Predicting weather forecast uncertainty with machine learning

2018 ◽  
Vol 144 (717) ◽  
pp. 2830-2841 ◽  
Author(s):  
Sebastian Scher ◽  
Gabriele Messori
Keyword(s):  
Machine Learning ◽  
Weather Forecast ◽  
Forecast Uncertainty

Numerical Weather Forecast Post-processing with Ensemble Learning and Transfer Learning

2020 ◽  
Author(s):  
Yuwen Chen ◽  
Xiaomeng Huang
Keyword(s):  
Machine Learning ◽  
Transfer Learning ◽  
Ensemble Learning ◽  
Weather Forecasting ◽  
Weather Forecast ◽  
Processing Temperature ◽  
Forecast Errors ◽  
Post Processing ◽  
Numerical Weather ◽  
One Year

<p>Statistical approaches have been used for decades to augment and interpret numerical weather forecasts. The emergence of artificial intelligence algorithms has provided new perspectives in this field, but the extension of algorithms developed for station networks with rich historical records to include newly-built stations remains a challenge. To address this, we design a framework that combines two machine learning methods: temperature prediction based on ensemble of multiple machine learning models and transfer learning for newly-built stations. We then evaluate this framework by post-processing temperature forecasts provided by a leading weather forecast center and observations from 301 weather stations in China. Station clustering reduces forecast errors by 24.4% averagely, while transfer learning improves predictions by 13.4% for recently-built sites with only one year of data available. This work demonstrates how ensemble learning and transfer learning can be used to supplement weather forecasting.</p><p></p>

scholarly journals Understanding emergency manager forecast use in severe weather events.

2018 ◽  
pp. 95-105 ◽  
Author(s):  
Sean Ernst ◽  
Daphne LaDue ◽  
Alan Gerard
Keyword(s):  
Critical Incident ◽  
Information Needs ◽  
Weather Forecast ◽  
Severe Weather ◽  
Relational Trust ◽  
Forecast Uncertainty ◽  
Emergency Managers ◽  
Emergency Manager ◽  
Uncertainty Information ◽  
The Impact

For Emergency Managers (EMs), preparations for severe weather have always relied on accurate, well-communicated National Weather Service (NWS) forecasts. As part of their constant work to improve these forecasts, the NWS has recently begun to develop impact-based products that share forecast uncertainty information with EMs, including the Probabilistic Hazard Information (PHI) tool. However, there is a lack of research investigating what forecast uncertainty information EMs understand, and what information needs exist in the current communication paradigm. This study used the Critical Incident Technique to identify themes from incidents involving weather forecast information that went well, or not so well, from the perspective of the EMs responding to them. In total, 11 EMs from a variety of locales east of the Rockies were interviewed—six of whom were county-level, two city, two state, and one from a school district. We found that EMs sought increased forecast detail as a potential event approached in time and built relational trust in the NWS through repeated interactions. EMs had difficulty preparing for events when they did not have details of the expected impacts, or the likelihood of those impacts, for their regions. In summary, EMs are already starting to work in an uncertainty-friendly frame and could be responsive to the impact details and increased forecaster relations proposed with the PHI tool.

Airport Capacity Prediction with Explicit Consideration of Weather Forecast Uncertainty

2016 ◽  
Vol 24 (1) ◽  
pp. 18-28 ◽  
Author(s):  
Rafal Kicinger ◽  
Jit-Tat Chen ◽  
Matthias Steiner ◽  
James Pinto
Keyword(s):  
Weather Forecast ◽  
Forecast Uncertainty ◽  
Airport Capacity ◽  
Explicit Consideration ◽  
Capacity Prediction

Changes to intrinsic weather forecast uncertainty in one scenario of extreme future climate

2016 ◽  
Vol 142 (698) ◽  
pp. 2102-2118 ◽  
Author(s):  
J. G. McLay ◽  
C. A. Reynolds ◽  
E. Satterfield ◽  
D. Hodyss
Keyword(s):  
Weather Forecast ◽  
Future Climate ◽  
Forecast Uncertainty
Sign in / Sign up

Export Citation Format

Share Document