Icing Detection over East Asia from Geostationary Satellite Data Using Machine Learning Approaches

Even though deicing or airframe coating technologies continue to develop, aircraft icing is still one of the critical threats to aviation. While the detection of potential icing clouds has been conducted using geostationary satellite data in the US and Europe, there is not yet a robust model that detects potential icing areas in East Asia. In this study, we proposed machine-learning-based icing detection models using data from two geostationary satellites—the Communication, Ocean, and Meteorological Satellite (COMS) Meteorological Imager (MI) and the Himawari-8 Advanced Himawari Imager (AHI)—over Northeast Asia. Two machine learning techniques—random forest (RF) and multinomial log-linear (MLL) models—were evaluated with quality-controlled pilot reports (PIREPs) as the reference data. The machine-learning-based models were compared to the existing models through five-fold cross-validation. The RF model for COMS MI produced the best performance, resulting in a mean probability of detection (POD) of 81.8%, a mean overall accuracy (OA) of 82.1%, and mean true skill statistics (TSS) of 64.0%. One of the existing models, flight icing threat (FIT), produced relatively poor performance, providing a mean POD of 36.4%, a mean OA of 61.0, and a mean TSS of 9.7%. The Himawari-8 based models also produced performance comparable to the COMS models. However, it should be noted that very limited PIREP reference data were available especially for the Himawari-8 models, which requires further evaluation in the future with more reference data. The spatio-temporal patterns of the icing areas detected using the developed models were also visually examined using time-series satellite data.

[1]  W. Paul Menzel,et al.  CLOUD TOP PROPERTIES AND CLOUD PHASE ALGORITHM THEORETICAL BASIS DOCUMENT , 2002 .

[2]  Jungho Im,et al.  Estimation of daily maximum and minimum air temperatures in urban landscapes using MODIS time series satellite data , 2018 .

[3]  Barry E. Schwartz,et al.  An Hourly Assimilation–Forecast Cycle: The RUC , 2004 .

[4]  Misako Kachi,et al.  Sea surface temperature from the new Japanese geostationary meteorological Himawari‐8 satellite , 2016 .

[5]  Brian Brisco,et al.  Wetland classification in Newfoundland and Labrador using multi-source SAR and optical data integration , 2017 .

[6]  J. Im,et al.  Machine learning approaches for forest classification and change analysis using multi-temporal Landsat TM images over Huntington Wildlife Forest , 2013 .

[7]  Chang-Hoi Ho,et al.  Space observations of cold-cloud phase change , 2010, Proceedings of the National Academy of Sciences.

[8]  R. A. Roebeling,et al.  Evaluation of Cloud-Phase Retrieval Methods for SEVIRI on Meteosat-8 Using Ground-Based Lidar and Cloud Radar Data , 2008 .

[9]  J. Im,et al.  Detection of tropical cyclone genesis via quantitative satellite ocean surface wind pattern and intensity analyses using decision trees , 2016 .

[10]  Gilles Louppe,et al.  Understanding variable importances in forests of randomized trees , 2013, NIPS.

[11]  J. R. Jensen,et al.  Delineation of climate regions using in-situ and remotely-sensed data for the Carolinas , 2008 .

[12]  Barry E. Schwartz,et al.  The Quantitative Use of PIREPs in Developing Aviation Weather Guidance Products , 1996 .

[13]  John R. Jensen,et al.  Vegetation Cover Analysis of Hazardous Waste Sites in Utah and Arizona Using Hyperspectral Remote Sensing , 2012, Remote. Sens..

[14]  Patrick Minnis,et al.  Near-Real Time Satellite-Retrieved Cloud and Surface Properties for Weather and Aviation Safety Applications , 2014 .

[15]  Jia Zong,et al.  Algorithm Theoretical Basis , 1999 .

[16]  Jungho Im,et al.  Detection of Convective Initiation Using Meteorological Imager Onboard Communication, Ocean, and Meteorological Satellite Based on Machine Learning Approaches , 2015, Remote. Sens..

[17]  Roy Rasmussen,et al.  A Numerical Weather Model’s Ability to Predict Characteristics of Aircraft Icing Environments , 2017 .

[18]  Tim Appelhans,et al.  Improving the accuracy of rainfall rates from optical satellite sensors with machine learning — A random forests-based approach applied to MSG SEVIRI , 2014 .

[19]  J V Tu,et al.  Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. , 1996, Journal of clinical epidemiology.

[20]  Michael E. Hodgson,et al.  Building type classification using spatial and landscape attributes derived from LiDAR remote sensing data , 2014 .

[21]  Scott A. Shappell,et al.  Developing Proactive Methods for General Aviation Data Collection , 2010 .

[22]  Gregory Thompson,et al.  Using Satellite Data to Reduce Spatial Extent of Diagnosed Icing , 1997 .

[23]  Koreen Millard,et al.  On the Importance of Training Data Sample Selection in Random Forest Image Classification: A Case Study in Peatland Ecosystem Mapping , 2015, Remote. Sens..

[24]  B. Brown,et al.  Intercomparison of In-Flight Icing Algorithms. Part I: WISP94 Real-Time Icing Prediction and Evaluation Program , 1997 .

[25]  Jean-Marie Carrière,et al.  A climatological study of surface freezing precipitation in Europe , 2000 .

[26]  Ben C. Bernstein,et al.  An Inferred Climatology of Icing Conditions Aloft, Including Supercooled Large Drops. Part II: Europe, Asia, and the Globe , 2009 .

[27]  M. Politovich.  Aircraft Icing Caused by Large Supercooled Droplets , 1989 .

[28]  J. R. Koehler,et al.  Modern Applied Statistics with S-Plus. , 1996 .

[29]  R. J. Kind,et al.  Experimental and computational simulation of in-flight icing phenomena , 1998 .

[30]  David Serke,et al.  Use of X-band radars to support the detection of in-flight icing hazards , 2009 .

[31]  Robert M. Rauber,et al.  An Explanation for the Existence of Supercooled Water at the Top of Cold Clouds , 1991 .

[32]  Duk-jin Kim,et al.  Retrieval of Melt Ponds on Arctic Multiyear Sea Ice in Summer from TerraSAR-X Dual-Polarization Data Using Machine Learning Approaches: A Case Study in the Chukchi Sea with Mid-Incidence Angle Data , 2016, Remote. Sens..

[33]  Jungho Im,et al.  Estimation of Fugacity of Carbon Dioxide in the East Sea Using In Situ Measurements and Geostationary Ocean Color Imager Satellite Data , 2017, Remote. Sens..

[34]  Jungho Im,et al.  Detection of deterministic and probabilistic convection initiation using Himawari-8 Advanced Himawari Imager data , 2016 .

[35]  Patrick Minnis,et al.  Real-time cloud, radiation, and aircraft icing parameters from GOES over the USA , 2004 .

[36]  Jungho Im,et al.  Detection of Tropical Overshooting Cloud Tops Using Himawari-8 Imagery , 2017, Remote. Sens..

[37]  Dimos Poulikakos,et al.  Mechanism of supercooled droplet freezing on surfaces , 2012, Nature Communications.

[38]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[39]  Timothy J. Schmit,et al.  Advanced infrared sounder subpixel cloud detection with imagers and its impact on radiance assimilation in NWP , 2014 .

[40]  Patrick Minnis,et al.  Determining the Flight Icing Threat to Aircraft with Single-Layer Cloud Parameters Derived from Operational Satellite Data , 2012 .

[41]  M. Ahn,et al.  An exploratory study of cloud remote sensing capabilities of the Communication, Ocean and Meteorological Satellite (COMS) imagery , 2007 .

[42]  J. Im,et al.  Drought assessment and monitoring through blending of multi-sensor indices using machine learning approaches for different climate regions , 2016 .

[43]  Barbara G. Brown,et al.  Current Icing Potential: Algorithm Description and Comparison with Aircraft Observations , 2005 .

[44]  Kohei Arai,et al.  Aerosol data assimilation using data from Himawari‐8, a next‐generation geostationary meteorological satellite , 2016 .

[45]  J. Im,et al.  Meteorological drought forecasting for ungauged areas based on machine learning: Using long-range climate forecast and remote sensing data , 2017 .

[46]  S. Ha,et al.  Machine learning approaches to coastal water quality monitoring using GOCI satellite data , 2014 .

[47]  Yong-Sang Choi,et al.  Algorithm Theoretical Basis Document For Cloud Optical Thickness , 2014 .

[48]  Cheng Da,et al.  Preliminary assessment of the Advanced Himawari Imager (AHI) measurement onboard Himawari-8 geostationary satellite , 2015 .

[49]  Jungho Im,et al.  A Volumetric Approach to Population Estimation Using Lidar Remote Sensing , 2011 .

[50]  Lauchlan H. Fraser,et al.  A comparison of geographic datasets and field measurements to model soil carbon using random forests and stepwise regressions (British Columbia, Canada) , 2017 .

[51]  Shihong Du,et al.  Mining parameter information for building extraction and change detection with very high-resolution imagery and GIS data , 2017 .

[52]  Kenneth Benoit,et al.  Linear Regression Models with Logarithmic Transformations , 2011 .

[53]  Gerald Forkuor,et al.  Landsat-8 vs. Sentinel-2: examining the added value of sentinel-2’s red-edge bands to land-use and land-cover mapping in Burkina Faso , 2018 .

[54]  Jungho Im,et al.  Landfast sea ice monitoring using multisensor fusion in the Antarctic , 2015 .

[55]  Amir AghaKouchak,et al.  Extended contingency table: Performance metrics for satellite observations and climate model simulations , 2013 .

[56]  Barbara G. Brown,et al.  Intercomparison of In-Flight Icing Algorithms. Part II: Statistical Verification Results , 1997 .

[57]  W. Paul Menzel,et al.  INTRODUCING THE NEXT-GENERATION ADVANCED BASELINE IMAGER ON GOES-R , 2005 .

[58]  Jungho Im,et al.  Forest Biomass and Carbon Stock Quantification Using Airborne LiDAR Data: A Case Study Over Huntington Wildlife Forest in the Adirondack Park , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[59]  George A. Isaac,et al.  Preliminary results from the Alliance Icing Research Study (AIRS) , 2001 .

[60]  Steven Platnick,et al.  Polarized view of supercooled liquid water clouds , 2016 .