I. The land-atmosphere water flux across plant, ecosystem, global and social scales. II. GIS and spatial analysis for environmental justice and wildlife

I. Evapotranspiration is a major component of the terrestrial water balance, and is central to the processes and models of global circulation and climate change, ecosystem carbon cycling, floods, droughts and irrigation. Over the entire land surface of the globe, approximately two-thirds of rainfall returns to the atmosphere as evapotranspiration, making evapotranspiration the largest single component of the hydrological cycle. Two main problems exist, however, of which the solutions are the objectives of this thesis: (1) evapotranspiration is very difficult to measure and predict, especially at synoptic spatial scales; (2) the processes that control evapotranspiration are weighted inconsistently across spatial scales. My methodology includes a combination of sap flow for transpiration in individual trees and shrubs, modeling as applied to eddy covariance measurements for ecosystems, remote sensing for the land-atmosphere flux over continents and the globe, and intellectual property rights and common property resources theories for data sharing. I found that nocturnal transpiration occurs in many species, which violates many long-held leaf-scale assumptions about daytime-dependency of photosynthesis-transpiration coupling; simple models tend to better predict ecosystem evapotranspiration measurements than do more theoretically accurate, but complex models, which are subject to greater sources of uncertainty; evapotranspiration can be estimated accurately using only remote sensing and a simple model of evapotranspiration with sophisticated ecophysiological constraints; and, that payment is always required in data sharing, but that the currency and value of payment in academia are associated with publication, authorship and acknowledgment. This thesis contributes to advancement in the science, understanding, and prediction of a major component in the water budget.;II. Spatial point patterns tend not to be random, but functions of larger processes that control behaviors and locations of people, institutions and wildlife. It is the goal here to identify and quantify the scales and dimensions at which clusters occur, and link these results to policy and management decisions that control these scales. Through internships at the U.S. Environmental Protection Agency and NASA Ames Research Center, I applied this methodology to projects on environmental justice and wildlife management. The environmental justice aspect focuses on West Oakland, California, where a low-income, high-minority community was situated among a dense distribution of point source air toxics polluters. The wildlife management aspect focuses on the threatened burrowing owl (Athene cunicularia), which was managed with conservation areas of varying sizes set aside on federal land. Both parts shared the same methods, a combination of geographic information systems (GIS), first-order intensity distributions, and second-order Ripley's K analysis; the West Oakland project included air modeling. At West Oakland, I quantified a disproportionate clustering of point source polluters, identified the potential exposure from the most dangerous polluter, and addressed the issue of non-point source air toxics. At NASA Ames, I identified the scale and timing of owl clusters, and the spatial pattern response to conservation areas of matching versus disaggregate scales. The policy implications of my environmental justice results were that the major West Oakland point source polluter was shut down. The management implications of my burrowing owl analysis were informed decisions on conservation area sizes.