THE EFFECTS OF CHANGING WATER AVAILABILITY
ON THE PHOTOSYNTHETIC RESPONSE OF PLANTS IN
THE SOTOL-GRASSLANDS OF BIG BEND NATIONAL PARK, TEXAS
by
ERIN M. WALKER, B.S.
A THESIS
IN
BIOLOGY
Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE
Approved
Chairperson of the Committee
Accepted
)ean of the Graduate School
August, 2003 ACKNOWLEDGEMENTS
First, I would like to thank my major professor Dr. David Tissue for all of the time effort, and guidance he provided during the pursuit of my degree. 1 would also like to thank the other members of my committee. Dr. John Zak and Dr. Nancy Mclntyre for their help in shaping my final document, and to especially thank Dr. Mclntyre for the hours she devoted to helping perform the statistical analysis. This project would never have been possible without the hard work and dedication of my partner in crime, Traesha Robertson. You were by my side through thick and thin, good times and bad. Thank you for the countless hours of hard work in the field, when you helped me with my project when you did not have to, and for driving all those miles. I would also like to thank April Nesbit, Amber Nagay, Cecilia Hernandez, Kari Maren, and Jen Reisinger for all their hard work in the field, and all the other members of Dr. Zak's lab that helped haul water up to the site. Finally, I would like to thank all of my friends and family for their love and support over the last three years. Thank you. Mom and Dad, for always supporting my love of science. Rick, thank you for being my "voice of reason," my love, and my best friend. I could never have gotten this far without you. TABLE OF CONTENTS
ACKNOWLEDGEMENTS ii
ABSTRACT iv
LIST OF TABLES v
LIST OF FIGURES vii
CHAPTER
I. INTRODUCTION 1
Rainfall in Deserts 1 Physiological Response to Pulses 2 Utilization of Pulse Resources 3 Competition for Pulse Resources 4 Water Partitioning 5 Photosynthetic Pathway 6 Rooting Depth 9 Functional Types 9 North American Deserts 13 Big Bend National Park 14
II. SITE DESCRIPTION, EXPERIMENTAL DESIGN, AND MEASUREMENTS 16
Site Description 16 Experimental Design 18 Measurements 23
III. RESULTS 29 Experiment 1: Plant response to an overall increase in precipitation 29 Experiment 2: Plant response to changes in the timing and magnitude of precipitation pulses 31
ill IV. DISCUSSION 61 Seasonal Variation in Ts, gs and And 61 Response of gs and A„et to T's 62 Relationship between Watering Treatment and Asat 64 Summary and Conclusions 64
LITERATURE CITED 67
IV ABSTRACT
The Hadley Climate Model 11 predicts that Big Bend National Park will receive a 25% increase in both summer and winter rainfall over the next 100 years and that seasonal rainfall patterns will shift from frequent, small storm events to fewer, large storm events. The physiological responses of sotol {Dasylirion leiophyllum) and side- oats grama (Bouteloua curtipendula) to future predicted rainfall patterns were examined over the summer of 2002 through the winter of 2003 in the sotol-grasslands of the Pine Canyon Watershed in Big Bend National Park. Stomatal conductance and photosynthesis were measured for plants that received a 25% increase in seasonal precipitation in either the summer, winter, both the summer and winter, or that received no increase in seasonal rainfall. Stomatal conductance and photosynthesis were also measured for plants that received small, frequent rainfall pulses, rainfall pulses of moderate size and frequency, and large, infrequent rainfall pulses. The 25% increase in seasonal rainfall had no effect on stomatal conductance and photosynthesis for either D. leiophyllum or B. curtipendula, but manipulating rainfall pulses size and frequency did affect the physiological response of the two species. The increase in stomatal conductance and photosynthesis for D. leiophyllum when it received larger, less frequent rainfall pulses, and the favorable response of 5. curtipendula to rainfall pulses of moderate size and frequency suggests that the response of both species is dependent on their rooting habit. Because of its shallow roots, B. curtipendula relied on numerous small pulses to maintain soil water availability in the upper-most soil layers, whereas D. leiophyllum required larger pulses that percolate into its deeper root zone. These results suggest that both species will have a favorable response to the future rainfall patterns predicted by the Hadley Climate Model II. LIST OF TABLES
Repeated measures ANOVA for the effect of experimental treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received an additional 25% in seasonal rainfall. 38
R , equations, and p-values of the linear regression analysis performed between best-fit estimates of mean photosynthetic rate (i^mol CO2W/S) and soil water potential (MPa). 43
R , equations, and p-values of the linear regression analysis performed between best fit estimates of mean conductance rate (mmol HaO/m Is) and soil water potential (MPa). 44
Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the summer season in August 2002. 47
Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the winter season in March 2003. 48
Repeated measures ANOVA for the effect of experimental treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received rainfall pulse manipulations. 51
PROC GLM ANOVA for the effects of model, species, month, and treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received rainfall pulse manipulations. 51
R^, equations, and p-values of the linear regression analysis performed between best fit estimates of (a) mean photosynthetic rate ()j.mol C02/m^/s) and soil water potential (MPa). 56
R^, equations, and p-values of the linear regression analysis performed between best fit estimates of mean conductance rate (mmol HaO/m^/s) and soil water potential. 57
VI 10 Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the summer season in August 2002 60
11 Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the winter season in February 2003. 61
Vll LIST OF FIGURES
Experiment 1: Daily precipitation received by the individual plots during the study period, from June 2002 to March 2003. 27
Experiment 2: Daily precipitation received by the rainout plots during the study period, from June 2002 to February 2003. 29
Total monthly precipitation received at the Panther Junction weather station during the study period from January 2002 to February 2003. 35
Mean monthly minimum and maximum air temperature recorded by the Panther Junction weather station during the study period from January 2002 to February 2003. 35
Response ofD. leiophyllum for (a) stomatal conductance and (b) photosynthetic rates when given a 25% increase in seasonal rainfall during the summer season (June to September 2002) and winter season (December 2002 to March 2003). 36
Response B. curtipendula for (a) stomatal conductance and (b) photosynthetic rates when given a 25% increase in seasonal rainfall during the summer season (June to September 2002) and winter season (December 2002 to March 2003). 37
Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (no additional water), additional winter water, additional summer water and additional water in the summer and winter experimental treatments during the summer season. 39
Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (no additional water), additional winter water, additional summer water and additional summer and winter water experimental treatments during the winter season 40
Vlll CHAPTER 1 INTRODUCTION
Rainfall in Deserts Desert ecosystems make up a significant portion of terrestrial ecosystems; approximately one third of the Earth is covered by arid or semi-arid lands and are characterized by high temperature and radiation levels, low relative humidity, and precipitation input (Brown 1994, Phillips and Comus 2000, Huenneke et al. 2002). Perhaps the most defining characteristic of deserts is the amount of precipitation, with most receiving less than 150 mm of rain per year (Noy-Meir 1973). Precipitation is the primary abiotic factor that controls the activity of biological processes in all arid ecosystems (Noy-Meir 1973, Ehleringer et al. 1991). Deserts are characterized by low, highly variable precipitation levels (Noy-Meir 1973). Rainfall usually occurs as discrete events that are infrequent and unpredictable (Noy-Meir 1973). This ephemeral supply of water is defined as a "pulse" (Noy-Meir 1973, Whitford 2001, Weltzin and Tissue 2003). While water availability from rainfall pulses can last from hours to weeks, the pulse characteristics are dependent on many factors, including temperature, size of the rain event, soil type, and evaporative demand (Noy-Meir 1973, Whitford 2001). Organisms in these water-controlled systems have evolved many adaptations that enable them to utilize pulsed water resources and to survive the drought that occurs between pulses (Ehleringer et al. 1999, Gebauer 2000, Schwinning 2001). In general, during long dry periods that occur between pulses, the ecosystem remains in an inactive state. A rain event of a significant size will stimulate biological activity until the available water is exhausted. As water becomes more limited, activity will decline until the inactive state is reached and the system awaits the next pulse (Noy-Meir 1973). Though all desert organisms depend on water, primary producers are extremely dependent on water input for growth and reproduction (Ehleringer et al. 1991). Generally, plants respond to precipitation pulses with increased photosynthesis, growth. and reproduction, but such responses are quite variable and depend on many factors, including rooting habits, phenology and pulse size (Schwinning and Ehleringer 2001).
Phvsiological Response to Pulses During inter-pulse periods when water availability is low, plants may enter an inactive state to conserve water and allow survival (Noy-Meir 1973). Once increased soil moisture triggers the active state, there are several aspects of drought stress recovery that a plant must undertake before normal physiological functions can resume (Noy-Meir 1973, Yanetal. 2000). The first step to water stress recovery involves restoration of water uptake by the roots (North and Nobel 1992, Hopkins 1999, Yan et al. 2000). As drought conditions intensify, hydraulic conductivity of the roots decreases (Sperry 2000). This loss of water transport is often caused by the increased suberization of roots in response to drought stress (Yan et al. 2000). Rain roots that are produced within hours of a rainfall event are less suberized than established roots and can facilitate water uptake as hydraulic conductivity is re-established in the older roots (Nobel 1988). Another limitation to initial water uptake after a rainfall pulse is the occurrence of xylem cavitations (Sperry 2000). When large water deficits occur, it becomes increasingly difficult for the plant to maintain intact water transport from the soil to the leaves, and air embolisms can form inside the xylem tissue that block water flow (Hopkins 1999, Sperry 2000). These air pockets in the vascular system must be repaired before water transport can resume (Sperry 2000). After root water uptake has been re-established and xylem cavitations have been repaired, leaf re-hydration can occur (Kramer and Boyer 1995). Stomatal conductance increases following the recovery of leaf water potential (Sperry 2000). Stomatal regulation allows the plant to reduce water loss through transpiration, maintain turgor, and avoid dehydrative damage to cells (Sperry 2000). Stomatal closure is not necessarily regulated by leaf water potential but is often initiated by root detection of decreasing soil moisture (Hopkins 1999). When roots sense soil drying, they release the hormone abscisic acid (ABA), which is transported through the xylem to the guard cells and causes the stomates to close (Hopkins 1999). Stomatal closure reduces water vapor loss but also excludes CO2 from entering the leaf, thus, also reducing photosynthesis (Hopkins 1999). When sufficient soil moisture is present, stomates can re-open and photosynthesis can take place (Sperry 2000). The time it takes for a plant to respond to a rain event and resume physiological activity at pre-drought levels depends on several factors, such as root depth, precipitation pulse size and season (Ehleringer et al. 1999). Studies have shown that plants with shallow root systems, such as grasses, respond quickly to pulses; plant water potential and photosynthesis increase dramatically within days of receiving rainfall (Sala and Lauenroth 1982, Schwinning et al. 2002). On the other hand, deeper-rooted species, such as woody shrubs, tend to respond more slowly to precipitation pulses (BassiriRad et al. 1999). Shrubs may show a rapid increase in plant water potential, but the photosynthetic response can either lag days or weeks behind or never increase at all (Yan et al. 2000). In the last several years, there has been more focus on how arid and semi-arid plants respond to precipitation pulses, but many of these studies focus on large-scale patterns such as primary productivity and community composition (Ehleringer et al. 1999). Though some work has been done at the leaf level to investigate how photosynthesis responds to pulses, most experiments involve woody shrubs (BassiriRad et al. 1999, Gebauer et al. 2002). Information on the photosynthetic responses of plants in other fianctional groups is lacking. Short-term responses, such as photosynthesis, can ultimately be linked with responses at other levels, such as community and ecosystem levels, to gain a comprehensive understanding of desert plant and ecosystem processes.
Utilization of Pulse Resources Because moisture input in deserts is supplied in discrete pulses, plant available resources are highly available for only a brief period after the pulse (Gebauer et al. 2002). The rainfall pulses cause soil moisture to be quite variable in time and space, and arid land plants have evolved many adaptations to utilize different aspects of this variable water availability (Schwinning and Ehleringer 2001). Noy-Meir (1973) developed the '^pulse-reserve" hypothesis to explain how desert systems respond to rainfall pulses. While most environmental factors in arid ecosystems are fairly constant, precipitation has a discontinuous input. These discrete rainfall pulses cause biological processes to also occur in discrete pulses. The "pulse-reserve" hypothesis states that a production pulse will be triggered by a rain event of a sufficient size. Some of this production pulse is diverted into some form of a reserve. When a rain event triggers the next production pulse, this pulse is initiated from the reserve produced from the previous pulse. Noy-Meir (1973) gives an example for an annual plant: a precipitation pulse stimulates germination and growth of the plant. Most of this production is lost to consumption and mortality, but some production is stored in the seeds. The seeds serve as the reserves from which the next germination and production event can arise. Perennial plants have a similar response, where a rain pulse causes growth and reproduction. Perennial plants can divert production not only into seeds but also into storage organs and compounds in leaves. Future rain pulses trigger growth from seed reserves but also from reserves contained in the plant itself Noy-Meir (1973) states that this ability of plants to shift between an inactive state to an active state is highly advantageous in an environment with such unpredictable rainfall. It also explains how a system with such extreme short-term variability can maintain long-term stability.
Competition for Pulse Resources When plants are forced to share resources that are limited in supply, interference reactions arise between individuals, resulting in competition (Teughels et al. 1995). Plants can compete for many resources, such as light, nutrients, and water (Teughels et al. 1995). In deserts, light is usually not limited due to low cloud cover and high radiation levels (Noy Meir 1973). Nutrient resources, especially nitrogen, are only available for plant uptake when the soil is wet but may become depleted when soil moisture levels are high for extended periods (BassiriRad et al. 1999). These periods are associated with large rainfall pulses that are rare in frequency in desert systems (Sala and Lauenroth 1982). In most instances water availability, not nitrogen, is the critical factor that controls the growth and development of desert plants (Briones et al. 1998). Because water is so scarce in these ecosystems, it is expected that plants will compete for this limited resource, but the extent to which competition occurs in arid systems is debatable (Briones et al. 1998, Gebauer et al. 2002). Shreve (1942) suggested that plants in deserts may not compete with each other because the harsh environmental conditions would prevent the existence of plant densities that are high enough to result in interference between neighbor plants for soil resources. Later, Connell (1975) stated that reduced plant establishment from high seedling and juvenile mortality contributes to low plant densities that curb the role of competition. Despite these predictions, several studies (e.g. Ehleringer 1984, Fowler 1986, Sala et al 1989, Briones et al. 1998, Gebauer et al. 2002) have demonstrated the existence of interspecific competition, and it is now generally accepted that desert plants compete for water resources (Ehleringer et al 1999). Though the effects of pulse and inter-pulse periods on the competitive relationships of plants are poorly understood, it is thought that competitive interactions will vary as available resources fluctuate (Ehleringer et al. 1999, Gebauer et al. 2002). Competition will be highest during pulses when resource levels are high, decline as resources diminish, and play a small role between pulses (Goldberg and Novoplansky 1997). The level of competition is related to pulse frequency. Frequent pulses result in intense competition between plants, whereas less frequent pulses result in longer inter- pulse periods where survival is more dependent on coping with drought stress than on competing for water (Goldberg and Novoplansky 1997).
Water Partitioning Soil water in deserts is a highly heterogeneous resource, being differentiated in many dimensions (Noy-Meir 1973). Plants do not use every aspect of the soil water profile but instead specialize in using only a subset of the components of the soil water continuum (Schwinning et al. 2002). These adaptations allow niche differentiation that partitions limited water resources, reduces competition, and allows the co-existence of different species (Dodd et al. 1998, Gebauer and Ehleringer 2000). Two characteristics have been identified as being important in allowing a plant to use temporally and spatially distinct zones of soil moisture: photosynthetic pathway and rooting depth (Kemp 1983, Donovan and Ehleringer 1994).
Photosvnthetic Pathwav All plants conduct photosynthesis to use energy from the sun to fix inorganic CO2 into organic carbon in the form of sugars (Hopkins 1999). The complex chemical reactions that make up photosynthesis can be accomplished in different ways by different metabolic pathways (Hopkins 1999). The C3 pathway is ultimately the pathway all plants must use to fix carbon, but the C4 pathway evolved as a variation to this original pathway (Hopkins 1999). The C3 pathway incorporates carbon solely by the use of the Calvin cycle (Hopkins 1999). CO2 is fixed directly by Ribulose-l-5-bisphosphate-carboxylase oxygenase (Rubisco), and photosynthetic rates are strongly dependent on the internal CO2 concentration in the leaf mesophyll cells (Pearcy and Ehleringer 1984, Hopkins 1999). Because O2 can occupy the same binding site on Rubisco as CO2, competitive inhibition of carbon fixation occurs as intracellular CO2 concentrations decrease (Pearcy and Ehleringer 1984, Hopkins 1999). Instead of solely relying on the Calvin cycle to fix carbon, C4 plants posses a host of anatomical, biochemical and physiological characteristics that evolved as a CO2 concentrafing mechanism (Pearcy and Ehleringer 1984, Hopkins 1999, Sage 1999). The photosynthetic apparatus in the leaf is separated physically into two different tissues: the mesophyll and bundle sheath cells (Hopkins 1999). The C4 pathway utilizes the enzyme PEP carboxylase in the mesophyll cells to incorporate CO2 into C4 acids that are then transported into the bundle sheath cells (Pearcy and Ehleringer 1984, Hopkins 1999). There, CO2 is decarboxylated from the C4 acid and is fixed by Rubisco. The transport of CO2 from the mesophyll cells to the bundle sheath cells, as a C4 acid, results in the concentrafion of CO2 inside the bundle sheath cells that prevents O2 inhibition of Rubisco activity. Ultimately, this results in efficient carbon fixation at low atmospheric CO2 levels that is generally insensitive to O2 concentration (Pearcy and Ehleringer 1984). These fundamental differences between the two pathways have important consequences for physiological response to environmental conditions (Sage 1999). For example, the photosynthetic response of each pathway is temperature dependent (Pearcy and Ehleringer 1984). The optimum temperature for each pathway reflects the difference in temperature optima for the enzyme that is responsible for the carboxylation reaction (Pearcy and Ehleringer 1984). The C4 pathway utilizes PEPase, which generates photosynthesis rates that increase between the temperatures of 20 and 40°C (Pearcy and Ehleringer 1984). Photosynthetic rate in a C4 plant is generally not adversely affected by higher temperatures because the bundle sheath cells can maintain high intercellular CO2 concentrations that favor the activity of the carboxylase reaction (Pearcy and Ehleringer 1984, Hopkins 1999). The evolution of C4 plants in a hot, dry environment resulted in PEPase functioning most efficiently at higher temperatures (Sage 1999). On the other hand, photosynthetic rates in C3 plants are often temperature independent at moderate temperatures, but photosynthetic rates do not increase dramatically with increasing temperature (Pearcy and Ehleringer 1984). As temperatures increase, diffusion of CO2 within the leaf decreases, and carbon gain becomes limited by decreased efficiency of Rubisco that results in increased oxygenase activity and photorespiration (Pearcy and Ehleringer 1984). Whereas photosynthesis in C3 plants decreases in response to high temperatures, sensitivity to low temperatures limits C4 photosynthesis (Hopkins 1999, Sage 1999). Decreases in carbon gain begin to occur at temperatures below 20°C in C3 plants, and most C4 species will senesce in response to chilling (Pearcy and Ehleringer 1984). C3 and C4 plants also exhibit a difference in stomatal conductance that results in differing water use efficiencies (Pearcy and Ehleringer 1984). Water use efficiency (WUE) is defined as the ratio of the photosynthetic rate to the rate of transpiration (Pearcy and Ehleringer 1984). Since the C4 pathway can concentrate CO2 inside the leaf tissues, photosynthesis becomes CO2 -saturated at low CO2 concentration; therefore, increased stomatal conductance would result in increased water loss but not increased photosynthetic rates (Pearcy and Ehleringer 1984, Hopkins 1999). Thus, C4 plants generally maintain a high carbon gain at low levels of transpiration that result in a high WUE (Pearcy and Ehleringer 1984, Hopkins 1999). In C3 plants, CO2 saturation of photosynthesis does not occur until higher intercellular CO2 concentrations are reached, so higher stomatal conductances are required to saturate the site of carboxylation with CO2. Higher conductance rates lead to increased transpiration relative to photosynthetic rate, which translates to a lower WUE. In general, C4 plants have a higher ratio of photosynthesis to water loss than C3 plants, but water use efficiency also depends on the evaporative demand of the environment. The C4 pathway is more efficient at higher tempferatures that promote significant water loss in C3 plants, but a C3 plant active during cooler temperatures can have comparable or higher WUE than the C4 pathway due to lessened evaporative demand and lower rates of transpiration (Pearcy and Ehleringer 1984). It has been demonstrated that there is a seasonal displacement of activity between plants with C3 and C4 photosynthetic pathways in hot deserts (Kemp 1983, Pearcy and Ehleringer 1984). Each pathway confers different physiological advantages depending on temperature and the timing of rainfall events (Kemp 1983, Martin et al. 1991, Smith et al. 1997). The C4 pathway allows these plants to have enhanced survival under hot conditions and greater drought tolerance during the many dry inter-pulse periods that occur between summer rainstorms (Kemp 1983, Sage 1999). Because the C4 pathway confers a high photosynthetic capacity and a high WUE at high temperatures, species with this pathway can utilize and will benefit the most from precipitation that falls during the hot summer months (Kemp 1983, Pearcy and Ehleringer 1984). Since C3 photosynthesis can occur throughout the year, C3 plants can best take advantage of rain that falls during the winter and spring seasons since the pathway has optimal photosynthetic rates and water use efficiencies at cooler temperatures (Pearcy and Ehleringer 1984). Seasonal separation of activity due to temperature dependent physiological functioning allows temporal partitioning of soil moisture and resuhs in niche differentiation among desert plants (Pearcy and Ehleringer 1984).
Rooting Depth In arid lands, water availability at the different soil depths is dependent on the frequency in which they are wetted by rainfall and rate of water loss by evaporation and soil texture (Noy-Meir 1973). Small pulses of rain wet the shallow layers of soil (0-30 cm), but these pulses are transient and evaporate quickly (Noy-Meir 1973). Large pulses wet the deeper layers (30 cm +) where evaporative demand is reduced and soil moisture levels become more stable (Noy-Meir 1973). Plants have adapted to utilize these spatially different soil moisture zones by adaptations in their rooting depth and distribufion (Donovan and Ehleringer 1994). The use of spatially different water sources allows the co-existence of plants with different rooting depths as put forth by Walter (1971) in his "two-layer hypothesis." He proposed that soil water was partitioned into either the shallow or deep soil layers. Plant roots were likewise partitioned with the grasses having shallow roots and woody shrubs having longer, deeper roots. Most rainfall wets only the upper soil layers, making water solely available to the grasses. However, some rain, especially that in larger storms, does infiltrate into the deeper soil layers, making water primarily accessible to the woody shrubs. The grasses only utilize surface water and the shrubs primarily use deeper water do not compete directly for soil moisture resources and are able to co-exist. Many studies support Walter's "two-layer hypothesis" and have shown that this hypothesis is applicable to desert systems (Sala et al. 1989, Dodd et al. 1998, Gebauer et al. 2002).
Functional Tvpes Ultimately, water acquisition is dependent on plant functional type, which integrates aspects of physiology and phenology (Dodd et al.l998). These suites of character types affect root and shoot functions and determine how a plant takes up water from specific soil layers at certain times of the year (Noy-Meir 1973, Kemp 1983, Ehleringer et al. 1999, Schwinning and Ehleringer 2001). Plant functional type represents an optimal phenotype that best utilizes a specific soil moisture pattern, and three phenotypes emerge as the dominant functional groups in desert plant communities: shallow-, deep-, and even-rooted plants (Ehleringer et al. 1999, Schwinning and Ehleringer 2001). In arid lands, shallow rooted plants are represented by annual species, succulents, and herbaceous perennials, such as bunchgrasses (Beatly 1974, Ehleringer et al. 1999). This group's strategy is to use the small, short-lived rainfall pulses that wet the upper layers of the soil (Dodd et al. 1998, Weltzin and Tissue 2003). To maximize use of pulse resources, these plants must maintain a functional root system that can respond quickly to the availability of soil moisture (Ehleringer et al. 1999). They accomplish this by distributing their roots exclusively in shallow soil layers where soil water potential is highest during pulses (Schwirming and Ehleringer 2001). Herbaceous annual and perennial species must support high photosynthetic rates while pulse water is available. This results in a decreased root: shoot ratio (Ehleringer et al. 1999). A small amount of shallow root biomass is sufficient to supply the plant with adequate water for transpiration so the plant can allocate more resources to aboveground biomass that increases carbon gain (Schwirming and Ehleringer 2001). High rates of gas exchange are also achieved by possessing leaves with high conductance that have sensitive stomatal control to prevent desiccation as water resources diminish (Ehleringer et al. 1999, Weltzin and Tissue 2003). In addition to these traits, annual and perennial herbaceous plants also often possess the C4 pathway (Ehleringer et al. 1999). This photosynthetic pathway confers optimal photosynthetic rates and increased water use efficiency at higher temperatures (Kemp 1983, Sage 1999). Since the majority of the small rainfall events occur during the hot summer season, it is generally advantageous for shallow-rooted plants to possess this pathway (Kemp 1983). The combination of all of these characteristics results in plants with an opportunistic approach to water acquisition that allows them to complete their growth or life cycle solely from small pulse resources during a relatively short period of time (Noy-Meir 1973).
10 Succulents, such as cacti, also possess a shallow root system but respond differently to rainfall pulses than herbaceous species (Nobel 1988). Cacti possess a host of morphological and physiological adaptations that allow them to utilize small, frequent rainfall pulses (Nobel 1988, Schwinning and Ehleringer 2001) They possess succulent stem tissues that can store water for use during the dry inter-pulse periods to sustain physiological activity (Nobel 1988, Schwinning and Ehleringer 2001). Unlike herbaceous species that use the C3 or C4 pathway, succulents utilize the CAM photosynthetic pathway to accomplish carbon assimilation. This photosynthetic pathway partitions the light and dark reactions of photosynthesis temporally, in which the plant opens its stomates at night to assimilate carbon, which is then stored as malic acid (Nobel 1988, Hopkins 1999). Carbon fixation of the stored carbon takes place during the day while the stomates are closed (Hopkins 1999). Since the stomates are only open at night, the plant experiences greatly reduced rates of evaporation and loses less water to transpiration (Nobel 1988, Hopkins 1999). Though succulents can maintain greater water availability by water storage and reduced water loss, these characteristics do not necessarily result in rapid carbon gain (Hopkins 1999). The CAM pathway does not generally support photosynthetic rates as high as those commonly found in C4 plants, and as a resuh, carbon gain and growth is often slow in succulent plants (Hopkins 1999). Unlike herbaceous species that complete their growth or life cycle within one season's time, succulents may take years and centuries to accomplish the same task (Nobel 1988). Species exploiting water in deeper soil layers, such as shrubs, can access a more stable, longer lasting water resource that is replenished by the larger, more infrequent pulses (Dodd et al. 1998). These plants take a more "slow but steady" approach to water uptake, and the characteristics of this group reflect this strategy (Weltzin and Tissue 2003). Deep-rooted plants rely on stored inter-pulse moisture, and they must increase allocation to roots to resource (Schwinning and Ehleringer 2001). This increases the plant's root: shoot ratio, which improves plant water status but may limit carbon gain due to decreased allocation to leaf area (Schwirming and Ehleringer 2001). Smaller root: shoot ratios maximize carbon gain when soils are water saturated (Schwinning and
11 Ehleringer 2001). Though deeper soils have a more stable water supply, this water is rarely present at saturating levels (Noy-Meir 1973). As a result, higher amounts of root biomass are needed to take up lower but consistent levels of moisture to maximize carbon gain in the long term (Schwinning and Ehleringer 2001). Plants with deep roots do not have to maintain high rates of gas exchange, and as a result, they tend to have lower leaf conductance and less sensitive stomatal control (Ehleringer et al. 1999, Weltzin and Tissue 2003). The majority of plants in this group also possess the C3 pathway (Ehleringer et al. 1999). Though this photosynthetic pathway has reduced photosynthetic rates during the high temperatures of summer and decreased water use efficiency, it can operate at temperatures lower than the C4 pathway, and thus, it is beneficial for carbon assimilation throughout the year (Kemp 1983, Hopkins 1999). This suite of traits allows this long-lived functional type to use deeper water resources at a slow, well-regulated rate to maintain steady but lower levels of gas exchange and growth (Noy-Meir 1973, Ehleringer et al. 1999). Plants living in arid environments face the tradeoffs of either investing root biomass in shallow soils to utilize pulse resources, or placing their roots in deep soils to use stored moisture (Schwinning and Ehleringer 2001). Though there are two functional types that specialize in being one or the other, there is another functional type that has a dimorphic root system that has significant root biomass in both soil layers (Noy-Meir 1973, Ehleringer et al. 1999, Weltzin and Tissue 2003). This intermediate group has characteristics in common with both shallow- and deep-rooted plants (Ehleringer et al. 1999). They have high allocation to roots, which results in a high root:shoot ratio characteristic of the deep-rooted fiinctional type, but have the high stomatal control of the shallow-rooted functional type (Ehleringer et al. 1999, Weltzin and Tissue 2003). These characteristics allow the dimorphic root type to be a versatile water user that can switch its utilization between ephemeral rainfall pulses and deeper stored seasonal moisture depending on source availability, though high allocation to root production limits the benefits of this strategy (Schwinning and Ehleringer 2001).
12 The costs and benefits of each of these strategies varies with pulse size, timing and environmental conditions (Ehleringer et al. 1999). One generalist functional type that can take up water anywhere at any time has not evolved because the different characteristics of each functional type are associated with tradeoffs that prevent a plant from performing equally well under all possible conditions (Ehleringer et al. 1999, Schwinning and Ehleringer 2001). Each specific utilization strategy balances these tradeoffs, and enables the plant to best use the water resource characteristic of its environment (Schwinning and Ehleringer 2001).
North American Deserts North America is home to four major deserts located in the western and southwestern United States: the Sonoran, Mojave, the Great Basin and the Chihuahuan Deserts (Brown 1994). The Great Basin is the northernmost of the four, encompassing western Utah from the Wasatch Moutains to the Sierra Nevada in southern California, and from the Snake River valley in southeast Idaho south to the Colorado River (Mares 1999). The Great Basin Desert is characterized by high elevations, a bi-modal rainfall period, and is considered the largest and only cold desert in North America (Ehleringer et al. 1991, Mares 1999, Phillips and Comus 2000). The Mojave Desert is found in southeastern California, southern Nevada and northwestern Arizona and is considered the smallest North American desert (Mares 1999). The Mojave is characterized mainly by its shrub-dominated vegetational cover and winter rainy season (Mares 1999, Phillips and Comus 2000). The Sonoran Desert is an extensive arid region west of the Sierra Madre Occidental, covering the western half of Mexico and Baja California north to southeastern California and southwestern Arizona (Mares 1999). The Sonoran Desert is characterized by bi-seasonal rainfall and is considered the hottest of the four deserts (Mares 1999, Phillips and Comus 2000). Lastly, the Chihuahuan Desert is a warm inland desert that is centered in north central Mexico, extending into southern New Mexico and southwest Texas between the Sierra Madre Occidental to the west and the Sierra Madre Oriental to the east (Brown 1994, Mares 1999). The Chihuahuan Desert covers
13 approximately 500,000 square kilometers and is considered the largest of the three warm deserts found in North America (Brown 1994, Mares 1999). This desert is mainly dominated by shrubs but also has the highest cactus density in the United States (Mares 1999). The Chihuahuan Desert is home to three National Parks, including Guadalupe, Carlsbad Caverns and Big Bend National Park.
Big Bend National Park Arid land ecosystems are some of the most sensitive areas to changes caused by human-related activities or by shifts in climatic conditions (Ehleringer et al. 1999, Schlesinger et al. 1990). One such sensitive area is Big Bend National Park, a protected region of the Chihuahuan Desert in southwest Texas. Big Bend is the seventh-largest National Park in the United States and is home to many plant species of concern. Big Bend has also been listed as the United States' most endangered National Park, and studies have shown that chemical deposition from increasing air pollution has already impacted the park's ecosystems. Due to global climate change caused by rising CO2 concentrations in the earth's atmosphere, seasonal rainfall patterns in Big Bend may be altered. The Hadley Climate Model II predicts that the Big Bend area will receive a 25% increase in both summer and winter rainfall over the next 100 years (Easterling et al. 2000). In addition to a total increase in rainfall, seasonal rainfall patterns are predicted to shift from frequent, small storm events to fewer, large storm events (Easterling et al. 2000). Vegetation in arid regions like Big Bend National Park is highly dependent on rainfall, and any alteration in seasonal water availability could cause shifts in plant community structure (Weltzin and McPherson 2000). Since water has such a significant impact on desert plant communities, these changes in precipitation pulse patterns are likely to have a more prominent role in altering ecosystems in Big Bend in the future than other aspects of global climate change, such as dry deposition of pollutants or increasing CO2 concentration (Ehleringer et al. 1999). Understanding how arid ecosystems respond to climate change is not only important for future management of lands in Big Bend
14 National Park but for all desert regions since approximately 30% of the earth's surface is covered by arid or semi-arid ecosystems (Huenneke et al. 2002).
15 CHAPTER 11 SITE DESCRIPTION, EXPERIMENTAL DESIGN, AND MEASUREMENTS
Site Description
Studv Site Big Bend National Park is located in the Trans-Pecos region of southern Texas, adjacent to the border with Mexico. The park covers an area of over 3,240 km^ and represents the largest portion of the Chihuahuan Desert that is protected by the National Park System. My study took place in the Pine Canyon Watershed, located in the Chisos Mountains (103° lO" W, 29° 5" N) of Big Bend. The watershed extends in an easterly direction for 19 kilometers, covers 78 km^ of habitat with an elevational gradient from 790 to 2100 m. It encompasses five vegetation zones, with xeric scrub at the lowest elevation to oak-pine forests at the summit of the Chisos Mountains. The study site is located at an elevation of 1526 m within the sotol-grassland vegetation zone. This upland grassland is dominated by sotol (Dasylirion leiophyllum), side-oats grama (Bouteloua curtipendula), brown spine prickly pear (Opuntia phaeacantha), and nolina (Nolina texana). Pine Canyon Watershed and the sotol-grasslands were chosen as the study area because it is a designated Natural Research Area by the park, and it has been the site for previous studies. The watershed is also currently a part of a long-term project to monitor the impact of human activities on several National Park watersheds (Herrman and Stottlemeyer 1991, Herrman et al. 2000).
Model Plants
Dasylirion leiophyllum, and Bouteloua curtipendula were used as the model plants in this experiment. Bouteloua curtipendula is a perennial bunchgrass that possesses the C4 pathway (Sage 1999). It is a warm-season grass that grows 30 to 40 cm in height, and our excavations of the root system showed that it has numerous, fine roots that extend from 0 cm to about 8 cm below ground. Previous studies have showoi that warm-season desert grasses can resume physiological activity within 12 to 24 hours after
16 a rain event (Sala and Lauenroth 1982, Ehleringer 1991). Dasylirion leiophyllum is a C3 perennial evergreen rosette plant (Kemp and Gardetto 1982). Our excavations revealed that this large plant has a coarse fibrous root system that start approximately 8 cm below ground and extend 20 to 30 cm into the soil. The roots do not begin immediately under the soil surface because sotol possesses a partially buried stem. Studies show that shrub response to rainfall may be delayed by as many as 3 to 5 days following a rain event (Gebauer and Ehleringer 2000, Gebauer et al 2002). Since D. leiophyllum seemed to have a growth form that was intermediate between grasses and shrubs, I expected that its response time would be intermediate as well. Preliminary measurements showed that sotol exhibited increases in stomatal conductance and photosynthetic rates within one to two days after watering. Dasylirion leiophyllum and B. curtipendula were chosen as study subjects because they represent the dominant C3 and C4 plants in our research area and possess different phenologies and rooting patterns.
Environmental Characteristics
Rainfall. Average precipitation for the Chihuahuan Desert ranges from 200 mm to more than 300 mm per year, falling predominantly as highly intense storm events from summer thunderstorms (Brown 1994). Though 50 to 70% of the total yearly precipitation is received during a discrete summer rainfall period, the Chihuahuan Desert does receive an additional 20 to 30% of its rainfall as low intensity precipitation from Pacific frontal storms in the winter (Brown 1994). Big Bend National Park is also characterized by this bi-modal but unequal rainfall regime. Rainfall records for the Sotol-grasslands in Pine Canyon from 1996 to 2000 indicated that the average yearly rainfall for that area totaled 374 mm, with 134 mm falling during the summer months of June, July, and August, and 51 mm falling during the winter months of December, January, and February. Temperature. Like most deserts, high temperatures are characteristic of the Chihuahuan Desert, and summer temperatures of over 40° C are common (Brown 1994). Despite its southerly location, winters in the Chihuahuan Desert can be quite cold (Phillips and Comus 2000). There are no barriers to protect it from Arctic air masses, and
17 freezing temperatures can be expected in the winter, especially in the northern reaches of the desert (Brown 1994). Big Bend is located roughly in the central portion of the Chihuahuan Desert and experiences both high summer temperatures and cold winter temperatures. The mean high summer temperature is 33.7° C and the mean summer low is 19.3° C. Winter temperatures average a high of 17.3° C and a low of 2.4° C.
Experimental Design
Experiment 1: Plant response to an overall increase in seasonal precipitation Introduction. Water is the primary limitation to plant growth and development in desert ecosystems (Noy-Meir 1973). The Hadley Climate Model II predicts that within 100 years the Big Bend Area will experience a 25% increase in both summer and winter rainfall (Easterling et al. 2000). Though much is known about how plants cope with drought conditions, less is known about plant physiological responses to periods of favorable soil resource availability (BassiriRad et al. 1999). Increased availability of this limited resource may cause greater plant carbon gain, result in increased productivity, and lead to changes in community structure (BassiriRad et al. 1999, Schwinning et al. 2002). The objective of this experiment was to determine the effect of a 25% increase in both summer and winter rainfall on the physiological activity of D. leiophyllum and B. curtipendula plants. Predictions. Plant photosynthefic activity is closely tied to water availability (Hopkins 1999). When soil water availability becomes limiting, leaf stomates close to reduce water vapor loss, but this action also excludes CO2 from entering the leaf, reducing carbon assimilation (Hopkins 1999). When sufficient soil water is available, stomates re-open and photosynthetic activity resumes (Sperry 2000). Based on this information, I hypothesized that plants that received an additional 25% in precipitation would exhibit greater rates of stomatal conductance and photosynthesis than plants receiving ambient precipitation, in response to increased water availability. The seasonality of the additional rainfall would also impact photosynthetic activity. Since B. curtipendula is a C4 plant, it would be physiologically more active in the summer season
18 due to the higher temperature optima of the C4 pathway. Also, since B. curtipendula is dormant during the winter, it is phenologically restrained from using any precipitation increase that would occur during that season (Huxman et al. 2003). D. leiophyllum, which is a C3 and has an evergreen habit, would be able to maintain photosynthetic activity during the winter and utilize additional winter rainfall. Thus, B. curtipendula was expected to maintain the greatest physiological response when given a 25% increase in precipitation during the summer, whereas D. leiophyllum would maintain the greatest physiological response when given a 25% increase in precipitation during the winter. Plots. For this experiment, 24 plots were set up in the sotol-grasslands, with each plot measuring 1 meter by 0.5 meters. A metal frame that was buried approximately 5 centimeters into the ground defined each plot. The plots were divided between the two plant species of interest so that 12 plots contained one D. leiophyluum plant and 12 plots contained several bunches oi B. curtipendula plants. Other species of forbs and grasses were also present in the plots. Watering Regime. The predicted changes in future rainfall by the Hadley Climate Model II were simulated by adding 25% more water to the plots. There were 12 plots for each species, and each was treated with one of four possible watering treatments: three plots received 25% more water only in the summer (S), three plots received 25% more water only in the winter (W), three plots received 25% more water in both the summer and the winter (SW), and three plots did not receive any additional water (C) (i.e., these plots received only natural rainfall and were considered the control plots). For the summer 2002 watering period, the amount of additional water was based on average monthly rainfall recorded in the sotol-grasslands in Pine Canyon from 1996 to 2000. The average amount of total summer rainfall was determined, and 25% of that amount was added to the plots. For the winter 2003 watering period, the method for calculafing the amount of additional water was changed. Instead of using a fixed water addition that was based on historic rainfall totals, the new water addition amount was determined by measuring the cumulative amount of natural rainfall received by the site during the experimental period. To calculate the amount of exfra water, the amount of
19 precipitation between day 330 (first day of winter) and day 60 (last day of winter) was measured by rain gauges at the site, and 25% of that total was then added to the plots. This method provided the same relative increase in natural rainfall (25%), but it was dependent on the actual rainfall for that period. The watering treatment was modified because it was found that the fixed amount method produced reasonable additions during an average rainfall year, but would not produce accurate additions during wetter or drier than average years. The new method, based on observed rainfall totals during the experimental period, better captured the annual rainfall variability that occurs in this desert system. Watering Amounts. During the summer 2002 watering period, the extra water was added as three simulated storms scheduled at the begirming, middle, and end of the historical summer rainfall period (Figure 4). A total of 33 mm of extra water was added to the summer and summer/winter plots, with 11 mm added on July 30, August 14, and September 16, 2002. For the July and August dates, all water was hauled to the site in backpacks and then applied to the plots using watering cans. By mid-September, three water tanks capable of holding a total of 4000 L had been installed at the site. The Big Bend Fire Department filled all the tanks with water from their pump truck. The tank water was the source of water for the September watering event. The winter watering event took place March 1, 2003, with a total of 6.8 mm of water added to the winter and summer/winter plots (Figure 4). The water was added in a single event near the end of the historical rainfall period, with water from the tanks applied to the plots using watering
cans.
Experiment 2: Plant response to changes in the timing and magnitude of precipitation pulses
Introduction. In desert systems, physiological activity, especially photosynthesis, is dependent on water availability (Noy-Meir 1973). The Hadley Climate Model II predicts that rainfall patterns will shift from frequent, small pulse events to larger, more infrequent pulse events (Easterling et al. 2000). Differences in photosynthetic pathway.
20 rooting habit and phenology determine plant physiological response to rainfall pulse input, such that different functional groups of plants specialize in utilizing specific rainfall pulse patterns (Schwinning et al. 2001). Changes in rainfall pulse patterns may favor one functional group over another and lead to shifts in community structure (Weltzin and McPherson 2000). The object of this experiment was to determine the effects of different size pulses (small, medium and large) that occurred with varying frequency (high frequency (ambient), moderate frequency, and low frequency) on the physiological response of D. leiophyllum and B. curtipendula in both the summer and winter seasons. Predictions. It has been shown that shallow rooted plants are specialized to use small, high frequency rainfall pulses, whereas plants with deeper roots rely on larger but less frequent pulses (Schwinning et al. 2001). Based on this knowledge, we expected that B. curtipendida, which had quite shallow roots (8 cm), would exhibit greater stomatal conductance and photosynthetic activity under the ambient (control) treatment where the plants were exposed to small but frequent pulses. We expected that D. leiophyllum, which had a deeper root system (30 cm), would display the greatest stomatal conductance and photosynthetic activity in response to the medium and large pulses that fell more infrequently. In addition to rooting habit, photosynthetic pathway will also affect photosynthetic response (Hopkins 1999). Since B. curtipendula is a C4 plant, peak photosynthetic activity will occur in the summer, when hot temperatures will be most favorable for C4 photosynthesis; however, D. leiophyllum is a C3 plant, so its peak photosynthetic period should be during the cooler temperatures of winter. Thus, B. curtipendula should show the greatest physiological response to the high-frequency, small-amount ambient pulses during the summer season and D. leiophyllum should have the greatest response to the medium and large pulse treatments during the winter. Plots. This experiment consisted of 15 plots, with each plot measuring 1.0 meter by 1.0 meter. Each plot contained one D. leiophyllum plant in the center of the plot, which was surrounded by B. curtipendula plants. All plots contained other species of forbs and grasses. Rainout shelters were buitt over 10 of the 15 plots to exclude all
21 natural precipitation. When the experiment was first established for the summer 2002 measurement period, the shelters consisted of 1.0 meter by 1.0 meter wooden frames covered with a clear, UV resistant PVC plastic film; three-foot long re-bar poles pounded into the ground around the plot supported the frames (Owens 2001). Before the winter measurement period of 2003, the previous rainout shelters were replaced with larger shelters. The new shelters consisted of 2.0 meter by 2.0 meter wooden frames covered with the same clear film. The shelter provided a greater rain-free zone around the plots. Five of the plots were left uncovered and received natural rainfall (i.e., these were considered the control plots). Watering Regime. To simulate the predicted change in rainfall patterns by the Hadley Climate Model II, all natural rainfall was excluded from the sheltered plots which were watered by hand. This allowed us to change the timing and magnitude of the precipitation that the plots received; five plots received large-sized pulses at a low frequency and five plots received moderate-sized pulses at a moderate frequency. An additional five plots were uncovered and received natural rainfall, generally in frequent, small pulses. Experimental manipulation of pulse size and frequency took place during both the summer and winter seasons. For both seasons, the amount of water received by the plots was based on a 4-year average (1996-2000) of monthly rainfall amounts for Pine Canyon. Pulse size and frequency were based on a 3-year average (1999-2001) of recorded daily rainfall amounts for the sotol-grasslands. The twelve-week summer season (June, July, August) received an average of 134 mm of rain with an average of two rainfall events per week, based on the 1996-2000 and 1999-2001 rainfall records. Therefore, the large-pulse plots received a total of three, 44-mm events, with watering occurring once every four weeks. The moderate pulse plots received a total of six, 22- mm events; plots were watered once every two weeks. The control plots that received natural precipitation received a total of 134.6 mm for the entire twelve-week summer season with an average of two, 5.8-mm rainfall events every week (Figure 5). The winter season (December, January, February) received an average of 51mm of precipitation with an average of one rainfall event every 4 weeks, based on the 1996-
22 2000 and 1999-2001 rainfall records. Therefore, the large pulse plots received one, 51- mm event for the entire season (twelve weeks). The moderate pulse plots received a total of two, 25.5-mm events; plots were watered once every six weeks. The control plots that received natural precipitation received a total of 20 mm of rainfall for the entire season with an average of one, 5-mm event occurring once every three weeks (Figure 5).
Measurements
Photosynthesis Leaf level stomatal conductance (gs) and photosynthesis (Anet) measurements were taken with a Licor 6400 portable photosynthesis system. All measurements were taken with a clear-top chamber (6 cm'^) that utilized ambient light levels. Ambient temperature and humidity levels were also used, but CO2 levels inside the cuvette were set to 370 ppm using a CO2 mixer. The leaf was placed inside the cuvette and allowed to equilibrate for approximately five minutes before a reading was recorded. Since both plant species have long, narrow leaves, several leaves were lined up and enclosed in the chamber to maximize leaf area inside the chamber. During the summer months, photosynthetic measurements were taken between 800 and 1000 hrs CST before stomatal closure occurred due to the heat. During the winter months, photosynthetic measurements were taken from 1000 to 1200 hrs CST to allow the plants to become warm enough to initiate photosynthetic activity. All measurements were taken one day before and one day after watering events in the months of June, July, August, September, and December 2002 and in January, February and March 2003.
Light Curves Light (AQ) curves were taken to evaluate photosynthetic capacity of Dasylirion leiophyllum (sotol), and Bouteloua curtipendula (side-oats grama). The AQ curves were measured with the Licor 6400 using the closed-top chamber (6 cm^) with the red-blue light source, ambient humidity levels, and CO2 levels inside the cuvette set to 370 ppm.
23 The light levels used to determine the AQ curves were set at 2000, 1500, 1000, 800, 600, 500, 400, 300, 200, 150. 100, 50, 0 famol photons m"^ s\ starting at the highest irradiance and decreasing to the zero light level. As with the other physiological measurements, several leaves were lined up and enclosed in the chamber to maximize leaf area inside the chamber. AQ curves were used to determine if the photosynthetic capacity at saturating light (Asat), apparent quantum efficiency (AQE), light compensation point (LCP), and light saturation point (LSP) of the two species had changed as a resuh of the watering treatments. One AQ curve was taken for each species under each treatment, for both the summer and winter seasons. The AQ curves were taken at the end of each season, the day before the last watering treatment was applied and one day after watering. The AQ curves were analyzed using Photosyn Assistant (Dundee Scientific, Dundee, Scotland, UK). For each curve, this software uses equations to relate the effect of changing light levels to the plant's photosynthetic rate and use mathematical equations to calculate the above parameters (Parsons and Ogston 1995).
Soil Moisture Percent volumetric water content (VWC) of the soil was measured using a Fieldscout TDR 100 soil moisture meter (Spectrum Technologies). The probe measured the water content using two, 12 cm rods that were inserted into the soil in the center of the plot. All readings were taken at the same time as all gas exchange measurements, one day before and one day after watering. Since soil water potential (4^s) is the preferred unit for measuring soil moisture content in physiological studies, the % VWC readings were converted to MPa (Corry 1987). Soil was collected from the site and water was added to create samples of varying moisture levels. The samples were first measured by the Fieldscout TDR to give the % VWC of each sample, then each sample was then analyzed using an Aqualab thermocouple psychrometer that gave readings in terms of water activity (aw). Soil water activity was then converted to ^s using the equation (Corry 1987):
24 T=(8.314nna,v)/V,, where T is water potential measured in Mega pascals (MPa), T\s absolute temperature (°K), aw is the water activity reading given by the Aqualab, and Vw is partial molar volume of water (which is assumed to be 1.8 x 10"^ mVmol). The % for each sample was then compared to the corresponding % VWC measurement made by the Fieldscout TDR, and the results were graphed to yield an equation (y= 2.7381x +8.8571) to determine Ts for each % VWC measured by the Fieldscout TDR.
Environmental Parameters Several micro-meterological stations are located within the Pine Canyon Watershed, including one station located within the sotol-grassland site. This station is located approximately 100 meters from the study site, but data were not available from this station due to equipment damage caused by a bear. Rainfall and temperature readings (Figures 6 and 7) were instead gathered from the Panther Junction weather station which is approximately 8 kilometers northwest of the study site.
Statistical Analysis A univariate repeated measures ANOVA was used to test the effects of month, treatment, and species on soil water potential, stomatal conductance, and photosynthesis. When the repeated measures ANOVA showed significant effects, it was followed by a PROC GLM ANOVA and Tukey's HSD tests. The relationship between photosynthesis and stomatal conductance to soil water potential were evaluated by least squares linear regression analysis, and a univariate ANCOVA was used to determine whether there were significant differences between slopes of the regressions to determine if the response of photosynthesis and stomatal conductance had significantly changed. All treatment effects were considered significant when p < 0.05. The repeated measures ANOVA and PROC GLM ANOVA were performed using SAS (Version 8.2, SAS Inc., Gary, NC, USA, 2001) and the linear regressions and ANCOVA were performed using SPSS (Version 10.0, SPSS Inc., Chicago, IL, USA, 2003).
25 50 Control plots
40
E E, c 30 g to Q. O CD 20
10
0 I ill AL J_i_L 50 Winter water addition
40
E _E c 30
Q. O 0) 20
10
uJl ul J_i_f June July Aug Sept Dec Jan Feb Mar 2002 2003
Figure 1. Experiment 1: Daily precipitation received by the individual plots during the study period, from June 2002 to March 2003. * Denotes rainfall received as an experimental water addition, not natural rainfall.
26 50 Summer water addition
40
E 30 .9 "•*—B» •g. o 20 0)
10
LL u J_L_L 50 Summer/winter water addition
40
30 o
'g. o 20
* * 10
I., it i_i_L June July Aug Sept Dec Jan Feb Mar 2002 2003
Figure 1. (Continued) Experiment 1: Daily precipitation received by the individual plots during the study period, from June 2002 to March 2003. * Denotes rainfall received as an experimental water addition, not natural rainfall.
27 60 Control
50
c 40 E c o
a t 30 d o (D D. 20
10
0 •l.li J I I 60 Medium pulse
50
E E, 40 c o 30
20
10
0 60 Large Pulse
50
E 40 c g 30 Q. o 20 QI
10
May June July Aug Sept Dec Jan Feb 2002 2003
Figure 2. Experiment 2: Daily precipitation received by the rainout plots during the study period, from June 2002 to February 2003.
28 CHAPTER 111
RESULTS
Experiment 1: Plant response to an overall increase in seasonal precipitation Environmental data Total rainfall during the summer study period (July, August, and September) was 168.7 mm, almost exactly the same as an average summer season (169 mm). Winter (December, January, and February) rainfall totaled 27.2 mm, 61% less than the average winter amount (51 mm) (Figure 3). Air temperature during the study period remained close to seasonal averages (Figure 4).
Seasonal variation in Angt and g^ A repeated measures analysis of variance (ANOVA) was used to examine the effects that the watering treatments had on soil water potential (^'s), stomatal conductance (gs) and net photosynthesis (Anet)- This test determined that a 25% increase in seasonal precipitation had no significant effect (p > 0.05) on ^s, gs, or Anet over time for either plant species (Table 1). Although gs and Anet of both D. leiophyllum and B. curtipendula were significantly different between the summer and winter seasons (Figures 5 and 6), this difference was not affected by the different watering treatments.
Relationship between ^.,, gs and Anet Linear regression analysis and analysis of co-variance (ANCOVA) was used to examine the response of stomatal conductance (gs) and photosynthesis (Anet) to soil water potential (*Ps )• All regressions displayed a positive relationship: gs and Anet rates for both species increased as ^s increased (Table 3 and 4). The ANCOVA demonstrated that during the summer season the watering treatments did result in significantly different slopes for gs rates for D. leiophyllum (p=0.005) (Figure 7a). The S and SW watering treatments caused gs for D. leiophyllum
29 plants to become 40% more responsive to ^s than plants in the control (C) treatments that received no additional water. Though the response of gs changed, there was no corresponding treatment effect (p= 0.133) on the adjustment of the response of Anet to ^s (Figure 7b). During the winter, there was no significant treatment effect on the response of gs or Anet to 4^s (p> 0.05) (Figure 8a and 8b). For B. curtipendula during the summer season, the additional summer water did not have a significant effect on the response of gs (p= 0.255), but did result in a significant difference (p= 0.009) between the slopes for Anet response to Ts (Figure 9a and 9b). Anet for B. curtipendula plants that received additional summer water showed a mixed response: Anet for the summer (S) treatment became 12% less responsive while Anet for the summer/winter (SW) treatment became 28% more responsive to *Ts compared to the treatments that did not receive a 25% increase in summer rainfall. During the winter the response of both gs and Anet to ^I's were significantly affected by the watering treatments (p< 0.05): gs became 75% more responsive for the winter (W) treatment, 80% more responsive for the S treatment, and 57% more responsive for the SW treatment as compared to the C treatment (Figure 10a), while Anet for the W treatment became 85% more responsive, 81% more responsive for the S treatment, and 115% more responsive for the SW treatment, relative to the C treatment (Figure 10b).
Relationship between watering treatment and A^ Light (AQ) curves were used to calculate several photosynthetic parameters, such as photosynthetic capacity at saturating light (Asat ), apparent quantum efficiency (AQE), light compensation point (LCP), and light saturation point (LSP) (Table 5 and 6). Since only one curve was taken for each species in each treatment for each season, means for these calculated parameters could not be determined; therefore statistical analyses were not conducted, but trends were noted. During the summer season, trends indicated that for both D. leiophyllum and B. curtipendula Asat tended to be higher after watering took place than before watering (Figure 11). The AQ curves also indicated that the addition of 25% more summer rainfall
30 increased Asat for plants in the S and SW treatments to a greater degree than for plants that did not receive the additional summer rainfall (W and C) treatments (Figure 11). Again, for both species in the winter, plants tended to have a higher Asat after watering than before receiving a watering treatment, and the plants that received an increase of 25% in winter rainfall (W and SW treatments) had a higher Asat than plants that did not receive the additional water (S and C treatments; Figure 12).
Experiment 2: Plant response to changes in the timing and magnitude of precipitation pulses
Environmental data Total rainfall during the summer study period (June, July, and August) was 134.6 mm, almost exactly the same as an average year (134 mm). Winter (December, January, and February) rainfall totaled 20.2 mm, 61% less than the average winter amount (51 mm) (Figure 3). Air temperature during the study period remained close to seasonal averages (Figure 4).
Seasonal variation in Anet and gs A repeated measures ANOVA was used to examine the effects that the pulse manipulations had on *Ps, gs, and Anet. This test inducated that, for either species, there was no significant difference (p> 0.05) in Ts between the different pulse treatments (Table 7). However, for both D. leiophyllum and B. curtipendula, the pulse manipulations had a significant effect on gs, and Anet (p< 0.05) (Figure 15 and 16). The effect of the different pulses on the Anet of Z). leiophyllum displayed marginal significance (p= 0.0657), but it suggests that a slight effect may have occurred (Table 7). A PROC GLM ANOVA and Tukey's HSD were run to further examine the significant effects found in the repeated measures ANOVA (SAS Institute Inc. 2001). The PROC GLM ANOVA indicated that month and pulse treatment significantly affected gs, while
31 species, month, and pulse treatment significantly affected Anet (Table 8). For both plant species, the Tukey's HSD determined that the medium pulse treatment had the greatest effect on gs rates, but all three pulse treatments had the same effect on Anet Although a significant effect was found by the PROC GLM ANOVA (p= 0.0186), the more conservative Tukey's HSD determined that there was enough variance in Anet to remove the significant effect of the pulse manipulations found by the earlier, and less robust test. The Tukey's HSD also determined that the month of August had the greatest effect on gs rates while July had the greatest effect on Anet rates for both D. leiophyllum and B. curtipendula.
Relationship between Relationship between T.. gc and An^ Linear regression analysis and analysis of co-variance (ANCOVA) were used to examine the response of gs, and Anet to Ts. All regressions displayed a positive relationship: gs and Anet for both species increased as Ts increased (Table 9 and 10). The ANCOVA indicated that for D. leiophyllum during the summer season, the watering treatments did result in significantly different slopes for gs (p=0.016), with gs for the medium pulse treatment becoming 89% less responsive and gs for the large pulse treatment becoming 62% less responsive to Ts (Figure 17a). Although the watering treatments affected gs, there was no adjustment in the response of Anet to Ts (p= 0.104) (Figure 17b). During the winter, there was no significant change in the response of gs (p=0.408) or Anet (p=0.05) to Ts (Figure 19a and 19b). For B. curtipendula during the summer season, there was no significant change in the response of gs (p= 0.646) or Anet to Ts (p=0.538) (Figure 18a and 18b), nor was there a significant change during the winter for the response of gs (p=0.139) or Anet (p= 0.769) to Ts (Figure 19a and 19b).
32 Relationship between watering treatment and A^ Light (AQ) curves were used to calculate several photosynthetic parameters, such as photosynthetic capacity at saturating light (Asat ), apparent quantum efficiency (AQE), light compensation point (LCP), and light saturation point (LSP) (Table 15 and 16). Since only one curve was taken for each species in each treatment for each season, means for these calculated parameters could not be determined; therefore statistical analyses were not conducted, but trends were noted. For both species during the summer, trends indicate that Asat tended to increase after a pulse treatment than before the watering treatment was applied. The large pulse treatment tended to generate a higher Asat for D. leiophyllum and while the medium pulse treatment increased Asat for B. curtipendula (Figure 20). In the winter, Asat tended to be higher after a pulse treatment was applied than before the water addition. For D. leiophyllum, the large pulse treatment increased Asat while B. curtipendula showed little to no activity during the winter.
33 E
c o
Q, O
Q.
o
Jan Feb Mar April May June July Aug Sept Oct Nov Dec Jan Feb 2002 rainfall during study period 2003
Figure 3. Total monthly precipitation received at the Panther Junction weather station during the study period from January 2002 to February 2003.
«— Mean max tempi 35 *— Mean min temp I
30
25 o CD rs 20 CC 0) Q. 15
10
5
0 Feb April June Aug Oct Dec Feb 2002 Temperature during study period 2003
Figure 4. Mean monthly minimum and maximum air temperature recorded by the Panther Junction weather station during the study period from January 2002 to February 2003.
34 160 (a)
140
20 - 0 -* 1 1 1 1 1 1 "k "k ic * June July Aug Sept Oct Nov Dec Jan Feb Mar 2002 2003 12 (b) 10 (I en f- CO C/3 (1) C\J E r >. O C/J O o -•—o • o ? :; Q- E 3 4 Control Winter 2 - i Summer • Summer/winter 0 J L J L _i I I * * * * June July Aug Sept Oct Nov Dec Jan Feb Mar 2002 2003 Figure 5. Response ofD. leiophyllum for (a) stomatal conductance and (b) photosynthetic rates when given a 25% increase in seasonal rainfall during the summer season (June to September 2002) and winter season (December 2002 to March 2003). Values are mean + SE for three plants per treatment. * Denotes when a watering addition took place. 35 250 (a) 200 a) Co" m T^ ' a CJ 3 CM T3 E 150 f o O O X ! (Ti — E() 100 E om a w 50 June July Aug Sept Oct Nov Dec Jan Feb fMa r 2002 2003 Control Winter Summer Summer/winter in V e>n> O^ o O o "O Q. E 3 f I J !_ J L "^ -k -k -k June July Aug Sept Oct Nov Dec Jan Feb Mar 2002 2003 Figure 6. Response B. curtipendula for (a) stomatal conductance and (b) photosynthetic rates when given a 25% increase in seasonal rainfall during the summer season (June to September 2002) and winter season (December 2002 to March 2003). Values are mean + SE for three plants per treatment. * Denotes when a watering addition took place. 36 Table 1. Repeated measures ANOVA for the effect of experimental treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received an additional 25% in seasonal rainfall. Repeated measures occur on month. Treatments are control (ambient rainfall), 25% additional summer rainfall, 25% additional winter rainfall, and 25% additional summer and winter rainfall. Effects are considered significant when p< 0.05. Species Soil Water Potential Stomatal Photosynthesis Conductance D. leiophyllum F3=2.17 F3=2.14 F3= 0.90 p= 0.1691 p= 0.1419 p= 0.4817 B. curtipendula F3=2.14 F3= 1.37 F3= 2.33 p= 0.1730 p= 0.3195 p= 0.1509 37 (a) 160 (D O .-^ C T CO en 120 P O 10 (b) en • en CM CD ,_ en o o *-' 4 - o o £ i 2 - _L JL _L -2.5 -2 -1.5 -1 -0.5 Soil Water Potential (MPa) Figure 7. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (no additional water), additional winter water, additional summer water and additional water in the summer and winter experimental treatments during the summer season. Values are mean + SE for three plants per treatment. Linear regression analysis was performed between best-fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 38 100 (a) 80 CD O --^ d T CO en •4-t CM 60 O ex o X S "o 40 11 -4—• —B— Control 20 —• Winter —^- Summer --•- Summer/winter 11 10 9 (f) 1 CO U) OJ sz E 8 -2.5 -2 -1.5 -1 Soil Water Potential (MPa) Figure 8. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (no additional water), additional winter water, additional summer water and additional summer and winter water experimental treatments during the winter season. Values are mean + SE for three plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 39 JUU (a) 250 - (D O C ^^-7 ^ CO en 200 - C) (i ^/f '^ 3 f- n c () o C\ 150 o X - — CO •5 il CO h b F too o CO -B— Control 50 - -• Winter -^- Summer 1 1. -•- Summer/winter 12 10 ^ ^ en CO 8 e/j CO CD E r >> 0 b r0n 0 0 0 n E /| 3 -2 -1.5 -1 0.5 Soil Water Potential (MPa) Figure 9. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for B. curtipendula under the control (no additional water), additional winter water, additional summer water and additional water in the summer and winter experimental treatments during the summer season. Values are mean + SE for three plants per freatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 40 200 r a) o 150 - c v CTi ta l "o CO 1- b F o CO e— Control • Winter ^- Summer •- Summer/winter I 4 (b) • 3.5 • . 3 -- • <• ,' en CM 2.5 - CD 'c i • .< • ^ 2 ~- ^ o O T .1 ^ y' o o i k • ^ E 1.5 - • ^ Q_ ~~~~ ~ —^ « i k ~7 • ~ ^. • ^ k 0.5 - [ • [ 1 —1 II 1 -^' 0 ' , ^,',' • ^ * 1 1 1 1 -3.5 -2.5 -2 -1.5 -1 -0.5 0 0.5 Soil Water Potential (MPa) Figure 10. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for B. curtipendula under the control (no additional water), additional winter water, additional summer water and additional sunimer and winter water experimental treatments during the winter season. Values are mean + SE for three plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 41 Table 2. R , equations, and p-values of the linear regression analysis performed between best-fit estimates of mean photosynthetic rate (\imo\ CO2W/S) and soil water potential (MPa). * Denotes a significant difference (p< 0.05) between the slopes of the different treatments as determined by the ANCOVA analysis. Different letters next to the equation denote which slopes are significantly different; equations with the same letter do not have significantiy different slopes, equations with different letters have significantly different slopes. Treatment Species Season R^ Equation P-value Control D. leiophyllum Summer 0.115 y= 1.86X + 6.62 0.133 Winter D. leiophyllum Summer 0.211 y=3.56x+11.2 Summer D. leiophyllum Summer 0.159 y=2.7x + 7.61 Summer/winter D. leiophyllum Summer 0.265 y=3.96x + 9.88 Control D. leiophyllum Winter 0.463 y=1.93x + 8.75 0.05 Winter D. leiophyllum Winter 0.010 y= 0.97x + 8.8 Summer D. leiophyllum Winter 0.253 y=2.01x+10.7 Summer/winter D. leiophyllum Winter 0.041 y=1.52x + 8.76 Control B. curtipendula Summer 0.642 y= 1.89x + 7.6a 0.009* Winter B. curtipendula Summer 0.291 y=3.51x + 9.8b Summer B. curtipendula Summer 0.009 y= 1.66x + 5.9a Summer/winter B. curtipendula Summer 0.057 y=2.6x+ll.l b Control B. curtipendula Winter 0.123 y=0.11x + 0.5a 0.000* Winter B. curtipendula Winter 0.358 y= 0.73x + 2.6 b Summer B. curtipendula Winter 0.984 y=0.57x+1.7b Summer/winter B. curtipendula Winter 0.686 y=1.24x + 3.3c 42 Table 3. R". equations, and p-values of the linear regression analysis performed between best fit estimates of mean conductance rate (mmol HiO/m^/s) and soil water potential (MPa). * Denotes a significant difference (p< 0.05) between the slopes of the different treatments as determined by the ANCOVA analysis. Different letters next to the equation denote which slopes are significantly different; equations with the same letter do not have significantly different slopes, equations with different letters have significantly different slopes. Treatment Species Season R^ Equation P-value Control D. leiophyllum Summer 0.902 y=^0.03 x + 0.15a 0.005* Winter D. leiophyllum Summer 0.402 y= 0.01X +0.12 b Summer D. leiophyllum Summer 0.514 y=• 0.05x + 0.03 c Summer/winter D. leiophyllum Summer 0.502 y=• 0.05x + 0.04 c Control D. leiophyllum Winter 0.407 y= 0.02X + 0.01 0.617 Winter D. leiophyllum Winter 0.233 y== 0.02x + 0.01 Summer D. leiophyllum Winter 0.735 y== 0.007X + .005 Summer/winter D. leiophyllum Winter 0.005 y== 0.02x + 0.01 Control B. curtipendula Summer 0.000 y== 0.08x + 0.35 0.255 Winter B. curtipendula Summer 0.543 y== 0.03x + 0.25 Summer B. curtipendula Summer 0.111 y= O.Olx + 0.17 Summer/winter B. curtipendula Summer 0.319 y== 0.02x + 0.19 Control B. curtipendula Winter 0.832 y== 0.01x + 0.04a 0.047* Winter B. curtipendula Winter 0.580 y== 0.04x +0.15 b Summer B. curtipendula Winter 0.109 y== 0.05x + 0.02 b Summer/winter B. curtipendula Winter 0.582 y== 0.02x + 0.01 c 43 20 (a) • A • 15 A •^ •« • 8 >« en o 10 • • 2 o o I 5 "o "5 SI Q. E ,4 g A 3 Control before «5^ Control after o Winter before n • Winter after A Summer before A Summer after O Summer/winter before 35 • Summer/winter after (b) 30 • A , , 25 • en en A (/> CM • E A th e 20 c en O O U 15 O SI • , A _ Q_ m o Z! 10 • A • ^ O fl A • i ^ a 5 0 ii h _L 500 1000 1500 2000 2500 Quantum flux (mol photons m "^s"') Figure 11. Representative light (AQ) curves of (a) D. leiophyllum and (b) B. curtipendula at the end of the summer season 2002 under the control (no additional water), additional winter, summer, and both summer and winter rainfall treatments before and after the final watering event. 44 16 (a) 14 12 •^c:« I E 10 c ^ >> O 8 S O 6 D °- I O 4 O o 9 O A 2 o o o A A Control before 0 Hm- A_4_ _L_ Control after J_ o Winter before 12 • Winter after (b) A Summer before A Summer after O Summer/winter before 10 Summer/winter after en en 8 0) Cvl CD E r CM O b os y O O "o • • • ^^•aSS > 8 t 500 1000 1500 2000 2500 Quantum Flux (mol photons m '^s"') Figure 12. Representative light (AQ) curves of (a) D. leiophyllum and (b) B. curtipendula at the end of the winter season 2003 under the control (no additional water), additional winter, summer, and both summer and winter rainfall treatments before and after the final watering event. 45 Table 4. Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the summer season in August 2002. Watering Species •^sat AQE LCP LSP Treatment Before watering Control D. leiophyllum 8.0 0.062 1 139 Winter D. leiophyllum 14.9 0.039 13 387 Summer D. leiophyllum 12.5 0.065 9 202 Summer/winter D. leiophyllum 14.2 0.076 7 193 Control B. curtipendula 17.7 0.029 144 742 Winter B. curtipendula 19.4 0.044 87 529 Surrmier B. curtipendula 21.8 0.044 101 592 Summer/winter B. curtipendula 16.5 0.037 79 527 After watering Control D. leiophyllum 15.2 0.041 45 417 Winter D. leiophyllum 11.9 0.059 5 207 Summer D. leiophyllum 16.5 0.019 21 858 Summer/winter D. leiophyllum 20.4 0.054 12 387 Control B. curtipendula 26.0 0.037 47 477 Winter B. curtipendula 24.1 0.045 94 621 Summer B. curtipendula 35.7 0.053 60 111 Summer/winter B. curtipendula 46.4 0.056 74 894 46 Table 5. Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the winter season in March 2003. Watering Species Asat AQE LCP LSP Treatment Before watering Control D. leiophyllum 4.9 0.005 11 139 Winter D. leiophyllum 6.1 0.017 32 387 Summer D. leiophyllum 3.5 0.003 25 202 Summer/winter D. leiophyllum 5.2 0.005 14 193 Control B. curtipendula 1.4 0.002 244 742 Winter B. curtipendula 0.8 0.001 431 529 Summer B. curtipendula 1.0 0.002 191 592 Summer/winter B. curtipendula 1.7 0.004 167 527 After watering Control D. leiophyllum 18.8 0.191 32 417 Winter D. leiophyllum 11.2 0.024 68 207 Summer D. leiophyllum 10.5 0.023 24 858 Summer/winter D. leiophyllum 14.3 0.039 52 387 Control B. curtipendula 9.35 0.036 113 477 Winter B. curtipendula 0.2 0.001 65 621 Summer B. curtipendula 2.1 0.017 115 727 Summer/winter B. curtipendula 12.2 0.022 133 894 47 200 p (a) m o 150 t_ (0 ^ o (/) 13 (N •a E o O O 100 X m —I) •*—» E E om a C/3 50 i I _L JL _L * A A *AA*AA*AA *A A May June July Aug Sept Oct Nov Dec Jan Feb 2002 2003 12 r (b) Small pulse Medium pulse 10 \ Large pulse 8 si s en 0) CM b c: _, >> H o Oo -»—o • o sz E CL \ ^ \ 1 -lbUi - JI I1 I L J L *AA*AA*AA*AA *A May June July Aug Sept Oct Nov Dec Jan Feb 2002 2003 Figure 13. Response ofD. leiophyllum for (a) stomatal conductance and (b) photosynthesis during the summer season (June to August 2002) and winter season (December 2002 to February 2003). Values are mean + SE for five plants per treatment. * Denotes when the medium pulse treatment plots were watered. ^ Denotes when the large pulse treatment plots were watered. 48 350 (a) 300 a) to_ 150- m T- 3o CVJ E 200 o O O X en — 150- .«—• E( ) E om a Oi 100 50 ID 0 -9 10 1 1 _L ± *A^ A *AA*AA *AA *A A Nov Dec Jan Feb May June July Aug Sept Oct 2003 12 2002 r (b) Small pulse 10 Medium pulse Large pulse 8 en T en elon ' O O o 4 sz E a. 3 2 m ^ • 0 ! -2 J L J L J I I L *A A *AA*AA *AA *A A Nov Dec Jan Feb May June July Aug Sept Oct 2003 2002 Figure 14. Response ofB. curtipendula for (a) stomatal conductance and (b) photosynthesis during the summer season (June to August 2002) and winter season (December 2002 to February 2003). Values are mean + SE for five plants per treatment. * Denotes when the medium pulse treatment plots were watered. ^ Denotes when the large pulse treatment plots were watered. 49 Table 6. Repeated measures ANOVA for the effect of experimental treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received rainfall pulse manipulations. Repeated measures occur on month. Treatments are small pulse, medium pulse, and large pulse. Effects are considered significant (*) when p< 0.05. Species Soil Water Potential Stomatal Photosynthesis Conductance D. leiophyllum F3=1.57 F3= 9.67 F3= 3.44 p= 0.2479 p= 0.0038* p= 0.0657 B. curtipendula F3=2.13 F3= 69.70 F3= 6.08 p= 0.1613 p= 0.0001* p= 0.0167* Table 7. PROC GLM ANOVA for the effects of model, species, month, and treatment on soil water potential, stomatal conductance, and photosynthesis for plants that received rainfall pulse manipulations. Species are D. leophyllum and B. curtipendula. Treatments are small pulse, medium pulse, and large pulse. Effects are considered significant (*) when p< 0.05. Soil Water Potential: Model F9= 35.65 p= 0.0001* Species F,= 0.01 p= 0.9053 Month F6= 46.11 p= 0.0001* Treatment F2= 22.12 p= 0.0001* Stomatal Conductance: Model F9= 8.93 p= 0.0001* Species F,= 0.25 p= 0.6204 Month F6= 9.80 p= 0.0001* Treatment F2= 10.68 p= 0.0001* Photosynthesis: Model F9= 46.18 p= 0.0001* Species Fi= 33.08 p= 0.0001* Month F6= 62.42 p= 0.0001* Treatment F2= 4.02 p= 0.0182* 50 250 r 200 - CD O —^ C T CD en .^- CM 150 - |o O c\ O X o too - "cd E E o E 00 B— Control •• Medium pulse ^- Large pulse en csj CD •,- o O D. E 3 -2.5 -2 -1.5 -1 Soil Water Potential (MPa) Figure 15. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (ambient rainfall), medium pulse treatment, and large pulse treatment during the summer season. Values are mean + SE for five plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 51 bUU (a) 700 - []_ 0 -——^"^^ o ,—. 600 - [] ^^^-^ i C T CO en [] _ 500 - |o , - -' O cv 400 O X u"" * "cc "o 300 "CO E E E 9nn -B- Control 100 -• Medium pulse -^- Large pulse 1 L_, 1 8 r- ~ (b) y/ [] 7 - [ ] yt J rn 1 6 - • U) OJ 1 1 1 1 1 1 1 -3.5 -2.5 -2 -1.5 -1 -0.5 0 Soil Water Potential (MPa) Figure 16. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for D. leiophyllum under the control (ambient rainfall), medium pulse treatment, and large pulse treatment during the winter season. Values are mean + SE for five plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 52 (f) CVJ 'E o X •5 E E ^ E CM 2o ^o & i -2.5 -2 -1.5 -1 -0.5 Soil Water Potential (MPa) Figure 17. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for B. curtipendula under the control (ambient rainfall), medium pulse treatment, and large pulse treatment during the summer season. Values are mean + SE for five plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 53 40 35 ^ 30 en 'E 25 o CO 20 X -1 mo l 15 E, 10 5 4 p -B— Control 3.5 - -• Medium pulse -^- Large pulse 3 - en CM c 2 - ^'^ O g o 1.5 - Q- 5 1 - -L 0.5 - 0 I -I- I -3.5 -2.5 -2 -1.5 -1 -0.5 0 Soil Water Potential (MPa) Figure 18. Relationship between (a) stomatal conductance and (b) photosynthesis and soil water potential for B. curtipendula under the control (ambient rainfall), medium pulse treatment, and large pulse treatment during the winter season. Values are mean + SE for five plants per treatment. Linear regression analysis was performed between best fit estimates of mean photosynthetic rate and soil water potential and mean conductance rate and soil water potential. ANCOVA was then used to evaluate if the slope of the regressions were significantly different. 54 Table 8. R\ equations, and p-values of the linear regression analysis performed between best fit estimates of (a) mean photosynthetic rate (^mol COa/mVs) and soil water potential (MPa). * Denotes a significant difference (p< 0.05) between the slopes of the different treatments as determined by the ANCOVA analysis. Different letters next to the equation denote which slopes are significantly different; equations with the same letter do not have significantly different slopes, equations with different letters have significantly different slopes. Treatment Species Season Equation P-vaJue Control D. leiophyllum Summer 0.096 y=4.36x+ 14.8 0.104 Medium Pulse D. leiophyllum Summer 0.228 y=2.01x + 9.89 Large Pulse D. leiophyllum Summer 0.855 y=3.01x+11.6 Control D. leiophyllum Winter 0.0544 y=2.34x+ 11.1 0.968 Medium Pulse D. leiophyllum Winter 0.005 y=0.06x + 5.02 Large pulse D. leiophyllum Winter 0.143 y= 0.096x + 5.77 Control B. curtipendula Summer 0.044 y=3.80x+ 12.1 0.538 Medium Pulse B. curtipendula Summer 0.311 y= 1.81X + 8.51 Large Pulse B. curtipendula Summer 0.685 y=3.01x-FlO.O Control B. curtipendula Winter 0.129 y= 1.02X + 4.75 0.769 Medium Pulse B. curtipendula Winter 0.297 y=0.17x+ 1.68 Large Pulse B. curtipendula Winter 0.013 y= 0.006X + 0.02 55 Table 9. R , equations, and p-values of the linear regression analysis performed between best fit estimates of mean conductance rate (mmol H2OW/S) and soil water potential. * Denotes a significant difference (p< 0.05) between the slopes of the different treatments as determined by the ANCOVA analysis. Different letters next to the equation denote which slopes are significantly different; equations with the same letter do not have significantly different slopes, equations with different letters have significantiy different slopes. Treatment Species Season R^ Equation P-value Control D. leiophyllum Summer 0.976 y= 0.09x + 0.28 a 0.016* Medium Pulse D. leiophyllum Summer 0.616 y=0.01x +0.12 b Large Pulse D. leiophyllum Summer 0.814 y=0.03x +0.13 b Control D. leiophyllum Winter 0.296 y= 0.008x + 0.08 0.408 Medium Pulse D. leiophyllum Winter 0.098 y= 0.002x + 0.59 Large pulse D. leiophyllum Winter 0.037 y=0.01x +0.065 Control B. curtipendula Summer 0.445 y=0.07x + 0.31 0.646 Medium Pulse B. curtipendula Summer 0.648 y=0.057x + 0.21 Large Pulse B. curtipendula Summer 0.252 y=0.021x + 0.15 Control B. curtipendula Winter 0.240 y=0.017x + 0.06 0.139 Medium Pulse B. curtipendula Winter 0.268 y=0.001x + 0.02 Large Pulse B. curtipendula Winter 0.154 y= 0.006X + 0.02 56 14 (a) 12 k & i • • _ 10 % i D •-. ^en B • • en CM • ^ E 8 • s c ~ 5 >> O • o O 6 O "o « o ^ 31 4 1 2 0 1 1 1 1 1 D Control before • Control after 14 IV\ O Medium before (b; • Medium after A Large before 12 A Large after A A A A 10 1 s 4 •-. ^ tf) CM A • • o o .^ E 8 • 1 D • A D D tn >< • A o O 6 a • i D D "- 3 4 • 2 1 • 0 1 1 1 1 1 1 500 1000 1500 2000 2500 3000 Quantum Flux (mol photons m 2^-s h) Figure 19. Representative light (AQ) curves of (a) D. leiophyllum and (b) B. curtipendula at the end of the summer season 2002 under the control (ambient rainfall), medium pulse and large pulse treatments before and after the final watering event. 57 u (a) 8 A A A A m _cn CD 6 A A • >> A en O o o a o 4 • a "5 ft CL a E 8 3 i % 2 e A 0 n 1 1 1 J_ 1 Control before Control after (b) Medium before Medium after A Large before A Large after 0.8 •;? en 0.6 >c ^ en o O o O "5 0.4 .c CL E 3 0.2 a o 8 _l_ 200 400 600 800 1000 1200 Quantum flux (mol photons m '^s"') Figure 20. Representative light (AQ) curves of (a) D. leiophyllum and (b) B. curtipendula at the end of the winter season 2003 under the control (ambient rainfall), medium pulse and large pulse treatments before and after the final watering event. 58 Table 10. Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the summer season in August 2002 Watering Species •^sal AQE LCP LSP Treatment Before watering Control D. leiophyllum 13 0.0632 31.8 237 Medium pulse D. leiophyllum 14.1 0.054 35.3 297 Large pulse D. leiophyllum 15.7 0.0677 26.2 258 Control B. curtipendula 7.72 0.0428 30.4 211 Medium pulse B. curtipendula 12.5 0.0495 57.9 311 Large pulse B. curtipendula 13.6 0.0421 96.2 420 After watering Control D. leiophyllum 12.3 0.0562 10.9 229 Medium pulse D. leiophyllum 12.1 0.0505 13.7 252 Large pulse D. leiophyllum 14.7 0.0702 16.2 226 Control B. curtipendula 8.55 0.0316 26.9 298 Medium pulse B. curtipendula 12.1 0.0355 66.2 408 Large pulse B. curtipendula 13.2 0.034 93.6 482 59 Table 11. Photosynthetic parameters Asat (photosynthetic capacity), AQE (apparent quantum efficiency), LCP (light compensation point), and LSP (light saturation point) calculated from the light curves taken at the end of the winter season in February 2003. Watering Species Asat AQE LCP LSP Treatment Before watering Control D. leiophyllum 6.29 0.037 14.0 184 Medium pulse D. leiophyllum 4.5 0.0232 1.91 196 Large pulse D. leiophyllum 9.48 0.141 53.9 726 Control B. curtipendula 0.209 0.000321 408 1059 Medium pulse B. curtipendula 0.103 0.000392 182 445 Large pulse B. curtipendula 0.285 0.000125 581 2856 After watering Control D. leiophyllum 8.37 0.0454 28.4 213 Medium pulse D. leiophyllum 5.53 0.065 23.8 109 Large pulse D. leiophyllum 13.7 0.0289 47.7 523 Control B. curtipendula 1.41 0.0103 132 269 Medium pulse B. curtipendula 0.438 0.000591 389 1129 Large pulse B. curtipendula 1.1 0.00126 332 1206 60 CHAPTER IV DISCUSSION Seasonal Variation in 4^., g^^ruj_A^^^ When D. leiophyllum and B. curtipendula were given a 25% increase in cumulative seasonal precipitation, there was no evidence that this additional water atfected gs or Anet for either species. However, B. curtipendula displayed high gs and Ane, during the summer, but little to no activity during the winter, while D. leiophyllum maintained gsand Anet activity throughout the year. The differential seasonal response may be in part due to the difference in phenology of these plants (Beatly 1974) while D. leiophyllum possesses an evergreen habit and may be physiologically active year-round (Kemp and Gardetto 1982), B. curtipendula is a summer-active perennial that is dormant during the winter (Kemp 1983). The difference in photosynthetic pathway between the two species may also be responsible for the difference in responses. D. leiophyllum possesses the C3 pathway which, under suitable environmental conditions, can maintain photosynthetic activity year-round, but B. curtipendula possesses the C4 pathway that limits or inhibits photosynthesis at lower temperatures due to the pathway's sensitivity to cold (Pearcy and Ehleringer 1984). Previous studies have found that the most immediate response to precipitation pulses is an increase in photosynthesis and stomatal conductance (Huxman et. al 2003). In this experiment the pulse manipulations did significantly affect gs and Anet for D. leiophyllum and B. curtipendula; the medium pulse treatment generated the highest gs. Although gs increased, there was no corresponding increase of Anet • The medium pulse treatment increased stomatal opening, allowing for greater diffusion and availability of CO2 in the leaf, yet carbon assimilation rates did not increase (Pearcy and Ehleringer 1984). Other studies have found photosynthesis to be less responsive than stomatal conductance to an increase in soil water availability after drought (Yan et al. 2000). Previous studies (BassiriRad et al. 1999, Ehleringer et al. 1999, Gebauer and Ehleringer 2000, Schwinning et al. 2002) have shown that photosynthetic response may not be 61 initiated until three to five days after a rainfall pulse. Since 1 recorded Ane, rates only one day after watering, 1 may have missed any delayed plant response. Photosynthesis may also not necessarily increase after a rainfall event because plants do not use pulse water for carbon gain only (Huxman et al. 2003). Plants that experience severe drought or have limited drought tolerance use pulse water to alleviate stress and repair tissues damaged by dehydration (Huxman et al. 2003). Photosynthetic response for D. leiophyllum may not have increased, despite greater water availability, due to damage to the photosynthetic apparatus during leaf dehydration and the utilization of water for tissue repair (Hopkins 1999, Ehleringer et al. 1999, Huxman et al. 2003). In addition, high summer leaf temperatures at the site (27-36 °C) may have inhibited photosynthesis in D. leiophyllum (C3 ) while low winter leaf temperatures (10-13 °C) recorded at the site may have inhibited photosynthesis in for B. curtipendula, which possess the C4 pathway (Sage 1999). Neither the 25% increase in seasonal precipitation nor the rainfall pulse manipulations produced significant differences in soil water potential (4^s) between the different treatments. This lack of effect may be due to the equipment used to measure soil moisture levels. The Fieldscout TDR had probes that reached only 12 cm into the soil. Since rainfall evaporates much more quickly within the upper 12 cm of the soil, the soil moisture levels recorded may not have been representative of soil moisture availability for the entire root zone (Noy-Meir 1973). A measure of leaf water potential in stead of "Ps may have been a better indicator of plant water status, but it was not used in this study. Response of g.; and Angt to 4^q When D. leiophyllum received a 25% increase in cumulative summer rainfall, gs became more responsive to changes in ^s, resulting in the plant having high gs rates at high Ts, but rapidly declining at moderate Ts- This suggests that the additional rainfall heightened stomatal sensitivity to changes in water availability such that D. leiophyllum exhibited high gs when water was readily available, but the stomates rapidly closed as ^^s 62 decreased (Huxman et al. 2003). Although gs became more sensitive to ^V,, it did not result in a significant change in the response of Ane, • The 25% rainfall increase did not result in an adjustment in the response of gs for B curtipendula; previous studies found that during the summer, a rainfall event as small as 5 mm would stimulate an increase in gs within 12 to 24 hours of receiving rainfall (Sala and Laurenroth 1982). Photosynthesis showed a mixed response: Ane, for B curtipendula for plants in the SW treatment became more responsive to changes in Ts but became less responsive for the S treatment. Based on these mixed responses and lack of corresponding adjustment in the response of gs, it appears that the 25% increase in cumulative summer rainfall did not effectively alter the Anet response ofB. curtipendula. Manipulating pulse size and frequency for D. leiophyllum caused gsto become less responsive to changes in Ts as pulse size increased and frequency decreased. Stomatal conductance became less responsive to changes in Ts, resulting in the plant maintaining positive gs at low Ts but reduced gs at high Ts- The positive gs at low Ts indicated that stomates remained open during the dry inter-pulse periods when the plants did not receive any water. D. leiophyllum is able to store small amounts of water within its leaf base, and the large and medium pulse treatments may have provided water pulses that were big enough to stimulate water storage that was then used during the inter-pulse drought (Correll and Johnston 1996). For B. curtipendula the pulse manipulations caused no significant change in the response of either gs or Anet to changes in Ts; this may be explained by restraints placed on physiological responses by the phenology of 5. curtipendula. Since B. curtipendula is only active during the summer, it must produce leaves, flower and set seed before the summer season ends (Kemp 1983, Correll and Johnston 1996). Stomatal conductance and photosynthesis need to remain highly responsive in order to take advantage of any rainfall event that occurs to maximize carbon gain for growth and reproduction (Schwinning and Ehleringer 2001). Highly responsive stomatal conductance and photosynthesis is also needed to conserve water as soil moisture is depleted since B. curtipendula has little to no water storage ability (Ehleringer et al. 1999). 63 Relationship between Watering Treatment and A. isat In both species in both watering experiments during the summer and winter, trends indicated that photosynthetic capacity (Asat) was higher after receiving a watering treatment than before water was applied. This supports that watering increased Asat increased, resulting in potentially higher carbon gain when light and water resources were not limiting (Pearcy and Ehleringer 1984). For both D. leiophyllum and B. curtipendula, plants that received the 25% increase in seasonal rainfall generated a higher Asat than plants that did not receive the additional water. During times of limited water availability, leaf dehydration may have caused the chloroplasts to become increasingly resistant to the diffusion of CO2, reducing the photosynthetic capacity (Huxman et al. 2003). Supplemental water restored leaf hydration in the D. leiophyllum and B. curtipendula plants, thereby sustaining higher Asat- Under the pulse manipulations, the large pulse treatment resulted in the highest Asat for D. leiophyllum, while the medium pulse treatment generated the highest Asat for B. curtipendula. Water from the large pulse may have been stored and then used to sustain a high Asat during the inter-pulse dry period (Correll and Johnston 1996). The moderately sized pulses may have resulted in a higher Asat for B. curtipendula since surface water may not have evaporated as quickly as that received in the small pulses. This pulse pattern may have maintained adequate leaf hydration to sustain a high Asat since B. curtipendula has little to no water storage capacity and depends on a consistent water supply to maintain physiological activity (Correll and Johnston 1996, Ehleringer et al. 1999). Summary and Conclusions In this study I investigated the effects of changes in future rainfall patterns, as predicted by the Hadley Climate Model II, on the dominant plants of the sotol-grasslands for Big Bend National Park. I tested how a 25% increase in total rainfall and a shift in pulse size and magnitude in both the summer and winter seasons would affect the physiological response ofD. leiophyllum and B. curtipendula. Based on the knowledge 64 of the model plant's physiology and phenology and on past studies of the responses of plants in arid ecosystems to water availability, 1 hypothesized that D. leiophyllum would be most responsive to a 25% increase in cumulative winter rainfall and a shift to rain falling in larger, less frequent pulses in the winter. I also hypothesized that B. curtipendula would be most responsive to a 25% increase in cumulative summer rainfall, and ambient pulse conditions, where rain falls as small but frequent pulses, in the summer. The results did not support my hypothesis for D. leiophyllum. The 25% increase in cumulative rainfall had no effect on the physiological response ofD. leiophyllum in either season, and though D. leiophyllum did respond to the medium and larger, less frequent pulses, the treatments had a much greater effect in the summer than in the winter. Though D. leiophyllum is a C3 plant, photosynthetic pathway did not play a role in governing seasonal physiological response. Most likely, D. leiophyllum maintained summer activity by limiting photosynthetesis to the morning when cooler temperatures favor C3 photosynthesis (Pearcy and Ehleringer 1984). Rooting depth may have been the major factor that influenced the physiological response ofD. leiophyllum to the watering treatments. Since the larger, less frequent pulses generated a greater response, this suggests that D. leiophyllum favored the deeper soil water penetration generated by the larger pulses that wet roots deeper in the soil profile. The results did not support my hypothesis that B. curtipendula would be responsive to a 25% increase in cumulative rainfall; the additional 25% did not affect physiological response in either season. Although B. curtipendula did respond to the pulse treatments during the summer as predicted, the medium pulse treatment generated the greatest physiological response, not the small pulse treatment. As predicted, B. curtipendula displayed only significant photosynthetic activity during the summer because of its phenology and photosynthetic pathway. The C4 pathway limits winter activity due to its sensitivity to cold temperatures (Sage 1999). B. curtipendula is also phenologically constrained to grow, flower and reproduce during the summer growing season only, preventing winter activity (Correll and Johnston 1996). Rooting depth was 65 also another important factor that influenced B. curtipendula's response to the pulse treatments. The medium pulse rather than the larger pulse resulted in a greater photosynthetic response since it wet the shallow soil layers where B. curtipendula's roots are located, but still fell frequently enough to maintain favorable activity. The medium pulses may also have been more favorable than the small pulses since they may not have evaporated as quickly. The results of this study indicate that the co-dominant plants of the sotol- grasslands, D. leiophyllum and B. curtipendula, can potentially respond to the future changes in rainfall as predicted by the Hadley Climate Model II. Since there was no significant response to the increase in cumulative rainfall, the change in storm size and frequency seems to be the more important aspect of the model with regard to the physiological response of desert plants. The results of my experiment support the findings of other studies that have found that total precipitation amount is not often associated with increased physiological activity and carbon gain, but the timing and variability of the rainfall pulses that supply that precipitation has a greater effect on physiological response (Ehleringer et al. 1999, Huxman et al. 2003). 66 LITERATURE CITED BassiriRad, H., D.C. Tremmel, R.A. Virginia, J.F. Reynolds, A.G. de Soyza, and M.H. Brunell. (1999) Short-term patterns in water and nitrogen acquisition by two desert shrubs following a simulated summer rain. Plant Ecology 145: 27-36 Beatly, J. (1974) Phenological events and their environmental triggers in Mojave desert ecosystems. Ecology 75: 922-932 Briones, O.. C. Montana, and E. Ezcurra. (1998) Competition intensity as a function of resource availability in a semiarid ecosystem. Oecologia 116: 365-372 Brown, D.E. (1994) Biotic Communities: Southwestern United States and Northwestern Mexico. University of Utah Press, Salt Lake City pp. 169-179 Cohen, D. (1970) The expected efficiency of water utilization in plants under different competition and selection regimes. Israel Joumal of Botany 19: 50-54 Connel, J.H. (1975) Some mechanisms producing structure in natural communities: a model and evidence from field experiments. In: Cody M.L., Diamond J.L. (eds) Ecology and evolution of communities. Belknap, Cambridge, Mass. pp. 460-490 Correll, D. and M. Johnston. (1996) Manual of the Vascular Plants of Texas. University of Texas at Dallas Press, Dallas pp. 248, 404-405 Corry, J.E.L. (1987) Relationship of water activity to fungal growth. In: Beuchat, L.R. (ed) Food and Beverage Mycology. New York, Von Nostrand Reinhold pp. 55-56 Dodd, M.B., W.K. Lauenroth, and J.M. Welker. (1998) Differential water resource use by herbaceous and woody plant life-forms in a shortgrass steppe community. Oecologia 117: 504-512 Donovan, L.A., and J.R. Ehleringer. (1994) Water stress and use of summer precipitation in a Great Basin shrub community. Functional Ecology 289-297 Easterling, D.R., G.A. Meehl, C. Parmesan, S.A. Changnon, T.R. Karl, and L.O. Means. (2000) Climate extremes: observations, modeling, and impacts. Science 289: 2068-2074 Ehleringer, J.R. (1984) Intraspecific competitive effects on water relations, growth, and reproduction in Enceliafarinosa. Oecologia 63: 153-158 67 Ehleringer J.R., S.L. Phillips, W.S.F. Schuster, and D.R. Sandquist. (1991) Differential ufihzatton of summer rains by desert plants. Oecologia 88: 430-434 Ehleringer, J.R., S. Schwinning, and R. Gebauer. (1999) Water use in arid land ecosystems in: Advances in Plant Physiological Ecology. M. Press editor. Blackwell Science, Oxford, UK. pp. 347-365 Fernandez, R.J., and J.F. Reynolds. (2000) Potential growth and drought tolerance of eight desert grasses: lack of a trade-off? Oecologia 123: 90-98 Fowler, N. (1986) The role of competition in plant communities in semiarid regions. Annual Review of Ecology and Systematics. 17: 89-110 Gebauer, R.L.E., S. Schwinning, and J.R. Ehleringer. (2002) Interspecific competition and resource pulse utilization in a cold desert community. Ecology 83: 2602-2616 Gebauer. R.L.E., and J.R. Ehleringer. (2000) Water and nitrogen uptake patterns following moisture pulses in a cold desert community. Ecology 81: 1415-1424 Goldberg, D., and A. Novoplansky. (1997) On the relative importance of competition in unproductive environments. Joumal of Ecology 85: 409-418 Herrmann, R. and R. Stottlemeyer. (1991) Long term monitoring for environmental change in U.S. National Parks: a watershed approach. Environmental Monitoring and Assessment 17: 51-65 Herrmarm, R., R. Stottlemeyer, J.C. Zak, R.L. Edmonds, and H. Van Miegroet. (2000) Biogeochamical effects of global change on U.S. National Parks. Joumal of the American Water Resources Association. 36: 337-346 Hopkins, W.G. (1999) Introduction to Plant Physiology. John Wiley and Sons, Inc New York, pp. 189-212 Huenneke, L.F., J.P. Anderson, M. Remmenga, and W.H. Schlesinger. (2002) Desertification alters patterns of aboveground net primary production in Chihuahuan ecosystems. Global Change Biology 8: 247-264 Huxman, T.R., d. Tissue, K. Snyder, J. Leffler, K. Ogle, W. Pockman, D. Sanquist, and D.G. Williams. (2003) Precipitation pulses and carbon dynamics in semi-arid ecosystems. In review. Jackson, R.B., J. Canadell, J.R. Ehleringer, H.A. Mooney, O.E. Sala, and E.D. Schulze. (1996) A global analysis of root distributions for terrestrial biomes. Oecologia 108:389-411 68 Kemp, P.R. (1983) Phenological patterns of Chihuahuan Desert plants in relations to the timing ot water availability. Journal of Ecology 71: 427-436 Kemp. PR. and P.E. Gardetto. (1982) Photosynthetic pathway types of evergreen rosette plants (Lihaceae) of the Chihuahuan Desert. Oecologia 55; 149-156 Kramer P.J. and J.S. Boyer. (1995) Water Relations of Plants and Soils. Academic Press, San Diego. Long, S^P. (1999) Environmental Responses. In: Sage, R.F. and R.K. Monson (Eds.) C4 Plant Biology. Academic Press, New York, pp. 226-229 Maragni, L.A., A.K. Knapp, and C.A. McAlliater. (2000) Pattems and determinants of potential carbon gain in the C3 evergreen Yucca glauca (Liliaceae) in a C4 grassland. American Joumal of Botany 87(2) 230-236 Mares, M.A. (1999) Encyclopedia of Deserts. University of Oklahoma Press, Norman pp. 112-115 Martin, C.E., F.S. Harris, and F.J. Norman. (1991) Ecophysiological responses of C3 forbs and C4 grasses to drought and rain on a tallgrass prairie in Northeastern Kansas. Botanical Gazette 152(3) 257-262 Monson, R.K., M.R. Sackschewsky. and G.J. Williams III. (1986) Field measurements of photosynthesis, water use efficiency, and growth in Agropyron smithii (C3) and Bouteloua gracilis (C4) in the Colorado shortgrass steppe. Oecologia 68: 400-409 Nobel, P.S. (1988) Environmental Biology of Agaves and Cacti. Cambridge University Press, New York North, G.B. and P.S. Nobel. (1992) Drought-induced changes in hydraulic conductivity and structure in roots of Freocactus acanthodes and Opuntia ficus-indica. Plant Physiology 89: 264-268 Noy-Meir, I. (1973) Desert ecosystems: environment and producers. Aimual Review of Ecology and Systematics. 4: 25-51 Owens, M.K. (2001) Approaches and techniques of rainfall manipulation. In: Changing Precipitation Regimes and Terrestrial Ecosystems: A North American Perspective. J.F. Weltzin and G.R. McPherson (eds). Pp. 104-128 Parson, R. and S.A. Ogston. (1995) Photosyn Assistant: Windows software for analysis of photosynthesis Version 1.1. Dundee Scientific, Scotiand UK, pp. 5-11 69 Pearcy, R.W. and J. Ehleringer. (1984) Comparative ecophysiology of C3and C4plants. h'lant. Cell, and Environment. 7, 1-13 Phillips, S.J., and P.W. Comus. (2000) A Natural History of the Sonoran Desert. Anzona-Sonora Desert Museum Press, Tuscon SAS Institue Inc. (2001). SAS/STAT" Software: Change and Enhancements for Release 8.2. SAS Institute Inc. Cary, NC Sage, R.F. (1999) Why C4 photosynthesis? In: C4 Plant Biology (eds. R.F. Sage and R.K. Monson), Academic Press, San Diego, pp. 3-14 Sala, O.E., R.A. Golluscio, W.K. Lauenroth, and A. Soriano. (1989) Resource partitioning between shrubs and grasses in the Patagonian steppe. Oecologia 81: 501-505 Sala, O.E., and W.K. Lauenroth. (1982) Small rainfall events: an ecological role in semiarid regions. Oecologia 53: 301-304 Schlesinger, W.H., J.F Reynolds, G.L. Cunningham, L.F. Heunneke, W.M. Jarrell, and R.A. Virginia. (1990) Biological feedbacks in global desertification. Science 148: 339-347 Schmitt, A.K., C.E. Martin, V.S. Loeschen, and A. Schmitt. (1993) Mid-summer gas exchange and water relations of seven C3 species in a desert wash in Baja California, Mexico. Joumal of Arid Environments 24: 155-164 Schwinning, S., K. Davis, L.Richardson, and J.R. Ehleringer (2002). Deuterium enriched irrigation indicates different forms of rain use in shrub/grass species of the Colorado Plateau. Oecologia 130: 345-355 Schwinning, S., and J.R. Ehleringer. (2001) Water use trade-offs and optimal adaptations to pulse driven arid ecosystems. Joumal of Ecology 89: 464-480 Shreve, F. (1942) The desert vegetation of North America. Botanical Review. 8: 195-246 Smith, S.D., R.K. Monson, and J.E. Anderson. (1997) Physiological Ecology of North American Desert Plants. Springer-Verlag, New York, pp. 45-55 Sperry, J.S. (2000) Hydraulic constraints on plant gas exchange. Agricultural and Forest Meteorology 104: 13-23 70 Teughels. H., 1 Nijs, P. Van Heke, and I. Impens. (1995) Competition in a global change environment: the importance of different plant traits for competitive success. Joumal of Biogeography. 22: 297-305 Walter, H. (1971) Natural savannahs as a transition to the arid zone. In: Ecology of Tropical and Subtropical Vegetation. Oliver <& Boyd, Edinburgh, pp 238-265 Weltzin, J.F., and D.T. Tissue. (2003) Resource pulses in arid environments- pattems of rain, patterns of life. New Phytologist 157: 167-173 Weltzin, J.F., and G.R. McPherson. (2000) Implications of precipitation redistribution for shifts in temperate savanna ecotones. Ecology 81(7) 1902-1913 Whitford, W. (2002) Ecology of Desert Systems. Academic Press, New York, pp. 44-95 Yan, S., C. Wan, R.E. Sosebee, D.B. Wester, E.B. Fish, and R.E Zartman. (2000) Response of photosynthesis and water relations to rainfall in the desert shmb cresote bush (Larrea tridentata) as influenced by municipal biosolids. Joumal of Arid Environments 46: 397-412 71 PERMISSION TO COPY In presenting this thesis in partial fixlfiUmento f the requirements for a master's degree at Texas Tech University or Texas Tech University Health Sciences Center, I agree that the Library and my major department shall make it freely available for research purposes. Permission to copy this thesis for scholarly purposes may be granted by the Director of the Library or my major professor. It is tmderstood that any copying or publication of this thesis for financial gain shall not be allowed without my fiortherwritte n permission and that any user may be hable for copyright infringement. Agree (Permission is granted.) Student Signature Date Disagree (Permission is not granted.) Student Signature Date