Terrestrial ecosystems are critical to human welfare and regulating Earth’s life support systems but many gaps in our knowledge remain regarding how terrestrial plant communities respond to changes in climate or human actions. I used field experiments distributed across three dryland ecosystems in North America to evaluate the consequences of changing precipitation and physical disturbance on plant community structure and function. Evidence from experiments and observational work exploring both plant community composition and ecological processes suggest that physical disturbance and precipitation reductions can reduce the diversity and function of these dryland ecosystems. Specifically, I found that aboveground net primary productivity could be reduced in an interactive manner when precipitation reductions and physical disturbance co-occur, and that within sites, this reduction in productivity was greater when growing-season precipitation was low. Further, I found that these dryland plant communities, commonly dominated by highly drought-resistant shrubs and perennial grasses, were not capable of compensating for the absence of these dominant shrubs and perennial grasses when they were removed by disturbance, and that precipitation reductions (as predicted to occur from anthropogenic climate change) exacerbate these gaps. Collectively, the results of the field experiment suggest that current management paradigms of maintaining cover and structure of native perennial plants in dryland systems are well founded and may be especially important as climate variability increases over time. Evaluating how these best management practices take place in the real world is an important extension of fundamental ecological research. To address the research-management gap in the context of dryland ecosystems in the western US, I used a set of environmental management plans and remotely sensed data to investigate how ecosystem services in drylands are accounted for, both as a supply from the land base and as a demand from stakeholders. Focusing on a less-investigated land base in the United States–areas owned and managed by the Department of Defense–I explored how ecosystem services are produced by this unique land management arrangement even if they are not explicitly managed for under current management schemes. My findings support a growing body of evidence that Department of Defense lands represent a valuable conservation opportunity, both for biodiversity and ecosystem services, if management regimes fully integrate the ecosystem services concept.

Dioecious plants often display sexual segregation in habitat preference and trait expression due to contrasts in reproductive costs. Females may be maladapted to environments with limited available resources, or habitats where resources are diminishing due to climate change. Reduced fitness in female individuals compared to males could lead to skewed sex ratios and reduce population fitness of dioecious species, including one of the most widely distributed dioecious tree species in North America, Acer negundo. The goal of this study was to evaluate how climate warming and drought may enhance sexual segregation in productivity and physiological stress in A. negundo. To address this goal, I measured radial growth and carbon isotope ratios (δ13C) in tree-ring cellulose of 22-year male and female A. negundo trees growing in a common garden in Salt Lake City, UT. The trees were originally transplanted as one-year old cuttings from a nearby site that was 6.5 °C cooler that the common garden. I hypothesized that 1) δ13C would be lower (more negative) in late growth that is formed during the hottest months of the growing season in males than in females, and during years with no supplemental watering, indicating lower stress from heat and drought in males than in females. And 2) radial growth would be greater in males under warm, well-watered conditions and the addition of drought will exacerbate the difference between males and females. To test these hypotheses, cores were extracted from the main stem of nine male and nine female trees with an increment borer. Annual growth was measured on each core and cellulose was extracted to measure annual δ13C ratios. Males had a 0.63‰ lower mean δ13C than females in years after supplemental water had ceased (p = 0.03) and a 4.12 mm wider radial growth compared to females while irrigated (p = 0.02). Although these data did not support my hypotheses per se, results nevertheless indicate that females are more likely to be maladapted to climate warming and drought to a greater extent than males. If so, a combination of drought and heat stress may have deleterious impacts on the population fitness of Acer negundo and other similar dioecious tree species.

Flowering phenology offers a sensitive and reliable biological indicator of climate change because plants use climatic and other environmental cues to initiate flower production. Drylands are the largest terrestrial biome, but with unpredictable precipitation patterns and infertile soils, they are particularly vulnerable to climate change. There is a need to increase our comprehension of how dryland plants might respond and adapt to environmental changes. I conducted a meta-analysis on the flowering phenology of dryland plants and showed that some species responded to climate change through accelerated flowering, while others delayed their flowering dates. Dryland plants advanced their mean flowering dates by 2.12 days decade-1, 2.83 days °C-1 and 2.91 days mm-1, respectively, responding to time series, temperature, and precipitation. Flowering phenology responses varied across taxonomic and functional groups, with the grass family Poaceae (-3.91 days decade1) and bulb forming Amaryllidaceae (-0.82 days decade1) showing the highest and lowest time series responses respectively, while Brassicaceae was not responsive. Analysis from herbarium specimens collected across Namibian drylands, spanning 26 species and six families, revealed that plants in hyper-arid to arid regions have lower phenological sensitivity to temperature (-9 days °C-1) and greater phenological responsiveness to precipitation (-0.56 days mm-1) than those in arid to semi-arid regions (-17 days °C-1, -0.35 days mm-1). The flowering phenology of serotinous plants showed greater sensitivity to both temperature and precipitation than that of non-serotinous plants. I used rainout shelters to reduce rainfall in a field experiment and showed that drought treatment advanced the vegetative and reproductive phenology of Cleome gynandra, a highly nutritional and medicinal semi-wild vegetable species. The peak leaf length date, peak number of leaves date, and peak flowering date of Cleome gynandra advanced by six, 10 and seven days, respectively. Lastly, I simulated drought and flood in a greenhouse experiment and found that flooding conditions resulted in higher germination percentage of C. gynandra than drought. My study found that the vegetative, and flowering phenology of dryland plants is responsive to climate change, with differential responses across taxonomic and functional groups, and aridity zones, which could alter the structure and function of these systems.

Xylem conduits, a primary feature of most terrestrial plant taxa, deliver water to photosynthetic tissues and play a critical role in plant water relations and drought tolerance. Non-succulent woody taxa generally follow a universal rate of tip-to-base conduit widening such that hydraulic resistance remains constant throughout the plant stem. Giant cacti inhabit arid regions throughout the Americas and thrive in water-limited environments by complimenting water-storing succulent tissues with resource-efficient Crassulacean Acid Metabolism. Considering these adaptations, the objectives of this study were threefold: 1) determine whether xylem conduits in columnar cacti follow universal scaling theory as observed in woody taxa; 2) evaluate whether xylem hydraulic diameter is inversely correlated with xylem vessel density; and 3) determine whether xylem double-wall thickness-to-span ratio and other hydraulic architectural traits are convergent among phylogenetically diverse cactus species. This thesis investigates the xylem anatomy of nine cactus species native to the Sonoran Desert of Arizona and Mexico, the tropical dry forests of southern Mexico, and the Alto Plano region of Argentina. Soft xylem tissues closest to the stem apex underwent a modified polyethylene glycol treatment to stabilize for sectioning with a sledge microtome. Across all species: hydraulic diameter followed a basipetal widening rate of 0.21 (p < 0.001), closely matching the universal rate of 0.20 for woody taxa; and xylem vessel density was inversely correlated with both length from stem apex (p < 0.001) and hydraulic diameter (p < 0.001). Double-wall thickness-to-span ratio had little to no significant correlation with either length from stem apex or hydraulic diameter. There was no significant difference in hydraulic architectural trait patterns between phylogenetically diverse species with various stem morphologies, nor was there a significant correlation between conduit widening rates and volume-to-surface-area ratios. This study demonstrates that giant cacti follow similar internal anatomical constraints as non-succulent woody taxa, yet stem succulence and water storage behavior in cacti remain separate from internal hydraulic architecture, allowing cacti to thrive in arid environments. Understanding how cacti cope with severe water limitations provides new insights on evolutionary constraints of stem succulents as they functionally diverged from other life forms.

Drylands cover over 40% of the Earth’s surface, account for one third of global carbon cycling, and are hotspots for climate change, with more frequent and severe droughts coupled with deluges of novel magnitude and frequency. Because of their large terrestrial extent, elucidating dryland ecosystem responses to changes in water availability is critical for a comprehensive understanding of controls on global aboveground net primary productivity (ANPP), an important ecosystem service. The focus of this dissertation is to investigate cause-effect mechanisms between altered water availability and ecosystem processes in dryland ecosystems. Across a network of experimental rainfall manipulations within a semiarid Chihuahuan Desert grassland, I examined short- and long-term dynamics of multiple ecosystem processes—from plant phenology to nitrogen cycling—in response to directional precipitation extremes. Aboveground, I found herbaceous plant phenology to be more sensitive in greenup timing compared to deep-rooted, woody shrubs, implying that precipitation extremes will disproportionately affect grass-dominated compared to woody ecosystems. Surprisingly, after 14 years of experimentally adding water and N, I observed no effect on ANPP. Belowground, bulk soil N dynamics remained stable with differing precipitation amounts. However, mineral associated organic N (MAOM-N) significantly increased under chronic N inputs, indicating potential for dryland soil N sequestration. Conversely, the difference between low- and high-N soil N content may increase a drawdown of N from all soil N pools under low-N conditions whereas plants source N from fertilizer input under high-N conditions. Finally, I considered ecosystem-level acclimation to climate change. I found that N availability decreased with annual precipitation in space across continents, but it posed initially increasing trends in response to rainfall extremes at the Jornada that decreased after 14 years. Mechanisms for the acclimation process are thus likely associated with differential lags to changes in precipitation between plants and microorganisms. Overall, my dissertation demonstrates that examining linkages between multiple ecosystem processes, from aboveground phenological cycles to belowground N cycling dynamics, can provide a more integrative understanding of dryland response to climate change. Because dryland range is potentially expanding globally, water limited systems provide a unique and critical focus area for future research that revisit and revise current ecological paradigms.

The southwestern US will experience more frequent heat waves, prolonged droughts, and declining water supply. Riparian ecosystems are particularly at-risk under climate change predictions, but little is known about the thermal tolerance of plant species inhabiting these ecosystems. Populus fremontii, a pioneer and foundation tree species in riparian ecosystems throughout the southwest, is of concern given its importance in driving community structure and influencing ecosystem processes. This study compared leaf thermal tolerance across populations of P. fremontii to determine if local adaptation affects leaf thermal tolerance. I hypothesized that warm-adapted (low-elevation) populations would have larger leaf thermal tolerance thresholds, thermal safety margins, and thermal time constants than cool-adapted (high-elevation) populations. I expected warm-adapted populations to maintain lower maximum leaf temperatures due to local adaptation affecting leaf thermal regulation. Using a common garden at the warm edge of this species’ range, I measured leaf thermal tolerance metrics in eight populations spanning a 1,200 m elevational gradient. Data collection occurred in May, during mild air temperatures, and in August, during high air temperatures. The first two metrics were leaf thermal tolerance thresholds. The critical temperature (Tcrit) is the temperature at which the electron transport capacity of PSII is disrupted. T50 is the temperature at which the electron transport capacity decreases to 50%. The next metric was thermal safety margins (TSMs), which reflect a leaf’s vulnerability to reaching thermal tolerance thresholds. TSMs are the difference between either Tcrit or T50 and an experienced environmental variable such as leaf or air temperature. The last metric was the thermal time constant (?), which is a trait that represents how quickly leaf temperatures respond to changes in air temperatures. Tcrit, T50, and ? were not correlated with elevation regardless of season, suggesting that acclimation or phenotypic plasticity is affecting these metrics. Conversely, TSMs using maximum leaf temperature were negatively correlated with elevation in August because warm-adapted populations maintained lower maximum leaf temperatures. These findings suggest that warm-adapted populations are locally adapted to maintain cooler leaf temperatures, which is critical for their future survival since they do not maintain higher leaf thermal tolerance thresholds than cool-adapted populations.

Physical and structural tree measurements are applied in forestry, precision agriculture and conservation for various reasons. Since measuring tree properties manually is tedious, measurements from only a small subset of trees present in a forest, agricultural land or survey site are often used. Utilizing robotics to autonomously estimate physical tree dimensions would speed up the measurement or data collection process and allow for a much larger set of trees to be used in studies. In turn, this would allow studies to make more generalizable inferences about areas with trees. To this end, this thesis focuses on developing a system that generates a semantic representation of the topology of a tree in real-time. The first part describes a simulation environment and a real-world sensor suite to develop and test the tree mapping pipeline proposed in this thesis. The second part presents details of the proposed tree mapping pipeline. Stage one of the mapping pipeline utilizes a deep learning network to detect woody and cylindrical portions of a tree like trunks and branches based on popular semantic segmentation networks. Stage two of the pipeline proposes an algorithm to separate the detected portions of a tree into individual trunk and branch segments. The third stage implements an optimization algorithm to represent each segment parametrically as a cylinder. The fourth stage formulates a multi-sensor factor graph to incrementally integrate and optimize the semantic tree map while also fusing two forms of odometry. Finally, results from all the stages of the tree mapping pipeline using simulation and real-world data are presented. With these implementations, this thesis provides an end-to-end system to estimate tree topology through semantic representations for forestry and precision agriculture applications.

Climate change is making the arid southwestern U.S. (“Southwest”) warmer and drier. Decreases in water availability coupled with increases in episodic heat waves can pose extraordinary challenges for native riparian tree species to persist in their current ranges. However, the morpho-physiological mechanisms that these species deploy to cope with extreme temperature events are not well understood. Specifically, how do these species maintain leaf temperatures within a safe operational threshold in the extreme conditions found across the region? Morpho-physiological mechanisms influencing intraspecific local adaptation to thermal stress were assessed in Populus fremontii using two experimental common gardens. In a common garden located near the mid-point of this species’ thermal distribution, I studied coordinated traits that reflect selection for leaf thermal regulation through the measurement of 28 traits encompassing four different trait spectra: phenology, whole-tree architecture, and the leaf and wood economic spectrum. Also, I assessed how these syndromes resulted in more acquisitive and riskier water-use strategies that explained how warm-adapted populations exhibited lower leaves temperatures than cool-adapted populations. Then, I investigated if different water-use strategies are detectable at inter-annual temporal scales by comparing tree-ring growth, carbon, and oxygen isotopic measurements of cool- versus warm-adapted populations in a common garden located at the extreme hottest edge of P. fremontii’s thermal distribution. I found that P. fremontii’s adaptation to the extreme temperatures is explained by a highly intraspecific specialized trait coordination across multiple trait scales. Furthermore, I found that warmer-adapted populations displayed 39% smaller leaves, 38% higher midday stomatal conductance, reflecting 3.8 °C cooler mean leaf temperature than cool-adapted populations, but with the tradeoff of having 14% lower minimum leaf water potentials. In addition, warm-adapted genotypes at the hot edge of P. fremontii’s distribution had 20% higher radial growth rates, although no differences were detected in either carbon or oxygen isotope ratios indicating that differences in growth may not have reflected seasonal differences in photosynthetic gas exchange. These studies describe the potential effect that extreme climate might have on P. fremontii’s survival, its intraspecific responses to those events, and which traits will be advantageous to cope with those extreme environmental conditions.

Drylands make up more than 45% of the Earth’s land surface and are essential to agriculture and understanding global carbon and elemental cycling. This thesis presents an analysis of atmospheric relative humidity (RH) and temperature (T) as they impact soil moisture and water content at two dryland sites. In particular, this thesis assesses the likelihood and impact of non-rainfall moisture (NRM) sources on dryland soils. This work also includes a discussion of the development and testing of a novel environmental sensing network, using custom nodes called EarthPods, and recommendations for the collection of future data from dryland sites to better understand NRM events in these regions. An analysis of weather conditions at two drylands sites suggest that nighttime RH is frequently high enough for NRM events to occur. Thesis results were unable to detect changes in soil water content based on historical weather data, likely due to instrument limitations (depth and sensitivity of soil moisture probes) and the small changes in soil moisture during NRM events. However, laboratory tests of EarthPod soil moisture sensors indicated strong sensitivity to T. Characterization of these T sensitivities provide opportunities to calibrate and correct soil moisture estimates using these sensors in the future. This work provides the foundation for larger biogeochemical sampling campaigns focusing on NRM in dryland systems.