Evaluation and Demonstration of Evapotranspiration-Based Irrigation Controllers FDACS Contract UF Project

Transcription

1 Evaluation and Demonstration of Evapotranspiration-Based Irrigation Controllers FDACS Contract UF Project Phase II Year 4 Task 2 and 3 Final Report By Michael D. Dukes 1 Faculty Investigator Sydney Park Brown 2 Faculty Co-Investigator Amy L. Shober 3 Faculty Co-Investigator Cecilia Romero 1 Post Doctoral Investigator And Stacia Davis 1 Graduate Student Investigator Department 1 Environmental Horticulture Department 2 Soil and Water Science Department 3 Institute of Food and Agricultural Sciences Submitted to Florida Department of Agriculture and Consumer Services February 22, 211

2 Year 4: Final Report Page 2 of 43 Table of Contents Executive Summary... 3 Introduction... 4 Materials and Methods... 4 Treatment Groups Summary... 4 ET Controller Installation... 5 AMR Evaluation... 5 Weather Stations... 5 Theoretical Irrigation Requirement... 6 Data Analysis... 7 Results and Discussion... 7 Weather Stations and Data Collection... 7 Evapotranspiration... 7 Rainfall... 8 Irrigation Water Use... 8 Apollo Beach... 8 Riverview... 9 Valrico... 1 Turfgrass Quality Conclusions References Appendix A Appendix B Appendix C Appendix D Appendix E Appendix F... 43

3 Year 4: Final Report Page 3 of 43 Executive Summary Due to limited water resources, new methods must be explored to reduce water use in irrigated urban landscapes. Phase II of this study involved testing one brand of ET controller in actual residential settings to determine if potential water savings can be achieved under real world conditions. Thirty-six cooperators were selected for this study and were located throughout Apollo Beach, Riverview, and Valrico areas. Cooperators were separated into two treatment groups: 21 cooperators were outfitted with Toro Intelli-sense ET controllers (ET controller group) and the remaining 15 were asked to maintain their current irrigation practices (comparison group). A weather station was installed at each of the three treatment locations with data collection beginning on 5 November 28 for Riverview and Valrico locations and 29 January 29 for the Apollo Beach location. Weekly water meter data collection occurred from 26 September 28 through 3 January 29 until automatic meter reading devices were installed on the household water meters. Sub-hourly water meter data collection continued through 3 September 21 and is included in this report. An ET controller would produce significant savings if used by a homeowner at least as much or higher than the theoretical irrigation requirement. Accordingly, a homeowner that traditionally irrigated more than 18 inches annually should benefit from using an ET controller. However, most of the cooperators in both treatment groups only irrigated a fraction of the theoretical requirement and similarly to or less than their historical average, partially due to municipal water restrictions (4 hours only 1 d/wk) that were in place from 3 April 29 through 3 June 29. ET controller homes were exempt from day of the week restrictions, but were asked to change their irrigation start times to midnight. For all three treatment locations, both the comparison and ET controller groups irrigated less than their historical average calculated using five years of water use records (21-26). Generally, the ET controllers did not increase historical irrigation patterns, with 12% to 15% reductions compared to historical average water use in Apollo Beach and Riverview areas, respectively, even though well-watered conditions were maintained when evapotranspiration was high and rainfall was low. In Valrico, ET controllers did not produce water savings, but only increased water use by 19% and maintained turfgrass quality ratings well above minimum acceptability. The comparison group irrigated between 2% and 36% less than their own historical average and many in this group irrigated significantly less than the theoretical irrigation requirement. For some of these cooperators, turfgrass quality suffered as a result with one cooperator requiring complete sod replacement. Between the treatment groups, the ET controller group showed increased average irrigation applications in Riverview and Valrico, averaging 189% and 19% more than the comparison group. Conversely, the ET controller group had 25% water savings in the Apollo Beach area. Irrigation application at each treatment location was independently normalized by the historical average and the theoretical requirement. Using the historical average, ET controllers increased irrigation application at all three locations: 4% in Apollo Beach, 22% in Riverview, and 48% in Valrico. The ET controller group applied more irrigation at two locations when normalizing by the theoretical irrigation requirement with increases of 141% and 16% in Riverview and Valrico, respectively. In contrast, the third location had reduced irrigation application of 12% by the ET controller group after normalization by the theoretical requirement. Due to reductions in irrigation of the comparison group, turfgrass quality generally suffered as a result. It is possible that the perception of the drought and increased watering

4 Year 4: Final Report Page 4 of 43 restrictions during spring 29 could have promoted deficit irrigation practices and a shift in priorities from good landscape quality to reducing irrigation. Introduction New methods for outdoor water conservation are needed due to increasingly limited water resources. Over half of total fresh water is used for irrigation in the United States (Hutson et al. 24). In one study, it was found that 64% of residential water use was used for irrigation in Central Florida (Haley et al. 27). As a result, new methods must be explored to reduce water use in irrigated urban landscapes. Phase I of this study determined that the Toro Intelli-sense ET controller balanced irrigation control with plant water requirements - applying the least water while still maintaining acceptable turfgrass quality. Phase II of this study tested this ET controller in actual residential settings to determine if potential water savings and turfgrass quality could be achieved under real world conditions. Homeowners were identified and recruited from utility records dating from 21 to 26 as moderate to high water users, located in the upper 5 th percentile of the county potable water meter records. During this historical period, the selected cooperators averaged 28 mm (8 inches) to 1,698 mm (67 inches) annually. The process used to identify the high water users can be found in the Year 2 Task 4 report. Homeowners were recruited from January through March 28; this process was described in the Year 3 Task 1 report. Of the 58 homes that received on-site evaluations, 36 homes remain in the study. These homes were identified and separated into treatment groups as described in the Year 3 Task 4 report. Homeowners were separated into two groups where 21 homes received ET controllers and rain sensors while the remaining 15 homes maintained control over their pre-existing irrigation timer. Initial data collection for the study occurred from 26 September 28 through 16 January 29 and was presented in the Year 3 Task 6 report. Continued data collection through 26 November 29 can be found in the Year 4 Task 1 interim report. This final report contains all data collected from 26 September 28 through 3 September 21 and satisfies the Year 4 Tasks 2 and 3 requirements as described in the scope of work. Materials and Methods Treatment Groups Summary Treatment locations were selected by determining the communities in Hillsborough County that averaged the highest historical outdoor water use. This analysis led to the selection of the following communities: Apollo Beach, Riverview, and Valrico. Homeowners in each community were solicited to join the study as cooperators and many of the respondents received free irrigation evaluations to determine if they were viable candidates. However, only 36 homes were selected from the three locations: 13 from Apollo Beach, 8 from Riverview, and 15 from Valrico. Cooperators were separated into one of two treatment groups for each location where one group received an ET controller and rain sensor while the other group was asked to maintain the normal irrigation practices. There were 21 cooperators in the ET controller group: 7 in Apollo Beach, 5 in Riverview, and 9 in Valrico. The remaining cooperators were selected for the comparison group.

5 Year 4: Final Report Page 5 of 43 ET Controller Installation Toro Intelli-sense controllers were installed at each of the 21 homes selected as part of the ET controller group. Each controller was installed by a professional landscape contractor over a three week period ending 5 August 28. Every controller was programmed by project personnel on the day of installation using zone-specific landscape characteristics recorded during the irrigation evaluation. Cooperators were given basic instructions on viewing the irrigation schedule and changes they could make to adjust the irrigation amount. Follow-ups occurred only if the cooperator contacted the research group for help. AMR Evaluation Automatic meter reading devices (AMRs) were installed on the potable water meters (3 homes), or irrigation meters (6 homes) if available, to monitor water use at sub-daily intervals. Installation occurred from 3 January 29 to 2 February 29. Hourly intervals were originally chosen as the time step. However, it was later changed to fifteen minute intervals to help distinguish indoor from outdoor use. Irrigation water use was determined from the total household water use by removing any water that was less than the smallest application rate determined during the irrigation evaluation. For example, a homeowner who has four irrigation zones could have application rates ranging from 6.5 gpm to 12.9 gpm. It is unlikely that a homeowner would have an indoor appliance that uses more than 6.5 gpm for a constant 15 minutes (a total of 97.5 gallons). Therefore, any water use less than the threshold of 97.5 gallons for every 15 minute period was removed with the assumption that anything more than this amount was irrigation. It is possible that threshold volume could vary for many reasons. For example, the application rate determined during the irrigation evaluation could be different than the normal application rate. This would occur due to leaks and clogs found during the evaluation that were later repaired or pressure differences between peak and off-peak times. As a result, water use for each cooperator was adjusted on an individual basis by a factor between.5 and 1.5 (5% to 15% of threshold). The adjustment to the threshold occurred based on indications of irrigation. When many consecutive large volumes of water occurred, assumed to be irrigation, with a few 15 minute periods under the threshold, then the threshold was lowered. Conversely, if independent instances of large water use occurred, the threshold level was increased without removing obvious irrigation events. The average per capita indoor water use ranged from 2 gal/day (W1426) to 228 gal/day (W632), averaging 63 gal/day, which is slightly less than the average found in a United States study covering twelve cities (Mayer et al. 1999). Indoor water use was determined daily by subtracting outdoor water use from total household water use. This indoor water use was averaged into one monthly average for each cooperator. The average indoor water use value was then used to remove indoor water use from total household water use to determine the historical average outdoor water use and the outdoor water use during the period of manual data collection before the AMRs were installed (26 September 28 through 16 January 29). Weather Stations Weather stations were installed in the vicinity of each study location to provide data for determination of the theoretical irrigation water requirement. The Riverview weather station was installed at the Riverview Park and Civic Center in Riverview, FL and the Valrico weather station was installed at the Valrico Wastewater Treatment Facility in Valrico, FL. Both the Riverview and Valrico weather stations were installed on 5 November 28. The Apollo Beach

6 Year 4: Final Report Page 6 of 43 weather station was installed on 29 January, 29 at the Tampa Electric Company Manatee Viewing Center. This weather station was moved on 31 August 21 to the Apollo Beach Fire Station located on Golf and Sea Blvd, Apollo Beach, FL. Weather stations require open, irrigated areas that are difficult to find in urban areas. Due to the extended range between the weather stations and the cooperators, additional rain gauges were installed in Apollo Beach and Valrico. These rain gauges allow better approximations of localized rainfall events for certain cooperators. Theoretical Irrigation Requirement An ideal soil water balance, referred to as the theoretical irrigation requirement, was calculated for the same study period using procedures similar to the SWAT protocol (Irrigation Association 28). The daily soil water balance was calculated by the following equation: MB i =MB i-1 -ET C +R E +I E [1] where MB is the soil water content (inches) on day i or i-1, ET C is the plant specific evapotranspiration (inches) on day i, R E is effective rainfall (inches) on day i, and I E is effective irrigation (inches) on day i. MB represents the water storage level in the soil profile on any given day. This level fluctuates from zero to a maximum value, termed root zone working water storage (RZWWS), determined from available water capacity for each cooperator (Table 1). The RZWWS is equivalent to readily available water in conventional irrigation terminology. Irrigation was scheduled to occur when the MB fell at or below zero and the total daily depth applied was equal to the amount of water required to refill the zone to RZWWS. Reference evapotranspiration (ET O ) was calculated daily using the weather data by the ASCE-EWRI standardized reference evapotranspiration equation (ASCE-EWRI 25). Daily ET C was then calculated from ET O by applying a crop coefficient (K C ) using the following equation (ASCE-EWRI 25): ET C =K C *ET O [2] Turfgrass K C values chosen for the theoretical soil water balance model were interpolated between north-central and south Florida warm season turfgrass values since the study location is in between these two regions (Jia et al. 29; Table 2). Net rainfall is defined as the depth of total rainfall that enters the soil profile and is considered available for plant use. It was assumed that daily net rainfall is 8% of total daily rainfall (Irrigation Association, 28). Effective rainfall is calculated from net rainfall and is the depth that causes the MB to reach RZWWS after ET C has been lost for the day. Excess rainfall (that which exceeds soil storage capacity) is assumed lost due to surface runoff or deep percolation. The theoretical irrigation requirement was calculated for individual cooperator sites based on the available water capacity determined from the dominant soil type and the efficiency determined from on-site uniformity testing results. The dominant soil type was found using the web soil survey (Soil Survey Staff 29) and average available water capacity was listed for each soil type as inches of water per sixty inches of soil (Table 1). The available water capacity for the root zone of each cooperator was calculated assuming a root zone depth of six inches to

7 Year 4: Final Report Page 7 of 43 correspond with the ET controller settings. The RZWWS was then calculated as 5% of the root zone available water capacity assuming this depth could be depleted while maintaining wellwatered conditions. Data Analysis Monthly irrigation application compared between treatments was performed using Statistical Analysis Software (SAS, Cary, NC). The glimmix procedure was chosen to calculate the least squares means between treatments at each location to determine if irrigation application amounts were statistically different at a 95% confidence level. Irrigation application was not distributed normally and required a log transformation for statistical analysis. Multiple comparisons can be made from the irrigation data collected over the study period. Irrigation application that actually occurred can be compared between treatments, to historical averages, and the theoretical irrigation requirement. However, direct comparisons may be misleading due to large differences in cumulative totals. For example, 2% reduction from 1,5 mm would be much more significant compared to a 2% reduction from 5 mm. To alleviate this problem and to provide prospective to the results, there is an additional comparison of irrigation application between treatments where cumulative irrigation was normalized by the historical average as well as the theoretical requirement. Normalization was calculated as: [3] where I NH is the normalized irrigation amount (mm) based on the historical average, I A is the actual irrigation applied (mm), and I H is the historical average irrigation (mm). Additionally, normalization based on the theoretical irrigation requirement was also be performed using the following equation: [4] where I NT is the normalized irrigation amount (mm) based on the theoretical irrigation requirement, I A is the actual irrigation applied (mm), and I TIR is the theoretical irrigation requirement (mm). Results and Discussion Weather Stations and Data Collection The weather data represent the time period of 6 November 28 through 3 September 21 for the Riverview and Valrico weather stations and 3 January 29 through 3 September 21 for the Apollo Beach weather station. Weather data collection includes the following parameters: solar radiation, relative humidity, temperature, wind speed, and rainfall. The figures used for quality assurance are located in Appendix A, B, and C for Apollo Beach, Riverview, and Valrico, respectively. Evapotranspiration The Apollo Beach and Valrico locations had similar ET O totals of 2,795 mm (11 inches) and 2,817 mm (111 inches), respectively, for a period of twenty months (Fig. 1). The Riverview

8 Year 4: Final Report Page 8 of 43 weather station had slightly less ET O, totaling 2,568 mm (11 inches). These totals indicate that there was higher ET than an average Florida year, which can total 122 cm (48 inches) over twelve months in the northern portion of south Florida (Abtew 23). Rainfall Rainfall totaled 2,413 mm (95 inches) at the Apollo Beach weather station, 1,946 mm (77 inches) at the Riverview weather station, and 2,43 mm (96 inches) at the Valrico weather station (Fig. 2). The rainfall total for the Valrico weather station was unreasonably low from June 29 through November 29 where the rain gauge was obstructed and not able to read rainfall depths accurately. During this period, rainfall measured using the alternate rain gauge was substituted at this location. Otherwise, the rainfall at all locations tracked relatively well to each other indicating that there were relatively few localized rainfall events over the study period. Rainfall totals illustrate four distinct periods of weather where irrigation scheduling techniques would change. The rainfall from February 29 until mid-may 29 totaled approximately 16 mm (4.2 inches) indicating that the spring season experienced extremely dry conditions where irrigation would be needed. Additionally, there was another dry period from mid-september 29 through mid-march 21 that was also considered dry conditions, totaling 24 mm (9.4 inches). The periods of the study period from mid-may 29 through mid- September 29 and mid-march 21 through September 21 had much more frequent rainfall amounts totaling 992 mm (39. inches) and 925 mm (36.4 inches), respectively. Irrigation scheduling becomes complicated when trying to incorporate rainfall due to the unpredictability of forecasting and the timing of events for an accurate daily soil water balance. Over irrigation would most likely occur during periods of frequent rainfall. Irrigation Water Use Cooperator irrigation water use was compared to their average historical outdoor water use and their theoretical irrigation requirement to determine changes in irrigation application based on treatment differences or variations from irrigation needs. Apollo Beach Historically in Apollo Beach, the comparison group irrigated more than the ET controller group (Fig. 3). Comparison homes generally irrigated as they normally would for the first part of the study period, but significantly decreased irrigation application beginning in October 29, totaling 1,158 mm (45.6 inches) over 24 months. The ET controller group, irrigating 865 mm (34. inches), applied 25% less than the comparison group but was not significantly different (Pvalue =.4532). The ET controller homes irrigated 4% more than the comparison homes when normalized for historical irrigation trends (Table 4). The ET controller group generally responded to weather conditions in the beginning of the study with low irrigation application during the winter months when irrigation demand was low and increased irrigation application in spring 29 to accommodate increased demand due to high ET and little rainfall. Increased water savings during winter months compared to other seasons is consistent with results found from the Phase I field study (Davis et al. 29). Though the difference is not as distinct, there was also a decrease in irrigation application during winter 21. The comparison group irrigated similarly to their historical trend at the beginning of the study and actually increased their monthly irrigation application above their historical average

9 Year 4: Final Report Page 9 of 43 for spring 29 when there was a high climatic demand. However, the comparison homes decreased their irrigation application more than the ET controller homes during the rainy summer 29 and continued to drastically decrease their irrigation application through winter and spring 21 thus resulting in a large amount of irrigation savings by this treatment group. During the hot and dry spring 29, the ET controllers were able to maintain similar irrigation application to the historical trend. As a result, irrigation savings during the winter months as a result of the ET controllers created an overall reduction compared to the historical average. According to the average theoretical irrigation requirement, the comparison group required more irrigation throughout the spring and summer months than the ET controller group (Fig. 4). However, the cumulative irrigation requirement totals were similar between the treatments. The comparison homes over-irrigated in the winter and spring 29 months compared to what was required, but quickly reduced irrigation to apply 34% less than the theoretical requirement over the 2-month period. The ET controller group also applied more irrigation than what was required, but to a lesser extent, with a reduction of 41% compared to the theoretical requirement. Normalizing by the theoretical requirement showed that the ET controller group was able to save 12% more than the comparison group (Table 4). The theoretical irrigation requirement for cooperators in the Apollo Beach comparison group ranged from 856 mm (34 inches) to 1,724 mm (68 inches), cumulatively. In this treatment group, three cooperators historically irrigated almost twice as much as the calculated theoretical landscape needs (Fig. 5). Two out of those three cooperators, W639 and W6345, decreased their irrigation application even more by irrigating 61% and 16%, respectively, less than the theoretical irrigation requirement while the third cooperator, W721, had some overirrigation at the beginning of the study but generally followed the theoretical irrigation requirement. One cooperator, W62, decreased their irrigation application from both the theoretical requirement and their historical average by 53% and 34%, respectively. The remaining two cooperators for the comparison group were theoretically required to apply more than what they historically applied and increased their actual irrigation application during the study by 25% to 3%. The theoretical requirement for the ET controller group in Apollo Beach ranged from 77 mm (3 inches) to 1,531 mm (6 inches) (Fig. 6). The ET controller applied 41% to 45% less than the theoretical requirement for all cases except one where there was only a 12% decrease. There were two cooperators, W6434 and W712, which historically irrigated similarly to the theoretical requirement, where the ET controller reduced irrigation application by 22% and 35%. Despite overall reductions compared to the theoretical requirement, irrigation application was increased from the historical average by 213% for W6432 indicating that ET controllers may not be suitable for residences that already implement deficit irrigation practices. Riverview The Riverview comparison group was comprised of low historical irrigators. These homes irrigated less than their historical trends by 3%, totaling 295 mm (11.6 inches) over the 24-month period (Fig. 7). The ET controller group maintained irrigation application totals slightly less (15%) than the historical average. The ET controller group historically irrigated more than the comparison group, so it was not surprising to see such high irrigation application for the study period by the ET controller group, totaling 851 mm (33.5 inches) and 189% more than the comparison group. Irrigation application between treatment groups was large enough that they were considered significantly different (P-value =.388). Even though the ET

10 Year 4: Final Report Page 1 of 43 controller group applied so much more irrigation, normalizing irrigation application by the historical average showed that the ET controller resulted in 22% more irrigation compared to historical trends (Table 4). Similar to the Apollo Beach results, the ET controller reduced irrigation during both winter periods within the study. Most of the irrigation application occurred during spring and summer months when irrigation demand is high. Both groups of homes had the same trend in the theoretical irrigation requirement where winter irrigation application is low and spring irrigation application is high (Fig. 8). Comparing directly to the theoretical irrigation requirement, the comparison group reduced irrigation application by 8% whereas the ET controller group reduced irrigation application by 52%. Normalizing by the theoretical requirement indicated that the ET controller group applied 141% more than the comparison group (Table 4). The theoretical irrigation requirement for the comparison group ranged from 1,17 mm (4 inches) to 1,687 mm (66 inches) (Fig. 9). Only W1122 historically irrigated less than the theoretical irrigation requirement whereas the other two cooperators had historical irrigation application similar to the theoretical irrigation requirement. However, all three homes irrigated much less than both the theoretical irrigation requirement and the historical average with total irrigation application ranging from 165 mm (7 inches) to 383 mm (15 inches) over a 23 month testing period. Only W1426 showed turfgrass quality ratings that were negatively affected by low levels of irrigation use. Historically, these cooperators increased irrigation during the spring season due to high ET and less rainfall. However, during this study, they generally did not increase their irrigation water use during any season. It is likely that these cooperators chose to irrigate less than normal due to the drought pressure and enforcement of watering restrictions during this period. There was a wide range of average historical water use by the ET controller group. Two cooperators, W11243 and W11518, historically irrigated similarly to the theoretical irrigation requirement (Fig. 1). Both of these cooperators irrigated much less than the historical average by using an ET controller with reductions of 77% and 66% for W11243 and W11518, respectively. One cooperator, W11528, was able to irrigate similar to the historical average, applying 621 mm (24 inches) over 23 months. The remaining two cooperators irrigated more than their historical average, totaling 1,129 mm (44 inches) and 1,118 mm (44 inches) for W6919 and W7434, respectively. These two cooperators most likely applied more irrigation than their historical average due to the ET controllers maintaining well-watered conditions when these cooperators allowed deficit conditions to exist prior to ET controller installation. Valrico Cumulative water use by the comparison group in Valrico, totaling 785 mm (31 inches), was 36% less than the historical average (Fig. 11). Irrigation generally did not vary from season to season indicating that this group did not update their irrigation schedules regularly. Similarly to the other ET controller groups, the Valrico ET controller group irrigated less than the historical average during the winter months, but increased water application for the hot and dry spring season with a cumulative total of 935 mm (37 inches). The ET controller group irrigated 19% more than the comparison group because of its response to spring weather conditions, but the difference was not large enough to be considered significantly different (P-value =.223). When normalizing by the historical average, the ET controller group applied 48% more irrigation than the comparison group (Table 4).

11 Year 4: Final Report Page 11 of 43 The theoretical irrigation requirements were similar between treatments indicating that the cooperators in these groups are more evenly matched in terms of landscape conditions (Fig. 12). Both treatments irrigated less than the theoretical requirement by 57% and 5% for the comparison and ET controller groups, respectively. When normalizing by the theoretical requirement, the ET controller group applied 16% more than the comparison group (Table 4). The comparison group varied in average historical irrigation application from 821 mm (32 inches) to 2,119 mm (83 inches) (Fig. 13). W115 historically irrigated only 29% of the theoretical requirement, but still managed to reduce irrigation application to 4% of the historical average. W152 and W1614 had historical averages similar to the theoretical requirement and irrigation application generally followed the same trends. W151 applied the same amount as the historical average, but applied 76% less than the theoretical requirement. Both W1949 and W221 historically irrigated more than the theoretical irrigation requirement, but only applied 2% and 32% of the theoretical landscape needs. Thus, these cooperators had significant water savings at the expense of their landscape quality with ratings falling below minimum acceptability. The cumulative theoretical irrigation requirement for the ET controller group ranged from 962 mm (38 inches) to 2,329 mm (92 inches) over the study period (Fig. 14). Irrigation application by W2212 followed a similar trend to the historical water use, but still applied 77% more than the theoretical irrigation requirement. Other than this cooperator, all cooperators in the ET controller group applied significantly less than the theoretical requirement. W118 was the only one in the ET controller group to have significant savings compared to both the historical average and the theoretical requirement, saving 81% of irrigation water use. W157 historically irrigated similarly to the theoretical requirement, but reduced actual water use by 36%. All other cooperators irrigated similarly to or slightly less than the historical average. Significant savings could not occur because the cooperators were already irrigating much less than the theoretical requirement prior to the study. Turfgrass Quality Turfgrass quality ratings were performed at least once per season since the beginning of the study beginning with the irrigation evaluations in spring 28. Turfgrass quality ratings from spring 28 through winter were presented in the Year 3 Task 6 Report. Average turfgrass quality ratings for both treatments in Apollo Beach were at least the minimum acceptable quality rating of 5 with the comparison group averaging a 6 (Table 5). Cooperators in the comparison group generally maintained their turfgrass quality over time. There was some decline in quality for a few cooperators in the ET controller group in fall 29. This occurred similarly during the previous fall and turfgrass quality rebounded in subsequent seasons without problems. One known cooperator saw a decline in turfgrass quality, but was able to prove that the lawn maintenance service was at fault and the decline was not related to irrigation (i.e. ET controller). Overall average turfgrass quality for Riverview averaged a 6 rating for the comparison group and a 7 rating for the ET controller group in 29 (Table 6). In 21, both treatments declined in turfgrass quality by 1 score rating to 5 for the comparison group and 6 for the ET controller group. Cooperators in the comparison group seemed to have trouble maintaining turfgrass quality during the winter-time that is as high as during the summer months. This could be due to the turfgrass entering the winter season under stressful conditions from lack of irrigation causing a slow green-up in the spring. The ET controller group maintained consistent ratings throughout the study period with one cooperator increasing over time from an

12 Year 4: Final Report Page 12 of 43 unacceptable rating to acceptable in both 29 and 21. Thus, the ET controller group showed a trend for increased turf quality as a result of more irrigation. The cooperators in Valrico had turfgrass quality ratings averaging a 5 for the comparison treatment and a 7 for the ET controller treatment in 29 and averaging a 6 for the comparison treatment and a 7 for the ET controller treatment in 21 (Table 7). Both treatments generally maintained their landscape quality ratings over both years with some fluctuations. The W221 cooperator in the comparison group steadily declined from an almost perfect turfgrass quality rating to below minimally acceptable, but increased to above acceptable quality again by fall 21. Cooperators in the ET controller group rarely approached the minimally acceptable mark with only a few instances of quality ratings less than 5. It appears that the ET controlled irrigation increased turfgrass quality at two of the three study locations. Photos of the treatment homes on the rating day representing fall 21 are located in Appendix D, E, and F for the Apollo Beach, Riverview, and Valrico locations, respectively. There may be a correlation between drought conditions and the irrigation practices of the comparison cooperators. During the period where most homeowners would increase their time clock runtimes for high evapotranspiration demand and little rainfall, more stringent watering restrictions were enacted and the neighboring City of Tampa banned all automatic irrigation. These extreme actions taken due to the water shortage situation that was heavily advertised on the local and sometimes national news may have influenced homeowner behavior and possibly allowing landscape quality to decline to save water. Additionally, the Southwest Florida Water Management District has run an educational campaign over the course of this study period, including television and radio advertisements, to encourage outdoor water conservation. Conclusions On average, for all three treatment locations, both the comparison and ET controller groups irrigated at or less than historical averages despite the large irrigation demand due to high evapotranspiration and little rainfall during most of the study. Most cooperators irrigated less than the theoretical irrigation requirement as well as their historical trend. Thus according to the overall averages, the ET controllers did not increase outdoor water use compared to the historical and theoretical irrigation totals, even though they are designed to maintain well-watered conditions. Fourteen of the 21 cooperators using ET controllers showed some water savings compared both to the theoretical requirement and historical averages. Out of those 14 cooperators, 7 of them had significant reductions. Based on the historical averages of these 14 cooperators, an ET controller should produce significant savings if used by a homeowner traditionally irrigating more than 457 mm (18 inches) per year in this region. According to the results of this study, if the cooperators historically irrigated similarly to the theoretical irrigation requirement then irrigation savings would have been substantial. However, most of the cooperators in this study only irrigated a small portion of the theoretical requirement on average. The comparison group generally irrigated much less than their historical average and the theoretical irrigation requirement. It is possible that the perception of the drought and increased watering restrictions during spring 29 could have promoted deficit irrigation practices and a shift in priorities from perfect landscape quality to reducing irrigation and expense. ET controllers are designed to maintain well-watered conditions to ensure proper landscape quality and health. However, landscape quality is a subjective measurement and not

13 Year 4: Final Report Page 13 of 43 always important for homeowners who prioritize other factors such as water costs or resource management (e.g. saving water while in a drought). References Abtew, W. (23). Evapotranspiration Estimation for South Florida. P. Bizier and P. A. DeBarry (eds.). Proceedings of the World Water and Environmental Resources Congress 23 and Related Symposia, ASCE. ASCE EWRI (25). The ASCE Standardized Reference Evapotranspiration Equation. Technical Committee report to the Environmental andwater Resources Institute of the American Society of Civil Engineers from the Task Committee on Standardization of Reference Evapotranspiration. ASCE EWRI, 181 Alexander Bell Drive, Reston, VA , 173 pp. Davis, S.L., Dukes, M.D., and Miller, G.L. (29). Landscape irrigation by evapotranspirationbased irrigation controllers under dry conditions in Southwest Florida. Agricultural Water Management 96(29): Haley, M.B., Dukes, M.D., and Miller, G.L. (27). Residential irrigation water use in Central Florida. Journal of Irrigation and Drainage Engineering 133(5): Hutson, S. S., Barber, N. L., Kenny, J. F., Linsey, K. S., Lumia, D. S., and Maupin, M. A. (24). Estimated use of water in the United States in 2. United States Geological Survey Circular p. Irrigation Association [IA] (28). Smart Water Application Technology (SWAT) Turf and Landscape Irrigation Equipment Testing Protocol for Climatologically Based Controllers: 8 th Testing Protocol. Irrigation Association, Falls Church, VA. Available at: Accessed 2 September 28. Jia, X., Dukes, M.D., and Jacobs, J.M. (29). Bahiagrass crop coefficients from eddy correlation measurements in central Florida. Journal of Irrigation Science 28(1): Mayer, P.W., DeOreo, W.B., Opitz, E.M., Kiefer, J.C., Davis, W.Y., Dziegielewski, B., and Nelson, J.O. (1999). Residential end uses of water. American Water Works Association Research Foundation. Denver, CO. Soil Survey Staff, Natural Resources Conservation Service, United States Department of Agriculture. Web Soil Survey. Available at: Accessed on 28 December 29.

14 Year 4: Final Report Page 14 of 43 Table 1. Cooperator-specific data used to calculate the theoretical irrigation requirement. Location Treatment Group Cooperator Code Available Water Holding Capacity 1 Readily Available Water 2 Efficiency 3 Fraction Turfgrass 4 (in./6 in.) (in.) (%) (m 2 /m 2 ) Apollo Beach Comparison W Apollo Beach Comparison W Apollo Beach Comparison W Apollo Beach Comparison W Apollo Beach Comparison W Apollo Beach Comparison W Apollo Beach ET W Apollo Beach ET W Apollo Beach ET W Apollo Beach ET W Apollo Beach ET W Apollo Beach ET W Apollo Beach ET W Riverview Comparison W Riverview Comparison W Riverview Comparison W Riverview ET W Riverview ET W Riverview ET W Riverview ET W Riverview ET W Valrico Comparison W Valrico Comparison W Valrico Comparison W Valrico Comparison W Valrico Comparison W Valrico Comparison W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Valrico ET W Available water holding capacity (AWHC) was selected using the dominant soil type for each home from the Hillsborough County Soil Survey. 2 Readily available water (RAW) is defined as the depth of water to be depleted before irrigation was required and was calculated as RAW = AWHC * RZ * MAD where RZ is the root zone (6 inches) and MAD is the maximum allowable depletion (5%). 3 Efficiency values were determined from on-site uniformity testing of 1 zone during the irrigation evaluation and calculated using low half distribution uniformity. Additionally, these same efficiency values were used during ET controller programming. 4 It was assumed that the theoretical requirement is the total irrigation necessary if the entire landscaped area was turfgrass. However, it was assumed that ornamentals were established and required no irrigation so only the percentage

15 Year 4: Final Report Page 15 of 43 of turfgrass would require irrigation. Table 2. Crop coefficients were selected for South Florida (Jia et al 29) and used to adjust daily reference evapotranspiration in the theoretical model. Month Crop Coefficient January.45 February.45 March.65 April.8 May.9 June.75 July.7 August.7 September.75 October.7 November.6 December.45

16 Year 4: Final Report Page 16 of 43 Table 3. Monthly rainfall totals for weather stations and rain gauges. Month Apollo Beach Riverview Valrico Nov 28 NA* Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep *The total rainfall for the month was not available due to equipment malfunction or the month was prior to equipment installation.

17 Year 4: Final Report Page 17 of 43 Table 4. Comparison of irrigation application for each treatment location to the historical average and the theoretical requirement. Different letters indicate a significant difference at the 95% confidence level. Treatment Irrigation Application 1,2 (mm) Historical Average (mm) Normalized Irrigation, I NH (mm/mm) Irrigation Application 3 (mm) Theoretical Requirement (mm) Normalized Irrigation, I TIR (mm/mm) Apollo Beach Comparison 1,158 a 1, ,28.66 ET Controller 865 a 1, Difference (%) 4-12 Riverview Comparison 295 b ET Controller 851 a Difference (%) Valrico Comparison 785 a 1, , ET Controller 935 a Difference (%) When comparing to the historical average, irrigation application was summed from October 28 through September 21 totaling 24 months. 2 Monthly irrigation application using all months summed in the column were compared statistically to determine significant differences greater than.5. 3 When comparing to the theoretical irrigation requirement, irrigation application was summed from the installation date of the weather station through September 21, totaling 2 months in Apollo Beach and a 23 months in the other locations.

18 Year 4: Final Report Page 18 of 43 Table 5. Turfgrass quality ratings for Apollo Beach, FL. Homeowner Treatment Spring Summer Fall Winter Average 29 W6345 Comparison NA 1 NA NA NA NA W62 Comparison W1333 Comparison W6333 Comparison W721 Comparison W639 Comparison Comparison Group W6432 ET W627 ET W6619 ET W632 ET W712 ET W6434 ET W45 ET ET Group W6345 Comparison NA NA NA -- 3 NA W62 Comparison W1333 Comparison W6333 Comparison W721 Comparison W639 Comparison Comparison Group W6432 ET W627 ET W6619 ET W632 ET W712 ET W6434 ET W45 ET ET Group NA is an abbreviation for not applicable and was used for a homeowner who does not have turfgrass viewable from the road. 2 Turfgrass was replaced by cooperator for reasons unrelated to the study. 3 Date of quality rating is outside of the scope of this report.

19 Year 4: Final Report Page 19 of 43 Table 6. Turfgrass quality ratings for Riverview, FL. Homeowner Treatment Spring Summer Fall Winter Average 29 W1122 Comparison W1426 Comparison W1151 Comparison Comparison Group W11518 ET W11528 ET W7434 ET W11243 ET W6919 ET ET Group W1122 Comparison W1426 Comparison W1151 Comparison Comparison Group W11518 ET W11528 ET W7434 ET W11243 ET W6919 ET ET Group Date of quality rating is outside of the scope of this report.

20 Year 4: Final Report Page 2 of 43 Table 7. Turfgrass quality ratings for Valrico, FL. Homeowner Treatment Spring Summer Fall Winter Average 29 W221 Comparison W151 Comparison W115 Comparison W152 Comparison W1614 Comparison W1949 Comparison Comparison Group W118 ET W2212 ET W157 ET W22 ET W187 ET W2215 ET W1948 ET W123 ET W121 ET ET Group W221 Comparison W151 Comparison W115 Comparison W152 Comparison W1614 Comparison W1949 Comparison Comparison Group W118 ET W2212 ET W157 ET W22 ET W187 ET W2215 ET W1948 ET W123 ET W121 ET ET Group Date of quality rating is outside of the scope of this report. 2 Cooperator re-sodded landscape. 3 Irrigation system was broken and cooperator waited many months to have it fixed causing decreased quality.

21 Year 4: Final Report Page 21 of Cumulative ETo (mm) /29 5/9 8/17 11/25 3/5 6/13 9/21 Date (29 21) Apollo Beach Riverview Valrico Figure 1. Cumulative reference evapotranspiration (ETo) for weather stations located in Apollo Beach, Riverview, and Valrico beginning when data became available from all three stations.

22 Year 4: Final Report Page 22 of Cumulative Rainfall (mm) /29 5/9 8/17 11/25 3/5 6/13 9/21 Date (29 21) Apollo Beach Riverview Valrico Figure 2. Cumulative rainfall for weather stations located in Apollo Beach, Riverview, and Valrico beginning when data became available from all three stations.

23 Year 4: Final Report Page 23 of 43 Cumulative Monthly Irrigation (mm) /1 12/3 2/4 4/8 6/1 8/12 1/14 12/16 2/17 4/21 6/23 8/25 Date (28 21) Monthly Rainfall (mm) Rainfall (mm) Comparison ET Controllers Historical Comparison Historical ET Controllers Figure 3. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly historical outdoor water use in Apollo Beach, FL.

24 Year 4: Final Report Page 24 of 43 Cumulative Monthly Irrigation (mm) /1 4/5 6/7 8/9 1/11 12/13 2/14 4/18 6/2 8/22 Date (28 21) Monthly Rainfall (mm) Rainfall (mm) Comparison ET Controllers Theoretical Comparison Theoretical ET Controllers Figure 4. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly theoretical outdoor water use in Apollo Beach, FL.

25 Year 4: Final Report Page 25 of 43 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cooperator Theoretical Cooperator Historical Average 4 2 W639 W6333 W1333 W62 Cumulative Monthly Irrigation (in) W721 W6345 1/1 4/11 7/2 1/28 2/5 5/16 8/24 1/1 4/11 7/2 1/28 2/5 5/16 8/24 Date (29-21) Date (29-21) Figure 5. Cumulative monthly irrigation for cooperators in the comparison treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Apollo Beach, FL.

26 Year 4: Final Report Page 26 of 43 Cumulative Monthly Irrigation (in) W45 W6432 Cumulative Monthly Irrigation (in) Cooperator W6434 Theoretical Cooperator Historical Average W632 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) W627 W6619 W712 1/1 4/11 7/2 1/28 2/5 5/16 8/24 Date (29-21) 1/1 4/11 7/2 1/28 2/5 5/16 8/24 Date (29-21) Figure 6. Cumulative monthly irrigation for cooperators in the ET controller treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Apollo Beach, FL.

27 Year 4: Final Report Page 27 of 43 Cumulative Monthly Irrigation (mm) /1 12/3 2/4 4/8 6/1 8/12 1/14 12/16 2/17 4/21 6/23 8/25 Date (28 21) Monthly Rainfall (mm) Rainfall (mm) Comparison ET Controllers Historical Comparison Historical ET Controllers Figure 7. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly historical outdoor water use in Riverview, FL.

28 Year 4: Final Report Page 28 of 43 Cumulative Monthly Irrigation (mm) Monthly Rainfall (mm) 2 11/1 1/3 3/7 5/9 7/11 9/12 11/14 1/16 3/2 5/22 7/24 9/25 Date (28 21) Rainfall (mm) Comparison ET Controllers Theoretical Comparison Theoretical ET Controllers Figure 8. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly theoretical outdoor water use in Riverview, FL. 35

29 Year 4: Final Report Page 29 of 43 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cooperator W1151 Theoretical Cooperator Historical Average W1426 W /1 2/9 5/2 8/28 12/6 3/16 6/24 Date (28-21) 11/1 2/9 5/2 8/28 12/6 3/16 6/24 Date (28-21) Figure 9. Cumulative monthly irrigation for cooperators in the comparison treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Riverview, FL.

30 Year 4: Final Report Page 3 of 43 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) W11243 W6919 Cooperator Theoretical 11/1 2/9 5/2 8/28 12/6 3/16 6/24 Cooperator Historical Average Date (28-21) W /1 2/9 5/2 8/28 12/6 3/16 6/24 W11528 W7434 Date (28-21) Figure 1. Cumulative monthly irrigation for cooperators in the ET controller treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Riverview, FL.

31 Year 4: Final Report Page 31 of 43 Cumulative Monthly Irrigation (mm) /1 12/3 2/4 4/8 6/1 8/12 1/14 12/16 2/17 4/21 6/23 8/25 Date (28 21) Monthly Rainfall (mm) Rainfall (mm) Comparison ET Controllers Historical Comparison Historical ET Controllers Figure 11. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly historical outdoor water use in Valrico, FL.

32 Year 4: Final Report Page 32 of 43 Cumulative Monthly Irrigation (mm) Monthly Rainfall (mm) 2 11/1 1/3 3/7 5/9 7/11 9/12 11/14 1/16 3/2 5/22 7/24 9/25 Date (28 21) Rainfall (mm) Comparison ET Controllers Theoretical Comparison Theoretical ET Controllers Figure 12. Cumulative monthly outdoor water application for each treatment group compared to the cumulative average monthly theoretical outdoor water use in Valrico, FL. 35

33 Year 4: Final Report Page 33 of 43 Cumulative Monthly Irrigation (in) W115 W1614 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cooperator Theoretical Cooperator Historical Average W152 W1949 W151 W221 11/1 2/9 5/2 8/28 12/6 3/16 6/24 11/1 2/9 5/2 8/28 12/6 3/16 6/24 Date (28-21) Date (28-21) Figure 13. Cumulative monthly irrigation for cooperators in the comparison treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Valrico, FL.

34 Year 4: Final Report Page 34 of 43 Cumulative Monthly Irrigation (in) W2215 W22 Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) Cumulative Monthly Irrigation (in) W157 W2212 Cooperator Theoretical W187 Cooperator Historical Average 2 W121 11/1 2/9 5/2 8/28 12/6 3/16 6/24 W1948 W123 W118 11/1 2/9 5/2 8/28 12/6 3/16 6/24 Date (28-21) Date (28-21) Figure 14. Cumulative monthly irrigation for cooperators in the ET controller treatment group compared to the theoretical irrigation requirement specific to the cooperator s location and landscape characteristics for Valrico, FL.

35 Year 4: Final Report Page 35 of 43 Appendix A. Data collected from the weather station located in Apollo Beach, FL. 35 Solar Radiation (MJ/m 2 /d) /27 5/7 8/15 11/23 3/3 6/11 9/19 Date (29 21) Rso=(KB+KD)Ra Rso=.75*Ra Rs Figure A-1. Solar radiation and clear-sky solar radiation 12 Relative Humidity (%) /27 5/7 8/15 11/23 3/3 6/11 9/19 Date (29 21) Maximum Relative Humidity Minimum Relative Humidity Figure A-2. Measured daily maximum and minimum relative humidity

36 Year 4: Final Report Page 36 of Temperature ( C) /27 5/7 8/15 11/23 3/3 6/11 9/19 Date (29 21) Maximum Temperature Minimum Temperature Figure A-3. Measured daily maximum and minimum temperature 16 Average Wind Speed (m/s) /27 5/7 8/15 11/23 3/3 6/11 9/19 Date (29 21) Average Wind speed Maximum Wind speed Figure A-4. Measured daily maximum and average wind speed

37 Year 4: Final Report Page 37 of 43 Appendix B. Data collected from the weather station located in Riverview, FL. 35 Solar Radiation (MJ/m 2 /d) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Rso=(KB+KD)Ra Rso=.75*Ra Rs Figure B-1. Solar radiation and clear-sky solar radiation 12 Relative Humidity (%) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Maximum Relative Humidity Minimum Relative Humidity Figure B-2. Measured daily maximum and minimum relative humidity

38 Year 4: Final Report Page 38 of Temperature ( C) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Maximum Temperature Minimum Temperature Figure B-3. Measured daily maximum and minimum temperature 16 Average Wind Speed (m/s) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Average Wind speed Maximum Wind speed Figure B-4. Measured daily maximum and average wind speed

39 Year 4: Final Report Page 39 of 43 Appendix C. Data collected from the weather station located in Valrico, FL. 35 Solar Radiation (MJ/m 2 /d) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Rso=(KB+KD)Ra Rso=.75*Ra Rs Figure C-1. Solar radiation and clear-sky solar radiation 12 Relative Humidity (%) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Maximum Relative Humidity Minimum Relative Humidity Figure C-2. Measured daily maximum and minimum relative humdity

40 Year 4: Final Report Page 4 of Temperature ( C) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Maximum Temperature Minimum Temperature Figure C-3. Measured daily maximum and minimum temperature 16 Average Wind Speed (m/s) /6 2/14 5/25 9/2 12/11 3/21 6/29 Date (28 21) Average Wind speed Maximum Wind speed Figure C-4. Measured daily maximum and average wind speed

41 Year 4: Final Report Page 41 of 43 Appendix D. Landscape photos of cooperators in Apollo Beach, FL taken on a turfgrass quality rating day, 13 August 21. ET Controller Treatment Group W45 W6432 W712 W6434 W6619 W632 W627 Comparison Treatment Group W62 W6333 W1333 W639 W721 W6345

42 Year 4: Final Report Page 42 of 43 Appendix E. Landscape photos of cooperators in Riverview, FL taken on a turfgrass quality rating day, 13 August 21. ET Controller Treatment Group W11518 W11243 W11528 W7434 W6919 Comparison Treatment Group W1151 W1122 W1426

43 Year 4: Final Report Page 43 of 43 Appendix F. Landscape photos of cooperators in Valrico, FL taken on a turfgrass quality rating day, 13 August 21. ET Controller Treatment Group W121 W118 W22 W1948 W123 W2212 W187 W157 W2215 Comparison Treatment Group W221 W1949 W1614 W152 W151 W115