Longevity of turf response to urea, coated urea, and blends

Transcription

1 Longevity of turf response to urea, coated urea, and blends K. Carey, A.J. Porter, K.S. Jordan and E.M. Lyons Department of Plant Agriculture and the Guelph Turfgrass Institute, University of Guelph, Ontario. The objective of this research project was to quantify turf response to a one-time application of urea-based fertilizers on Kentucky bluegrass turf on a soil rootzone. Data collected included the duration and strength of the color response following application of the tested products, turf quality, uniformity, and density, and resistance of the turf to disease and other stresses. MATERIALS / METHODS The treatments consisted of the sponsor s products at specified rate and application program (Tle 1). An unfertilized check treatment was also included. Treatments were applied to 1 x 3 m plots of Kentucky bluegrass turf maintained as a home-lawn type turf on the research ranges at the Guelph Turfgrass Institute (mowing at 40 mm, irrigation to prevent stress) (Figure 1). Treatments were replicated four times in a randomized complete block design. Treatments were applied June 7, 2011 according to the recommended programs. Color response of the turf to treatments was assessed pre-treatment, and then on a weekly basis, both visually and using instrumental color (canopy reflectance normalized-difference vegetation index using an Ntech Greenseeker). Uniformity of the color response was assessed visually using a scale of 1 to 9 (1=dead, 9=ideal, 5=acceptle). Plots were rated for turf quality, density and uniformity. Clippings were collected at 2, 4, 6, 8, 10, 12, 14, and 16 weeks after treatment, dried and weighed to determine shoot dry matter accumulation per unit area. Soil temperature at 5 cm depth was monitored with Spectrum WatchDog data loggers, and reported as daily mean. Other stresses were measured as they occurred (disease, weed, drought). Spring greenup will be assessed in April An anecdotal photographic record of the experiment was kept. All measurements were analyzed by appropriate statistical analyses (general linear models). RESULTS Environmental data. Daily air and soil temperatures for June October 2011 are presented in Figures 2 and 3. Visual ratings. There were significant differences in visual ratings of turf colour by 6 Tle 1. Treatments Trt # Description 1. Untreated control % urea 3. 15% XCU, 85% urea 4. 30% XCU, 70% urea 5. 50% XCU, 50% urea % XCU 7. AAT DS w/ 40% XCU 8. 15% SCU, 85% urea 9. 30% SCU, 70% urea % SCU, 50% urea % SCU 12. Lesco w/ 40% SCU Application program All treatments were applied once (June 7, 2011) at 1 lb / 1000 sq ft actual N (4.88 g m -2 ) Figure 1. Plot area July 2, 2011 (25 ). Guelph Turfgrass Institute 2011 Annual Research Report 1

2 (Tle 2). The differences persisted until 5 weeks after treatment, but by 7 weeks after treatment they were no longer significant. Other visual performance ratings (quality, uniformity, and density) generally did not differ significantly, though the mean value for untreated control plots was low. Canopy reflectance. The canopy reflectance (normalized-difference vegetation index) data collected with the Greenseeker gave a very precise picture of the response to the treatments. Index values were calculated both as the raw NDVI values and as values corrected by subtracting the value of the untreated control to Tle 2. Visual ratings of treated plots. Treatment pre Colour 100% SCU c 7.00 a 7.50 a 7.75 a a 100% urea c 7.25 a 7.75 a a 100% XCU c a 7.75 a % SCU, 85% urea c 7.25 a 7.50 a a 15% XCU, 85% urea c 7.25 a 7.75 a a 30% SCU, 70% urea a % XCU, 70% urea bc a a 50% SCU, 50% urea a 7.25 a 7.75 a a 50% XCU, 50% urea c 7.00 a 8.25 a 8.00 a a AAT DS w/ 40% XCU a 7.50 a b Lesco w/ 40% SCU c a a Untreated control bc 5.25 b 6.13 b 6.50 b msd p= NS NS 0.71 Quality 100% SCU a % urea % XCU % SCU, 85% urea % XCU, 85% urea a % SCU, 70% urea a % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea a AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control b msd p= NS NS NS 1.18 NS NS NS Uniformity 100% SCU % urea % XCU % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control msd p= NS NS NS NS Density 100% SCU % urea % XCU % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control msd p= NS NS NS NS 1 Visual ratings 0-10, 10 = best, 6 = acceptle. Means of 4 replicates; means within columns followed by the same letter are not significantly different (Tukey s HSD test, p=). 2 Guelph Turfgrass Institute 2011 Annual Research Report

3 remove background variation, since the NDVI value is affected by mowing, moisture status, and other factors in addition to nitrogen status. Figure 4 shows the pattern of change of the raw NDVI values (averaged across all plots) and the NDVI values (averaged across all non-control plots) during the experiment C=10 C=37.8 C=50 C=75 NDVI mean NDVI mean Figure 4. Changes in overall mean NDVI (black) and NDVI corrected to remove control value (blue) during the experiment. There were significant differences in canopy reflectance among the treatments beginning 6 and lasting until 41 (Tle 3). By 13 all treatments had significantly larger canopy reflectance values than the untreated control. Generally there were no strong patterns among the fertilizer treatments in NDVI values per se. The NDVI values, when plotted over time, allowed some differentiation among the fertilizer treatments in terms of release characteristics as detected by canopy reflectance. Replicate mean values of NDVI were tested against various curves to determine which functions had potential to adequately describe the responses. The online curve fitting and surface fitting web site at was used to investigate families of curves. One of the best functions to fit the data was a compound exponential function NDVI = 4*A * e (-/ C) *(1-e (-/C) ), in which there are two fitted parameters: A, which varies with maximum NDVI, and C, which varies with days to maximum NDVI (Figure 5). The suitility was judged based on the combination of goodness of fit, minimum number of parameters, and interpretility of the parameters A=.11 A=.2274 Figure 5. Families of curves of the function NDVI = 4* A * e (-/C) * (1 - e (-/C) ) illustrating the effects of varying the parameters A and C. The curve with A= and C=37.8 is the actual curve fitted to the 100% SCU treatment. The NDVI values for each treatment were fitted to these curves using GraphPad Prism, and the estimates of A and C for each treatment were compared using ANOVAs. The parameter estimates of the fitted curves are shown in Tle 4 and Figure 6, and the fitted curves are shown in Figures Tle 4. Multiple comparisons of estimated parameters for fitted curves of NDVI. Treatment A C 100% SCU 66 a de 100% urea a 100% XCU c 15% SCU, 85% urea a 15% XCU, 85% urea c 30% SCU, 70% urea b c 30% XCU, 70% urea bcd 50% SCU, 50% urea b c 50% XCU, 50% urea cde AAT DS w/ 40% XCU b Lesco w/ 40% SCU e 1 Parameters followed by the same letter are not significantly different (Tukey s Multiple Comparison Test, p=) A=.5 A=1 Guelph Turfgrass Institute 2011 Annual Research Report 3

4 Tle 3. Canopy reflectance in treated plots Treatment pre % SCU a % urea a a a % XCU b 15% SCU, 85% urea a a % XCU, 85% urea a a a % SCU, 70% urea a a a a a 30% XCU, 70% urea a a % SCU, 50% urea a a a % XCU, 50% urea a a a b AAT DS w/ 40% XCU a a a a Lesco w/ 40% SCU a a a Untreated control b b b b c msd p= NS NS NS NS NS % SCU a a a a a a a % urea a a a a a a a % XCU a a a a a a a a 15% SCU, 85% urea a a a a a b b a a b 15% XCU, 85% urea a a a a a a a % SCU, 70% urea a a a a a a a % XCU, 70% urea a a a a a a a % SCU, 50% urea a a a a a a a % XCU, 50% urea a a a a a a a AAT DS w/ 40% XCU a a a a a a a a a Lesco w/ 40% SCU a a a a a a a Untreated control b b b b b c c b b c msd p= % SCU a a a a a % urea a a a % XCU a a a a a a % SCU, 85% urea a a b b a % XCU, 85% urea a a a a a % SCU, 70% urea a a a a a % XCU, 70% urea a a a a a % SCU, 50% urea a a a a a % XCU, 50% urea a a a a a a AAT DS w/ 40% XCU a a a a a a Lesco w/ 40% SCU a a a a a a Untreated control b b c c b b b b msd p= NS NS % SCU % urea % XCU % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control msd p= NS NS NS NS NS NS NS NS NS NS % SCU % urea % XCU % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control msd p= NS NS NS NS NS NS NS NS NS NS % SCU % urea % XCU % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU Lesco w/ 40% SCU Untreated control msd p= NS NS NS NS NS NS NS NS 1 Normalized-difference vegetation index: mean of 4 replicates; means within columns followed by the same letter are not significantly different (Tukey s HSD test, p=). Readings in bold are from dates with significant treatment effects. 4 Guelph Turfgrass Institute 2011 Annual Research Report

5 100% urea 15% XCU 85% urea 30% XCU 70% urea 50% XCU 50% urea 100% XCU AAT DS w/ 40% XCU 15% SCU 85% urea 30% SCU 70% urea 50% SCU 50% urea 100% SCU Lesco w/ 40% SCU NDVI maximum (A) Figure 6. Parameters A and C for curves fitted to fertilizer response as estimated by NDVI. Parameter estimates are all significantly different except where a common letter is present on the bars (Tukey s multiple comparison test, p=). a b b b a c bcdcde a c c c de Days to NDVI maximum (C) e - 100% SCU 100% XCU 100% urea - 30% SCU 70% urea 30% XCU 70% urea Figure 7. Curves fitted to fertilizer response as estimated by NDVI. See Tle 4 for estimates of A (max NDVI) and C Figure 9. Curves fitted to fertilizer response as estimated by NDVI. See Tle 4 for estimates of A (max NDVI) and C 15% SCU 85% urea 15% XCU 85% urea - Figure 8. Curves fitted to fertilizer response as estimated by NDVI. See Tle 4 for estimates of A (max NDVI) and C - 50% SCU 50% urea 50% XCU 50% urea Figure 10. Curves fitted to fertilizer response as estimated by NDVI. See Tle 4 for estimates of A (max NDVI) and C (days to max NDVI). Points are means of 4 replicates; curves Guelph Turfgrass Institute 2011 Annual Research Report 5

6 - AAT DS w/ 40% XCU Lesco w/ 40% SCU Figure 11. Curves fitted to fertilizer response as estimated by NDVI. See Tle 4 for estimates of A (max NDVI) and C Shoot growth. Clippings were collected periodically to estimate shoot dry matter accumulation. Plots were mowed to 40 mm and then 5 7 days later clippings were collected from a 0.35 m 2 strip (Figure 12). Although there were differences in growth, the noisiness of the data meant that the differences were only significant on the date 4 weeks after application when the maximum fertilizer release was happening (Tle 5). The general pattern of increase and decline in growth rates was similar to the response curves as estimated by NDVI. Comparing the plot means for growth with NDVI shows this relationship (Figure 13), and NDVI appears to be a good proxy for shoot growth under these conditions. Figure 12. Clipping collection: border strips were mowed short (<40 mm) and clippings were then collected from a 0.94 m strip lengthwise in each plot using a Gardena electric reel mower set at 42 mm height of cut NDVI Figure 13. Relationship between increase in canopy reflectance and shoot growth as estimated by dry matter accumulation. Points are plot means; all clipping collections data are plotted. Dry matter accumulation is log-transformed to give a linear relationship. Large red points are the data from 27, which was the only date when the differences in growth were statistically significant. Tle 5. Dry matter accumulation. Treatment g m % SCU % urea % XCU a % SCU, 85% urea % XCU, 85% urea % SCU, 70% urea % XCU, 70% urea a % SCU, 50% urea % XCU, 50% urea AAT DS w/ 40% XCU a Lesco w/ 40% SCU Untreated control b msd p= NS 2.12 NS NS NS NS NS 1 Clippings collected from 0.38 x 0.94 m strip of each plot, mowed at 42 mm after 5-7 days of growth. log 10 growth (g m -2 ) Guelph Turfgrass Institute 2011 Annual Research Report

7 DISCUSSION AND CONCLUSIONS All treatments gave a significant improvement in colour and growth compared to the untreated control. The fertilizer effects were observle within a week after treatment by the canopy reflectance data, and persisted in significant amounts until 7 weeks after treatment. The average gain of fertilized treatments over control was out 2 ranks on the visual colour rating scale (6 to 8), or out 0.07 units on the canopy reflectance index. The untreated control plots were at an acceptle colour and quality level (>5) through most of the trial. There was no strong or consistent pattern date by date distinguishing the fertilizer treatments from one another, either in visual ratings, or canopy reflectance, or growth. Using the release curves fitted to the seasonal pattern of NDVI suggests that the ranking of the fertilizer treatments for strength of response (A) was 100% SCU, 15% SCU < 85% urea, 100% XCU, 30% XCU/70% urea, Lesco w/ 40% SCU < 15% XCU/85% urea, 50% XCU/50% urea < 100% urea < 50% SCU/50% urea < 30% SCU/ 70% urea < AAT DS w/ 40% XCU. Similarly, the ranking of the treatments for days to maximum release (C) was 100% urea < 15% SCU/85% urea < AAT DS w/ 40% XCU, 100% XCU < 50% SCU/50% urea, 30% SCU/ 70% urea < 15% XCU/85% urea < 30% XCU/ 70% urea < 50% XCU/50% urea < 100% SCU < Lesco w/ 40% SCU. Differences in growth were only apparent when fertilizer response was near its maximum, out 4 weeks after treatment. At this point the treatments with the highest growth rate had out an 80% increase over the untreated control. Sponsor: Agrium Advanced Technology Guelph Turfgrass Institute 2011 Annual Research Report 7