STAS: Adaptive Selecting Spatio-Temporal Deep Features for Improving Bias Correction on Precipitation

Numerical Weather Prediction (NWP) can reduce human suffering by predicting disastrous precipitation in time. A commonly-used NWP in the world is the European Centre for medium-range weather forecasts (EC). However, it is necessary to correct EC forecast through Bias Correcting on Precipitation (BCoP) since we still have not fully understood the mechanism of precipitation, making EC often have some biases. The existing BCoPs suffers from limited prior data and the fixed Spatio-Temporal (ST) scale. We thus propose an end-to-end deep-learning BCoP model named Spatio-Temporal feature Auto-Selective (STAS) model to select optimal ST regularity from EC via the ST Feature-selective Mechanisms (SFM/TFM). Given different input features, these two mechanisms can automatically adjust the spatial and temporal scales for correcting. Experiments on an EC public dataset indicate that compared with 8 published BCoP methods, STAS shows state-of-the-art performance on several criteria of BCoP, named threat scores (TS). Further, ablation studies justify that the SFM/TFM indeed work well in boosting the performance of BCoP, especially on the heavy precipitation.

[1]  Crayton J. Yapp,et al.  A model for the relationships between precipitation D/H ratios and precipitation intensity , 1982 .

[2]  F. Mesinger Bias Adjusted Precipitation Threat Scores , 2008 .

[3]  Luca Delle Monache,et al.  A Kalman-filter bias correction method applied to deterministic, ensemble averaged and probabilistic forecasts of surface ozone , 2008 .

[4]  Thomas M. Hamill,et al.  Verification of TIGGE Multimodel and ECMWF Reforecast-Calibrated Probabilistic Precipitation Forecasts over the Contiguous United States* , 2012 .

[5]  D. Stephenson Water Resources Management , 2003 .

[6]  Yi Li,et al.  Deformable Convolutional Networks , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[7]  Bin Wang,et al.  Conditional nonlinear optimal perturbation and its applications , 2003 .

[8]  R. D. Ward MONTHLY WEATHER REVIEW. , 1907, Science.

[9]  Juan Song,et al.  Learning Spatiotemporal Features Using 3DCNN and Convolutional LSTM for Gesture Recognition , 2017, 2017 IEEE International Conference on Computer Vision Workshops (ICCVW).

[10]  Xun Xu,et al.  C3AE: Exploring the Limits of Compact Model for Age Estimation , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  James B. Elsner,et al.  Weather and Forecasting , 2007 .

[12]  Yiqun Liu,et al.  Towards a Precipitation Bias Corrector against Noise and Maldistribution , 2019, ArXiv.

[13]  F. Pellicciotti,et al.  Advances in Water Resources , 2017 .

[14]  James Zijun Wang,et al.  Facial Aging and Rejuvenation by Conditional Multi-Adversarial Autoencoder with Ordinal Regression , 2018, ArXiv.

[15]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[16]  Witold F. Krajewski,et al.  Analysis and modeling of spatial correlation structure in small-scale rainfall in Central Oklahoma , 2006 .

[17]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Liqun Hu,et al.  Kinetic and thermodynamic modeling of Portland cement hydration at low temperatures , 2017, Chemical Papers.

[19]  Renate Hagedorn,et al.  Probabilistic Forecast Calibration Using ECMWF and GFS Ensemble Reforecasts. Part II: Precipitation , 2008 .

[20]  George P. Petropoulos,et al.  WRF Dynamical Downscaling and Bias Correction Schemes for NCEP Estimated Hydro-Meteorological Variables , 2015, Water Resources Management.

[21]  Soroosh Sorooshian,et al.  Calibration of Probabilistic Quantitative Precipitation Forecasts with an Artificial Neural Network , 2007 .

[22]  Bellie Sivakumar,et al.  Evaluation of Quantitative Precipitation Predictions by ECMWF, CMA, and UKMO for Flood Forecasting: Application to Two Basins in China , 2018 .

[23]  C. Willmott,et al.  Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance , 2005 .