Modeling and Forecasting the GPS Zenith Troposphere Delay in West Antarctica Based on Different Blind Source Separation Methods and Deep Learning

Tropospheric delay is an important error source in global positioning systems (GPS), and the water vapor retrieved from the tropospheric delay is widely used in meteorological research such as climate analysis and weather forecasting. Most zenith tropospheric delay (ZTD) models are presently used as positioning corrections, and few models are used for the estimation of water vapor, especially in Antarctica. Through two blind source separation algorithms (principal component analysis (PCA) and independent component analysis (ICA)), a back-propagation (BP) neural network and a deep learning technique (long short-term memory (LSTM) network), we establish an hourly high-accuracy ZTD model for GPS meteorology using the GPS-ZTD from 52 GPS stations in West Antarctica. Our results show that under the condition in which the principal components (PCs) and independent components (ICs) remain fixed after decomposition, the mean accuracy of the models for West Antarctica using PCA or ICA are better than 10 mm. Compared with the ZTDs from the nonmodeling stations, the mean root mean square (RMS) of the PCA and ICA models are 9.3 and 8.9 mm, respectively, and the correlation coefficients between the GPS-ZTD and model-ZTDs all exceed 90%. The accuracy of the ICA model is slightly higher than that of the PCA model, and the ICs of the ICA model show more consistent spatial responses. The six-hour forecast is the best among the forecast results, with a mean correlation coefficient of 90.6% and a mean RMS of 7.2 mm using GPS-ZTD. The long-term forecast result is significantly inaccurate, as the correlation coefficient between the 24-h forecast and GPS-ZTD is only 63.2%. Generally modest results have been achieved (HSS ≤ 0.38). Furthermore, the forecast accuracy in coastal areas is lower than that in inland areas. Our study confirms that the combined use of ICA and deep learning in ZTD modeling can effectively restore the original signals, and short-term forecasting can be effectively used in GPS meteorology. However, further development of the technology is necessary.

[1]  I. Shapiro,et al.  Geodesy by radio interferometry: Effects of atmospheric modeling errors on estimates of baseline length , 1985 .

[2]  Heng Tao Shen,et al.  Principal Component Analysis , 2009, Encyclopedia of Biometrics.

[3]  T. Nilsson,et al.  GPT2: Empirical slant delay model for radio space geodetic techniques , 2013, Geophysical research letters.

[4]  F. Alshawaf,et al.  On the Statistical Significance of Climatic Trends Estimated From GPS Tropospheric Time Series , 2018, Journal of Geophysical Research: Atmospheres.

[5]  Wei Li,et al.  A new global zenith tropospheric delay model IGGtrop for GNSS applications , 2012 .

[6]  R. Langley,et al.  UNB Neutral Atmosphere Models : Development and Performance , 2006 .

[7]  Hermann Ney,et al.  LSTM Neural Networks for Language Modeling , 2012, INTERSPEECH.

[8]  Robert Weber,et al.  Development of an improved empirical model for slant delays in the troposphere (GPT2w) , 2015, GPS Solutions.

[9]  Fei Li,et al.  Spatiotemporal Filtering and Noise Analysis for Regional GNSS Network in Antarctica Using Independent Component Analysis , 2019, Remote. Sens..

[10]  R. Langley,et al.  Limiting Factors in Tropospheric Propagation Delay Error Modelling for GPS Airborne Navigation , 1996 .

[11]  Zhilu Wu,et al.  A simplified GNSS tropospheric delay model based on the nonlinear hypothesis , 2017, GPS Solutions.

[12]  Witold Rohm,et al.  Ground-based GNSS ZTD/IWV estimation system for numerical weather prediction in challenging weather conditions , 2014 .

[13]  Z. Jing.  Establishment of a new global model for zenith tropospheric delay based on functional expression for its vertical profile , 2014 .

[14]  G. Blewitt,et al.  Harnessing the GPS Data Explosion for Interdisciplinary Science , 2018, Eos.

[15]  T. Schneider.  Analysis of Incomplete Climate Data: Estimation of Mean Values and Covariance Matrices and Imputation of Missing Values. , 2001 .

[16]  Yuei-An Liou,et al.  New versions of the BDS/GNSS zenith tropospheric delay model IGGtrop , 2014, Journal of Geodesy.

[17]  Wayan Suparta,et al.  Modeling of zenith path delay over Antarctica using an adaptive neuro fuzzy inference system technique , 2015, Expert Syst. Appl..

[18]  Shifei Ding,et al.  An optimizing BP neural network algorithm based on genetic algorithm , 2011, Artificial Intelligence Review.

[19]  Yunbin Yuan,et al.  IGGtrop_SH and IGGtrop_rH: Two Improved Empirical Tropospheric Delay Models Based on Vertical Reduction Functions , 2018, IEEE Transactions on Geoscience and Remote Sensing.

[20]  Johannes Böhm,et al.  VMF3/GPT3: refined discrete and empirical troposphere mapping functions , 2017, Journal of Geodesy.

[21]  Jun Yan,et al.  Separation of Sources of Seasonal Uplift in China Using Independent Component Analysis of GNSS Time Series , 2019, Journal of Geophysical Research: Solid Earth.

[22]  Enrico Serpelloni,et al.  Blind source separation problem in GPS time series , 2016, Journal of Geodesy.

[23]  T. Herring,et al.  GPS Meteorology: Remote Sensing of Atmospheric Water Vapor Using the Global Positioning System , 1992 .

[24]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[25]  Richard B. Langley,et al.  UNB3m_pack: a neutral atmosphere delay package for radiometric space techniques , 2008 .

[26]  Otto Hyvärinen,et al.  A Probabilistic Derivation of Heidke Skill Score , 2014 .

[27]  Yuanxi Yang,et al.  Spatiotemporal filtering for regional GPS network in China using independent component analysis , 2017, Journal of Geodesy.

[28]  AbdiHervé,et al.  Principal Component Analysis , 2010, Essentials of Pattern Recognition.

[29]  B. Liu,et al.  Regional spatio-temporal zenith tropospheric delay modelling using independent component analysis , 2018, Survey Review.

[30]  Yibin Yao,et al.  An ERA5‐Based Model for Estimating Tropospheric Delay and Weighted Mean Temperature Over China With Improved Spatiotemporal Resolutions , 2019, Earth and Space Science.

[31]  Yehuda Bock,et al.  Spatiotemporal filtering using principal component analysis and Karhunen-Loeve expansion approaches for regional GPS network analysis , 2006 .

[32]  David Bekaert,et al.  Tracking the weight of Hurricane Harvey’s stormwater using GPS data , 2018, Science Advances.

[33]  D. Wiens,et al.  POLENET Seismic and GPS Network in West Antarctica , 2008 .

[34]  Yibin Yao,et al.  ITG: A New Global GNSS Tropospheric Correction Model , 2015, Scientific Reports.

[35]  Pedro M. Valero-Mora,et al.  Determining the Number of Factors to Retain in EFA: An easy-to-use computer program for carrying out Parallel Analysis , 2007 .

[36]  H. S. Hopfield.  Two- quartic tropospheric refractivity profile for correcting satellite data , 1969 .

[37]  Yibin Yao,et al.  An improved global zenith tropospheric delay model GZTD2 considering diurnal variations , 2016 .

[38]  Yao Yi.  A new global zenith tropospheric delay model GZTD , 2013 .

[39]  E. Wei,et al.  Global model of zenith tropospheric delay proposed based on EOF analysis , 2017 .

[40]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[41]  J. Saastamoinen.  Atmospheric Correction for the Troposphere and Stratosphere in Radio Ranging Satellites , 2013 .

[42]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..