Automated, Unsupervised Inversion of MultiwavelengthLidar Data With TiARA: Assessment of RetrievalPerformance of Microphysical Parameters UsingSimulated DataApplied Optics
- Detlef Mueller, Eduard Chemyakin, Alexei Kolgotin, Richard Ferrare, Chris Hostetler, and Anton Romanov
- received 09/28/2018; accepted 01/02/2019; posted 01/04/2019; Doc. ID 346995
- Abstract: We evaluate the retrieval performance of the automated, unsupervised inversion algorithm TiARA(Tikhonov Advanced Regularization Algorithm) which is used for the autonomous retrieval of microphysicalparameters of anthropogenic and natural pollution particles. TiARA (version 1.0) has been developedin the past 10 years and builds on the legacy of a data-operator controlled inversion algorithm that is usedsince 1998 for the analysis of data from multiwavelength Raman lidar. The development of TiARA hasbeen driven by the need to analyze in (near) real-time large volumes of data collected with NASA LangleyResearch Center’s HSRL-2 (High-Spectral-REsolution Lidar). HSRL-2 was envisioned as part of theNASA ACE (Aerosols-Clouds-Ecosystems) mission in response to the NAS Decadal Study (DS) missionrecommendations 2007. TiARA could thus also serve as inversion algorithm in the context of a futurespace-borne lidar. We summarize key properties of TiARA on the basis of simulations with monomodallogarithmic-normal particle size distributions which cover particle radii from approximately 0.05 – 10 mm.The real and imaginary parts cover the range from non-absorbing to highly light-absorbing pollutants.Our simulations included up to 25% measurement uncertainty. The goal of our study is to provide guidancewith respect to technical features of future space-borne lidars if such lidars will be used for retrievalsof microphysical data products, absorption coefficients, and single-scattering albedo. We investigated theimpact of two different measurement-error models on the quality of the data products. We also obtainedfor the first time a statistical view on systematic and statistical uncertainties if a large volume of data isprocessed. Effective radius is retrieved to 50% accuracy for 58% of cases with an imaginary part up to 0.01iand up to 100% of cases with an imaginary part of 0.05i. Similarly, volume concentration, surface-areaconcentration and number concentrations are retrieved to 50% accuracy in 56-100% of cases, 99-100% ofcases, and 54-87% of cases, respectively, depending on the imaginary refractive index. The numbers representmeasurement uncertainties of up to 15%. If we target 20% retrieval accuracy, the number of casesthat fall within that threshold are 36-76% for effective radius, 36-73% for volume concentration, 98-100%for surface-area concentration, and 37-61% for number concentration. That range of numbers again representsa spread in results for different values of the imaginary part. The real part should be retrieved toapproximately 0.075 or better. At present we obtain an accuracy of (on average) 0.1 for the real part. A casestudy from ORACLES is used to illustrate data products obtained with TiARA.