A neural network aerosol-typing algorithm based on lidar data

Abstract. Atmospheric aerosols play a crucial role in the Earth's system, but their role is not completely understood, partly because of the large variability in their properties resulting from a large number of possible aerosol sources. Recently developed lidar-based techniques were able to retrieve the height distributions of optical and microphysical properties of fine-mode and coarse-mode particles, providing the types of the aerosols. One such technique is based on artificial neural networks (ANNs). In this article, a Neural Network Aerosol Typing Algorithm Based on Lidar Data (NATALI) was developed to estimate the most probable aerosol type from a set of multispectral lidar data. The algorithm was adjusted to run on the EARLINET 3 β + 2 α ( + 1 δ ) profiles. The NATALI algorithm is based on the ability of specialized ANNs to resolve the overlapping values of the intensive optical parameters, calculated for each identified layer in the multiwavelength Raman lidar profiles. The ANNs were trained using synthetic data, for which a new aerosol model was developed. Two parallel typing schemes were implemented in order to accommodate data sets containing (or not) the measured linear particle depolarization ratios (LPDRs): (a) identification of 14 aerosol mixtures (high-resolution typing) if the LPDR is available in the input data files, and (b) identification of five predominant aerosol types (low-resolution typing) if the LPDR is not provided. For each scheme, three ANNs were run simultaneously, and a voting procedure selects the most probable aerosol type. The whole algorithm has been integrated into a Python application. The limitation of NATALI is that the results are strongly dependent on the input data, and thus the outputs should be understood accordingly. Additional applications of NATALI are feasible, e.g. testing the quality of the optical data and identifying incorrect calibration or insufficient cloud screening. Blind tests on EARLINET data samples showed the capability of NATALI to retrieve the aerosol type from a large variety of data, with different levels of quality and physical content.

[1]  Larry D. Travis,et al.  T-matrix computations of light scattering by large spheroidal particles , 1994 .

[2]  Alexandros Papayannis,et al.  Using Raman-lidar-based regularized microphysical retrievals and Aerosol Mass Spectrometer measurements for the characterization of biomass burning aerosols , 2015, J. Comput. Phys..

[3]  Hiroaki Kuze,et al.  An intercomparison of lidar‐derived aerosol optical properties with airborne measurements near Tokyo during ACE‐Asia , 2003 .

[4]  Takashi Shibata,et al.  Free tropospheric aerosol backscatter, depolarization ratio, and relative humidity measured with the Raman lidar at Nagoya in 1994-1997: contributions of aerosols from the Asian Continent and the Pacific Ocean , 2000 .

[5]  Tatsuro Tsukamoto,et al.  Characterization of Asian dust and Siberian smoke with multi‐wavelength Raman lidar over Tokyo, Japan in spring 2003 , 2004 .

[6]  L. Alados-Arboledas,et al.  Microphysical characterization of long-range transported biomass burning particles from North America at three EARLINET stations , 2016 .

[7]  Mohamed F. Tolba,et al.  Enhancement of OMI aerosol optical depth data assimilation using artificial neural network , 2012, Neural Computing and Applications.

[8]  Albert Ansmann,et al.  Optical properties of long-range transported Saharan dust over Barbados as measured by dual-wavelength depolarization Raman lidar measurements , 2015 .

[9]  V. Freudenthaler,et al.  Saharan dust contribution to the Caribbean summertime boundary layer –a lidar study during SALTRACE , 2016 .

[10]  R. Ferrare,et al.  Aerosol classification using airborne High Spectral Resolution Lidar measurements – methodology and examples , 2011 .

[11]  M. Vaughan,et al.  Separating mixtures of aerosol types in airborne High Spectral Resolution Lidar data , 2013 .

[12]  V. Freudenthaler,et al.  Depolarization ratio profiling at several wavelengths in pure Saharan dust during SAMUM 2006 , 2009 .

[13]  B. Weinzierl,et al.  Aerosol classification by airborne high spectral resolution lidar observations , 2012 .

[14]  V. Loiko,et al.  Neural networks for aerosol particles characterization , 2016 .

[15]  N. Mahowald.  Aerosol Indirect Effect on Biogeochemical Cycles and Climate , 2011, Science.

[16]  R. Wolke,et al.  An assessment of aerosol optical properties from remote-sensing observations and regional chemistry–climate coupled models over Europe , 2017 .

[17]  B. Gross,et al.  Aerosol hygroscopic models based on in situ measurements and lidar retrievals , 2007 .

[18]  A. Ansmann,et al.  Volcanic aerosol layers observed with multiwavelength Raman lidar over central Europe in 2008–2009 , 2010 .

[19]  V. Freudenthaler,et al.  Dual-wavelength linear depolarization ratio of volcanic aerosols: Lidar measurements of the Eyjafjallajökull plume over Maisach, Germany , 2012 .

[20]  Dietrich Althausen,et al.  Modification of Local Urban Aerosol Properties by Long-Range Transport of Biomass Burning Aerosol , 2018, Remote. Sens..

[21]  Isabella Morlini,et al.  Artificial neural network estimation of rainfall intensity from radar observations , 2000 .

[22]  A. Ansmann,et al.  Microphysical particle parameters from extinction and backscatter lidar data by inversion with regularization: theory. , 1999, Applied optics.

[23]  Christoph U. Keller,et al.  Use of neural networks in ground-based aerosol retrievals from multi-angle spectropolarimetric observations , 2014 .

[24]  Anil K. Jain,et al.  Statistical Pattern Recognition: A Review , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[25]  A. Smirnov,et al.  AERONET-a federated instrument network and data archive for aerosol Characterization , 1998 .

[26]  J. Jimenez,et al.  Aerosol optical properties in the southeastern United States in summer – Part 1: Hygroscopic growth , 2015 .

[27]  Darren L. Jackson,et al.  Predicting near-surface atmospheric variables from Special Sensor Microwave/Imager using neural networks with a first-guess approach , 2010 .

[28]  A. Ansmann,et al.  Aerosol-type-dependent lidar ratios observed with Raman lidar , 2007 .

[29]  Benjamin Thomas,et al.  Retrieving simulated volcanic, desert dust and sea-salt particle properties from two/three-component particle mixtures using UV-VIS polarization lidar and T matrix , 2013 .

[30]  Alexandra Tsekeri,et al.  Satellite retrieval of aerosol microphysical and optical parameters using neural networks: a new methodology applied to the Sahara desert dust peak , 2014 .

[31]  N. Mahowald,et al.  Aerosol Deposition Impacts on Land and Ocean Carbon Cycles , 2017, Current Climate Change Reports.

[32]  P. Koepke,et al.  Optical Properties of Aerosols and Clouds: The Software Package OPAC , 1998 .

[33]  V. Freudenthaler,et al.  EARLINET: towards an advanced sustainable European aerosol lidar network , 2014 .

[34]  R. Engelmann,et al.  North-south cross sections of the vertical aerosol distribution over the Atlantic Ocean from multiwavelength Raman/polarization lidar during Polarstern cruises , 2013, Journal of geophysical research. Atmospheres : JGR.

[35]  A. Ansmann,et al.  Microphysical particle parameters from extinction and backscatter lidar data by inversion with regularization: simulation. , 1999, Applied optics.

[36]  I. Riipinen,et al.  Particulate matter, air quality and climate: Lessons learned and future needs , 2015 .

[37]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..

[38]  L. Mona,et al.  Systematic lidar observations of Saharan dust over Europe in the frame of EARLINET (2000-2002) , 2008 .

[39]  M. Mishchenko,et al.  Reprint of: T-matrix computations of light scattering by nonspherical particles: a review , 1996 .

[40]  M. Perrone,et al.  Mediterranean aerosol typing by integrating three-wavelength lidar and sun photometer measurements , 2016, Environmental Science and Pollution Research.

[41]  O. Dubovik,et al.  Variability of aerosol and spectral lidar and backscatter and extinction ratios of key aerosol types derived from selected Aerosol Robotic Network locations , 2005 .

[42]  P. Waterman,et al.  SYMMETRY, UNITARITY, AND GEOMETRY IN ELECTROMAGNETIC SCATTERING. , 1971 .

[43]  D. Müller,et al.  Characterization of fresh and aged biomass burning events using multiwavelength Raman lidar and mass spectrometry , 2013 .

[44]  Iwona S. Stachlewska,et al.  Effect of Heat Wave Conditions on Aerosol Optical Properties Derived from Satellite and Ground-Based Remote Sensing over Poland , 2017, Remote. Sens..

[45]  R. Engelmann,et al.  Volcanic ash over Scandinavia originating from the Grímsvötn eruptions in May 2011 , 2012 .

[46]  Doina Nicolae,et al.  Experimental techniques for the calibration of lidar depolarization channels in EARLINET , 2017 .

[47]  Wei Gong,et al.  Measurement and Study of Lidar Ratio by Using a Raman Lidar in Central China , 2016, International journal of environmental research and public health.

[48]  Iwona S. Stachlewska,et al.  Temporal variations in optical and microphysical properties of mineral dust and biomass burning aerosol derived from daytime Raman lidar observations over Warsaw, Poland , 2017 .

[49]  G. Vaughan,et al.  Transport of Canadian forest fire smoke over the UK as observed by lidar , 2018, Atmospheric Chemistry and Physics.

[50]  R. Draxler,et al.  NOAA’s HYSPLIT Atmospheric Transport and Dispersion Modeling System , 2015 .

[51]  V. Freudenthaler,et al.  Characterization of Saharan dust, marine aerosols and mixtures of biomass-burning aerosols and dust by means of multi-wavelength depolarization and Raman lidar measurements during SAMUM 2 , 2011 .

[52]  M. Kahnert,et al.  Observations of the spectral dependence of linear particle depolarization ratio of aerosols using NASA Langley airborne High Spectral Resolution Lidar , 2015 .

[53]  D. Nicolae,et al.  AEROSOL CHARACTERIZATION BASED ON CHEMICAL COMPOSITION AND OPTICAL PROPERTIES , 2016 .

[54]  Albert Ansmann,et al.  Vertically resolved separation of dust and smoke over Cape Verde using multiwavelength Raman and polarization lidars during Saharan Mineral Dust Experiment 2008 , 2009 .

[55]  Frank Arnold,et al.  A Backward Modeling Study of Intercontinental Pollution Transport using Aircraft Measurements , 2003 .

[56]  Zhaoyan Liu,et al.  Extinction-to-backscatter ratio of Asian dust observed with high-spectral-resolution lidar and Raman lidar. , 2002, Applied optics.

[57]  P. D. Girolamo,et al.  APPLICATION OF RANDOMLY ORIENTED SPHEROIDS FORRETRIEVAL OF DUST PARTICLE PARAMETERS FROM MULTIWAVELENGTH LIDAR MEASUREMENTS , 2010 .

[58]  R. Engelmann,et al.  HETEAC: The Aerosol Classification Model for EarthCARE , 2016 .

[59]  Daniele Bortoli,et al.  February 2017 extreme Saharan dust outbreak in the Iberian Peninsula: from lidar-derived optical properties to evaluation of forecast models , 2018 .

[60]  C. Bretherton,et al.  Improving our fundamental understanding of the role of aerosol−cloud interactions in the climate system , 2016, Proceedings of the National Academy of Sciences.

[61]  B. Holben,et al.  An AERONET-based aerosol classification using the Mahalanobis distance , 2016 .

[62]  Detlef Müller,et al.  Multi-wavelength Raman lidar, sun photometric and aircraft measurements in combination with inversion models for the estimation of the aerosol optical and physico-chemical properties over Athens, Greece , 2012 .

[63]  V. Freudenthaler,et al.  Mineral dust observed with AERONET Sun photometer, Raman lidar, and in situ instruments during SAMUM 2006: Shape‐independent particle properties , 2010 .

[64]  P. Gupta,et al.  Particulate Matter Air Quality Assessment using Integrated Surface, Satellite, and Meteorological Products , 2009 .

[65]  S. Patade,et al.  Aerosol–Cloud Interaction in Deep Convective Clouds over the Indian Peninsula Using Spectral (Bin) Microphysics , 2017 .

[66]  L. Mona,et al.  CALIPSO climatological products: evaluation and suggestions from EARLINET , 2015 .

[67]  Albert Ansmann,et al.  European pollution outbreaks during ACE 2: Optical particle properties inferred from multiwavelength lidar and star-Sun photometry , 2002 .

[68]  V. Freudenthaler,et al.  Optical and microphysical properties of smoke over Cape Verde inferred from multiwavelength lidar measurements , 2011 .

[69]  K. Moorthy,et al.  Radiative effects of natural aerosols: A review , 2005 .

[70]  S. Trippetta,et al.  Fine aerosol particles (PM1): natural and anthropogenic contributions and health risk assessment , 2016, Air Quality, Atmosphere & Health.

[71]  H. Horvath,et al.  UV-VIS-NIR spectral optical properties of soot and soot-containing aerosols , 2003 .

[72]  Albert Ansmann,et al.  Vertical profiling of Saharan dust with Raman lidars and airborne HSRL in southern Morocco during SAMUM , 2009 .

[73]  Doina Nicolae,et al.  Assessment of aerosol's mass concentrations from measured linear particle depolarization ratio (vertically resolved) and simulations , 2013 .

[74]  Strengths and limitations of the NATALI code for aerosol typing from multiwavelength Raman lidar observations , 2018 .

[75]  M. Wendisch,et al.  Microphysical particle parameters from extinction and backscatter lidar data by inversion with regularization: experiment. , 2000, Applied optics.

[76]  V. Pont,et al.  Research on aerosol sources and chemical composition: Past, current and emerging issues , 2013 .

[77]  Alexander Smirnov,et al.  Multiangle Imaging SpectroRadiometer global aerosol product assessment by comparison with the Aerosol Robotic Network , 2010 .

[78]  Toshiyuki Murayama,et al.  Aerosol lidar ratio characteristics measured by a multi-wavelength Raman lidar system at Anmyeon Island, Korea , 2007 .

[79]  R. Engelmann,et al.  Aerosol profiling with lidar in the Amazon Basin during the wet and dry season , 2012 .

[80]  V. Freudenthaler,et al.  The 16 April 2010 major volcanic ash plume over central Europe: EARLINET lidar and AERONET photometer observations at Leipzig and Munich, Germany , 2010 .

[81]  Yuan Wang,et al.  Review of Aerosol–Cloud Interactions: Mechanisms, Significance, and Challenges , 2016 .

[82]  U. Wandinger,et al.  Profiling of aerosol microphysical properties at several EARLINET/AERONET sites during the July 2012 ChArMEx/EMEP campaign , 2015 .

[83]  L. Mona,et al.  Lidar Measurements for Desert Dust Characterization: An Overview , 2012 .

[84]  R. Reynolds,et al.  The NCEP/NCAR 40-Year Reanalysis Project , 1996, Renewable Energy.

[85]  T. Eck,et al.  An analysis of AERONET aerosol absorption properties and classifications representative of aerosol source regions , 2012 .

[86]  S. Kreidenweis,et al.  The Microphysical Roles of Lower-Tropospheric versus Midtropospheric Aerosol Particles in Mature-Stage MCS Precipitation , 2017 .

[87]  Josef Gasteiger,et al.  Modelling lidar-relevant optical properties of complex mineral dust aerosols , 2011 .

[88]  D. Nicolae,et al.  DETECTION OF LOCAL WEATHER EVENTS FROM MULTIWAVELENGTH LIDAR MEASUREMENTS DURING THE EARLI 09 CAMPAIGN , 2011 .

[89]  Retrieval of the boundary layer height from active and passive remote sensors. Comparison with a NWP model , 2014, Acta Geophysica.

[90]  Ernest Weingartner,et al.  Effects of relative humidity on aerosol light scattering: results from different European sites , 2012 .

[91]  L. Alados-Arboledas,et al.  Optical and microphysical properties of fresh biomass burning aerosol retrieved by Raman lidar, and star‐and sun‐photometry , 2011 .

[92]  James R. Drummond,et al.  Depolarization calibration and measurements using the CANDAC Rayleigh-Mie-Raman lidar at Eureka, Canada , 2017 .

[93]  T. Eck,et al.  Variability of Absorption and Optical Properties of Key Aerosol Types Observed in Worldwide Locations , 2002 .

[94]  J. Jimenez,et al.  Aerosol optical properties in the southeastern United States in summer – Part 2: Sensitivity of aerosol optical depth to relative humidity and aerosol parameters , 2015 .

[95]  L. Alados-Arboledas,et al.  Monitoring of the Eyjafjallajökull volcanic aerosol plume over the Iberian Peninsula by means of four EARLINET lidar stations , 2011 .

[96]  Josef Gasteiger,et al.  Volcanic ash from Iceland over Munich: mass concentration retrieved from ground-based remote sensing measurements , 2010 .

[97]  A. Stohl,et al.  Raman lidar observations of aged Siberian and Canadian forest fire smoke in the free troposphere over Germany in 2003 : Microphysical particle characterization , 2005 .

[98]  V. Freudenthaler,et al.  EARLINET correlative measurements for CALIPSO: First intercomparison results , 2010 .

[99]  C. Zerefos,et al.  Smoke injection heights from agricultural burning in Eastern Europe as seen by CALIPSO , 2010 .

[100]  D. Winker,et al.  The CALIPSO Automated Aerosol Classification and Lidar Ratio Selection Algorithm , 2009 .

[101]  M. Wendisch,et al.  Optical and microphysical characterization of biomass‐ burning and industrial‐pollution aerosols from‐ multiwavelength lidar and aircraft measurements , 2002 .

[102]  C. Zerefos,et al.  Optical properties of different aerosol types: seven years of combined Raman-elastic backscatter lidar measurements in Thessaloniki, Greece , 2009 .

[103]  M. Vaughan,et al.  Aerosol classification from airborne HSRL and comparisons with the CALIPSO vertical feature mask , 2013 .

[104]  Barbara J. Gaitley,et al.  An analysis of global aerosol type as retrieved by MISR , 2015 .

[105]  L. Mona,et al.  Multi year sun-photometer measurements for aerosol characterization in a Central Mediterranean site , 2012 .

[106]  Wayne C. Welch,et al.  Airborne high spectral resolution lidar for profiling aerosol optical properties. , 2008, Applied optics.