Summary

Click to view as a slide show

There are no magic "technofixes" to solve the problem of too many people and too little resources, but science and common sense together teach us what works or what ought to.  Ecologically appropriate landscaping--plant materials and selection strategy--must vary according to the climate of the region (e.g., humid vs. xeric) and human use.  Although "drought tolerance" is a popular misnomer for plant materials in the landscape, there are excellent varieties such as turfgrasses which serve us, while surviving mostly on natural rainfall. Turfgrasses are being developed with reduced need for irrigation; this might not "save" water but it is ecologically appropriate and imperative.
If we suspend thinking of the new grasses as products, and see them as a process, then we can do a better job of integrating the parts of our future landscape.  Once species and cultivar have been selected, establishment technique either conspires to destroy it, or promotes the plant's--and peoples'--success in this environment.  Maintenance of established landscapes can create a vicious cycle of pest problems, thatch accumulation, excess mowing requirement, and too frequent irrigation need.  Understanding the physics and biology of water conservation can save other resources, and is fun.
Finally, considering the gross inadequacy of most landscape sprinkler systems, plant variety selection will have relatively little impact on water conservation compared with the obvious:  design and manage the irrigation system so that it provides even coverage.

Statement of Problem
Landscape plant selection is a tool for water conservation, but it has its share of mythology.  This article briefly covers the theory of drought resistance in plants, variety selection, and its application to landscape water conservation.  Everyday Florida examples help us to understand the facts behind the theory.
Two assumptions are obvious and must be accepted: (1) landscape plants consume water and survive within the limits of moisture from nature (rain) and from management (irrigation) and (2) water applied in excess of this plant use results in runoff to canals and ponds, or percolation to the surficial aquifer.
Research has also shown that: (1) landscape plants vary in the water that they use, and/or their survival in drought, according to species and cultivar, and according to environmental conditions and (2) landscape soil moisture storage is limited by soil characteristics and on the depth and efficiency of roots.
There are apparently four methods by which managers could affect the use of water in the landscape: (1) change their watering practices; (2) select and use different species or varieties of plants, e.g., (a) varieties which use less water or (b) varieties with deeper root systems and larger available soil moisture supply; (3) modify the soil or topography to increase the available soil moisture; and (4) change the microenvironment to encourage lower evapotranspiration.  This article deals principally with method #2, although it emphasizes the importance of and the interaction among all four methods.

Drought resistance and water use
Drought resistance is "the generic term used to cover a range of mechanisms whereby plants withstand periods of dry weather" (Paleg and Aspinall, 1981).  In this broad, simple definition, drought resistant plants are the ones that survive irrigation curtailment, even during dry weather.
Drought tolerance is the ability to withstand a drying stress which penetrates plant tissue, while drought avoidance is the ability to avoid the drying stress (i.e., desiccation).  Most higher plants, including most landscape plants, are intolerant of desiccation.  Examples of drought tolerance occur in lower plants, e.g., fungi, some pteridophytes (e.g., resurrection fern), and the dried seed.  Otherwise, higher plants "normally remain turgid . . . therefore possess drought avoidance." (Levitt, 1980).  Except for a few rare adaptations, it is generally incorrect to refer to drought tolerance of landscape plants, because for all purposes it is nonexistent.
There are, however, many examples of drought avoidance as a mechanism of drought resistance.  Some plants are water saving drought avoiders.  Cacti conserve water, and therefore resist drought, by exchanging gases at night.  Bahiagrass conserves water during dry spells, and thereby resists drought, by reducing leaf area.  Bahiagrass can go into permanent wilt, lose all its leaves, and still come back.  It has protected stems which, although not tolerating desiccation, are well protected, and will initiate new leaves when rain returns.  St. Augustinegrass can avoid drought stress and conserve water, by wilting temporarily.  However, St. Augustinegrass does not come back from permanent wilt.  Unlike bahiagrass, St. Augustinegrass which is allowed to defoliate will die.
Deep rooting, is a drought avoidance mechanism within the general category of resistance (Levitt, 1980).  The deeply rooted plant may be considered a water spender.  Although ". . . it may appear contradictory to propose water spending as an adaptation to drought . . ." (Levitt, 1980), this mechanism is obviously successful.  Examples include mesquite, which grows in the desert southwestern United States, and a wide diversity of Florida natives and exotics.  Under proper establishment, many landscape plants can survive permanently with no irrigation, because their root systems are sufficiently extensive to maintain turgidity.  Levitt (1980) provides several examples of grasses which are drought avoidant due to greater water absorption.  Bahiagrass also fits into this water-spending, drought-resistant category.  It has the deepest roots of any warm-season turfgrass used in south Florida.
A deep-rooting, water-spending drought resistance mechanism provides a practical tool for water conservation.  In a few words, you don't have to irrigate.  The portion of irrigation water that is generally lost to direct evaporation, can be conserved by the use of non-irrigated landscape plants.  Some additional savings might come from a reduction in the luxuriant, water-demanding canopy that has been shown (Kneebone and Pepper, 1984) to result from excessive irrigation.  Using drought avoidant, deeply rooted plants can allow for survival of the landscape, while allowing for self regulation.  Does this mean that water has been saved?
No! In the immediate sense, a dead landscape saves water, while a surviving landscape must transpire.  Are drought avoidant plants appropriate for the Florida environment? Yes, they have been grown here for thousands of years.

Water in the Florida landscape
Florida is not a xeric region, and yet water availability is critical.  Although rainfall is sufficient annually to grow almost anything, its seasonally variability puts a stress on plants with shallow root systems, e.g., closely mowed turfgrass.  In Fort Lauderdale the 61 inches of annual rainfall more than satisfies the 44 inches of evapotranspiration (ET) for well-watered St. Augustinegrass.  Rainfall is seasonally variable, however, and about 32 inches of irrigation annually are needed to make up for the seasonal deficit between evapotranspiration and rainfall.  Drought is typically most severe in April and May, but non-irrigated St. Augustinegrass can be damaged at any time of the year, if irrigation is not available.
Because of more variable rainfall, seasonal variability of water stress is much greater in southern Florida than in northern Florida.  Soil moisture storage is least in southern Florida, due to the sandy soils.  For example, at turf plots in Fort Lauderdale, there is only 3% soil moisture, by volume, in the top foot of soil.  In the absence of rain or irrigation, and in full sun, most cultivars of St. Augustinegrass wilt 3 to 6 days after the last saturating rainfall.  Although there is no absolute limit to the effective root zone, the estimated cumulative root-available moisture reserve for St. Augustinegrass (based on days to wilt) is only about 15 mm (1/2 inch).  The frequent need for irrigation does not mean that turf areas need or use more water than deeply rooted trees or other vegetation types, but it shows that turf is vulnerable due to its more limited moisture reserve.  In the deeper sands, where plant available soil moisture is less than 2% by volume, deeper root systems would not be an overwhelming advantage.  Doubly the root system to 2 feet, from 1 foot, would only add about 6 mm moisture, or about 1 day of transpiration.  So where are deeply rooted, drought avoidant plants likely to get water?
Dig almost anywhere in Florida and you find moisture.  Low lying entisols along the southeast coast have an artificially drained water table which is often 4 to 6 feet below the ground level.  The widespread use of centrifugal pumps for lawn irrigation is proof that water is close to the surface.  This is verified by the fact that almost every canal is filled with water to within about 5 feet of the ground level.  The groundwater is often associated with a limestone substrate.  In southern Dade County the limestone is frequently exposed, but is interrupted by "solution holes" filled with fine particles.  In central Broward County the same pattern occurs, but the rock is several feet below the ground level.  At Fort Lauderdale Research and Education Center, the rock undulates from about 1 foot below the surface to sand-filled solution holes which extend 6 feet deep or deeper.  The surficial aquifer at this location is at 4.5 feet.  The rock and water table together are important, because roots of turfgrasses and other plants are found attached to the rock, extending in a yellowish marl layer down to the water table.  The marl layer is always wet, even after prolonged drought, and almost certainly provides upward capillary movement of water.  The proximity of the water table and rock associated with fine particles means that many deeply rooted landscape plants rarely need to be irrigated.  Their roots are the only pumps that they need.
The same generalization is true for another major soil region of Florida, but for a different reason.  The spodosols of central Florida have a zone of slow percolation, a "hardpan", which perches the water table nearer the plant roots.  In this region many sod farms efficiently produce turfgrass with no overhead irrigation (except possibly right after plug planting).  Home lawns in this region can be cared for also, in some cases, with minimal irrigation.  In coastal ridges throughout Florida, the water situation is not so favorable.  But in the major part of most urban areas (Jacksonville, Tampa Bay, Orlando, and South Florida) the net irrigation requirement should be more than adequate, even if our estimates were based on full-sun exposure (not universal in urban areas).
If persons frequently irrigate their lawns and other landscape areas, it does not necessarily mean that the plants need that much water, especially in Florida.  In all of Florida there is a rich tapestry of native vegetation which obviously has survived through wet-dry cycles with no irrigation.  Exotic vegetation has been introduced, some with even greater drought avoidance than the native plants.  Such exotic and native vegetation, turfgrasses and trees, can already be grown in many cases with no supplemental irrigation.  While the development of more drought avoidant vegetation is important, it appears to be equally essential: (1) to clearly understand the ecological and water relations of already available landscape plants; and (2) to apply common-sense in their maintenance.

Limited turf areas?
It is no accident that grasslands predominate in areas receiving less than adequate rainfall (Fig. 1).  While survival characteristics (e.g., intercalary meristems) are important, total water use differences are associated, as well.  In general, forested regions have about 45% higher transpiration than grasslands (based on Larcher, 1980).  Sawgrass, the native vegetation covering the Everglades, is more demanding of water than turfgrass.  Clayton (1949) reported 68 inches ET from sawgrass, compared with 54 inches for bahiagrass.
Unfortunately, there has been meager work on water use of woody plants, compared with the wealth of information on grasses. There is, however, no physiological basis on which to speculate that limiting turf areas, and replacing them with trees or shrubs, would conserve water. Because of simple thermodynamics, trees would be expected to actually use more water, thus confirming studies of real ecosystems (Fig. 1). Transpiration is driven principally by total radiant energy flux. The heat of vaporization is 580 calories per gram of water. This same value is true for transpiration from a tree, a turf, or a wet sponge. Because of the constant heat of vaporization, and the unchangeable radiant energy from the sun, a green growing canopy will consume about the same amount of water per unit area, regardless of whether it is covered by trees or turf. The openness of a tree canopy (causing lack of resistance to diffusion), and its roughness (causing turbulent air flow) would be the only obvious difference in the two vegetations' evapotranspiration. For these reasons, we would expect a tree canopy to transpire slightly more than a turf canopy. This is true, based on ecology, even though it refutes the popular mythology, that "turfgrass wastes water". People waste water.
There have been other alternatives proposed for reducing turfgrass areas, and thereby "save" water. One recommendation has been to replace turfgrass with gravel in order to reduce transpiration (South Florida Water Management District, 1978). This would be effective in a shortsighted way, but would be environmentally unsound, because it would involve other problems, e.g., weed control. Wood chips are also effective means of reducing evapotranspiration, but are temporary because they decompose and become overrun with weeds. Any surface which does not transpire or evaporate will contribute to heat buildup. It has been shown that wood chips underneath trees increase tree evapotranspiration, compared with an understory of turfgrass (James Beard, personal communication). The reason for this is obvious; the increased heat accumulated by the woodchips radiates back up to the undersides of tree leaves. As stated previously, the heat of vaporization of water is the same for turf and trees. The environmental air conditioning of trees and turf is proportional to their evapotranspiration. The old adage says, "there ain't no such thing as a free lunch."
When arguments to reduce turfgrass areas as a means of water conservation have run up against irrefutable laws of physics, other facts have had to be changed to agree with the theory.
One example involves the supposed maintenance inefficiency of turfgrass compared with groundcover. In this argument (currently being developed by the South Florida Water Management District), it costs more to maintain turfgrass areas than it does to maintain groundcover areas (Teets, Xeriscape Florida 90, Orlando, 20 September 1990). The major assumption in this argument is that a typical 5,000-square-foot lawn requires mowing 36 times per year, at 6 hours per mowing, at $22.50 per hour. By multiplication, it can be shown that it costs $4860 per year just to mow a typical lawn. For anyone who has a lawn, it is obvious that these values are inflated by many times. As an example, the estimated cost of just one cutting of a typical lawn would be $135, or about as much as an inexpensive lawnmower! Published data (Van Zomeren, 1983) shows that the minimum rate of professional lawn cutting on a college campus is actually 22,000 square feet per hour. The figures of the South Florida Water Management District overestimate the time factor by 2,540 %.

Beneficial turfgrass, healthy environment
From an environmental viewpoint, rather than limiting turfgrass areas, it would be more appropriate to expand turfgrass areas, in some cases. Examples include church and swap-shop parking lots, swales which collect stormwater runoff, slopes subject to erosion, and areas near buildings. It would be so much better to use turfgrass to filter particulates and oily residues from asphalt, than to run that water out to the nearest oceangoing canal. Turfgrass areas can facilitate the movement of cool, mosquito-abating breezes. This is important near houses, especially with the increased risk of St. Louis encephalitis. Turfgrass areas are increasingly more important in personal security, because of the risk of home invasion in most urban areas of Florida. Turfgrass areas also provide safety and sanitation, for outdoor play activities, for movement of people and vehicles, and for inspection of sensitive infrastructures, e.g., bridge embankments. While other environmental plants provide some of these benefits, a sod-forming turfgrass is most dependable, for these situations. Turfgrass areas blend in with other vegetation types, and provide a transition for the use by butterflies, birds, and mammals, including people who come to observe.

Water quality and water quantity impact on the landscape
Water quantity is closely related to water quality. Ultimately, this is the urgent reason behind landscape water conservation. In low lying coastal areas, if the pressure or "head" of freshwater is relaxed, the aquifer can be penetrated by salt water. Because fresh water is lighter than salt water, there is a small margin of safety. The porosity of the underlying rock, and other factors (presence of canals and water control structures) can affect the rate of saltwater intrusion. Hydrogeologically, Florida is like a sponge filled with freshwater, floating in a buoyant ocean of salt. As long as the inflow of freshwater considerably exceeds the use, water quality is adequate. During dry periods there is a risk of reduced water quality.
In southern Florida, the previously mentioned landscape factors work together with geographic factors to make landscape water use a very major concern: greater seasonal variability in plant water stress, sandy soils, closeness to the ocean, the presence of large population centers, and the demand for luxuriant, drought sensitive vegetation. These factors explain the great involvement of public agencies (e.g., water management districts) in landscape water conservation in southern Florida.
During extended dry periods water users must restrict consumption. Landscape irrigation traditionally receives relatively greater restriction than some other water uses, and such restriction has the potential for harming landscape plants. At Fort Lauderdale, commonly grown Floratam St. Augustinegrass goes into permanent wilt (i.e., remains wilted in the morning) only 19 days after irrigation curtailment. Thereafter, the turf loses about 3% coverage per day. St. Augustinegrass, mainly Floratam, represents about 234 thousand acres of turfgrass in the South Florida Water Management District alone. Water curtailment to the millions of St. Augustinegrass lawns could result in great damage. Landscape water use is important because it is large (an estimated 200 billion gallons irrigation requirement in south Florida St. Augustinegrass lawns) and because of the potential for great harm when it is curtailed.

Breeding for drought resistance
Previous research on drought resistance has emphasized physiological characteristics of plant tissue. Examples of such research include the selection for varieties with fewer stomates. It was found that such selection was ineffective in reducing water use, because it was associated with an increase in stomate size. Leaf resistance has been shown to be a minimal factor in St. Augustinegrass turf water loss, and so there is little basis on which to expect that selection for stomate characteristics would be effective.
Possibly, the fascination for physiological mechanisms of drought resistance has been in the belief that science can change the underlying processes by which plants grow. In fact, the search for other magic bullets has been repeatedly unsuccessful. Compilation of some few studies which have compared species (Table 1) shows that the maximum species variation (6.2 mm day-1 for buffalograss vs. 7.4 mm day-1 for St. Augustinegrass) would only result in 16% savings even if you could grow buffalograss successfully in Florida. While species differences are so inconsequential, variations among cultivars within species have been even more elusive. The only major demonstrated differences in water use among grasses have been studies which compared C3 species e.g., tall fescue with C4 species, e.g., St. Augustinegrass (Biran et al., 1981). The result has been that cool-season, C3 species have been shown to use about 45% more water than warm-season, C4 species. While this is important, it is not practically relevant, because we wouldn't grow tall fescue in Florida.
The reasons that the water use approach to resistance breeding is inappropriate for the Florida setting are several: (1) differences in water use efficiency are rarely observed; (2) in notable instances these differences contradict obvious drought survival differences; and (3) most of this research has been done in western states where total water supply is limited.
A practical method for selecting drought avoidance in Florida grasses is to grow them under conditions of limited, or no, irrigation. This has been shown (Busey, 1989) to be useful in selecting drought resistant polyploid St. Augustinegrasses, and has resulted in the recognition of FX-10, which survives for extended periods with no irrigation. In the first two years of total curtailment of irrigation, FX-10 had very little loss of stand, compared with Floratam, which was almost completely wiped out. The grass has a patent pending, and the first plant patent applied for by the University of Florida. FX-10 is not a grass which never has to be irrigated. After 2.5 years with no irrigation at Fort Lauderdale, several FX-10 plots have lost substantial coverage. If anything, it would be important to have an efficient irrigation system for the few times a year that one would need it. The number of required irrigations per year would vary greatly with microenvironment.
The potential advantages of FX-10 require continued evaluation before a blanket recommendation can be made. However, FX-10 has additional advantages besides drought resistance. FX-10 is moderately resistant to the Floratam-killing chinch bug (Busey, 1990) and is more shade tolerant than either Bitterblue or Floratam. This shade tolerance could be quite important for urban areas, in view of the increasing tree canopy, and the desire of persons to enjoy turf, but protect themselves from sun exposure. FX-10 is lower growing than Floratam. (It is not known how low growth habit might translate into mowing energy requirement.) FX-10 retains a distinctively more bluish color than Floratam, which would be an esthetic alternative. While FX-10 is no "wonder grass", it appears to be a dependable alternative to Floratam. The technology was transferred to the Florida Sod Growers Cooperative, which has the exclusive rights to produce FX-10. Currently there are 22 growers statewide who are producing FX-10. The grass was distributed for experimental purposes to extension faculty in 34 Florida counties, as well as private cooperators.

Educational opportunities - a common-sense approach
Even before more drought resistant grasses become available, there are many common-sense things that can be done to conserve water in the landscape.

Here are some examples:
1. Treat the landscape (including turf) as a process, not a series of products. Encourage the sensitive awareness of landscapes as ecological communities. This should include an awareness of the benefits of turfgrass when it is used appropriately, as well as other vegetations when they are used appropriately. This should also include the treatment of other organisms as components of an ecosystem, not as pests. Avoid reliance on "technofixes". Landscape biology can help in conservation of water and other scarce resources (e.g., petroleum) and is fun.
2. Turn it off. Homeowners could reduce irrigation use by relying on "single- event" timers, rather than more technologically complex systems. While moisture sensors are very compatible with commercial turf irrigation, it is easier for most homeowners to use the bioassay, "let your lawn tell you when to water."
3. Reduce nitrogen and phosphorus fertilization to the minimal amount needed to prevent weed encroachment. Excessive reliance on color cues results in excessively high application rates. This is associated with premature wilt, greater risk of chinch bug and sod webworm infestation, and excessive cutting requirements.
4. Avoid hard curbs and hill mounds. Structures such as these direct rain water to storm drains, canals, and, very quickly, the ocean.
5. Properly establish turfgrass and other landscape areas. This involves proper irrigation design and installation, careful grading, and timely weed control. The best, most well adapted plant material, will die or perform very poorly if it is root bound, poorly pruned, contaminated with nematodes or torpedograss, or planted over buried debris.
6. If it dies, grow something else. Many plants, including exotics and natives, are poorly adapted to the disturbed soils of urban areas. Consider this a good selection opportunity.
7. Suspend the adoption of buzzwords. Acronyms and other catch phrases are an effective marketing tool. Until Florida horticulturists can agree on the message that is trying to be conveyed, it would be wise to withhold the acronyms. The healthy landscape is a process, not a product.

References
Biran, I., B. Bravdo, I. Bushkin-Harav, and E. Rawitz. 1981. Water consumptions and growth rate of 11 turfgrasses as affected by mowing height, irrigation frequency, and soil moisture. Agron. J. 73:85-90.
Busey, P. 1996. Wilt avoidance in St. Augustinegrass germplasm. HortScience 31:1135-1138
Busey, P. 1990. Polyploid Stenotaphrum germplasm: Resistance to the polyploid damaging population southern chinch bug. Crop Sci. 30:588-593.
Busey, P. and J. H. Parker. 1991. Energy conservation and efficient turfgrass maintenance. in: Waddington, D. V., R. C. Shearman, and R. N. Carrow (eds.). Turfgrass science (revised)., American Society of Agronomy, Madison, WI. (In press).
Casnoff, D. M., R. L. Green, and J. B. Beard. 1989. Leaf blade stomatal densities of ten warm-season perennial grasses and their evapotranspiration rates. Proc. Sixth Int. Turfgrass Research Conference 129-131.
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.
Kim, K. S. 1983. Comparative evapotranspiration rates of thirteen turfgrasses grown under both nonlimiting soil moisture and progressive water stress conditions. M. S. Thesis, Texas A & M University, College Station, TX. 64 p.
Kneebone, W. R. and I. L. Pepper. 1982. Consumptive water use by sub- irrigated turfgrasses 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.
Larcher, W. 1980. Physiological plant ecology. Springer-Verlag, Berlin. 2nd edition.
Levitt, J. 1980. Responses of plants to environmental stresses. Volume II: Water, radiation, salt, and other stresses. Academic Press, New York. 606 p.
Paleg, L. G. and D. Aspinall. 1981. Drought resistance in plants. Academic Press, Sydney, Australia. 492 p.
South Florida Water Management District. 1978. Landscaping: Water conservation. South Florida Water Management District. 12 p.
Stewart, E. H., J. E. Browning, and E. O. Burt. 1967. Effect of depth to water table and plant density on evapotranspiration rate in southern Florida. Trans ASAE 10:746-747.
Thayer, R. L., Jr. 1982. Public response to water-conserving landscapes. HortScience 17:562-565.