceedings of the 53rd Annual Michigan Turfgrass Conference. East Lansing. MI: january 18-19, 1983. East Lansing : Michigan State University. Danneberger, T.K. and 1M. Vargas Jr. 1984. Annual bluegrass seedhead emergence as predicted by degree-day accumulation. Agronomy journal 76:756-758. Danneberger, T.K., B.E. Branham, and J.M. Vargas. 1987. Mefluidide applications for annual bluegrass seedhead suppression based on degree-day accumulation. Agron ]. 79:69-71. Dernoeden, RH. 1984. Four-year response of a Kentucky bluegrass-red fescue turf to plant growth retardants. Agron. ]. 76(5):p. 807-813. Diesburg, KL. 2000. Growth regulators boost density in different ways: More tillers vs. more leaves per tiller: Products offer distinct results. Golf Course Manage. 68(4):p. 61-63. Eggens, ].L., and C.P.M. Wright. 1985. Kentucky bluegrass and annual bluegrass response to ethephon]. Am. Soc. Hortic. Sci. 110(5):p. 609-611. 123 Eggens, ].L., and DP. Ormrod. 1982. Creeping bentgrass, Kentucky bluegrass and annual bluegrass seed germination response to elevated temperature. Guelph Turfgrass Inst Res. Rep. p. 5. ' Eggens, ].L., C.P.M. Wright, and K. Carey. 1989. Use of Mefluidide for Turf Overseeding. HortScience. 24(2):p. 300-302. Ellis, WM. 1973. The breeding system and variation of populations of Poa annua L. Evolution 27:656-662. Engel, RE. 1967. Temperatures required for germination of annual bluegrass and colonial bentgrass. Golf Superintendent. 35(9):20, 23. ' Fidanza, M.A., P.H. Dernoeden, and M. Zhang. 1996. Degree-days for predicting smooth crabgrass emergence in cool-season turfgrasses. Crop Sci. 36(4):p. 990-996. Gadgil, M. and O.T. Solbrig. 1972. The concept of r- and K-selection: evidence from wild flowers and some theoretical considerations. Am. Nat, 106, 14. Gaussoin, RE. and B.E. Branham. 1989. Influence of cultural factors on species dominance in a mixed stand of annual bluegrass/creeping bentgrass. Crop Sci. 29(2):480-484. Gelernter, W. and LJ. Stowell. 2001. Advances in Poa seedhead management: Ethephon, a new tool in Poa annua seedhead suppression, has been used successfully on the West Coast. Golf Course Manage. 69(10):p. 49-53. Gelernter, W.D., and L.]. Stowell. 2000. Annual bluegrass seedhead management using ethephon and ethephon plus trinexapac-ethyl on golf course greens and fairways. Annu. Meet Abstr. p. 155. Gibeault, VA. 1971. Perenniality in Poa annua L. Ph.D. dissertation, Oregon State University, Corvallis, OR. Gibeault, VA and N .R. Goetze. 1972. Annual meadowgrass. journal of the Sports Turf Research Institute 48:9-19. Gillmore, EC. and ].S. Rogers. 1958. Heat units as a method of measuring maturity in corn. Agron. ]. 50:611-615. 124 Grime, ].P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111:1169-1194. Harper, ].C. 1965. History of Poa annua. Turf Conference Proceedings. University of Massachusetts. pp. 5-10. Heng, DA, and DB. White. 1969. Investigations into the Activity of Ethrel (2- Chloroethylphosphonic acid) in the Growth of Poa Pratensis. Annu. Meet. Abstr. 61:p. 54. Hovin, A.W. 1957. Variations in annual bluegrass. The Golf Course Reporter 25(7): 18-19. Hovin, A.W. 1958. Reduction of self-pollination by high night temperature in naturally self-fertilized Poa annua. Agronomy journal 50:369. Huff, DR 1999. For richer, for Poa: cultivar development of greens-type Poa annua. USGA Green Section Record 37(1):11-14. Hutchinson, CS. and GB. Seymour. 1982. The journal of Ecology, Vol. 70, N o. 3, (Nov., 1982). pp. 887-901 Inguagiato, I.C., ].A. Murphy, and 8.8. Clarke. 2008. Anthracnose severity on annual bluegrass influenced by nitrogen fertilization, growth regulators, and verticutting. Crop Sci. 48(4):p. 1595-1607. Jagschitz, ]A. 1984. Suppression of Poa annua L. seedheads in turfgrass areas. Proc. Annu. Meet. Northeast. Weed Sci. Soc. 38:p. 305-306. johnson, PG and DB. White. 1997a. Vernalization requirements among selected genotypes of annual bluegrass (Poa annua L.). Crop Science 37:1538-1542. ]ohnson, PG and DB. White. 1997b. Vernalization requirements among selected genotypes of annual bluegrass (Poa annua L.). Crop Science 37:1538-1542. johnson, P.G., B.A. Ruemmele, P. Velguth, D.B. White, and RD. Ascher. 1993. An overview of Poa annua reproductive biology. ln ternational Turfgrass Society Research journal 7:798-804. 125 ]uhren, M., W. Noble and F.W. Went. 1957. The standardization of Poa annua as an indicator of smog concentrations. I. Effects of temperature, photoperiod and light intensity during growth of test plants. Plant Physiology 32:576-586. Kaminski, IE, and PH. Dernoeden. 2007. Seasonal Poa annua L. seedling emergence patterns in Maryland. Crop Sci. 47(2):775-781. Kane, R., and L. Miller. 2003. Field testing plant grth regulators and wetting agents for seedhead suppression of annual bluegrass. [Online]USGA Turfgrass Environ. Res. Online. 2(7):p. [1-9]. Koch, W. 1968. Environmental factors affecting the germination of some annual grasses. Proc. of the 9th British Weed Control Confi 9:14-19 Kopec, D.M., J. Gilbert, and M. Bates. 2004. Application of proxy PGR for Poa seed head suppression, 2000. Turfgrass Landscape Urban IPM Res. Summ. p. [1-8]. Koshy, T.K. 1968. Evolutionary origin of Poa annua L. in light of karyotypic studies. Canadian journal of Genetic Cytology 10:112-118. Koshy, T.K. 1969. Breeding systems in annual bluegrass, Poa annua L. Crop Science 9:40-43. Law, R.A., D. Bradshaw and PD. Putwain. 1977. Life history and variation in Poa annua. Evolution 31:233-246. Lush, W.M. 1988a. Biology of Poa annua in a temperate zone golf putting green (Agrostis stolonifera/Poa annua) I. The above ground population. journal of Applied Ecology 25:977-988. Lush, W.M. 1988b. Biology of Poa annua in a temperate zone golf putting green (Agrostis stolonifera/Poa annua) II. The seed bank. journal oprplied Ecology 25:989-997. Matteson, ].W., E.S. Ratcliffe, and D. R. Pauly. 1976. EMBARK Plant Growth Regulator- A New Suppressant for Control of Grass Growth. Annu. Meet. Abstr. 68:p. 6. McElroy, 15., RH. Walker, and GR. Wehtje. 2004. Annual bluegrass (Poa annua) populations exhibit variation in germination response to temperature, photoperiod, and fenarimol. Weed Science. 52:47-52. 126 Nannfeldt, JA. 1937. The chromosome numbers of Poa. sect. Ochlopoa A and Gr. and their taxonomic significance. Botaniska Notiser 1937:238-257. Parups, E.V., and WE. Cordukes. 1977. Growth of turfgrasses as affected by Atrinal and Embark. HortScience. 12(3):p. 258-259. Petrovic, A.M., R.A. White, and M. Kligerman. 1985. Annual bluegrass growth and quality as influenced by treatments of grth retardants and wetting agents. Agron. J. 77(5):p. 670-674. Renney, A]. 1964. Preventing Poa annua infestations. In Proceedings of the 18th Annual Northwest Turfgrass Conference. Gleneden Beach, Oregon. pp. 3-5. Roberts, HA. and PM. Feast. 1973. Emergence and longevity of seeds of annual weeds in cultivated and undisturbed soil. journal of Ecology 10:133-143. Saruckhan, J. 1974. Studies on plant demography: Ranunculus repens L., R bulbosus L. and R acris L.: II Reproductive strategies and seed population dynamics. journal ofEcology 62(1):151-177. Schmidt, R.E., and SW. Bingham. 1977. Chemical grth regulation of 'Baron' Kentucky bluegrass. Agron. J. 69(6):p. 995-1000. Schott, P.E., H. Will, and H.H. Nolle. 1980. Turfgrass grth reduction by means of a new plant growth regulator. Int. Turfgrass Soc. Res. J. p. 325-328. Scott, S.J., R.A. Jones, and WA. Williams. 1984. Review of data analysis methods for seed germination. Crop Sci. 24:1192-1199. Shem-Tov, S. and SA. Fennimore. 2003. Seasonal changes in annual bluegrass (Poa annua) germinability and emergence. Weed Science. 51:690-695. Stowell, L]. and W. Gelernter. 2000. Converting poa greens to bentgrass. [Online] Available at http: / /www.paceturf.org/ PTRI / Documents /Weeds /poa2bent96.pdf. San Diego, California: PACE Turfgrass Research Institute Timm, G. 1965. Biology and systematics of Poa annua. FeitschriftfurAckerund Pflanzenbau 122(3): 267-294. Tutin, T.G. 1952. Origin of Poa annua. Nature 169:160. 127 Tutin, T.G. 1957. A contribution to the experimental taxonomy of Poa annua L. Wastonia 4:1-10. Uva, R.H., J.C. Neal, and J.M. DiTomaso. 1997. Weeds of the Northeast. Cornell University Press. Ithaca, N .Y. p. 78 Vargas, J.M. Jr. and A]. Turgeon. 2004. Poa annua: Physiology, culture and control of annual bluegrass. John Wiley & Sons, Hoboken, NJ. Warwick, 3.1. 1979. The biology of Canadian weeds. 37. Poa annua L. Canadian Journal of Plant Science. 59:1053-1066. Warwick, SJ. and D. Briggs. 1978a. The genecology of lawn weeds. I. Population differentiation in Poa annua L. in a mosaic environment of bowling green lawns and flower beds. New Phytol. 81(3):p. 711-723. Warwick, SJ. and D. Briggs. 1978b. The genecology of lawn weeds. II. Evidence for disruptive selection in Poa annua L. in a mosaic environment of bowling green lawns and flowerbeds. New Phytol. 81(3):p. 725-737. Watschke, T.L., and J.A. Borger. 2005. Seedhead suppression on putting greens. Proc. Annu. Meet. Northeast. Weed Sci. Soc. 59:p. 107. Watschke, T.L., D.J. Wehner, and J.M. Duich. 1977. Initial and residual effects of growth regulators on a Pennstar-Flyking Kentucky bluegrass blend. Proc. Annu. Meet. Northeast. Weed Sci. Soc. 31:p. 378-389. White, D.B., D. Heng, T.B. Bailey, and L. Foote. 1969. Chemical regulation of growth in turfgrass. Proc. Int. Turfgrass Res. Conf. 1:p. 481-492. Wu, L., I. Till-Bottraud, and A. Torres. 1987. Genetic differentiation in temperature- enforced seed dormancy among golf course populations of Poa annua L. New Phytol. 107(3):623-631. Yang, S., J. Logan, and D.L. Coffey. 1995. Mathematical formulae for calculating the base temperature for growing degree-days. Agricultural and Forest Meteorology. 74:61-74. 128 llllllljllilll“ 63 5 53 "‘iiliil 1293