Austin Green Building Program Analysis:  The Effects of Water-Related Green Building Features on Residential Water Consumption

 

Audrey Tinker, Richard Burt, Sherry Bame, and Michael Speed

Texas A&M University

College Station, Texas

 

The green building movement has grown tremendously in the last decade.  A major impetus for this movement was the establishment of the Austin Green Building Program.  It created the first U.S. green rating system to give direction and recognize builders for efforts to make residences more sustainable.  Today, almost 20 similar programs are functioning throughout the country.  While these programs are assumed to provide benefit to both homeowners and the environment, little research has been conducted on the effect they are actually having in regards to water conservation.  This study investigates the results of installing either water-conserving features or ideas into homes registered with the Austin Green Building Program.  It was hoped that if these items were shown to reduce water consumption, the Program would receive recognition and other programs would be able to use their rating system as a model.  If items were shown to be ineffective, then further investigations would need to be conducted to find the reasons and new water-saving strategies devised.

In actuality, a mix of findings emerged.  One of the most important findings was that the effect of water-conserving features varies considerably by builder.  Also of interest is the relatively small number of water-related features builders are incorporating.  Finally, only a few items were found to consistently be associated with reduced water consumption while others had either no or varied effects or were even associated with increased water consumption.  While these findings are interesting, further research will no doubt be required to provide definite confirmation.

Key Words:  green building programs, water-conservation, residential conservation programs

Need for Conservation 

Water supplies around the world are in jeopardy as populations increase and sources dwindle.  Current projections estimate that 3 billion people will live in areas classified as under water stress or scarcity by the year 2025, up over 6 times from the year 2000 level of 480 million (Gleick, Burns, Chalecki, Cohen, Cushing, Mann, Reyes, Wolff, & Wong, 2002).  The United States shares this problem of increasing water scarcity.   Projections estimate that by 2050, almost 40% of the U.S. population will live in areas that will either need to be conserving water or developing new water sources to cope with water shortages (House Committee, 2003). 

Texas is one of the largest (top four) water consuming states in the U.S. (Wagner & Kreuter, 2002).   With a population expected to double by 2050, water needs are expected to increase from 4.23 to 7.06 million acre-feet/year (National Wildlife Federation, Environmental Defense, and Lone Star Chapter of the Sierra Club, 2001).   To complicate the issue, conventional fresh water supplies in the State are already 75%-80% developed (Texas Water Development Board, 1995).  This is why the Governor’s task force in 2000 identified limited water supply as one of the two “most serious natural resource issues facing Texas today” (Wagner & Kreuter, 2002).   

The purpose of this study was to determine the effects of (1) socioeconomic factors (home size, lot size, appraised value and pools), (2) atmospheric factors (temperature, rainfall and evaporation) and water-related features from the Austin Green Building Program single-family checklist on residential water consumption.  Austin, Texas was selected for this study for two reasons.  First, the city has recently experienced water crises as the city’s source aquifer has experienced greater than average declines.  Additionally, Austin has the oldest and one of the largest Green Building Programs for residential construction in the country.   The idea and development of green building programs is spreading quickly across the country with 19 residential green builder programs functioning and seven additional programs in the development stages (NAHB, 2002).  Little research has been done, however, on the homes involved with these programs.  It was therefore of interest to investigate homes registered with the Program to determine if findings vary from prior residential studies and if conservation techniques purported in the Program (Xeriscaping, rainwater catchment systems, natural vegetation…) significantly reduce water consumption.   

Austin Green Building Program 

Developed in the early 90’s as the first U.S. program to promote sustainable residential construction, the Austin Green Building Program’s (AGBP) mission is to “accelerate the integration of sustainable building products and practices with mainstream building through marketing, education, and technology transfer.” (ACEEE, 2003).  One of the main ways this is accomplished is through the Program’s “green” rating system.  The Program rates new homes and remodels on a scale of one-to-five stars, with more stars indicating a greater number of green features or design considerations.  Features and designs are subdivided into five main categories:  (1) energy efficiency, (2) water efficiency, (3) materials efficiency, (4) health and safety and (5) community (City of Austin, 2001).  Points are given for incorporating items from a few or all of the five categories.  Besides rating homes, the AGBP also provides consultation services and marketing support for members, technical seminars for designers, a directory of Green Building professionals and a resource library for members (City of Austin, 2001).   

The Program is gaining immensely in popularity.  In 2002, 57% of all new homes built in the Austin area were rated by the Program.  Also, all new city-supported housing is now required to meet the 2-star compliance level or higher (ACEEE, 2003).  Program success can also be assessed from the growth in similar programs throughout the nation.   

While reports vary concerning the profitability of green building, almost all involved builders believe the AGBP rating system point values are based on environmental impact (80%) and that the Program makes a positive environmental difference in the community (Tinker, Kreuter, Bame, & Burt 2003).  The largest builder involved in the Program reported that their commitment to green building in the Austin area has resulted in “increased sales, market position, and customer satisfaction.” (City of Austin, 2002).  While support for the Program may be high in the community and among builders, little research has been done regarding the water conservation aspects of the AGBP and their resulting effect.  Evidence of conservation effort effectiveness could assist in strengthening the Program (Mayer, DeOreo, Opitz, Kiefer, Davis, Dziegielewski, & Nelson, 1999).    The findings of this study will measure correlations between green home water consumption and AGBP water-related features. 

Prior Water Assessment Techniques 

Previous studies have employed a variety of techniques to assess features or conditions that significantly affect water consumption.  Engineering models have been used to predict how much water items or processes will use (Baumann, Boland, & Hanemann, 1998).  Technological advances have enabled researchers the ability to monitor water usage directly as it occurs.  While flow-trace devices do provide accurate data, cost and obtrusiveness can be a problem, and thus are limited to small case studies (Mayer, et al., 1999).    

Although few in situ studies have been conducted on water saving features, more substantial research has been completed on other conditions or items that effect residential water consumption.  In regards to weather for example, rainfall and temperature have proven to be significant variables affecting water use in a number of studies that utilized multiple regression techniques (Anderson, Miller, and Washburn, 1980; Morgan and Smolen, 1976; Hansen and Narayanan, 1981).  For example, a study conducted by Fourt (1958) in 1955 found that rainfall days, average number of people per meter and cost of water were significant factors affecting consumption (R² of .839) when 21 large cities were surveyed (Grima, 1972).  However, in studies conducted by Linaweaver, Gyer, and Wolff (1967) and Haver and Winter, (1963) climatic factors were not found to be highly significant (Grima, 1972).    Evapotranspiration (ET) has also been found to affect outdoor water consumption in a variety of studies (Danielson et al., 1980, Duble, 1997; Mayer, 1995; Stadjuhar, 1997; Aquacraft, Inc., 1997).   

Aside from weather-related findings, homes with pools were found to use more than twice the exterior water as those without.  Findings have also indicated that outdoor water use was positively correlated with home and lot square footage (Mayer, et al., 1999).    Income or economic level has also been held as a significant factor influencing water consumption because it is generally assumed that those with higher incomes have more water-utilizing appliances or features (such as pools) and often have larger lots for irrigating (Grima, 1972).   These items, as well as the AGBP conservation-related items will be tested to determine the effect each has on residential water use by Austin residents in 1998-2002. 

Data

The first data required included monthly temperature, rainfall and evaporation data for the two-year period of March 2001 to March 2003.  The second data consisted of monthly water consumption records (gallons per month) for each home registered in the Green Builder database on record at the Travis County Appraisal District.   The span of the monthly records was from March 2001 to March 2003, corresponding with weather information gathered.  The third data needed was the square footage of each home and yard, the appraised value and the existence of a pool for each home registered with the Green Builder Program that had a record at the Appraisal District.  Finally, records of the green features included in each home along with information on the builder and star-rating were required.   

The Location of the Data

Monthly temperature and rainfall information was gathered from the National Weather Service’s Southern Regional Headquarters website.  Evaporation data was gathered from the Ft. Worth District’s Reservoir Control Office of the U.S. Army Corps of Engineers website for Lake Travis/Marshall Ford Dam.   Monthly water consumption records were obtained from the Austin Energy department.  Information on the square footage of the homes and yards, appraised value and the existence of a pool for the green residences were then obtained by typing in each address at the Travis County Appraisal District and recording the information on site.  Finally, records on green water-related features, builder and star rating information for each home were obtained from Austin Energy’s Green Building Program database zip disk received in person from Austin Energy.  The list of water-related green features is included in Appendix A.  Identifying information was deleted prior to obtaining the data except for addresses that were used to merge the diverse data sets.  Once merged, address information was deleted.  All findings are reported in aggregate figures only to ensure anonymity of residences.  Builder information was considered public information by the AGBP.  

Data Analysis

Temperature and evaporation were averaged monthly, while rainfall was totaled for each month.  Home square footage, land square footage, appraised value, pool information, green features, builder, star-rating and household monthly gallons consumed were entered according to each residence built between 1998 and 2001 with a 10% verification of coding the data.   Stepwise linear regression was used to analyze significant relationships associated with monthly water use.  Stepwise regression is a statistical search method which “develops a sequence of regression models, at each step adding or deleting an X (independent) variable.  The criterion for adding or deleting an X (independent) variable can be stated equivalently in terms of error sum of squares reduction, coefficient or partial correlation, T* statistic, or F* statistic.” (Neter, Kutner, Nachtsheim, & Wasserman,1996). Only those models with p-values less than .05 were considered further. Standardized residuals were saved for the model and then, to control for equal variance, only those cases in which the absolute value of the residuals was +/-4 were retained.  This reduced the number of cases by less than 1% (N=16,477 from 16,605).   Multicollinearity of the independent variables was analyzed using the Variance Inflation Factor (VIF) with values over 10 considered problematic.  Therefore, models were only selected if  VIF’s for all variables were under 10.   

Of the thirty-three builders participating in the Austin Green Building Program during the study period with corresponding available home water consumption records, there were 5 strata constructed based on the number of green homes, (1) Builder 1 with twenty-nine small builders with 24 or fewer homes totaling 169 homes, (2) Builder 2 with 65 homes, (3) Builder 3 with 79 homes, (4) Builder 4 with 115 homes, and (5) Builder 5 with 402.   The “Builder 1” category was created to represent smaller builders and to allow for easier comparison with five, in lieu of 33 groups. 

Results 

Descriptive statistics were first calculated for the data and are summarized in Table 1. 

The large variance in monthly water consumption was evidenced in the spread from a minimum of 100 gallons monthly per household to a maximum of 57,700 gallons.  Half of the green-built households used 7,900 gallons/month or less.  Approximately 20% of households consumed 4,000 gallons or less a month and approximately 20% used 15,000 or more.  Less than 1% of households used over 38,000 gallons in a month. 

Climatic conditions experienced during the study period were then compared to average Austin conditions to determine if the two year study period was indicative of normal weather patterns.  Using Pearson correlation coefficients, temperature for the two year study period correlated over 98% with historic Austin levels and evaporation correlated almost 93%.  However, rainfall was negatively correlated (-.14) for the study period’s rainfall vs. historic Austin rainfall (42.28 inches per year for the study period as opposed to an average of 33.42 from 1856-2003) (National Weather Service, 2003).  The differences are illustrated in Figure 1 below.  Thus, the inverse relationship between rainfall and water consumption found in prior studies may vary with respect to this particular study.

Appraised values of the green-built homes ranged from $42,365 to $952,004 per home.  When frequencies were analyzed, just over 12% of homes were valued at or below $100,000.  Almost 50% were between $200,000 and $300,000, which is illustrated by the large spike in Figure 2 below.  Finally, just over 11% of homes were appraised at or above $350,000. This distribution includes 47% of all green homes in the greater Austin area built between 1998 and 2001, however, it is not known how these values compare to non-green builder residences. 

Only 2% of study homes had square foot areas of 1,000s.f. or less and less than 2% were at or above 3,450s.f.  The mean home square footage was 2,186.  Lot square footages ranged from

3,300 to 30,056 with almost 60% between 6,000 and 9,000s.f..  Figure 3 illustrates this distribution. While the maximum sized lot was 30,056s.f., only 1% of lots were above 17,500s.f. 

 The distribution of star-ratings ranged from 1 to 4 stars for homes registered by the Austin Green Building Program from the period of 1998-2001 (a 5-star rating was available, but for all 7 homes registered with the rating, residuals were higher than 4).  Findings show that almost 75% of homes registered achieved the two-star rating which requires point totals of 60 to 89.  Less than 10% achieved a higher rating, with no 5-star homes.  This distribution is reflected in Table 2 below.  

The frequency of using water-conservation features listed as builder options in the Austin Green Building Program home rating sheet was then investigated.  A few items were very popular, including utilization of site trees (81.7% of the homes), Xeriscape (66% of homes) and efficient dishwashers, (65% of study homes) however, other features were used less than 25% of the time.   For example, less than 1% of homes included solar heating, efficient clothes washers, rainwater catchment systems, efficient irrigation systems, drip irrigation and graywater systems.  While other options were incorporated more, their inclusion was still minimal.  Table 3 provides frequency percentages for inclusion for each of the AGBP water-related options. 

Next, SPSS was employed to calculate the most effective model describing the factors affecting water consumption using stepwise regression.  The following formula was chosen as that with the highest R² value without any variance inflation factors over 10.  The resulting R² was 0.224 and all variables were significant below the .05 level. 

Y = -12445.996B₀ + 170.476B₁ + .0104B₂ + 351.521B₃ + .209B₄ + 3660.119B₅ + 1.428B₆ + 1021.443B₇ + 1884.920B₈ + 487.716B₉ -4954.909B₁₀ + 2648.481B₁₁ + 7113.801B₁₂ -3288.587B₁₃ + 6884.856B₁₄ -1454.845B₁₅ -2248.672B₁₆ -1424.801B₁₇ + 1498.224B₁₈ -1630.345B₁₉

Where:

Negative relationships were of considerable interest, with the following green builder variables found to decrease consumption:  an efficient irrigation system, a remodel, minimal use of turf, gutters directed towards vegetation, Dillo Dirt amendment (EPA-certified soil conditioner comprised of wastewater sludge that is anaerobically digested and composted) (Water and Wastewater, 2001) and an efficient dishwasher.  Unexpectedly, the following Program variables were found to correlate with increased water use:  a water heater located close to distribution sources, low-water turf varieties, saving site trees, use of an efficient clothes washer, and retaining natural vegetation.    As found in several previous studies, temperature, value, home and lot square footage, evaporation and the existence of a pool were directly related to water consumption as well.  However, rainfall also increased consumption which is counterintuitive and in opposition to prior findings.  The builder variable was also found to significantly affect water consumption, but further investigation is needed on which builders in particular are positively and negatively associated with water consumption. 

One should keep in mind that because the R² value was fairly low, significant variation is left unexplained.  Figure 4 illustrates this variance with predicted values and 95% confidence intervals in the center of the graph, 95% prediction intervals at the far top and bottom and actual mean monthly water consumption in the spiked areas. Therefore, this model, while achieving similar results (R² values) as previous studies, should be used with caution.   

 Therefore, regression analysis was utilized to develop formulas for each builder group to determine if the same variables were significant for each.  Table 4 below illustrates the findings for and between each builder. 

             *Bold indicates significant at p<.05

Extreme differences between homebuilder groups are evident in the above table.  One observation is that the small builder group incorporated a wider variety of water-conserving features into its homes.  It is unclear, however, if this relates to the greater variety of builders and styles in the small builder (Builder 1) group or if small builders actually incorporate more water saving features.  Findings for the group however, do vary substantially from prior study population findings as a whole.  For instance, ensuring shade, installing a porch, using site trees, creating a light colored exterior, installing pervious paving and topsoil amendment all appeared to significantly reduce water consumption where these items had not in the general grouping.  As before however, the use of Dillo Dirt amendment, an efficient dishwasher, a remodel, minimal use of turf, and gutters directed towards vegetation, decreased water usage.  Also of interest were the findings that solar heating, multiple dwellings, and 15 year old + utilities related to increased consumption, although upon consideration, this may seem reasonable.  Finally, within the small builder group, it appears that two-star homes have the greatest impact in reducing water consumption.   

Among the other builders, few variables stood out as significantly affecting water consumption.  This may be because some builders include certain variables in every one of their homes and thus within group comparisons cannot be made to determine if the item is effective.  Regardless, in Builder Group 2, homes with porches related to less water consumption, while light colored homes and saving trees related to increased water use.  For Builder Group 3, multiple dwellings were the only item found to reduce consumption and utilizing trees on site increased consumption for Builder Group 4.  Also evident in Table 4 is that the large builders incorporate very few water-related features.  This may be because most achieve a maximum star rating of 1 or 2 and therefore fewer items are actually required to qualify.  Additionally, they may focus their efforts on areas besides water conservation, such as energy and materials due to the fact that there is no minimum point requirement for each of the five categories.     

Between group findings were also worthy of note.  For groups one, two and four, appraised value was found to negatively relate to consumption.  This is in contrast to previous findings from other studies and to the findings for the groups as a whole.  Also, weather had significantly different magnitudes of effect between builders such that each degree in temperature rise increased water use by 48.5 gallons for builder 2 and over 234 gallons for builder 5.  However, each inch of evaporation increased water consumption for Builder 2’s homes by over 29,000 gallons, while Builder 5’s homes only increase 3,634 gallons per inch.  Only for Builder 3, did home square footage and consumption have a negative relationship and for Builder 4, lot square footage and consumption had a negative relationship.        

Discussion 

Several significant findings resulted from this study.  One of the most important is the fact that different conditions and green features have distinct effects depending on the builder.  This may be due to the manner builders install features, what types of products they install, their overall designs or other characteristics of their homes.  Whatever the cause, it means that the assumption should not be made that certain green features are ineffective in all cases, for it may just depend on who is installing the item and what other features are included.  Study results also indicated that large builders in this study incorporated very few water-saving features in their homes.  The reasons for this may include a majority of 1 and 2 star ratings for large builders, thus indicating that fewer AGBP checklist items were attained overall.  Also, it may reveal that large builders are focusing their attention on other categories such as Energy, Materials or Health and Safety in lieu of Water as they may view there is greater customer concern over these issues.    

Overall, it was found that Dillo Dirt amendment, minimal use of turf and gutters directed towards vegetation were associated with decreased water consumption in both formulas.  Consistently related to increased water consumption were the green variables of low-water turf, saving site trees, use of an efficient clothes washer, and retaining natural vegetation. Conclusive results regarding clothes washers should not be made however, because less than 1% of homes incorporate these fixtures.  As far as non-green variables, temperature, rainfall, evaporation, appraised value, home square footage, lot square footage and pools all had a positive relationship with water consumption.  Rainfall’s positive affect was a surprise as most prior research has found it to have a negative relationship to water consumption.  In this study, the reason may be due to abnormal monthly rainfall during the study period and overall higher average rainfall, which correlated only slightly and also negatively with Austin historic averages.          

Finally of interest is the relatively small effect green features had overall in reducing the total variability in monthly water consumption.  The R² of the regression model with temperature, rainfall, size, value and pool characteristics included and no green features was .207.  The highest R² including green features was .224.  Thus, only .017 of the water consumption variability was described by the green features.  Thus, some variables not investigated in this study or purely homeowner use habits must account for the largest portion of water variability.  Due to the low R² achieved, the model should be used with caution in predicting future water consumption.   

Recommendations 

To determine the true effects of more green water-related features, a sample population that incorporated a greater number of these items would be beneficial.  For instance, less than 1% of study population homes included such items as rainwater catchment systems, graywater reclamation systems, and drip irrigation systems.  These however, may significantly reduce water consumption.  Even though the total number of homes included in the study was quite large, either a population with greater variety or a study of homes that are known to include these items versus homes that do not would be more beneficial.      

Also, a comparison of green to non-green homes could provide additional information on the effectiveness of these features.  These homes may have owners that are more highly concerned with water conservation or may all have better plumbing or fixtures than the general population.  Thus, a comparison with non-rated homes would allow the total story to be told.     

Finally, a study of an area that did not include one dominant builder would also be beneficial.  Because one builder constructed approximately 50% of the homes rated by the Program, their homes and features included may have had an undue influence on the findings for the group as a whole.   With a more evenly distributed builder population, one builder’s effects would be dissipated.   

References 

American Council for an Energy-Efficient Economy (n.d.).  Residential New Construction Exemplary Program.  Retrieved June 25, 2003 from:  http://aceee.org/utility/5egrnbldgaustin.pdf 

Anderson, R.L., Miller, T.A. & Washburn, M.C. (1980).  Water Savings from Lawn Watering Restrictions During a Drought Year.  Water Resources Bulletin, 16, 4, 642-645. 

Aquacraft, Inc. (1997).  Project Report:  Evaluation of Reliability and Cost Effectivness of Soil Moisture Sensors in Extended Field Use.  Boulder, Colorado:  Aquacraft, Inc. 

Baumann, D.D., Boland, J.J. & Hanemann, W.M. (1998).  Urban Water Demand Management and Planning  McGraw-Hill, Inc.:  New York. 

City of Austin (n.d.). Our residential program for architects, builders, owners and more. Retreived June 5, 2001, from http://www.ci.austin.tx.us/greenbuilder/programs_res.htm 

City of Austin (2002, January).  Kudos to Newmark Homes.  Green Building Program Newsletter.  Retrieved June 25, 2003 from:  http://www.ci.austin.tx.us/greenbuilder/n1_jan_mb.htm 

City of Austin (2001).  Green Building Program.  Retrieved June 25, 2003 from:  http://www.ci.austin.tx.us/greenbuilder 

Danielson, R.E., Feldhake, C.M., and Hart, W.E. (1980).  Water Requirements for Urban Lawns.  Fort Collins, Colorado:  Colorado Water Resources Institute. 

Duble, R.L. (1997).  Water Management on Turfgrasses.  Texas Agricultural Extension Service.  College Station, Texas:  Texas A&M University. 

Gleick, P., Burns, W.C.G., Chalecki, E.L., Cohen, M., Cushing, K.K., Mann, A.S., Reyes, R., Wolff, G.H., & Wong, A.K. (2002).  The World’s Water:  The Biennial Report on Freshwater Resources. Island Press:  Washington. 

Grima, A.P., (1972). Residential Water Demand:  Alternative Choices for Management.  Toronto:  University of Toronto Press. 

Fourt, L. (1958).  Forecasting the Residential Demand for Water, Agricultural Economics seminar paper at the University of Chicago. 

Hansen, R.D. & Narayanan, R. (1981).  A Monthly Time Series Model of Municipal Water Demand.  Water Resources Bulletin, 17, 4, 578-585. 

Haver, C.B. & Winter, J.R. (1963).  Future Water Supply in London:  An Economic Appraisal and Supplementary Comments Re Economic Analysis of Future Water Supply of London.  London, Ontario:  Public Utilities Commission. 

Linaweaver, F.P., Geyer, J.C., and Wolff, J.B. (1967).  A Study of Residential Water Use.  Washington, DC:  BPO. 

Mayer, P.W. (1995).  Residential Water Use and Conservation Effectiveness:  A Process Approach.  Master’s thesis.  University of Colorado, Boulder. 

Mayer, P.W., DeOreo, W.B., Opitz, E.M., Kiefer, J.C., Davis, W.Y., Dziegielewski, B. & Nelson, J.O. (1999). Residential End Uses of Water.  AWWA Research Foundation and American Water Works Association:  Denver, Colorado. 

Morgan, W.D. & Smolen, J.C. (1976).  Climatic Indicators in the Estimation of Municipal Water Demand.  Water Resources Bulletin, 12, 3, 511-518.   

NAHB Research Center, Inc. (2002).  Summary of Existing Green Building Programs.  Upper Marlboro, Maryland:  NAHB Research Center. 

National Weather Service (2003).  Monthly/Annual/Average Precipitation - Austin Mueller/Mabry, Texas (1856 - December 2003).  Retrieved from:  http://www.srh.noaa.gov/ewx/html/cli/aus/ausmonpcpn.htm 

National Wildlife Federation, Environmental Defense, and Lone Star Chapter of the Sierra Club, (2001).  Texas Water Matters.  Retrieved from:  http://www.texaswatermatters.org/conservation.htm 

Neter, J., Kutner, M.H., Nachtsheim, C.J. & Wasserman, W. (1996).  Applied Linear Statistical Models, Fourth Edition.  WCB/McGraw Hill:  Boston, Massachusetts. 

Prasifka, D.W., (1994).  Water Supply Planning.  Krieger Publishing Company:  Malabar, Florida. 

Stadjuhar, L.E. (1997).  Outdoor Residential Water Use.  Master’s Thesis.  University of Colorado, Boulder. 

Texas Environmental Profiles (n.d.)  Retrieved June 25, 2003 from:  http://www.texasep.org/html/wqn/wqn_7pay.html. 

Texas Legislature (2003). House Bill Nos. 2660, 2663, and 1094.  Retrieved June 25, 2003 from Texas Legislature Online at: http://www.capitol.state.tx.us/tlo 

Texas Water Development Board (1995).  Homeowner’s Guide to Water Use and Conservation.  Austin, Texas.

Tinker, A., Kreuter, U., Bame, S. & Burt, R. (2003).  Contractor Motivation and Trends in Austin’s Green Builder Program.  Unpublished doctoral dissertation, Texas A&M University, College Station, Texas.

U.S. Committee on Transportation and Infrastructure (2003).  Water Officials Urge Congress to Implement Improvements in Water Supply Management.  Retrieved from:  http://www.house.gov/transportation/press/press2003/release103.html 

Wagner, M. & Kreuter, U. (2002).  Groundwater supply in Texas:  A review, and the role of landowner associations.  Unpublished manuscript.   

Water and Wastewater Utility: Dillo Dirt, (2001). Retrieved April 20, 2003 from http://www.ci.austin.tx.us/water/dillo.htm  

Appendix A 

Water-Related Green Features