IRRIGATION OF HIGH MAINTENANCE TURF USING THE ARIZONA DEPARTMENT OF WATER RESOURCES WATER DUTY: EVALUATION OF TURF PERFORMANCE AND THE POTENTIAL FOR SOIL SALINIZATION by Brian Stephen Whitlark Thesis Submitted to the Faculty of the DEPARTMENT OF SOIL, WATER AND ENVIRONMENTAL SCIENCE In Partial Fulfillment of the Requirements For the Degree of MASTER OF SCIENCE In the Graduate College THE UNIVERSITY OF ARIZONA 1999 2 STATEMENT OF AUTHOR This thesis has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library. Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department of the Dean of the Graduate College when in his or her judgment proposed used of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author. APPROVAL BY THESIS DIRECTOR This thesis has 'een approved on the date shown below: 7//7 Paul W. Brown Specialist, Soil, Water and Environmental Science IKte 3 ACKNOWLEDGEMENTS I would like to express my sincere thanks and utmost respect to Dr. Paul Brown. We spent countless hours revising and reviewing this thesis, and I want to thank him for his patience and time. I would also like to thank Dr. Donald Post and Dr. Tom Thompson for their guidance and friendly comments during my career here at the University of Arizona. Thanks are also extended to Dr. Peter Wierenga -- always there to lend a helping hand and provide keen advice. I want to thank Dr. James Riley for getting me started in teaching. I am grateful to Teles Machibya for providing tremendous help in weekly data analysis and installing the potable pump system. I thank Randy Post and Blake Elliot for helping with weekly lysimeter data collection. I want to sincerely thank two members of my family that aided in a portion of the statistical analysis and mathematical computations. My Uncle, David Whitlark, was instrumental in providing me with a method to statistically analyze the Kc data. My father, Don Whitlark, helped with various computations throughout the study. Lastly, but certainly not least, I want to send a sincere thank you to the faculty and staff at the Karsten Desert Turfgrass Research Facility. Dr. David Kopec provided helpful advice on turf management; namely the late winter overseeding process. Alison Maricic was kind enough to allow me to utilize the laboratory for storing and analyzing water samples. Dr. Ken Marcum provided advice on turfgrass salinity issues. And finally, I want to thank Jeff Gilbert. I do not think I would have completed this thesis without the help of Jeff. He went out of his way to help me whenever I needed it, and was always there to share his experience and knowledge in the area of turfgrass and irrigation management. 4 DEDICATION I can't think of anyone who deserves this dedication more than my family. Not only has their financial support been greatly appreciated, their emotional support has brought me through the thick and thin. I truly am in great debt to them for their tremendous support and guidance. Thank You! If I forget to show my appreciation in the future, please refer to this manuscript...at least then I will know someone will have referred to it! 5 TABLE OF CONTENTS LIST OF FIGURES 7 LIST OF TABLES 11 LIST OF EQUATIONS 12 ABSTRACT 13 CHAPTER 1 INTRODUCTION 15 CHAPTER 2 LITERATURE REVIEW History of Water Use in Southern Arizona Turf Water Duties Within The Tucson AMA Evaporation and Turf Water Use Evapotranspiration Measurement Lysimetry Irrigation Water Assessments Soil Water Depletion Meteorological Assessment of ET Empirical Methods Reference ET and Crop Coefficients Turf Consumptive Use Studies Salinity and Turfgrass Culture Assessing Soil Salinity Direct Methods Time Domain Reflectometry 18 18 22 28 33 34 38 38 39 40 41 42 50 54 54 56 CHAPTER 3 MATERIALS AND METHODS Lysimeter Facility and Operation Turf and Irrigation Management Water and Salt Balance Procedure Subsurface Salinity Assessment Crop Coefficient (Kc) Assessment 64 64 70 76 78 81 CHAPTER 4 RESULTS AND DISCUSSION Water Balance Salt Balance Assessment of Salinity Profiles Turf Quality Biomass Accumulation and Turf Growth Rates 82 82 95 104 108 115 6 CHAPTER 4 RESULTS AND DISCUSSION (continued) Crop Coefficients 122 CHAPTER 5 CONCLUSIONS Water Balance Salt Balance Turfgrass Quality Biomass Accumulation and Turfgrass Growth Rates Crop coefficients Scenario Testing 136 136 137 138 138 139 140 APPENDIX A IRRIGATION WATER ANALYSIS 142 APPENDIX B MONTHLY WATER BALANCE DATA 143 APPENDIX C TIME DOMAIN REFLECTOMETRY Comparison of ECa and ECw Comparison of Neutron Probe 0v and TDR Ov 156 156 158 APPENDIX D MANAGEMENT OF LYSIMETER DATA GAPS 163 APPENDIX E LYSIMETER PERFORMANCE AND RECOMMENDATIONS Lysimeter Performance Lysimeter Recommendations 165 165 168 APPENDIX F CAMPBELL SCIENTIFIC DATALOGGER PROGRAM 170 REFERENCES 197 7 LIST OF FIGURES Figure Page 1. Percentage of water use in the Tucson Active Management Area supplied by mined groundwater (GW), renewable groundwater and effluent during 1997. Figure 19 courtesy of Arizona Department of Water Resources 2. Map showing the location and irrigation water source of the 32 golf courses within the Tucson Active Management Area. Courtesy of the Arizona Department of 21 Water Resources 3. Typical TDR output trace used to assess soil volumetric water content. Label A identifies point where pulse signal enters the soil. Label B identifies point where the 58 reflected signal has returned to the soil entry point 4. Typical TDR output trace used to assess bulk soil electrical conductivity. Arrows 60 labeled VT and V R depict the transmitted and reflected voltages, respectively. 5. Plan view of Karsten Desert Turfgrass Research Facility in Tucson, AZ. Turfed areas are shown in green. The lysimeters are located in the middle of the facility. 65 Labels W and E identify the west and east lysimeters, respectively 6. Lysimeter scale response for day 170. The vertical axis depicts the depth of water in mm relative to the midnight volume on day 170. Time of day is presented in 68 decimal format as the fraction of the current day of year 7. Monthly inputs and outputs of water for the west and east lysimeters for the period August 1997 to September 1998. Inputs (irrigation + precipitation) and outputs 84 (ETa + drainage) are presented as an equivalent depth of water in mm 8. Percent difference in monthly turf ET between the west and east lysimeters plotted as a function of the percent difference in monthly turf growth rate between the west 90 and east lysimeters 9. Electrical conductivity (EC) of the drainage water obtained from the east and west 99 lysimeters over the course of the study 10. Components of the salt balance for the east and west lysimeters during the period August 1997 to September 1998. Monthly inputs of salt from irrigation in grams are presented as positive values on the y-axis. Monthly outputs of salt from drainage in grams are presented as negative values on the y-axis. 102 8 11. Mean bulk soil electrical conductivity (ECa) obtained at depths of 0.5, 1.0 and 2.0 m 105 in the west lysimeter for the period 24 July through 17 Sep. 1998 12. Mean bulk soil electrical conductivity (ECa) obtained at depths of 1.0 and 2.0 ni in 106 the east lysimeter for the period 24 July through 17 Sep. 1998 13. NTEP quality ratings of west and east lysimeters over the course of the study. 109 Rating of 6 indicates acceptable quality turf 14. Turfgrass quality and the total amount of water supplied to west lysimeter turf by 111 month for the period August 1997-September 1998 15. Turfgrass quality and the total amount of water supplied to east lysimeter turf by 111 month for the period August 1997-September 1998 16. Monthly mean turfgrass quality ratings from both lysimeters plotted as a function of water supplied to the turf. Water supplied is presented as a percentage of ETa. 112 The line depicted represents the least squares regression line 17. Mean monthly turfgrass quality ratings plotted as a function of bulk soil electrical conductivity obtained at the 0.5 m depth in the west lysimeter. The line depicted is 114 the least squares regression line 18. Mean monthly turfgrass quality ratings plotted as a function of bulk soil electrical conductivity obtained at the 1.0 m depth in the east lysimeter. The line depicted is 114 the least squares regression line 19. Turf biomass collected from the west and east lysimeters over the course of the 116 study. Biomass is presented in grams for both lysimeters ?O. Monthly growth rates in g/m2 /day for turf grown on the east and west lysimeters 116 between August 1997 and September 1998 21. Turfgrass growth rates and the total amount of water supplied to both lysimeters by 118 month for the period August 1997-September 1998 22. Monthly mean turfgrass growth rates for both lysimeters plotted as a function of water supplied to the turf Water supplied is presented as a percentage of ETa. The 120 line depicted represents the least squares regression line 9 23. Mean monthly turfgrass growth rates plotted as a function of bulk soil electrical conductivity obtained at the 0.5 m depth in the west lysimeter. The line depicted is 121 the least squares regression line 24. Mean monthly turfgrass growth rates plotted as a function of bulk soil electrical conductivity obtained at the 1.0 m depth in the east lysimeter. The line depicted is 121 the least squares regression line 25. Comparison of the turf growth rates obtained from the previous study (1996) with 123 those obtained in the current study (1998) for the west lysimeter 26. Comparison of the turf growth rates obtained from the previous study (1996) with 123 those obtained in the current study (1998) for the east lysimeter 27. Mean monthly crop coefficients (Kcs) for overseeded intermediate ryegrass grown on the east lysimeter. The seasonal mean Kc value is depicted by the horizontal 126 arrow 28. Mean monthly crop coefficients (Kcs) for overseeded intermediate ryegrass grown on the west lysimeter. The seasonal mean Kc value is depicted by the horizontal 126 arrow 29. Mean monthly crop coefficients (Kcs) for bermudagrass grown on the east 127 lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow 30. Mean monthly crop coefficients (Kcs) for bermudagrass grown on the west 127 lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow 31. Mean daily crop coefficients (Kcs) by month for intermediate ryegrass grown on the east lysimeter. Error bars represent the standard deviation of the mean daily Kc .129 32. Mean daily crop coefficients (Kcs) by month for intermediate ryegrass grown on the west lysimeter. Error bars represent the standard deviation of the mean daily Kc 129 33. Mean daily crop coefficients (Kcs) by month for bermudagrass grown on the east 130 lysimeter. Error bars represent the standard deviation of the mean daily Kc 34. Mean daily crop coefficients (Kcs) by month for bermudagrass grown on the west 130 lysimeter. Error bars represent the standard deviation of the mean daily Kc 35. Coefficient of variation of mean daily crop coefficients computed using all days in a 131 month and only dry days (no rain) 10 36. Daily values of turf ET from the east lysimeter versus daily ETo for the period November 1997 through April 1998. The line depicted is the least squares 133 regression line with intercept forced through zero 37. Daily values of turf ET from the west lysimeter versus daily ETo for the period November 1997 through April 1998. The line depicted is the least squares 133 regression line with intercept forced through zero 38. Daily values of turf ET from the east lysimeter versus daily ETo for the period May through September 1998. The line depicted is the least squares regression line with 134 intercept forced through zero 39. Daily values of turf ET from the west lysimeter versus daily ETo for the period May through September 1998. The line depicted is the least squares regression line with 134 intercept forced through zero 40. Soil solution salinity (ECw) plotted as a function of bulk soil salinity (ECa) for the period August 1997 to September 1998 for the west lysimeter at the 1.0 m depth. 157 The line depicted represents the least squares regression line 41. Soil solution salinity (ECw) plotted as a function of bulk soil salinity (ECa) for the period August 1997 to September 1998 for the east lysimeter at the 1.0 m depth. 157 The line depicted represents the least squares regression line 42. Soil solution salinity (ECw) plotted as a function of bulk soil salinity (ECa) for the period August 1997 to September 1998 for both lysimeters at all depths. The line 159 depicted represents the least squares regression line 43. Volumetric water content (theta) derived from time domain reflectometry (TDR) plotted as a function of theta derived from neutron attenuation for the period August 1997 to September 1998 for both lysimeters at all depths. The line depicted represents the least squares regression line 161 11 LIST OF TABLES Table Page 1. Salinity tolerance of commonly used cool season and warm season turfgrasses in 55 southern Arizona 2. Fertilizer application dates, rates and the type of fertilizer applied to both east and 72 west lysimeters over the course of the study 3. Seasonal crop coefficients (Kcs) applied during summer (15 June-13 Oct.) and winter (5 Nov.-4 June) seasons and Kcs applied during fall overseed (14 Oct.-4 75 Nov.) and spring transition (5 June-14 June) 4. Components of the water balance by month in equivalent depth of water for the east and west lysimeters for the period August 1997-September 1998. Totals represent 83 the sum of individual monthly values for the 12 months ending 30 Sept. 1998 5. Maximum (Tmax) and minimum temperatures (Tmin), precipitation (ppt.) and reference evapotranspiration (ETo) measured over the course of the study by the AZMET weather station located at the Karsten Desert Turfgrass Research Facility. Temperatures are reported as monthly means and precipitation and ETo are reported as totals. Departure from normal (+ / - Normal) compares monthly values to mean values for the period 1987-1997 86 6. Total salt inputs (through irrigation) in grams of salt applied, total salt outputs removed through drainage over the course of the study, calculated from October 1997 to September 1998. Values indicate totals for the study period. Total water applied indicates the summation of rainfall and irrigation. The leaching fraction indicates the total drainage divided by the total water applied 96 7. Month-by-month assessment of the salt balance for the east lysimeter over the course of the study. Monthly totals indicate either salt input through irrigation, salt output removed through drainage or the net balance between inputs and outputs. 101 8. Month-by-month assessment of the salt balance for the west lysimeter over the course of the study. Monthly totals indicate either salt input through irrigation, salt output removed through drainage or the net balance between inputs and outputs .101 9. Monthly crop coefficients computed using ETa data obtained from the east and west lysimeters, and reference ET data provided by the Arizona Meteorological Network. 124 12 Equation LIST OF EQUATIONS Page 1. Water Balance 33 2. Modified water balance (solved for ET) 33 3. Energy balance 39 4. Modified energy balance (solved for latent heat) 40 5. Bowen ratio 40 6. Crop coefficient 41 7. Leaching fraction 52 8. Leaching requirement 53 9. Dielectric constant 57 10. Volumetric water content 57 11. Expression of bulk soil electrical conductivity related to the bulk liquid phase 59 conductivity and the bulk surface conductivity 12. Bulk soil electrical conductivity related to the saturation percentage 59 13. Derivation of bulk soil electrical conductivity from TDR output 59 14. Relationship between ECa and ECw for unsaturated soils 61 15. Empirical relationship for the transmission coefficient 62 16. Time domain reflectometry pulse travel time 80 17. Statistical model used to assess Kc values 81 18. East lysimeter rainfall adjustment 163 19. West lysimeter rainfall adjustment 163 13 ABSTRACT Water is an essential resource that requires careful management at all golf courses located in southern Arizona. The Arizona Department of Water Resources, through its enforcement of irrigation water duties, is forcing the golf industry to reduce water usage. The objective of this study was to evaluate turfgrass performance and the potential for soil salinization, when high maintenance desert turf systems are irrigated in accordance with the present Tucson area water duty of 1.4 ha-m/ha/yr (4.6 ac-ft/ac/yr). Two large weighing lysimeters supporting year-round turf systems consisting of bermudagrass (Cynodon dactylon x transvalensis (L.) pers.) overseeded with intermediate ryegrass (Lolium multiflorum x perenne) were irrigated at rates not to exceed the ADWR water duty using either low salinity (EC = 0.25 dS/m) groundwater or higher salinity effluent water (EC = 1.0 dS/m). Irrigation treatments were initiated in August 1997 and continued through September 1998 and consisted of applying water daily at rates set by applying appropriate crop coefficients to values of reference evapotranspiration generated by an on-site weather station. Soil moisture and salinity regimes were monitored weekly using the lysimeter subsurface sampling system and time domain reflectometry (TDR). Water percolating below the root zone was quantified and sub-sampled to facilitate assessment of leaching fractions and total lysimeter salt balance. For the year ending 30 Sept. 1998, each lysimeter received —1729 mm (68 in.) of water comprised of 1296 mm (51 in.) of irrigation water and 433 mm (17 in.) of precipitation. Turfgrass evapotranspiration (ET) totaled 1419 mm (56 in.) for the lysimeter irrigated with 14 groundwater (east lysimeter) and 1466 mm (58 in.) for the lysimeter irrigated with effluent (west lysimeter). Approximately 421 and 311 mm of drainage water was removed from the east and west lysimeters, respectively, establishing leaching fractions of 0.24 in the east lysimeter and 0.18 in the west lysimeter. Salts accumulated in both lysimeters over the course of the study; however, the substantial amount of drainage did not allow for salts to accumulate to harmful levels. Turfgrass performance, as quantified by turf quality and growth was acceptable or better during most months of the study. Crop coefficients (Kcs) were slightly higher than previous years, however, Kcs compared favorably to previous research at the study site. Summer Kcs averaged 0.79 and were significantly higher than winter Kcs that averaged 0.73. Turf irrigated with effluent produced better quality turf and used slightly more water than turf irrigated with groundwater. These data indicate that the present ADWR water duty of 1.4 ha-m/ha/yr (4.6 acre-ft/acre/yr) is adequate to replace turfgrass evapotranspiration and provide for leaching of salts when rainfall exceeds normal amounts, assuming no water loss due to irrigation and plumbing inefficiencies. 15 CHAPTER 1 INTRODUCTION Water is an essential production input at all golf courses located in southern Arizona. Golf courses in the Tucson area require large quantities of irrigation water due in large part to the arid climate which allows for year round production of turf yet provides limited precipitation to offset high rates of evaporative demand. Both economic and regulatory pressures are driving the golf industry to conserve water. Economic pressures result from the increasing cost of water, while the regulatory pressures come from the Arizona Department of Water Resources (ADWR) through its implementation and enforcement of the 1980 Groundwater Management (GWMA). The primary objective of the GWMA is to reduce groundwater overdraft in the five Active Management Areas (AMAs) -- geographical areas designated for intense management due to the large imbalance between the groundwater consumed and the dependable supply. Total water use in the Tucson AMA presently exceeds annual groundwater recharge by approximately 50 percent. This chronic overuse of groundwater has reduced the static levels of local aquifers by approximately 50 m and increases the threat of land subsidence. Groundwater overdraft in the Tucson and other AMAs has led ADWR to develop water conservation plans for all industries and municipalities that use substantial amounts of groundwater. An important aspect of the conservation plans established for Tucson area turf facilities is the water duty which represents the maximum amount of 16 groundwater a turf facility may use for irrigation. The present water duty for Tucson turf facilities provides 1.4 ha-m/ha (4.6 acre-ft/acre) of water per year to an 18 hole golf course with 36.5 ha of turf There is growing pressure from the public and ADWR to reduce this duty further in future years. The golf industry and developers claim that the current duty is inadequate and a further reduction in the water duty will reduce turf quality and/or turf acreage, and possibly lead to soil salinization. Recent research conducted at the Karsten Desert Turf Facility in Tucson lends credence to industry concerns (Brown and Kopec, 1997; Brown et al. 1996). These studies indicate the water duty is adequate to support high quality turf provided the entire duty can be applied to the turf (no inefficiencies in application) and rainfall runs near normal. However, these studies were unable to address industry concerns regarding: (1) application inefficiencies inherent in all commercial irrigation systems, (2) the adequacy of the duty to provide sufficient water for salinity management and (3) the adequacy of the duty in dry years. The study described in this thesis was designed to address the industry concerns regarding the adequacy of the present turf water duty to (1) generate acceptable turf performance in both wet and dry years and (2) provide adequate water to facilitate the necessary leaching of salts from the turf root zone. Large weighing lysimeters supporting year round green turf were irrigated with local groundwater and effluent at rates not to exceed ADWR's water duty of 1.4 ha-m/ha/yr. Turf performance, turf water requirements and the soil water and salinity balances were monitored to address the 17 aforementioned industry concerns regarding the adequacy of current turf water duties. The results presented in this thesis summarize the findings from the first year of a planned three year study. 18 CHAPTER 2 LITERATURE REVIEW History of Water Use in Southern Arizona Groundwater management in Arizona has been a recurring problem since the 1930s. The major problem lies in the imbalance between the amount of water consumed, and the amount that is naturally or artificially recharged. About 60% of the total water consumed in Arizona comes from underground aquifers, and, each year, Arizonans consume approximately 305,750 ha-m (2.5 million acre-ft) more groundwater than is recharged. Within the Tucson AMA, total water demand has climbed to 38,402 ha-m (314,000 acre-ft) per year. More than 50% of that water demand is served by mined groundwater (Figure 1; ADWR, 1997). Groundwater overdraft has lowered the static level of underground aquifers by as much as 52 m (170 ft) since 1940, causing land subsidence and water quality problems (ADWR, 1984). In order to deter groundwater overdraft, the Arizona legislature enacted the Groundwater Management Act (GWMA) in 1980. The GWMA established five Active Management Areas (AMA's) including Tucson, Pinal, Phoenix, Santa Cruz and Prescott. These are geographical areas where serious groundwater overdraft is occurring. The GWMA created the state agency Arizona Department of Water Resources (hereafter referred to as "the Department" or ADWR) to develop and implement all state water legislation and secure long term water supplies for Arizona's AMA's (Eden and Wallace, 1992). In the Tucson, Phoenix, Prescott and Santa Cruz AMA's, ADWR's primary 19 Renewable GW (42.60%) 134,000 AF Mined GW P.79%) 169,000 AF 10,138 AF of CAP water delivered to agriculture through in-lieu recharge is legally accounted for as mined groundwater. Figure 1. Percentage of water use in the Tucson Active Management Area supplied by mined groundwater (GW), renewable groundwater and effluent during 1997. Figure courtesy of Arizona Department of Water Resources. 20 management objective is to achieve "safe-yield" by 2025 through a series of water management plans. The management goal for the Pinal AMA is to preserve the agricultural industry for as long as feasible while preserving groundwater for future nonirrigation uses. Safe-yield is defined as "a groundwater management goal which attempts to achieve and thereafter maintain a long term balance between the annual amount of groundwater withdrawn in an Active Management Area and the annual amount of natural and artificial recharge in the AMA." (ADWR, 1984). The Department is presently employing the second management plan which is scheduled to end in 1999; the third management plan will be implemented in the year 2000. Within these management plans are conservation programs for agriculture, municipalities, and industrial use. Turf-related facilities are classified as industrial water users; thus turf facilities must comply with plans set forth under the Industrial Conservation Program. The 32 golf courses located within the Tucson AMA (TAMA) used 1,976 ha-m (16,153 acre-ft) of water (groundwater and effluent) to irrigate 1.309 ha (3,233 acres) of turf in 1995. Thirty five percent of the total water demand was satisfied with effluent water which was supplied to thirteen golf facilities. Figure 2 depicts the locations of golf facilities within the TAMA (ADWR, 1997). The Department reported the average application rate for all golf courses in 1995 was 1.42 ha-m/ha (4.7 acre-ft/acre) per year; golf courses using only groundwater for irrigation used on average 1.52 ha-m/ha (5 acreft/acre; David Johnson, ADWR, 1998, personal communication). Water use by the golf 21 Figure 2. Map showing the location and irrigation water source of the 32 golf courses within the Tucson Active Management Area. Courtesy of the Arizona Department of Water Resources_ 22 industry is expected to grow in subsequent years. The Department projects an additional 23 golf courses will be built within the TAMA by 2025, increasing water use by turfrelated facilities to approximately 3,914 ha-m (32,000 acre-ft), compared to the 1,976 ham used in 1995. Up to 85% of total water demand by turf facilities is projected to be met by alternative water sources in 2025, the primary one being effluent. Turf Water Duties Within The Tucson AMA The first management (FMP) plan instituted by ADWR covered a ten year period ending in 1990. The Industrial Conservation Program of the FMP included a catch all conservation requirement stating, For industrial uses including industrial uses within the exterior boundaries of the service area of a city, town, private water company or irrigation district, the program shall require use of the latest commercially available conservation technology consistent with reasonable economic return. (ADWR, 1984) Conservation requirements were to be met by 31 Dec. 1986. One such requirement involved a cap on the amount of water per unit area allocated to golf courses within the TAMA. This water allotment is referred to as a water duty. The initial water duty instituted for the FMP was 1.52 ha-m/ha (5.0 acre-ft/acre) per year for existing turfrelated facilities and 1.46 ha-m/ha (4.8 acre-ft/acre) per year for new turf-related facilities (ADWR, 1984). An existing turf-related facility was one in existence or in operation before the effective date of the FMP. Additional water (1.76 ha-m/ha; 5.8 acre-ft/acre) was allocated for filling/refilling surface water bodies. Also, turf-related facilities had the 23 option to apply for a modification of the total water allotment if technical difficulties caused by using effluent water were evident. The Department used five factors to derive turf facility water duties for the FMP (ADWR, 1984): 1. Number of acres in water-demanding landscaping; 2. Number of acres of water surface; 3. Type of vegetation grown; 4. Germination requirements of new turf; 5. Efficiency of water application. The Department points out that the most important factor with respect to water use is the number of acres with water demanding landscape. Although ADWR does not regulate the total acres a turf facility manages, they do regulate the total water allocation. A new turf-related facility under the FMP would receive 3.06 ha-m (24.8 acre-ft) of water per golf hole. This total water allotment was derived from the product of 1.46 ha-m/ha (4.8 acre-ft) x 2 turf-ha (5 turf-acres) per golf hole plus 0.10 ha-m (0.8 acre-ft) allocated for surface water. Most golf courses have an eighteen hole layout approximating ninety acres of irrigated turf. Thus, ADWR regulations allowed for approximately 55 ha-m (446 acre-ft) of water for golf courses regulated as new turf-facilities under the FMP (David Johnson, ADWR, 1998, personal communication). Golf course designers and managers realize the implications the number of irrigated acres have on the amount of water applied per unit area of turf. Hence, the trend for golf courses in the southwest has been to subscribe to the "target golf' design philosophy which involves reduced turf area 24 compared to golf courses using traditional designs. A reduction in turfed acreage allows for greater amounts of water applied per unit area. The Department listed the type of vegetation and the water requirements for new vegetation as factors impacting the determination of the water duty for the FMP. Most turf-related facilities within the Tucson AMA manage bermudagrass as the dominant grass species. Since bermudagrass has a winter dormancy period, many facilities overseed with ryegrass to provide for year-round green turf. The Department did not openly recognize the water requirements of ryegrass in developing the water duties, however, bermudagrass consumptive use was recognized and listed as 1.1 ha-m/ha (3.5 acre-ft/acre). Newly turfed areas were to receive additional water for germination and establishment (overseed areas do not qualify as new turf). Existing facilities received 1.82 ha-ni/ha (6.0 acre-ft/acre) and new facilities received 1.76 ha-ni/ha (5.8 acre-ft/acre) for newly turfed areas in addition to their annual allotment. The efficiency of the water applied was also listed as a factor affecting the derivation of the water duty. The Department realizes that irrigation systems vary in efficiency, however, additional water is not allocated for turf managers to account for irrigation non-uniformities and plumbing losses. Absent from the above list is (1) the excess water required for establishment of winter overseeded turf and general turf maintenance, and (2) the excess water needed to obtain an adequate leaching fraction. ADWR has a provision to provide additional water for leaching. The provision is based on standard procedures for computing a leaching 25 fraction, however, the provision requires that the irrigation water contains a minimum dissolved salt content of 1000 mg/liter. The leaching fraction requirement will be discussed in more detail under salinity concerns. The second management plan (SMP) was instituted in 1990 and will continue through the year 1999 (ADWR, 1991). Development of the SMP involved a reevaluation of the water duties set by the FMP. The section of the SMP related to turf water duties was developed based on a review of scientific literature, surveys, and on site visits. Sources for the literature review primarily came from southern California and southern Arizona including Erie et al., 1981; Kneebone and Pepper, 1982, 1984; and Marsh et al., 1980a. In reviewing the aforementioned sources, the following information was considered. • Actual water figures collected from over 400 turf-related facilities regulated under the first management plan • The consumptive use of the grass species most frequently used • Water application efficiency achievable with currently available technology • Evaporative losses from lakes based on pond-evaporation data • Management practices and technologies currently in place • Conservation potential associated with additional technologies, practices and design alternatives • Germination requirements for establishing new turf Again, absent from this information is (1) additional water required for winter overseed establishment and general turf maintenance, and (2) excess water needed to obtain an 26 adequate leaching fraction. Leaching allotments remained unchanged and were available only if the TDS of irrigation water exceeded 1000 mg/l. In most cases water duties were reduced during the SMP. Existing and new Tucson area facilities that qualified for the first management plan were required to meet a maximum water allotment of 1.4 ha-m/ha (4.6 acre-ft/acre) per year by the year 2000, whereas new turf facilities that qualified for the second management plan had to meet the 1.4 ha-ni/ha per year water duty by the year 1992 (ADWR, 1991). An added contingency to the SMP involved segregating low water use landscaped areas from high water use turf areas. The SMP allowed 0.18 ha-m (1.5 acre-ft) of water for low water use landscaped areas. Also, as an incentive to increase effluent use, the Department provides allotment modifications (incentives) based on the percent water use met by effluent. The effluent incentive states that each 0.123 ha-m (1.0 acre-ft) of effluent used by the turf facility for landscape watering purposes counts as 0.117 ha-m (0.95 acre-ft) of water if effluent accounts for 50-89% of the total water used, or the total allotment (whichever is less) for the year. The Department counts each 0.123 ha-m of effluent used by the turf facility as 0.111 ha-m (0.90 acre-ft) of water if the total effluent used by the turf facility for the year is greater than 90% of their total water use or their total water allotment, whichever is less. Although the effluent incentive does not change the total water allotment, a using >90% effluent water to meet their water demands could apply up to acre-ft) per year as opposed to 1.4 ha-m (4.6 acre-ft) per year. facility 1.55 ha-m (5.1 27 The third management plan (currently in draft form) will be implemented in the year 2000 and continue through 2010 (ADWR, 1998). Water duties projected for the third management plan (TMP) remain at 1.4 ha-m/ha (4.6 acre-ft/acre) for turfed acres, 1.76 ha-ni/ha (5.8 acre-ft/acre) for water surface acres and 0.46 ha-ni/ha (1.5 acre-ft/acre) for low water use landscaped areas. The maximum annual water allotment is the same as the SMP at 2.93 ha-m (23.8 acre-ft) of water per golf hole. Similar to the previous plans, the draft of the TMP includes "allotment additions" which allow for turf facilities to receive additional water for specific cases. These specific cases include allotments for a reduction in turfed acreage, establishment of newly turfed area (does not qualify as winter overseeding), revegetation of low water use plants, filling bodies of water, leaching of salts and allotment modifications based on effluent use. Due to the underutilization of effluent, the effluent incentive was modified in the third management plan. The maximum water allotment did not change, however, each 0.123 ha-m of effluent water used will count as 0.09 ha-m (0.7 acre-ft) when turf facilities are in compliance with the maximum annual water allotment (ADWR, 1998). It is debatable whether some of these special case water allotments should be included in the maximum annual water allotment. The Department stresses the need for increased use of effluent and Central Arizona Project (CAP) water in the future if conservation requirements are to be met. Long-term use of effluent and CAP water will likely lead to build-up of salts and will require additional water to flush those salts below the root zone. The Department recognizes that further research is required to quantify the 28 long-term impacts of using water high in salts and to determine its affects on future water application rates. Evaporation and Turf Water Use It is evident that water conservation is imperative to secure long term water supplies to ensure future growth and continued commercial success of southern Arizona's golf industry. It is therefore important to understand the water requirements and salinity tolerance of high maintenance turf systems in order to implement water duties that encourage water conservation yet provide acceptable turf quality. Evapotranspiration (ET) is defined as the loss of water from vegetation through the combined processes of soil evaporation and plant transpiration. Effective irrigation management therefore requires an understanding of ET in order to determine the amount and timing of irrigation. Consumptive use includes ET plus the water retained in the plant tissue (Jensen et al., 1990). Under most circumstances the water in the plant tissue is nominal relative to daily ET and does not change appreciably from one day to the next, thus ET and consumptive use are essentially equal and the terms are often used interchangeably. Both environmental and biological factors combine to influence ET (Beard, 1985). Environmental factors include soil water content and meteorological conditions which provide the energy and transport mechanisms necessary to evaporate and transfer water vapor from the plant-soil system. Biological factors include the plant type and the stage of development. 29 Soil moisture is the most important environmental factor impacting ET since without soil moisture there is no water available for evapotranspiration (Kramer and Boyer, 1995). When soil moisture is non-limiting, ET is a function of meteorological factors or evaporative demand and biological factors such as size and type of plant, however, when soil moisture levels decline below field capacity ET becomes much more dependent on the soil moisture availability. Studies by Kneebone and Pepper (1982, 1984) have shown that turf ET is directly related to available soil moisture. Four meteorological variables - - net radiation, air temperature, windspeed and relative humidity impact evaporation and thus contribute to evaporative demand. Net radiation is derived from a combination of solar and thermal components and represents a key source of energy necessary for evaporation. The solar component of net radiation is made up of incoming radiation from the sun termed solar radiation minus that portion of solar radiation that is reflected from the surface. Net thermal radiation is described as a balance between the incoming longwave radiation emitted by the earth's atmosphere outgoing longwave radiation which is emitted by the earth's surface (Kramer and and Boyer, 1995). Net radiation is simply the sum of the solar and thermal components. On most days solar radiation is the dominant component of net radiation; thus, solar radiation is the primary meteorological factor affecting ET when soil moisture is increased available (Kneebone et al., 1992). Feldhake et al., (1983) found that turf ET in linearly with solar radiation. Solar radiation is most intense during the summer months is at its peak during this period. the northern hemisphere, especially in the desert, thus ET 30 Air temperature and horizontal wind movement provide the means by which sensible heat is transferred between the leaf and the environment. Sensible heat represents a source of energy for evaporation and moves to or from vegetation by the processes of diffusion, convection and advection (Kramer and Boyer, 1995). Diffusion occurs in unstirred air due to differences in air temperature between the plant and surrounding air. Convection refers to the sensible heat transfer by the bulk movement of air due to temperature-induced density differences in the air. Advection occurs due to bulk movement of air caused by horizontal wind movement (Rosenberg, 1974). Wind can also decrease the leaf boundary layer which increases the rate at which water is lost from the plant and transported to the surrounding air (Kramer and Boyer, 1995). Humidity impacts ET by establishing the vapor gradient from the leaf surface to the atmosphere. This gradient is referred to as the vapor pressure deficit (VPD), and is a function of canopy and air temperature, and atmospheric vapor pressure. High temperatures and low atmospheric humidities result in high VPDs (Brown, 1999a), which in turn lead to accelerated rates of ET. Lower temperatures and higher humidities reduce VPD and lessen ET. The biological factor that has the greatest potential influence on ET is the plant type or plant species. Grass species can use upwards of 12.5 mm per day when grown under high evaporative conditions and provided with ample soil moisture (Kneebone and Pepper, 1984), whereas native desert plant species can live on rainfall alone (approximately 30 cm per year). Differing grass species vary in consumptive use values. 31 In Tucson, AZ, annual consumptive use values for Tifgreen bermudagrass, St. Augustinegrass and tall fescue were determined to be 1635 mm, 1657 mm and 1817 mm, respectively, under non-limiting soil-moisture conditions (Kneebone and Pepper, 1982). Cool season grasses typically have higher rates of ET of than warm season grasses when grown under similar conditions (Biran et al., 1981; Kneebone and Pepper, 1982; and Feldhake et al., 1983). Feldhake et al. (1983) determined that Kentucky bluegrass and tall fescue which are cool season grasses utilized over 20% more water than bermudagrass and buffalograss which are warm season grasses. Biran et al. (1981) studied the effects of mowing height, irrigation frequency and soil moisture on ET and found that the single most important factor affecting ET was whether the grass was cool season or warm season. They observed the ET of the cool season grasses were 45% higher than ET for warm season grasses. Another biological factor affecting ET is the relative growing activity of the plant. A dormant plant with little or no green foliage will use considerably less water than an actively growing plant. Bermudagrass and buffalograss represent typical southwest turfgrasses that display a distinct seasonal (winter) dormancy that impacts ET. Under a deep and infrequent irrigation regime (irrigation applied only when turf showed symptoms of wilt), mean annual consumptive use for bermudagrass over a three year period was 879 mm (Garrot and Mancino, 1994). However, when bermudagrass is overseeded with a cool season grass to sustain green turf during winter months, annual consumptive use values in excess of 1500 mm have been reported (Brown et al., 1998; 32 Kneebone and Pepper, 1982; Kopec et al. 1991). Grass that is growing vigorously will have higher ET rates than a slow growing grass. K_rogman (1967) observed a direct relationship between ET and turf growth rates, which resulted from increased rates of nitrogen (N) fertility. Likewise, Feldhake et al. (1983, 1984) reported that Kentucky Bluegrass produced higher growth rates and used 13% more water when subjected to an adequate N fertility regime of 4 kg of N/1000 m 2 /month compared to a deficit level of 4 kg of N/1000 m2 /year. Other biological factors affecting ET include turf height and density. Increased mowing heights will increase ET (Shearman and Beard, 1973, Biran et al., 1981, and Johns et al., 1983). Biran et al. (1981) found that increasing the mowing height from 3 to 6 cm resulted in a significant increase in water consumption in cool season grasses and a slight increase in warm season grasses. Water consumption for cool season grasses increased by 29 and 25% for tall fescue and perennial ryegrass, respectively. Shearman and Beard (1973) determined that an increase in mowing height from 0.75 to 2.5 cm increased the ET by 50% in Pencross' creeping bentgrass. A reduction in turf density may actually increase ET rates. As turf density declines surface roughness increases, thus decreasing the leaf boundary layer resulting in a subsequent increase in turf ET (Kramer and Boyer, 1995). In Kentucky bluegrass, shoot density was determined to be negatively correlated with ET, while the vertical leaf extension rate was positively correlated with ET (Shearman, 1986). Ebdon and Petrovic (1998) determined that turf with a lower ET rate was associated with more horizontal leaf 33 orientation, narrower leaf texture, more lateral shoots per plant, shorter leaf blades and sheath, and more leaves per shoot than the higher water use turf. Evapotranspiration Measurement Various methods exist for measuring ET (Kneebone et al., 1992). Common methods include (1) utilizing lysimeters to measure the amount of water lost by evaporation and transpiration, (2) measuring the amount of water applied to field crops, (3) measuring the change in soil water content over a period of time, and (4) meteorological techniques. The first three methods listed utilize the water balance method to measure ET. The water balance equation is given below: P + I — ET — R — D = ASWS [Eq.1] where P is precipitation, I is irrigation, R is runoff, D is drainage out of the soil profile and ASWS is the change in soil water storage. Rearrangement of Equation  provides the following: ET = P + I —D — R — ASWS [Eq.2] which allows computation of ET provided one can quantify P, I, D, R and ASWS. the most Quantitation of all the variables on the right side of Equation  will provide accurate appraisal of ET, however, if terms R, D and ASWS are assumed negligible, an al., estimation of ET can be obtained simply by quantifying terms I and P (Marsh et 1980a). 34 Lysimetry A lysimeter is a receptacle that contains soil as the medium by which a particular plant species can be irrigated and fertilized in a contained system in order to assess water use requirements (Tanner, 1967). Lysimeters have the potential to measure every parameter in the water balance equation, thus they serve as an excellent tool for measuring ET. Lysimeters have been used extensively to provide information on water quality (Kirkham et al., 1984), root growth (Dugas et al., 1985), and validation and quantitation of evapotranspiration (Tovey et al., 1969; Kneebone and Pepper, 1982, 1984). There are two general types of lysimeters in use today: percolation and weighing lysimeters (Kneebone et al. 1992). Percolation lysimeters rely on the change in water storage, or the inflow minus the outflow to determine ET. The inflow-outflow method relies on measurement and quantitation of the inflow: precipitation (P) and irrigation (I), and the outflow: drainage (D) to estimate ET. Percolation lysimeters measure ET most effectively over time periods of a week or longer due to the lag time between irrigation and the resulting leachate obtained from percolation. Tovey et al. (1969) used percolation lysimeters and employed the inflow-outflow method to measure actual ET of lawngrass and bermudagrass. Percolation lysimeters were used by Kneebone and Pepper (1984) to measure luxury water use on and the effect of very high water application rates on consumptive bermudagrass use. Gee et al. (1994) and deep percolation. A used percolation lysimeters to monitor shallow water movement 35 variant of the standard percolation lysimeter is one that uses Marriotte siphons to deliver subirrigation. Kneebone and Pepper (1982) used this technique to evaluate consumptive use of turfgrasses. The subirrigation technique relies on measurement of the loss of water from the Marriotte siphon container rather than the difference between water applied and water lost to drainage. In contrast to percolation lysimeters, weighing lysimeters are capable of quantifying accurate ET measurements over much shorter time intervals. Weighing lysimeters measure changes in mass that can be attributed directly to plant water use. Weighing lysimeters were introduced in order to perform precise measurements of precipitation, irrigation, evaporation and drainage events (Tanner, 1967) The simplest weighing lysimeters (micro-lysimeters) consist of pots weighed at regular intervals under different water regimes. Micro-lysimeters (-20 cm in depth) can be used to measure potential or maximum ET rates. These micro-lysimeters are commonly constructed of PVC pipe and are filled with calcined clay or a fine silica sand to facilitate drainage, yet provide for adequate soil moisture-holding capacity (Kopec et al., 1988). Rogowski and Jacoby (1977) measured and observed distribution patterns of ET using micro-lysimeters in order to develop a model for predicting ET in the field. Micro-lysimeters were used by Kopec et al. (1988) to measure ET of tall fescue turf ET under field conditions. Johns et al., (1983) used mini-lysimeters to measure potential of St. Augustinegrass utilizing the water balance method. 36 A slightly more complicated weighing lysimeter was utilized by Feldhake et al. (1983, 1984) to measure turf ET where lysimeters were constructed of large PVC pipe and inserted into PVC cylinders set into the turf. Even more complicated (and more expensive) weighing lysimeters are constructed of an inner and an outer lysimeter box fabricated from steel. The inner lysimeter box rests on a scale to provide for frequent monitoring of lysimeter mass changes. Sensitivity of these lysimeter systems can typically resolve better than 0.05 mm (0.002 in.) of water. Dugas et al. (1985) used sensitive weighing lysimeters to study crop root densities and compare lysimeter ET to calculated Penman ET. Weighing lysimeters were installed at the Department of Energy's Hanford site in Washington state to determine the water balance at hazardous waste sites, and to measure ET and calculate crop coefficients (Kirkham et al., 1984). Lysimeters, while considered an excellent tool for measuring ET, are prone to error, if not managed or operated correctly. Some sources of error include evaporative and vegetative areas on the lysimeter, height of the lysimeter rim above the surrounding area, thickness of the rim, height and density differences between the lysimeter vegetation and surrounding vegetation, inadequate fetch surrounding the tanks, and shallow lysimeter soil profiles (Allen et al., 1991) An accurate appraisal of the evaporative area on the lysimeter is essential to determine ET. Evapotranspiration measurements will be erroneously high if lysimeter vegetation is allowed to overhang beyond the lysimeter rim (Allen at al., 1991). For example, the mini-lysimeters used by Kopec et al. (1988) had a radius of 10 cm. If 37 lysimeter vegetation was allowed to grow beyond the lysimeter rim extending the effective radius (that which is covered by vegetation) by only 1 cm, a 17% error in ET measurement would result. Overhang errors typically decrease when lysimeter surface area increases; thus, most weighing lysimeters are designed with a large surface area. A lysimeter rim that is too tall or too wide can block wind, absorb heat or reflect radiation to the lysimeter vegetative surface and thus impact ET (Allen et al., 1991). Increased vegetation height on the lysimeter surface can increase ET compared to surrounding vegetation by increasing radiative transfer. A decrease in the density of lysimeter vegetation relative to surrounding vegetation may increase ET due to a decrease in the leaf boundary layer (increased wind transfer). Inadequate fetch results in increased advective transfer (edge-effect). Shallow lysimeters can affect rooting depth, soil-water availability and thermal conditions of the soil (Marek et al., 1988; Allen et al., 1991). Deep weighing lysimeters have been developed to avoid these sources of error. Dugas et al. (1985) developed lysimeters 2.7 m in depth to provide adequate room for root development. Large weighing lysimeters , 2.5 m in diameter and 4.0 m deep were installed at the University of Arizona Karsten Desert Turfgrass Research Facility to assess water use and deep percolation in a deep soil profile. (Young et al., 1996). Lysimeters have been used extensively in the past and will continue to play an integral role in determining ET of crops in the future. Environmental conditions must be taken into account when managing and interpreting lysimeter data. The problems which 38 beset lysimeters can be deterred by proper design, installation, management and operation. Irrigation Water Assessments Evapotranspiration can be estimated through quantitation of applied irrigation water. In this case, terms P and I of Equation  are quantified, while terms R, D, and ASWS are assumed to be negligible. Evapotranspiration is assumed to equal the quantity of water applied to the turf. Accurate irrigation is required to minimize R and D and the technique is best suited for monthly and seasonal time-frames to minimize the impact of changes in ASWS on ET. In irrigated turf systems, a significant portion of the applied water may drain from the upper soil profile, which is neglected in this method, thus ET evaluation is not as quantitative as those methods that measure all components of the water balance equation. Marsh et al. (1980) used this procedure to assess consumptive use of two warm season grasses and two cool season grasses in California; tensiometers were used to control irrigation and minimize D. Soil Water Depletion The most widely used method of measuring ET is to monitor the change in soil water content over time (Meyer and Gibeault, 1986, 1987; Frank et al. 1987). Evapotranspiration is estimated from Eq.  by making soil water depletion measurements to obtain ASWS and assuming R and D, I and P are known or negligible. Soil water depletion techniques are commonly used to estimate ET of row crops, 39 especially in arid regions where rain is uncommon and irrigations are infrequent. Typically, soil water measurements are obtained two to three days after irrigation, then again one day prior to the following irrigation. The lag time between irrigations is commonly one to three weeks, which allows ample time for additional drainage to occur, however, this method often neglects drainage during this period. Several researchers have used this method to estimate turf water use. Erie et al. (1981) used this method to determine consumptive use of bermudagrass while Garrot and Mancino (1994) used this procedure to determine crop coefficients of bermudagrass. Frank et al. (1987) used this technique to determined the ET of crested wheatgrass, intermediate wheatgrass, and western wheatgrass in order to assess the water use efficiency of the varieties. Meteorological Assessment of ET The most commonly used meteorological method for measuring turfgrass ET involves use of the energy balance method: + H +1E + G + aA --= 0 [Eq.3] where 1Z, is the net radiation, H is the sensible heat exchange with the atmosphere, lE is the latent heat exchange with the atmosphere, G is the heat exchange with both vegetation and soil, and aA is the energy utilized in photosynthesis and respiration processes within plants (Beard, 1985). Transpiration occurs due to energy supplied by the atmosphere that creates a free energy gradient in the water from the evaporating surface of the leaf through the plant and down to the root hairs (Rosenberg, 1974). The energy supplied by the atmosphere satisfies the latent heat parameter of the energy balance equation. The 40 energy utilized in photosynthesis and respiration (aA) is usually very small compared to the other components, thus aA is commonly ignored. Using the convention that lE is positive when there is a loss of lE from the surface, Equation  can be rearranged to give: lE = + H - G [Eq. 4] Measurements of Rn, H and G can be used in conjunction with Eq.  to obtain lE which is transformed to ET by multiplying lE by the latent heat of vaporization. A more common means of using Eq.  to estimate ET involves the introduction of the Bowen Ratio, [2. (Bowen, 1926). The Bowen Ratio is a measure of the ratio of sensible to latent heat flux (H/lE). Substitution of p into Eq.  provides: lE = (Rn - G) / (1 - p) Eq.  The p is obtained by measuring the thermal and vapor gradients above the crop canopy. Measurements of (3, Rn and G provide an effective meteorological procedure for estimating ET. Most meteorological methods of ET assessment require that measurements be made in large fields to ensure measured meteorological parameters reflect the underlying surface. Thus, energy balance methods are best suited for measuring ET of homogeneous vegetative surfaces grown under non-limiting soil moisture conditions (Tanner, 1967) Empirical Methods Commonly, ET derived from experimental studies is compared or referenced to meteorological strategies for estimating and predicting ET. Empirical modeling 41 procedures are a function of meteorological demand which differs from region to region, thus local calibration is essential to obtain an accurate estimate of ET (Brown, 1999b). Empirical methods are grouped into combination methods (i.e. Penman-Monteith, Businger-van Bavel), radiation (Jensen-Haise), temperature (Blaney-Criddle), and evaporation pan (U.S. Weather Bureau Class A and Bureau of Plant Industry; Kneebone and Pepper, 1982). The measurement frequency to estimate ET ranges from seasonal to daily measurements depending on the empirical modeling approach being used. The most common empirical model for estimating ET today is the Penman equation (Penman, 1948). The Penman equation utilizes meteorological data to generate estimates of reference crop ET (ETo). The four weather parameters used by the Penman equation are (1) solar radiation, (2) wind, (3) temperature, and (4) humidity. Numerous variations of the Penman equation exist to estimate ETo (Brown, 1999). Reference ET and Crop Coefficients Reference crop ET (ETo) is commonly determined from a continuous stand of cool season grass, 8-15 cm in height, grown under non-limiting soil moisture conditions (Brown, 1999). The ETo must then be modified to provide ET estimates for each plant species that differs from the reference crop. Plant species will differ in stage of development, height, density and overall physiology, which will affect ET, as previously discussed. A crop coefficient (Kc) is used to convert ETo to a "crop-specific" ET (ETa) ETa = Kc * ETo [Eq. 6] 42 Actual turf ET will commonly be less than ETo due to limited soil moisture, difference in turf species and lower mowing height, thus turf Kcs less than one are common. Turf Consumptive Use Studies Estimates of consumptive use and Kc values vary widely due to regional climatic differences, variance in turf height and species, fertilization rate, irrigation method, and different procedures for estimating ETo. Climatic differences dominate the factors affecting ET. Bermudagrass ET, grown in an arid environment was reported to use, on average, 8.7 mm/day (Kneebone and Pepper, 1982). In contrast, much lower ET rates (4.0 mm/day) were reported for bermudagrass grown in North Carolina (Van Bavel, 1966). In Texas, St. Augustinegrass used 12.2 mm of water per day (Johns et al., 1983), whereas in Arizona, ET from St. Augustinegrass averaged 9.6 mm/day (Kneebone and Pepper, 1982). Although cultural management techniques in the above studies accounted for some variance in water use, climatic differences were the primary reason for the wide range of ET rates. Variance in turf species can also have an impact on turf ET. A study by Qian et al. (1996) in Manhattan, KS determined the mean consumptive use of 'Meyer' zoysiagrass (Zoysia japonica Steud.), `Midlawn' bermudagrass [Cynodon dactylon (L.) Pers. x transvaalensis Burtt-Davy] and 'Mustang' tall fescue (Festuca arundinacea Schreb.) was 5.9, 4.9 and 6.6 mm/day, respectively in a one-year study. A three year study by Marsh et al., (1980a) at the California South Coast Field Station determined the consumptive use for warm season grasses to be lower than cool season grasses. Annual 43 consumptive use values for warm season grasses totaled 693 mm for bermudagrass and 635 mm for St. Augustinegrass, whereas cool season grass consumptive use values were 1072 mm and 991 mm for tall fescue and Kentucky bluegrass, respectively. Plant nutrition also impacts turf ET. Improved plant nutrition increases leaf area, and leaf vertical elongation rate, thus increasing turf water use. A study by Danielson et al. (1981), determined that Kentucky bluegrass used 10% more water when supplied with an adequate amount of N fertilizer as compared to the same grass without any application of fertilizer. In Colorado, Feldhake et al. (1984) showed that Kentucky bluegrass (Poa pratensis L. var. `Merion'), grown under deficit N fertility conditions (4kg N/1000 m 2 /yr) utilized 13% less water and experienced a decrease in maximum ET compared to the same turf grown under adequate N levels of (4kg N/1000 m 2 /month). The authors attribute lower ET under deficit N fertilization to slower growth (shorter grass) rates and lighter green turf that reflects more incoming solar radiation (higher albedo). Higher mowing heights typically produce higher rates of ET. Increased leaf radiation surface area resulting from higher mowing heights increases absorption of solar al. (1983) found that and lessens boundary layer resistance, thus increasing ET. Johns et ET of St. Augustinegrass grown in a controlled environment chamber increased by 30% when mowing height was increased from 5 to 8 cm. Biran et al. (1981) found that bermudagrass grown in growth chambers under conditions representative of semiarid 44 climates used 23% more water when mowing height changed from 3 to 6 cm. Consumptive use of ryegrass, grown under the same conditions increased by 25%. Amount and frequency of irrigation can also have significant affects on turf ET. Bermudagrass (Cynodon dactylon 'Suwannee') consumptive use was 16% lower when irrigated two to three times in a 2-week span (dry treatment), compared to six to eight irrigations per 2-week period (wet treatment). Every irrigation event for both treatments supplied water between 20 and 40% in excess of soil water depletion, thus the wet treatment received considerably more water than the dry treatment over the 9-week study period. The same study showed a decrease of 24% in consumptive use for perennial ryegrass (Lolium perenne Tennfine') under the dry treatment compared to the wet treatment (Biran et al., 1981). Total consumptive use of kikuyugrass grown in the arid climate of Israel ranged from a low of 324 mm to a high of 789 mm depending on the amount and frequency of irrigation. Irrigation amounts were designed to refill the soil profile and irrigation frequencies ranged from every 30 days to every 7 days. Consumptive use increased by 60% when irrigation frequency changed from once every thirty days to once every seven days. Differences in consumptive use between fertilized and non-fertilized plots were less than 10% (Mantell, 1966). significant The above discussion emphasizes that climatic differences can have a impacts on ET rates from turfgrasses. In order to normalize ET over differing climates, universal reference crop ET (ETo) is used in conjunction with crop coefficients (Kcs). 45 Assuming the reference crop is managed similarly and the method to calculate ET is the same, Kcs are applicable across different climates. However, ET estimation procedures vary from region to region, as does the management of the reference crop, thus restricting Kc use to areas where local ETo data is accessible. Procedures to estimate ET have shown significant error when compared to one another. Evapotranspiration estimates derived from a Class A pan (Kps) in Irvine, California ranged from 0.52 to 0.63 for warm and cool season grasses, respectively. However, Kcs for the same grasses were determined to be 0.65 and 0.79, respectively when referenced to ETo computed from weather data collected by CIMIS (California Irrigation Management Information System) at the same study site (Meyer and Gibeault, 1986). much Crop coefficients for bermudagrass grown in Reno, Nevada differed by as as 17% depending on the method by which ETo was estimated. The seasonal bermudagrass Kc for evaporation from a from Class A pan was .73, whereas the Kc calculated from the Penman method was .90 (Tovey et al., 1969). coefficients for five A study recently completed in Tucson, Arizona develop crop data different forms of the Penman equation used in concert with meteorological This study measured from an on site AZMET (Arizona Meteorological Network). with developed both summer and winter Kcs for bermudagrass turf overseeded 1996. Winter Kcs intermediate ryegrass during a three year period between 1994 and 46 ranged from a low value of 0.55 to a high of 0.86, and summer Kcs ranged from 0.65 to 0.86, depending upon the form of Penman equation used (Brown et al., 1998). The aforementioned studies demonstrate the diverse array of factors that combine to influence actual turf ET and estimation of ETo and Kcs. Actual turf ET and estimates of ETo vary primarily due to: • regional climatic differences • variance in turf species • fertility management • varying turf height • different watering regimes • different procedures and models for estimating ETo A summary of these studies emphasizes that locally derived meteorological data employed with appropriate use of a model to estimate ETo is essential; otherwise, significant errors in water application will result. Of greatest relevance to this investigation are those data derived either locally or from regions with a very similar climate. Also of great importance are those studies that with a implement similar turf management systems; that is, bermudagrass overseeded cool season ryegrass. Under the extreme environment of Kuwait, short grass (Al-Nakshabandi, (bermudagrass) was grown in lysimeter tanks to measure potential ET 1983). Consumptive use values ranged from 4 mm per day to 11 mm averaged 7.3 mm per day over an entire year. The cumulative per day and annual consumptive use 47 was reported to be 2650 mm. Estimates of bermudagrass ET via the Penman equation (2193 mm) and Jensen-Haise (2088 mm) method were considerably lower than the actual ET values, producing crop coefficients well in excess of 1.0. Turf Kcs in excess of 1.0 suggest possible errors in ET assessment. The high ET estimates can be attributed in part to the relatively small fetch area (10 m x 10 m) surrounding the lysimeter, which leads to high levels of sensible heat advection from surrounding areas. Turf height was maintained fairly high at 5-10 cm, which may cause higher ET rates. Also, the lysimeter tanks were raised 10 cm above the soil surface which can increase the radiant energy load and sensible heat advection, thereby increasing turf ET (Allen et al., 1991). Kneebone and Pepper (1982) measured consumptive use of bermudagrass, zoysiagrass, St. Augustinegrass and tall fescue in Tucson, Arizona using lysimeters that employed a Marriotte siphon, subsurface irrigation system. The Marriotte siphon system was used to deliver water to the bottom of the lysimeters to maintain water tables at certain heights, according to high or low management regimes. High and low management regimes included maintaining water tables at 35 and 45 cm below the soil surface, respectively throughout the first year, and 30 and 40 cm for the following year. High management plots were fertilized monthly with urea at 4.9 g N1m 2 compared to bimonthly treatments at the same rate for low management. High management plots were overseeded each fall with annual ryegrass to simulate high maintenance turf on golf fairways. The mean annual consumptive use values for bermudagrass and zoysiagrass were 1654 mm and 1310 mm for high and low management, respectively. The authors 48 concluded that consumptive use of warm season grasses ranged from 50-80% of Class A pan evaporation depending on management intensity, whereas the consumptive use of cool season grass ranged from 60-85% of pan evaporation. These ET estimates could be inaccurately high because the subsurface irrigation system maintained a high water table, (35 cm below soil surface at high management) thus creating a non-limiting soil moisture environment. As previously discussed, soil moisture significantly impacts ET. The subsurface irrigation system employed by Kneebone and Pepper did not produce a typical diurnal variation in soil moisture that would be expected from sprinkler irrigation, thus some differences in ET estimates could be expected between this study and other studies employing sprinkler irrigation. Consumptive use studies were conducted in Tempe and Mesa, Arizona by Erie et al., (1981). Consumptive use was determined from periodic gravimetric soil samples taken at one-foot intervals throughout the root zone. Bermudagrass consumptive use was determined to be 1105 mm during the period between mid April to mid October (approximately 6 mos.). These estimates appear somewhat higher than those obtained over similar periods in other studies and may result from a high mowing height or errors in assessment of drainage. Turf was moved every four weeks at a height of 3.8 cm (1.5 in) (Garrot and Mancino, 1994), creating more surface area and thus higher ET rates. Gravimetric soil samples were taken throughout the root zone, however, it is unclear if adjustments were made for water lost below the root zone, which is inevitable with the 49 flood irrigation system used in the study. Failure to account for drainage would result in an overestimation of ET. Kopec et al. (1991) determined water use and Kc values from small bucket lysimeters for bermudagrass and perennial ryegrass in Tucson, Arizona. Reference ET was determined by a modified Penman method using automated weather station data (AZMET) located approximately 10 km from the study site. Average consumptive use values determined over four years were 678 mm for perennial ryegrass (October-April), and 871 mm for bermudagrass, resulting in an average annual water application of 1525 mm (61 in.). Under optimal growing temperatures for ryegrass, Kc values ranged from .73-.77, while bermudagrass Kcs ranged from .72-.75 under summer desert conditions. Brown et al. (1996) examined the quality and growth rate of Tifway bermudagrass in Tucson, AZ irrigated with tertiary effluent water daily and every three days at levels of 60, 70 and 80% of ETo. Management was employed to produce fairway quality turfgrass, and the bermudagrass was overseeded with perennial ryegrass each winter. Bermudagrass overseeded with perennial ryegrass irrigated at 60% of ETo did not produce acceptable quality turf. Acceptable quality turf resulted when irrigation was applied at 70-80% of ETo, suggesting an appropriate Kc for both winter and summer turf rests in the range of 0.7 to 0.8. In more recent study, Brown et al. (1998) determined crop coefficients for Tifway bermudagrass overseeded with "Froghair" intermediate ryegrass using large weighing lysimeters irrigated with local groundwater and tertiary treated wastewater. Actual turf 50 ET determined from the lysimeters was compared with ETo computed from five different forms of the Penman equation. As previously discussed, different forms of the Penman equation produced a wide variance in Kcs. However, the modified Penman equation used in Arizona by AZMET showed the lowest coefficient of variation in terms of month-tomonth changes in Kc over the course of the year. Bermudagrass Kc averaged 0.76 for the period May through September, while the Kc for overseeded ryegrass averaged 0.72 for the winter months (October-April). Salinity and Turfgrass Culture It is evident from the previous section that large amounts of irrigation are required to maintain high maintenance turf in arid regions such as southern Arizona. As the domestic and industrial demand for potable water have increased, so to have water costs and legislation restricting water use. Thus, turf facilities are being forced to water under deficit irrigation regimes or turn to alternative water sources. This poses a problem of salinity that is becoming increasingly relevant in arid regions. Salinity danger arises under conditions of: (1) insufficient drainage; (2) irrigation with water containing high levels of soluble salts; (3) deficit irrigation; and (4) arid regions where soils contain appreciable amounts of soluble salts. Salt-affected soils can be classified under saline or sodic soils, or a combination of the two. Saline soils have an electrical conductivity > 4 dS/m as measured by saturated extract (ECe), and an exchangeable sodium percentage (ESP) < 15. Sodic soils have an ECe <4 dS/m, and an ESP > 15. Saline-sodic soils have both an ECe > 4, and an ESP > 15. (Singer and 51 Munns, 1991). Saline soils generally have a pH of 8.5 or less and are well flocculated, whereas sodic soils have a pH above 8.5, and exhibit poor structural characteristics, which can significantly reduce infiltration rates (Harivandi et al., 1992). Salt affected soils negatively impact germination and growth of turfgrass. Visual symptoms of plant stress are typically the earliest indicators of physiological changes or adaptations, yet their evaluation tends to be subjective and error prone. Salt stressed plants exhibit symptoms similar to phosphorus deficiency, including smaller, darkergreen leaves, decreased shoot:root ratio, and decreased tillering (Maas and Hoffman, 1977). High levels of salinity can lead to a delay, or slow development of seed germination (Beard, 1982). As salt damage progresses, leaf tips may die, leaves may fall from the plants, and turf density may decrease (Turgeon, 1996). Rooting depth may decrease, however, root growth may increase relative to shoot growth in certain salttolerant turfgrasses (Marcum, 1994). Devitt (1989) determined that total root length and root density decreased with increased salinity in bermudagrass. Peacock et al. (1993) determined that bermudagrass, under salt stress, exhibited decreased stomatal conductance, stolon length, biomass, and reduced leaf water potential. Reductions in growth and yield of turfgrasses associated with salinity are due to osmotic effects (Hanson et al., 1993). Reductions in osmotic potential of soil water decreases the soil water potential, thus reducing the availability of soil water to the plant. Low osmotic potentials produced by excessive salinity cause the plant to divert valuable energy normally used for plant growth to mechanisms required to alter cell osmotic 52 potential. Lower cell osmotic potentials can decrease tissue water content, top growth, root density, leaf turgor, and photosynthetic rate (Dudeck et al., 1993; Devitt et al., 1993). While osmotic stress is most important in overall growth reductions, excessive accumulation of ions in plant tissue can also be detrimental. Sodium (Na*), chlorine (C1 - ), and boron (B ) can accumulate in plant leaves causing toxic effects. Sodium can k compete with other cations, changing the overall plant nutritional balance. Dudeck and Peacock, (1985) determined that tissue Na+ and Cl contents increased with increasing - salinity, while calcium, magnesium, and potassium decreased. Excessive accumulation of Na+ and Cl can result in leaf tip burn and decreased turf density (Marcum, 1994). - Boron toxicity can be seen on tips of older leaves as yellowing and drying (Ayers and Westcott, 1989). The aforementioned salinity problems can be avoided, or lessened by employing intelligent management practices. A few management strategies include: (1) application of irrigation water over and above ET to facilitate leaching; (2) providing adequate drainage to ensure leaching is possible; and (3) selection of salt-tolerant turfgrasses. Leaching involves applying water in excess of ET, in order to prevent salt accumulation in the root zone. The fraction of water applied that percolates below the root zone is known as the leaching fraction (LF) (Ayers and Westcot, 1989). (LF) depth of w ater leach ed below t he root zo ne depth of w ater appli ed at the surface [Eq. 7] The larger the leaching fraction, the less salt accumulation in the root zone. 53 A related calculation is the leaching requirement (LR), which estimates the LF required to sustain crop yield performance for a given EC of irrigation water (ECiw; Ayers and Westcot, 1989): LR = ECiw 5 (ECe) - ECiw [Eq. 8] where ECe is the soil salinity in dS/m associated with a particular crop yield level and ECiw is in units of dS/m. Ayers and Westcot (1989) indicate bermudagrass will produce maximum yields when ECe is 6.9 dS/m or less. To achieve 100% yield with bermudagrass utilizing irrigation water of 1 dS/m, Equation  indicates the leaching requirement should be 3% (i.e., three percent of the applied water must pass through the root zone in order to leach enough salts to sustain 100% yield potential). LR =100 x ldSm -1 — 3% , 5 0.9dSm -1 ) — ldSm -1 Ryegrass, which is slightly less salt tolerant requires a leaching requirement of 4% to achieve 100% yield using the same irrigation water. It is important to utilize salt tolerant turfgrasses in areas of salinity danger. The relative tolerance to salinity varies widely with plant species. In the past twenty years, a research has identified many salt-tolerant turfgrasses (Harivandi et al., 1992). Although few species can tolerate highly saline conditions, their characteristics may not be danger is desirable for high maintenance turf systems. In arid regions, where salinity most likely, warm season grasses such as bermudagrass, Seashore paspalum, and St. season grasses Augustinegrass are used on golf courses during summer months. Cool 54 such as annual ryegrass, perennial ryegrass, and tall fescue are used during winter months. Harivandi et al. (1992) rated these grasses for salinity tolerance (Table 1). The warm season grasses common to arid turf systems are all rated as tolerant to salinity stress. The cool season grasses are slightly less tolerant to salinity with perennial ryegrass rated moderately tolerant, annual ryegrass rated moderately sensitive and tall fescue rated moderately tolerant. Assessing Soil Salinity Direct Methods Numerous methods exist for determining soil salinity. Numerical values for classification of saline and sodic soils and the relative tolerances of turfgrasses are commonly derived from measurements of ECe by the saturation extract method. Plant response is highly correlated with the ion concentration of the soil solution, rather than the salts contained in the soil (Marcum, 1994). Thus the EC of the soil water (ECw) will affect the physiological functions of the plant. For this reason, the saturation extract is the most widely used method of assessing soil salinity. The saturated paste involves collecting a soil sample from the field, drying the sample, and saturating with distilled water. After a saturated paste is achieved, the soil water sample is extracted via vacuum extraction, centrifugation, or gravity filtration (Bower and Wilcox, 1965). Various methods exist for creating the soil water paste. One method involves saturating the soil to the maximum amount without any water loss (field capacity). This method is rather subjective, thus soil pastes using 1:1 or 1:5 soil:water ratios are being used. 55 Table 1. Salinity tolerance of commonly used cool season and warm season turfgrasses in southern Arizona. Warm-season Cool-Season Name R Name R Annual M Bermudagrass T Seashore T Perennial M I Tall fescue M T St. ' i --The rating reflects the general difficulty in establishment and maintenance at various salinity levels. It in no way indicates that a grass will not tolerate higher salinity levels with good growing conditions and optimum care. The ratings are based on soil (ECe) levels of: Sensitive (S) = <3 dS/m, moderately sensitive (MS) = 3-6 dS/m, moderately tolerant (MT) = 6-10 dS/m, tolerant (T) = >10 dS/m (adopted from Harivandi et al., 1992). 56 Electrical Conductivity can also be measured by direct extraction of the soil water via porous suction cups. Various problems are associated with this method. The soil water obtained by solution samplers may not be representative of the entire water-filled pore space (Ward et al., 1994). The solution may extract the water from large pores (under low tension), and exclude the water in smaller pores (held under high tension). The water in the small pores held at the solid/liquid interface is considered immobile, thus the ion mobility associated with this water is far less than ions located in the central pore water. The decreased ion mobility associated with the water held in small pores can affect ECw measurement (Mualem and Friedman, 1991). Also, solution samplers are not effective when soil matric potentials are low (low 0v) (Heimovaara et al., 1995). The measurement of ECw via suction samplers is similar, but does not equal the ECe measured from saturation extract (Rhoades et al., 1989). The ECe value is considered to be somewhat lower (-2x) than the ECw, (Ayers and Westcot, 1989) due to dilution of ions with distilled water. Time Domain Reflectometry Time domain reflectometry (TDR) is a less tedious method of assessing soil salinity. Numerous authors have found TDR to produce rapid, reliable, and routine et al., 1982; measurements for in situ volumetric water content (Topp et al., 1980; Topp Dasberg and Dalton, 1985), and more recently, techniques have been developed to obtain probes both bulk soil electrical conductivity (ECa) and water content using the same TDR (Dalton et al., 1984; Topp et al., 1988 and Zegelin et al., 1989). 57 In theory, measurement of water content and bulk soil electrical conductivity using TDR is independent of soil texture, structure, density, and temperature, and no calibration is required. A cable tester (TDR) measures the propagation velocity of a highfrequency pulse through parallel metal probes inserted in the soil. The time interval required for the electromagnetic pulse to travel from the point at which the probe enters the soil to the end of the probes and reflect back to the point of entry in the soil varies with transmission line length and the dielectric constant (K). The expression to define K is given: K = [ct/21] 2[Eq. 9] where c is the velocity of light in a vacuum, t is the transit time, and 1 is the path length (Dalton et al., 1990). Changes in K are primarily dictated by changes in 0„, regardless of soil type. Topp et al. (1980) developed an empirical relationship between K and O v : Ov = -5.3 * 10' + 2.92 * 10' K - 5.5 * 10 4 K2 + 4.3 * 10 6 K3 [Eq. 10] This empirical equation is still widely used, however, a site specific calibration curve is needed to obtain very accurate results for peat and heavy clay soils (Dalton and Poss, 1990). The travel time is measured in distance units on the cable tester screen (Figure 3). This distance can be read directly from the screen or digitized and transferred to a computer. Direct readings can be subjective and error prone. A computer algorithm can be used to determine the tangent of the two distance points to accurately calculate the 58 V o a I Time (ns) 1 Figure 3. Typical TDR output trace used to assess soil volumetric water content. Label A identifies point when pulse signal enters the soil. Label B identifies point where the reflected signal has returned to the soil entry point. 59 dielectric constant (Yanuka et al., 1988; Nadler et al., 1991). Numerous computer programs are now available to determine moisture content and salinity (e.g., Evett, 1994). The TDR measures bulk soil salinity independent of soil moisture. The water content affects the horizontal distance of the TDR trace, (Figure 3) whereas, the salinity of the soil affects the attenuation of the signal (Figure 4). The bulk soil salinity (ECa) is comprised of the bulk liquid-phase conductivity (ECb) associated with the salts contained in the liquid-filled pores, and the bulk surface conductivity (ECs), which is associated with the salts contained at the solid/liquid interface (salts derived from exchangeable cations adsorbed on the cation exchange sites of clay surfaces). The relationship is given by Rhoades et al., 1976): ECa = ECb + EC s [Eq. 11] Rhoades et al. (1990) reported that ECs values for typical arid land soils of the southwestern United States may be estimated from the water saturation percentage (SP) of mineral soils. The following equation relating ECs to SP was developed for Arizona and southern California soils by Rhoades et al (1990). ECs = 0.019 SP — 0.434 [Eq. 12 1 The bulk soil electrical conductivity (ECa) is given by Dalton et al. (1984) in terms of the transmitted voltage (V T), the reflected voltage (V R) (Figure 4), the length of the waveguide (L) and the dielectric constant (K): ECa = / 120nL (ln * V T/V R) [Eq. 13] 60 — V o 1 t a g e ITime (ns) I Figure 4. Typical TDR output trace used to assess bulk soil electrical conductivity. Arrows labeled VT and VR depict the transmitted and reflected voltages, respectively. 61 The transmitted voltage is the voltage that enters the waveguide, and V R is the voltage reflected from the end of the probe. The soil salinity attenuates the transmitted voltage pulse, VT, and reduces the amplitude of the reflected voltage V R (Dalton et al., 1990). Values of VR can represent different amplitudes depending on where the signal is read, (i.e., after partial reflection, from the start of the probe, after reflection at the end of the probe and reflectance over a long time) (Figure 4). Nadler et al., (1991) determined that V R should be measured after a very long time, when it approaches a constant value, thus avoiding multi-reflection interferences caused by impedance discontinuities. The ECa value is very low compared to ECw values obtained by solution samplers and ECe via saturation extract. The low ECa results because rock material and soil solids have low electrical conductivities and thus act as insulators. Several models have been developed to convert ka to the more useful value of the saturation extract ECe or soil solution ECw. Rhoades et al. (1976) developed a model to estimate ECw from ECa using Equation  and the assumption that ECb is a linear function of ECw. Rhoades et al., (1976) reworked Equation  to determine the relationship between ECa and ECw for unsaturated soils: ECa ECw ev T (0v) + ECs [Eq. 14] where ev, T(0v), and ECs are the soil volumetric water content, soil water transmission coefficient, and the soil solid phase conductivity, respectively. The transmission coefficient ( 1) accounts for the tortuous nature of the current flow through the geometric 62 matrix of the soil pores and the decrease in mobility of ions at the solid/liquid interface. The transmission coefficient can be described by the empirical equation: T = a Ov + b [Eq. 15] where a and b are constants and 0v is the volumetric water content. Constants a and b will vary due to soil mineralogy and texture. The Rhoades model is but one of many developed to estimate ECw from ECa. Other models developed include Mualem and Friedman (1991), Heimovaara et al. (1995) and Persson (1997). These models serve as useful predictive tools for monitoring changes in soil salinity over time, however, the models do not provide accurate estimations of ECw over wide ranges of soil salinity. Achieving a strong relationship between ECa and ECw is difficult due to the different media by which each parameter is obtained. Bulk soil salinity is measured via TDR, which measures the conductivity of both the mobile and immobile soil solution plus the soil solid phase. In contrast, only the mobile portion of the soil solution is measured by solution samplers when obtaining ECw. Therefore, an accurate relationship solutions between ECa and ECw can only be made when the mobile and immobile soil are in diffusional equilibrium (Persson, 1997). An additional source of error relating ECa to ECw occurs when either ECa or ECw values are very low. The relationship between ECa and ECw at very low salinity values is non-linear (Nadler and Frenkel, 1980). An independent study by (1989) revealed that ECw values measured with an ac bridge Zegelin et al. compared favorably with 63 ECa values derived from TDR (within ± 10%) provided ECa exceeds 1.0 dS/m. A study by Persson (1997) concluded the accuracy of ECa and ECw measurements increases with ECa. Error associated with the ECw estimation was <10%, provided ECa values were higher than 0.6 dS/m. Dasberg and Dalton (1985) determined ECa values below 0.18 dS/m did not compare favorably with ECw values obtained from the same samples. 64 CHAPTER 3 MATERIALS AND METHODS Lysimeter Facility and Operation This study was conducted in the field at the Large Weighing Lysimeter Facility located at the University of Arizona Karsten Desert Turfgrass Research Facility (KDTRF) in Tucson, AZ. The field research plots at KDTRF are located on 2.2 ha of bottom land that resides just south of the Rillito River at an elevation of 717 m above mean sea level. The soil at the KDTRF is classified as an Agua sandy loam (coarse-loamy over sandy or sandy-skeletal, mixed, calcarious, thermic Typic Torrifluvent). The Large Weighing Lysimeter Facility is centrally located within the KDTRF and 4.0 m (Figure 5). Two large, cylindrical weighing lysimeters, 2.5 m in diameter deep, filled with a Vinton fine sand (sandy, mixed, thermic Typic Torrifluvent), were used in this study. The lysimeters, labeled (W) and (E) in Figure 5, referred to as the west and east lysimeters, respectively. The remove coarse fragments in excess of 2 mm prior to will hereafter be Vinton soil was sieved to placement in the lysimeters. The soil was then packed to a uniform bulk density of 1.5 Mg/m 3 and 1.49 Mg/in 3 in the west and of the east lysimeters, respectively. Particle size analyses conducted on samples lysimeter soil revealed the Vinton soil was 90% sand (35% fine sand, 35% medium sand), 7% silt and 3% clay on a weight basis. Saturated hydraulic conductivity of the 65 65m 86m 95m 90m IN her Station Karsten Facility North 11 Figure 5. Plan view of the Karsten Desert Turfgrass Research Facility, Tucson, AZ. Turfed areas are shown in green. The lysimeters are located in the middle of the facility. Labels W and E identify the west and east lysimeters, respectively. The arrows and associated distances represent the distance from the lysimeters to the turf facility boundary. Map not drawn to scale. 66 lysimeter soil was determined to be 150 cm/day using the constant head method of Klute and Dirksen (1986). The lysimeters are located within a uniform 918 in' block of irrigated turf (Figure 5). The area outside the immediate vicinity of the lysimeters consists of irrigated turf plots interspersed with plot alleyways and cart paths and thus does not provide a uniform turf fetch which is preferred for evapotranspiration studies. Turf occupies approximately two-thirds of this surrounding area; the remaining area is allocated to the alleyways and paths. The length of fetch in the upwind direction is most critical for ET studies. Wind flow at the KDTRF is periodic in nature, moving up river (from W or NW) during the day and flowing down river (from E) at night. Fetch lengths west and east of the lysimeter area measured 86 m and 95 m, respectively. The two lysimeters rest on truck scales (Model FS-8, Cardinal Scale, Webb City, MO) equipped with loadcells (Model Z-100, Cardinal Scale) which can resolve changes in lysimeter mass of approximately ±100 g (0.02 mm depth of water) under controlled calibration conditions. However, potential errors from electronic drift, wind and mechanical limitations suggest a realistic daily resolution is between 0.5 and 1.0 kg (0.10.2 mm of water). Loadcell calibration was checked twice over the course of the study by comparing loadcell output following addition and subtraction of known weights. al., Additional details on lysimeter construction and layout are described in Young et (1996). 67 Both loadcells are connected to a datalogger (Model CR-7, Campbell Scientific, Inc., Logan, UT) programmed to sample the lysimeter mass once every ten seconds and compute and output mean values of lysimeter mass at ten-minute intervals (see Appendix F). Data are retrieved from the CR-7 using a personal computer and the communications software Telcom (Version 6E, Campbell Scientific, Inc., copyright 1990). Data are downloaded to a disk and subsequently entered into a spreadsheet program (Quattro Pro, version 7.0-Pro) for ET analysis. Evapotranspiration is determined on a daily basis by analyzing the diurnal pattern of lysimeter mass change. Changes in mass in kg are converted to an equivalent depth of water in mm by converting the mass change to an equivalent volume of water in crn3 then dividing by the surface area of the lysimeter in crn2 . The resulting equivalent depth of water in cm is multiplied by 10 to convert the units to mm. An example follows for a mass change of 29 kg which occurred on day 170 (19 June, 1998) for the east lysimeter (see Figure 6). Mass change = 29 kg = 29000 cm 3 water Lysimeter Area = 49087 cin2 Equivalent Depth of Water = 29000 crn 3 / 49087cm' = 0.59 cm = 5.9 mm is Figure 6 depicts lysimeter mass change for day 170. The vertical axis in Figure 6 converted to units of mm of water using the conversion from kg of mass change to mm of water described above, and the midnight mass value is assigned a value of 0 mm. Day 170 runs from point (A) through point (F) in Figure 6. The abrupt increase in lysimeter 68 0 I.LI. 63 in c' IL 2 OL I. a I-- >. co 0 9*OL I. V 2 a$ E '7) >, ›, 06 U) ctl o to — co d c co I ci) E p U) c o o. u) V OL I. cu cL a) To t.) 0 ,_ a) a1 E '7) 4 >, _1 Z OL I. 0 (Ni I C . n q- ' -- cd 0 — — (5 ô 0 ci o'OL l• co N -a- 00 WW o ti! Jalem Jo todaa 'T CO N 69 mass results from irrigation (points B to C). Removal of drainage water produces the abrupt decrease in lysimeter mass (points D to E). The more gradual loss of lysimeter mass between points C and D and E and F represents ET. Evapotranspiration is assumed to be negligible during irrigation events because irrigation is applied over a period of thirty minutes or less during the pre-dawn hours. Linear interpolation is used to estimate ET during daytime irrigation events and when drainage water is removed. The mass change during such events is estimated by determining the slope of the line representing the change in ET per unit time during the period thirty minutes (three data points) prior to and thirty minutes following the irrigation or drainage event. The average slope of these two lines is computed and used to estimate the rate of ET per unit time during the irrigation or drainage event. The ET is then estimated by multiplying the rate of ET per unit time by the total time elapsed during the event. Total ET for the day is determined by summing the changes in equivalent depth of water as depicted by points A to B (+0.26 mm), C to D (-2.55 mm), and E to F (-4.48 mm) to the estimated ET occurring between points D to E (-0.41 mm). The resulting summation is -7.18 mm. ET is reported as a positive value when there is a loss of water; thus, ET is reported as 7.18 mm. Drainage water reaching the bottom of the lysimeter is extracted via a vacuum pump attached to a series of ceramic suction candles (25.4 mm OD, 610 mm long, Soil Measurement Systems, Tucson, AZ) located 25 cm above the bottom of the lysimeters. The vacuum pump applies a suction of approximately 150 kPa for a period of four hours 70 daily to remove the drainage water. Drainage water is stored in containers situated on the tank scale, thus daily drainage events do not impact lysimeter mass. Drainage water is removed and quantified weekly using both volumetric and gravimetric procedures. Volumetric procedures involve reading the depth gauge on each drainage tank, then converting the depth to volume using the linear calibration equation for the depth gauge. Gravimetric procedures simply use the change in lysimeter mass obtained before and after the drainage water is removed from the tank (see Figure 6). Drainage volumes are converted to an equivalent depth of water using the procedure described previously. A subsample of the drainage water is acquired during each drainage event. Subsamples are placed in a plastic container and then stored in a freezer at -28 °C for later analysis. Turf and Irrigation Management Tifway (Tifton 419) bermudagrass, (Cynodon dactylon x transvalensis (L.) pers.), sprigged during the summer of 1994 over the entire lysimeter facility (lysimeters and was surrounding turf), served as the summer turf surface for the study. The bermudagrass mowed two to three times per week at a height of 1.9-2.2 cm. Both lysimeters and the surrounding turf block were overseeded on 13 Oct. 1997 with `Froghair' intermediate was allowed ryegrass (Lolium multiflorum x perenne). Prior to overseeding, bermuda turf to increase in height (approx. 4 cm), then was mowed to about lcm (scalped). Thatch was then removed by raking the surface and disposing of the resulting material. Intermediate ryegrass was overseeded at a rate of 733 kg/ha (15 lbs/1000 ft 2 ) and worked into the soil surface to ensure soil contact. During the two-week period required for 71 germination and establishment of the overseeded turf, daily irrigation amounts of approximately 4.1 mm/day were applied using six application periods to ensure the soil surface remained moist. Once established, overseeded turf was mowed one to two times per week at a height of 2.2-2.5 cm. The lysimeters were overseeded again on 26 Jan. 1998 due to a thinning stand of ryegrass. Froghair intermediate ryegrass was again used, however, the seed was pregerminated in water to hasten the germination process. The turf was de-thatched in order to open the canopy to ensure proper seed-soil contact. Each lysimeter was top-dressed with soil identical to that contained in the lysimeters (Vinton soil). Topdressing was applied at a depth of 3.2 mm (1/8 in.) over the surface of each lysimeter. Irrigation was applied at a daily rate of 2.7 mm/day using three application periods to ensure adequate surface soil moisture for germinating seed. This irrigation schedule ran from 27 January through 4 February. Lysimeter turf was fertilized every two to four weeks using liquid fertilizer (see Table 2 for fertilizer rates and dates of application). The fertilization regime before and N. after 2 Feb. 1998 differed to facilitate a companion study on the fate of fertilizer Applications of fertilizer N prior to 2 Feb. 1998 were made using ( 15 NH 4 ) 2 SO 4 . Potassium, phosphorus and micronutrients were supplied prior to 2 Feb. 1998 using 1/2 strength Hoaglands solution (Hoagland and Arnon, 1950) containing 5.5 g of potassium 72 Table 2. Fertilizer rates and dates applied throughout the study. Date 97/07/07 Fertilization Rates and Dates Applied Rate Fertilizer West East Kg/ha Kg/ha 12.1 24.3 (NI-14)2SO4 ° 97/07/29 (NH4)2 S 4 18.2 15.2 97/08/18 (NH4)2SO4 24.3 15.2 97/09/10 97/10/21 97/10/22 97/10/24 98/10/28 98/10/29 97/11/07 97/11/14 97/11/26 97/12/14 97/12/29 98/01/12 98/02/02 98/03/09 98/04/05 98/04/24 98/05/12 98/06/05 98/06/24 98/07/11 98/08/08 (NH4)2SO4 P and K P and K 24.28 2liters 2liters 21.25 2liters 2liters 18.21 2liters 18.21 2liters 18.21 2liters 12.2 12.2 12.2 12.2 24.4 24.4 24.4 24.4 24.4 15.18 2liters 2liters 15.18 2liters 2liters 12.14 2liters 12.14 2liters 12.14 2liters 12.2 12.2 0 6.3 5.1 7.8 10.5 15.3 11.7 , (NH4)2SO4 P and K P and K (NH4)2SO4 P and K (NH4)2SO4 P and K (NH4)2SO4 P and K Peters Solution Peters Solution Peters Solution Peters Solution Peters Solution Peters Solution Peters Solution Peters Solution Peters Solution 73 After 2 Feb. 1998 fertilization was accomplished by applying 24 kg/ha (0.5 lb./1000 ft2 ) of a standard soluble 20-20-20 liquid fertilizer (Peter's Solution; Grow More Inc., Gardena, CA) at regular intervals (Table 2). A total of 2 1 of fertilizer solution was applied to each lysimeter on each date of fertilization. Similar amounts of N were applied to both lysimeters over the course of the study. However, the amount of fertilizer N applied to each lysimeter was adjusted to compensate for differences in N concentration in the respective irrigation waters (See Appendix A for complete water analysis). At each fertilization event, the amount of N required by the turf was reduced by the amount of N supplied in the irrigation water since the previous fertilization. The turf quality on each lysimeter was rated weekly using the USDA National Turfgrass Evaluation Program (NTEP) standards by which turf color, density, and overall quality were assessed independently on a 1-9 scale. An NTEP rating of 6 or higher represents a turf of acceptable quality or better. Turf growth rate was determined by collecting turf clippings during mowing events and drying to a constant weight at 40°C. Growth rate in units of g/m2 /day was computed by dividing total clipping dry weight by the surface area of the lysimeter and the number of days between mowing events. Irrigation water was supplied to each lysimeter by four low trajectory Rain Bird 1804 Series pop-up sprinkler heads (Rain Bird Sales, Inc. Glendora, CA) installed in a was monitored square pattern with a head spacing of 3.65 m (12 ft). Precipitation rate periodically using metal catch cans and averaged 5.3 cm/hr. System operating pressure 74 was maintained at approximately 207 l(Pa (30 psi) utilizing pressure regulators. Irrigation uniformity was determined on a regular basis using Christiansen's Coefficient (Christiansen, 1942) and ranged from 89 to 95%. Irrigation was controlled by a Rainbird Maxi-5 irrigation control system. Irrigations were applied in the pre-dawn hours to minimize (1) problems with wind-induced distortion of sprinkler patterns and (2) losses due to evaporation. Irrigation amounts were not adjusted for irrigation nonuniformity. Each lysimeter received irrigation water from a different source. The west lysimeter was irrigated with tertiary effluent water from the Pima County wastewater treatment plant. The ECiw (EC of irrigation water) of the effluent water was approximately 1.0 dS/m and contained approximately 13 ppm (mg/1) NO 3 -N. The east lysimeter received potable groundwater, which has an ECiw of approximately 0.25 dS/m and a NO 3 - N concentration of 3 ppm (mg/1) (see Appendix A for water analysis). Irrigation treatments consisted of applying each water source to a single lysimeter at annual rates not to exceed the 1.4 halm/ha turf water duty presently required by ADWR in the Tucson AMA. Irrigation was applied daily with application rates determined by applying the crop coefficients presented in Table 3 to values of reference evapotranspiration (ETo) computed by the Maxi-5 weather station located approximately 10 m south of the lysimeters. The Maxi-five system computes ETo using the procedure described by Jensen and Haise (1963). These ETo values were reduced by 10% prior applying the Kcs in Table 3 to make ETo compatible with values generated by AZMET (Brown, 1998). Irrigation was diminished or eliminated by the Maxi-Five Irrigation 75 Table 3. Crop coefficients (Kcs) applied during summer (15 June-13 October) and winter (5 November-4 June) seasons and Kcs applied during fall overseed (14 October-4 November) and spring transition (5 June-14 June). Kc Time Period 0.76 15 Jun- 13 Oct. 1.10 14 Oct. - 4 Nov. 0.72 5 Nov - 4 June 1.00 5 Jun. - 14 Jun. 76 System following precipitation events. Treatments were initiated on 21 July 1997 and continued through 30 Sept. 1998. Difficulties with regulating the pressure of the irrigation system resulted in somewhat inaccurate water applications prior to mid-August 1997. The Kcs in Table 3 generate an annual water application of approximately 1.55 ha-m/ha (5.1 acre-ft/acre) in a normal year if precipitation is not considered. On average, precipitation adds approximately 0.3 ha-m/ha, which should lower application rates below 1.4 ha-m/ha if precipitation is used efficiently. Plans were developed to reduce the Kcs if rainfall and weather conditions indicated annual water use would exceed 1.40 ham/ha. However, the ensuing year produced above normal precipitation and plans to reduce Kcs were never implemented. Water and Salt Balance Procedure All parameters of the water balance were assessed daily over the course of the study. Irrigation was calculated from lysimeter mass gain during irrigation events. Precipitation was estimated from lysimeter mass gain during rainfall events. Lysimeter measured precipitation was used in lieu of rainfall measured by the AZMET weather station due to greater accuracy obtained with lysimeter measured precipitation [see was Appendix D (Equations 17 and 18) for raingage precipitation adjustment]. Drainage determined from the loss in lysimeter mass during drainage events, and evapotranspiration was determined from lysimeter mass changes as described previously. Runoff was assumed negligible in this study. Daily values of each component of the 77 water balance were collected and stored in a spreadsheet program (Microsoft Excel, version 7.0), and then summarized at the end of each month. Water balances for each month are presented in Appendix B. An annual water balance was developed for the period October 1997 to September 1998 by summing the water balance parameters from each of these twelve months. The change in soil water storage was determined for the entire year (October 1997 - September 1998) by computing the difference between the total inputs to each lysimeter (irrigation and precipitation) and the total outputs from each lysimeter (turf ET and drainage). The salt balance of each lysimeter was assessed on a monthly basis by quantifying the amount and EC of drainage and irrigation waters, respectively. The EC of drainage and irrigation water was measured using a YSI Model 35 Conductance Meter (Yellow Springs Instrument Co., Yellow Springs, OH). The cell constant (k) was 1.0/cm and probe resolution was ±0.01 dS/m using the appropriate display range. Cell accuracy was determined using ten prepared solutions ranging from 0.33 dS/m to 15.24 dS/m. Cell calibration confirmed the cell accuracy of ±1%. Water samples were thawed and allowed to equilibrate at room temperature prior to measurement. Salt input to each lysimeter was determined by multiplying ECiw by 640 to obtain the total dissolved solids (TDS) in mg/1, then multiplying TDS by the volume of irrigation water applied during each irrigation. An example follows for 3.0 cm of applied effluent irrigation water (ECiw 1.0 dS/m): TDS Eciw * 640 1.0 dS/m * 640 640 mg/1 78 Volume of Water = lysimeter area * depth of irrigation =49087 cm' * 3 cm = 147,261 cm' = 147.26 1 Total Salt Input = 640 mg/1* 147.26 1= 94,247 mg = 94.2 g The water contacting the lysimeters from rainfall was assumed to have negligible TDS, thus salt iput from rainwater was assumed to be zero. The amount of salt removed in the drainage water was determined by computing the grams of salt per drainage event. An example follows for a drainage event that occurred on 3 Sep. 1998, where 34.8 1 of drainage water was quantified with an EC of 2.14 dS/m. TDS2- drainage water EC * 640 = --- 2.14 dS/m * 640 mg/1= 1369.6 mg/1 - Volume Drained = 34.8 1 Total Salt Lost = 1369.6 mg/1* 34.8 1= 47662 mg = 47.7 g The balance of salt applied versus salt removed in drainage was determined on a monthly basis by computing the difference between the grams of salt applied and the grams of salt removed. The annual salt balance for the period October 1997 to September 1998 was computed by subtracting the total mass of salt drained from the lysimeters from the total mass of salt applied in irrigation water. The leaching fraction was computed for the entire year (October 1997 - September 1998) by dividing the total drainage water by the total amount of water applied (irrigation + rain). Subsurface Salinity Assessment The lysimeters are equipped with 96 external sampling ports that facilitate installation and utilization of solution samplers, tensiometers, thermocouples and time 79 domain reflectometry (TDR) probes. The sampling ports on each lysimeter are divided into three vertically aligned sets. Adjacent sets of ports are separated by an angular distance of 120 0 . In the vertical dimension, sampling ports are separated by 50 cm over the depth range of 100 cm to 350 cm. An additional set of sampling ports was available at a depth of 50 cm in the west lysimeter to facilitate monitoring near the soil surface. Time domain reflectometry was used to assess vertical profiles of volumetric soil water content (0v) and bulk soil electrical conductivity (ECa) at weekly intervals. Horizontal TDR probes were available in sampling ports at each depth increment. The TDR probes (Dynamax, Inc., Houston TX) consist of three parallel 0.3 cm OD waveguides, 20 cm in length and spaced 3 cm apart. The probes were connected through RG 58/U, 50S 2 coaxial cable to a TDR cable tester (Model 1502, Tektronix Corp., - Beaverton, OR). Cable tester settings were set as follows for the Ov measurements: 0.5 Ft/div for horizontal control, 100 mp/div for vertical control and 0.99 for the cable dielectric constant. The vertical control was compressed to 200 mp/div to obtain the transmitted voltage, V T , and the reflected voltage, V R , necessary for assessment of ECa. A personal computer was used to sample and analyze the output from the cable tester via the computer serial port. The computer program, TACQ, (Time Domain Reflectometry (TDR) Automatic Data Collection Program, Copyright 1993, Vadose Zone Equipment Company, version of 7-24-94, provided by Steven Evett, USDA-ARS) was used to collect and analyze TDR traces for travel times, apparent dielectric constants and soil volumetric water contents. This program measures the path length of the signal in 80 distance units (d), and using a known pulse velocity (V) calculates the travel time (t) in nano-seconds (Dalton et al., 1984): t= d/V [Eq. 16] The pulse transit time is determined from the point at which the pulse signal enters the soil (point A in Figure 3) to the point at which the transmitted pulse signal has returned to the point of entry in the soil (point B in Figure 3). Direct measurements may introduce error into locating the points of inflection, thus the TACQ computer program is used to determine these points by locating the local maximum and minimum points on the output trace. Equations  and  are then used to yield the dielectric constant (K) and volumetric water content (ev), respectively. K = [ct/2L] 2 [Eq. 9] ev = -5.3 * 10 -2 + 2.92 * 10 -2 K - 5.5 * 10 4 K2 + 4.3 * 10 6 K3[Eq. 10] where c and L are the speed of light and probe length, respectively. Bulk soil electrical conductivity was analyzed manually using the method described by Dalton et al., (1984) which requires measurements of the transmitted and reflected voltage (Figure 4; VT and V R). Bulk soil electrical conductivity was computed from Equation  using V T , V R , K obtained from Equation  and probe length (L). ECa = K 1/2 / 1207cL (ln VT V R) [Eq. 13] It should be noted that VR was determined after it approaches a constant value (Figure 4), according to Nadler et al., (1991). 81 Crop Coefficient (Kc) Assessment Crop coefficients were computed on a monthly basis by dividing the monthly sum of the lysimeter ET by the sum of ETo as computed by AZMET. Seasonal Kcs for summer (May-Sep) and winter (Nov-Apr) were calculated as the average of the monthly Kc values during the appropriate time period. To assess day-to-day variation in Kcs, Kc values were calculated each day by dividing lysimeter ET by ETo. From these data, a mean Kc value and standard deviation were computed on a monthly basis. The Kc values were also analyzed to determine statistical significance of seasonal and locational (east vs. west lysimeter) affects on Kc. To accomplish this, a regression model was used. The daily Kc values were entered into a spreadsheet program (Microsoft Excel, version 7.0) for each lysimeter and for both summer and winter seasons. Next, a dummy code was assigned to season (summer = 1, winter = 0) and to location (west = 1, and east = 0). These codes were then regressed (as independent variables) against Kc using the model described by Johnston (1984): Kc = + a i L i + p i s, + u [Eq. 17] Statistical model usewhere the parameter u represents the expected Kc for the east lysimeter during the winter season, oc i represents the differential effect for L i (west lysimeter location Kc) compared with L2 (east lysimeter location Kc), the f3 i parameter measures the differential effect for S i (summer season Kc) compared with S2 (winter season Kc) and u is the random error in the model. 82 CHAPTER 4 RESULTS AND DISCUSSION The objectives of this study were to evaluate the adequacy of the ADWR water duty to meet turfgrass ET requirements, to provide for fairway quality turf, and to evaluate any potential for salinity damage to turfgrass under this irrigation regime. These objectives were assessed by evaluating the lysimeter water balances, salt balances, turf quality, biomass accumulation, turf growth rates, and crop coefficients. Water Balance The components of the water balance for each lysimeter are presented by month in Table 4 and Figure 7. While August and September 1997 are presented in Table 4, the water balance for both months was not considered representative of the irrigation regime employed by this study due to residual effects from the previous study. Irrigation was applied in the previous study to create non-limiting soil moisture conditions, thus three months prior irrigation was supplied well above ETa (approx. 120% of ETa). The to October 1997 served as a transitional period during which the soil water dynamics of the lysimeters approached a new equilibrium representative of the irrigation regime soil employed in this study. High levels of drainage and large decreases in lysimeter water storage in August and September 1997 provide ample evidence of lysimeter consistent adjustment to the new irrigation regime. October drainage decreased to levels with monthly drainage measured later in this study; thus, October was selected as the 83 tili Z — In LU cp 0 41.( LO• .1- 00 0 LO L41 cD LO Cn T- v- 14/ ..., 6 6 • cri 6 6 6 ,- '''. c,:.:2 6 6 ,,r cv c\I '7 4 co co 4 ''' .;co cv 6 E --,_ cv 6 .7 = , c,.; E r-. Lo. _>,1 ...-, ) ZW < 0 rA 1 H co () (1) LIJ CT.) 'Jo co ill (.9 • (7) E d d 0 < fi' E CO 01 ' ' Z iY , <o co I H ci. C.) u) i3 't 0 03 -cr cs) c) h. ''- Y- C) Cr). 03 CI • 06 (D c,i 6 CV cri r...: CD 6 4 6 c‘- . CO C \ I N- 1 .4- 't 1- 4 CV 7 .4 ) Q) ,- o) ..--, cs, Ct/ . 7) ›., 0 Q) 6. a — 0 .8 0 1-. a) 06 t'-'3 -I-. cd CIN • ,--n 0 n-, 1--, a) CO CO E cd 06 co E Lo OE/ ..-x 1-u) Lu NT T UJ V) w a) N (1) Cn ...-. (J E c(..,\jj r....: (1.- co '-' -I "0 0 E c:Ts ..._., N- 't . a.) *(7) a _i ,..C1 ,-, o o "o T=-14 o o cv ra. = Q) a) 17. o) ,--, CD LU E ai CV , cri •E r .4,— —.... CO r— a) -c-',0I >N .-,•54 ... cr\ ,_, ar on c, Q) -., 4-4 oe N. c '-E-' .,.. " W .M E ' 141 LU E ._ o 4—> cn cn (I) o _0 E E un ....... - . co c )^ _I Vs o iz 6 6 (1) N. as .c- co 7.6. == cD CL (0 1= CL I‘- ct w .174 CI) 0 o ,, o 0., o o (1.) al 7,5 C\J. T- 14/ CD UD CD CD CD ..4 -.\-j 6 6 ci ,i: csi ,-* Lri -4: 6 ,r CO CA CO CO CO CA CO ,4- CA T- - CO LU (1) T- T'- CO CD 01 CD co CD. to co co ‘._ ci (.6 d 6 6 .- c‘i 4 c<1 01 03 CD r- r- 0/ ,T- Nt T- r- co uo Ul CD T- T- ,n ,n Tn T T- T I..... LI/ Nt 0/ T. 03 CD U1 CV UP .-- ,r CO CD CD OD CA CD r- 0D 'Cr CV T.,-) co cq CA CA CO CV CO CO 7." T. 0q (\I 14/ 03. CA CO cn cv a) CO c—• Lo' 00• ..4-• r-• 7c: N ce) .1. L° CD CA 'Tr• r-• 03• Y-. c-• c-. 6 0D CD Ul cl- CD CO co 0., r r r c\i r r r "Iz.- LLI a) › 6 N: 6 1.41 4 CA in to 0\ (-1 n1 03 CD CA n,.. - _Z cn a) Lo E ' c° E cT? i. --- ,- (c) CI) D a. .. L. (d —' C° ....: . N". a) •Fs E 'A ,47 ..., a) cC E ._ E cn ›.., E a, — cZ't — 6. (4 . Q) r I') 5 -0 c i cy; cr) - a) r,cu -_,_, , > cci ... 03• CD CO CD 0 01 CO CD 04 cD co CA CD ,..6 r--: 4 cci c i d 6 ci ci cfi cti co r-: cl- YCD 01 CA vl0) 1-- Y- OD LID. CD• CO• CD. CO. CD. r1....: T. 03 rn • .„.4: ,- CV CA Ul l's -co -a'cr CO 03 ."' CD '- r- CO P '-' y- U c\ 1 r .,-. „..- CV A-- LO r-cCD co r- oo 04 T. • CD UD 01 cD 01 OD 1.0 cs4 cD 01 01 CD 6 ui 6 • • LU CO ,- CO 00 • N- CO C) C) , r CD 01 CA y- y- OD UP CD .1- c- CD 04 UD 'Cl- r...7 co ,r. Up 6 co CD 0 o, co r ID r • r) ri c'''. • • a' cri cn 06 c) 01 c\I r- ill CD 00 Ul ,r ,r ,- N.- T_ ,- T— c— cy ,-- T. CD CL it < co N- rCD CD = 'é -. 0 2 6) 6_ = o < (.0 To .A. r- r- CO 03 03 CO 03 03 co CO CO c si oi c ,» 0,.) c ,» cs).1.,:::: cn Q) O) cn, a)....... cy) 0 c ..0 67, 8:5 c= —= = cll ,-..., > o a) cs o r- OZO—) u_ 2 < 2 -, -) < u) 73_ 84 Water Inputs (Irrigation + Rain) and Outputs (ETa + Drainage ) for Both Lysimeters r-. 33 O) D. (0 0 West Inputs o a) 0 Q. IL . ) -3 • East Inputs 0 West Outputs 0 East Outputs Figure 7. Monthly inputs and outputs of water for the west and east lysimeters for the period August 1997 to September 1998. Inputs (irrigation + precipitatioin) and outputs (ETa + drainage) are presented as an equivalent depth of water in mm. 85 month where the water balance first reflected the irrigation regime used in this study. The twelve months from October 1997 - September 1998 were therefore chosen for assessment of the lysimeter water balances in this study. Weather was one factor that significantly influenced the lysimeter water balance during the course of this study (Table 5). Weather conditions during the first two months of the study (October and November) were typical of the fall season in Tucson. The expected El Nirio winter weather pattern developed in December bringing below normal temperatures and ETo, and rainfall that exceeded normal by approximately 50%. Above normal temperatures and dry weather returned in January, only to be replaced again by the cool, wet El Nirio weather pattern in February. February rainfall totaled 73% above normal while ETo ran 29% below normal. Wet weather continued through April, producing below normal temperatures and ETo. Rainfall in March and April exceeded the normal by 42% while ETo was reduced 11% relative to normal. The summer season and began cool and dry. Only 0.24 mm of precipitation was reported in May and June temperatures in both months averaged below normal. The summer monsoon season developed in July. July rainfall exceeded normal by 58% and the increased cloudiness continued in August; and humidity reduced July ETo 8% below normal. Monsoon flow however, August rainfall was 37% below normal and ETo totaled near normal. September 1998 concluded the period of study with slightly above normal precipitation and normal levels of Eto. 86 Table 5. Maximum (Tmax) and minimum temperatures (Tmin), precipitation (ppt.) and reference evapotranspiration (ETo) measured over the course of the study by the AZMET weather station located at the Karsten Desert Turfgrass Research Facility. Temperatures are reported as monthly means and precipitation and ETo are reported as totals. Departure from normal (+ / - Normal) compares monthly values to mean values for the period 19871997. Date Tmax °C 08/97 09/97 10/97 11/97 12/97 01/98 02/98 03/98 04/98 05/98 06/98 07/98 08/98 09/98 37 36 28 24 16 20 17 22 24 31 36 37 37 36 +/Normal Tmax °C 0 +1 -2 +1 -3 +1 -4 -1 -4 -2 -2 -1 0 +1 Tmin °C 23 21 11 4 1 2 3 6 7 12 16 22 23 19 +/Normal Tmin °C +2 +3 0 0 0 +1 +1 +2 0 0 -1 0 +2 +1 Rain mm 46.5 50.8 8.1 15.0 74.2 4.8 97.3 37.1 12.5 0.0 0.0 85.6 29.0 23.6 +/Normal Rain mm +0.5 +35.2 -10.7 -3.5 +38 -30.9 +71 +10.8 +4 -6.1 -0.8 +50 -17 +8.1 ETo mm 198.9 176.8 163.8 90.7 53.6 73.9 67.3 127.5 182.4 237.2 274.3 220.5 213.6 188.0 +/Normal ETo mm -4.3 -11.2 +16.5 -0.7 -7.4 +2.8 -26.7 -17.3 -20.8 -6.6 +5.1 -18.3 +10.4 -1.0 87 Clearly, the most striking meteorological feature of the year of study was the above normal winter rainfall caused by the El Nirio weather pattern. This wet winter was the main reason twelve-month rainfall totals exceeded normal by —30% and annual ETo totaled 3% below normal. Temperatures followed a pattern typical of wet years with maximum temperatures running slightly below normal, but minimum temperatures running above normal. Irrigation is an important component of the water balance of any irrigated turf system. An important objective of this study was to apply the same amount of water to each lysimeter, and Table 4 reveals this goal was accomplished. Differences in applied irrigation between lysimeters varied by less than 4% in all months except for December 1997, and March and June 1998. In December 1997, the east lysimeter received 18% more water from irrigation than the west lysimeter. This 18% difference actually translates to <3 mm in irrigation water due to the very wet December weather. The difference in irrigation in March 1998 can be attributed in part to a leaking sprinkler which reduced the precipitation rate of the west lysimeter irrigation system by about 10% for the month. The difference in June irrigation was due to a malfunction in the potable pump system supplying water to the east lysimeter, resulting in two days where no irrigation was applied to the east lysimeter. The overall pattern of irrigation followed a typical seasonal pattern with the lowest amounts applied in the months of December through February, and the highest amounts applied during the spring and summer months. Winter irrigations were reduced 88 due to above normal precipitation. Overall, the total amount of irrigation applied between 1 Oct. 1997 and 30 Sept. 1998 was 1,294 and 1,297 mm for the west and east lysimeters, respectively. Precipitation was measured by monitoring mass changes of the lysimeter during precipitation events. Rainfall measured by the lysimeters was always slightly higher (-15%) than that measured by the AZMET weather station (see Appendix D). The lysimeters provide a more accurate estimate of precipitation because (1) the lysimeter mass gain is a direct measurement of precipitation, (2) the surface area of the lysimeter is far greater than the limited collection area of the raingage, and (3) the raingage is mounted 3 m above the ground and is prone to loss of precipitation due to wind turbulence near the gage. As expected, rainfall events recorded from the mass gain by each lysimeter were very similar. The difference in precipitation between the east and west lysimeters was always less than 1 mm and usually less than 0.5 mm. Total rainfall measured by the west and east lysimeters was 435 mm and 431 mm, respectively. Periods of high rainfall occurred during the winter months of December, February and March. These three months contributed over 50% of the precipitation for the 12 months ending 30 September 1998. However, monsoon rains produced the highest monthly precipitation total - - —99 mm in July 1998. Low amounts of precipitation occurred during the summer months of May and June, when rainfall during the twomonth period totaled 0.2 mm, and in January when precipitation totaled —6 mm. 89 Turf ET followed a typical seasonal pattern with minimal ET during the winter months and peak ET during the summer months. Comparatively, ET from the east and west lysimeters varied slightly over the course of the study. Total annual ET was 1466 mm and 1419 mm on the west and east lysimeters, respectively. While it is difficult to assess statistically whether the 3% greater ET observed on the west lysimeter is real, there are some compelling reasons for believing the difference is significant. Perhaps the most important factor supporting higher ET on the west lysimeter is the fact that turf growth rates were higher on the west lysimeter. This was most evident during the summer months of May and June when ET from the west lysimeter exceeded that of the east lysimeter by —10%. During these months turf growth rates on the west lysimeter exceeded those of the east lysimeter by 26%. Figure 8 depicts the percent difference in monthly lysimeter ET (west vs. east) as a function of the percent differences in turf growth rate. It is evident in Figure 8 that higher growth rates are related to higher turf ET. The slope of the regression line in Figure 8 was subjected to a statistical analysis which tested the null hypothesis that the slope did not differ from zero. The null hypothesis was rejected at p < 0.05 adding further support to the fact that higher ET rates are related to higher rates of biomass for accumulation. Higher ET and growth rates have been reported on the west lysimeter associated with the the past four years. Brown et al. (1998) suggested improved nutrition Several use of effluent as a water source was the cause of higher growth rates and ET. 90 14 rn cvl 111 12 • y = 0.24x + 0.30 42 0.) .,> r 2 = 0.71 10 ra Tu iI r) Ec —1 VI 0 0 . _. 0 LL.1 e ._ 0 a, an =na c c) - cu u ,.. a) -30 -20 -10o 10 20 30 • • o_ Percent Change in Growth Rate—West Lysimeter Relative to East Lysimeter Figure 8. Percent difference in monthly turf ET between the west and east lysimeters plotted as a function of the percent difference in monthly turf growth rate between the west and east lysimeters. 40 50 91 other studies have observed similar relationships between increased turfgrass growth rates and ET (Krogman 1967; Feldhake et al. 1983, 1984). A second possible reason for higher ET in the west lysimeter may relate to the difference in thatch in the two lysimeters. Higher turf growth rates have generated more thatch in the west lysimeter. Visual observations indicated the heavier thatch retarded infiltration of irrigation water, thus leaving more water at/near the surface where it would more readily evaporate. The seasonal and/or annual ET totals measured during this study were slightly less than totals obtained from previous Arizona studies. Kneebone and Pepper (1982) found the mean annual consumptive use of bermudagrass overseeded with ryegrass was 1,654 mm when the grass was well watered and fertilized for optimal performance. Kopec et al. (1991) determined the annual consumptive use of bermudagrass overseeded with ryegrass to be 1,525 mm. Erie et al., (1981) determined the consumptive use for bermudagrass during the summer months was 1,005 mm. Bermudagrass ET obtained from this study ranged from 868 mm (east lysimeter) to 906 mm (west lysimeter). The lower ET totals observed during this study resulted in part from the wet year which lowered overall evaporative demand. Other likely causes for lower ET observed from this study were the lower mowing heights employed in this study and differences in irrigation procedure. Lower mowing heights often result in lower water use rates (Shearman and Beard, 1973; Biran et al., 1981; and Johns et al., 1983), while different 92 irrigation procedures can alter the soil evaporation component of ET (Kneebone and Pepper, 1984). Drainage during the period of study varied in response to total water inputs as expected. Drainage decreased during the fall period (October - November) due to minimal rainfall, then increased to higher levels during the winter months (January April) as water inputs greatly exceeded ET. Lack of rainfall in May and June significantly reduced drainage in June. Monsoon rains in July and August resulted in a subsequent increase in east lysimeter drainage during July and west lysimeter drainage in July and August. For the year ending 30 September 1998, drainage from the east lysimeter totaled 421 mm or 26% more than the 311 mm of drainage obtained from the west lysimeter. The large difference in drainage results from two factors: (1) a higher drainage rate on the east lysimeter due to residual effects of the previous study and (2) a lower turf ET rate on the east lysimeter. Approximately 56% of the difference in total annual drainage occurred during the months of October and November of 1997 when east lysimeter drainage far exceeded that of the west lysimeter. Subsequent lysimeter analyses of soil water measurements revealed the water content of the east lysimeter was slightly higher than the west lysimeter at the onset of the study and thus was the reason for higher levels of drainage. The higher soil moisture of the east lysimeter early in the study is believed to be a residual effect of the previous study where irrigation rates were not controlled with the same accuracy as achieved in this study. The remaining 46% of the difference in 93 drainage was due to lower ET on the east lysimeter. Similar amounts of water were applied to both lysimeters during this study; however, annual ET on the east lysimeter was 47 mm below ET on the west lysimeter. The soil water balance equation indicates that such a scenario should lead to the increased level of drainage observed in the east lysimeter. The monthly change in soil water storage for both lysimeters (Table 4) reflects the difference between the inputs (irrigation + rain) and outputs (ET + drainage) and thus fluctuates between positive and negative values. Soil water storage declines during periods with limited rainfall (e.g. May and June 1998) and rises during periods with significant rainfall (e.g. February 1998). For the year ending 30 September 1998 the change in soil water storage was -48 and -111 mm for the west and east lysimeters, respectively. Negative storage values for the year indicate lysimeter soil moisture was still adjusting to the new irrigation regime imposed at the start of this study. Assuming a lysimeter soil depth of 3750 mm, the net annual decrease in soil water storage of -48 and -111 mm equate to an average decline of Ov of 1.3% and 2.9% for the west and east lysimeters, respectively. It is unlikely the higher soil moisture in the east lysimeter impacted the ET and performance of the turf because soil moisture in the root zone should have adjusted quite quickly. However, the residual effects of the previous study's irrigation regime clearly impacted drainage and leaching characteristics. For the year ending 30 September 1998 both lysimeters received — 1,730 mm (68 in.) of water comprised of 1,300 mm (51 in.) of irrigation water and — 430 mm (17 in.) 94 of precipitation (Table 4). According to historical data collected by AZMET over the past twelve years (Table 5), normal precipitation for the study site averages 274 mm (10.8 in.) annually. Thus, precipitation during the period from October 1997 to September 1998 exceeded normal rainfall by 156 mm (6.1 in.). Reference evapotranspiration totaled 1,893 mm (74.5 in.) during the period of study, which is 63 mm (2.5 in.) below historical ETo data derived at the KDTRF. Actual turf ET totaled 1,466 mm (1.46 ha-m) for the west lysimeter and 1,419 mm (1.42 ha-m) for the east lysimeter during the period of study (Table 4). These ET values are slightly lower than normal, which can be attributed to the excess precipitation and the increased cloud cover encountered during this study. Drainage totaled 311 mm and 421 mm for the west and east lysimeters, respectively. These drainage values were quite high due to (1) residual soil water draining from the previous irrigation regime, and (2) excessive rainfall that percolated below the turf rooting zone. Clearly, the drainage of residual water from the previous study overstates the drainage one would expect from the irrigation regime imposed by this study. If one assumes the change in soil water storage represents this residual drainage, then a good estimate of true drainage under this irrigation regime is obtained by adding the change in soil water storage to the drainage measured in this study. Such a computation provides drainage estimates of 263 min and 311 mm for the west and east lysimeters, respectively. The overall water balance shows that during a year where precipitation exceeded normal by 156 mm, total water inputs from irrigation and rainfall exceeded turf water use by more than 250 mm. The amount of irrigation water applied totaled —1295 mm which 95 is —8% below the water duty imposed by ADWR. Given the level of drainage observed in this study, one can conclude the ADWR water duty was sufficient to provide for turf ET and leach salts from the rooting zone during this wetter than normal twelve-month period. However, it is important to recognize that total irrigation applied represents 92% of the ADWR water duty which leaves only 8% of the water allocation for use in general turf maintenance, or to offset losses due to irrigation and plumbing inefficiencies. Salt Balance One objective of this study was to evaluate the potential for soil salinization when irrigating at or below the ADWR water duty. Table 6 presents an assessment of the salt balance in each lysimeter for the year ending 30 Sept. 1998. This period was chosen in order to best represent the irrigation regime employed in this study. Important aspects to consider when determining a salt balance are (1) the electrical conductivity of the applied irrigation water (2) the total amount of water supplied in irrigation and precipitation, and (3) the total amount and electrical conductivity of the drainage water. From these data, the amount of salt applied through irrigation and the total amount of salt removed by drainage can be estimated, and the leaching fraction calculated. It is evident in Table 6 that both lysimeters received similar amounts of irrigation and rainfall. However, the amount of salt applied to the west lysimeter (4,054 g) was approximately four times higher than that applied to the east lysimeter (1,017 g) due to the different EC of the irrigation water. The EC of the irrigation water was 1.0 dS/m and 0.25 dS/m for the west and east lysimeters, respectively. However, if rainfall is included and assumed to have an 96 Table 6. Total salt inputs (through irrigation) in grams of salt applied, total salt outputs removed through drainage over the course of the study, calculated from October 1997 to September 1998. Values indicate totals for the study period. Total water applied indicates the summation of rainfall and irrigation. The leaching fraction indicates the total drainage divided by the total water applied. Grams of Salt Applied East West 1,017 4,054 Grams of Salt Drained East West 1,870 871 Total Water Applied (mm) West East 1,729 1,729 Drainage Water (mm) West East 311 421 Leaching Fraction West East 0.24 0.18 97 EC of 0.0 dS/m, the effective EC of the water received by the lysimeters was 0.75 dS/m and 0.18 dS/m for the west and east lysimeters, respectively. High amounts of drainage were obtained from both lysimeters with the east lysimeter drainage (421 mm) exceeding that from the west lysimeter (311 mm) by 26%. Annual leaching fractions, computed by dividing drainage by total water inputs, also were quite high at 0.24 and 0.18 for the east and west lysimeters, respectively. The discussion of lysimeter water balances clearly showed that residual effects from the previous study increased drainage in this study above what could be expected from the irrigation and precipitation regime encountered between 1 Oct. 1997 and 30 Sept. 1998. When drainage was adjusted to remove the effects of the previous study, the annual totals were 311 mm and 268 mm for the east and west lysimeters, respectively. These adjusted totals suggest more realistic leaching fractions for the study were 0.18 for the east lysimeter and 0.16 for the west lysimeter. The high leaching fractions were due to a combination of factors including (1) water inputs from irrigation and precipitation were substantially in excess of turf ET, and (2) high hydraulic conductivity of the sandy lysimeter soil which supported rapid downward movement of excess soil water that accumulated during precipitation events. The salt balance of both lysimeters was positive for the period of study. Inputs of salt to the east lysimeter totaled 1,017 g while drainage water removed only 871 g. The salt balance of the west lysimeter was even more positive with salt inputs totaling 4,054 g and salt removal accumulating to 1,870 g. The accumulation of salt in the east lysimeter 98 is minimal and reflects both the low salt content of the irrigation water and the high level of leaching. The accumulation of salt in the west lysimeter seems quite high considering the significant leaching fraction established. However, this high rate of salt accumulation can partly be explained by the change in soil moisture regimes that occurred during the transition from the previous study to the current study. In the previous study, excessive irrigation rates provided for leaching fractions near 0.50, whereas the leaching fractions for this study were reduced by more than half to — 0.18. The change in leaching fractions dictated by this study has created a higher potential equilibrium status with respect to soil salinity and thus explains the accumulation of salts in both lysimeters. Dean et al. (1996) observed that 18 months were required after a change in irrigation before soil salt profiles reached steady state. One might expect the time required to reach a steady state salt balance to be even longer in the 4 m deep lysimeters. Implications of the transition to a new soil water equilibrium status are (1) the ratio of the salt applied to the amount of salt drained will decrease, and (2) the EC of the drainage water will rise over time until the new equilibrium is established, then the salinity of the drainage water will remain fairly constant. The trend toward increasing EC in the drainage water is clearly shown in Figure 9, where the drainage water EC is plotted over the course of this study for both lysimeters. Since the onset of the study, drainage water EC has increased by 40% and 18% in the west and east lysimeters, respectively. Ayers and Westcott (1989) provide procedures for estimating the soil water EC for various leaching fractions and irrigation 99 Drainage Water EC over Time East & West Lysimeter 2.4 . A ..1 alb& A 1.9 I ll 0.9 l ..; 0.4 r- " % ‘.•.„, A 1 MA vw Y7 ,• • ‘.."•.. .... ..... „..•, • ••••".••. ••• • . nsA v ••.., •• n ' ii ii II I} i IIIIIIMMIIIIIIMM MIIIIIiIIMM HMI ill to N CO CO CO 0) 0) to to (;) co c6 2 < 6 e, ON WEST 4 EAST Figure 9. Electrical conductivity (EC) of the drainage water obtained from the east and west lysimeters over the course of the study. 100 water salinities. They indicate drainage water EC should approach 5.6 dS/m in the west lysimeter when the ECiw = 1.0 dS/m and the leaching fraction is 18%, and 1.0 dS/m in the east lysimeter when the ECiw = 0.25 dS/m and the leaching fraction is 22%. As previously indicated, rainfall effectively reduced the EC of the irrigation water by 25%. If precipitation levels that occurred during this study were to continue, drainage water in the lysimeters should continue to increase to levels approximating 4.2 dS/m and 0.8 dS/m in the west and east lysimeters, respectively. The EC of the west lysimeter drainage water was 2.1 dS/m as of 30 Sept. 1998, suggesting soil profile adjustment to the new irrigation regime is not complete. In contrast, the EC of the east lysimeter drainage was 0.8 dS/m at the conclusion of the study - - a value that would imply a near equilibrium salt balance is already established in the east lysimeter. It is important to note than normal fluctuation in precipitation will eliminate the possibility of developing a true equilibrium in the lysimeters. For example, if precipitation levels were to decrease to normal and leaching fractions decline to about 0.10, the drainage water EC from the lysimeters should increase to levels closer to 10.0 dS/m for the west and 2.5 dS/m for the east lysimeter. Tables 7 and 8 and Figure 10 present a month-by-month assessment of inputs, outputs and the net balance of salts in each lysimeter for the period between October 1997 and September 1998. Both the tables and the Figure clearly show the overall salt balance differed significantly among the two lysimeters. The salt balance of the east lysimeter was negative during the fall and winter (October through March) because: (1) 101 Table 7. Month-by-month salt Balance over the course of the study for the east lysimeter. Monthly totals indicate either salt input through irrigation, salt output removed through drainage or the net balance between inputs and outputs. Salt Balance -- East Lysimeter Salt Input Salt Output Net Balance Month (+ or -) grams grams Aug-97 Sep-97 Okt-97 Nov-97 Des-97 Jan-98 Feb-98 Mar-98 Apr-98 Mei-98 Jun-98 Jul-98 Aug-98 Sep-98 138.3 75.1 93.2 59.7 9.1 40.6 5.6 65.9 79.5 143.3 181.9 114.3 115.4 108.1 329.3 155.7 93.7 102.2 44.9 65.2 44.5 67.6 70.5 78.3 44.7 79.0 115.6 64.4 -191.0 -80.6 -0.5 -42.4 -35.8 -24.6 -38.9 -1.8 9.0 65.0 137.3 35.3 -0.2 43.7 Table 8. Month-by-month salt Balance over the course of the study for the west lysimeter. Monthly totals indicate either salt input through irrigation, salt output removed through drainage or the net balance between inputs and outputs. Salt Balance -- West Lysimeter Salt Input Salt Output Net Balance Month (+ or -) grams grams Aug-97 Sep-97 Okt-97 Nov-97 Des-97 Jan-98 Feb-98 Mar-98 Apr-98 Mei-98 Jun-98 Jul-98 Aug-98 Sep-98 541.8 310.6 372.9 248.4 27.8 157.5 22.8 235.6 324.3 574.7 788.6 439.7 443.2 418.2 564.7 263.2 177.6 154.4 141.7 191.5 189.5 161.1 181.6 181.2 96.8 98.9 150.5 144.7 -22.8 47.4 195.3 94.0 -114.0 -34.0 -166.6 74.4 142.7 393.5 691.8 340.9 292.6 273.6 102 Salt Inputs (Irrigation) and Outputs (Drainage) for Both Lysimeters (3.> 1 A I 3 8 3.0 -3 u_a) 33 D. CO ,M1A../ g800 CV) L. cn ..S. 600 . g 400 1 -.. 13= 200 0 - OS --n—r 1 .17. -200 .. 1111 4 iiii ki.„ 4 .-400 0= c v1 Cl) ( -£300 El Salt Input(East) 0 Salt Input(West) • Salt Output(East) 0 Salt Output(East) Figure 10. Components of the salt balance for the east and west lysimeters during the period August 1997 to September 1998. Monthly inputs of salt from irrigation in grams are presented as positive values on the y-axis. Monthly outputs of salt from drainage in grams are presented as negative values on the y-axis. 103 water supply was well above ET values, producing increased drainage, and (2) significant rainfall reduced salt input as nearly salt-free rainfall replaced irrigation as the source of water. In contrast, the salt balance in the west lysimeter was positive from September through November and during March. The west lysimeter salt balance turned negative only during the winter months of December through February when salt inputs decreased due to low irrigation rates and drainage increased as a result of heavy precipitation. The normal combination of low precipitation and high evaporative demand resulted in high levels of irrigation in April, May and June 1998. Salt inputs increased dramatically and drainage slowed, producing a positive salt balance in both lysimeters. Arrival of the monsoon rains in July lowered irrigation rates due to rainfall and reduced evaporative demand. However, the salt balance for both lysimeters remained positive through September 1998, with the exception of the east lysimeter in August when inputs and outputs were essentially equal. While Tables 7 and 8 represent but one year of data, the results provide some clear guidance on when leaching is most likely to occur. The winter months are clearly a good time for leaching because ETa is low and rainfall is common. Natural leaching can be expected during those winter months when rainfall is above normal or excessive irrigation is applied. Another obvious time for leaching is during overseeding when increased irrigation is required to germinate the winter grass. It appears the increase in October irrigation that accompanied overseeding did create a leaching event in the east 104 lysimeter. However, there is no strong indication that irrigation during overseeding produced an increase in leaching from the west lysimeter. The most difficult time to leach is during the summer months when ETa is high. It is evident in Tables 7 and 8 that during extended dry periods such as in May and June, salts can be expected to accumulate due to high rates of irrigation needed to replace turf ET. However, the annual monsoon weather pattern in July and August typically brings increased atmospheric humidity and greater rainfall, thus providing potential leaching events. The amount of salt removed from both lysimeters increased during the summer months due to an increase in drainage. In most months, however, the lysimeter salt balances remained positive suggesting the summer rains did not purge salts from the lysimeters. One must be careful to not interpret the positive salt balances as indicating summer rains have no leaching potential. As indicated earlier, the salt balance of the lysimeters is in a readjustment phase. Once the adjustment is complete, summer rains will likely produce negative salt balances in the lysimeters. In situ measurements of soil salinity using time domain reflectometry provide evidence of summer leaching and are described below. Assessment of Salinity Profiles Time domain reflectometry provided an effective means for evaluating the changes in salinity throughout the soil profile over time. Figures 11 and 12 depict the salinity profiles at depths of 0.5, 1.0 and 2.0 m for the west lysimeter and 1.0 and 2.0 m for the east lysimeter during the course of the study. Figures 11 and 12 represent the bulk 6 105 • • • • 26-deS-L — 26 - deS - 01, —86-6my-Lz — 26 - 6nV - 90 • • — 96 - In 1-- CZ • • — 26 In (-so - • • — 26 - unr -17Z • • • • • • • • • • — 26 - un r - 7-4 • • • • • • • • — 26 - ! 9 W - LZ — 26-PlA,1-2 f — 261e1A1 - 1,0 t — 96 - KIV - L — — 26 - AV - E0 — 96 -JeVg - 0 Z • • • 26 - MN - 90 - 26 - CiaJ - OZ • - • • 90 H - lier - 60 L6-sea-2Z • • - 26 - lier - EZ • • 26 - Cia —L6-saa-zi. • L6 - A 0 N - 9Z — L6 - A 0 N -17 — L6 -1)10 - 1,2 • — L6 - 1)10 - L — L6 - 1310 - 20 • - L6 - deS - 61, L6 - deS - 90 L6 - 6nV - I,Z L6- 6 nv- z o L6 1 n r-vz N 6 o 6 Lr) 6 (0 C, ) 6 Cr) (y) O. C,)N. 6 6 w/SP e03 CO N LC) N o ô ô ô 6 6 6 6 • N 106 6-d9s-z[ 26-dos- O l. 26 -6 nV - LZ 26 -6 n V - 90 26 - In r - EZ 26 - In r - 80 86-unr-frz 86 - Uni - IL 26 - PIN - LZ 26 - ! 1 - S l• 86 - LAI - [0 26 -1 d V - L - 86 - AV - E0 26 - JelAI - 0Z — 26 - -leVg - 90 — 26 - [email protected] - 0Z — 26 - qad - 90 — 26 - mr - CZ - 86 - Ur - 6O L6 - seC1 - 2Z • — L6 - s 8 CI - Z L6 - An N - 9Z - L6 - A 0 N -171• — L6 - 4310 - 1.2 - L6 - 4)10 - L — L6 - 1,10 - 20 — L6 - daS - 6 - 2.6 - d 8 S - S0 — L6 -6 nV - LZ L6 -6 nV - LO • I op o) sC7I cn 0 N N ô ô D 6 o o ci 6 6 ci ci 6 6 6 6 6 ci ci N 0) • d- 00 0 o 00 •:1- CID WISP B03 L6-Inr-tz 107 soil salinity (ECa) at varying depths, where each depth represents the mean ECa derived from the three angular sampling positions on the lysimeters. As expected, the most discernible changes in ECa occurred above one meter where plant roots are actively extracting soil water. Fluctuations in salinity over time reflect dry periods requiring high levels of irrigation which result in increases in salinity (if not applied in excess of turf ET) and significant rainfall periods which result in leaching events (decreases in salinity). Comparatively, the most significant changes in ECa over time were observed in the west lysimeter where salt loading from irrigation was highest. Bulk soil salinity in both lysimeters followed a slight decreasing trend from August through October 1997, perhaps reflecting impacts of excessive irrigation from the previous system. As drainage decreased ECa increased in November and into the early part of December 1997. In the latter part of December, rainfall (76 mm) significantly decreased the ECa in both lysimeters. Minimal rainfall in January 1998 (only 5 mm) did not provide significant leaching, and ECa increased slightly in both lysimeters. However, excessive February rainfall provided substantial leaching and reduced ECa in both lysimeters. The next significant increase in ECa occurred in June on the west lysimeter, following two dry months where turf water requirements were met by irrigation. Bulk soil EC increased 53% at 0.5 m relative to values observed during the wet winter period of early March. The ECa data appear to reflect and/or confirm the findings presented earlier (Table 8) which show very positive monthly salt balances in May and June. A 108 similar rapid increase in ECa was not observed in the east lysimeter. The failure to observe such an increase in ECa likely reflects the lower net increase in salts in the east lysimeter and the fact that ECa measurements were not available in the east lysimeter at the 50 cm depth. Monsoon rainfall began on 5 July, and ECa at 50 cm in the west lysimeter declined precipitously as salts leached below the 0.5 m region. This salt laden pulse reached the 1.0 m depth about ten days later, producing a very substantial peak in ECa at 1.0 m in mid-July. As the July rains continued, the pulse of salts moving with the percolating water moved beyond the 1.0 m region and produced an increase in ECa at the 2.0 m depth in late July and early August. Although the monsoon season did provide substantial leaching, Figures 11 and 12 show that ECa in the west lysimeter did not decrease to levels comparable to those in the wet winter months. These results are in complete agreement with the salt balance data presented in Tables 7 and 8. While the monthly balances show an increase in the amount of salts leached in July and August, the net salt balance of the lysimeters remained positive throughout the summer. Turf Quality An important objective of this research was to determine whether irrigating at levels equal to or below the ADWR water duty would produce acceptable quality turf. On average, turfgrass quality was acceptable or better during most of the study (Figure 13). There were three periods where turf quality slipped below acceptable levels for the study period. The first period followed overseeding in October and was not unexpected. 109 Quality Ratings Over Time for the East and West Lysimeter 9 7 i n Accep. 4— 3 N- 1"-- 0)0) D > 0 0 (3? 0 (;) co 03 CO CO CO CO CO CO CO CO CO CO CO CO 03 CO 03 CO CO CO CO i a? Cri ai ai cl? ai ai a? ai a) CY) ai ai a a.' ci) Ci) CL CL C C5 .C1 -‘5 735 w w w " cs_ sa sa as °) z 0 —) 6 0!) 6 4 6 °? °? °? u_ 2 < < 2 (;-) N o West 4 < < 0 N (.6 6 0 N East Figure 13. NTEP quality ratings of west and east lysimeter turf over the course of the study. Rating of 6 indicates acceptable quality turf 110 A second period of decreased turf quality occurred in late November and December, when temperatures turned too cold to support adequate ryegrass growth. A nitrogen deficiency was also evident during this period as indicated by a slight yellowing of the older leaves on the turf plant. This period of poor turf quality lasted for nearly a month and was resolved when the lysimeters were re-seeded in mid-January 1998. The third period of significant turf quality reduction occurred during the transition from ryegrass to bermudagrass in early July. Summer transition to bennudagrass is often slow at the KDTRF due to cold night temperatures that develop as a result of cold air drainage from higher elevation areas surrounding the facility. Summer transition at KDTRF is often delayed until the nights warm in response to the monsoon. Mean turf quality ratings over the course of the study were 6.2 and 5.8 for the west and east lysimeters, respectively. The quality ratings of west lysimeter turf were significantly higher than quality ratings for the east lysimeter turf (p = 0.029), a finding consistent with previous studies. The improved quality of the west lysimeter turf is presumably due to improved plant nutrition that results from daily irrigation with effluent water. The quantity of water supplied to the turf did not impact turf quality. Figures 14 and 15 provide mean turf quality ratings and the amount of water supplied to the turf as a percentage of ETa for each month of the study. Turfgrass quality did not improve when a greater percentage of water was applied, suggesting water was not a limiting factor. This fact is clearly shown in Figures 16, where monthly quality ratings show no 111 Turfgrass Quality vs Supplied Water West Lysimeter —8 300 — —7 250 — 6 cTs - 200 — ô 150 — !NT 73 a 100 — l =CL 7 • (/) —2 50 — o 0 I Aug- Sep- Old- Nov- Des- Jan- Feb- Mar- Ap - Mel- Jun- Jul-98 Aug- Sep98 98 98 98 98 98 98 97 98 97 97 97 97 — --M—Quality = Supplied Water Figure 14. Turfgrass quality and the the total amount of water supplied to west lysimeter turf by month for the period August 1997- September 1998. Turfgrass Quality vs Supplied Water East Lysimeter —8 300 — —7 250 — -6 200 — -5 —4 150 — —3 100 — —2 50 — o —1 • Aug- Sep- Old- Nov- Des- Jan- Feb- Mar- Apr- Mel- Jun- Jul- Aug- Sep98 98 97 97 97 97 97 98 98 98 98 98 98 98 C=I Supplied Water 4-- Quality — Figure 15. Turfgrass quality and the the total amount of water supplied to east lysimeter turf by month for the period August 1997- September 1998. 0 112 Turfgrass Quality vs. Supplied Water 8— 7— * 6— • • 5— • 4— 3— 2— y = -0.09x + 6.13 r2 = 0.00 1— 0 0 50 100 150 200 250 Water Supplied (%ETa) Figure 16. Monthly mean turfgrass quality ratings from both lysimeters plotted as a function of water supplied to the turf. Water supplied is presented as a percentage of ETa. The line depicted represents the least squares regression line. 300 350 113 distinguishable relationship with the level of supplied water. It is important to realize that the level of water supplied to the turf was not below the expected turf ET in most months, thus water stress induced reductions in quality should not have been expected. Under the objectives of this study, it was important to determine whether bulk soil salinity (ECa) levels would hinder turfgrass quality. In order to investigate this objective, turfgrass quality was regressed against ECa for the east and west lysimeters (Figures 17 and 18). Bulk soil EC data in Figures 17 and 18 are from the 0.5 m depth in the west lysimeter and the 1.0 m depth in the east lysimeter. These depths were chosen in order to best represent ECa near the turf root zone. Figures 17 and 18 suggest that no discernible relationship existed between turf quality and ECa. The least squares regression lines presented in Figures 17 and 18 were subjected to a statistical analysis which tested the null hypothesis that the slopes did not differ from zero. In both cases, the null hypothesis was not rejected, indicating the regression line slopes did not differ from zero. Thus it was concluded that ECa did not accumulate to levels capable of hindering turfgrass quality. The lack of relationship between turf quality and ECa was not surprising. Water inputs were typically in excess of ETa during the course of the study, which helped minimize the buildup of salt in the root zone. The EC of the lysimeter drainage water also suggests there should be no relationship between ECa and turf quality. Drainage EC was well below levels deemed dangerous to turfgrass performance (Ayers and Wescott, 1989). 114 Turfgrass Quality vs. Bulk Soil Salinity West Lysimeter (0.5 m depth) 8— • • •• y = -25.89x + 6.74 • •• 12 = 0.02 7 -6— • •• • • 5 — 4 — 3 0.000 0.005 0.025 0.020 0.015 0.010 0.030 Bulk Soil EC dS/m Figure 17. Mean monthly turfgrass quality ratings plotted as a function of bulk soil electrical conductivity obtained at the 0.5 m depth in the west lysimeter. The line depicted is the least squares regression line. 8 Turfgrass Quality vs. Bulk Soil Salinity East Lysimeter (1.0 m depth) - 7— 0 y = 98.06x + 4.14 "'" " 12=0.04 6 . 47 0 0 0. 5 — 4 — 3 0 0.005 0.01 0.015 0.02 Bulk Soil EC dS/m Figure 18. Mean monthly turfgrass quality ratings plotted as a function of bulk soil electrical conductivity obtained at the 1.0 m depth in the east lysimeter. The line depicted is the least squares regression line. 0.025 115 Biomass Accumulation and Turf Growth Rates Biomass accumulation and growth rate provide a second means of assessing turfgrass performance under the irrigation regime employed in this study. Biomass accumulation, as determined by collecting, drying and weighing turf clippings after each mowing event is presented in Figure 19. The two large peaks in biomass accumulation occurred during the de-thatching and scalping process required prior to the fall and winter overseed events. Figure 20 depicts the mean monthly turf growth rates obtained from both lysimeters over the course of the study. Biomass collected during overseed preparation was removed from the data presented in Figure 20. Figure 20 indicates bermudagrass growth rate declined in September 1997 prior to overseeding in October. In October and November, growth rate increased significantly due to ryegrass germination and growth. Turf growth rate and stand density then decreased considerably in December due primarily to cold temperatures. Reseeding of the lysimeters in midJanuary resulted in improved growth rates in late January and in February. Growth rates continued to increase into March as temperatures were favorable for ryegrass growth. In April, growth rates decreased slightly, but rebounded in May to levels similar to that of March. The slow transition from ryegrass to bermudagrass caused growth rates to decrease in June. Similar to previous years, growth rates increased in July and August during the annual monsoon weather pattern, and then declined as temperatures cooled in September. • • 2 116 Collected Biomass Over Time 1600 Fall Overseed 1400 — 1200 — 1000 — S 800 • 600 — • • cl;) a' ) OD (3) OD CO Cr.) al ) Q oo cr? 3 — cr) 0 6 z 6 -- o - 3 (44 17N OD II) 7 0 Clipping Weights East —Clipping Weights West Figure 19. Turf biomass collected from the west and east lysimeters over the course of the study. Biomass is presented in grams for both lysimeters. Turfgrass Growth Rate East and West Lysimeters 6— Aug- Sep- Old- Nov- Des- Jan- Feb- Mar- Apr Mei- 97 97 97 97 97 98 • Growth Rate-West 98 98 98 98 98 98 D Growth Rate-East Figure 20. Monthly growth rates in g/m2 /day for turf grown on the east and west lysimeters between August 1997 and September 1998. 98 117 Comparatively, the turf on the west lysimeter had a higher growth rate than the east lysimeter turf over the course of the study. Turf on east lysimeter grew at a higher rate only during the months of January and August 1998. The total biomass accumulation during the period (October 1997-September 1998) was 5,498 g and 6,314 g for the east and west lysimeters, respectively. Average turf growth rates for the east and west lysimeters were 1.87 g/m2/day and 2.22 g/m 2 /day, respectively over the course of the study. Turf growth rates were compared with the supply of water provided to the turf as a means of assessing whether water was a factor impacting growth (Figure 21). One would expect that low amounts of water supplied to the turf would result in decreased growth rates. However, the turf received water in excess of its ET requirement in every month during the study except April, thus water stress induced reductions in turfgrass growth rates were not expected. Turfgrass growth rates were lowest during the months September of 1997, December and January. Figure 21 suggests that growth rates in September were low due to the low amount of water applied. However, September is typically a month where temperatures decrease and bermudagrass growth slows. Ryegrass growth rate was low in December 1997 and January 1998 due to a weakening turf stand and cooler temperatures. If water stress was limiting growth one would expect ample or even excessive water applications to produce the highest growth rates. This phenomenon, termed luxury rates water use, was shown by Kneebone and Pepper (1984). Periods of high growth 118 Turfgrass Growth Rate vs Supplied Water – 350 6.00 – Aug- Sep- Okt- Nov- Des- Jan- Feb- Mar- Ap - Mel- Jun- Jul Aug- Sep97 97 97 97 97 98 98 98 98 98 98 98 98 98 Growth Rate-West —•—WEST Supplied Water Growth Rate-East - — EAST Supplied Water Figure 21. Turfgrass growth rates and the total amount of water supplied to both lysimeters by month for the period August 1997September 1998. 119 occurred in November 1997 and March, May, July and August 1998. However, the water supplied during these months was not comparatively higher than other months (Figure 21). An increase in water application did prove beneficial during overseeding events. The relationship between growth rate and water supply was further evaluated by comparing growth rate against the level of supplied water reported as a percentage of ETa (Figure 22). The least squares regression line presented in Figure 22 implies a reduction in turf growth rate with higher rates of supplied water. However, a statistical evaluation of the regression line determined that the slope was not significantly different from to zero (the null hypothesis was rejected), indicating this negative relationship between growth and water supply is likely not real. Regression analysis was used to determine if growth was related to ECa. Monthly turf growth rates were regressed on ECa (Figures 23 and 24). Bulk EC data for the lysimeters were again taken from the 0.5 m depth in the west lysimeter and the 1.0 m depth in the east lysimeter. Results from these analyses reveal no relationship between growth and ECa. Statistical assessment of the slopes of the regression lines revealed neither slope was significantly different from zero. It is not surprising that growth rate appears unrelated to ECa. Bulk soil EC and soil solution EC peaked in early summer at 0.03 dS/m and 3 dS/m, respectively. The soil solution EC would convert to an extract ECe value of — 1.5 dS/m. Ayers and Westcott (1989) indicate ECe values of 6.9 dS/m are required to reduce bermudagrass yield. 120 Turfgrass Growth Rate vs Supplied Water 6y = -0.75x + 3.04 r2 = 0.09 5- en 4- - 3 - m 2 - t 1- m 181 It II 1 - Ei 0 0 o 50 100 150 200 250 300 Supplied Water (% of ETa) Figure 22. Monthly mean turfgrass growth rates from both lysimeters plotted as a function of water supplied to the turf. Water supplied is presented as a percentage of ETa. The line depicted represents the least squares regression line. 350 121 Turfgrass Growth Rate vs. Bulk Soil Salinity West Lysimeter (0.5 m Depth) 6— 5— • y = 17.16x + 1.86 r = 0.00 2 4— •• • • 3— 2 — • • 1 — • 0 0.000 • 0.005 0.025 0.020 0.015 0.010 0.030 Bulk Soil EC dS/m Figure 23. Mean monthly turfgrass growth rates plotted as a function of bulk soil electrical conductivity obtained at the 0.5 m depth in the west lysimeter. The line depicted is the least squares regression line. Turfgrass Growth Rate vs. Bulk Soil Salinity East Lysimeter (1.0 m Depth) 6— ro • "E "i3) 5— y = -265.49x + 6.43 12 = 0.11 cu 4 — rts 2 •m u' :th 3 - 2 — •%<<to • 1— • • o 0 0.005 0.01 0.015 0.02 Bulk Soil EC dS/m Figure 24. Mean monthly turfgrass growth rates plotted as a function of bulk soil electrical conductivity obtained at the 1.0 m depth in the east lysimeter. The line depicted is the least squares regression line. 0.025 122 Harivandi et al., (1992) lists bermudagrass as tolerant to salinity stress (Table 1), indicating that bermudagrass is difficult to establish or maintain at ECe values > 10dS/m. Soil salinity during this study did not reach these levels, thus bermudagrass yield did not decline due to salinity stress. Turf growth rates were compared with those obtained by Brown et al. (1998) to further evaluate whether irrigating within the limits of the ADWR water duty would hinder turf growth. Growth rates reported by Brown et al. (1998) were for turf irrigated well above turf ET and the ADWR water duty. Growth rates observed during the current study exceeded those from the previous study during vigorous growth periods except in the month of April (Figures 25 and 26). The lower growth rates observed during the month of April for this study could be attributed to abnormally cool temperatures. Mean temperature in April 1998 was 16°C. Comparatively, mean temperature was 4°C higher during the previous study (April 1996). The higher growth rates during this study indicate the irrigation regime employed in this study coupled with above normal rainfall did not hinder turfgrass growth. Improved nitrogen fertility and better cultural management of the turf are possible reasons for the improved turf performance in this study. Crop Coefficients Turfgrass crop coefficients (Kcs) are presented in Table 9 for each month of the study. Crop coefficients ranged from a low of 0.63 in December 1997 to a high of 0.83 in 123 Growth Rate--Previous Study vs Current Study West Lysimeter 3.5 0.5 Bermudagrass • Previous Study < Mar D Current Study Figure 25. Comparison of the turf growth rates obtained from the previous study (1996) with those obtained in the current study (1998) for the west lysimeter. Growth Rate--Previous Study vs Current Study East Lysimeter 3.00 1.00 0.50 Jul Bermudagrass • Previous Study CI Current Study Figure 26. Comparison of the turf growth rates obtained from the previous study (1996) with those obtained in the current study (1998) for the east lysimeter. 124 Table 9. Monthly crop coefficients computed using ETa data obtained from the east and west lysimeters, and reference ET data provided by the Arizona Meteorological Network. MONTHLY CROP COEFFICIENTS ARIZONA PENMAN EQUATION DATE EAST LYS. WEST LYS. Aug-97 0.80 0.83 Sep-97 0.78 0.81 0.71 Oct-97 0.68 0.79 Nov-97 0.78 0.66 0.63 Dec-97 0.66 0.68 Jan-98 0.80 0.83 Feb-98 0.77 0.75 Mar-98 0.75 0.72 Apr-98 0.83 0.77 May-98 0.80 0.70 Jun-98 0.79 0.79 Jul-98 0.82 0.83 Aug-98 0.75 0.74 Sep-98 125 August 1997 and May 1998. The low Kcs obtained in December and January coincide with the period of declining turf quality and thus may underestimate the Kcs appropriate for higher quality turf. It is important to note, however, that low Kcs were observed during these same months in the previous study. Brown et al. (1998) reported December Kcs ranged from 0.60 to 0.71 over a three-year period. It therefore appears factors other than low turf quality can reduce Kcs at this time of year. Other possible reasons for low winter Kcs may be turf response to frost and/or low temperature conditions, or a weakness in the Penman estimates of ETo at that time of year. Seasonal Kc values were developed by averaging the monthly Kcs for the respective winter and summer periods. Winter Kcs were 0.73 and 0.74 for the east and west lysimeters (Figures 27 and 28), respectively and compared favorably with the winter Kc value of 0.72 obtained by Brown et al. (1998). Summer Kcs averaged 0.77 for the east lysimeter (Figure 29) and 0.80 for the west lysimeter (Figure 30), again supporting previous research that determined the summer Kc to be 0.77. Overall, Kcs obtained in this study were slightly higher than Kcs obtained in the previous study (Brown et al. 1998). Whether the higher Kcs are truly different in a statistical or practical sense is debatable; however, both weather and higher turf growth rates resulting from improved nitrogen nutrition could produce an increase in Kcs. The relationship between higher turf growth rates and higher ET was described in an earlier section of this chapter. Turf growth rates in this study were higher than those reported in the previous study, and may be the cause of the higher Kcs. Brown et al. (1998) also 126 WINTER CROP COEFFICIENT: EAST LYSIMETER ARI7ONA PENMAN FOUATION 1.00 1!!!!!!!!!!!"! 0.80 0.73 3 0.60 u. tu 0 c.) 0. 0.40 201n., O ce 0.20 0.00 Nov-97 Des-97 Jan-98 Feb-98 Mar-98 Apr-98 TIME PERIOD 1997 1998 - Figure 27. Mean monthly crop coefficients (Kcs) for overseeded intermediate ryegrass grown on the east lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow. WINTER CROP COEFFICIENT: WEST LYSIMETER ARIZONA PENMAN EQUATION 1.00 0.80 2 u_ 77.!!!!!!!Zr 0.74 •-nn11•1=1. 0.60 u_ Ui O C.) a. 0.40 0.20 1n11n•=1, . j11n1•n•n• 0.00 Nov-97 Des-97 Jan-98 Feb-98 Mar-98 TIME PERIOD 1997 1998 - Figure 28. Mean monthly crop coefficients (Kcs) for overseeded intermediate ryegrass grown on the west lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow. Apr-98 127 SUMMER CROP COEFFICIENT: EAST LYSIMETER ARIZONA PENMAN EQUATION 1.00 0.80 —7 ••••••••••••• 0.77 0.60 0.40 0.20 1 0.00 Mei-98 Jun-98 Jul-98 Aug-98 Sep-98 TIME PERIOD 1997-1998 Figure 29. Mean monthly crop coefficients (Kcs) for bermudagrass grown on the east lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow. SUMMER CROP COEFFICIENT: WEST LYSIMETER ARIZONA PENMAN EQUATION 1.00 i— z -10• 0.80 ei E, 0.60 u_ - 0 0.40 O re o 0.20 0.00 Mei-98 Jun-98 Jul-98 Aug-98 TIME PERIOD 1997-1998 Figure 30. Mean monthly crop coefficients (Kcs) for bermudagrass grown on the west lysimeter. The seasonal mean Kc value is depicted by the horizontal arrow. Sep-98 0.80 128 found that wet weather conditions produced higher Kc values because the AZMET Penman Equation underestimates ETo during cloudy periods. This tendency was again observed in this study and is best examined in the wet month of February. Crop coefficients obtained in February were the highest of the winter season and approached the peak Kcs for the entire year which occurred in August 1998 - - another humid, cloudy month. Daily Kcs were examined to provide some assessment of the day-to-day variability in Kcs. Means and standard deviations of the daily Kcs were computed for each lysimeter and season, and are presented in Figures 31-34. It is clear from the standard deviations presented in Figures 31-34 that daily Kcs are more variable in winter than summer. Coefficients of variation (CVs) for monthly Kcs, computed by dividing Kc standard deviations by the mean monthly Kc values, clearly reveal the differences in winter and summer Kc variation (Figure 35). Winter Kc CVs averaged 0.21 and 0.19 for the east and west lysimeters, respectively, whereas summer Kc CVs averaged only 0.08 for the east and 0.09 for the west lysimeter. Some of this greater variation in daily Kcs is due to weaknesses in the Penman estimates of ETo during wet conditions. This weakness is evident during the wet month of February when Kc CVs reached their highest levels of similar increase the year at 0.33 and 0.26 for the east and west lysimeters, respectively. A in Kc CVs is evident in July when rainfall was well above normal. Lower day-to-day month variation in Kcs is evident during dry months. Crop coefficient CVs in May, a with no precipitation, averaged about 0.05. 129 Winter Crop Coefficient East Lysimeter 0.73 Feb-98 Mar-98 Apr-98 Figure 31. Mean daily crop coefficients (Kcs) by month for intermediate ryegrass grown on the east lysimeter. Error bars represent the standard deviation of the mean daily Kc. Winter Crop Coefficient West Lysimeter 0.74 Nov-97 Dec-97 Jan-98 Feb-98 Mar-98 Figure 32. Mean daily crop coefficients (Kcs) by month for intermediate ryegrass grown on the west lysimeter. Error bars represent the standard deviation of the mean daily Kc. Apr-98 130 Summer Crop Coefficient East Lysimeter 0.77 May-98 Jun-98 Jul-98 Aug-98 Sep-98 Figure 33. Mean daily crop coefficients (Kcs) by month for berrnudagrass grown on the east lysimeter. Error bars represent the standard deviation of the mean daily Kc. Summer Crop Coefficient West Lysimeter 0.80 May-98 Jun-98 Jul-98 by month for Figure 34. Mean daily crop coefficients (Kcs) bars bermudagrass grown on the west lysimeter. Error Kc. mean daily represent the standard deviation of the 131 Comparison of the Monthly Coefficient of Variation for Each Lysimeter 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 ClEast Lys. • West Lys. 0 East Lys. (no rain) 0 West Lys. (no rain) Figure 35. Coefficient of variation of mean daily crop coefficients computed using all days in a month and only dry days (no rain). 132 To further investigate the impact of rain on Kc stability, CVs were computed for monthly Kcs developed using only dry days (days with rain excluded). Figure 35 provides the comparison of the CVs obtained from all days with CVs computed using only dry days. The CVs based on dry days decline significantly in months with substantial rainfall (December, February, March and July). Regression analysis was used to further assess whether Kcs differed among lysimeters and seasons. Daily values of turf ET from each lysimeter were regressed on daily totals of ETo for the winter and summer periods. Figures 36 and 37 present the data and the resulting least squares regression lines for the winter months, and Figures 38 and 39 provide the same information for the summer months. The y-intercept was set equal to zero for all regression lines to facilitate assessment of line slopes (dETa/dETo) which provide another means of estimating seasonal Kc values. A good linear relationship was obtained between turf ET and ETo during the winter months (Figures 36 and 37). Winter Kcs based on regression line slopes were 0.72 and 0.74 for the east and west lysimeters, respectively. These values agree quite closely with seasonal Kcs developed from monthly totals of turf ET and ETo (Figures 27 and 28). The relationship between turf ET and ETo was a bit more scattered during the summer months as evidenced by the lower r2 associated with the least squares regression and lines (Figures 38 and 39). Summer Kcs based on regression line slopes were 0.76 developed from this 0.80 for the east and west lysimeters, respectively. Again, the Kcs 133 Lysimeter Turf ET vs Reference ET East Lysimeter -- Winter Regression (Nov-Apr) 7- o 7 1 8 Figure 36. Daily values of turf ET from the east lysimeter versus daily ETo for the period November 1997 through April 1998. The line depicted is the least squares regression line with intercept forced through zero. Lysimeter Turf ET vs Reference ET West Lysimeter -- Winter Regression (Nov-Apr) Reference ET (mm) Figure 37. Daily values of turf ET from the west lysimeter versus daily ETo for the period November 1997 through April 1998. The line depicted is the least squares regression line with intercept forced through zero. 9 10 134 Lysimeter Turf ET vs Reference ET East Lysimeter -- Summer Regression (May-Sept) 9.0 • 8.0 7.0 n•• 6.0 ••• 5.0 •n 4.0 3.0 2.0 1.0 n• 0.0 10.0 8.0 6.0 4.0 2.0 0.0 12.0 Reference ET (mm) Figure 38. Daily values of turf ET from the east lysimeter versus daily ETo for the period May through September 1998. The line depicted is the least squares regression line with intercept forced through zero. Lysimeter Turf ET vs Reference ET West Lysimeter -- Summer Regression (May-Sept) 9 8 7 6 5 4 3 2 1 0 o 2 4 6 8 10 Referecne ET (mm) Figure 39. Daily values of turf ET from the west lysimeter versus daily ETo for the period May through September 1998. The line depicted is the least squares regression line with intercept forced through zero. 12 135 regression analysis agree quite closely with Kcs developed from monthly totals of turf ET and ETo (Figures 29 and 30). Observed differences in Kcs among seasons and lysimeters were evaluated using the regression procedure described by Johnston (1984). The difference of 0.05 between summer (0.78) and winter Kc values (0.73) was found to be highly significant at p < .0003. However, assessment of the Kc difference of 0.017 between the west (0.75) and east lysimeters (0.73) was less clear cut. The Kc difference was significant only if p was greater than .012, suggesting the difference was only marginally significant. 136 CHAPTER 5 CONCLUSIONS This study was designed to address industry concerns regarding the adequacy of the present ADWR water duty to provide (1) sufficient water for culture of acceptable quality turf produced under high maintenance conditions in both wet and dry years and (2) adequate water to leach the necessary amount of salts below the turf root zone. Soil water and salinity balances, turf quality and growth rates, and turf water requirements were monitored in order to address the aforementioned industry concerns regarding the adequacy of the ADWR water duty. Water Balance Water balance data indicate that during a year where precipitation exceeded normal by 156 mm, total water inputs from irrigation and rainfall exceeded turf water use by more than 250 mm. Turf irrigated with effluent (west lysimeter) used 3% more water then turf irrigated with local groundwater (east lysimeter), presumably due to (1) higher turf growth rates resulting from improved plant nutrition and (2) greater thatch accumulation which retarded water infiltration, thus leaving more water in the thatch where it could more readily evaporate. The amount of irrigation water applied totaled —1295 mm which is —8% below the water duty imposed by ADWR. Given the level of drainage observed in this study, one can conclude the ADWR water duty provided sufficient water to satisfy turf ET and leach salts from the root zone during this wetter 137 than normal twelve-month period. However, it is important to recognize that total irrigation applied represents 92% of the ADWR water duty which leaves only 8% of the water allocation for use in general turf maintenance, or to offset losses due to irrigation and plumbing inefficiencies. Salt Balance Both lysimeters received similar amounts of irrigation and rainfall, however, salt input to the west lysimeter was about four times higher than that applied to the east lysimeter due to differences in total dissolved solids of the two irrigation waters. A year of excessive rainfall and residual effects from the previous study resulted in leaching fractions of 0.24 and 0.18 for the east and west lysimeters, respectively. Although leaching fractions were high, the salt balance of both lysimeters remained positive. The positive salt balances reflect the adjustment of soil salinity to a new equilibrium status caused by the current irrigation regime which applied less water to the lysimeters (relative to previous study). Drainage water electrical conductivity at the conclusion of the the study indicates salt accumulation in the west lysimeter is still increasing, however, EC of east lysimeter drainage water implies a near equilibrium salt balance is already established. Salinity profiles indicated salt accumulation is most likely to occur during summer months when turf water requirements are high. Leaching is most efficient during winter months when turf water requirements are low and rainfall is common. Assessment of the salinity profiles also indicate that late summer monsoon rainfall and overirrigation during establishment of overseeded winter turf provide potential leaching events. 138 Turfgrass Quality On average, turfgrass quality was acceptable or better during most of the study. Turf irrigated with effluent rated higher in quality than turf irrigated with groundwater; a result observed in previous studies. Improved plant nutrition resulting from daily irrigation with effluent water is the likely cause for this improvement in turf quality. Turf quality declined due to slow germination following overseeding, during periods with cold temperatures and during spring transition. The level of water supplied to the turf was not below the expected turf ET in most months, thus turf quality was not hindered due to lack of soil moisture. Bulk soil salinity did not accumulate to levels capable of reducing turf quality. Biomass Accumulation and Turfgrass Growth Rates Biomass accumulation and turf growth rates were monitored as a second means of assessing turf performance under the irrigation regime employed in this study. Growth rates were highest following overseeding, during spring when temperatures increased to levels favorable for ryegrass growth and during summer monsoon. Growth slowed during periods of freezing/cold temperatures and spring transition. Turf irrigated with effluent grew faster than turf irrigated with groundwater - - a finding that agrees with of ET in most previous research at the study site. Water was supplied to the turf in excess Growth rates months, thus growth rates did not decline due to water induced stress. obtained during this study were higher than those obtained in the previous lysimeter study 139 that utilized higher irrigation rates, indicating the irrigation regime employed in this study, coupled with above normal precipitation, did not hinder turfgrass growth. Bulk soil EC and soil solution EC did not accumulate to levels capable of jeopardizing turf growth rates. Crop Coefficients Crop coefficients (Kcs) averaged 0.73 and 0.74 for winter ryegrass irrigated using groundwater and effluent, respectively. Bermudagrass Kcs averaged 0.77 when irrigated with groundwater and 0.80 when irrigated with effluent. These Kcs compared favorably but were slightly higher than Kcs obtained in previous research at the study site. Possible reasons for the higher Kcs in this study are (1) turf growth rates were higher than those reported in the previous study and (2) wet weather conditions which cause the AZMET Penman Equation to underestimate ETo. Summer Kc values were statistically higher (+0.05) than average winter Kc values (p = .0003). Turf irrigated with effluent produced higher Kc values (+0.017) than turf irrigated with groundwater, however this relationship was considered only marginally significant. The fact that higher Kcs were observed with effluent irrigation suggests irrigation with effluent may result in higher turf water use superior rates than similar turf irrigated with potable water. The higher growth rates and turf quality (both of which were shown to produce higher water use rates) on the west effluent lysimeter strengthen the argument of higher water use from turf irrigated with water. 140 Although these results represent but one year of a planned three-year study, the findings show that the water duty set by ADWR provides sufficient water for production of acceptable turf and leaching of salt during a year of above normal precipitation. The remaining two years of the study will address the adequacy of the ADWR water duty during years with different and likely lower levels of precipitation. Scenario Testing Although findings from this study represent data collected during a year of excessive precipitation, one can use historical ETo and precipitation data to project the adequacy of the ADWR water duty in normal and dry years. Assuming an annual Kc of 0.74, one can estimate an ET value for bermudagrass turf overseeded with ryegrass in a normal and dry year. Normal annual ETo for the past twelve years as given by AZMET is 1956 mm (77 in.), and using an annual Kc of 0.74 gives an estimated annual turf ET value of 1448 mm (57 in.). The driest year in the AZMET record, 1989, produced an annual ETo value of 2184 mm (86 in.). Multiplying the 1989 ETo value by 0.74 provides a dry season estimation of turf ET of 1626 mm (64 in.). The ADWR water duty totals to —1400 mm, thus the ETa during a normal year and a dry year could exceed the water duty by —48 and —226 mm, respectively. Normal rainfall for the past twelve years as given by AZMET was 273 mm (10.7 in.), and for a dry year (1989), rainfall totaled 176 mm (6.9 in.). Therefore, during a normal year of precipitation, 18% of the average rainfall would be needed in addition to the ADWR water allotment to provide sufficient water to meet ETa requirements. Data from this 141 study indicated the turf utilized —33% of the total annual precipitation, thus the value of 18% does not seem unreasonable. During a dry year, however, ETa could exceed the ADWR water duty by —226 mm, which is higher than the total precipitation expected during a dry year (176 mm); thus, compliance with the ADWR water duty would result in deficit irrigation. These computations, though only approximations, show that irrigation within ADWR water duty is achievable under a normal year of precipitation and ETo. However, under a dry year, irrigation within ADWR guidelines will be a difficult task for turf managers. It should be strongly noted that these values are extrapolations based on one year of data, thus these values are not based on actual data and should not be construed thereof. It is also important to note that the water use requirements presented above were computed based on the assumption that 100% of the water applied will reach the turf. The water duty allocated by ADWR represents the amount of water recorded by the pumping station, not the amount of water the turf receives. Some loss of water is unavoidable during transit from the pumping station to the turf due to inefficiencies inherent in all irrigation systems (e.g. leaks, evaporation). From the perspective of the turf facility, these inefficiencies represent a loss of water for irrigation. Thus, turf facility compliance with ADWR water duties may relate to facility's ability to minimize these system inefficiencies. 142 APPENDIX A IRRIGATION WATER ANALYSIS The table below provides the components of the water analysis for the lysimeter irrigated with reclaimed water (west lysimeter) and the lysimeter irrigated with groundwater (east lysimeter). Attribute Groundwater Reclaimed Bicarbonate *2.12 meq/1 *4.88 meq/1 Calcium *1.77 meq/1 *3.68 meq/1 Magnesium *0.18 meq/1 *0.90 meq/1 Sodium *1.03 meq/1 *6.36 meq/1 Nitrate-N *0.05 meq/1 *0.21 meq/1 Sulfate-S 0.31 meq/1 4.13 meq/1 Boron 0.03 meq/1 0.01 meq/1 Chloride 0.59 meq/1 2.96 meq/1 Total Salts 160 mg/1 640 mg/1 PH Conductivity 6.8 7.5 0.25 dS/m 1.0 dS/m SAR 1.0 3.3 Adj. SAR 1.4 6.9 , • ,- , values reported from the most current water analysis teoruary 143 APPENDIX B MONTHLY WATER BALANCE DATA The following data represent each of the lysimeter monthly water balances during the period October 1997 to September 1998. Irrigation, rainfall, evapotranspiration (ET) and drainage events are recorded and summed on a daily basis. Monthly summation of these data are presented in mm at the bottom of each table. 144 October 1997 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Drainage Date Irrigation ET Irrigation Rain Date 4.01 0.22 1 4.27 0.10 1 4.67 3.02 4.01 2 4.34 2.91 4.22 2 5.32 4.14 3.26 3 10.90 3.76 3.38 3 4.95 4.33 4 4.50 4.12 4 4.93 4.04 5 4.62 4.05 5 4.94 4.14 6 4.74 4.20 6 2.57 0.69 0.00 7 2.34 0.65 0.00 7 4.16 0.00 8 3.73 0.00 8 3.79 3.24 9 3.63 2.93 9 6.46 5.42 0.00 10 19.21 4.83 0.00 10 4.79 3.77 11 4.32 3.77 11 3.60 4.42 12 3.46 4.38 12 3.73 6.99 13 3.69 7.17 13 4.03 10.47 14 3.82 11.69 14 3.56 3.85 15 3.93 4.18 15 5.46 2.89 16 6.29 3.60 16 9.33 4.50 3.40 17 8.15 4.25 3.72 17 3.17 4.35 18 2.99 4.49 18 3.41 3.73 19 3.25 3.25 19 3.10 3.91 20 3.09 4.06 20 3.67 3.40 4.16 21 3.60 3.38 4.15 21 1.83 3.79 3.48 22 1.72 3.50 3.32 22 2.26 0.06 2.36 23 2.19 0.12 2.00 23 9.74 4.75 8.44 24 9.02 3.94 6.91 24 3.78 2.76 25 3.56 2.89 25 2.98 3.88 26 3.05 3.83 26 1.36 4.42 27 1.41 4.39 27 2.27 5.31 28 1.97 5.15 28 2.83 4.44 29 2.86 4.32 29 0.66 4.08 4.56 30 6.50 3.96 4.52 30 4.08 4.16 31 3,96 4.12 31 Drainage ET Rain Irrigation Drainage ET Irrigation Rain 31.51, 116.82 10.96 118.99 53.78 118.91 10.56 112.07 Sum(mm) 145 November 1997 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Date Irrigation Drainage Rain ET Irrigation Date 4.08 1 4.16 3.96 4.12 1 4.08 4.16 3.96 2 4.12 2 4.08 4.16 3.96 3 4.12 3 2.13 6.15 4 2.13 4 6.27 2.79 2.42 5 2.72 2.34 5 1.20 1.72 2.34 4.07 6 7.10 1.72 2.38 3.73 6 3.70 8.23 7 3.70 6.50 7 3.70 3.11 8 3.70 3.11 8 3.02 3.85 9 2.74 3.22 9 2.47 0.00 10 2.24 0.00 10 0.65 6.23 2.69 11 6.03 0.60 2.61 11 0.39 0.00 12 0.40 0.00 12 1.56 0.00 13 1.60 0.00 13 8.50 2.87 4.60 0.00 14 11.73 4.64 2.57 0.00 14 1.99 0.20 15 2.00 0.18 15 2.24 0.20 16 2.18 0.14 16 2.06 0.20 17 1.91 0.16 17 1.99 0.55 18 2.04 0.49 18 5.85 2.04 1.91 19 15.40 2.06 1.81 19 2.37 3.02 20 2.26 2.77 20 2.40 2.97 21 2.43 2.53 21 2.62 3.05 22 2.65 3.05 22 2.49 3.30 23 2.87 3.30 23 2.84 3.56 24 2.87 3.56 24 2.40 3.81 25 2.43 3.81 25 8.52 1.90 5.74 26 13.32 2.13 6.82 26 2.23 0.90 2.06 27 0.90 2.31 2.06 27 1.68 1.98 28 1.66 1.87 28 1.75 1.91 29 1.72 1.83 29 1.05 1.81 30 0.99 1.75 30 31 31 Drainage Irrigation Rain ET Drainage ET Irrigation Rain 24.07 70.31 13.45 79.27 47.55 76.27 13.29 71.17 Sum(mm) 146 December 1997 EAST Lysimeter Irrigation Date , 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Sum(mm) , WEST Lysimeter mm Drainage Drainage Date Irrigation Rain ET Rain ET 0.00 12.79 1.83 1 0.00 1.82 12.61 0.00 13.75 0.07 2 0.06 0.00 13.71 1.43 0.00 3 1.10 0.00 1.94 0.00 4 1.85 0.00 8.23 1.70 0.00 5 10.19 2.04 0.00 1.36 0.00 5.54 6 1.16 0.00 5.52 0.00 15.14 0.00 7 0.00 15.15 0.00 0.00 9.02 0.59 8 0.00 8.98 0.24 1.79 0.00 9 1.69 0.00 1.59 0.00 10 1.49 0.00 1.81 0.00 11 1.73 0.00 2.26 2.16 0.00 12 3.73 2.19 0.00 1.16 0.00 13 1.12 0.00 1.61 0.00 14 1.55 0.00 1.32 0.00 15 1.36 0.00 1.61 0.00 16 1.34 0.00 1.45 0.00 17 1.36 0.00 3.42 1.51 0.00 18 2.49 1.77 1.93 2.55 2.24 0.66 19 2.76 2.18 0.52 0.90 0.00 20 0.77 0.00 1.10 1.81 0.98 21 1.23 1.79 1.08 0.08 18.68 0.00 22 0.00 18.63 0.08 1.02 0.00 23 0.73 0.00 0.45 7.35 0.00 24 0.00 7.46 0.51 1.20 0.00 25 0.98 0.00 0.98 0.00 26 1.04 0.00 1.14 0.00 27 0.96 0.00 9.15 1.06 0.00 28 6.82 1.41 0.00 0.83 3.50 29 1.02 4.12 1.16 0.00 30 1.08 0.00 1.63 0.00 31 1.49 0.00 Drainage Irrigation Rain ET Drainage ET Irrigation Rain 23.06 35.31 86.32 8.86 23.23 11.71 86.03 33.87 mm • 147 January 1998 WEST Lysimeter EAST Lysimeter mm mm Date Irrigation Rain ET Drainage Date Irrigation Rain ET 1.20 0.00 3.63 1.42 0.00 1.30 0.00 1.07 1.03 1.30 1.13 2.16 1.83 0.00 0.59 1.24 1.89 0.87 1.92 8.21 1.36 0.81 0.93 2.13 2.65 1.10 0.00 1.20 1.59 1.89 1.56 0.95 1.54 2.09 1.72 1.06 8.25 1.79 1.71 1.68 2.17 1.66 1.96 1.87 1.84 1.73 1.96 2.08 1.77 1.57 1.94 8.11 1.71 1.98 2.21 2.04 1.88 1.98 1.90 2.02 1.82 7.86 0.86 0.88 3.15 2.12 7.27 2.37 2.31 1.99 2.38 Drainage Irrigation Rain ET 36.06 51.82 6.01 50.30 Sum(mm) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 0.00 0.00 0.00 0.94 0.94 2.18 0.00 1.88 1.89 0.90 1.94 2.59 0.00 1.20 2.16 0.92 1.78 0.86 1.94 1.87 1.92 1.83 1.90 1.94 1.91 2.00 1.89 1.85 1.83 7.54 0.84 1.98 2.49 2.32 Irrigation Rain 50.26 5.97 Drainage 1.22 1.36 1.30 1.10 1.18 1.79 1.41 0.72 1.32 0.72 1.28 1.43 1.45 1.23 1.64 1.76 1.49 1.60 1.65 1.67 2.33 1.55 1.58 2.28 1.90 2.00 1.84 0.78 3.09 2.32 1.94 ET 48.93 2.34 4.91 7.15 7.68 7.60 Drainage 29.68 148 February 1998 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Drainage Date Irrigation Irrigation Rain ET Date 2.43 1 2.57 2.49 2.34 1 2.27 2 2.66 2.22 2.60 2 2.06 4.32 2.95 3 2.06 4.32 2.98 3 0.75 0.00 19.81 4 0.73 4 0.00 19.81 1.85 0.00 0.25 5 1.20 0.00 0.25 5 8.31 2.44 0.00 6 1.96 2.10 0.00 6 2.30 7 0.00 2.42 0.00 7 0.00 8.89 0.53 8 0.00 8.89 0.61 8 0.00 12.95 0.53 9 0.00 12.95 0.64 9 2.04 0.00 10 2.16 0.00 10 2.36 0.00 11 2.55 0.00 11 2.38 0.00 12 2.53 0.00 12 4.44 1.73 0.00 13 5.03 1.56 0.00 13 1.77 0.00 14 2.10 0.00 14 0.00 13.46 2.22 15 1.96 0.00 13.46 15 2.47 0.00 16 2.69 0.00 16 0.39 19.56 0.00 17 0.51 19.56 0.00 17 0.73 4.32 0.00 18 0.76 4.32 0.00 18 1.98 0.00 19 2.00 0.00 19 7.86 0.43 7.87 0.00 20 7.84 0.00 7.87 0.48 20 1.32 0.00 21 1.57 0.00 21 2.36 0.00 22 2.61 0.00 22 2.85 0.00 23 3.02 0.00 23 2.40 2.79 0.00 24 0.00 2.79 2.51 24 1.94 3.05 0.00 25 0.00 3.05 2.00 25 2.61 0.00 26 2.83 0.00 26 9.86 3.48 0.00 27 10.25 3.86 0.00 27 2.67 0.00 28 2.93 0.00 28 29 29 30 30 31 31 Drainage Irrigation Rain ET Drainage Irrigation Rain ET 30.47 54.18 97.27 7.29 25.08 7.15 97.27 55.87 Sum(mm) 149 March 1998 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Irrigation Rain Date Rain ET Drainage Irrigation Date 2.74 1 1.79 2.51 1.80 1 2.70 2.77 2 2.47 2 2.30 2.94 2.80 3 2.79 2.28 3 3.63 2.63 4 3.59 4 3.04 3.63 3.63 5 3.59 4.14 5 4.75 3.80 3.59 0.41 6 5.46 3.38 4.05 0.41 6 3.44 0.77 3.61 7 0.77 3.73 4.26 7 2.45 2.49 8 2.61 2.96 8 2.84 2.61 9 2.88 3.07 9 2.85 2.67 10 2.93 3.14 10 3.54 0.18 3.81 11 3.85 0.14 3.54 11 3.96 3.75 12 3.66 4.24 12 8.54 2.67 4.62 13 7.15 2.43 5.24 13 1.54 5.87 2.57 14 1.55 5.87 2.73 14 0.00 0.86 2.08 15 1.96 0.00 0.88 15 2.34 0.00 16 2.24 0.00 16 2.19 1.89 17 2.11 2.16 17 3.32 1.83 18 3.30 2.08 18 3.34 0.00 19 3.23 3.24 19 10.23 4.86 2.79 20 6.91 4.65 2.87 20 4.18 3.77 21 4.01 4.14 21 4.20 3.91 22 3.97 4.30 22 3.89 4.71 23 3.67 5.30 23 5.21 3.66 24 5.15 4.09 24 6.01 3.75 25 5.91 4.09 25 0.49 5.52 11.31 26 0.73 11.12 4.71 26 4.83 3.25 0.00 27 13.95 3.52 0.00 27 4.40 0.00 4.17 28 0.00 4.18 4.75 28' 0.00 16.13 0.57 29 0.62 0.00 16.13 29 1.62 0.00 4.10 30 1.32 0.00 3.81 30 2.87 0.00 31 2.95 0.00 31 Drainage Irrigation Rain ET Drainage ET Irrigation Rain 28.35 97.55 43.80 75.17 33.47 84.08 43.31 95.75 Sum(mm) 150 April 1998 WEST Lysimeter EAST Lysimeter mm mm Drainage Drainage Date Irrigation Rain ET Irrigation Rain ET Date 3.26 0.00 12.83 1 3.12 0.00 12.88 1 3.14 0.00 0.90 2 3.08 0.00 0.90 2 9.70 3.75 0.00 3 10.63 3.59 0.00 3 3.99 0.00 4 3.91 0.00 4 4.32 0.00 5 4.44 0.00 5 3.30 0.26 0.00 6 3.32 0.00 0.26 6 3.54 1.12 2.57 7 3.57 1.12 2.28 7 3.91 0.00 8 3.87 0.00 8 7.21 3.97 0.00 9 8.17 4.36 0.00 9 3.62 1.68 10 3.54 1.61 10 5.34 3.88 11 5.35 3.88 11 3.75 7.19 12 3.59 5.54 12 4.30 4.95 13 3.12 4.33 13 4.38 0.00 14 4.46 0.00 14 3.28 0.00 15 2.95 0.00 15 4.43 3.59 16 4.35 3.52 16 8.47 5.02 7.40 17 8.78 4.58 7.33 17 4.80 4.89 18 4.42 4.75 18 4.68 8.92 19 4.80 8.68 19 5.05 4.64 20 5.02 4.83 20 4.80 5.59 21 4.62 5.52 21 6.03 5.49 22 5.72 5.40 22 6.82 5.53 23 6.57 5.82 23 6.87 6.95 6.64 24 6.99 6.25 6.64 24 4.14 0.37 5.99 25 4.07 6.42 0.39 25 3.30 4.69 26 3.08 4.67 26 4.86 3.84 27 4.60 3.69 27 6.20 4.73 28 6.24 4.85 28 5.93 5.62 29 5.79 5.74 29 5.54 5.66 30 5.55 5.95 _ 30 31 31 Drainage Irrigation Rain ET Drainage Irrigation Rain ET 32.25 136.40 15.48 103.49 34.57 101.45 15.55 131.93 Sum(mm) 151 May 1998 WEST Lysimeter EAST Lysimeter mm mm Date Irrigation Rain ET Drainage Date Irrigation Rain ET 5.47 5.54 7.40 6.13 5.68 5.92 5.87 6.06 5.62 5.99 5.56 4.95 5.87 5.60 4.73 11.35 4.66 5.81 5.58 4.79 6.93 5.85 6.01 6.84 6.52 5.32 0.24 5.34 4.97 4.95 4.73 6.29 5.58 5.93 6.15 5.72 5.73 6.15 6.01 6.68 6.05 6.62 4.59 6.07 6.32 6.97 5.34 _ 6.36 6.27 . 5.74 6.72 5.24 5.70 6.13 5.46 6.93 6.52 8.19 6.19 6.72 5.97 6.58 6.34 6.46 6.66 6.27 5.70 6.48 Drainage ET Rain Irrigation 39.55 182.45 0.24 182.87 Sum(mm) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Drainage 5.91 6.13 5.96 6.09 6.07 5.24 5.62 4.66 5.85 7.24 6.67 6.57 5.69 5.64 6.38 6.53 6.20 6.49 6.61 5.28 6.82 6.88 6.47 6.06 7.08 7.34 5.79 7.41 7.46 7.62 6.96 5.26 6.93 5.66 5.19 5.34 4.91 5.68 4.89 6.46 5.62 4.73 5.22 6.78 5.25 0.24 5.14 4.52 5.66 5.50 5.78 6.33 6.94 6.69 6.19 5.19 5.15 6.74 6.43 5.98 6.14 7.09 5.82 7.07 6.80 6.57 6.79 7.02 Drainage Irrigation Rain ET 30.06 183.40 0.24 196.72 152 June 1998 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Irrigation Drainage Date Irrigation Rain ET Date 7.20 1 8.02 6.32 6.56 1 7.83 8.04 2 6.93 6.40 2 7.59 8.38 3 6.82 7.66 3 6.85 5.68 6.94 4 2.89 5.92 6.54 4 6.72 9.88 5 5.64 10.67 5 7.67 8.16 6 6.88 8.56 6 7.28 10.01 7 6.58 9.15 7 6.43 9.49 8 5.65 9.47 8 7.56 8.92 9 6.51 8.86 9 5.30 9.82 10 4.93 9.07 10 5.50 6.08 8.05 11 7.37 6.39 8.05 11 8.21 9.49 12 7.19 9.31 12 7.97 10.60 13 6.26 10.53 13 7.40 8.38 14 6.26 9.02 14 7.17 8.04 15 6.21 7.62 15 8.30 7.37 16 7.43 7.31 16 7.74 8.15 17 6.61 7.97 17 3.12 8.00 7.46 18 7.29 5.83 7.68 18 7.08 7.25 19 6.69 7.46 19 7.94 6.84 20 7.09 7.64 20 8.10 8.06 21 6.99 0.00 21 5.87 0.00 22 5.89 0.00 22 7.09 16.15 23 6.45 9.78 23 2.36 7.87 7.96 24 4.16 6.98 8.23 24 8.09 7.82 25 6.84 7.90 25 7.63 8.02 26 6.74 8.33 26 7.12 8.77 27 6.15 8.58 27 7.80 8.25 28 6.90 7.52 28 8.16 9.31 29 6.82 8.33 29 6.87 8.03 30 6.70 8.05 30 31 31 Drainage Irrigation Rain ET Drainage Irrigation Rain ET 17.83, 219.75 0.00 251.66 21.61 232.25 0.00 194.60 Sum(mm) 153 July 1998 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Irrigation Date ET Drainage Irrigation Rain Date 4.09 7.55 7.98 1 7.57 3.89 8.17 1 6.47 5.40 2 6.52 5.97 2 6.33 8.31 3 6.15 8.41 3 5.03 7.19 0.16 4 4.85 7.46 0.18 4 1.39 6.03 27.38 5 1.30 6.68 27.22 5 1.49 0.00 15.58 6 1.59 0.00 15.36 6 4.36 18.80 0.00 7 4.69 18.46 0.00 7 5.09 0.00 0.84 8 5.56 0.00 0.84 8 1.91 5.46 0.00 9 5.36 5.67 0.00 9 5.81 0.00 10 6.05 0.00 10 5.62 1.75 2.67 11 5.87 1.69 2.67 11 7.80 5.24 12 7.33 5.54 12 7.48 7.03 13 7.42 7.37 13 7.62 7.05 14 7.50 7.35 14 7.29 6.91 15 7.12 7.68 15 2.12 8.55 7.25 0.31 16 3.57 8.26 0.26 7.09 16 6.59 7.76 2.81 17 6.75 8.17 2.81 17 4.99 1.41 5.42 18 5.07 1.32 5.26 18 6.10 1.79 4.40 19 6.18 1.73 4.77 19 5.72 5.54 20 5.85 5.77 20 4.38 3.71 4.83 21 4.44 3.69 5.22 21 1.36 1.67 11.45 22 1.40 1.79 11.43 22 3.97 5.03 0.51 0.00 23 9.17 4.74 0.00 0.47 23 6.27 0.00 24 6.07 0.00 24 6.07 2.75 25 5.91 2.79 25 6.46 6.40 26 6.40 6.46 26 7.03 6.29 27 6.98 6.25 27 7.03 6.27 28 7.13 6.74 28' 5.23 7.05 29 5.41 7.35 29 4.46 6.02 5.54 30 13.91 6.07 5.56 30 3.48 6.27 11.92 31 3.25 6.32 11.82 31 Drainage ET Rain Irrigation Drainage 'Irrigation Rain ET 16.55 175.10 99.34 140.33 35.90 145.86 98.26 175.10 Sum(mm) 154 August 1998 WEST Lysimeter EAST Lysimeter mm mm Irrigation Rain Date 1 2 3 4 5 6 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 0.00 1.89 5.64 6.46 0.00 1.79 6.52 6.40 5.62 1.73 4.87 3.67 4.60 5.95 0.00 3.85 6.52 2.73 5.54 5.38 4.64 7.60 6.42 6.97 5.60 5.72 5.72 6.29 4.85 6.80 7.54 ET 147.31 5.26 5.02 6.46 6.72 6.74 6.33 5.67 5.32 4.90 4.13 5.88 5.18 5.95 5.10 6.07 6.06 4.45 5.32 6.22 5.62 6.53 6.65 6.10 5.17 5.44 5.81 5.65 5.20 7.02 6.70 5.46 7.84 0.22 3.99 3.42 1.00 9.94 3.32 0.59 0.16 0.20 3.54 Irrigation Rain Sum(mm) Drainage 34.22 ET 178.13 Date Irrigation Rain 14.26 12.06 9.94 8.64 Drainage 44.90 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 ET Drainage 5.30 4.97 6.40 6.65 6.76 6.24 6.02 5.19 5.07 4.06 5.85 4.98 6.00 5.19 5.99 5.89 4.34 5.26 5.96 5.92 6.61 6.05 6.11 4.94 5.87 5.56 6.00 5.01 6.61 4.89 4.45 0.00 1.81 5.46 6.11 8.09 0.00 5.44 1.67 6.27 0.08 6.48 3.99 5.36 1.87 3.36 4.69 3.26 6.13 1.00 4.30 5.36 10.80 0.00 3.50 3.30 6.17 2.59 5.36 6.17 0.57 5.58 0.16 4.13 7.19 6.09 6.64 0.20 5.40 5.50 6.78 5.46 3.77 6.54 4.79 5.96 7.89 Drainage ET Irrigation Rain 24.52, 141.43 35.32 174.14 155 September 1997 WEST Lysimeter EAST Lysimeter mm mm Drainage ET Rain Irrigation Date Drainage Irrigation Rain ET Date 5.70 0.10 1 5.07 0.10 1 5.30 5.90 2 6.60 5.66 2 7.06 4.86 5.15 3 4.28 8.16 5.90 3 2.36 5.32 5.22 4 2.38 6.05 5.46 4 1.98 0.00 18.03 5 1.94 0.00 17.97 5 4.83 0.00 6 4.56 0.00 6 5.36 0.00 0.20 7 5.28 0.00 0.22 7 5.56 0.90 8 5.41 0.94 8 5.09 5.28 9 5.06 5.32 9 4.48 5.10 5.42 10 7.54 4.63 5.58 10 3.84 4.30 11 3.72 4.36 11 4.81 3.81 12 4.76 3.63 12 4.88 4.53 13 4.86 4.73 13 5.15 5.54 14 5.10 5.74 14 5.42 9.19 15 5.33 9.02 15 5.58 3.38 3.95 16 5.28 3.38 4.42 16 5.71 4.05 4.56 17 5.80 3.78 4.38 17 4.71 5.34 18 4.77 5.60 18 4.79 5.46 19 4.81 5.64 19 5.87 4.83 20 5.81 5.19 20 4.55 6.29 21 4.50 6.42 21 4.35 5.32 22 4.42 5.60 22 4.22 4.73 23 4.33 4.93 23 4.92 4.77 4.36 24 3.87 4.91 4.64 24 4.60 5.64 25 4.41 5.79 25 4.42 5.38 26 4.45 5.46 26 3.97 5.48 27 4.24 5.62 27 4.38 5.52 28 4.28 5.81 28 4.76 5.48 29 4.80 5.72 29 4.79 5.79 30 4.65 5.79 30 31 31 Drainage Irrigation Rain ET Drainage ET Irrigation "Rain 22.17, 140.05 26.93 133.47 25.37 138.04 27.03 138.42 Sum(mm) 156 APPENDIX C TIME DOMAIN REFLECTOMETRY Time domain reflectometry (TDR) was used to assess bulk soil salinity (ECa) and volumetric water content Ov using the procedures described in Chapter 2. Independent assessments of soil water EC (ECw) were determined from soil water samples obtained from the lysimeter solution samplers while independent measurements of Ov were obtained using neutron attenuation. Bulk soil EC and ECw were compared over the course of the study as were measurements of 0v made using TDR and neutron attenuation. The results of these comparisons are provided below. Comparison of ECa and ECw Bulk soil EC is useful to determine changes in soil and soil water electrical EC of the conductivity over time. However, the standard for assessing soil salinity is the the saturated extract (ECe), or the soil water salinity itself. In order to determine relationship between the bulk soil salinity (ECa) and the soil water salinity (ECw), the two were compared at three different depths (1.0, 2.0 and 3.0 m) for both the east and west lysimeters. ECw at any given Overall, a strong relationship did not exist between ECa and west lysimeter and is depth. The best relationship occurred at the 1 m depth in the the best fit to presented in Figure 40. A third order polynomial provided Figure 40. A more typical comparison of ECw versus ECa the data in is provided in Figure 41 157 ECw vs. ECa for the West Lysimeter -- 1.0 m depth 4- y = 250149x 3 - 26367x 2 + 948.87x - 8.82 = 0.71 3.5 3- • 2.5 -o LU • • • 2 - •• • • n•• 1.5 1 0.5 T 0 0.005 0.01 0.015 I I I I I 0.025 0.02 0.035 0.03 I I 0.04 0.045 0.05 ECa dS/m Figure 40. Soil solution salinity (ECw) plotted as a function of bulk soil salinity (ECa) for the period August 1997 to September 1998 for the west lysimeter at the 1.0 m depth. The curve depicted represents the least squares regression curve. ECw vs. ECa for the East Lysimeter 1.0m depth 1.2 1 y = 5E+06x 3 - 246850x 2 + 4471.2x - 26.161 0.8 LLI r2 = 0.23 0.6 0.4 0.2 O 0.000 0.005 0.010 0.015 0.020 ECa function of bulk Figure 41. Soil solution salinity (ECw) plotted as a 1998 for soil salinity (ECa) for the period August 1997 to September curve depicted represents the east lysimeter at the 1.0 m depth. The the least squares regression curve. 0.025 158 where ECw appears unrelated to ECa at the 1.0 m depth in the east lysimeter. The closest relationship between ECa and ECw was obtained in the west lysimeter at the 1.0 m depth (r2 =0.71) where overall salinity levels and the change in salinity with time were greatest. Poor ECa versus ECw relationships were obtained at low levels of salinity, and especially where salinity does not change a great deal over time. A much better relationship developed between ECa and ECw when data from both lysimeters and all depths were combined (Figure 42). The combined data set offers a greater range of ECa and ECw values. A relationship between ECa and ECw is apparent in Figure 42, however, the r2 value (0.68) is somewhat lower than observed in other studies. The low soil salinity values encountered in this study may explain the poor relationship between ECa and ECw. Rhoades et al. (1976) observed a good relationship between ECa and ECw when ECw ranged from 2.5 and 56 dS/m (ECa ranged from 0.31 to 1.62 dS/m). However, Nadler et al. (1982) showed that models were deficient at salinity levels below the range tested by Rhoades et al. Persson, (1997) showed similar error at low salinity levels using Rhoades model, and three other models used to compute ECw from ECa. Comparison of Neutron Probe Av and TDR Ov A recent alternative to measuring the change in soil water storage over time with the neutron probe or gravimetric analysis is time domain reflectometry (TDR). Time is less domain reflectometry provides a smaller sampling volume than the neutron probe, 159 Soil Solution Salinity vs. Bulk Soil Salinity 54.5 y = 151.41x -1.69 r2 = 0.68 4- • 3.5 32.5 2 1.5 1 0.5 0.00 0.01 0.01 0.02 0.02 0.03 0.03 0.04 0.04 0.05 ECa dS/m as a function of bulk Figure 42. Soil solution salinity (ECw) plotted to September 1998 for soil salinity (ECa) for the period August 1997 both lysimeters at all depths. The line depicted represents the least squares regression line. 0.05 160 destructive than gravimetric soil sampling and is able to measure the change in soil water storage near the soil surface (Dasberg and Dalton, 1985). Profiles of lysimeter Ov were obtained using both TDR and neutron attenuation on 9 days during the course of this study. Measurements of Ov were obtained at intervals of 0.5 m over the depth range of 0.5 m to 3.5 m. Figure 43 compares Av obtained from TDR with Ov derived from neutron attenuation. A least squares regression equation was fit to the data in Figure 43. Both the non-zero intercept and the slope of 1.16 suggest the two procedures for measuring Ov do not provide similar values of Ov. On average, Ov measured with TDR was 0.02 m 3 /m3 lower than Ov measured by neutron probe. This discrepancy among the methods could be due to several factors, including: (1) differences in the soil volume sampled, (2) differing sampling locations for the TDR and neutron attenuation, and (3) inaccurate calibration of the neutron probe. Differences in sampling volume may have caused bias between Ov measured with TDR and neutron attenuation. The TDR sample volume is restricted to an area approximately within the three-wire probes, whereas the neutron probe measures sample in a volumes ranging from a radii of approximately 0.16 m in a saturated soil to 0.7 m soil near oven dryness (Van Bavel et al., 1956). Sampling locations also differed for TDR and neutron attenuation. Time domain from the reflectometry probes are located at each of the three angular positions 500 mm locations for inner lysimeter wall and the probes are 200 mm in length, thus the sampling TDR range from 500 mm to 700 mm from the inner lysimeter wall. In contrast, the 161 TDR Volumetric Water Content vs. Neutron Probe 0.15 • y= 1.17x - 0.04 0.14 r2 = 0.48 0.13 0.12 - - •to • 0.11 — ••• • • • • 0.10 0.09 I 0.08 0.08 0.09 0.1 0.11 0.12 0.13 0.14 0.15 Neutron Probe - Theta Figure 43. Volumetric water content (theta) derived from time domain reflectometry (TDR) plotted as a function of theta derived from neutron attenuation for the period August 1997 to September 1998 for both lysimeters at all depths. The line depicted represents the least squares regression line. 0.16 162 sampling location for neutron attenuation extends from the neutron access tube located directly in the center of the lysimeters in a spherical pattern. Differences in Av based on sampling location could create systematic errors; however one would expect these differences to eventually impact turf performance. Turf performance did not vary over the lysimeter. Site specific calibration of the neutron probe is necessary to obtain accurate measurements of Ov. Inaccurate calibration may result in measurement error, which may have been the case in this study. Calibration of the neutron probe involved recording count ratios through the entire lysimeter soil profile on 12 dates and then comparing the mean change in profile count ratio to the change in soil water storage (ASWS) as given by the lysimeter mass change between any two sampling dates. This calibration method is better suited for studies where ASWS is large. Unfortunately, the ASWS was not large in this study and the probe calibration may well be in error. 163 APPENDIX D MANAGEMENT OF LYSIMETER DATA GAPS Lysimeter data were lost or in error on 17 days between October 1997 and September 1998. Missing data occurred on 2, 12 and 3 days in the months of October 1997, November 1997 and March 1998, respectively. The water balance equation, rearranged to solve for ET, (Equation ), was used to calculate the ET during these periods of missing data: ET = P + I —D — ASWS [Eq. 2] For each data gap, the amount of irrigation (I) applied was estimated from irrigation run times or by multiplying the applicable Kc by ETo if run times were not available. Any precipitation (P) that fell during data gaps was recorded by the on-site weather station. Raingage precipitation was adjusted upward using Equations  and  based on previous work that indicated lysimeter precipitation always exceeded precipitation recovered by the raingage. East Lysimeter Rain [mm] = 1.148 (raingage rainfall (mm)) + 0.093 [Eq. 18] West Lysimeter Rain [mm] = 1.149 (raingage rainfall (mm)) + 0.081 [Eq. 19] in Drainage (D) was quantified manually and converted to an equivalent depth soil water storage mm of water and runoff (R) was assumed equal to zero. The change in (ASWS) was determined from the change in lysimeter weight during periods of missing data. ET was then computed from these values. 164 The above procedure did not provide acceptable values for missing data during 12 and 13 Nov. 1997. It was assumed an incorrect irrigation amount was the reason the procedure failed; thus, on these days ET was estimated by multiplying the mean monthly Kc value by the ETo. The water balance equation was then solved for I using known values of P, R, and ASWS and estimated ET. 165 APPENDIX E LYSIMETER PERFORMANCE AND RECOMMENDATIONS Lysimeter Performance During the past two years we have struggled with many aspects of lysimeter operation. Effective evaluation of the ADWR water duty required an efficient irrigation system. An initial test of the lysimeter irrigation system (June 1997) revealed unstable water pressure as a cause for inconsistent irrigation of the east lysimeter. A shallow well pump system was installed to ensure adequate water pressure in an attempt to solve the problems with east lysimeter irrigation.. Following installation of the pump, frequent precipitation tests were run using low lying soil cans to decrease water loss due to receptacle deflection, thus decreasing potential for error. The irrigation coefficient of rates uniformity (CU) was quite good for both lysimeters (>90%), however, precipitation were not equal. A difference in static pressure between the two lysimeter irrigation systems was identified as the cause of differing precipitation rates. In-line pressure regulators were installed at 0.207 MPa (30psi) in the irrigation lines of each lysimeter on and the CU for both 22 Aug. 1997 and by 4 Sep. 1997, precipitation rates were equal, lysimeters was 93%. easy task. Thatch Growing high maintenance turf on the lysimeters is not an build-up has created a slight mound on the lysimeter, which has made It is important to mow the surrounding turf and lysimeter turf mowing difficult. at the same height to avoid circular pattern on the lysimeter, ET measurement error. If mowing is done following a 166 traffic stress due to the weight of the mower wheels becomes evident on the turf To avoid this, turf immediately surrounding the lysimeter is mowed prior to mowing the lysimeter turf. The mower can then be driven across the lysimeter and the surrounding turf, while only clipping the lysimeter turf One must be careful, however, not to scalp the lysimeter turf, as the lysimeters are slightly raised above the surrounding area. Serious consideration should be given to re-leveling the lysimeter surfaces at the conclusion of this study. In order to decrease any heterogeneity in the soil, these lysimeters were filled with vinton sand (93% sand). However, this soil also has a very low cation exchange capacity and turf nutrient deficiencies will result unless frequently fertilized. Previous studies have indicated that improved turf performance is associated with effluent irrigation. However, irrigation with different water sources presents the problem of a bias in nutrition between the west and east lysimeters. West lysimeter fertilizer applications are decreased to account for the nitrogen in effluent water. However, greater daily nutrition from effluent has produced increased growth, quality, ET and thatch build-up compared to the east lysimeter. Thatch build-up has increased on both lysimeters to produce a slight mound. A in ET lysimeter crop that is raised above the surrounding area can cause errors measurement (Allen et al., 1991). Fortunately, turf is maintained at very low mowing heights (-2.5 cm), thus minimizing error. 167 Turf growing over the lysimeter rim creates two sources of measurement error for ET. Growth beyond the lysimeter rim expands the evaporative surface of the lysimeters and leads to overestimation of ET. Growth of turf onto adjacent soil areas can inhibit lysimeter movement, thus affecting change in lysimeter mass. Thatch, rhizomes and roots can get stuck or grow under and through the nylon flashing covering the tank rim, which can also inhibit tank movement. Regular trimming of turf along the outer rim of the lysimeters is required to minimize these errors. Also, the nylon flashing should be replaced every four to six months in order to deter this potential error. During overseeding, ryegrass seed is planted to produce green turf during winter months. Unfortunately, the local bird population loves to feed on ryegrass seed. Thus, during the fall overseeding process in 1997, Jeff Gilbert and I built a "bird-away" structure for each lysimeter. The structure consisted of PVC pipe connected to form a square area of approximately 13.3 m 2 (144 ft 2 ) and 36 cm high with a nylon net strewn across to keep birds away. Water has appeared on the lysimeter floor following significant rainfall events. A small amount of water seeps in from under the lysimeter well floor, however, the majority of water entered the facility from an electrical conduit that supplies power to the facility. south and 25 A break in this conduit was discovered in an in-ground electrical box 20 m has since been sealed, and m west of the lysimeter well stairway. This electrical conduit soil levies were constructed to inhibit water from entering the box. Facility flooding has not been a problem since. 168 When heavy rains occur or when excessive irrigation is necessary, timely removal of lysimeter drainage water is essential to avoid leakage from the lysimeters. If water within the lysimeters accumulates to a height above the lower lysimeter sampling ports, hydrostatic pressure can force open the stoppers resulting in lysimeter water loss. Thus it is essential to ensure the drainage pumps are running properly, and drainage is removed frequently. During monsoon, lightning can be a problem. In early September 1997, lightning struck very near the lysimeter well. An electrical surge destroyed the computer, and the datalogger program and memory were scrambled. Two surge protectors have since been installed to protect the computer and datalogger. One other problem we encountered and solved was a mysterious change in lysimeter weight upon entering and exiting the lysimeter well. The lysimeter well door must be left open while the lysimeter ventilation system is functioning, otherwise air pressure will build up and slightly lift the tanks (enough to affect lysimeter weight change). Lysimeter Recommendations few concerns need to In order to better evaluate the lysimeter salinity profiles, a closer to the soil be addressed. Installation of time domain reflectometry (TDR) probes zone salinity surface at depths of 0.15 m and 0.35 m would significantly improve root zone measurement. Previous research lends support for the need to monitor upper root salinity. Garrot and Mancino (1994) determined that —55 % of soil water use by turf 169 plants came from the top 0.30 m of soil. The east lysimeter needs TDR probes installed at these depths, and at 0.5 m for consistency with the west lysimeter. Finally, it would seem wise to use the same irrigation water on both lysimeters. Our data have shown that irrigation with effluent water will increase growth rates and turf quality compared to the same soil and turf system irrigated with potable water. Use of the same water might eliminate the difference in turf quality and water use observed in this study and allow two replicate treatments for any future study. 170 APPENDIX F CAMPBELL SCIENTIFIC DATALOGGER PROGRAM The following provides the Campbell CR7 datalogger program used to obtain lysimeter mass change data. Program: Flag Usage: Input Channel Usage: Excitation Channel Usage: Continuous Analog Output Usage: Control Port Usage: Pulse Input Channel Usage: Output Array Definitions: Table 1 Programs * 1 01: 2 Sec. Execution Interval Resolution 01: P78 1: 1 2: High Resolution P89 If X<=>F 01:32 X Loc 02: 1 03:0 04: 99 F Call Subroutine 99 171 3: P86 Do 01: 1 4: Call Subroutine 1 P86 Do 01: 2 Call Subroutine 2 05: P86 Do 01: 3 06: P92 Call Subroutine 3 If time is 01:0minutes into a 02: 10 minute interval 03: 98 Call Subroutine 98 07: P92 If time is 01:0minutes into a 2: 10 minute interval 3: 10 Set high Flag 0 (output) 08: P77 Real Time 01: 111 Day,Hour-Minute, Seconds 09: P71 Average 01:3Reps 02: 3 10: P71 Loc Average 01:7Reps 172 02: 7 Loc 11: P92 If time is 01:0minutes into a 2: 360 minute interval 3: 99 Call Subroutine 99 Page 2 Table 1 12: P End Table 1 * 2 Table 2 Programs 01: 60 Sec. Execution Interval 01: P78 High Resolution 1: 1 2: P86 Do Call Subroutine 4 01: 4 P86 Do 3: 01: 5 4: Resolution Call Subroutine 5 P86 Do 1: 7 05: P92 Call Subroutine 7 If time is 01:0minutes into a 2: 60 minute interval 03:10Set high Flag 0 (output) 173 6: P77 Real Time 01: 111 Day,Hour-Minute,Seconds 7: P71 Average 1: 19 Reps 2: 14 Loc 08: P * 3 End Table 2 Table 3 Subroutines 01: P85 Beginning of Subroutine 1: 99 Subroutine Number 2: P89 If X<=>F 01:32 X Loc 02:1 03:0F 04:30 Then Do 3: P30 Z=F 01:1F 02:32Z Loc : 4: P95 End 05: P30 Z=F 01:5000 F 02:36Z Loc : 174 6: P30 Z=F 01:0 F 02:35 Z Loc : Page 3 Table 3 7: P6 Full Bridge 1: 1 Rep 2: 6 500 mV slow Range 03: 1 IN Card 04:1 IN Chan 05:1 EX Card 06:1 EX Chan 7: 1 Meas/EX 8: 2500 mV Excitation 09:1 Loc : 10: 2423.8 Mult 11:0Offset 08: P6 Full Bridge 1: 1 Rep 2: 6 500 mV slow Range 03: 1 IN Card 04:2 IN Chan 175 05:1EX Card 06:2EX Chan 7: 1 Meas/EX 8: 2500 mV Excitation 09:2Loc : 10:2488 Mult 11:0Offset 9: P95 End 10: P85 Beginning of Subroutine 01: 98 Subroutine Number 11: P35 Z=X-Y 01:3X Loc 02:1Y Loc 03:33Z Loc : 12: P35 Z=X-Y 01:4X Loc 02:2Y Loc 03:34Z Loc : 13: P93 Case 01:32Case Loc 14: P83 If Case Location < F 176 01:-99 F 02: 95 Call Subroutine 95 15: P83 If Case Location < F 01:99 F 02: 97 Call Subroutine 97 16: P95 End 17: P95 End Page 4 Table 3 18: P85 Beginning of Subroutine 01: 97 Subroutine Number 19: P89 If X<=>F 01:33 X Loc 02:3 >= 03: 120 F 04: 96 Call Subroutine 96 20: P89 If X<—>f 01:34 X Loc 02:3 >= 3: 120 F 4: 96 Call Subroutine 96 21: P95 End 177 22: P85 Beginning of Subroutine 01: 96 Subroutine Number 23: P21 Analog Out 01:1EX Card 02: 1 CAO Chan 03:36 mv Loc 24: P22 Excitation with Delay 01:1EX Card 02:8EX Chan 3: 0 Delay w/EX (units=.01sec) 4: 200 Delay after EX (units=.01sec) 05: 0 25: P21 mV Excitation Analog Out 01:1EX Card 02: 1 CAO Chan 03:35 mv Loc 26: P30 Z=F 01:-999 F 02:32 Z Loc : 27: P95 End 28: P85 Beginning of Subroutine 178 01: 95 Subroutine Number 29: P89 If X<=>F 01:33 X Loc 02:4 3: 120 F 4: 94 Call Subroutine 94 30: P89 If X<=>F 01:34 X Loc 02:4 3: 120 F 4: 94 Call Subroutine 94 Page 5 Table 3 31: P95 End 32: P85 Beginning of Subroutine 01: 94 Subroutine Number 33: P21 Analog Out 01:1 EX Card 02: 1 CAO Chan 03:36 mv Loc 34: P22 Excitation with Delay 01:1 EX Card 179 02:8 EX Chan 3: 0 Delay w/EX (units=.01sec) 4: 200 Delay after EX (units=.01sec) 05: 0 mV Excitation 35: P21 01:1 Analog Out EX Card 02:1CAO Chan 03:35 mv Loc 36: P30 Z=F 01:1F 02:32Z Loc : 37: P95 End 38: P85 01: 1 39: P6 Beginning of Subroutine Subroutine Number Full Bridge 1: 1 Rep 2: 4 50 mV slow Range 03:1 IN Card 04: 1 IN Chan 05:1 EX Card 06: 1 EX Chan 180 7: 1 Meas/EX 8: 2500 mV Excitation 09:3 Loc : 10: 2423.8 Mult 11:0Offset 40: P89 If X<=>F 01:3 X Loc 02:4 < 03: -99998 F 04:30 Then Do Page 6 Table 3 41: P6 Full Bridge 01:1 Rep 02:5 150 mV slow Range 3: 1 IN Card 4: 1 IN Chan 05: 1 EX Card 06:1 EX Chan 7: 1 Meas/EX 8: 3000 mV Excitation 09:3 Loc : 181 10: 2423.8 Mult 11:0Offset 42: P89 If X<=>F 01:3 X Loc 02:4 03: -99998 F 04:30 Then Do 43: P6 Full Bridge 01:1 Rep 02: 6 500 mV slow Range 03:1IN Card 04: 1 IN Chan 05:1EX Card 06:1EX Chan 7: 1 Meas/EX 8: 2500 mV Excitation 09:3 Loc : 10: 2423.8 Mult 11:0Offset 44: P95 End 45: P95 End 01: 182 46: P95 End 47: P85 2 Beginning of Subroutine Subroutine Number 48: P6 Full Bridge 1: 1 Rep 2: 4 50 mV slow Range 03:1IN Card 04:2IN Chan 05:1EX Card 06:2 EX Chan 7: 1 Meas/EX 8: 2500 mV Excitation 09:4Loc : 10:2488 Mult 11:0 Offset Page 7 Table 3 49: P89 If X<=>F 01:4X Loc 02:4< 03: -99998 F 04:30 Then Do 183 50: P6 Full Bridge 1: 1 Rep 2: 5 150 mV slow Range 03: 1 IN Card 04:2 IN Chan 05:1 EX Card 06:2 EX Chan Meas/EX 7: 1 8: 3000 mV Excitation Loc : 09:4 10: 2488 Mult 11:0Offset 51: P89 If X<=>f X Loc 01:4 2: 4 < 3: -99998 F 04:30 Then Do 52: P6 Full Bridge 1: 1 Rep 2: 6 500 mV slow Range 03:1 IN Card 184 04:2IN Chan 05:1EX Card 06:2 EX Chan 7: 1 Meas/EX 8: 2500 mV Excitation 09:4 Loc : 10:2488 Mult 11:0Offset 53: P95 End 54: P95 End 55: P95 End 56: P85 01: 3 Beginning of Subroutine Subroutine Number 57: P17 Panel Temperature 01:1IN Card 02:6 Loc : Page 8 Table 3 58: P14 Thermocouple Temp (DIFF) 1: 1 Rep 2: 2 5000 uV slow Range 03:1IN Card 185 04:3IN Chan 5: 1 Type T (Copper-Constantan) 6: 6 Ref Temp Loc 07:5Loc : 8: 1 Mult 9: 0 Offset 59: P14 Thermocouple Temp (DIFF) 01:7Reps 02: 2 5000 uV slow Range 03:1IN Card 04:4IN Chan 05: 1 Type T (Copper-Constantan) 06:6 Ref Temp Loc 07:7Loc : 8: 1 Mult 9: 0 Offset 60: P95 End 61: P85 01: 4 62: P6 Beginning of Subroutine Subroutine Number Full Bridge 01:8Reps 186 02: 1 1500 uV slow Range 03:2IN Card 04: 1 IN Chan 05:1EX Card 6: 1 EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 14 Loc : 10: 1 Mult 11:0Offset 63: P89 If X<=>F 01:14 X Loc 2: 4 3: -99998 F 04:30 Then Do Page 9 Table 3 64: P6 Full Bridge 01:8Reps 02: 2 5000 uV slow Range 03:2IN Card 04:1IN Chan 187 05:1EX Card 06:1EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 14 Loc : 10: 1 Mult 11:0Offset 65: P89 If X<—>F 01:14X Loc 02:4 03: -99998 F 04:30 Then Do 66: P6 Full Bridge 01:8Reps 02:315 mV slow Range 03:2IN Card 04:1IN Chan 05:1EX Card 6: 1 EX Chan 7: 1 Meas/EX 08: 5000 mV Excitation 188 9: 14 Loc : 10: 1 Mult 11:0Offset 67: P89 If X<=>F 01:14 X Loc 02:4 03: -99998 F 04:30 Then Do 68: P6 Full Bridge 01:8Reps 02: 4 50 mV slow Range 03:2IN Card 04:1IN Chan 05: 1 EX Card 06:1EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 14 Loc : 10: 1 Mult 11:0Offset 69: P95 End 189 70: P95 End 71: P95 End Page 10 Table 3 72: P95 End 73: P85 01: 5 Beginning of Subroutine Subroutine Number 74: P6 Full Bridge 01:4Reps 02: 2 5000 uV slow Range 03:2 IN Card 04:9 IN Chan 5: 1 EX Card 6: 1 EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 22 Loc : 10: 1 Mult 11:0Offset 75: P89 If X<—>F 01:22 X Loc 02:4 < 190 03: -99998 F 04:30 Then Do 76: P6 Full Bridge 01:4Reps 02: 2 5000 uV slow Range 03:2IN Card 04:9IN Chan 05:1EX Card 6: 1 EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 22 Loc : 10: 1 Mult 11:0Offset 77: P89 If X<=>F 01:22 X Loc 02:4 < 03: -99998 F 04: 30 Then Do 78: P6 Full Bridge 01:4Reps 191 02: 3 15 mV slow Range 03:2IN Card 04:9IN Chan 05:1EX Card 6: 1 EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 22 Loc : 10: 1 Mult 11:0Offset Page 11 Table 3 79: P89 If X<=>F 01:22 X Loc 02:4 < 03: -99998 F 04: 30 Then Do 80: P6 Full Bridge 01:4Reps 02: 4 50 mV slow Range 03:2IN Card 04:9IN Chan 192 05:1EX Card 6: 1 EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 22 Loc : 10: 1 Mult 11:0Offset 81: P95 End 82: P95 End 83: P95 End 84: P95 End 85: P85 01: 7 86: P6 Beginning of Subroutine Subroutine Number Full Bridge 01:6Reps 02:1 1500 uV slow Range 03:3IN Card 04: 1 IN Chan 05:1EX Card 06:3EX Chan 07: 1 Meas/EX 193 8: 5000 mV Excitation 9: 26 Loc : 10: 1 Mult 11:0Offset 87: P89 If X<=>F 01:26 X Loc 02:4 < 03: -99998 F 04:30 Then Do Page 12 Table 3 88: P6 Full Bridge 01:6Reps 02: 2 5000 uV slow Range 03:3IN Card 04: 1 IN Chan 05: 1 EX Card 06:3EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 26 Loc : 10: 1 Mult 194 11:.0Offset 89: P89 If X<=>F 01:26 X Loc 02:4 03: -99998 F 04:30 Then Do 90: P6 Full Bridge 01:6Reps 02: 3 15 mV slow Range 03:3IN Card 04: 1 IN Chan 05: 1 EX Card 06:3EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 26 Loc : 10: 1 Mult 11:0Offset 91: P89 If X<=>F 01:26 X Loc 02:4 195 03: -99998 F 04:30 Then Do 92: P6 Full Bridge 01:6Reps 02: 4 50 mV slow Range 03:3IN Card 04: 1 IN Chan 05: 1 EX Card 06:3EX Chan 7: 1 Meas/EX 8: 5000 mV Excitation 9: 26 Loc : 10: 1 Mult 11:0Offset 93: P95 End 94: P95 End 95: P95 End Page 13 Table 3 96: P95 End 97: P * 4 End Table 3 Mode 4 Output Options 196 1: 1 (Tape OFF) (Printer ON) 2: 2 Printer 9600 Baud * A Mode 10 Memory Allocation 1: 36 Input Locations 2: 64 Intermediate Locations * C Mode 12 Security 1: 00 Security Option 2: 0000 Security Code 197 REFERENCES Arizona Department of Water Resources. 1984. Management plan for the first management period. Arizona Department of Water Resources, Phoenix AZ. Arizona Department of Water Resources. 1991. Management plan for the second management period. Arizona Department of Water Resources, Phoenix AZ. Arizona Department of Water Resources Internet Address. 1997. Tucson AMA figures [Online]. Available at http://vvww.adwr.state.az.us/AZWater Info/InsideAMAs/tucimages.html#s1 (verified 3 Apr. 1999). Arizona Department of Water Resources. 1998. Draft of the third management plan. Arizona Department of Water Resources, Phoenix AZ. Allen, R.G., W.O. Pruitt, and M.E. Jensen. 1991. Environmental requirements of lysimeters. Proceedings of the International Symposium on Lysimetry: Honolulu, Hawaii, July 23-25, 1991. New York, N.Y. ASCE. Al-Nakshabandi, G.A. 1983. The potential ET of short grass (Cynadon dactylon) as related to the estimated potential evaporation and ET from meteorological data in the coastal region of Kuwait. Journal of Arid Env. 6:33-38. Ayers, R.S., and D.W. Westcott. 1989. Water quality for agriculture. FAO #29 irrigation and drainage paper. Rev 1. United Nations. Rome, Italy. 174p. Beard, J.B. 1982. Turf management for golf courses. Macmillan Publishing Co. 866 Third Ave. New York, N.Y. Beard, J.B. 1985. An assessment of water use by turfgrasses. Turfgrass water conservation, Editors: Gibeault, V.A. and Cockerham, S.T. [Oakland CA]; Coop. Extension, Univ. CA, Division of AG and Natural Resources, c1985. Biran, I., B. Bravado, I. Bushkin-Harav, and E. Rawitz. 1981. Water consumption and growth rate of eleven turfgrasses as affected by mowing height, irrigation frequency, and soil moisture. Agron. J. 73:86-90. Bowen, I.S. 1926. The ratio of heat losses by conduction and by evaporation from any water surface. Phys. Rev. 27:779-787. 198 Bower, C.A., and L.V. Wilcox. 1965. Soluble salts. In C.A. Black et al. (ed.) Methods of soil analysis. Agronomy 9:933-951. Am Soc. of Agron., Inc., Madison, Wis. Brown, P.W. 1998. AZMET computation of reference crop evapotranspiration [Online]. Available at http://ag.arizona.edu/AZMET/et2.htm (verified 3 Apr. 1999). Brown, P.W. 1999a. Turf water management series: I basics of evaporation and evapotranspiration. Extension bulletin. In press. Brown, P.W. 1999b. Confused about weather station ET? Perhaps you should be. ET and crop coefficient data from various weather stations are not interchangeable [Online]. Available at http://usga.com/green/ (verified 3 Apr. 1999). Brown, P.W. and D.M. Kopec. 1997. Applicable case review studies relevant to turfgrass water use issues. Report submitted to Turfgrass Technical Advisory Committee of Phoenix AMA, ADWR. Brown, P.W., J.J. Gilbert, and D.M. Kopec. 1996. Report to Arizona Department of Water Resources: Turfgrass irrigation scheduling using weather based estimates of evapotranspiration for high and low traffic turfs. Contract No. CA94TUI03-00. Brown, P.W., T. Thompson, D.M. Kopec, M.H. Young, P. Wierenga, and B.S Whitlark. 1998. Turfgrass irrigation with municipal effluent: nitrogen fate, turf crop coefficients and water requirements. Final project report submitted to the United States Golf Association. Christiansen, J.E. 1942. Irrigation by sprinkling. University of California Agricultural Experiment Station, Berkley CA. Bull. 670. Dalton, F.N., and J.A. Poss. 1990. Soil water content and salinity assessment for irrigation scheduling using time domain reflectometry: principles and applications. USDAARS. U.S. Salinity Laboratory, Riverside, CA. Dalton, F.N., J.A. Poss, T.J. Heimovaara, R.S. Austin, W.J. Alves, and W.B. Russell. 1990. Principles, techniques and design considerations for measuring soil water content and soil salinity using time-domain reflectometry. U.S. Salinity Lab. Res. Rep. 122. U.S. Salinity Lab., Riverside, CA. Dalton, F.N., W.N. Herkelrath, D.S. Rawlins, and J.D. Rhoades. 1984. Time domain reflectometry: Simultaneous measurement of soil water content and electrical conductivity with a single probe. Science (Wash, DC)224:989-990. 199 Danielson, R.E., C.M. Feldhake, and W.E. Hart. 1981. Urban lawn irrigation and management practices for water saving with minimum effect on lawn quality. Office of Water Res. And Technology Proj. No. A-043-Colo., Fort Collins, CO. U.S. Dept. of the Interior, Washington, DC. Dasberg, S. and F.N. Dalton 1985. Time domain reflectometry field measurements of soil water content and electrical conductivity. Soil Soc. Am. J. 49:293-297. Dean, D.E., D.A. Devitt, L.S. Verchick and R.L. Morris. 1996. Turfgrass quality, growth, and water use influenced by salinity and water stress. Agron. J. 88:844-849. Devitt, D.A. 1989. Bermudagrass response to leaching fractions, irrigation salinity, and soil types. Agron. J. 81:893-901. Devitt. D., D. Bowman, and P. Schulte. 1993. Response of Cynadon dactylon to prolonged water deficits under saline conditions. Plant Soil 148:239-251. Dudeck, A.E., and C.H. Peacock. 1985. Effects of salinity on seashore paspalum turfgrasses. Agron. J. 77:47-50. Dudeck, A.E., C.H. Peacock, and J.C. Wildmon. 1993. Physiological and growth responses of St. Augustinegrass cultivars to salinity. HortScience 28(1):46-48. Dugas, W.A., D.R. Upchurch, and J.T. Ritchie. 1985. A weighing lysimeter for evaporation and root measurements. Agron. J. 77:821-825. Ebdon, J.S. and A.M. Petrovic. 1988. Morphological and growth characteristics of lowand high-water use Kentucky bluegrass cultivars. Crop Sci. 38:143-152 Eden, S., and Wallace, M.G. 1992. Arizona water: information and issues. Water Resources Research Center. The University of Arizona. Erie, L.J., French, 0.F., Bucks, D.A., and Harris, K. 1981. Consumptive use of water by major crops in the southwestern U.S. United States Dept. of Ag., Conservation research report No. 29, 42p. Evett, S. 1994. Time domain reflectometry (TDR) automatic data collection program, Vadose Zone Equipment Company, Version 7-24-94. Feldhake, C.M., R.E. Danielson, and J.D. Butler. 1983. Turfgrass evapotranspiration. I. Factors influencing rate in urban environments. Agron. J. 75:824-830. 200 Feldhalce, C.M., R.E. Danielson, and J.D. Butler. 1984. Turfgrass ET II. Responses to deficit irrigation. Agron. J. 76:85-89. Frank, A.B., R.E. Barker, and J.D. Berdahl. 1987. Water-use efficiency of grasses grown under controlled and field conditions. Agron. J. 79:541-544. Garrot, D.J., and C.F. Mancino. 1994. Consumptive water use of three intensively managed bermudagrasses growing under arid conditions. Crop Sci. 34:215-221. Gee. G.W., P.J. Wierenga, B.J. Andraski, M.H. Young, M.J. Fayer, and M.L. Rockhold. 1994. Variations in water balance and recharge potential at three western desert sites. Soil Sci. Soc. Am. J. 58:63-72. Hanson, B., S.R. Grattan, and A. Fulton. 1993. Agriculture salinity and drainage. University of California Irrigation Program. University of California Davis. Harivandi, M.A., J.D. Butler, and L. Wu. 1992. Salinity and turfgrass culture. In Turfgrass Agronomy. Monograph 32. Chpt. 6. Am. Soc. of Agron., Inc., Madison, WI. Heimovaara, T.J., A.G. Focke, W. Bouten, and J.M. Verstraten. 1995. Assessing temporal variations in soil water composition with time domain reflectometry. Soil Sci. Soc. Am. J. 59:689-698. Hoagland, D.R. and D.I. Arnon. 1950. The water-culture for growing plants without soil. Calif Agr. Expt. Sta. Circular. 347 (Revised). Jensen, M.E., and H.R. Haise. 1963. Estimating evapotranspiration from solar radiation. Journal of Irrigation and Drainage, 89:15-41. Jensen, M.E., R.D Burman,. and Allen, R.G. 1990. ET and irrigation water requirements. Am. Soc. of Eng. New York, New York 10017. Johns, D., J.B. Beard, and C.H.M. Van Bave!. 1983. Resistances to ET from a St. Augustine turf canopy. Agron J. 75:419-422. Johnston, J. 1984. Econometric Methods. Third Edition. McGraw-Hill Book Company. McGraw-Hill Inc. Kirkham, R.R., G.W. Gee, and T.L. Jones. 1984. Weighing lysimeters for long-term water balance investigations at remote sites. Soil Sci. Soc. Am. J. 48:1203-1205. 201 Klute, A., and C. Dirksen. 1986. Hydraulic conductivity and diffusivity: Laboratory Methods. P. 687-734. In Klute (ed.) Methods of Soil Analysis. Part 1. Second Edition. Agronomy Monograph 9. Am. Soc. of Agron., Inc., Madison, WI. Kneebone, W.R., D.M. Kopec, and C.F. Mancino. 1992. Water requirements and irrigation, in turfgrass, D.V. Waddington, R.N. Carrow, and R.C. Shearman, eds. In Turfgrass Agronomy. Monograph 32:441-467. Kneebone, W.R. and I.L. Pepper. 1982. Consumptive use by sub-irrigated turfgrass under desert conditions. Agron. J. 74:419-423. Kneebone, W.R. and I.L. Pepper. 1984. Luxury water use by bermudagrass turf. Agron. J. 76:999-1002. Kopec, D.M., P.W. Brown, C.F. Mancino, and J.J. Gilbert. 1991. Water use and crop coefficients for warm and cool season turfgrasses in the desert. Agron. Abstracts. Kopec, D.M., R.C. Shearman, and T.P. Riordan. 1988. Evaporation of tall fescue turf. HortScience 23(2):300-301. Kramer, P.J. and J.S. Boyer. 1995. Water relations of plants and soils. Academic Press, San Diego. Krogman, K.K. 1967. Evapotranspiration by irrigated grass as related to fertilizer. Can. J. Plant Sci. 47:281-287. Maas, E.V. and G.J. Hoffman. 1977. Crop salt tolerance - current assessment. Journal of Irrigation and Drainage Division. ASCE. June 1977 115-135. Mantell, A. 1966. Effect of irrigation frequency and nitrogen fertilization on growth and water use of a kikuyugrass lawn (Pennisetum clandestinum Hochst.). Agron. J. 58:559-561. Marcum, K. 1994. Salt-tolerant mechanisms of turfgrasses. Golf Course Mngt. Sept 1994. 55-59. Marek, T.H., A.D. Schneider, T.A. Howell, and L.L. Ebeling. 1988. Design and construction of large weighing monolithic lysimeters. Trans ASAE. 31(2):477-484. Marsh, A.W., R.A. Strohman, S Spaulding, V.B. Youngner, and V.A. Gibeault. 1980a. Determining the water requirements of various turfgrass species. Irrig. Assoc. Ann. Tech. Conf. Proc. 1:34-45. 202 Marsh, A.W., R.A. Strohman, S. Spaulding, V.B. Youngner, and V.A. Gibeault,. 1980b. Turfgrass Irrigation Research at the University of California: warm and cool season grasses tested for water needs. Irrigation Journal. p. 20-33 Meyer, J.L. and V. A. Gibeault. 1986. Turfgrass performance under reduced irrigation. Ca Agriculture. Jul-Aug 1986 P. 17-20. Meyer, J.L. and V.A. Gibeault. 1987. Turfgrass Performance when underin igated. Applied Agricultural Research Vol. 2, p. 117-119. - Mualem, Y., and S.P. Friedman. 1991. Theoretical prediction of electrical conductivity in saturated and unsaturated soil. Water Resources Res., 27(10) 2771-2777. Nadler, A. 1982. Estimating the soil water dependence of the electrical conductivity soil solution/electrical conductivity bulk soil ratio. Soil Sci. Soc. Am. J., 46:722-726. Nadler, A. and H. Freakel. 1980. Determination of soil solution electrical conductivity from bulk soil electrical conductivity measurements by the four-electrode method. Soil Sci. Soc. Am. J. 44:1216-1221. Nadler, A., S. Dasberg, and I. Lapid. 1991. Time domain reflectometry measurements of water content and electrical conductivity of layered soil columns. Soil Sci. Soc. Am. J. 55:938-943. Peacock, C.H., A.E. Dudeck, and J.C. Wildmon. 1993. Physiological and growth responses of St. Augustine cultivars to salinity. HortScience. 28(1):46-48. Penman, H.L. 1948. Natural evaporation from open water, bare soil and grass. Roy. Soc. London, Proc. Ser. A. 193:120-146. Persson, M. 1997. Soil solution electrical conductivity measurements under transient conditions using time domain reflectometry. Soil Sci. Soc. Am. J. 61:997-1003. Qian, Y.L., J.D. Fry, S.C. Wiest, and W.S. Upham. 1996. Estimating turfgrass evapotranspiration using atmometers and the Penman-Monteith model. Crop Sci. 36:699-704. Rhoades, J.D. 1990. Determining soil salinity from measurements of electrical conductivity. Commun. In soil Sci. Plant Anal., 21 (13-16) 1887-1926. 203 Rhoades, J.D., N.A. Manteghi, P.J. Shouse, and W.J. Alves. 1989. Soil electrical conductivity and soil salinity: New formulations and calibrations. Soil Sci. Soc. Am. J. 53:433-439. Rhoades, J.D., P.A.C. Raats, and R.J. Prather. 1976. Effects of liquid-phase electrical conductivity, volumetric water content, and surface conductivity on bulk soil EC. Soil Sci. Soc. Am. 40:651-655. Rogowski, A.S. and E.L. Jacoby, Jr. 1977. Assessment of water loss patterns with microlysimeters. Agron. J. 69:419-424. Rosenberg, N.J. 1974. Microclimate: The biological environment. John Wiley and Sons, Inc. Shearman, R.C. 1986. Kentucky bluegrass cultivar evapotranspiration rates. HortScience 21: 455-457. Shearman, R.C. and J.B. Beard. 1973. Environmental and cultural pre-conditioning effects on water use ratio of Agrostis palustris Huds. cultivar `Pencross'. Crop Sci. 13:424427. Singer, M.J., and D.N. Munns.. 1991. Soils: An introduction. Second Edition. Macmillan Publishing Company. Tanner, C.B. 1967. Measurement of ET: In Irrigation of Agricultural lands, R.H. Hagan, H.R. Haise and T.W. Edminster, eds. Agron. Monogr. 11:534-545. Am. Soc. of Agron., Inc., Madison, Wis. Topp, G.C., J.L. Davis, and A.P. Annan. 1980. Electromagnetic determination of soil water content: Measurements in coaxial transmission lines. Water Resources Res. 16:574-582. Topp, G.C., J.L. Davis, and A.P. Arman. 1982. Electromagnetic determination of soil water content using time domain reflectometry: Applications to wetting fronts and steep gradients. Soil Sci. Soc. Am. J. 46: 672-678. Topp, G.C., M. Yanuka, W.D. Zebchuk, and S. Zegelin. 1988. Determination of electrical conductivity using time domain reflectometry: Soil and water experiments in coaxial lines. Water Resour. Res. 24(7) 945-952. Tovey, R., J.S. Spencer, and D.C. Muckel. 1969. Turfgrass evapotranspiration. Agron. J. 61:863-867. 204 Turgeon, A.J. 1996. Turfgrass management, Fourth Edition. Prentice-Hall, Inc., Princeton, New Jersey. Van Bavel, c.h.M. 1966. Potential evaporation: the combination concept and its experimental verification. Water Resour. Res. 2(3) 455-468. Van Bavel, C.H.M., N. Underwood, and R.W. Swanson,. 1956. Soil moisture measurement by neutron moderation. Soil Sci. 82:29-41. Ward, A.L., R.G. Kachanoski, and D.E. Elrick. 1994. Laboratory measurements of solute transport using time domain reflectometry. Soil Sci. Soc. Am. J. 58:1031-1039. Yanuka, M., G.C. Topp, S. Zegelin, and W.D. Zebchulc. 1988. Multiple reflection and attenuation of time domain reflectometry pulses: Theoretical considerations for applications to soil water. Water Resour. Res. 24(7):939-944. Young, M.H., P.J. Wierenga, and C.F. Mancino. 1996. Large weighing lysimeters for water use and deep percolation studies. Soil Sci. 161:491-501. Zegelin, S.J., I. White, and D.R. Jenkins. 1989. Improved field probes for soil water content and electrical conductivity measurement using time domain reflectometry. Water Resources. Res. 25:112367-2376.
* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project