Soil-water-use Characteristics of Precision-irrigated Buffalograss and Kentucky Bluegrass

J. Ryan Stewart, Department of Horticulture, Iowa State University, 106 Horticulture Hall, Ames 50011-1100; Roger Kjelgren and Paul G. Johnson, Department of Plants, Soils, and Biometeorology, Utah State University, 4820 Old Main Hill, Logan 84322-4820; Michael R. Kuhns, Department of Forest, Range, and Wildlife Sciences, Utah State University, 5230 Old Main Hill, Logan 84322-5230.

Corresponding author: Roger Kjelgren. rkjel@mendel.usu.edu

Abstract

As landscape water conservation becomes more important in the American West, pubic interest in using low water-use turfgrasses is increasing. Little is known about soil water extraction characteristics that contribute to low water use. We investigated how buffalograss (Buchloë dactyloides (Nutt.) Engelm) and Kentucky bluegrass (Poa pratensis L.), considered to be low and high water-use species, respectively, extract soil water in terms of rooting depth and use of available water. Leaf canopy temperature, air temperature, and vapor pressure deficit (VPD) were measured, and a relationship developed between leaf canopy temperature minus air temperature (T_L-T_A) and VPD for each species under well-watered conditions. This regression was significant within each species but not different between the two species. During soil drying, changes in soil water content were tracked to incipient water stress where T_L-T_A increased above the well-watered T_L-T_A:VPD relationship. Within seven days of soil drying, Kentucky bluegrass reached incipient water stress when nearly 50% of the total water was depleted in its 0.6-m-deep root zone. Buffalograss, however, reached incipient water stress after 22 days of soil drying, when it had depleted nearly 60% of soil water to a 0.9-m depth. Ninety-four percent of the Kentucky bluegrass root system was in the top 0.3 m of the soil compared to 62% for buffalograss. These results present rooting depth and water extraction patterns of these species that can be used to determine more precise irrigation scheduling in the West.

Introduction

Over half of the treated municipal and industrial water for many cities in the high-elevation, arid western United States is applied to amenity landscape vegetation (15). Turfgrass requires significant water inputs, in part because the turf species used are more suited to high rainfall regions of the eastern United States and northern Europe (15) and because of the large area devoted to turfgrasses. Kentucky bluegrass, the most commonly used turfgrass in much of the western United States, requires frequent irrigation to maintain appearance during hot, arid summers. Buffalograss is a sod-forming species native to the Great Plains region of North America and is promoted as a low-water-use alternative to Kentucky bluegrass (19). Buffalograss is known to use water more efficiently than most cool-season grasses in part due to its C₄ metabolism (24). Additionally, its more glaucous foliage can potentially help maintain lower leaf canopy temperatures and transpiration (14), and a reportedly deeper root system allows buffalograss to extract water from a larger soil volume (12,16,26) than Kentucky bluegrass.

Although there are species differences in water use, additional water savings for turfgrass management is achieved through precision irrigation through which turfgrass is allowed to deplete soil water to the point of incipient water stress within its root zone before irrigation (9). Incipient water stress is the point at which the turfgrass begins to show signs of water stress such as leaf firing and chlorosis, and thus can be used to schedule when to apply water. The amount of water within the rooting profile, a function of rooting depth, can be used to schedule the amount of water to apply. This water-budget approach maximizes efficient use of water within the turfgrass rooting profile.

Little is known, however, about the interaction of incipient water stress and rooting characteristics of buffalograss for irrigation scheduling in a managed landscape situation. Deep rooting potential has been correlated to greater drought avoidance among and within a number of species (5,21). However, there is little information on the amount of water turfgrass species can deplete at incipient water stress. Leaf canopy temperature offers a relatively clear and easily determined signal of water stress because when stomata close, transpirational cooling decreases and leaf canopy temperature increases relative to air temperature (13). Using leaf canopy temperature to detect water stress for irrigation has been applied to turfgrass in other studies (11,17).

The first objective of this study was to characterize the relationship between foliage and air temperatures of buffalograss and Kentucky bluegrass under well-watered conditions and during a period without irrigation to determine the soil water content at the point of incipient water stress for both species. The second objective of this study was to characterize and compare root depth and density of Kentucky bluegrass and buffalograss.

Measurement of Water Use and Soil-Water Extraction

This study was conducted at the Greenville Research Farm in North Logan, Utah, from 6 July to 31 August 1999 on a Millville silt loam (coarse-silty, carbonatic, mesic Typic Haploxerolls). In spring 1998, twenty plots, each 36 m², were arranged in a randomized complete block design. Eight plots were seeded with ‘Tatanka’ buffalograss, eight with Kentucky bluegrass (varieties were not stated on the seed source), and four plots were non-vegetated controls covered with coarse wood mulch 0.1 m deep so that changes in soil water content without plant extraction could be followed. A sprinkler system with gear-drive heads was installed in each of the 16 plots with one sprinkler head at each of the four corners of each plot. After establishment in 1998, in 1999 the plots were fertilized with N at 100 kg/ha/year in two applications between early June and mid-July using ammonium sulfate. Plots were mowed every 7 days to a height of 8 cm, and the clippings were removed. Soil phosphorous and potassium were adequate.

Through late July 1999, the plots were kept well watered by irrigating every 3 days to replace lost soil water, estimated based on reference evapotranspiration (3). A distribution-uniformity test was conducted to determine average application rate and the correction factor used to account for areas of uneven application (18). Distribution uniformity was 63%, averaged over all the plots.

Beginning in June 1999, leaf canopy temperature (T_L) was measured using infrared thermocouple transducers (model IRT-S; Apogee Instruments, Logan, Utah) suspended 1 m above the canopy at each of the turfgrass plots (Fig. 1), allowing a field-of-view of 2 m in diameter. The transducers were connected to a data logger (model CR10X; Campbell Scientific, Logan, Utah) via a multiplexer (model AM25T; Campbell Scientific), scanned every 10 seconds and canopy temperature was logged hourly. Vapor pressure deficit (VPD) of the air and air temperature (T_A) was also recorded with a VPD/TA sensor (model CS500; Campbell Scientific) located 1.5 m above the ground in the center of the study area and also logged hourly. Reference evapotranspiration (ET_o) was calculated from an automated weather station 0.3 km away using the Penman-Monteith UN-FAO-56 equation (1).

Fig. 1. Positioning of the infrared thermocouple transducer.

Hourly leaf canopy temperature data from the month of July were collected under full-sun conditions for the hours between noon and 3 p.m. (MDST) under maximum daily insolation concurrent with VPD to develop a predictive relationship under well-watered conditions. Hourly data were stratified by days after irrigation, time of day, then T_L minus T_A (T_L-T_A) was regressed on the paired VPD. The resulting regression model was used to predict well-watered turf T_L-T_A at a given VPD for comparison against T_L-T_A values under water-stressed conditions after we initiated dry-down treatments. This formulation is similar to the simplified version of the crop-water-stress index (13), which was previously applied to turfgrass (11).

Irrigation stopped on 28 July and leaf canopy temperature, air temperature, and air VPD under maximum insolation were monitored daily until the dry-down was terminated. Each turfgrass plot also was visually rated three times each week during the dry-down based on turfgrass quality (1 = worst possible quality, 10 = best possible quality) (22) and wilting as observed by color and general appearance.

During this period without irrigation, soil water content was measured using a Hydroprobe neutron depth moisture gauge (model 503 DR; Campbell Pacific Nuclear, Martinez, CA) in access tubes (0.04-m-diameter PVC pipe) located 1.5 m from the center of each plot and installed vertically to a depth of 2 m in 0.3-m increments. Measurements were collected during predawn on 29 July (to allow for drainage from irrigation the previous day), and on 4, 12, 18, 23, and 30 August. Water use by both species was measured as the change in volumetric water content for a given time period. A volumetric water-content sensor (HydroSense CS620; Campbell Scientific) was used to measure soil water content in the top 0.15 m of soil, as the neutron moisture gauge cannot measure volumetric water content close to the soil surface (20). Soil water content was monitored twice each week with the neutron moisture gauge and the water-content sensor to determine the equivalent depth of soil water remaining after extraction by both species at incipient water stress. Equivalent water depth for each soil increment depth, D_e, is:

D_e = D_v (Equation 1)

where D is soil increment (m) and _v is volumetric water content.

Incipient water stress for both species was determined as the point where the actual foliage minus air temperature difference began to deviate from the predicted foliage minus air temperature difference as a result of reduced transpirational cooling due to stomatal closure. Soil water content associated with incipient water stress was calculated by summing the water content for all soil increments where there was a detectable change:

Fractional equivalent water depth at incipient water stress	= 1 -	D_v - D_wp	(Equation 2)

		D_fc - D_wp

where _wp is water content at wilting point, and _fc is water content at field capacity. Field capacity (at -0.03 MPa) and wilting point of the soil (at -1.5 MPa) were 0.25/m·m and 0.10/m·m for this particular soil type, respectively (20).

In mid-October 2000, vertical root cores were taken with a 0.15-m soil corer (0.08-m diameter) to a depth of 1.2 m in 0.15-m increments in each plot to measure rooting depth. Soil cores were placed in plastic bags and stored at 4°C until the roots were separated from the soil in a root washer 7 days later. Root samples were dried at 60°C for 12 h and then weighed.

We performed regression analysis on T_L-T_A versus VPD data (SigmaPlot ver. 8.0, SPSS Inc., Chicago, IL). Equality of slopes of regression lines was compared with a t-test. Slopes for the first day after irrigation were evaluated, and then species slopes were compared. Soil water content during the period with irrigation was analyzed with a one-way analysis of variance (ANOVA) that compared the two species. Root mass was also analyzed with a one-way ANOVA comparing species (SigmaStat, SPSS Inc., Chicago, IL). Means were separated based on the Tukey-Kramer test (SigmaStat, SPSS Inc., Chicago, IL).

Leaf Minus Air Temperature and Quality

Non-water-stressed T_L-T_A and VPD were negatively correlated for buffalograss (Fig. 2A) and Kentucky bluegrass (Fig. 2B) on the second day after an irrigation of four well-watered periods. As VPD reached maximum during mid-day (peak insolation around solar noon), transpiration increased, cooling the leaves relative to the air. Data from the day after an irrigation were used because we found that excessive surface evaporation from the turfgrass canopy of both species on the day of an irrigation altered the T_L-T_A slopes. Although buffalograss had a lower intercept and slope than Kentucky bluegrass, the regression of T_L-T_A versus VPD slopes for the species were not different (t = 0.28, df = 160), indicating the two species responded similarly to changing climatic conditions during the day. Some of the scatter in the data may have resulted from the sensors being placed 1.5 m above the turfgrass plots.

Fig. 2. Non-water-stressed baselines of buffalograss (A) and Kentucky bluegrass (B) determined from second-day irrigation data of four well-watered periods at solar noon during July 1999. Each symbol represents an individual data point.

After irrigation was ceased on 28 July, there was no precipitation and minimal cloud cover during the entire dry-down period. Under these conditions, T_L-T_A was a good indicator of water stress and species differences in drought resistance. Actual T_L-T_A of Kentucky bluegrass began to diverge from predicted T_L-T_A on 5 August, 7 days after irrigation was suspended (Fig. 3A), indicating its point of incipient water stress inducing stomatal closure and increased foliage temperature above the predicted T_L-T_A baseline (Fig. 3A). The visual rating of Kentucky bluegrass also markedly decreased shortly after irrigation ceased (Fig. 3A). The response of buffalograss differed significantly from that of Kentucky bluegrass (Fig. 3B). Actual T_L-T_A of buffalograss diverged from predicted values on 15 August, but did not remain consistent and was not reflected in a decline in turfgrass quality. Only near the end of the study period did buffalograss begin to show a slight decline in visual quality (Fig. 3B) correlating to incipient water stress which occurred on 20 August (Fig. 3B). However, actual T_L-T_A more clearly diverged from predicted on 23 August, and while visual rating declined during these three days, it was still quite visually acceptable. These data suggest that buffalograss is more able to maintain verdure under mild water stress.

Fig. 3. The difference in leaf canopy temperature versus air temperature (T_L-T_A) at solar noon and turfgrass rating (1 = worst possible quality, 10 = best possible quality) plotted against time of buffalograss (A) and Kentucky bluegrass (B) during a period without irrigation that began on 29 July 1999 and ended on 31 August 1999. Vertical bars represent standard error of the mean (standard error is smaller than symbol when error bar is not shown). Values are means of eight replications per species.

The relationship between T_L-T_A and VPD is more useful in arid climates, such as our study site in Utah, than in more humid climates (2,10). However the higher slope and more positive intercept observed in our work, compared to other investigations (13,23,25), may be due to environmental conditions (L. E. Hipps, personal communication; 11). The differences may also be due to air temperature and humidity not being measured at the turf canopy surface. In our work, these were measured at 1.5 m above the surface. The lowest T_L-T_A for Kentucky bluegrass in our work was -8°C at a VPD of 5 kPa (Fig. 2B), compared to -4°C at VPD of 5 kPa in Kansas (25). Since we used only one VPD/T_A sensor and did not want to place it over just one species, which may differ in surface heating during the day, we used a higher measurement point to obtain an integrated measure of VPD in the atmospheric layer relative to the canopy surface. The disadvantage of this sensor location was the measurement did not as closely represent the evaporative conditions at the turf surface.

Soil-water Extraction

Kentucky bluegrass had extracted nearly 50% of the available soil water within its root zone (to a depth of 0.6 m) when it experienced incipient water stress (4 August) (Fig. 4). By contrast, buffalograss reached incipient water stress on 20 August when it had extracted 56% of available soil water within its apparently deeper root zone (0.9 m) as compared to bluegrass, and 60% by 23 August (Figs. 4C and 5). This is consistent with other work that found buffalograss forestalls drought with deep roots that enhance water uptake in a larger volume of soil (12). In the top 0.15 m, bluegrass had a lower initial water content that was ostensibly due to evapotranspiration loss on 28 July. Subsequent depletion in the top soil layer during the dry-down was more rapid than buffalograss, but after three weeks both had depleted soil water down to about 6 mm of water, well below the wilting point for this soils (Fig. 4A).

Fig. 4. Changes in equivalent water depth at different soil depths during a period without irrigation that began on 29 July 1999 and ended on 31 August 1999, of buffalograss and Kentucky bluegrass. There was no significant extraction of water at depths of 0.9 to 1.8 m. Vertical bars represent SE of the mean (SE is smaller than symbol when error bar is not shown). The arrows at 4 August represent the incipient water stress point of Kentucky bluegrass. The arrows at 23 August represent the incipient water stress point of buffalograss. Values are means of eight replications per species and four replications of the control treatment.

Kentucky bluegrass extracted slightly less water than buffalograss at depths of 0.15 to 0.3 and 0.3 to 0.6 m (32% versus 38% of available water content) during the period without irrigation (Figs. 4B, 4C). Buffalograss, however, extracted significantly more water at depths of 0.6 to 0.9 m compared to bluegrass after three weeks of no irrigation (Fig. 4D). Equivalent water content under bluegrass at the depth of 0.6 to 0.9 m differed little from that of the mulched plots until late in the dry-down period, indicating minimal water extraction (Fig. 4D). Equivalent water depth of the mulched plots dropped slightly at depths of 0 to 0.15 m halfway into the period without irrigation, but did not change over time at any other depth. Below 0.9 m there were no differences among any treatment in equivalent water depth, indicating no extraction of soil water at those depths (data not shown).

Data with different Kentucky bluegrass varieties may be somewhat different than the results presented in this work due to differences in water-use efficiency and rooting. However, water-use differences within species are relatively small and inconsistent. Much larger differences occur between species (6).

During the dry-down period, buffalograss used 71 mm and Kentucky bluegrass used 53 mm of water, compared to a reference (ET_o) of 156 mm. Less total water use by Kentucky bluegrass during the study period was due to it becoming dormant after the first week, resulting in browning foliage and poor turfgrass quality that is less able to transpire. After the point of incipient water stress of Kentucky bluegrass, soil water extraction did continue deeper in the soil, using approximately 30% of ET_o. This water use did not help appearance as the turf was dormant, but this water use evidently helped maintain vitality in roots and the crown tissue.

Soil-water extraction patterns closely fit the patterns of root mass measured in the soil profiles. Kentucky bluegrass had greater root mass than buffalograss in the upper 0.15 m of soil (Fig. 5). Root mass was similar between the species at depths of 0.15 to 0.45 m, but buffalograss had significantly greater root mass at depths of 0.45 to 1.2 m (Fig. 5). While variation among root mass at depths between 0.9 and 1.2 m was not normally distributed, these root mass means are biologically significant because any roots at this depth indicate potential ability to take up water. No roots of Kentucky bluegrass were observed at these depths.

Fig. 5. Relative root mass percentage at different soil depths (A, B, C, D, and E) of roots of buffalograss and Kentucky bluegrass extracted in mid-October 2000 (* = significantly different at P < 0.05 level). Values are means of eight replications per species.

Buffalograss clearly distributes its roots differently and shows a different pattern of water use than does Kentucky bluegrass (27). Consistent with other work (4,8), we found that 95% of the Kentucky bluegrass root system was concentrated in the upper 0.3 m of soil. However, under the soil conditions of this study the Kentucky bluegrass root system was more truncated than reported elsewhere, with nearly 80% of its root system in the upper 0.15 m. Kentucky bluegrass has been reported to have deeper roots than shown in this work (7), but the reasons for this shallow root zone are not known, as no hardpan layer or other root-growth barrier exists in the soil we used. Although the roots were extracted in 2000, this high root concentration may explain the rapid decline in soil water in the upper layer during the period without irrigation in 1999. While bluegrass had depleted nearly 50% of the water in its root zone of 0 to 0.6 m, it appears it received the signal to go dormant from the top 0.3 m, where it had depleted 70% of the available water after seven days. The fewer bluegrass roots below 0.3 m were apparently not capable of supplying enough water to avoid water stress. Kentucky bluegrass continued to lose water during the period without irrigation, but after the onset of water stress, the consumption of deeper soil water did not benefit plant appearance and was probably enough to maintain a dormant state.

In contrast, buffalograss extended roots down to 0.9 m, as has been previously observed (16,26). The roots were deeper and less concentrated in the upper 0.15 m than were roots of Kentucky bluegrass. With only 45% of its root mass in the upper 0.15 m, buffalograss did not deplete the subsurface layer as rapidly; instead it drew upon water deeper in the soil, apparently through the dry-down.

Conclusions and Recommendations

While the results on leaf canopy temperature are only from one year, they are consistent with other results on these species (5,21). Thus, we have confidence that the fraction of extractable water at incipient water stress and rooting-depth values are valid and usable for precision-irrigation scheduling under the high-elevation, arid western United States conditions. These results for extractable water fraction and rooting depth can be used in irrigation scheduling of landscapes elsewhere in the western U.S. to calculate the total amount of water in bluegrass and buffalgrass root zones for given soil. When total amount of water in the root zone is combined with turf water loss estimated from ET_o, irrigation timing and amount can be reasonably calculated. Rooting depth would have to be adjusted in root zone truncated by an impervious layer or sharp change in texture, but the extractable water fraction values would likely still be applicable. Bluegrass could be considered a somewhat drought-resistant turf by virtue of going dormant more rapidly than buffalograss by draining soil water in its top 0.3 m of root depth. However, its continued use of water during a drought period, albeit at a lower rate, provides no aesthetic benefit and it is water that would need to be replaced to regenerate the turf after a drought period. While buffalograss used more water at each soil layer and deeper into the soil during the dry-down period, water that would also need replacing during post-drought regeneration, it is able to maintain a more acceptable appearance and thus would be a more preferable turf species where extended periods without water may occur.

Acknowledgments

This research was supported in part by the Utah Agricultural Experiment Station and the Center for Water Efficient Landscaping.

Literature Cited

1. Allen, R. G., Pereira, L. S., Raes, D., and Smith, M. 1998. Crop evapotranspiration: Guidelines for computing crop water requirements. UN Food and Agr. Org. Irr. and Drain. Paper 56. Rome, Italy.

2. Andrews, P. K., Chalmers, D. J., and Moremong, M. 1992. Canopy-air temperature differences and soil water as predictors of water stress of apple trees grown in a humid, temperate climate. J. Am. Soc. Hortic. Sci. 117:453-458.

3. Ashcroft, G. L., Jensen, D. T., and Brown, J. L. 1992. Utah climate. Utah Climate Center, Utah State Univ., Logan.

4. Beard, J. B. 1973. Turfgrass: Science and Culture. Prentice-Hall, Englewood Cliffs, N.J.

5. Bonos, S. A., and Murphy, J. A. 1999. Growth responses and performance of Kentucky bluegrass under summer stress. Crop Sci. 39:770-774.

6. Ebdon, J. S., Petrovic, A. M., and Zobel, R. W. 1998. Stability of evapotranspiration rates in Kentucky bluegrass cultivars across low and high evaporative environments. Crop Sci. 38:135-142.

7. Ervin, E. H., and Koski, A. J. 1998. Drought avoidance aspects and crop coefficients of Kentucky bluegrass and tall fescue turfs in the semiarid west. Crop Sci. 38:788-795.

8. Gist, G. R., and Smith, R. M. 1948. Root development of several common forage grasses to a depth of eighteen inches. Agron. J. 40:1036-1042.

9. Goldhamer, D., and Snyder, R. 1989. Irrigation scheduling: A guide to efficient on-farm water management. Univ. of Calif. Pub. 21454, Berkeley.

10. Hatfield, J. L. 1990. Measuring plant stress with an infrared thermometer. HortScience 25:1535-1538.

11. Horst, G. L., O’Toole, J. C., and Faver, K. L. 1989. Seasonal and species variation in baseline function for determining crop water stress indices in turfgrass. Crop Sci. 29:1227-1232.

12. Huang, B. 1999. Water relations and root activities of Buchloe dactyloides and Zoysia japonica in response to localized soil drying. Plant Soil 208:179-186.

13. Idso, S. B., Jackson, R. D., Pinter, P. J., Jr., Reginato, R. J., and Hatfield, J. L. 1981. Normalizing the stress-degree-day parameter for environmental variability. Agr. For. Meteorol. 24:45-55.

14. Jefferson, P. G., Johnson, D. A., and Asay, K. H. 1989. Epicuticular wax production, water status and leaf temperature in Triticeae range grasses of contrasting visible glaucousness. Can. J. Plant Sci. 69:513-519.

15. Kjelgren, R. K., Rupp, L., and Kilgren, D. 2000. Water conservation in urban landscapes. HortScience 35:1037-1040.

16. Klingenberg, J. P. 1992. Evaluation, genetic variation, and selection for improvement of a seeded, turf-type buffalograss population. (Abstr.) Ph.D. diss. 93-14407, University of Nebraska, Lincoln.

17. Martin, D. L., Wehner, D. J., and Throssell, C. S. 1994. Models for predicting the lower limit of the canopy-air temperature difference of two cool season grasses. Crop Sci. 34:192-198.

18. McCarty, L. B. 2001. Best Golf Course Management Practices. Prentice-Hall, Inc. Upper Saddle River, NJ.

19. Mintenko, A. S., Smith, S. R., and Cattani, D. J. 2002. Turfgrass evaluation of native grasses for the northern Great Plains region. Crop Sci. 42:2018-2024.

20. Or, D. 1990. Irrigation management considering soil variability and climatic uncertainty. (Abstr.) Ph.D. diss. 90-34086, Utah State University, Logan.

21. Qian, Y. L., Fry, J. D., and Upham, W. S. 1997. Rooting and drought avoidance of warm-season turfgrasses and tall fescue in Kansas. Crop Sci. 37:905-910.

22. Skogley, C. R., and Sawyer, C. D. 1992. Field research, Pages 589-614 in: Turfgrass. D. Waddington, R. Carrow, and R. Shearman, eds. Amer. Soc. Agron., Madison, Wis.

23. Slack, D. C., Kopec, D. M., Brown, P. W., Matthias, A. D., and Jalali-Farahani, H. R. 1990. Evaluation and comparison of crop water stress index models for irrigation scheduling of bermudagrass turf. Proc. Natl. Irr. Symp. 3:725-730.

24. Taiz, L., and Zeiger, E. 1998. Plant Physiology. Sinauer Assoc., Sunderland, Mass.

25. Throssell, C. S., Carrow, R. N., and Milliken, G. A. 1987. Canopy temperature based irrigation scheduling indices for Kentucky bluegrass turf. Crop Sci. 27:126-131.

26. Weaver, J. E. 1958. Summary and interpretation of underground development in natural grassland communities. Ecol. Monogr. 28:55-78.

27. Youngner, V. B., Marsh, A. W., Strohman, R. A., Gibeault, V. A., and Spaulding, S. 1981. Water use and quality of warm-season and cool-season turfgrasses. Proc. Int. Turf. Res. Conf. 4:251-257.