UC Santa Barbara

Riparian woodlands are hotspots of productivity and biodiversity on dryland landscapes, yet riparian tree species are also extremely vulnerable to catastrophic hydraulic damage caused by hydroclimatic change. Root zone water subsidies from shallow groundwater facilitate the high levels of productivity seen in riparian woodlands. Shallow groundwater provides a persistent source of root zone soil moisture that is somewhat decoupled from the local precipitation regime. As a result, riparian tree species are able to avoid water stress and maximize productivity throughout the prolonged summer dry seasons that are common in the southwestern United States. Groundwater declines caused by extreme drought conditions threaten the health and function of dryland riparian woodlands. If groundwater elevations drop below the root zones of riparian tree species, it is likely that riparian woodlands will experience widespread stress and mortality. However, the sensitivity of riparian tree species to changes in root zone water availability remains poorly constrained. This dissertation combines remote sensing, flux tower measurements, and ecological modeling to examine vegetation cover, evapotranspiration, and photosynthesis in riparian woodlands under changing environmental conditions. It also identifies critical physiological thresholds that are needed to maintain ecosystem structure and function under anthropogenic climate change.

In the first chapter, I combine remote sensing data with measurements from groundwater monitoring wells to identify critical groundwater thresholds that are needed to maintain the health and function of riparian woodlands. The analysis examines the Santa Clara River in southern California, which experienced widespread groundwater declines during an extreme drought from 2012 to 2019. Spectral mixture analysis was used to estimate the fractional cover of green vegetation and non-photosynthetic vegetation (i.e., dead and woody plant material) in six riparian woodlands. The groundwater depth was characterized for each woodland using data from groundwater monitoring wells. The analysis revealed that riparian woodlands experience substantial decreases in green vegetation cover and substantial increases in dead/woody vegetation cover when the depth to groundwater exceeds ca. 5 m. The analysis also revealed a coherent spatial and temporal trend of riparian woodland mortality that proceeded downstream over six years and mirrored trends in groundwater elevation over space and time. The findings reveal that riparian woodlands depend on shallow groundwater access to maintain health and function, and that they experience substantial stress and mortality when they lose access to root zone water subsidies from shallow groundwater aquifers.

In the second chapter, I develop a novel theoretical model to predict leaf temperature as a function of evapotranspiration. The model reveals that the difference between leaf temperature and air temperature varies as a linear function of the evaporative fraction. The model also reveals that leaf temperature converges to air temperature when the evaporative fraction equals one. The model predictions were validated using flux tower measurements from a riparian woodland and an upland savanna in southeastern Arizona. The flux tower measurements reveal that evaporative cooling reduced leaf temperature by ca. 1-5 °C in the middle of the growing season. Evaporative cooling also resulted in a ca. 15% reduction in leaf respiration. The impact of evaporative cooling on leaf carbon cycling represents a novel connection between plant water and carbon cycles via leaf energy balance that has received little attention in literature.

In the third chapter, I develop a novel modeling framework to predict the vertical profiles of radiation and wind speed within forest canopies based on known physical principles. The predictions are then used to estimate the vertical profiles of leaf temperature and net photosynthesis for five flux tower sites spanning a large latitudinal gradient in North and South America. The model predictions reveal that leaf temperature decreases exponentially downward through forest canopies, and that the difference between top-of-canopy leaf temperature and bottom-of-canopy leaf temperature can exceed 7 °C. Unexpectedly, in many forest biomes, the highest levels of net photosynthesis often occur in the middle of the canopy. The analysis helps uncover the mechanistic basis for this behavior. The analysis also identifies the combinations of environmental conditions that result in critical leaf temperatures that cause irreversible damage to leaf photosynthetic infrastructure. Two different data sets provide evidence of trait coordination between leaf size and stomatal conductance to avoid critical leaf temperatures across global forest biomes.