Return to METRIA 9

Drought Resistance in Trees: An Overview of Mechanisms
and New Research with Iowa's Black Maples

William R. Graves

Associate Professor
Department of Horticulture
Iowa State University
Ames, Iowa 50011-1100

A paper from the Proceedings of the 9th Metropolitan Tree Improvement Alliance (METRIA) Conference held in Columbus, Ohio, August 8-10, 1996.

Drought and Its Relevance to Arborist

Drought is a meteorological term denoting an extended period without precipitation. The productivity of 25% of soils in the United States is constrained by drought, making the absence of reliable soil moisture the leading edaphic limitation on agricultural productivity. Drought regions were defined for the United States by Warrick et al. (1975). The midcontinental region between eastern Colorado and western Illinois represents a transition zone between the humid eastern region and the arid western zones. Serious droughts are more common in the midcontinental region than in the eastern region (Rosenberg, 1978). Historical weather data show that mean annual precipitation east of the Rocky Mountains generally increases from west to east, and from north to south in about the eastern third of the country.

In most agricultural contexts, drought is defined in terms of a negative impact on crops and livestock. Crop losses are a complex function of not only the quantity of precipitation, but also seasonal pattern of precipitation and the quantity of precipitation received per event. The type of crop and its inherent resistance to drought, the soil type and its water-holding capacity, and other climatic features, including humidity, temperature, and solar radiation, also influence the impact of drought. Also, plants stressed by drought may have diminished resistance to pestsand pathogens that can cause further damage. Although many variables are involved, the relative uniformity of crop genotypes and field conditions makes prediction of the impact of drought on agronomic crops relatively straightforward compared with anticipating the impact of drought on trees in urban settings.

As for farmers producing low-intensity agronomic crops, what constitutes a drought to individuals who produce and manage trees varies with tree species and with soil and atmospheric conditions. Major nursery-production areas of the United States are located where serious droughts are rare, and many nurseries are equipped with irrigation equipment. Hence, the impact of drought on ornamental crops often is most serious after plants are installed in the landscape. As considered in depth by Craul (1992), soils along city streets, in new housing developments, and at commercial construction sites are extremely variable. Often these soils are of poor physical and chemical quality and lack the capacity to retain water. The ratio of root system to shoot system is low for many newly installed trees. Irrigation of municipal trees is rare, and space for root systems to spread to areas where soil moisture supplies are favorable frequently is scant. Compounding these factors with the diversity of taxa used in ornamental landscapes makes it difficult to envision obtaining the goal of uniformly healthy urban forests comprised of individual trees that obtain near-maximal potential life spans.

A manageable approach to dealing with this problem is to plant the right tree in the right location. This sounds simple. But matching trees and sites requires thorough knowledge of site characteristics and the capacity of various tree species thrive under those conditions. An evaluation of the water economy of street trees in New York City found tree water deficits occurred less frequently than presumed, and that water deficits were more closely linked to high evaporative demand than to limited water supplies (Whitlow et al., 1992). Additional information on the water required by trees to remain undamaged, particularly information linked to aerial and edaphic planting site characteristics, is needed. For most tree taxa, our knowledge of drought resistance is limited to field observations in uncontrolled settings. Enough research has been done, however, to illustrate that drought resistance varies greatly among tree species and among genotypes (cultivars, ecotypes, etc.) within species. My discussion first will center on traits of trees that influence drought resistance. Then I will provide an overview of current research we are conducting at Iowa State University on the drought resistance of black maple (Acer nigrum Michx.f.).

Drought Resistance

Many factors at the whole-plant, tissue, cellular, and molecular levels influence the resistance of woody plants to drought stress. Review of all factors is beyond the scope of this presentation. Rather, I will highlight a few factors that may be among the most relevant. The term drought resistance will be used broadly here to encompass various mechanisms that allow trees to tolerate drought-induced stress or to avoid becoming stressed in the first place.

The impact of drought stress on agronomic crops often is described by growth reductions and yield loss, or by measures of the physiological status of the plant, such as water potential. I argue that many such gauges of stress impact are not relevant for trees in the landscape. In the majority of landscape situations, growth reductions, even substantial ones, are tolerable. And in many landscapes, reduced growth is an advantage due limitations on space. Likewise, reductions in leaf water potential during drought might reduce fruit size or seed production, but will this be a serious problem? Probably not. What matters most for the majority of species in most urban landscapes are tree survival and the maintenance of undamaged foliage. This accounts for the importance researchers like John Pair of Kansas State University have placed on evaluation of Acer taxa for occurrence of leaf tatter in the Plains states. A 50% reduction in stem elongation might go unnoticed by most observers of urban trees, but desiccation of 50% of the leaf surface of a street tree creates a serious visual impact.

Damage-causing stresses induced by water deficits can be avoided by trees in various ways. Among these so-called avoidance mechanisms are the development of high root-to-shoot dry mass or surface area ratios. Allocation of photosynthetic energy preferentially to root systems is a common response to drought. Root : shoot ratio of well-irrigated seedlings of thornless honey locust (Gleditsia triacanthos L. var. inermis Willd.) was 0.21, whereas the ratio for seedlings subjected to drought was 0.53 (Graves and Wilkins, 1991). Certain tree species develop extensive root systems as seedlings before shoot growth is favored regardless of soil moisture status. A large, fibrous root system maximizes the portion of available soil water that can be obtained, thereby extending the time between precipitation events that trees can avoid damage. Unfortunately, species that allocate resources to roots rather than shoots early during ontogeny may not be favored in the nursery industry, where trees that develop a strong and vigorous canopy rapidly are prized for economy of production.

Another root-related avoidance mechanism may be adjustment of the resistance encountered by water as it travels radially from the soil through the root toward the xylem conducting elements. This resistance can be substantial, as illustrated by the wilted leaves of plants in moist soil on dry, sunny days. The impermeable Casparian strip in endodermal cell walls necessitates passage of water through cell membranes. The chemical composition of those membranes can affect the resistance to water flow (Markhart et al., 1980). Graves et al. (1991) found that roots of tree-of-heaven [Ailanthus altissima (Mill.) Swingle] grown at 34C had lower hydraulic conductivity coefficients than roots grown at 24C; roots of thornless honey locust, showed no difference, however. There is also some evidence that drought affects resistance to water transport of roots of woody plants (Ramos and Kaufmann, 1979). Root hair development, which varies tremendously among woody taxa and with edaphic conditions, also may influence the capacity for roots to obtain adequate water.

Relatively little resistance is encountered as water travels axially through vascular tissue to leaves. Rates of water transport can be about 10 times higher in ring porous (e.g. species of Celtis, Fraxinus, Gleditsia, Quercus, Robinia, and Ulmus) than in diffuse porous (e.g. species of Acer, Betula, Fagus, Liriodendron, Platanus, and Pyrus) woody plants. Total transport may be similar, however, because fewer vascular elements (only the most distal) are engaged in transport in ring porous species. Foliar resistance to water loss can be a major source of drought resistance. Numerous foliar traits that influence water use are evident. These include total surface area and thickness of the lamina, cuticle thickness and composition, various stomatal characteristics, pubescence, lobing, and lamina orientation with respect to solar radiation and wind (stance), and cell size and density. Data of Abrams et al. (1990) illustrate how foliar traits vary in populations of green ash (Fraxinus pennsylvanica Marsh.). Leaves from a relatively dry environment in South Dakota were smaller, thicker, and had a greater mass per unit area than leaves from more easterly populations. Physiologically, laminae (and other tissues) may adjust osmotically or show changes in the partitioning of water between the symplast and apoplast. These may enhance the capacity of the tissue to maintain turgor at low water potentials. Among the ornamental genera in which osmotic adjustment can occur in response to drought are Prunus (Ranney et al., 1991), Magnolia (Nash and Graves, 1993), while the genera in which increases in apoplastic water have been found include Acer (Nash and Graves, 1993) and Pseudotsuga (Joly and Zaerr, 1987). Woody species with elastic cell walls may be able to maintain turgor by accumulating solutes passively during drought (Evans et al., 1992). The text by Jones (1992) provides much more detail on these topics.

A Few Examples of Variation in Drought-Resistance Traits of Tree Genera

Eastern and Mexican redbud. Foliar traits of eastern redbud (Cercis canadensis L.) from its broad geographical native range were described by Donselman and Flint (1982). Leaves from the western part of the range lacked acuminate tips, and thickness increased linearly with longitude of origin. Stomatal density increased with increasing net precipitation of the origin. In addition, leaves from western sources had more pubescence on the abaxial surface than leaves from the east. Subsequent work has shown Mexican redbud [C. canadensis L. var. mexicana (Rose) M. Hopk.], which is native to west Texas and Mexico, has leaves that are about twice as thick as leaves of eastern redbud (Tipton and White, 1995). Leaves of the Mexican variety also had thicker cuticles and contained more water per unit area than leaves of eastern redbud. Water loss per unit leaf area of the two taxa did not differ, but the smaller leaf area of Mexican redbud represented an adaptation to drought.

Red maples. Townsend and Roberts (1973) compared the effects of water stress on seedlings of red maple (Acer rubrum L.) grown from seed collected in wet and dry habitats in Virginia, West Virginia, New Jersey, and Georgia. Longer droughts were required to induce wilting of leaves of seedlings from dry sites than in those from wet sites. Plants from wet sites also transpired at higher rates than those from dry sites. In contrast, Abrams and Kubiske (1990) reported that red maple plants from bog environments in Pennsylvania had less foliar gas exchange than plants from nearby upland habitats as drought progressed. But, consistent with Townsend and Roberts (1973), bog plants had less negative shoot water potentials at incipient wilting than red maples from upland areas. No evidence was found for osmotic adjustment for red maples from most sources. Likewise, Nash and Graves (1993) did not observed osmotic adjustment in drought-stressed 'Franksred' (trademark Red Sunset) red maple.

Species of Prunus. Drought resistance of Prunus species was compared by Rieger and Duemmel (1992). The most xeric species, P. andersonii A. Gray (Nevada Desert almond), had the smallest leaves, the least surface area per unit of leaf mass, and the highest stomatal conductance when not drought stressed. It root system also was large relative to it leaf area. Water use efficiency (carbon dioxide assimilation/transpiration) increased during drought for P. andersonii and P. subcordata Benth. (Sierra plum), which is native to mesic and xeric areas of northern California. In contrast, water use efficiency of the relatively mesic P. maritima Marsh. (beach plum), P. persica (L.) Batsch. (peach), and P. tomentosa Thunb. (Nanking cherry) decreased during drought. Osmotic adjustment in response to drought was demonstrated in P. avium L. P. pseudocerasus Lindl. 'Colt' and P. cerasus L. 'Meteor' by Ranney et al. (1991). Variation between the cultivars in the leaf osmotic potential at full turgor indicated that cultivars with low osmotic potential could be selected and evaluated for superior drought resistance.

Ongoing Research on Black Maple at Iowa State University

Background. Sugar maples (Acer saccharum Marsh.) are highly valued trees for urban landscapes in eastern North America. But leaves of several cultivars are prone to damage from drought stress and wind, reducing the usefulness of sugar maple at stressful landscape sites, particularly in the midwestern and southern United States. Black maple has been treated both as a subspecies of sugar maple and as a separate species. George Ware of the Morton Arboretum reviewed the knowledge about the potential for black maple as an urban tree during a previous METRIA meeting. He described this taxon as "a taxonomic and ecological entity that is not simply and easily separable from other sugar maples, yet it represents a way of designating the populations of maples in Iowa and nearby areas." See Ware (1983) for a summary of interesting background information on black maple and taxa allied with eastern sugar maple. Noting that trees of black maple planted at the Morton Arboretum developed extensive root systems while top growth was slow, Ware proposed that black maples may be particularly well-adapted for urban conditions further east in North America.

Leaves of black maple differ from leaves of sugar maple by having primarily three lobes, stipules, pubescence on the abaxial surface, a leathery texture, and a droopy stance. Black maple is indigenous further west in North America than sugar maple. Populations are found as far west as Minnesota, Iowa, Kansas, and Arkansas. There are no sugar maples native in central Iowa. But black maple is common, and the Iowa State University campus has numerous fine specimens that are either indigenous or were planted on campus during the nineteenth century from local populations. There are reports of indigenous trees in South Dakota, but we have been unable to substantiate this. Its foliar traits and occurrence further west where rainfall is relatively low have led to speculation that black maple is more resistant to drought and heat stress than sugar maple. Yet only one cultivar of black maple, 'Greencolumn', is common in the nursery industry.

Iowa State University is located in Ames, near the center of the state in Story County. All indigenous hard maples in this area are black maples. Old trees of black maple can be found in various locations on campus, including Pammel Woods, a native forested area that forms the northwest corner of campus. Black maple is also a common street tree in older residential neighborhoods in Ames. From a moderate distance, the most distinguishing feature of the local black maples is the laminar stance. Although somewhat variable, the two side lobes of the three-lobed laminae often are positioned at 90 angles to the middle lobe, creating a stance that has been described as droopy. This leaf orientation is not seen on the sugar maples that have been imported to this area. It is reasonable to consider this laminar stance a putative drought resistance mechanism because it almost certainly buffers air disturbance around the stomates and adds substantially to the boundary layer resistance of the abaxial surface. A survey I conducted on July 15, 1996, of leaves of trees on campus revealed that leaves in the dense shade and relatively low temperatures in Pammel Woods had the erect leaf stance typical of sugar maples. This suggests that the lamina stance trait is somewhat plastic. The dense pubescence of black maples in central Iowa also would increase resistance to water loss, but we have no evidence that trichome density varies with site conditions.

An initial study in our lab quantified the observations by Ware (Graves, 1994). We compared three half-sib seedling populations of black maple from central Iowa with three half-sib seedling populations of hard maples from eastern Iowa or Minnesota. Seedlings from further west developed higher root-to-shoot ratios. Also, leaves of black maple from central Iowa had a higher specific mass (mass per unit surface area) than leaves of hard maples from eastern Iowa or Minnesota.

Foliar traits of hard maples from 71 to 94W longitude. Ware had noted that indigenous populations of sugar maple in northern Illinois display traits intermediate between black maple and eastern sugar maple. So we wished to expand our knowledge of traits of hard maples by examining leaves from a broader range. Graduate Research Assistant Rolston St. Hilaire defined a study area near the 43N latitude as far west as central Iowa (94W longitude) to as far east as southern Maine (71W longitude). This area was chosen primarily because it would include the most westerly populations of black maple. St. Hilaire identified 24 locations in the defined area (in the states of Iowa, Wisconsin, Michigan, New York, Vermont, New Hampshire, and Maine) where indigenous hard maples could be obtained. By working with cooperators for most locations, St. Hilaire received shoot samples from 10 trees per location. Cooperators were instructed to collect the samples from a group of trees that represented hard maples typical of their area. They were also asked to provide shoots exposed to direct sun to minimize variation due to degree of radiation exposure.

After rehydration, samples at prescribed areas on selected laminae were used to determine specific mass of leaves, and additional samples were fixed for subsequent microscopic analysis. Total surface area, partitioning of surface area among lobes, and density of trichomes then were determined for representative leaves. St. Hilaire also noted whether the foliar samples most closely resembled sugar maple or black maple based on lobing, texture, pubescence, stipules, stance, and other traits. He classified all samples from central Iowa as black maple. But both black maple and sugar maple were obtained from eastern Iowa, Wisconsin, and Michigan. All samples from New York, Vermont, New Hampshire, and Maine were considered sugar maple.

Data analysis has shown relationships between foliar traits and the longitude where the sample originated. For example, increases in total leaf surface area and the density of trichomes on the abaxial surface with increasing longitude were found. The largest and most pubescent laminae were found at the most-westerly locations in central Iowa. Leaves from central Iowa had nearly 1000 trichomes per square cm of leaf area and had surface areas as high as six times the area of the smallest leaves from the eastern United States. Some leaves from the eastern United States had fewer than 10 trichomes per square cm. Another finding was that the percentages of total lamina surface area partitioned in the various lobes varied with longitude. Specific mass differed among leaves from various sampling locations. Weather data obtained from meteorological references and that was provided by cooperators showed that mean annual precipitation decreased with increasing longitude. Thus, lamina size, trichome density, and the partitioning of lamina surface among lobes can be related to precipitation. Data on trichome morphology, stomate density, stomate size, total lamina thickness, and the width of the various cellular layers are now being collected. The entire study is being repeated in 1996 with additional locations within the defined geographical region. We plan to publish a complete report of our findings elsewhere.

Development of seedlings of black maple and sugar maple. Cooperators were asked to provide fruit from indigenous trees in the defined area during the autumn of 1995. Although there was poor seed set in parts of this area, we were able to obtain several groups of half-sib seed from three regions, central Iowa, eastern Iowa, and the eastern United States (eastern New York and New England). Samaras were stratified during the winter. Seeds from half-sib sources from each of the three regions were sown in plastic pots during March of 1996. Seed germination and seedling development progressed under greenhouse conditions.

The young plants were irrigated with a fertilizer solution every 2 days until May, when a drought-stress experiment was initiated. Seedlings from each of the various sources have now been assigned randomly to a drought and a control treatment. On the day treatments began, randomly chosen plants from each treatment group were destructively harvested to assess seedling traits at the time of treatment initiation. Shoot development was most advanced among seedlings from the eastern United States. Specific mass of lamina also differed among regions. Generally, data from this initial harvest are consistent with trends in the data from our pilot study (Graves, 1994). Also, these preliminary growth data appear to agree with observations by Ware (1983) of preferential allocation to the root system among young plants of black maple. The question of whether trends found in seedlings are consistent over time and with trends that might be observed in trees is an obvious concern. Consistent with our observations, Pair (1990) found trees of black maple (seedling origin and 'Greencolumn') had less height and caliper growth than trees of the other eight hard maple taxa he studied.

Literature Cited

Abrams, M.D. and M.E. Kubiske. 1990. Photosynthesis and water relations during drought in Acer rubrum L. genotypes from contrasting sites in central Pennsylvania. Funct. Ecol. 4:727-733.

Abrams, M.D., M.E. Kubiske, and K.C. Steiner. 1990. Drought adaptations and responses in five genotypes of Fraxinus pennsylvanica Marsh.: photosynthesis, water relations and leaf morphology. Tree Physiol. 6:305-315.

Craul, P.J. 1992. Urban soil in landscape design. John Wiley & Sons, New York.

Donselman H.M. and H.L. Flint. 1982. Genecology of eastern redbud (Cercis canadensis). Ecology 63:962-971.

Evans, R.D., R.A. Black, W.H. Loescher, and R.J. Fellows. 1992. Osmotic relations of the drought-tolerant shrub Artemisia tridentata in response to water stress. Plant Cell Environ. 15:49-59.

Graves, W.R. 1994. Development of seedlings of sugar maple and black maple irrigated at various frequencies. HortScience 29:1292-1294.

Graves, W.R. and L.C. Wilkins. 1991. Growth of honey locust seedlings during high root-zone temperature and osmotic stress. HortScience 26:1312-1315.

Graves, W.R., R.J. Joly, and M.N. Dana. 1991. Water use and growth of honey locust and tree-of-heaven at high root-zone temperature. HortScience 26:1309-1312.

Joly, R.J. and J.B. Zaerr. 1987. Alteration of cell-wall water content and elasticity in Douglass-fir during periods of water deficit. Plant Physiol. 83:418-422.

Jones, H.G. 1992. Plants and microclimate. 2nd ed. Cambridge Univ. Press, Cambridge.

Markhart, A.H. III, M.M. Peet., N. Sionit, and P.J. Kramer. 1980. Low temperature acclimation of root fatty acid composition, leaf water potential, gas exchange and growth of soybean seedlings. Plant Cell Environ. 3:345-441.

Nash, L.J. and W.R. Graves. 1993. Drought and flood stress effects on plant development and leaf water relations of five tree species native to bottomland habitats. J. Amer. Soc. Hort. Sci. 117:845-850.

Pair, J.C. 1990. KSU horticulture research center ornamentals testing program. Metropolitan Tree Improvement Alliance (METRIA) 7:31-34.

Ramos, C. and M.R. Kaufmann. 1979. Hydraulic resistance of rough lemon roots. Physiol. Plant. 45:311-314.

Ranney, T.G., N.L. Bassuk, and T.H. Whitlow. 1991. Osmotic adjustment and solute constituents in leaves and roots of water-stressed cherry (Prunus) trees. J. Amer. Soc. Hort. Sci. 116:684-688.

Rieger, M. and M.J. Duemmel. 1992. Comparison of drought resistance among Prunus species from divergent habitats. Tree Physiol. 11:369-380.

Rosenberg, N.J. (ed.) 1978. North American droughts. Westview Press, Boulder, Colorado.

Tipton, J.L. and M. White. 1995. Differences in leaf cuticle structure and efficacy among eastern redbud and Mexican redbud phenotypes. J. Amer. Soc. Hort. Sci. 120:59-64.

Townsend, A.M. and B.R. Roberts. 1973. Effect of moisture stress on red maple seedlings from different sources. Can. J. Bot. 51:1989-1995.

Ware, G.H. 1983. Acer saccharum subspecies nigrum: Meritorious midwestern maple. Metropolitan Tree Improvement Alliance (METRIA) 4:1-6.

Warrick, R.A., P.B. Trainer, E.J. Baker, and W. Brinkman. 1975. Drought hazard in the United States: A research assessment. Monograph of Inst. Behav. Sci. Univ. Colorado, Boulder.

Whitlow, T.H., N.L. Bassuk, and D.L. Reichert. 1992. A 3-year study of water relations of urban street trees. J. Appl. Ecol. 29:436-450.

Return to METRIA 9

Web Crafter: Anne S. Napier ~ Email:

Format updated July 24, 2009