UC Santa Cruz

The Earth’s critical zone (CZ) extends from the top of vegetation canopies to the base of weathered bedrock. Within this region, rock meets life, as water is stored and released to streams, groundwater and vegetation. However, understanding the complex interactions between vegetation, water cycling, and bedrock weathering remains a grand challenge in forecasting and managing freshwater resources and ecosystem health. The research presented here is motivated by the need to more accurately quantify how small changes in climate will alter the relationship between aboveground vegetation and belowground water cycling and subsurface weathering. It addresses this need by utilizing a natural experiment provided by aspect, or the direction a hillslope faces, which produces different amounts of direct solar radiation between adjacent hillslopes. Specifically,pole-facing slopes (north-facing in the northern hemisphere) receive less direct solar radiation compared to equator-facing slopes (south-facing in the northern hemisphere) which often lead to distinct vegetation structures (e.g. density, composition). Current models of how aspect mediates the relationship between vegetation, water cycling and the subsurface have been limited to snow-dominated landscapes with woody vegetation on either hillslope (e.g. trees). Furthermore, most studies have focused on the shallow soil, owing to the difficulty to directly characterize the physical and chemical properties of the weathered bedrock below. To overcome this limitation and explore how microclimates regulate critical zone form and ecohydrologic function, I established and continuously monitored Arbor Creek Experimental Catchment, a rain-dominated catchment that is dominated by oak trees on the pole-facing slope and only has grasses on the equator-facing slope.

In my first chapter, I investigated how the interplay between microclimates and vegetation type (e.g. grass versus tree) regulate the timing and magnitude of evapotranspiration (combined evaporation and plant water use). Within Arbor Creek Catchment, I combined oak tree sapflow, grass transpiration, and soil moisture monitoring with tree survey based evapotranspiration scaling, and remote sensing techniques to quanitfy hillslope-scale evapotranspiration. My research revealed that despite receiving less direct solar radiation, the pole-facing slope with oak trees has higher evapotranspiration than warmer equator-facing slopes. This research highlights the importance of adequately representing hillslope-scale vegetation dynamics to more accurately predict evapotranspiration, which is the largest and most difficult component of the terrestrial water cycle to constrain. Furthermore, these findings suggest that due to oak tree transpiration on pole-facing slopes, the subsurface root-zone water storage deficit may be higher (i.e. pole-facing slopes are drier) compared to the equator-facing slope with grasses which has important consequences for groundwater recharge and streamflow generation.

In my second chapter, I used deep drilling (6 - 40 m) to determine how differences in vegetation type (e.g. rooting structures and water use) influence the relationship between subsurface bedrock weathering and water storage dynamics between hillslopes with opposing aspects. Extensive geochemical, physical and hydrologic measurements within near-ridge boreholes on the pole-facing and equator-facing slope reveal a coupling between the depth and extent of vadose zone water storage and bedrock weathering. Specifically, the subsurface bedrock within the pole-facing slope has a higher degree of chemical and physical alteration compared to the equator-facing slope. This is likely driven by both ”bottom-up” and ”top-down” drivers including a higher degree of protolith fracturing, larger change in vadose zone water storage, and deeper extent to roots which may promote rock weathering through root growth in fractures and enhanced microbial processes.

Lastly, in my third chapter, I combined topographic analyses, soil pits, and seismic refraction measurements to characterize the shallow (top 6 m) subsurface structure(e.g. thickness, porosity) from the ridge tops to the stream between hillslopeswith opposing aspects. To determine the biophysical controls on shallow water storage and movement, I paired this physical characterization with soil moisture and transient, perched, groundwater measurements. My research showed that the average slope, soil, and saprolite thickness were similar between hillslopes with opposing aspects. The similarities in slope, soil thickness and saprolite thickness may also be a consequence of past weathering processes when this landscape was likely not precipitation-limited and woody vegetation was dominant across the catchment. Additionally, my results show a depth dependent relationship between aspect and subsurface water storage. Specifically, while the soil moisture suggested similarities in soil water storage capacity, between rain events the pole-facing slope soil remained wetter than the equator-facing slope soil due to less evaporative demand. However, we observed a higher occurrence of a perched, transient, groundwater response on the grass-dominated equator-facing slope, which was likely due to a lower root-zone storage deficit because of overall lower evapotranspiration.

Together, these dissertation chapters highlight the importance of past and present-day vegetation dynamics to mediate the effects of microclimate on water cycling and subsurface weathering. This work demonstrates the critical need to better refine vegetation type, rooting architecture, and water use patterns within Earth system models to more accurately predict hillslope-scale subsurface weathering and hydrologic processes. Furthermore, this research reveals the importance of root-zone waterstorage, below the soil and into weathered bedrock, to serve as an important control on hydrologic refugia, groundwater recharge, and ecosystem health within oak savannas.