Drylands compose approximately 40% of the earth’s land surface; a hallmark of these systems is low average soil moisture and infrequent precipitation (or irrigation) which limit biogeochemical cycling rates. However, following a re-wetting event, soil metabolism can recover within minutes, leading to discrete pulses of high rates of carbon dioxide and nitrogen oxide trace gas flux that can exceed those of more mesic systems. Although evidence suggests that moisture, temperature, and substrate availability are predominant abiotic predictors of soil metabolism, these mechanisms have not been well-described interactively and in the context of re-wetting pulses. In this dissertation, I utilized a unique network of soil chambers and trace gas analyzers to quantify patterns of carbon dioxide (CO2), nitrous oxide (N2O), and nitric oxide (NO) pulses driven by temperature-moisture-substrate interactions. In an urban shrubland context (Chapter 1), I investigated pulse consequences of exotic grass litter accumulation during shrubland-grassland ecosystem type conversion. I showed that rewetting pulses of CO2, N2O, and NO differ across seasonal wetting history, and that invasive grass litter may provide a labile C source that stimulates CO2 and N2O, but not NO, emissions from soils in this system. In a high-temperature agricultural context (Chapter 2), I explored the potential for subsurface drip irrigation to reduce soil trace gas emissions compared to traditional furrow irrigation management. I found substantial evidence that drip irrigation reduced water use, irrigation-induced pulses, and per-yield emissions of CO2, N2O, and NO from alfalfa and sudangrass fields. Particularly, the benefits of drip irrigation were strongest for fertilized sudangrass and in hot summer months, suggesting that irrigation is a strong control over emissions in optimal temperature and substrate conditions. In a desert context (Chapter 3), I tested interactive mechanisms limiting CO2 pulses by explicitly manipulating temperature and coupled carbon-nitrogen availability. I found evidence that C and N stoichiometry, not strictly limitation, determined CO2 pulse responses to soil re-wetting, and pulse responses to temperature were complicated by interactions with soil moisture immediately following rewetting. My dissertation research identifies a theme that temperature-moisture-substrate interactions drive pulse responses to dryland soil rewetting.

Urban vegetation represents a novel ecosystem where classical theories of vegetation ecology interact with systems of management and control not found in wildland areas. These interactions provide unique circumstances to test classical ecological constructs of how vegetation responds to climate while under the influence of urban actors. Cities can be hotter than their rural counterparts, include a diverse array of vegetation, and are directly and inadvertently treated with additional water and nitrogen resources. All these factors make urban regions novel common gardens to examine vegetation ecology. Furthermore, understanding the nexus of urbanization, vegetation, and climate will aid in quantifying ecosystem services as well as provide insight into how a diverse array of vegetation responds to multiple stressors. The research contained within this dissertation aims to explore how the dynamics of cities influence vegetation responses to extreme climates. To capture the many possible interactions, I explore urban vegetation ecology at multiple levels of organization, including the organismal, community, and ecosystem scale. Moreover, these studies examine both within the city and across city dynamics, comparing cities from different regional climates. I use a combination of ecophysiological traits, community diversity sampling, and remote sensor networks to understand the interactions of cities, climate, and plant-based ecosystem services. Overall, I find that the abundance of water resources in arid cities causes urban trees to decouple their carbon and water-use strategies and that decoupling is increased in desert climates. When comparing plant communities in parks in mesic and arid cities, taxonomic diversity was strongly driven by climate, but aspects of functional diversity were more determined by management practice. Across cities, the ecosystem service of vegetation-derived nocturnal cooling was tightly correlated to atmospheric aridity, highlighting the relationship between transpiration and ecosystem services. Taken together this dissertation connects how increases in water availability can result in shifts in plant function, community diversity, and resulting plant-derived ecosystem services

Urban ecosystems are defined by unique relationships between biological diversity and distribution, socio-demographics, and climate. These relationships occur in the context of a changing world: increasing temperatures, a shifting composition of urban residents, and rapid urbanization. It is likely that the relationships which structure the urban environment, such as that between temperature and income, or race-based exposure to heat and greenness, are not temporally stable. As urban temperatures and greenness have well known associations with morbidity, mortality, and mental health, non-stationarity of urban ecological relationships may pose significant challenges to the well-being of urban residents. I tested assumptions about non-stationarity by assessing the change through time of urban ecological relationships at three distinct spatial scales: the urban (single city) scale (chapter 1), at a continental scale (chapter 2), and at a global scale (chapter 3). At each spatial scale I used a multidecadal time series of satellite imagery, coupled with landcover and socio-demographic information, to assess how the dynamics of urban greenness, temperature, and vegetative cooling evolved through time and what factors mediated any change. In chapter 1, where I used Los Angeles, CA as a case study, I found large intra-urban variability in how greenness, temperature, and vegetative cooling changed over time. The consequence of these changes was a weakening of the effectiveness of income as a mediator of urban greenness and temperature (the luxury effect). I corroborated the multidecadal decline in the luxury effect in chapter 2, a continental-scale assessment where I looked at 52 cities from the conterminous United States. In this chapter I additionally compared the biophysical dynamics in every city with a nearby non-urban reference site. Urbanization weakened the relationship between the weather and landcover and the biophysical environment, where the most arid cities were entirely decoupled from precipitation. In chapter 3, I looked at a subset of 266 global cities from 82 countries, finding that the evolving relationship between greenness and temperature led to declines in the cooling effectiveness of urban vegetation. Overall, this dissertation highlights the non-stationarity of urban dynamics, as well as the importance of climatic context in understanding these urban relationships.