A random forest algorithm for nowcasting of intense precipitation events

Abstract Automatic nowcasting of convective initiation and thunderstorms has potential applications in several sectors including aviation planning and disaster management. In this paper, random forest based machine learning algorithm is tested for nowcasting of convective rain with a ground based radiometer. Brightness temperatures measured at 14 frequencies (7 frequencies in 22–31 GHz band and 7 frequencies in 51–58 GHz bands) are utilized as the inputs of the model. The lower frequency band is associated to the water vapor absorption whereas the upper frequency band relates to the oxygen absorption and hence, provide information on the temperature and humidity of the atmosphere. Synthetic minority over-sampling technique is used to balance the data set and 10-fold cross validation is used to assess the performance of the model. Results indicate that random forest algorithm with fixed alarm generation time of 30 min and 60 min performs quite well (probability of detection of all types of weather condition ∼90%) with low false alarms. It is, however, also observed that reducing the alarm generation time improves the threat score significantly and also decreases false alarms. The proposed model is found to be very sensitive to the boundary layer instability as indicated by the variable importance measure. The study shows the suitability of a random forest algorithm for nowcasting application utilizing a large number of input parameters from diverse sources and can be utilized in other forecasting problems.

[1]  G S Pankiewicz,et al.  Pattern recognition techniques for the identification of cloud and cloud systems , 2007 .

[2]  Juanzhen Sun,et al.  Nowcasting Thunderstorms: A Status Report , 1998 .

[3]  Mukand S. Babel,et al.  Using the mutual information technique to select explanatory variables in artificial neural networks for rainfall forecasting , 2015 .

[4]  Matthias Steiner,et al.  A machine-learning approach to finding weather regimes and skillful predictor combinations for short-term storm forecasting [presentation] , 2008 .

[5]  Lindsay J. Bennett,et al.  A review of the initiation of precipitating convection in the United Kingdom , 2006 .

[6]  Chih-Chiang Wei,et al.  Soft computing techniques in ensemble precipitation nowcast , 2013, Appl. Soft Comput..

[7]  Frank S. Marzano,et al.  Temperature and humidity profile retrievals from ground-based microwave radiometers during TUC , 2006 .

[8]  Y. Zhao,et al.  Comparison of decision tree methods for finding active objects , 2007, 0708.4274.

[9]  Somia A. Asklany,et al.  Rainfall events prediction using rule-based fuzzy inference system , 2011 .

[10]  A. K. Showalter,et al.  A stability index for thunderstorm forecasting , 1953 .

[11]  Nitesh V. Chawla,et al.  SMOTE: Synthetic Minority Over-sampling Technique , 2002, J. Artif. Intell. Res..

[12]  F. Marzano,et al.  Artificial neural-network technique for precipitation nowcasting from satellite imagery , 2006 .

[13]  A. Maitra,et al.  Prediction of convective events using multi-frequency radiometric observations at Kolkata , 2016 .

[14]  John K. Williams,et al.  Using random forests to diagnose aviation turbulence , 2013, Machine Learning.

[15]  S. Sorooshian,et al.  Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks , 1997 .

[16]  Pradip Kumar Pal,et al.  A New Technique for Temperature and Humidity Profile Retrieval From Infrared-Sounder Observations Using the Adaptive Neuro-Fuzzy Inference System , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[17]  Frank S. Marzano,et al.  A Neural Networks–Based Fusion Technique to Estimate Half-Hourly Rainfall Estimates at 0.1° Resolution from Satellite Passive Microwave and Infrared Data , 2004 .

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

[19]  Frédéric Fabry,et al.  The Spatial Variability of Moisture in the Boundary Layer and Its Effect on Convection Initiation: Project-Long Characterization , 2006 .

[20]  M. Rajeevan,et al.  Nowcasting severe convective activity over southeast India using ground‐based microwave radiometer observations , 2013 .

[21]  G. Visconti,et al.  A Neural Network Approach to Real-Time Rainfall Estimation for Africa Using Satellite Data , 2003 .

[22]  P. Chan.  Performance and application of a multi-wavelength, ground-based microwave radiometer in intense convective weather , 2009 .

[23]  J. R. Colquhoun.  A Decision Tree Method of Forecasting Thunderstorms, Severe Thunderstorms and Tornadoes , 1987 .

[24]  Jietai Mao,et al.  A Study of a Retrieval Method for Temperature and Humidity Profiles from Microwave Radiometer Observations Based on Principal Component Analysis and Stepwise Regression , 2011 .

[25]  R. Chakraborty,et al.  Nowcasting of tropical rain using dual frequency atmospheric brightness temperatures at Kolkata , 2012, 2012 5th International Conference on Computers and Devices for Communication (CODEC).

[26]  Nan Li,et al.  A new radar‐based storm identification and warning technique , 2012 .

[27]  Frank S. Marzano,et al.  Rainfall Nowcasting From Multisatellite Passive-Sensor Images Using a Recurrent Neural Network , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[28]  Paresh Chandra Deka,et al.  Support vector machine applications in the field of hydrology: A review , 2014, Appl. Soft Comput..

[29]  Richard K. Moore,et al.  Microwave Remote Sensing, Active and Passive , 1982 .

[30]  A. Maitra,et al.  Nowcasting of rain events using multi-frequency radiometric observations , 2014 .

[31]  Sancho Salcedo-Sanz,et al.  Accurate precipitation prediction with support vector classifiers: A study including novel predictive variables and observational data , 2014 .

[32]  Niklaus Kämpfer,et al.  Retrieval of tropospheric water vapour by using spectra of a 22 GHz radiometer , 2011 .

[33]  Chih-Chiang Wei.  Wavelet Support Vector Machines for Forecasting Precipitation in Tropical Cyclones: Comparisons with GSVM, Regression, and MM5 , 2012 .

[34]  Study of tropospheric delay over Indian region from MODIS, NCEP/NCAR data and ground based water vapor measurements at Kolkata , 2015 .