Weather Prediction Machine Learning Optimization


Figure 1. DICast system diagram.
Figure 1. DICast system diagram.

RAL is a leader in the development of intelligent weather prediction systems that blend data from numerical weather prediction models, statistical datasets, real time observations, and human intelligence to optimize forecasts at user– defined locations. The Dynamic Integrated Forecast System (DICast®) and the GRidded Atmospheric Forecast System (GRAFS) are examples of such technology (Figures 1 and 2).

Figure 2. GRAFS system design.
Figure 2. GRAFS system design.

DICast® is currently being used by three of the nation's largest commercial weather service companies. Applications of this technology continue to expand as there is a growing desire in industry to have fine–tuned forecasts for specific user–defined locations. This trend is clear in the energy, transportation, agriculture, and location–based service industries. RAL's expertise in meteorology, engineering, and applied mathematics and statistics is being utilized to address society's growing need for accurate weather information.


During this year significant research has been performed with machine learning techniques in an attempt to improve both short-term machine learning techniques, called StatCast, and DICast® forecasts in a project with the Kuwait Institute of Scientific Research (KISR). This system started using DICast® as the core forecast engine for a combined wind and solar forecasting system. This system will combine output from global numerical weather prediction models and a high-resolution version of WRF to produce custom forecasts for an extreme desert climate environment. The KISR project is a multi-stage machine learning methodology as StatCast, the machine learning based approached for wind and solar power predictions based on surface observations, is being developed for the KISR project for short-term predictions out to six hours, DICast® is utilized across time scales, and the Analog Ensemble (AnEn) machine learning approach produces a calibrated ensemble forecast.  The StatCast techniques are currently undergoing significant research and development to advance the state of the science for combining regime classification with supervised machine learning techniques.  The StatCast-Solar methodology is currently comparing regime-classification using k-means clustering to find statistical regimes representing cloud types and then artificial neural networks compared to implicit regime separation methods such as regression trees and random forests.  The StatCast-Wind methodology is based on using stability regimes, as classified by the Richardson number, to separate regimes and train artificial neural networks on each regime separately.  A manuscript for the StatCast-Solar is under development and presentations on these machine learning approaches to renewable energy prediction will be given at the 2020 AMS Conference in January.

More information on the KISR project can be found at [Link to]

RAL has advanced the application of machine learning to support wildfire prediction.  Atmospheric conditions, fuel type, and fuel moisture content (FMC) are critical factors controlling the rate of spread and heat release from wildland fires.  Commonly used wildland fire spread models have displaced significant sensitivity to FMC; therefore, having accurate FMC estimates to use as initial conditions is important.  The National Fuel Moisture Database provides sporadically updated information about FMC created by interpolating sparse manual samplings of live FMC and relatively sparse surface observations of dead FMC (by Remote Automated Weather Stations.  At present gridded FMC data set that can be assimilated in real-time in an operational system does not exist.  RAL built a real-time FMC database to use in WRF-FIRE coupled atmosphere wildland fire prediction model, which is a component of the Colorado Fire Prediction System.   The random forest based models predict live and dead FMC that results in more realistic, dynamic representation of fuel heterogeneity and in improved accuracy of wildland fire spread prediction.  This model runs daily and the output is displayed on the Operations tab on Improved Wildfire Spread Prediction.

Figure 3. Real-time predictions of the Dead Fuel Moisture Content updated daily on RAL’s FTP page.
Figure 3. Real-time predictions of the Dead Fuel Moisture Content updated daily on RAL’s FTP page.

These RAL forecast systems also continuing to push the envelope of advanced weather forecasting in the transportation sector. The Maintenance Decision Support System (MDSS) was adapted from its original focus on roadways to be used as a Runway Decision Support System for Denver International Airport (DIA). The system generates tuned weather forecasts and treatment recommendations for the runways at DIA. In addition, DICast® and a weather-tuned version of GRAFS form the backend weather engine used in both the FHWA and Colorado Pikalert Hazard Assessment forecast systems.


Areas of development for the next fiscal year include:

  • Extend machine learning techniques to other variables produced by DICast®.
  • Test Regime-Dependent Methodologies of predicting short-term solar and wind power generation as part of the KISR project.
  • Advance the application of machine learning in renewable energy prediction across timescales and climates.
  • Make improvements related to road temperature and precipitation forecasts in the MDSS.
  • Test the application of the Analog Ensemble on DICast® forecasts to produce probabilistic wind and solar power predictions using the Shakke Shuffle technique.
  • Utilize machine learning techniques for improving offshore wind energy modeling.