Exploring the Wide Net of Human Energy Systems: From Carbon Dioxide Emissions in China to Hydraulic Fracturing Chemicals Usage in the United States

Abstract

Human energy systems have profound and far-reaching impacts, from global to regional levels. This thesis examines two seemingly distinct issues that are intrinsically united by the growing human appetite for energy and commodities: Part One assesses carbon dioxide (CO2) emissions in China with implications for global climate; Part Two presents an analysis of hydraulic fracturing chemicals usage in the United States with implications for regional water quality. Part One presents a pilot study for Northern China examining the applicability of modeling tools used extensively in North American greenhouse gas inventory optimization studies to evaluate China’s CO2 emissions. China, the leading anthropogenic CO2 emitter, accounts for nearly 30% of the world’s CO2 emissions from energy consumption. However, the estimated magnitude and distribution of China’s CO2 emissions are highly uncertain. While optimization of CO2 emissions inventories in the United States benefits from dense observational networks and readily accessible atmospheric data, access to and availability of China’s data is restricted. In this study we use continuous hourly measurements of CO2 from a site in Northern China from 2005 to 2009 to compare three different estimates of anthropogenic CO2 emissions, one China-specific and two global. Intra-annual evaluation of CO2 emissions is further complicated by the strong vegetation signal during the growing season; we construct a simple biosphere model to quantify the vegetation contribution to CO2. The biosphere model demonstrates low bias as a prior at hourly timescales. At annual timescales, the global inventories generally report emissions that are lower than the regional inventory and observations by 30% to 70%. In Part Two we move to the United States where we analyze the composition and disclosure patterns in hydraulic fracturing fluid chemical records of varying quality and completeness collected between 2011 and 2015. We find rates of withheld ingredients are increasing, but find that certain methods of reporting encourage disclosure rates by up to four-fold. The studies in both parts are united by a common theme of uncertainty in sparse data; in both cases our results serve as important baselines for future analyses that would greatly benefit from access to larger, more complete data sets.