Repository logo

Browsing by Author "Kummerow, Christian D., committee member"

Now showing 1 - 10 of 10
  • Thumbnail Image
    ItemOpen Access
    Characteristics and organization of precipitating features during NAME 2004 and their relationship to environmental conditions
    (Colorado State University. Libraries, 2008) Pereira, Luis Gustavo P., author; Rutledge, Steven A., advisor; Johnson, Richard H., committee member; Kummerow, Christian D., committee member; Cifelli, Robert C., committee member; Chandrasekar, V., committee member
    The focus of this study is to examine the characteristics of convective precipitating features (PFs) during the 2004 North American Monsoon Experiment (NAME) and their precursor environmental conditions. The goal is to gain a better insight into the predictability and variability of warm season convective processes in the southern portion of the North American Monsoon core region. The organization and characteristics of PFs are evaluated using composite radar reflectivity images over the southern portion of the Gulf of California. The environmental conditions are assessed using satellite images and a plethora of atmospheric observational analysis maps, such as winds at multiple levels, upper-level divergence, vorticity, vertical air motion, moisture and vertical cross-sections. Our study reveals that most PFs occurred during the afternoon and evening over land, especially near the foothills of the Sierra Madre Occidental. The vast majority of the precipitating features (~95%) were small, isolated, unorganized, short-lived convective cells. Mesoscale convective systems (MCSs) made up only 5% of the PF population. Nonetheless, these large, long-lived, precipitating features were responsible for 72% of the total precipitation within the radar composite region. An analysis of the number and rainfall produced by these MCSs revealed that they were not constant from day to day, but rather, varied significantly throughout NAME. We found that MCSs were more frequent when the atmosphere is thermodynamically unstable and the wind shear or large-scale dynamics favors the development of organized convection. Lastly, we examined the synoptic conditions associated with episodes of above average MCS rainfall in the southern portion of the NAME core region. Tropical waves were found to be an essential source of moisture and instability in the region. We also found that transient upper-level inverted troughs interact with the upper-level anticyclone to produce a "North American Monsoon Jet Streak" that created favorable dynamical uplift and wind shear conditions for MCS development.
  • Thumbnail Image
    ItemOpen Access
    Development and fabrication of low-mass, low-power, internally-calibrated, MMIC-based millimeter-wave radiometers at 92 and 166 GHz
    (Colorado State University. Libraries, 2012) Lee, Alexander L., author; Reising, Steven C., advisor; Notaros, Branislav, committee member; Kummerow, Christian D., committee member; Kangaslahti, Pekka, committee member
    This thesis discusses the design, fabrication, and testing of two millimeter-wave internally calibrated MMIC based radiometers operating at 92 and 166 GHz. These laboratory prototype radiometers are intended to increase the technological maturity of the radiometer components and reduce the risk, development time and cost of deploying satellite based radiometers operating in the 90-170 GHz frequency range. Specifically, radiometers at similar frequencies are being considered on NASA's SWOT mission, planned for launch in 2020. The SWOT mission is an ocean altimetry mission intended to increase the Earth science community's knowledge of the kinetic energy in ocean circulation and mesoscale eddies as well as the vertical transport of heat and carbon in the ocean. These direct detection Dicke radiometers have two internal calibration sources integrated in the front end. These sources consist of a high excess noise ratio noise diode and a temperature controlled matched load. Internal calibration is a requirement on ocean altimetry missions to avoid the need for moving parts, which are necessary to accomplish external calibration. The index of refraction of the atmosphere depends on temperature and humidity. The variability of humidity in time and space is more difficult to measure and model than that of temperature. Changes in the index of refraction of the atmosphere add error to satellite based ocean altimetry measurements. Microwave radiometers have been used on altimetry missions to measure the amount of atmospheric water vapor, and this data is used to correct the altimetry measurements. Traditionally, microwave radiometers in the 18-37 GHz range have been used on these missions. However, due to the large instantaneous fields of view (IFOV) on the Earth's surface, land begins to encroach upon the radiometer's surface footprint at about 40 km from the coast. The emission from the land adds additional error to the radiometer measurements. The amount of error is unknown due to the highly variable emissivity of land. The addition of higher frequency millimeter-wave radiometers in the 90-170 GHz frequency range will reduce the IFOV on the Earth's surface and therefore enable atmospheric water vapor measurements closer to the coasts. The radiometers presented in this thesis are laboratory prototypes. They are intended to demonstrate new component technology and improve estimates of mass, volume, power consumption, and radiometric performance for future space-borne millimeter-wave radiometers.
  • Thumbnail Image
    ItemOpen Access
    Estimation of snow microphysical properties with application to millimeter-wavelength radar retrievals for snowfall rate
    (Colorado State University. Libraries, 2011) Wood, Norman Bryce, author; Stephens, Graeme L., advisor; Cotton, William R., committee member; Fassnacht, Steven R., committee member; Kummerow, Christian D., committee member; Matrosov, Sergey Y., committee member
    The need for measuring snowfall is driven by the roles snow plays providing freshwater resources and affecting climate. Snow accumulations are an important resource for ecological and human needs and in many areas appear vulnerable to climate change. Snow cover modifies surface heat fluxes over areas extensive enough to influence climate at regional and perhaps global scales. Seasonal runoff from snowmelt, along with over-ocean snowfall, contributes to freshening in the Arctic and high-latitude North Atlantic oceans. Yet much of the Earth's area for which snowfall plays such significant roles is not well-monitored by observations. Radar reflectivity at 94 GHz is sensitive to scattering by snow particles and CloudSat, in a near-polar orbit, provides vertically resolved measurements of 94 GHz reflectivity at latitudes from 82 N to 82 S. While not global in areal coverage, CloudSat does provide observations sampled from regions where snowfall is the dominant form of precipitation and an important component of hydrologic processes. The work presented in this study seeks to exploit these observations by developing and assessing a physically-base snowfall retrieval which uses an explicit representation of snow microphysical properties. As the reflectivity-based snowfall retrieval problem is significantly underconstrained, a priori information about snow microphysical properties is required. The approaches typically used to develop relations between reflectivity and snowfall rate, so-called Ze-S relations, require assumptions about particle properties such as mass, area, fallspeed, and shape. Limited information about the distributions of these properties makes difficult the characterization of how uncertainties in the properties influence uncertainties in the Ze-S relations. To address this, the study proceeded in two parts. In the first, probability distributions for snow particle microphysical properties were assessed using optimal estimation applied to multi-sensor surface-based snow observations from a field campaign. Mass properties were moderately well determined by the observations, the area properties less so. The retrieval revealed nontrivial correlations between mass and area parameters not apparent in prior studies. Synthetic testing showed that the performance of the retrieval was hampered by uncertainties in the fallspeed forward model. The mass and area properties obtained from this retrieval were used to construct particle models including 94 GHz scattering properties for dry snow. These properties were insufficient to constrain scattering properties to match observed 94 GHz reflectivities. Vertical aspect ratio supplied a sufficient additional constraint. In the second part, the CloudSat retrieval, designed to estimate vertical profiles of snow size distribution parameters from reflectivity profiles, was applied to measurements from the field campaign and from an orbit of CloudSat observations. Uncertainties in the mass and area microphysical properties, obtained from the first part of this study, were substantial contributors to the uncertainties in the retrieved snowfall rates. Snowfall rate fractional uncertainties were typically 140% to 200%. Accumulations of snowfall calculated from the retrieval results matched observed accumulations to within 13%, however, when allowances were made for snowfall with properties likely inconsistent with the snow particle model. Information content metrics showed that the size distribution slope parameters were moderately to strongly constrained by the reflectivity observations, while the intercept parameters were determined primarily by the a priori constraints. Results from the CloudSat orbit demonstrated the ability of the CloudSat retrieval to represent a range of scene-dependent Ze-S relations.
  • Thumbnail Image
    ItemOpen Access
    Examining chaotic convection with super-parameterization ensembles
    (Colorado State University. Libraries, 2017) Jones, Todd R., author; Randall, David A., advisor; Kummerow, Christian D., committee member; Van den Heever, Susan S., committee member; Schumacher, Russ S., committee member; Eykholt, Richard E., committee member
    This study investigates a variety of features present in a new configuration of the Community Atmosphere Model (CAM) variant, SP-CAM 2.0. The new configuration (multiple-parameterization-CAM, MP-CAM) changes the manner in which the super-parameterization (SP) concept represents physical tendency feedbacks to the large-scale by using the mean of 10 independent two-dimensional cloud-permitting model (CPM) curtains in each global model column instead of the conventional single CPM curtain. The climates of the SP and MP configurations are examined to investigate any significant differences caused by the application of convective physical tendencies that are more deterministic in nature, paying particular attention to extreme precipitation events and large-scale weather systems, such as the Madden-Julian Oscillation (MJO). A number of small but significant changes in the mean state climate are uncovered, and it is found that the new formulation degrades MJO performance. Despite these deficiencies, the ensemble of possible realizations of convective states in the MP configuration allows for analysis of uncertainty in the small-scale solution, lending to examination of those weather regimes and physical mechanisms associated with strong, chaotic convection. Methods of quantifying precipitation predictability are explored, and use of the most reliable of these leads to the conclusion that poor precipitation predictability is most directly related to the proximity of the global climate model column state to atmospheric critical points. Secondarily, the predictability is tied to the availability of potential convective energy, the presence of mesoscale convective organization on the CPM grid, and the directive power of the large-scale.
  • Thumbnail Image
    ItemOpen Access
    GREMLIN: GOES radar estimation via machine learning to inform NWP
    (Colorado State University. Libraries, 2023) Hilburn, Kyle Aaron, author; Miller, Steven D., advisor; Kummerow, Christian D., committee member; Barnes, Elizabeth A., committee member; Ebert-Uphoff, Imme, committee member; Alexander, Curtis R., committee member
    Imagery from the Geostationary Operational Environmental Satellite (GOES) has been a key element of U.S. operational weather forecasting since 1975. The latest generation, the GOES-R Series, offers new capabilities to support the need for high-resolution rapidly refreshing imagery for situational awareness. Despite the well demonstrated value to human forecasters, usage of GOES imagery in data assimilation (DA) for initializing numerical weather prediction (NWP) has been limited, particularly in cloudy and precipitating scenes. By providing a rich and powerful library of nonlinear statistical tools, artificial intelligence (AI) / machine learning (ML) enables new approaches for connecting models and observations. The objective of this research is to develop techniques for assimilating GOES-R Series observations in precipitating scenes for the purpose of improving short-term convective-scale forecasts of high-impact weather hazards. The hypothesis of this dissertation is that by harnessing the power of ML, the new GOES-R capabilities can be used to create equivalent radar reflectivity suitable for initializing convection in high-resolution NWP models. Chapter 1 will present a proof-of-concept that ML can be used as an observation operator for GOES-R to simulate Multi-Radar Multi-Sensor (MRMS) composite reflectivity data and thereby initialize convection in NOAA's Rapid Refresh and High-Resolution Rapid Refresh (RAP/HRRR). Development of the GREMLIN (GOES Radar Estimation via Machine Learning to Inform NWP) convolutional neural network (CNN) will be described. This includes the creation of a hierarchy of open source datasets, and will emphasize the importance of the neural network loss function in focusing the attention of the network on the most important meteorological features. Explainable AI (XAI) tools are applied to GREMLIN to discover three primary strategies employed by the network in making predictions, highlighting the unique ability of CNNs to utilize spatial context in satellite imagery. The results of retrospective Rapid Refresh Forecast System (RRFS) forecasts will be described, which show that GREMLIN can produce more accurate short-term forecasts than using real radar data over areas of the U.S. with poor radar coverage. In Chapter 2, the Interpretable GREMLIN model is developed to elucidate the nature of the spatial context utilized by CNNs to make accurate predictions. This clarity is accomplished by moving the inner workings of the CNN out into a feature engineering step and replacing the neural network with a linear regression model. This exposes the effective input space of the CNN and establishes well defined relationships between inputs and outputs, which provides guarantees on how the model will respond to novel inputs. Despite a 24x reduction in the number of trainable parameters, the interpretable model has similar accuracy as the original CNN. Using the interpretable model, five additional physical strategies missed by XAI are discovered. The pros and cons of interpretable model development and implications for generalizability, consistency, and trustworthy AI will be discussed. Finally, Chapter 3 will extend this research for the development of Global GREMLIN, discussing the challenges and opportunities. GREMLIN is validated for regimes outside of the training dataset, and regime dependence is quantified in terms of temperature and moisture. The impacts of additional predictors and advanced ML architectures, and the derivation of uncertainty estimates that will be needed for new DA approaches in RRFS, will be discussed. Current efforts to implement GREMLIN on NOAA's GeoCloud, which will make GREMLIN available to a broader base of users, will be described.
  • Thumbnail Image
    ItemOpen Access
    Interactions of arctic clouds, radiation, and sea ice in present-day and future climates
    (Colorado State University. Libraries, 2016) Burt, Melissa Ann, author; Randall, David A., advisor; Kreidenweis, Sonia M., committee member; Kummerow, Christian D., committee member; Betsill, Michele M., committee member
    The Arctic climate system involves complex interactions among the atmosphere, land surface, and the sea-ice-covered Arctic Ocean. Observed changes in the Arctic have emerged and projected climate trends are of significant concern. Surface warming over the last few decades is nearly double that of the entire Earth. Reduced sea-ice extent and volume, changes to ecosystems, and melting permafrost are some examples of noticeable changes in the region. This work is aimed at improving our understanding of how Arctic clouds interact with, and influence, the surface budget, how clouds influence the distribution of sea ice, and the role of downwelling longwave radiation (DLR) in climate change. In the first half of this study, we explore the roles of sea-ice thickness and downwelling longwave radiation in Arctic amplification. As the Arctic sea ice thins and ultimately disappears in a warming climate, its insulating power decreases. This causes the surface air temperature to approach the temperature of the relatively warm ocean water below the ice. The resulting increases in air temperature, water vapor and cloudiness lead to an increase in the surface downwelling longwave radiation, which enables a further thinning of the ice. This positive ice-insulation feedback operates mainly in the autumn and winter. A climate-change simulation with the Community Earth System Model shows that, averaged over the year, the increase in Arctic DLR is three times stronger than the increase in Arctic absorbed solar radiation at the surface. The warming of the surface air over the Arctic Ocean during fall and winter creates a strong thermal contrast with the colder surrounding continents. Sea-level pressure falls over the Arctic Ocean and the high-latitude circulation reorganizes into a shallow "winter monsoon." The resulting increase in surface wind speed promotes stronger surface evaporation and higher humidity over portions of the Arctic Ocean, thus reinforcing the ice-insulation feedback. In the second half of this study, we explore the effects of super-parameterization on the Arctic climate by evaluating a number of key atmospheric characteristics that strongly influence the regional and global climate. One aspect in particular that we examine is the occurrence of Arctic weather states. Observations show that during winter the Arctic exhibits two preferred and persistent states — a radiatively clear and an opaquely cloudy state. These distinct regimes are influenced by the phase of the clouds and affect the surface radiative fluxes. We explore the radiative and microphysical effects of these Arctic clouds and the influence on these regimes in two present-day climate simulations. We compare simulations performed with the Community Earth System Model, and its super-parameterized counterpart (SP-CESM). We find that the SP-CESM is able to better reproduce both of the preferred winter states, compared to CESM, and has an overall more realistic representation of the Arctic climate.
  • Thumbnail Image
    ItemOpen Access
    Near-cloud aerosol retrieval and three-dimensional radiative transfer using machine learning
    (Colorado State University. Libraries, 2021) Yang, Chen-Kuang, author; Chiu, Christine, advisor; Kummerow, Christian D., committee member; Miller, Steven D., committee member; Ebert-Uphoff, Imme, committee member
    According to the most recent report of the Intergovernmental Panel on Climate Change, aerosols remain one of the largest sources of uncertainty in estimating and interpreting the Earth's changing energy budget. To reduce the uncertainty, an advanced understanding of aerosol optical properties and aerosol-cloud interaction is needed, which has largely relied on (but is not limited to) passive satellite observations. Current aerosol retrieval methods require a separation between cloud-free and cloudy regions, but this separation is often ambiguous. Three-dimensional (3D) cloud radiative effects can extend beyond the physical boundaries and enhance the reflectance in adjacent cloud-free regions as far as 10 km from clouds. Aerosol optical properties cannot be accurately retrieved without considering the 3D cloud radiative effect in this so-called "twilight" or "transition" zone, which denotes the area between cloud-free and cloudy regions. Indeed, most contemporary retrievals discard these regions, making it impossible to estimate the aerosol radiative effects in this zone. To help break the deadlock, 3D cloud radiative effects must be incorporated into the retrieval methods, and two approaches are proposed in this work, both leveraging machine learning techniques. The first approach accounts for 3D cloud radiative effects by building a 3D shortwave radiative transfer emulator as the forward model for the retrieval methods. Our emulator was trained by cumulus scenes generated from large eddy simulations and radiation fields calculated from 3D radiative transfer, to predict downward and upward flux profiles at a 500 m horizontal resolution and 30 m vertical resolution. From a case drawn from the testing dataset, our emulator captures the spatial pattern of the surface downwelling flux (e.g., shadows and illuminations), and the associated PDF has a remarkable similarity to the synthetic truth. In addition, compared to 1D calculation, our 3D emulator improves the root-mean-square-error by a factor of 6. For the flux and heating rate profiles, our emulator is much superior to the 1D calculation for the cloudy column. The application of this 3D radiative transfer emulator to numerical weather modeling or large-eddy simulations type of model is beyond the scope of the current work to develop an aerosol retrieval algorithm, but the possibility exists to do so. While the promising results from the emulator make it possible to conduct 3D RT retrieval methods, this approach still faces ambiguity in separating cloud-free and cloudy pixels. Here, we present a new retrieval algorithm for aerosol optical depth (AOD) in the vicinity of clouds which contains two unique features. First, it does not require pre-separation of aerosols and clouds. Second, it incorporates 3D radiative effects, allowing us to provide accurate aerosol retrievals near clouds. The AOD retrieval uncertainty of this method in the cloud-free region is (0.0004 ± 4% AOD), which is much better than the (0.03 ± 5% AOD) retrieval uncertainty in NASA Aerosol Robotic Network (AERONET). This method shows skill of predicting AOD over the near-cloud regions, and its validity was confirmed by using one of the explainable artificial intelligence methods to demonstrate that the model's decisions are supported by radiative transfer theory. Finally, a case study using MODIS observations shed light on how this new method can be applied to real world observation, possibly leading to new scientific insight on aerosol structure and aerosol-cloud interaction.
  • Thumbnail Image
    ItemOpen Access
    Observations and simulations of the interactions between clouds, radiation, and precipitation
    (Colorado State University. Libraries, 2016) Naegele, Alexandra Claire, author; Randall, David A., advisor; Kummerow, Christian D., committee member; Ramirez, Jorge A., committee member
    The first part of this study focuses on the radiative constraint on the hydrologic cycle as seen in observations. In the global energy budget, the atmospheric radiative cooling (ARC) is approximately balanced by latent heating, but on regional scales, the ARC and precipitation are inversely related. We use precipitation data from the Global Precipitation Climatology Project and radiative flux data from the Clouds and the Earth's Radiant Energy System (CERES) project to investigate the radiative constraint on the hydrologic cycle and how it changes in both space and time. We find that the effect of clouds is to decrease the ARC in the tropics, and to increase the ARC in middle and higher latitudes. We find that, spatially, precipitation and the ARC are negatively correlated in the tropics, and positively correlated in middle and higher latitudes. In terms of the global mean, the precipitation rate and the ARC are temporally out-of-phase during the Northern Hemisphere winter. In the second part of this study, we use a cloud-resolving model to gain a deeper understanding of the relationship between precipitation and the ARC. In particular, we explore how the relationship between precipitation and the ARC is affected by convective aggregation, in which the convective activity is confined to a small portion of the domain that is surrounded by a much larger region of dry, subsiding air. We investigate the responses of the ARC and precipitation rate to changes in the sea surface temperature (SST), domain size, and microphysics parameterization. Both fields increase with increasing SST and the use of 2-moment microphysics. The precipitation and ARC show evidence of convective aggregation, and in the domain average, both fields increase as a result. While running these sensitivity tests, we observed a pulsation in the convective precipitation rate, once aggregation had occurred. The period of the pulsation is on the order of ten simulated hours for a domain size of 768 km. The sensitivity tests mentioned above were used to investigate the mechanism of the pulsation. We also performed an additional test with no evaporation of falling rain, which leads to no cold pools in the boundary layer. Our results show that the period of the pulsation is noticeably sensitive to microphysics and domain size. The pulsation disappears completely when cold pools are prevented from forming, which suggests a "discharge-recharge" mechanism.
  • Thumbnail Image
    ItemOpen Access
    Origins and impacts of tropopause layer cooling in tropical cyclones
    (Colorado State University. Libraries, 2020) Rivoire, Louis, author; Birner, Thomas, advisor; Knaff, John A., advisor; Bell, Michael M., committee member; Davis, Christopher A., committee member; Kummerow, Christian D., committee member; Venayagamoorthy, Subhas K., committee member
    Remote sensing data from GPS radio occultation reveal temperatures lower than climatological average over a layer several kilometers deep near the tropopause above tropical cyclones (TCs). This signal, here referred to as tropopause layer cooling (TLC), occurs primarily during TC intensification and on spatial scales of the order of 1000 km. TLC has been hypothesized to be the result of: 1) Adiabatic expansion in cloud tops that overshoot the local level of neutral buoyancy. 2) Long wave radiative effects near the cloud top. 3) Adiabatic expansion in the TC secondary circulation. The relative role of these mechanism has not been quantified yet, perhaps pertaining to the large uncertainties and relative lack of vertical resolution of observational data sets and numerical modeling studies near the tropopause. Given the complex relationships between the thermal structure of the upper troposphere and the TC secondary circulation, determining which mechanisms are at play is paramount. TLC is also expected to destabilize the upper troposphere to convection and allow clouds to reach higher altitudes, likely leading to subtle but consequential changes in the secondary circulation and associated latent heating vertical distribution. Low temperatures near the tropopause can lead to in situ formation of cirrus clouds, which impact the radiative budget in the tropical tropopause layer. Lastly, low temperatures above convective systems have been linked to dehydration of the stratosphere, prompting the question of the role of TCs on the climate. Mechanism 1 is discussed in light of existing literature and suggested to be of marginal importance. Mechanisms 2 and 3 are examined using a combination of observational and theoretical analysis, and numerical modeling. Radiative heating rates calculated using cloud properties retrieved by the A-train suggest that mechanism 2 may explain up to half of TLC in the inner core, but only marginal amounts of TLC at larger radii. While reanalysis data sets suggest that mechanism 3 may explain TLC, numerical simulations of TCs with higher resolution suggest that mechanism 3 does not act in a way consistent with the secondary circulation as is typically pictured, and may need to be revisited. Other mechanisms involving processes which violate gradient wind balance near the tropopause need to be formulated. Finally, feedbacks between TLC, cloud structure, and TC dynamics are examined using parcel theory and idealized simulations. Parcel theory predicts that the TC thermal structure exerts a positive feedback on cloud top height during intensification, especially when convective entrainment is taken into account. While idealized simulations capture this general behavior, they exhibit other complex, transient behaviors which indicate breaking points in the interaction between clouds and their thermal environment.
  • Thumbnail Image
    ItemOpen Access
    The impact of aerosols on space-based retrievals of carbon dioxide
    (Colorado State University. Libraries, 2015) Nelson, Robert R., author; O'Dell, Christopher W., advisor; Denning, A. Scott, committee member; Kummerow, Christian D., committee member; Lefsky, Michael A., committee member
    This work describes an investigation into the impact of aerosols on space-based retrievals of the column-averaged dry-air mole fraction of carbon dioxide (XCO2). It was initially hypothesized that a simplified non-scattering, or "clear sky", retrieval, which neglects scattering and absorption by clouds and aerosols, could potentially avoid errors and biases brought about by attempting to measure properties of clouds and aerosols when there are none present. Clear sky retrievals have the benefit of being orders of magnitude faster and potentially as accurate as "full physics" retrievals that attempt to gain information about clouds and aerosols. Real data from the Greenhouse Gases Observing Satellite (GOSAT) and simulated data from the Orbiting Carbon Observatory-2 (OCO-2) were analyzed to find conditions under which a clear sky retrieval might perform as well as a full physics retrieval. It was found that for real GOSAT data the clear sky retrieval performed relatively well over land but not as well over ocean. The opposite conclusion was found for simulated OCO-2 data: it performed well over ocean but poorly over land. For both real GOSAT data and simulated OCO-2 data, high levels of filtering were needed for the clear sky retrieval to be able to perform nearly as well as or better than the full physics retrieval for both land and ocean surfaces. Spectral residuals were then examined to determine if the clear sky algorithm's performance was tied to errors in the spectral fitting. It was found that the clear sky retrievals had larger residuals than the full physics retrievals but that reducing the clear sky residuals by allowing them to fit for a customized residual pattern did little to reduce the XCO2 errors. It was also shown that even very clear scenes can result in small but detectable clear sky residual patterns. A comparison of cloud and aerosol properties measured by the XCO2 retrieval algorithm to aerosol optical depths from the AErosol RObotic NETwork (AERONET) revealed that the algorithm is generally unable to accurately retrieve information about the amount of clouds and aerosols present. Using OCO-2 simulations, it was shown that the algorithm is also only somewhat able to retrieve the heights of the aerosol layers. Information retrieved about individual aerosol types was shown to be even less accurate. Finally, early work in this study prompted investigation into how sensitive the XCO2 retrieval algorithm is to the first guess of aerosol properties. χ² space was explored by varying the first guess of various aerosol parameters. It was revealed that the retrieved aerosol information and XCO2 values can be highly sensitive to the first guess of the state vector, indicating significant nonlinearity in the retrieval's forward model. Two main conclusions were derived from this work. The first is that an analysis of real GOSAT clear sky XCO2 retrievals and simulated OCO-2 clear sky XCO2 retrievals revealed that the clear sky algorithm is generally inferior to the full physics algorithm, except for when high levels of filtering are applied. The second conclusion is that the current aerosol parameterization leads to unacceptable levels of nonlinearity in the XCO2 retrieval. These results motivate further study to improve the retrieval algorithm's aerosol parameterization, either directly or by including additional information, which may result in an improvement of the retrieval algorithm's ability to accurately measure XCO2.