D. P. Edwards

National Center for Atmospheric Research, Boulder, Colorado, United States

Are you D. P. Edwards?

Claim your profile

Publications (126)235.87 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This review paper provides a framework for the application of the Observing System Simulation Experiment (OSSE) methodology to satellite observations of atmospheric constituents relevant for air quality. The OSSEs are experiments used to determine the potential benefit of future observing systems using an existing monitoring or forecasting system and by this can help to define optimal characteristics of future instruments. To this end observations from future instruments are simulated from a model representing the realistic state of the atmosphere and an instrument simulator. The added value of the new observations is evaluated through assimilation into another model or model version and comparison with the simulated true state and a control run.
    Atmospheric Environment 05/2015; 115. DOI:10.1016/j.atmosenv.2015.05.032 · 3.28 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Carbon monoxide (CO) is a key atmospheric compound that can be remotely sensed by satellite on the global scale. Fifteen years of continuous observations are now available from the MOPITT/Terra mission (2000 to present). Another fifteen and more years of observations will be provided by the IASI/MetOp instrument series (2007–2023>). In order to study long term variability and trends, a homogeneous record is required, which is not straightforward as the retrieved products are instrument and processing dependent. The present study aims at evaluating the consistency between the CO products derived from the MOPITT and IASI missions, both for total columns and vertical profiles, during a six year overlap period (2008–2013). The analysis is performed by first comparing the available 2013 versions of the retrieval algorithms, and second using a dedicated reprocessing of MOPITT CO profiles and columns based on the IASI a priori constraints. MOPITT v5T total columns are generally slightly higher over land (bias ranging from 0 to 13%) than IASI v20100815 data. When IASI and MOPITT data are retrieved with the same a priori constraints, correlation coefficients are slightly improved. Large discrepancies (total column bias over 15%) observed in the Northern Hemisphere during the winter months are reduced by a factor of 2 to 2.5. The detailed analysis of retrieved vertical profiles compared with collocated aircraft data from the MOZAIC-IAGOS network, illustrates the advantages and disadvantages of a constant vs. a variable a priori. On one hand, MOPITT agrees better with the aircraft profiles for observations with persisting high levels of CO throughout the year due to pollution or seasonal fire activity (because the climatology-based a priori is supposed to be closer to the real atmospheric state). On the other hand, IASI performs better when unexpected events leading to high levels of CO occur, due to the less constrained variance-covariance matrix.
    04/2015; 8(4):4095-4135. DOI:10.5194/amtd-8-4095-2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: We apply the Tropospheric Emission Spectrometer (TES) ozone retrieval algorithm to Infrared Atmospheric Sounding Instrument (IASI) radiances and characterise the uncertainties and information content of the retrieved ozone profiles. This study focuses on mid-latitudes for the year 2008. We validate our results by comparing the IASI ozone profiles to ozone sondes. In the sonde comparisons, we find a positive bias in the IASI ozone profiles in the UTLS region of up to 14% on average. For the described cases, the degrees of freedom for signal are on average 3.2, 0.3, 0.8, and 0.9 for the columns 0 km-top of atmosphere, (0-6) km, (0-11) km, and (8-16) km, respectively. We find that our biases with respect to sondes and our degrees of freedom for signal for ozone are comparable to previously published results from other IASI ozone algorithms. In addition to evaluating biases, we validate the retrieval errors by comparing predicted errors to the sample covariance matrix of the IASI observations themselves. For the predicted vs. empirical error comparison, we find that these errors are consistent and that the measurement noise and the interference of temperature and water vapour on the retrieval together mostly explain the empirically derived random errors. In general, the precision of the IASI ozone profiles is better than 20%.
    Atmospheric Measurement Techniques 12/2014; 7(12). DOI:10.5194/amt-7-4223-2014 · 3.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Measurements of Pollution in the Troposphere (MOPITT) Version 6 (V6) product for carbon monoxide (CO) incorporates several enhancements which will benefit many users of MOPITT data. V6 algorithm improvements are described in detail, and V6 validation results are presented. First, a geolocation bias related to the orientation of the MOPITT instrument relative to the TERRA platform was characterized and eliminated. Second, the variable a priori for CO concentrations for V6 is based on simulations performed with the chemical transport model Community Atmosphere Model with Chemistry (CAM-chem) for the years 2000–2009 instead of the model-derived climatology for 1997–2004 used for V5. Third, meteorological fields required for V6 retrieval processing are extracted from the MERRA (Modern-Era Retrospective Analysis For Research And Applications) reanalysis. Finally, a significant latitude-dependent retrieval bias in the upper troposphere in Version 5 products has been substantially reduced.
    Atmospheric Measurement Techniques 11/2014; 7(11):3623-3632. DOI:10.5194/amt-7-3623-2014 · 3.21 Impact Factor
  • 01/2014; 7(7):7013-7051. DOI:10.5194/amtd-7-7013-2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: [1] Validation results are reported for the MOPITT (Measurements of Pollution in the Troposphere) “Version 5” (V5) product for tropospheric carbon monoxide (CO) and are compared to results for the “Version 4” product. The V5 retrieval algorithm introduces (1) a method for reducing retrieval bias drift associated with long-term instrumental degradation, (2) a more exact representation of the effects of random errors in the radiances and, for the first time, (3) the use of MOPITT's near-infrared (NIR) radiances to complement the thermal-infrared (TIR) radiances. Exploiting TIR and NIR radiances together facilitates retrievals of CO in the lowermost troposphere. V5 retrieval products based (1) solely on TIR measurements, (2) solely on NIR measurements and (3) on both TIR and NIR measurements are separately validated and analyzed. Actual retrieved CO profiles and total columns are compared with equivalent retrievals based on in situ measurements from (1) routine NOAA aircraft sampling mainly over North America and (2) the “HIAPER Pole to Pole Observations” (HIPPO) field campaign. Particular attention is focused on the long-term stability and geographical uniformity of the retrieval errors. Results for the retrieved total column clearly indicate reduced temporal bias drift in the V5 products compared to the V4 product, and do not exhibit a positive bias in the Southern Hemisphere, which is evident in the V4 product.
    06/2013; 118(12). DOI:10.1002/jgrd.50272
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Practical implementations of chemical OSSEs (Observing System Simulation Experiments) usually rely on approximations of the pseudo-observations by means of a predefined parametrization of the averaging kernels, which describe the sensitivity of the observing system to the target atmospheric species. This is intended to avoid the use of a computationally expensive pseudo-observations simulator, that relies on full radiative transfer calculations. Here we present an investigation on how no, or limited, scene dependent averaging kernels parametrizations may misrepresent the sensitivity of an observing system. We carried out the full radiative transfer calculation for a three-days period over Europe, to produce reference pseudo-observations of lower tropospheric ozone, as they would be observed by a concept geostationary observing system called MAGEAQ (Monitoring the Atmosphere from Geostationary orbit for European Air Quality). The selected spatio-temporal interval is characterised by an ozone pollution event. We then compared our reference with approximated pseudo-observations, following existing simulation exercises made for both the MAGEAQ and GEOstationary Coastal and Air Pollution Events (GEO-CAPE) missions. We found that approximated averaging kernels may fail to replicate the variability of the full radiative transfer calculations. In addition, we found that the approximations substantially overestimate the capability of MAGEAQ to follow the spatio-temporal variations of the lower tropospheric ozone in selected areas, during the mentioned pollution event. We conclude that such approximations may lead to false conclusions if used in an OSSE. Thus, we recommend to use comprehensive scene-dependent approximations of the averaging kernels, in cases where the full radiative transfer is computationally too costly for the OSSE being investigated.
    Atmospheric Measurement Techniques 03/2013; 6(2):2413-2448. DOI:10.5194/amtd-6-2413-2013 · 3.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A current obstacle to the Observation System Simulation Experiments (OSSEs) used to quantify the potential performance of future atmospheric composition remote sensing systems is a computationally efficient method to define the scene-dependent vertical sensitivity of measurements as expressed by the retrieval averaging kernels (AKs). We present a method for the efficient prediction of AKs for multispectral retrievals of carbon monoxide (CO) and ozone (O3) based on actual retrievals from MOPITT on EOS-Terra and TES and OMI on EOS-Aura, respectively. This employs a multiple regression approach for deriving scene-dependent AKs using predictors based on state parameters such as the thermal contrast between the surface and lower atmospheric layers, trace gas volume mixing ratios (VMR), solar zenith angle, water vapor amount, etc. We first compute the singular vector decomposition (SVD) for individual cloud-free AKs and retain the 1st three ranked singular vectors in order to fit the most significant, orthogonal components of the AK in the subsequent multiple regression on a training set of retrieval cases. The resulting fit coefficients are applied to the predictors from a different test set of retrievals cased to reconstruct predicted AKs, which can then be evaluated against the true test set retrieval AKs. By comparing the VMR profile adjustment resulting from the use of the predicted vs. true AKs, we quantify the CO and O3 VMR profile errors associated with the use of the predicted AKs compared to the true AKs that might be obtained from a computationally expensive full retrieval calculation as part of an OSSE. Similarly, we estimate the errors in CO and O3 VMRs from using a single regional average AK to represent all retrievals, which has been a common approximation in chemical OSSEs performed to-date. For both CO and O3 in the lower troposphere, we find a significant reduction in error when using the predicted AKs as compared to a single average AK. This study examined data from the continental United States (CONUS) for 2006, but the approach could be applied to other regions and times.
    03/2013; 6(2):2751-2791. DOI:10.5194/amtd-6-2751-2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: A current obstacle to the observation system simulation experiments (OSSEs) used to quantify the potential performance of future atmospheric composition remote sensing systems is a computationally efficient method to define the scene-dependent vertical sensitivity of measurements as expressed by the retrieval averaging kernels (AKs). We present a method for the efficient prediction of AKs for multispectral retrievals of carbon monoxide (CO) and ozone (O3) based on actual retrievals from MOPITT (Measurements Of Pollution In The Troposphere) on the Earth Observing System (EOS)-Terra satellite and TES (Tropospheric Emission Spectrometer) and OMI (Ozone Monitoring Instrument) on EOS-Aura, respectively. This employs a multiple regression approach for deriving scene-dependent AKs using predictors based on state parameters such as the thermal contrast between the surface and lower atmospheric layers, trace gas volume mixing ratios (VMRs), solar zenith angle, water vapor amount, etc. We first compute the singular value decomposition (SVD) for individual cloud-free AKs and retain the first three ranked singular vectors in order to fit the most significant orthogonal components of the AK in the subsequent multiple regression on a training set of retrieval cases. The resulting fit coefficients are applied to the predictors from a different test set of test retrievals cased to reconstruct predicted AKs, which can then be evaluated against the true retrieval AKs from the test set. By comparing the VMR profile adjustment resulting from the use of the predicted vs. true AKs, we quantify the CO and O3 VMR profile errors associated with the use of the predicted AKs compared to the true AKs that might be obtained from a computationally expensive full retrieval calculation as part of an OSSE. Similarly, we estimate the errors in CO and O3 VMRs from using a single regional average AK to represent all retrievals, which has been a common approximation in chemical OSSEs performed to date. For both CO and O3 in the lower troposphere, we find a significant reduction in error when using the predicted AKs as compared to a single average AK. This study examined data from the continental United States (CONUS) for 2006, but the approach could be applied to other regions and times.
    Atmospheric Measurement Techniques 01/2013; 6(7):1633-1646. DOI:10.5194/amt-6-1633-2013 · 3.21 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Atmospheric carbon monoxide (CO) distributions are controlled by anthropogenic emissions, biomass burning, transport and oxidation by reaction with the hydroxyl radical (OH). Quantifying trends in CO is therefore important for understanding changes related to all of these contributions. Here we present a comprehensive record of satellite observations from 2000 through 2011 of total column CO using the available measurements from nadir-viewing thermal infrared instruments: MOPITT, AIRS, TES and IASI. We examine trends for CO in the Northern and Southern Hemispheres along with regional trends for Eastern China, Eastern USA, Europe and India. We find that all the satellite observations are consistent with a modest decreasing trend ∼-1% yr-1 in total column CO over the Northern Hemisphere for this time period and a less significant, but still decreasing trend in the Southern Hemisphere. Although decreasing trends in the United States and Europe have been observed from surface CO measurements, we also find a decrease in CO over E. China that, to our knowledge, has not been reported previously. Some of the interannual variability in the observations can be explained by global fire emissions, but the overall decrease needs further study to understand the implications for changes in anthropogenic emissions.
    Atmospheric Chemistry and Physics 01/2013; 13(2-2):837-850. DOI:10.5194/acp-13-837-2013 · 4.88 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The new Version 5 MOPITT (Measurements of Pollution in the Troposphere) product for carbon monoxide (CO) is the first satellite product to exploit simultaneous near-infrared and thermal-infrared observations to enhance retrieval sensitivity in the lower troposphere. This feature is important to air quality analyses and studies of CO sources. However, because of the influence of both thermal contrast and geophysical noise, the retrieval characteristics for this new multispectral product are highly variable. New V5 products for surface-level CO concentrations have been evaluated over the contiguous United States using both in situ vertical profiles and NOAA ground-based "Tall Tower" measurements. Validation results based on the in situ profiles indicate that retrieval biases are on the order of a few percent. However, direct comparisons with the Tall Tower measurements demonstrate that smoothing error, which depends on both the retrieval averaging kernels and CO variability in the lower troposphere, exhibits significant geographical and seasonal variability.
    Journal of Geophysical Research Atmospheres 07/2012; 117(D13):13306-. DOI:10.1029/2012JD017553 · 3.44 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Tropospheric O3 and CO are two important pollutants and play a major role in determining the oxidizing capacity of the troposphere. While observed correlations between O3 and CO from surface and aircraft measurements have been used in many studies to better understand the anthropogenic influence on O3, there have only been a few studies using global satellite data for this purpose. The correlation patterns provide also a valuable test for models to predict O3 photochemical production and global anthropogenic influence on O3. In this study, we propose to take advantage of the high spatio-temporal coverage of the Infrared Atmospheric Sounding Interferometer (IASI) instrument which provides a unique dataset for analyzing O3-CO correlations from simultaneous and vertically-resolved O3 and CO measurements. We examine global distributions of IASI O3-CO correlations in the lower, the middle and the upper troposphere during the spring (MAM) and the summer (JJA) 2008 to evaluate the continental influence on tropospheric ozone. Observed O3-CO correlations from IASI are also compared to the correlations simulated in the MOZART-4 global 3-D chemical transport model in order to test our understanding of anthropogenic influence on O3 and also to evaluate the photochemistry in the model. The feasibility for using the ratio dO3/dCO in estimating the O3 production is tested from IASI observations and compared to the MOZART-4 simulations. We further investigate the different sources of the correlations through analysis of CO and O3 from each CO and NOx emission sources tagged separately in the MOZART-4 simulations.
  • R. Drori · U. Dayan · D. P. Edwards · L.K. Emmons, · C. Erlick
    [Show abstract] [Hide abstract]
    ABSTRACT: Pollutants from global sources are known to affect the Eastern Mediterranean Shore (EMS). However, there has been no previous study explicitly locating the European sources, characterizing their transport pathways, and quantifying their contribution to local concentrations in the EMS. In the current study, spatially tagged carbon monoxide was used as a tracer for pollutant transport from Europe to the EMS over five consecutive years (2003-2007) using the global chemical transport model MOZART-4. The model results were compared against NOAA/GMD ground station data and remotely sensed data from the Terra/MOPITT satellite and found to agree well on monthly basis but do not agree on daily basis. On synoptic scale, there is agreement between MOZART and GMD during July to August. A budget analysis reveals the role of CO from hydrocarbon oxidation on CO concentration during summer. European anthropogenic emissions were found to significantly influence EM surface concentrations, while European biomass burning (BB) emissions were found to have only a small impact on EM surface concentrations. Over the five simulated years, only two European biomass burning episodes contributed more than 10 ppb to surface CO concentrations in the EM. CO enhancement in the EM during the summer was attributed to synoptic conditions prone to favorable transport from Turkey and Eastern Europe towards the EM rather than increased emissions. We attribute the apparently misleading association between CO emitted from European BB and CO enhancements over the EM to typical summer synoptic conditions caused by the lingering of an anticyclone positioned over the Western and Central Mediterranean Basin that lead to forest fires in the area. Combined with a barometric trough over the eastern part of the Mediterranean Basin, this generates a prevailing transport of air masses from Eastern Europe to the EMS. Synoptic scale variations are shown to change the transport pathways from Europe towards the EMS having an overall small affect. CO concentration over the EMS can be describe as having 3 components: the seasonal cycle, the cycle of CO produced from hydrocarbon oxidation and a synoptic variation.
    ATMOSPHERIC CHEMISTRY AND PHYSICS 01/2012; DOI:10.5194/acp-12-1067-2012 · 5.30 Impact Factor
  • M. N. Deeter · H. M. Worden · D. P. Edwards · J. C. Gille
    [Show abstract] [Hide abstract]
    ABSTRACT: The new MOPITT Version 5 product is the first satellite product for carbon monoxide (CO) to exploit multispectral observations. Together, near-infrared (NIR) and thermal-infrared (TIR) observations permit the retrieval of CO concentrations in the lower troposphere, near the actual sources. However, while this feature is potentially valuable for applications including chemical weather forecasting and inverse modeling, the retrieval characteristics of the new multispectral product (e.g., the retrieval averaging kernels) are highly variable. The underlying causes of this variability include thermal contrast and geophysical noise. Validating the new V5 product is challenging for several reasons. Compared to the TIR-only MOPITT product, random retrieval errors in the multispectral product are substantially larger. To compensate, stronger temporal and spatial averaging may be necessary. However, the temporal and spatial variability of CO in the planetary boundary layer (PBL) are much greater than in the free troposphere. Consequently, in-situ CO measurements made in the PBL may be poorly correlated with CO concentrations averaged over areas representative of MOPITT pixels (~ 22 km). Errors from this effect are only compounded as multiple MOPITT pixels are averaged together, particularly if there are significant regional sources. Previous MOPITT validation efforts have relied mainly on in-situ profiles measured from aircraft. While these are valuable, we are also exploring new sources of validation data. For example, we have begun comparing MOPITT retrieval results with CO in-situ measurements from the NOAA 'Tall Tower' network. Instruments on these towers measure CO concentrations at high temporal frequency and at multiple fixed altitudes within the PBL.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The Geostationary Coastal and Air Pollution Events (GEO-CAPE) mission is required to be able to measure ozone with two degrees of freedom for signal (DFS) in the troposphere, and to provide sensitivity in the lowest 2 km (lowermost troposphere, LMT). We explore the feasibility of using multispectral measurements from the ultraviolet (UV) to the thermal infrared (TIR) to improve LMT ozone retrievals compared to conventional retrievals employing only the UV or the TIR. Simulations spanning a range of ozone mixing ratios indicate that the combination of UV, visible and TIR significantly improves the total column DFS as well as the LMT DFS. We will present results both for clear sky scenarios and those containing realistic cloud and aerosol loadings.
  • A. Boynard · G. Pfister · D. P. Edwards
    [Show abstract] [Hide abstract]
    ABSTRACT: As a tracer of pollution and a precursor of tropospheric ozone, carbon monoxide (CO) is a key species for studying air quality. This pollutant is highly variable and the integration of observations and modelling can provide a better insight into, for example, pollution episode development and pollution transport on regional to global scales. In this work, we analyse CO concentrations simulated by the WRF chem regional model over the U.S. during the summer 2008 in order to characterize its geographical and temporal variations at the surface, in the lowermost troposphere and in the free troposphere. The model is evaluated by comparisons with surface measurements available from several EPA ground-based stations over the U.S. and satellite observations from Terra/MOPITT. CO tracers for different emission sources are also used to differentiate the variability due to dynamics, photochemistry and emissions and to assess the CO budget. Finally, we compare the representation of surface and planetary boundary layer CO concentration variability that would be captured by representative satellite sensor observations using the different CO spectral bands. These comprise thermal infrared measurements sensitive primarily to the free troposphere concentration, near-infrared measurements sensitive to the total column and combined multispectral measurements that would provide sensitivity to the lowermost troposphere concentration.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Carbon monoxide (CO) is an important trace gas for understanding air quality and atmospheric composition. It is a good tracer of pollution plumes and atmospheric dynamics. In this talk we describe the analysis of global and regional CO distributions as seen by both the MOPITT/Terra and IASI/MetOp missions, from mid-2007 up to now. Long term trends for IASI and MOPITT (version 3, 4 and 5) CO total columns over specific areas are presented, and show excellent agreement. CO distributions over polluted and clean regions are compared. Preliminary results of a study comparing the surface CO level for the two sounders over different cities will be presented, using MOPITT version 5 products that include the NIR and show a better sensitivity near the ground. Finally, this talk will also illustrate how the data processed at ULB-LATMOS are currently assimilated in the MACC project in near-real time in order to generate the CO pollution forecasts over Europe. Surface CO obtained after assimilation show better agreement with ground-based station measurements.
  • [Show abstract] [Hide abstract]
    ABSTRACT: Geophysical noise is caused by surface heterogeneity and field of view motionThe MOPITT retrieval algorithm now explicitly accounts for geophysical noiseAlgorithm changes enhance the sensitivity to CO near the surface
    Journal of Geophysical Research Atmospheres 08/2011; 116(D15). DOI:10.1029/2011JD015703 · 3.44 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we analyze tropospheric O3 together with HNO3 during the POLARCAT (Polar Study using Aircraft, Remote Sensing, Surface Measurements and Models, of Climate, Chemistry, Aerosols, and Transport) program, combining observations and model results. Aircraft observations from the NASA ARCTAS (Arctic Research of the Composition of the Troposphere from Aircraft and Satellites) and NOAA ARCPAC (Aerosol, Radiation and Cloud Processes affecting Arctic Climate) campaigns during spring and summer of 2008 are used together with the Model for Ozone and Related Chemical Tracers, version 4 (MOZART-4) to assist in the interpretation of the observations in terms of the source attribution and transport of O3 and HNO3 into the Arctic. The MOZART-4 simulations reproduce the aircraft observations generally well (within 15 %), but some discrepancies in the model are identified and discussed. The observed correlation of O3 with HNO3 is exploited to evaluate the MOZART-4 model performance for different air mass types (fresh plumes, free troposphere and stratospheric-contaminated air masses). Based on model simulations of O3 and HNO3 tagged by source type and region, we find that the anthropogenic pollution from the Northern Hemisphere is the dominant source of O3 and HNO3 in the Arctic at pressure greater than 400 hPa, and that the stratospheric influence is the principal contribution at pressures less 400 hPa. During the summer, intense Russian fire emissions contribute some amount to the tropospheric columns of both gases over the American sector of the Arctic. North American fire emissions (California and Canada) also show an important impact on tropospheric ozone in the Arctic boundary layer. Additional analysis of tropospheric O3 measurements from ground-based FTIR and from the IASI satellite sounder made at the Eureka (Canada) and Thule (Greenland) polar sites during POLARCAT has been performed using the tagged contributions. It demonstrates the capability of these instruments for observing pollution at Northern high latitudes. Differences between contributions from the sources to the tropospheric columns as measured by FTIR and IASI are discussed in terms of vertical sensitivity associated with these instruments. The first analysis of O3 tropospheric columns observed by the IASI satellite instrument over the Arctic is also provided. Despite its limited vertical sensitivity in the lowermost atmospheric layers, we demonstrate that IASI is capable of detecting low-altitude pollution transported into the Arctic with some limitations.
    Atmospheric Chemistry and Physics 08/2011; 11(8):23707-23760. DOI:10.5194/acpd-11-23707-2011 · 4.88 Impact Factor

Publication Stats

4k Citations
235.87 Total Impact Points

Institutions

  • 1997–2015
    • National Center for Atmospheric Research
      • Division of Atmospheric Chemistry
      Boulder, Colorado, United States
  • 2002–2014
    • National Research Center (CO, USA)
      Boulder, Colorado, United States
  • 2004–2006
    • University of Maryland, Baltimore County
      • Department of Physics
      Baltimore, Maryland, United States