Linking Weather Forecast Ensemble Spread to Forecast Risk and Confidence
Background
Potential Error Ranges of Forecasts can be determined through a number of methods.
One method, Ensemble Forecasting, include multiple forecasts with small variations in initial conditions or the way in which the model calculates certain process. Regions and times in a forecast in which the ensemble of forecasts differ significantly from each other indicate areas where a weather situation is
expected to be unpredictable.
Example of a forecast ensemble product showing the average temperatures forecasted by
an ensemble of forecasts and the standard deviations of the forecasted temperatures.
the areas with high spread show areas of high potential error in a forecast.
Another method is to determine the way to project forecast risk is to observe measurable features that we know tend to increase the complexity and unpredictability of short-term weather patterns. SDSMT and NATO has develop such a method call the "Confidence Index" (or "CI") in which complex weather features are identified and statistically linked of the resulting forecast error.
A year of forecast errors plotted against estimated
forecast risk or "confidence"
In previous work we have linked errors using the latter CI method to forecast errors. We now will be working with both the National Weather Service's forecast ensembles and those of a university numerical modeling consortium, the "Big Weather Web" team to associate CI-derived risk to ensemble spread and from there associate both to resulting model error.
Project Details
We will work with a student on validating a forecast model used in the Western South Dakota for real-time forecasts and decision making and also associate forecast error to forecast risk using the Confidence Index.
The student will also tour the Rapid City, National Weather Service where measurements are collected and forecasts are produced and also visit a remote South Dakota Mesonet station to understand what is measured and how the data is incorporated into a weather forecast model as well as the larger computational- and human-driven weather forecast process.
Skills Development
- Programming
- Statistics
- Numerical Analysis
- Weather Forecast Modeling
- Unix environment
Research Duties
Students will work with regional scale forecast ensemble output to collate ensemble means and spreads and associate them with CI scores calculated concurrently for those scenarios.
Impact of Research
We hope that this project will improve regional numerical weather prediction and assessing the risk of forecasts which can then be used to assign workflow of the larger forecast process (both the human meteorologists and their numerical tools). This project is also part of a larger project to assess the role of multiple regional forecasts in creating a multi-university/organization forecast ensemble.
Research Team
You will be working with Dr. Capehart.
Dr. Capehart has expertise in numerical weather prediction, regional climate modeling and hydrologic modeling, and will serve as the lead on the project.
You'll also be working with the larger Big Weather Web team, made up of 8 universities and National Center for Atmospheric Reserach.
AMS Talk: Capehart et al, 2013: A confidence index for numerical weather prediction forecasts, Special Symposium on Advancing Weather and Climate Forecasts: Innovative Techniques and Applications, Austin, TX, 7-9 January 2013, American Meteorological Society
https://ams.confex.com/ams/93Annual/webprogram/Paper218311.html
Powerpoint on the CI Concept : CI_seminar_2014-09-23.ppt
Powerpoint on the basics of NWP : SDSMT_REALTIME_MODEL_2012.pptx
For more details on this project, contact the lead investigator:
Dr. Bill Capehart
Atmospheric and Environmental Sciences
South Dakota School of Mines and Technology
Rapid City, SD 57702
Ph: +1 (605) 394-1994
Email: William.Capehart@sdsmt.edu
Comments (0)
You don't have permission to comment on this page.