Yonghua Wu

Remote Sensing Instruments Used for Measurement and Model Validation of Optical Parameters of Atmospheric Aerosols

With the dramatical climate changing that we are facing today, atmospheric monitoring is of major importance. Several atmospheric monitoring instruments are already being deployed for real-time surface-level retrieval of atmospheric composition, optical coefficients, particulate matter less than 2.5 μm in diameter (PM2.5), aerosol optical depth (AOD), and particle size distribution. However, these measurements are, in general, very cost intensive and can realistically be deployed only over very limited areas. Therefore, it is very important that advanced modeling methods be employed to fill these gaps and provide air quality predictions that can be used for forecasts as well as a better understanding of the interplay of meteorology, atmospheric emissions, and chemistry. In particular, for the New York State area, the New York State Department of Environmental Conservation uses Community Multiscale Air Quality (CMAQ) model to couple meteorology to local emissions, and there is intense interest in trying to assess the model performance beyond current surface network measurements. In particular, a deeper understanding of boundary layer processes can be made by experimentally exploring the vertical distribution model forecasts to better understand the underlying causes when model forecast results are not accurate. To this end, we develop a comprehensive Mie-scattering-based procedure including the effects of relative humidity that allows us to convert CMAQ aerosol distribution data into vertical-profile multiwavelength optical parameters that can be compared to column-integrated and vertical-profiling measurements to assess model performance and point to areas where the model is deficient. In particular, we make use of multiwavelength light detection and ranging (LIDAR), sunphotometer, ceilometer, and existing tapered element oscillating microbalance (TEOM) measurements to assess various vertical and column-integrated parameters of the CMAQ model under different stability conditions. In particular, we find that, for cases where the planetary boundary layer (PBL) is stable, the column-integrated AOD comparisons are in good agreement unlike the days with dynamic PBL. This is also consistent with observations that the TEOM PM2.5 trends are closely followed by the CMAQ model during these stable conditions. On the other hand, significant errors between the surface CMAQ PM2.5 and TEOM measurements can occur which can be traced to unphysically high particulate concentration profiles distributed too close to the surface not seen in ceilometer/LIDAR profiles. Finally, we note that, even for the stable cases, the multiwavelength optical depth data show that, for sufficiently low wavelengths, the column AOD in the model is underestimated, illustrating that there is a general underestimation of ultrafine (Aitken) particulates which can dramatically affect health.

Statistical comparison between Hysplit sounding and lidar observation of planetary boundary layer characteristics over New York City

The need to characterize in a robust way Planetary Boundary Layer (PBL) heights is crucial as in air quality forecast and transport models. In particular, incorrect determination of PBL heights can severely distort the surface air quality predictions such as PM2.5. Local properties and morphological features can influence PBL dynamics through local circulation phenomena such as the sea-breeze development as well as influences from the Urban Heat Island Canopy resulting in multiple layers that need to be resolved. In this paper, based on a combination of wavelet and image processing methods, we develop methods to quantify multilayer PBLs and assess their dynamics with meteorological measurements including temperature, wind and humidity profiles. In particular, meteorologically based PBL heights based on both the Potential Temperature Gradient and Richardson Number are compared against both lidar and ceilometer measurements. It is shown that in general, the Potential Temperature Gradient method is better correlated to the PBL dynamics. Meanwhile, the Hysplit model provides sounding data which can be used for comparison between actual sounding and lidar measurements. On the other hand, when strong atmospheric instability is present or layering develops, the comparison between different methods can provide information about the PBL internal structure. Further comparisons with air quality models such as MM5 are also made and illustrate the difficulty in these models properly predicting the PBL dynamics seen in urban areas.

Lidar application to monitoring emissions and transport of particulate pollution in urban environments with high temporal and spatial resolution

Attainment of National Ambient Air Quality Standard-NAAQS for exposure limits to air pollutants is of great concern to State and Local agencies and communities in the United State because of potential health impacts. This is particularly important and challenging in urban areas because of high population densities and complex terrain. Exceedances of NAAQS requires states to develop implementation plans to address them and as such, studying the horizontal and vertical distribution and mixing of pollutants is key to understanding their transport and evolution. In this study, vertical and scanning horizontal lidar measurements together with in situ observations from particulate matter and trace gas analyzers from state air quality networks are used to shed light on mechanisms that impact movement of aerosol, including emissions from power generating stations at periods of high electricity demand.

Applications of synergistic combination of remote sensing and in-situ measurements on urban monitoring of air quality

In this study, multiple remote sensing and in-situ measurements are combined in order to obtain a comprehensive understanding of the aerosol distribution in New York City. Measurement of the horizontal distribution of aerosols is performed using a scanning eye-safe elastic-backscatter micro-pulse lidar. Vertical distribution of aerosols is measured with a co-located ceilometer. Furthermore, our analysis also includes in-situ measurements of particulate matter and wind speed and direction. These observations combined show boundary layer dynamics as well as transport and inhomogeneous spatial distribution of aerosols, which are of importance for air quality monitoring.

Lidar, TEOM and Sunphotometer measured and model reconstructed atmospheric parameters

With the dramatically climate changing we are facing today atmospheric monitoring is of major importance. Several atmospheric monitoring instruments are used for measuring atmospheric composition, optical coefficients, PM2.5, aerosol optical depth, size distribution, PBL height and many other parameters. However an inexpensive method of determining these parameters is by use of models and one model that depicts the aerosol dynamics in the atmosphere is the Community Multi-scale Air Quality (CMAQ) model. Our paper is comparing the lidar, sunphotometer and TEOM measurements performed at City College of the City University of New York against CMAQ model.

Cloud optical depth measurement comparison between a Raman-Mie and Mie elastic lidar

In this paper, the properties of Low-level clouds are explored with a Raman-elastic lidar. In particular, we examine two complementary methods to measure thin cloud optical depth (COD). The first is direct integration of Raman Derived extinction while the second method utilizes a regression technique. We show that if we correct for aerosol influences the regression method for low cloud optical depth can be dramatically improved. Furthermore, estimates of extinction to backscatter ratio can be made within the cloud. We find that when the lidar ratio in cloud is averaged over the vertical extent, an S ratio on the order of 20 sr is found which is consistent with conventional water phase cloud droplet models.

Examination of hygroscopic properties of aerosols using a combined multiwavelength elastic-Raman lidar

Water vapor is an important greenhouse gas due to its high concentration in the atmosphere (parts per thousand) and its interaction with tropospheric aerosols particles. The upward convection of water vapor and aerosols due to intense heating of the ground leads to aggregation of water particles or ice on aerosols in the air forming different types of clouds at various altitudes. The condensation of water vapor on aerosols is affecting their size, shape, refractive index and chemical composition. The warming or cooling effect of the clouds hence formed are both possible depending on the cloud location, cover, composition and structure. The effect of these clouds on radiative global forcing and therefore on the short and long term global climate is of high interest in the scientific world. A major interest is manifested in obtaining accurate vertical water vapor profiles simultaneously with aerosol extinction and backscatter in the meteorological and remote sensing fields all around the globe in an effort to understand the hygroscopic properties of aerosols In previous work, simultaneous measurements of RH with backscatter measurements from a surface nephelometer were used to probe the hygroscopic properties of aerosols. However, most of these measurements were not able to probe the high RH domain since such high RH is rare for surface altitudes. For this reason, experiments using a 355 nm raman water vapor and aerosol lidar at the ARM site were used. Capable of providing simultaneous backscatter and RH profiles, and performing the experiments under low altitude cloud decks insured stable well mixed layers as well as probing RH profiles to above 95% which is required for the differentiation of different aerosol hygroscopic models.

GPS calibrated multiwavelength water vapor Raman lidar measurements to assess urban aerosol hygroscopicity

In this paper, we explore the possibility of determining thenature and variability of urban aerosol hygroscopic properties using multi-wavelength Raman lidar measurements at 355nm, as well as backscatter measurements at 532nm and 1064nm.. The addition of these longer wavelength channels allow us to more accurately validate the homogeneity of the aerosol layer as well as provide additional multiwavelength information that can be used to validate and modify the aerosol models underlying the hygroscopic trends observed in the Raman channel. In support of our hygroscopic measurements, we also discuss our calibration procedures for both the aerosol and water vapor profiles. The calibration algorithm we ultimately use for the water vapor measurements are twilight measurements where water vapor radiosonde data from the OKX station in NYS, are combined with total water vapor obtained from a GPS MET station. These sondes are then time correlated with independent near surface RH measurements to address any bias issues that may occur due to imperfect calibration due to lidar overlap issues and SNR limitations in seeing the water vapor at high altitudes.. In particular, we investigate the possibility of using ratio optical scatter measurments which eliminate the inherent problem of variable particle number and illustrate the sensitivity of different hygroscopic aerosols to these measurements. We find that the use of combine backscatter color ratios between 355 and 1064 together with the conventional extinction to backscatter ratio at 355nm should be able to improve retrieval of hygroscopic properties.

Atmospheric transport of Smoke and Dust Particulates and their interaction with the Planetary Boundary Layer as observed by multi-wavelength Lidar and supporting instrumentation

In this paper, we present results showing the usefulness of multi-wavelength lidar measurements to study the interaction of aerosols in the PBL with long range advected aerosol plumes. In particular, our measurements are used to determine the plume angstrom exponent, which allows us to differentiate smoke events from dust events, as well as partitioning the total aerosol optical depth obtained from a CIMEL sky radiometer between the PBL and the high altitude plumes. Furthermore, we show that only if the optical depth from the upper level plumes is taken into account, the correlation between the lidar derived PBL aerosol optical depth and surface PM2.5 is high. In addition, we also observe the dynamic interaction of high altitude plumes interacting with the PBL, resulting in a dramatic rise in surface PM10 concentrations without a corresponding dramatic rise in PM2.5 concentrations. These observations strongly suggest the deposition of large particulates into the PBL which is consistent with both lidar angstrom coefficient measurements and back-trajectory analysis. Finally, we investigate the correspondence between surface PM2.5 concentrations and optical backscatter coefficients as a function of altitude. To perform this study, our lidar system is replaced by a ceilometer (Vaisala CL-31) which can determine backscatter to near surface level. In particular, we confirm that near surface backscatter within the first 100 meters is a good proxy for PM2.5 but as altitude increases beyond 500 meters, the correlations degrades dramatically. These studies are useful in identifying the vertical length scales in which spaced based lidars such as Calipso can be used to probe surface PM2.5.

Raman-Mie lidar measurements of low and optically thin cloud

In this paper, we implement and compare two complementary methods for the measurement of low cloud optical depth with a Raman-Mie lidar over the metropolitan area of New York City. The first approach, based on the method of S. Young, determines the cloud optical depth by regressing the elastic signal against a molecular reference signal above and below the cloud. Due to high aerosol loading below and above the low cloud, correction for aerosol influence was necessary and achieved with the combined Raman-elastic returns. The second approach uses N2-Raman signal to derive cloud extinction profiles and then integrate them to determine optical depth. We find excellent agreements between these two retrievals for cloud optical depths as large as 1.5. Extinction-to-backscatter ratio within the low cloud is obtained and is shown to be consistent to values calculated from liquid water cloud model. The varied lidar ratios at cloud edge may imply the changes of cloud droplet size providing clues to the CCN seeding process. Furthermore, multiple-scattering effects on retrieving cloud optical depths are estimated by using an empirical model and specific lidar parameters.