Use of neural networks in ground-based aerosol retrievals from multi-angle spectropolarimetric observations

In this paper, the use of a neural network algorithm for the retrieval of the aerosol properties from ground-based spectropolarimetric measurements is discussed. The neural network is able to retrieve the aerosol properties with an accuracy that is almost comparable to that of an iterative retrieval. By using the outcome of the neural network as first guess in the iterative retrieval scheme, the accuracy of the retrieved fine- and coarse-mode optical thickness is further improved, while for the other parameters the improvement is small or absent. The resulting scheme (neural network + iterative retrieval) is compared to the original one (look-up table + iterative retrieval) on a set of simulated ground-based measurements, and on a small set of real observations carried out by an accurate ground-based spectropolarimeter. The results show that the use of a neural-network-based first guess leads to an increase in the number of converging retrievals, and possibly to more accurate estimates of the aerosol effective radius and complex refractive index.

[1]  J. Fischer,et al.  Effect of aerosol microphysical properties on polarization of skylight: sensitivity study and measurements. , 2006, Applied optics.

[2]  Pavel Litvinov,et al.  Aerosol properties over the ocean from PARASOL multiangle photopolarimetric measurements , 2011 .

[3]  P. Werbos,et al.  Beyond Regression : "New Tools for Prediction and Analysis in the Behavioral Sciences , 1974 .

[4]  Helmut Schiller.  Model Inversion by Parameter Fit Using NN Emulating the Forward Model - Evaluation of Indirect Measurements , 2006, IJCNN.

[5]  Zoran Obradovic,et al.  A statistical complement to deterministic algorithms for the retrieval of aerosol optical thickness from radiance data , 2006, Eng. Appl. Artif. Intell..

[6]  M. Chin,et al.  A review of measurement-based assessments of the aerosol direct radiative effect and forcing , 2005 .

[7]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[8]  Didier Tanré,et al.  Estimate of the aerosol properties over the ocean with POLDER , 2000 .

[9]  J. Thundiyil,et al.  Clearing the Air: A Review of the Effects of Particulate Matter Air Pollution on Human Health , 2011, Journal of Medical Toxicology.

[10]  William L. Smith,et al.  A principal component noise filter for high spectral resolution infrared measurements , 2004 .

[11]  Tony R. Martinez,et al.  The Inefficiency of Batch Training for Large Training Sets , 2000, IJCNN.

[12]  J. Hansen,et al.  Light scattering in planetary atmospheres , 1974 .

[13]  Sylvie Thiria,et al.  Use of a neuro-variational inversion for retrieving oceanic and atmospheric constituents from satellite ocean colour sensor: Application to absorbing aerosols , 2006, Neural Networks.

[14]  Shun-ichi Amari,et al.  Annealed online learning in multilayer neural networks , 1999 .

[15]  POTENTIAL RETRIEVAL OF AEROSOL PROPERTIES COMBINING THE MULTISPECTRAL , MULTIANGLE RESEARCH SCANNING POLARIMETER ( RSP ) MEASUREMENTS OF THE INTENSITY AND LINEAR POLARIZATION OF LIGHT , 2011 .

[16]  T. Eck,et al.  Spectral discrimination of coarse and fine mode optical depth , 2003 .

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

[18]  Shun-ichi Amari,et al.  On-Line Learning in Neural Networks: Annealed On-line Learning in Multilayer Neural Networks , 1999 .

[19]  O. P. Hasekamp,et al.  A linearized vector radiative transfer model for atmospheric trace gas retrieval , 2002 .

[20]  Jean-François Léon,et al.  Application of spheroid models to account for aerosol particle nonsphericity in remote sensing of desert dust , 2006 .

[21]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[22]  Petri Koistinen,et al.  Using additive noise in back-propagation training , 1992, IEEE Trans. Neural Networks.

[23]  Fouad Badran,et al.  Retrieval of aerosol type and optical thickness over the Mediterranean from SeaWiFS images using an automatic neural classification method , 2006 .

[24]  S. Thiria,et al.  Retrieving aerosol characteristics and sea-surface chlorophyll from satellite ocean color multi-spectral sensors using a neural-variational method , 2013 .

[25]  Martin Fodslette Meiller.  A Scaled Conjugate Gradient Algorithm for Fast Supervised Learning , 1993 .

[26]  Zoran Obradovic,et al.  Uncertainty Analysis of Neural-Network-Based Aerosol Retrieval , 2012, IEEE Transactions on Geoscience and Remote Sensing.

[27]  Christopher M. Bishop,et al.  Current address: Microsoft Research, , 2022 .

[28]  M. Mishchenko,et al.  Satellite retrieval of aerosol properties over the ocean using polarization as well as intensity of reflected sunlight , 1997 .

[29]  Zoran Obradovic,et al.  A Data-Mining Technique for Aerosol Retrieval Across Multiple Accuracy Measures , 2010, IEEE Geoscience and Remote Sensing Letters.

[30]  Michael D. King,et al.  A flexible inversion algorithm for retrieval of aerosol optical properties from Sun and sky radiance measurements , 2000 .

[31]  M. Mishchenko,et al.  Remote sensing of terrestrial tropospheric aerosols from aircraft and satellites , 2005, Journal of Physics: Conference Series.

[32]  U. Lohmann,et al.  Global indirect aerosol effects: a review , 2004 .

[33]  Didier Tanré,et al.  Statistically optimized inversion algorithm for enhanced retrieval of aerosol properties from spectral multi-angle polarimetric satellite observations , 2010 .

[34]  F. Maignan,et al.  Remote sensing of aerosols over land surfaces from POLDER‐ADEOS‐1 polarized measurements , 2001 .

[35]  Rich Caruana,et al.  Overfitting in Neural Nets: Backpropagation, Conjugate Gradient, and Early Stopping , 2000, NIPS.

[36]  Salvatore Stramondo,et al.  Volcanic ash detection and retrievals using MODIS data by means of neural networks , 2011 .

[37]  C. Rodgers.  Characterization and Error Analysis of Profiles Retrieved From Remote Sounding Measurements , 1990 .

[38]  W. Collins,et al.  Geoscientific Model Development Toward a minimal representation of aerosols in climate models : description and evaluation in the Community Atmosphere Model CAM 5 , 2012 .

[39]  Dianne P. O'Leary,et al.  The Use of the L-Curve in the Regularization of Discrete Ill-Posed Problems , 1993, SIAM J. Sci. Comput..

[40]  Beat Schmid,et al.  Polarimetric remote sensing of aerosols over land , 2009 .

[41]  S. Thiria,et al.  Use of a Neurovariational Inversion for Retrieving Oceanic and Atmospheric Constituents from Ocean Color Imagery: A Feasibility Study , 2005 .

[42]  Christoph U. Keller,et al.  Atmospheric aerosol characterization with a ground-based SPEX spectropolarimetric instrument , 2014 .

[43]  Sylvie Thiria,et al.  Atmospheric correction of MERIS data for case-2 waters using a neuro-variational inversion , 2012 .

[44]  Otto P. Hasekamp,et al.  Linearization of vector radiative transfer with respect to aerosol properties and its use in satellite remote sensing , 2005 .

[45]  S. Howell,et al.  Combined Retrievals of Boreal Forest Fire Aerosol Properties with a Polarimeter and Lidar , 2011 .

[46]  Jochen Landgraf,et al.  Retrieval of aerosol properties over land surfaces: capabilities of multiple-viewing-angle intensity and polarization measurements. , 2007, Applied optics.

[47]  Clive D Rodgers,et al.  Inverse Methods for Atmospheric Sounding: Theory and Practice , 2000 .

[48]  David Alexander,et al.  Volcanic ash in the atmosphere and risks for civil aviation: A study in European crisis management , 2013, International Journal of Disaster Risk Science.

[49]  C. Keller,et al.  Spectral modulation for full linear polarimetry. , 2009, Applied optics.

[50]  T. Eck,et al.  Accuracy assessments of aerosol optical properties retrieved from Aerosol Robotic Network (AERONET) Sun and sky radiance measurements , 2000 .

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

[52]  J. Hansen,et al.  Radiative forcing and climate response , 1997 .

[53]  C. Lee Giles,et al.  Overfitting and neural networks: conjugate gradient and backpropagation , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[54]  S. Roberts,et al.  Confidence Intervals and Prediction Intervals for Feed-Forward Neural Networks , 2001 .

[55]  Eric E. Smith,et al.  Uncertainty analysis , 2001 .

[56]  Helmut Schiller.  Model Inversion by Parameter Fit Using NN Emulating the Forward Model – Evaluation of Indirect Measurements , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.

[57]  Ah Chung Tsoi,et al.  Lessons in Neural Network Training: Overfitting May be Harder than Expected , 1997, AAAI/IAAI.

[58]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[59]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[60]  S. Sherwood,et al.  Aerosol Cloud-Mediated Radiative Forcing: Highly Uncertain and Opposite Effects from Shallow and Deep Clouds , 2013 .

[61]  Yoram J. Kaufman,et al.  Monitoring of aerosol forcing of climate from space: analysis of measurement requirements , 2004, Journal of Quantitative Spectroscopy and Radiative Transfer.

[62]  J. Nazuno.  Haykin, Simon. Neural networks: A comprehensive foundation, Prentice Hall, Inc. Segunda Edición, 1999 , 2000 .

[63]  Filipe Aires,et al.  Remote sensing from the infrared atmospheric sounding interferometer instrument 1. Compression, denoising, and first-guess retrieval algorithms , 2002 .

[64]  Ilan Koren,et al.  Smoke and Pollution Aerosol Effect on Cloud Cover , 2006, Science.