This dissertation explores the utility of high-resolution satellite carbon dioxide (CO2) and water vapor measurements for advancing climate treaty verification, for improving numerical weather prediction (NWP), and for understanding natural carbon cycling in the terrestrial biosphere. We present a series of Observing System Simulation Experiments (OSSEs) using a carbon-weather data assimilation (DA) system, where the state vector comprises weather variables (wind, temperature, humidity and pressure) and atmospheric CO2 mixing ratios. The system seeks the optimal fit between a suite of synthetic meteorological and satellite-based total column CO2 (XCO2) observations with forecasts from a global Earth system model. Given the incomplete observations and imperfect model, the simultaneous assimilation of weather and CO2 observations into our system yields the best approximation of atmospheric transport as well as its uncertainty, something not captured by other community carbon data assimilation and surface flux inversion systems which use a single realization of atmospheric transport. Our assimilation window is six hours, meaning that we have a time-evolving estimate of the atmospheric state, and its uncertainty (represented by the spread in the ensemble) at the resolution of six hours.

In Chapter 2, we employ this machinery to assess the capability of our carbon-weather DA system, along with satellite-borne XCO2 observations, to detect underreporting of CO2 emissions at the scale of a large country. In a series of OSSEs, we assimilate synthetic observations of XCO2 at the locations of (1) the Orbiting Carbon Observatory 2 (OCO-2) soundings and (2) a hypothetical observing system which observes globally at 1pm local time. Fossil fuel CO2 emissions are modified to have a -50% bias over China, but the observations are pulled from a model run where this bias is not present. We test whether the data assimilation system can detect the imposed bias by examining the near-surface innovation in CO2 mass in a method similar to the mass-balance inversion. We find that with the hypothetical observation strategy, we can recover half of the imposed bias, and that the ensemble mean of the near-surface CO2 tracks the truth during the daytime, but underestimates the truth during the unconstrained nighttime hours over the region of the imposed bias. For the OCO-2 strategy, we detect a signal at the location of the imposed bias that is obscured by problems such as observation coverage. We discuss potential additions to the observing system which could optimize the detection of biased emissions with our data assimilation machinery.

Chapter 3 presents results from OSSEs aimed at understanding the potential of OCO-2 total precipitable water (TPW) and XCO2 to improve weather forecasting capabilities. The hypothesis is that the time- and space-varying correlation between the satellite observable and wind in the Earth System Model could be used to improve the weather forecast where wind observations are sparse. We find that the TPW observations impact all meteorological state variables in the experiment, and that the XCO2 observations reduce weather forecast errors globally, and most significantly in the southern extratropics, in all meteorological fields except humidity. We conclude that both of these observation types from OCO-2 could serve as useful additions to the suite of observations assimilated by national weather forecasting centers.

In Chapter 4, we calculate global CO2 surface fluxes as a residual in the vertically-integrated CO2 tracer transport equation, using time-varying 3D-CO2 and meteorology reanalysis fields from a carbon-weather DA system that assimilates weather and XCO2 from the Atmospheric Infrared Sounder (AIRS). As AIRS XCO2 is weighted in the mid-troposphere, we find that the most significant impact on the surface flux calculation is in the tropics, especially over the Amazon and in the tropical Pacific, where intense convection mixes CO2 through the entire tropospheric column. We compare our posterior flux estimates to those made by CarbonTracker and find general sign agreement except in the Amazon region. Here we estimate a net annual sink of -0.26 PgC whereas CarbonTracker, which uses only surface observations, estimates a net annual source of about the same magnitude.

Evapotranspiration cools the land surface by consuming a large fraction of the net radiative energy at the surface. In forested regions, trees actively control the rate of transpiration by modulating stomatal conductance in response to environmental conditions, and species with different stomatal dynamics can affect the atmosphere in distinct ways. Using principal component analysis (PCA) and Markov chain Monte Carlo (MCMC) parameter estimation with direct, tree-level measurements of water use, we show that Douglas-firs (Pseudotsuga menziesii), a common evergreen needleleaf tree species in the Northern California Coast Range, decrease their transpiration sharply in the summer dry season in response to a dry root zone; and in contrast, broadleaf evergreen tree species, especially Pacific madrones (Arbutus menziesii), transpire maximally in the summer dry season because their transpiration is much less sensitive to a dry root zone and increases continually in response to increasing atmospheric evaporative demand. We scale up these tree-level observations to construct a bottom-up estimate of regional transpiration, and we use these regional estimates along with atmospheric models, one simple and one comprehensive, to quantify the potential impact of species transpiration differences on regional summertime climate. The atmospheric models suggest that these species differences in transpiration could affect the well-mixed atmospheric boundary layer temperature and humidity by 1-1.5 degrees C and 1 g/kg, respectively, and near-surface temperature and humidity by 1.5-2.5 degrees C and 2-3 g/kg, respectively. We further investigate the sensitivity of California climate to evapotranspiration by estimating the sensitivity of wind energy forecasts at a California wind farm to regional-scale perturbations in soil moisture using a regional atmospheric model. These tests show that forecasts at this particular farm are most sensitive to soil moisture in the Central Valley, and that changes in soil moisture on the order of previously-demonstrated errors in atmospheric reanalyses lead to wind energy forecast errors of 1-3 m/s, which can translate to differences in forecasted wind energy of 15-40% of a wind farm’s maximum rated power.

Spring and summer rainfall over East Asia is characterized by unique organized bands of precipitation, stretching from China to Korea and Japan in climatology, on top of the typical local convective systems. This banded precipitation has moderate to high intensity and can persist for several days to a few weeks. It undergoes a northward migration from May to August, starting from south China in Guangdong and Guangxi provinces to the Yangtze River Basin and eventually to north China and the Korean Peninsula. The persistent precipitation during this period is referred to as Meiyu in China, Baiu or Tsuyu in Japan, and Changma in Korea; these names capture the timing and the long-lasting nature of this precipitation. Meiyu and Baiu mean “plum rain”, denoting the season when the plum ripens; Changma means “long period of rain”. This rainy season is sometimes referred to as the East Asian summer monsoon, denoting the seasonally reversing flow that carries warm moist air from the Pacific and the Indian Ocean into continental East Asia.

Previously studies of banded precipitation in East Asia, or the East Asian rainbands, have focused on the total precipitation within specific regions, and have yielded little to no information on individual events or the exact position, shape, and size of the individual rainbands. In this work, I have developed a novel method to identify individual East Asian rainbands in the 6-hourly precipitation from ERA-interim reanalysis product from 1979-2018, by using criteria on precipitation threshold and longitudinal connectedness. This method employs a combination of image processing techniques, manual labelling, and deep neural networks to identify and catalog the long and narrow bands of precipitation. We find that the northward migration of the East Asian rainbands is continuous, and not step-wise, as suggested by previous studies that relied on rainfall data averaged over pentads or longer periods of time.

With our individually identified rainbands, we examine the atmospheric circulation associated with the rainbands to understand their dynamics and seasonal migration. We find that the mid-tropospheric northerly flow north of the East Asian rainbands to be the most important circulation feature that determines the latitudinal position of the rainbands. The seasonal climatology shows that rainbands are located at the convergence of northerly and southerly flows at the latitude of the greatest meridional gradient in moist static energy. On the days with rainbands identified, however, the northerly is much stronger than the seasonal mean while the southerly remains the same. We hypothesize that the southerly flow continually supplies warm moist air for the front, while the sporadic strong northerly triggers the frontogenesis. We have found no strong direct correlation between the latitude or strength of the zonal wind and jet stream to the latitudes of the identified East Asian rainbands. Our finding does not contradict previous studies that have associated the seasonal migrations of the rainbands with those of the jet stream, as they do happen simultaneously.

Using a simple one-layer barotropic quasi-geostrophic model of potential vorticity to guide our analysis of the ERA-interim data, we show that northerly winds crucial for the formation of the rainbands are part of the orographic Rossby wave excited by the Tibetan Plateau. The location and the speed of this northerly wind are determined by the mid-tropospheric westerlies upstream(west) of the Tibetan Plateau. In early spring, strong westerlies impinging on the Tibetan Plateau are likely to generate Rossby waves with long wavelengths and strong northerlies in East Asia, thus limiting the northward migration of the rainbands. In late spring and summer, the impinging westerlies weaken, the Rossby wavelengths shorten, and the northerlies over East Asia also weaken, permitting rainband migration northwards. On the interannual scale, the East Asian rainbands could covary with the northerly wind and also with the upstream zonal wind in some years. In other years, the correlation cannot be established, as other factors such as the subtropical high over North Pacific and the southerly moisture supply might play a role.

The westerly wind interacting with the Tibetan Plateau is part of the jet streams over East Asia, and we present a detailed characterization of the jet streams over this region. We have developed a method to identify in the 6-hourly reanalysis data the jet streams both on a 2-D surface and in the 3-D domain and to capture the subtropical jet and the eddy-driven jet. This method uses a threshold of 20 m/s for the wind speed and the connected-component algorithm. It can also distinguish the subtropical jet from the eddy-driven jet. Most importantly, the subtropical jet over Asia has a core at 200 hPa, and stays around 25°N in winter until April, when it starts to move northward smoothly to around 45°N in mid-summer. The eddy-driven jet has its core between 200 and 300 hPa. Its latitudinal position has great variability even within a month; it can be between 45°N and 60°N in winter, and between 50°N and 70°N in summer. Both the East Asian rainbands and subtropical jet migrate northward from spring to summer, but they are not perfectly collocated in latitude. The role of jet streams in the season of the rainbands is more on interacting with the Tibetan Plateau to generate Rossby waves downstream than on directly controlling the frontogenesis.

The Asian summer monsoon supplies around 3 billion people with much of their yearly supply of freshwater, necessary for human consumption as well as in agriculture and industry. In many regions, particularly along the Ganges River in India and in northern China, use of freshwater far exceeds natural recharge rates. Given the high population density of these regions, a substantial fraction of Asia's population is therefore critically sensitive to interannual changes in the supply of freshwater by the monsoon, as well as potential future change under 21st century warming. This dissertation focuses on understanding the atmospheric dynamics of the leading mode of July-August Asian Monsoon rainfall variability, which links two major subsystems: the South Asian and East Asian monsoons.

In summer, two distinct rainfall régimes are observed: Convective storms over India, Bangladesh and Nepal (the South Asian monsoon), and frontal rainfall over China, Japan and Korea (the East Asian monsoon). In addition, the Himalayas and other orography, including the Arakan Mountains, Ghats and Yunnan Plateau, create smaller precipitation domains separated by sharp gradients. My research has revealed a previously unrecognized mode of continental precipitation variability that spans both South and East Asia during July and August. A dipole between the Himalayan Foothills and the ��Monsoon Zone�� of central India dominates July-August interannual variability in South Asia, and is also associated in East Asia with a tripole between the Yangtze Corridor and North and South China. By performing lag-lead correlation of rainfall, I show that this covariation does not correspond to the spatial pattern of July-August storm tracks. Instead, I hypothesize that interannual change in the strength of moisture transport from the Bay of Bengal to the Yangtze Corridor across the northern Yunnan Plateau induces widespread precipitation anomalies. Abundant moisture transport along this route requires both cyclonic monsoon circulation over India and sufficient heating over the Bay of Bengal, which occurs only during July and August, as I also show by analyzing existing runs with a zoomed, nested version of the LMDZ5 model nudged to reanalysis.

In China, a growing body of work identifies a régime shift in rainfall occurring in the late 1970s known as the ``South Flood-North Drought.'' However, this phenomenon has not previously been described in terms of the complex seasonal cycle of the East Asian monsoon. During the peak interval of rainfall, known as Meiyu season, a persistent but meandering front (the ``��Meiyu Front''��) delivers strong bouts of rainfall over a narrow latitude band. The preferred latitude of this front shifts throughout the year and shifts abruptly from May to July. I have developed an image processing algorithm, the Rainband Detection Algorithm (RDA), that produces a 57-year (1951-2007) daily catalog of frontal rainfall events in China, reporting latitude, intensity and zonal extent. This result allows for a quantitative assessment of Meiyu change between 1951-1979 and 1980-2007, which has never previously been performed on an event-by-event basis. I find that the greatest change in frontal behavior has occurred during May, when events have strongly decreased over the Yangtze River valley. In addition, a statistically significant southward shift has occurred in July-August Meiyu events. The rainfall changes associated with South Flood-North Drought can be attributed specifically to changes in banded rainfall. A suite of alternative simple metrics of China rainfall cannot capture all of the characteristics of the South Flood-North Drought in one succinct measure. Another distinct change occurred between 1979-1993 and 1994-2007 featuring intensification of Meiyu rainbands, possibly reflecting a different causal mechanism. Our RDA catalog is available to other East Asian monsoon researchers.

The leading Asian monsoon mode described in Chapter 2 (hereafter All-Asia EOF1) not only describes a key mode of Asian monsoon variability, but is also associated with climate anomalies across the entire Northern Hemisphere. We find an association between strong All-Asia EOF1 years and the Circumglobal Teleconnection (CGT), a high-latitude standing wave pattern with wavenumber 5 or 6 implicated in Northern Hemisphere heat wave events, which we further investigate through compositing with JRA-55 reanalysis. JRA-55 captures changes in water vapor transport across the Yunnan Plateau associated with All-Asia EOF1 as hypothesized in Chapter 2, and suggests that they correspond to an anomalous band of diabatic heating of amplitude 100 W/m^2. We propose that this heating can force the CGT and participates in the phase-locking of the CGT, which is further amplified by the configuration of high topography. Over China and East Asia, All-Asia EOF1 is manifested as a meridional jet shift. Using the RDA catalog from Chapter 3, we test whether rainfall shifts across the entire East Asian monsoon also correspond to jet changes. We find that the South Flood-North Drought robustly corresponds to a southward shift in the East Asian tropospheric jet. Thus, although IPCC5 projections of the 21st-century Asian monsoon remain inconclusive, future changes may be more predictable by further studying the effects of global warming on Bay of Bengal moisture transport, the CGT and the East Asian jet.

Ecosystems are an integral part of climate, interacting with the atmosphere by modifying fluxes of energy, water, momentum, and carbon---processes whose importance varies spatially. This dissertation focuses on the study of ecoclimate -- unraveling the complex interactions between ecosystems and climate using a variety of tools including simple models, global scale carbon-climate models, and direct analysis of site-level data.

Using a global climate model we show that transpired water vapor from expanded vegetation in the Arctic contributes greenhouse warming with a magnitude of forcing larger than that from the associated changes in albedo. This mechanism has not been considered in previous studies and here we suggest that it may negate the negative climate feedbacks expected from the surface latent cooling associated with the expansion of deciduous forest.

Additionally we find that large-scale expansion of low albedo mid latitude forests in the model has the ability to increase Northern Hemisphere temperatures, moving the thermal equator northward and thus modifying global circulation patterns. We observe a northward shift of the intertropical convergence zone over Africa suggesting that extratropical vegetation may be partly responsible for the maintenance of precipitation over the Sahara inferred from proxy records of vegetation from ~6000 Years Ago. The ability of vegetation to effect remote circulation has implications for the role that plants may play in the current climate as well as in climates of the past and the future.

In an analysis of nighttime respiration across the eddy covariance Ameriflux tower network we find a residual signal in respiration resembing seasonal variations in substrate mass after accounting for the effects of temperature and moisture. The prevalence of a mass signal indicates that allowing temperature sensitivity to change over time unphysically attributes variations in mass and moisture to changes in temperature. Using a simple carbon cycle model we show that ecosystems with long turnover times are most sensitive to long period forcing and therefore it will be harder to detect or attribute changes in these systems. This suggests that it may be easier to detect and attribute changes in the Tropics compared to the Boreal forest.

The annual cycle is the dominant mode of climate variability for most aspects of climate that humans engage with. Trends in the amplitude and timing of the annual cycle have been observed in both the surface temperature record and the atmospheric carbon dioxide record, but causes and significances of these changes remain poorly understood – in part because we lack an understanding of the character and mechanisms of natural variability.

Here we show that the phase of the annual cycle of surface temperature over extratropical land shifted towards earlier seasons by 1.7 days between 1954-2007, and that this change is highly anomalous with respect to earlier variations, which we take as indicative of the natural range. Significant changes in the amplitude of the annual cycle are also observed between 1954-2007. The observed land phase trends are explained by the cumulative effects of orbital changes, aliasing of the tropical year by sampling of temperature on the Gregorian calendar and changes in atmospheric circulation described by the Northern Annular Mode and the Pacific North America Pattern. These effects also appear to explain the contemporaneous amplitude gain trends, though this is less unambiguous. Ocean trends, in contrast, appear more significant after taking into account natural sources of variability.

The annual cycle of atmospheric carbon dioxide is controlled by variability in the carbon cycling of the terrestrial biosphere. Trends in the annual amplitude, timing of annual carbon draw-down (t_drop) and timing of annual carbon rise (t_rise) have thus been used to argue for a more photosynthetically active terrestrial biosphere, earlier initiation of biological spring, and increased terrestrial respiration in autumn respectively. We show that changes in amplitude are significant, anomalous when compared against a model of natural variability, spatially coherent over broad latitude bands and related to climate variability. Changes in t_drop are neither significant, nor anomalous, and are coherent over only limited spatial scales, indicating t_drop is a poor metric for diagnosing large-scale changes in the terrestrial biosphere. The utility of t_rise variability for diagnosing changes in the terrestrial biosphere is more ambiguous.