Le Kuai, PhD
 Assistant Researcher
University of California, Los Angles
Joint Institute for Regional Earth System Science & Engineer
 
Jet Propulsion Laboratory
4800 Oak Grove Drive
Pasadena, CA 91109
Mail stop: 233-200 Email: kl (at) gps.caltech.edu

 


 
Education Professional Experience Other Working Experience Activities Honors Professional Memberships Research Descriptions

Accurate global measurements of surface fluxes of greenhouse gases (e.g. such as anthropogenic emission of CO2 from the populated LA basin and the uptake by the biosphere at the Amazon forest) and other atmospheric compositions are crucial for better understanding atmospheric changes related to natural process and anthropogenic activities. Besides in situ surface flux measurements on ground, tropospheric concentrations observed from ground and space are commonly used to infer the global surface fluxes. My primary interest is to develop robust retrieval algorithms based on the Bayesian optimal estimation for minor atmospheric species such as CO2, carbonyl sulfide (OCS), methane (CH4), and ozone (O3) from ground-based and space-based spectral measurements that are highly relevant to the global carbon cycle, air pollution, and climate change. I also have research experiences on atmospheric dynamics, such as solar cycle modulations on Quasi-Biennial Oscillation (QBO), the study of topical/polar intraseasonal variability of ozone, and numerical simulations of the tornado with its comparison to the field measurements. Some of these works were my graduate study and some of them were collaboration with other people.

Atmospheric CO2 retrievals

As more and more high-resolution CO2 become available due to both existing (e.g. NASA’s AIRS and TES, ESA’s IASI and SCIAMACHY, and JAXA’s GOSAT) and future (e.g. NASA’s OCO-2, ESA’s CarbonSat, and China’s TanSat) satellite instruments, the processing time of these huge datasets is one of the biggest problems in generating real-time retrieval products. While high spectral resolution measurements may be desired, the information of absorptions by a specific molecule at different wavelengths are usually degenerated and thus more spectral channels do not necessarily means more accurate retrievals of the tracer concentrations; sometimes more channels may even upset the retrieval algorithm, thereby introducing more errors. I demonstrated that a proper selection of a few tens of TCCON and TES spectral channels would not only speed up the process of the retrievals but also reduce the error in the retrieved CO2 due to uncertainties of surface pressure, temperature, and interference gases, such as H2O; see Kuai et al. (2010), J. Quant. Spectro. Rad. Trans., 111, 1296-1304 for details.

Ground-based spectral measurements with much higher accuracy and precision than those of satellite measurements have usually been used to validate satellite measurements. The Total Carbon Column Observation Network (TCCON) has been recently established to measure the CO2 absorption in the incoming solar Near Infra-Red (NIR) spectrum. This project was originally designed to measure the total column CO2 (TCO) only. But in my graduate studies, I showed that the precision of the measured spectra actually allows an estimation of tropospheric CO2 in ~ 3 bulk layers. I also showed that the TCCON TCO data subtracted by the free tropospheric CO2 measured by the spaceborne Tropospheric Emissions Spectrometer (TES) may provide a robust estimation of the CO2 seasonal cycle in boundary layer, as compared to the Southern Great Plains (SGP) aircraft profile monthly data in 2009. These vertically resolved CO2 profiles are extremely valuable for better characterizing the surface fluxes as well as the influence of tropospheric dynamics. I am planning to apply the same method to combine two satellite measurements such as TES and the JAXA’s Global Greenhouse Gas Observation by Satellite (GOSAT) to estimate the global distribution of the boundary layer CO2. Similar algorithm may also apply to other greenhouse gases such as methane; see Kuai et al. (2012), J. Quant. Spectro. Rad. Trans., 113, 1753-1761 for details.

Atmospheric OCS retrievals

OCS is the most abundant sulfur gas in the troposphere with a global averaging mixing ratio of about 500 part per trillion (ppt). Ocean is the major source, emitting OCS directly or its precursors, carbon disulfide and dimethyl sulfide. While OCS and CO2 uptake into leaves and soils through the same physical, diffusion, and pathway, the subsequent hydration reactions are reversible for CO2 only, i.e. some CO2 may be released back to the atmosphere by respiration. As a result, ecosystem and local eddy covariance studies for CO2 can only resolve the Net ecosystem exchange (NEE). On the other hand, the irreversible hydration reaction for OCS results in a one-way flux into the land biosphere. The resultant imbalance between the CO2/OCS budget can thus be employed as a proxy for partitioning the regional carbon flux into respiration and photosynthesis components, providing constraints on gross primary production (GPP). I have been retrieving free tropospheric OCS concentrations over ocean to understand its spatial and temporal distribution using TES IR spectra. This OCS product is validated against aircraft measurements and can be used to characterize the error budget and improve the constraints on the ocean flux. I plan to extend the TES OCS retrieval over land, which will serve as the first spaceborne OCS observations to further elucidate the impacts of anthropogenic OCS emissions to the air quality in megacities; see Kuai et al. (2013), Atmos. Meas. Tech., 6, 6975-7003 for details.

QBO and Solar Cycle

The Quasi-Biennial Oscillation (QBO) is an oscillation in the equatorial zonal wind in the stratosphere. The easterly and westerly components alternate with periods varying from about 24 to 30 months. Although QBO occurs in the stratosphere and origins in the equatorial region, it has wide impacts on tropospheric dynamics and remotely interactive with polar region. The variation of the QBO period has additional significance, especially with respect to the timing of its phase relative to the Northern Hemisphere (NH) winter, a phenomenon called seasonal synchronization. In both observational and modeling studies, I found that the QBO period has a tendency to synchronize with the Semi-Annual Oscillation (SAO) in the upper stratosphere. This explains that the currently observed average QBO period of 28 months is likely a result of intermittent jumps of the QBO period from four SAO to five SAO periods. In addition, the 11-year solar cycle is found to have a modulation on the tendency of the synchronization of QBO to SAO. Based on a modeling work, I found that the solar forcing would shift the distribution of QBO periods corresponding to 24 and 30 months. As a result, the record-averaged QBO period increases with the solar forcing in the statistical sense. This work serves as a basis for future studies of observations and modeling for understanding the variation of the QBO period related to other dynamic phenomenon and solar radiative forcing and its impact on the future climate; see Kuai et al. (2008), J. Atmos. Sci., 66, 1654-1664 and Kuai et al. (2009), J. Atmos. Sci., 66, 2418-2428.

Numerical simulation of Tornado

Tornados cause more than $1 billion in damages and over 80 deaths per year in the United States. One of the main types of damage is building collapse. Near-surface wind speeds in a tornado can exceed 100 m/s and cause significant damage, as the swirling winds exert greater loads on structures than straight-line. A better understanding of tornado-induced wind loads is needed to improve the design of typical structures to resist these winds. An accurate understanding of the loads requires knowledge of near-ground tornado winds. I was involved in a numerical simulation of tornados, which aimed to understand the near-ground flow field in tornados below 20−50 m, where radar data are not accurate. Tornado-induced wind loads on typical structures were determined so as to help design the structures to withstand F0-F2 tornados. The simulation examined the model sensitivity of various parameters that might affect laboratory-simulated tornados. The study found that the mesh size, inflow radius, and surface roughness are most important parameters for the simulated tornado dynamics. The vortex intensity or magnitude of the maximum tangential speed depends upon the input angular momentum or tangential speed distribution at the far field; see Kuai et al. (2008), Wind and Structures, 11, 75-96 for details.

Publications Selected Conference Presentations
Last Modified: PST