Hydraulic Lift: A Review

Jeff Harrison

Writer’s comment: First, I must thank my instructor Pam Demory for her expert help and instruction. Her in-class exercises, selection of readings, and enthusiasm for the subject matter provided me with excellent guidance for writing this paper. A review paper like “Hydraulic Lift: A Review” (written for the scientific journal Oecologia) is designed to bring the reader up to pace with our current understanding of a particular scientific topic. No original research is necessary, and all of the information and ideas presented are a synthesis of the work done by other researchers. The most difficult—and probably the most critical—part in writing this paper was choosing references that accurately represent the much larger volume of research that completely describes the subject. Once I had the appropriate references, writing the paper was mostly a technical effort that combined Pam’s teachings and a little creativity on my part.
—Jeff Harrison

Instructor’s comment: English 104E is a course in writing scientific papers for (primarily) scientific audiences. The culmination of the course is a review paper: a critical synthesis of the research on a topic or problem in the student’s area of interest. Jeff Harrison’s achievement in this review essay is impressive. Recently, many researchers have studied the phenomenon of hydraulic lift, seeking ways to make better use of limited water supplies, yet no review essay existed. Jeff read everything he could find, synthesized all the research, and organized the information to clearly convey the scope and significance of current thinking about hydraulic lift.
        Scientific writing challenges students to find a style that communicates clearly and concisely and at the same time is technically precise and professional, something that—as Jeff found out—requires painstaking sentence-level editing. I only fully realized just how good he had gotten at this when I began to read one of the articles he reviews in this paper—with its turgid, tangled, and unnecessarily abstract prose, I turned with relief to Jeff’s much more clearly-written paper.
—Pamela Demory


         Hydraulic lift, a process by which some plants solve the problem of water deficit in the upper soil horizons, has been the subject of many recent studies by plant ecologists. These studies show that some shrubs and trees have the ability to move water from deep in the soil profile, where water potentials are high, to more shallow regions where water potentials are low and root transpirational demands are extreme. Evidence suggests that this phenomenon occurs mainly at night and is driven by the water potential gradient existing between the upper and lower soil horizons. As a result of this gradient, water enters the deeper tap roots, moves up the root system, and effluxes from the shallower roots into the soil. One benefit of hydraulic lift is clear—a greater supply of water in the upper soil horizons to support daytime transpiration. However, researchers propose that hydraulic lift also benefits the plant in other ways. These include increased rates of mineralization in the rhizosphere and the maintenance of fine roots that would normally desiccate during periods of drought. On a larger scale, the fact that many plants parasitize hydraulically-lifted water may have significant effects on species distributions within plant communities. Finally, with so many plants being “potential hydraulic lifters,” it is likely that hydraulic lift plays an important role in regional water balances and should be included in any model describing landscape hydrology and/or ecology.


         Many of the anatomical and physiological traits that plants exhibit are thought to be an evolutionary adaptation to adverse ecological pressures. Some of the more noted examples are unique growth forms that reduce light competition, specialized stigmas and anthers resulting from coevolution with insects, and creative strategies of seed dispersal in response to various ecological pressures. In each of these cases, the adaptation increases the plant’s chances of survival by shifting the balance of ecological stresses in favor of the plant.
         One such adaptation that has received increasing attention in the literature is a phenomenon termed “hydraulic lift” (Richards and Caldwell 1987). Hydraulic lift (HL) is a unique solution to a problem experienced by many plants. The problem is that soils often lack the water reserves necessary to supply the numerous shallow roots of local flora with enough water to meet daytime transpiration demands. In response to this water deficit, some plants—and possibly a majority of all plants (Dawson 1993)—have adapted the ability to lift water from deep in the soil profile, where it is abundant and yet difficult to access, to shallower regions, where it is scarce and readily used by fine roots. This process of water redistribution, if prevalent throughout the plant kingdom, has the potential to radically alter not only water balances but also ecological processes throughout much of the terrestrial realm. To facilitate better understanding and management of regions where this phenomenon is prevalent, we must first attempt to qualify and quantify the extent to which HL occurs.

Hydraulic Lift

         Hydraulic lift uses the plant’s deep taproots to withdraw water from the wet subsoil. In larger species, groundwater may be tapped directly (Dawson 1993). This water then travels up the root system to the plant’s shallow roots, where it effluxes into the dry topsoil. Richards and Caldwell (1987) have shown that this process moves water through the soil profile at a much faster rate than can be explained by soil capillary action alone.
         During daytime transpiration, the greatest water potential gradient in the plant exists between the roots and the leaf stomata. As a result of this gradient, water moves from the roots and exits the transpiring leaves. When the stomata close—at night or during prolonged cloud cover—water continues to flow into the deeper taproots but no longer exits the leaves. This results in turgor pressure that increases water potential within the plant body. Finally, when the water potential in the shallow roots reaches a certain threshold above the water potential in the surrounding soil (this threshold varies among species), water begins to efflux from the roots and into the soil (Baker and van Bavel 1986; Richards and Caldwell 1987; Caldwell 1990; Dawson 1993).
         Although Richards and Caldwell (1987) were the first to thoroughly investigate and document the HL phenomenon, the potential for deep root water uptake and subsequent water efflux from shallow roots has been hypothesized for at least 60 years (Breazeale and Crider 1930; Volk 1947; Bormann 1957). More recent work on HL (Caldwell and Richards 1989; Corak et al. 1987; Dawson 1993) has quantified many of the parameters that govern the process, but much research is still needed in order to determine the extent to which HL occurs throughout the plant kingdom. The exact mechanisms that govern influx and efflux also need to be elucidated. Until recently, most researchers (Baker and van Bavel 1986; Richards and Caldwell 1987; Corak et al. 1987; Dawson 1993) worked under the assumption that HL is passively driven by matric potential gradients in the soil. However, this may not be completely true: Schwenke and Wagner (1992) have suggested that the plant may perform a more active role in the transport process.
         Most of what is known about water transport during HL comes from Baker and van Bavel (1986), Richards and Caldwell (1987), Caldwell and Richards (1989), Caldwell (1990), Corak et al. (1987), Dawson (1993 & 1995), and Xu and Bland (1993). In their original study of Artemisia tridentata (Big Sagebrush), Richards and Caldwell (1987) showed that diel fluctuations in the water potential of the upper soil layers can be attributed directly to the nocturnal efflux of water from the shallow roots of study plants. Caldwell and Richards (1989) showed that cloud cover can also stimulate HL. In their 1987 study, Richards and Caldwell discovered that daytime transpiration potentials dropped 25-50% after the plant’s stomata were forced to remain open during the previous night. By keeping the stomata open, they circumvented HL, causing the upper soil layers to remain dry—not replenished after the previous day’s transpiratory draw. It was their conclusion that HL contributed a significant volume of water to the topsoil and that Artemsia tridentata used this reserve extensively.


         Aside from increased water reserves in the upper soil layers, hydraulically lifted water (HLW) may provide other benefits to the plant. One of these is the maintenance of fine roots in the upper soil layers (Richards and Caldwell 1987; Caldwell and Richards 1989). Since these roots exist in the soil’s mineral rich zone and perform most of the plant’s nutrient uptake, it is essential that the plant maintain them. In times when desiccation may threaten the survival of smaller roots, HLW may provide the moisture necessary to sustain life and subsequent function. A second benefit of HLW may be increased rates of mineralization (Richards and Caldwell 1987; Caldwell and Richards 1989). In the soil, mineralization is facilitated by the presence of water and microbes that combine in the rhizosphere to degrade organic matter into essential nutrients. Without HLW, it is likely that rates of mineralization would slow and plants would experience various nutrient stresses.


         How widespread is this phenomenon among different species? Richards and Caldwell (1987) and Caldwell and Richards (1989) studied Artemisia tridentata and Agropyron desertorum (Crested Wheatgrass) and found that both conducted HL. However, since the sagebrush roots extend 2.2 meters into the subsoil and the grass roots extend to only 1.7 meters, it is not surprising that Artemisia—with its access to lower soil horizons—lifts significantly more water than Agropyron (Caldwell and Richards 1989). Exploring HL in a larger species, Dawson (1993 & 1995) studied Acer saccharum (Sugar Maples). He found that this species, when full grown, has the potential to reach the water table with its taproots. Dawson discovered that trees taller than ten meters hydraulically lift and transpire groundwater exclusively, while trees less than ten meters tend to transpire only shallow soil water. Wan, Sosebee, and McMichael (1993) found that even the shallow-rooted species Gutierrezia sorothrae (Broom Snakeweed) conducted HL to a small extent. Gutierrezia, rooted to a depth of only sixty centimeters, hydraulically lifts about fifteen percent of the water it transpires, roughly half of that estimated for Artemisia tridentata (Wan, Sosebee, and McMichael, 1993; Richards and Caldwell, 1987).
         What determines the extent to which HL occurs within a species? As already shown, a plant’s ability to conduct HL seems to correlate directly with the depth to which its roots penetrate the soil. This may be entirely a function of increasing water potential with depth or there may be other factors at work. For example, Xu and Bland (1993) found that water efflux from sorghum roots varied among plants of different genotypes. Although their results were not conclusive, this study suggests that genetic variation may affect HL potential within a species.


         What keeps other plants from parasitizing HLW? The answer seems to be . . . nothing. Many researchers have found that HLW is parasitized extensively by surrounding plants (Bormann 1957; Caldwell and Richards 1989; Caldwell 1990; Dawson 1993). In his work with tomato plants, Bormann (1957) found that the amount of water available to a parasite plant depends on the distance from the donor plant and the number of donor plants present. Dawson (1993) also found that as distance to the donor plant increases, the water available to the parasite decreases. In his work on HL in Acer saccharum, Dawson discovered that soil moisture varied most within 2.5 meters of the tree with measurable rewetting (from HL) occurring out to five meters. Within this rewetting zone, Dawson found the greatest productivity closest to the tree. The most successful of these parasites used HLW as sixty percent of their total daily transpired volume. Other species showed differential success in using HLW (Dawson 1993). Of the many potential parasites, relatively few have been tested for their ability to parasitize HLW (Caldwell 1990; Dawson 1993). Although this is the case, the potential benefit of parasitism suggests widespread occurrence and the need for further research.

Conclusion: Ecological Significance

         If HL and HLW parasitism are as ubiquitous as initial studies indicate, the effects on landscape ecology and regional water balance could be considerable. Dawson (1993) states that HL can be a significant life-sustaining mechanism in mesic as well as xeric environments. Although the need for HLW in xeric biomes may be more widespread in both space and time, biomes characterized by mesic conditions may also benefit from HLW during long periods of drought. In support of this, Corak et al. (1986) showed that parasitized water can significantly prolong the life of the parasite plant during times of water stress. This conclusion was also supported by Dawson (1993) when he revealed that HLW can negate the effects of drought on parasitic species rooted near Acer saccharum.
         To conclude, it is clear that HL can drastically alter both plant-plant and plant-water interactions. Because HL also changes soil water and groundwater distributions, hydrological processes may change enough to significantly impact regional water balances. With evapotranspiration, surface vegetation, erosion, and local microclimates all potentially altered, changes in water balance add one more factor to the increasing list of potential feedbacks resulting from HL. Until further investigation of these interactions is conducted, their relative magnitude and occurrence will remain speculation.
         How should future research on HL proceed? Further documentation of the phenomenon is probably the first step. Dawson (1995) suggests that the next step be an attempt at extrapolative modeling. By gathering data at the individual and community levels and developing models that accurately scale these local processes to regional responses, researchers can develop an economic as well as practical approach to predictive modeling. An obvious problem with this type of model is its inability to accurately simulate ecological processes. In response to this problem, Dawson (1995) suggests using a variety of scaling methods that are verified, validated, and then applied to an assortment of test species. Sampling a wide array of species and increasing the resolution of our extrapolative models will undoubtedly be a solid first step in determining the impact of HL on ecosystems across the globe.


Baker, J.H., C.H.M. van Bavel. 1986. Resistance of plant roots to water loss. Agronomy Journal 78, 641-644.

Bormann, F.H. 1957. Moisture transfer between plants through intertwined root systems. Plant Physiology 32, 48-55.

Breazeale, J.F., F.J. Crider. 1934. Plant association and survival, and the build-up of moisture in semi-arid soils. Arizona Agricultural Experiment Station Technical Bulletin 53.

Caldwell, M.M. 1990. Water parasitism stemming from hydraulic lift: a quantitative test in the field. Israel Journal of Botany 39, 395-402.

Caldwell, M.M., J.H. Richards. 1989. Hydraulic lift: water efflux from upper roots improves effectiveness of water uptake by deep roots. Oecologia 79, 1-5.

Corak, S.J., D.G. Blevins, S.G. Pallardy. 1987. Water transfer in an alfalfa/maize association. Plant Physiology 84, 582-586.

Dawson, T.E. 1995. Determining water use by trees and forests from isotopic, energy balance and transpiration analysis: the roles of tree size and hydraulic lift. Tree Physiology 16, 263-272.

Dawson, T.E. 1993. Hydraulic lift and water use by plants: implications for water balance, performance and plant-plant interactions. Oecologia 95, 565-574.

Richards, J.H., M.M. Caldwell. 1987. Hydraulic lift: substantial nocturnal water transport between soil layers by Artemisia tridentata roots. Oecologia 73, 486-489.

Schwenke, H., E. Wagner. 1992. A new concept of root exudation. Plant, Cell and Environment 15, 289-299.

Volk, G.M. 1947. Significance of moisture translocation from soil zones of low moisture tension to zones of high tension by plant roots. Journal of the American Society of Agronomy 39, 93-106.

Wan, C., R.E. Sosebee, R.L. McMichael. 1993. Does hydraulic lift exist in shallow rooted species? A quantitative examination with the half-shrub Gutierrezia sarothrae. Plant and Soil 153, 11-17.

Xu, X., W.L. Bland. 1993. Reverse water flow in sorghum roots. Agronomy Journal 85, 384-388.