Linear estimation of particle bulk parameters from multi-wavelength lidar measurements

[摘要] An algorithm for linear estimation of aerosol bulk properties such asparticle volume, effective radius and complex refractive index frommultiwavelength lidar measurements is presented. The approach uses the factthat the total aerosol concentration can well be approximated as a linearcombination of aerosol characteristics measured by multi-wavelength lidar.Therefore, the aerosol concentration can be estimated from lidarmeasurements without the need to derive the size distribution, which entailsmore sophisticated procedures. The definition of the coefficients requiredfor the linear estimates is based on an expansion of the particle sizedistribution in terms of the measurement kernels. Once the coefficients areestablished, the approach permits fast retrieval of aerosol bulk propertieswhen compared with the full regularization technique. In addition, thestraightforward estimation of bulk properties stabilizes the inversionmaking it more resistant to noise in the optical data.

Numerical tests demonstrate that for data sets containing three aerosolbackscattering and two extinction coefficients (so called 3β + 2α) the uncertainties in the retrieval of particle volume and surface areaare below 45% when input data random uncertainties are below 20%.Moreover, using linear estimates allows reliable retrievals even when thenumber of input data is reduced. To evaluate the approach, the resultsobtained using this technique are compared with those based on thepreviously developed full inversion scheme that relies on the regularizationprocedure. Both techniques were applied to the data measured bymultiwavelength lidar at NASA/GSFC. The results obtained with both methodsusing the same observations are in good agreement. At the same time, thehigh speed of the retrieval using linear estimates makes the methodpreferable for generating aerosol information from extended lidarobservations. To demonstrate the efficiency of the method, an extended timeseries of observations acquired in Turkey in May 2010 was processed usingthe linear estimates technique permitting, for what we believe to be thefirst time, temporal-height distributions of particle parameters.

[学科分类] 几何与拓扑