A thin veneer of soil mantles much of Earth’s surface yet the fundamental processes driving its production remain inconclusive. Untangling the roles of climate, lithology, and tectonics on soil production are challenging due to spatial variability of the weathered subsurface, locally stochastic surface processes, and the time and expense required to quantify rates of soil production. However, soil production rates are a valuable dataset in landscape evolution studies as they measure the rates at which chemical and mechanical weathering processes convert bedrock into mobile colluvial soil. Here I use terrestrial cosmogenic nuclides (TCN) to quantify rates of soil production across two landscapes with similar morphologies and climate to evaluate the role of tectonics in driving sediment production. The San Bernardino Mountains form the eastern part of southern California’s Transverse Ranges and can be structurally characterized by an elevated, low relief paleosurface drained by steep catchments incising in response to base-level fall. The morphologic transition between process domains provides a setting with nearly invariant lithologic and climatic conditions and a wide range of erosion rates. Soil production rates from each domain indicate that soil production is not solely determined by climate and lithology but dynamically adjusts to erosion rates. The Pinaleño Mountains in southeastern Arizona are a gneissic metamorphic core complex characterized by a high elevation, low relief surface drained by steep, flanking catchments. Following the cessation of Basin and Range tectonics ~5 Ma, the post-tectonic landscape is considered to be in a state of erosional decay. Soil production rates continue to scale with erosion rates and suggest that the mountain range is eroding 4–10x slower than the San Bernardino Mountains. Additionally, these rates ground a landscape evolution model to explain the enigmatic topography. In response to the COVID-19 pandemic, I created a digital capstone course to serve as an alternative to a geology field camp. I explore social, historical, and institutional contexts associated with a traditional field camp and contextualize my experience as an instructor spanning five iterations of a field course. Finally, I outline a digital course curriculum design and evaluate it through a sociotransformative constructivism (sTc) lens.

Carbonate minerals are susceptible to dissolution, so any signal from enhanced chemical erosion controlled by climate might be magnified in landscapes underlain by carbonate rocks. The reduction of sediment flux magnitude (an important factor in shaping landscapes) by chemical erosion could be one mechanism by which climate modifies landscape. This study aims to test landscape evolution models on carbonate terrain, and uses recently-developed cosmogenic 36-Chlorine techniques to measure long-term erosion rates combined with a compilation of previously published erosion rates from carbonate rocks. The stream power model of fluvial erosion describes erosion rate in terms of channel steepness and the erosional efficiency coefficient, which incorporates factors such as runoff variability, sediment flux, and bedrock erodibility. The compilation of 32 catchment-averaged erosion rates, including 4 new basin erosion rates, channel steepness, and mean annual precipitation (MAP), show slight sensitivity of erosion rates to MAP when channel steepness is accounted for. Carbonate river basins are not strongly sensitive to MAP in a singular, monotonic manner. The linear sediment transport law predicts that erosion is linearly related to hilltop curvature by a hillslope sediment transport coefficient, D. Previous work finds that vegetation and climate influence D. Since carbonate rocks may be more susceptible to climate, I test whether the topography and erosion rates indicate enhanced chemical erosion or D along a MAP gradient. I recalculate erosion rates, extract hilltop curvature, and calculate D for 50 carbonate hillslopes. Carbonate D values are uncorrelated with MAP, but peak in the 600-900 mm/yr range, a more complex pattern than simply enhanced chemical erosion or D with MAP. Arid regions can accumulate significant amounts of dust over millennia, which could overprint sediment loss due to chemical erosion. Using a mass balance framework, I quantify the chemical erosion and dust accumulation for two carbonate hilltop sites in central Arizona (site MT1; MAP = 450 mm/yr) and SE Spain (NQ; MAP 550 mm/yr) with cosmogenic 36-Chlorine and meteoric 10-Beryllium. More dust accumulated at MT1, the arid site, augmenting bedrock mass flux by 25%. Chemical erosion was greater at NQ, but accounted for >50% of mass loss at both sites. The effect of climate on carbonate rocks appears to be nuanced; future work should expand data coverage and investigate specific mechanisms of chemical erosion across climate gradients.

Both volcanic and tectonic landforms are surface expressions of the inner workings of a planet. On Earth, volcanism and crustal deformation are primarily surface expressions of plate tectonics. In contrast, the lunar crust has been deformed by solely endogenic processes following large impact events.The Procellarum KREEP (potassium (K), rare earth elements (REE), and phosphorus (P)) Terrane (PKT) is a thermally and chemically distinct geologic province on the Moon. Despite the wealth of remote sensing data, the origin and evolution of the PKT is poorly understood. This study focuses on floor-fractured craters and silicic magma genesis within the PKT. First, I present a detailed study of floor-fractured craters, including morphometric measurements using topographic datasets from the Lunar Reconnaissance Orbiter Camera (LROC), variations in temporal heat flow, lithospheric rheology and the locations of floor-fractured craters relative to impact basins. The overarching conclusion is viscous relaxation and magmatic intrusion are not necessarily mutually exclusive, as has been argued in earlier studies. This work also provides new evidence for the existence of the putative Procellarum basin. Next, with rhyolite-MELTS modeling, I demonstrate that fractional crystallization of KREEP basalt magmas is a plausible mechanism for generating silicic melts. The results suggest that following crystallization, the composition of the remaining ~30 wt.% liquids are consistent with returned lunar silicic fragments. Finally, using crater counting methods I tested the stratigraphic relationship between the floor-fractured crater, Hansteen, and the silicic volcanic landform, Mons Hansteen. Absolute model ages (AMAs) suggest that the basalts on the floor of Hansteen crater formed contemporaneously with Mons Hansteen, implying that bimodal volcanism might have played a role in silicic magma genesis on the Moon.

In the southwestern United States, water is a precious resource that influences landscapes and their respective ecosystems. Ephemeral lakes, known as playas, are drainage points for closed or endorheic basins and serve as important locations for plant productivity, biogeochemical processes, and groundwater recharge. In this study, I explore the hydrologic dynamics of eighteen (18) instrumented playas in the Jornada Basin of the Chihuahuan Desert with respect to the drivers of playa inundation and how their behaviors vary in space and time. To this end, I combine water level observations in playas with gauge-corrected radar precipitation estimates to determine hydrologic dynamics over the more than 6-year period of June 2016 to October 2022. Results indicate that all playa inundation events are associated with precipitation and that 76% of events occur during the warm season from April to September that is characterized by the North American monsoon. Mean annual runoff ratios in the playa catchments range from 0.01% to 9.28%. I observe precipitation depth and 60-minute intensity thresholds for playa inundation ranging from 16.1 to 71.3 mm and 8.8 to 40.5 mm/hr, respectively. Although playa inundation is typically caused by high rainfall amounts and intensities, other factors such as antecedent wetness conditions and the spatial variability of rainfall within the playa catchment also play a role. The magnitudes, durations, and occurrence of inundation events vary among playas, but their responses to precipitation generally agree with groupings determined based on their geological origin. Logistic and linear regressions across all playas reveal the relative importance of catchment variables, such as area, sand fraction, slope, and the percentage of bare ground. It is shown that larger catchment areas are strongly associated with a lower likelihood of inundation and higher precipitation thresholds for inundation. An analysis of precipitation data from 1916 to 2015 leads to the estimation of historical playa inundation and suggests that an increase has occurred in the frequency of large rainfall events that may be associated with increasing frequency of playa inundation. This study highlights the complex nature of playa inundation in the Jornada Basin, which can change over time in an evolving climate and landscape.

The presence of ices (H2O and CO2) and liquid water is key to the evolution ofmartian geology, with implications for the potential for past or extant life, and the future of robotic and human exploration on Mars. In this dissertation, I present the first direct evidence that the smooth deposits covering mid-latitude, martian pole-facing slopes are composed of shallow dusty H2O ice covered by desiccated material. To analyze this H2O ice, I developed the first validated radiative transfer model for dusty martian snow and glacier ice. I found that these ice exposures have < 1% dust in them, and discovered the lowest latitude detection of H2O ice on Mars, at 32.9°S. After observing the ice disappear, and new gully channels form, I proposed a model for gully formation. In this model, dusty ice gets exposed by slumping, leading to melting in the subsurface and channels eroding within the ice and the wall rock beneath. Access to liquid water within this ice could provide potential abodes for any extant life. Next, I developed novel methodology to search for CO2 frosts within the entire Thermal Emission Imaging System (THEMIS) infrared dataset and found that about half of all gullies overlap with CO2 frost detections. I also used the Thermal Emission Spectrometer (TES) water vapor retrievals to assess the formation and distribution of H2O frosts on Mars. Additionally, I used radar data from the Mars Advanced Radar for Subsurface and Ionospheric Sounding (MARSIS) instrument to investigate Mars’ ice-rich South Polar Layered Deposits (SPLD). I discovered radar signals similar to those proposed to be caused by a subglacial lake throughout the martian SPLD. Finally, I mapped martian polygonal ridge networks thought to represent fossilized remnants of ancient groundwater near the Perseverance rover landing site with the help of citizen scientists across a fifth of Mars’ total surface area and analyzed their thermophysical properties. All these studies highlight the key role that ices and liquid water have played in shaping Mars’ landscape through time, and provide an intriguing path forward in martian exploration and the search for alien life.

Accelerated climate and land use land cover (LULC) changes are anticipated to significantly impact water resources in the Colorado River Basin (CRB), a major freshwater source in the southwestern U.S. The need for actionable information from hydrologic research is growing rapidly, given considerable uncertainties. For instance, it is unclear if the predicted high degree of interannual precipitation variability across the basin could overwhelm the impacts of future warming and how this might vary in space. Climate change will also intensify forest disturbances (wildfire, mortality, thinning), which can significantly impact water resources. These impacts are not constrained, given findings of mixed post-disturbance hydrologic responses. Process-based models like the Variable Infiltration Capacity (VIC) platform can quantitatively predict hydrologic behaviors of these complex systems. However, barriers limit their effectiveness to inform decision making: (1) simulations generate enormous data volumes, (2) outputs are inaccessible to managers, and (3) modeling is not transparent. I designed a stakeholder engagement and VIC modeling process to overcome these challenges, and developed a web-based tool, VIC-Explorer, to “open the black box” of my efforts. Meteorological data was from downscaled historical (1950-2005) and future projections (2006-2099) of eight climate models that best represent climatology under low- and high- emissions. I used two modeling methods: (1) a “top-down” approach to assess an “envelope of hydrologic possibility” under the 16 climate futures; and (2) a “bottom-up” evaluation of hydrology in two climates from the ensemble representing “Hot/Dry” and “Warm/Wet” futures. For the latter assessment, I modified land cover using projections of a LULC model and applied more drastic forest disturbances. I consulted water managers to expand the legitimacy of the research. Results showed Far-Future (2066-2095) basin-wide mean annual streamflow decline (relative to 1976-2005; ensemble median trends of -5% to -25%), attributed to warming that diminished spring snowfall and melt and year-round increased soil evaporation from the Upper Basin, and overall precipitation declines in the Lower Basin. Forest disturbances partially offset warming effects (basin-wide mean annual streamflow up to 12% larger than without disturbance). Results are available via VIC-Explorer, which includes documentation and guided analyses to ensure findings are interpreted appropriately for decision-making.

Rivers in steep mountainous landscapes control how, where, and when signals of base-level fall are transmitted to the surrounding topography. In doing so rivers play an important role in determining landscape evolution in response to external controls of tectonics and climate. However, tectonics and climate often covary and understanding how they influence landscape evolution remains a significant challenge. The Hawaiian Islands, where tectonics are minimized but climate signals are amplified, provide an opportunity to better understand how signals of climate are recorded by landscapes. Focusing on the Hawaiian Islands, I examine (1) how variability in rock mass properties and thresholds in sediment mobility determine where waterfalls form or stall along the Nāpali coast of Kauaʻi, (2) I then extend these findings to other volcanoes to test if observed physical limits in flood size, climate, and volcano gradient can determine where waterfalls form, and (3) I explore how thresholds in river incision below waterfalls limit information about the influence of climate on river incision rates. Findings from this analysis show that waterfalls form or stall where the maximum unit stream power is at or below a critical unit stream power for bedrock river incision. Climate appears to have little effect in determining where these conditions are met but where waterfalls stall or form does record information about discharge-area scaling for global maximum observed floods. Below waterfalls the maximum incision depth for rivers on the island of Kauaʻi (which formed ~ 4-5 million years ago) is approximately proportional to the inverse square root of mean annual rainfall. Though maximum river incision depths for some of the younger volcanoes do not exhibit the same dependency on mean annual rainfall rates they are comparable to the maximum incision depths observed on Kauaʻi even though they are a quarter to one-tenth the age of Kauaʻi. Importantly, these patterns of incision can be explained by thresholds in sediment mobility as recorded by river longitudinal profiles and indicate that the Hawaiian Islands are dominated by threshold conditions where signals of climate are recorded in the topography through controls on incision depth but not incision rates.

The American Southwest is one of the most rapidly growing regions of the United States, as are similar arid regions globally. Across these landscapes where surface water is intermittent and variable, groundwater aquifers recharged by surface waters become a keystone resource for communities and are consumed at rates disproportional to recharge. In this study, I focus on the controls of runoff generation and connectivity at both hillslope and watershed scales along a piedmont slope. I also investigate the effects of plant phenology on hydrologic connectivity and runoff response at the hillslope scale during the summer monsoon season. To carry out this work, I combine existing hydrologic instrumentation, a new set of runoff plots with high-resolution monitoring, near-field remote sensing techniques, and historical datasets. Key analyses show that a rainfall intensity (I30) of 10 mm/hr yields runoff production at three scales (8, 12700, and 46700 m2). Rainfall, runoff, and soil moisture observations indicate a Hortonian (infiltration-excess) dominated system with little control imposed by antecedent wetness. Hydrologic connectivity analyses revealed that <15% of total rainfall events generate runoff at the hillslope scale. Of the hillslope events, only 20% of the runoff production leads to discharge at the outlet. Vegetation was observed to effect individual plot runoff response to rainfall. The results of this study show that 1) rainfall intensity is a large control on runoff production at all three scales (8, 12700, and 46700 m2), 2) proportions between bare and vegetated space effect runoff production at the hillslope scale, and 3) runoff connectivity decreases, and channel losses increase as you move downstream on an individual storm basis and across a 30-year historical record. These findings indicate that connectivity from the hillslope to outlet scale can be an evolving process over thehistorical record, reliant on both rainfall intensity, plant and bare soil mosaics, and available channel storage.

In arid and semiarid areas of the southwestern United States and northwestern México, water availability is the main control on the interactions between the land surface and the atmosphere. Seasonal and interannual variations in water availability regulate the response of water and carbon dioxide fluxes in natural and urban landscapes. However, despite sharing a similar dependance to water availability, landscape characteristics, such as land cover heterogeneity, landscape position, access to groundwater, microclimatic conditions, and vegetation functional traits, among others, can play a fundamental role in modulating the interactions between landscapes and the atmosphere. In this dissertation, I study how different landscape characteristics influence the response of water and carbon dioxide fluxes in arid and semiarid urban and natural settings. The study uses the eddy covariance technique, which calculates the vertical turbulent fluxes within the boundary layer, to quantify water, energy, and carbon dioxide fluxes within local patches. Specifically, the study focused on three main scopes: (1) how vegetation, anthropogenic activity, and water availability influence carbon fluxes in four urban landscapes in Phoenix, Arizona, (2) how access to groundwater and soil-microclimate conditions modulate the flux response of three natural ecosystems in northwestern México during the North American monsoon, and (3) how the seasonal hydrologic partitioning in a watershed with complex terrain regulates the carbon dioxide fluxes of a Chihuahuan Desert shrubland. Results showed a differential response of landscapes according to their land cover composition, access to groundwater or functional traits. In Chapter 2, in urban landscapes with irrigation, vegetation activity can counteract carbon dioxide emissions during the day, but anthropogenic sources from the built environment dominate the carbon dioxide fluxes overall. In Chapter 3, across an elevation-groundwater access gradient, low elevation ecosystems showed intensive water use strategies linked to a dependance to shallow or intermittent access to soil moisture, while a high elevation ecosystem showed extensive water use strategies which depend on a reliable access to groundwater. Finally, in Chapter 4, the mixed shrubland in complex terrain showed an evenly distributed bimodal vegetation productivity which is supported by an abundant water availability during wet seasons and by carry-over moisture in deeper layers of the soil during the dry season. The results from this dissertation highlight how different forms of water availability are responsible for the activity of vegetation which modulates land surface fluxes in arid and semiarid settings. Furthermore, the outcomes of this dissertation help to understand how landscape properties regulate the flux response to water availability in urban and natural areas.

Lava flow emplacement in the laboratory and on the surface of Mars was investigated. In the laboratory, the effects of unsteady effusion rates at the vent on four modes of emplacement common to lava flow propagation: resurfacing, marginal breakouts, inflation, and lava tubes was addressed. A total of 222 experiments were conducted using a programmable pump to inject dyed PEG wax into a chilled bath (~ 0° C) in tanks with a roughened base at slopes of 0, 7, 16, and 29°. The experiments were divided into four conditions, which featured increasing or decreasing eruption rates for either 10 or 50 s. The primary controls on modes of emplacement were crust formation, variability in the eruption rate, and duration of the pulsatory flow rate. Resurfacing – although a relatively minor process – is inhibited by an extensive, coherent crust. Inflation requires a competent, flexible crust. Tube formation requires a crust and intermediate to low effusion rates. On Mars, laboratory analogue experiments combined with models that use flow dimensions to estimate emplacement conditions and using high resolution image data and digital terrain models (e.g. THEMIS IR, CTX, HRSC), the eruption rates, viscosities, and yield strengths of 40 lava flows in the Tharsis Volcanic Province have been quantified. These lava flows have lengths, mean widths, and mean thicknesses of 15 – 314 km, 0.5 – 29 km, and 11 – 91 m, respectively. Flow volumes range from ~1 – 430 km3. Based on laboratory experiments, the 40 observed lava flows were erupted at 0.2 – 6.5x103 m3/s, while the Graetz number and Jeffrey’s equation when applied to 34 of 40 lava flows indicates eruption rates and viscosities of 300 – ~3.5 x 104 m3/s and ~105 – 108 Pa s, respectively. Another model which accounts for mass loss to levee formation was applied to a subset of flows, n = 13, and suggests eruption rates and viscosities of ~30 – ~1.2 x 103 m3/s and 4.5 x 106 – ~3 x 107 Pa s, respectively. Emplacement times range from days to centuries indicating the necessity for long-term subsurface conduits capable of delivering enormous volumes of lava to the surface.