Excellent Forecasts and Major Weather Disasters: Why Don’t We Do Better?

Presented on March 25, 2026, by

Professor Cliff Mass
Atmospheric and Climate Science
University of Washington

ABSTRACT
The ability to forecast damaging and deadly weather events has increased dramatically during the past decades due to improvements in the observational network, more skillful numerical models, and advances in statistical post-processing and machine learning.  But even with improved prediction skill, catastrophic damage and great loss of life continue.  This talk will explore this important issue, including a review of a dozen major weather events that illustrate this failure mode. The origins of the problem are varied, including inadequate communication, disorganized evacuations, critical infrastructure not properly constructed or maintained, forecast guidance that was not sufficiently specific or clear, or emergency response being late or inadequate.  This presentation will suggest specific actions for addressing the continuing great loss of life and property from well-predicted meteorological threats. For example, T\the National Weather Service should reorganize its office structure, a National Disaster Review Board (NDRB) should be established, greater emphasis is needed on increased infrastructure resilience to extreme weather events, greater priority should be given to educating the public about extreme weather, and the Federal government should improve local weather prediction capabilities.

BIO

https://atmos.uw.edu/~cliff/ShortBio.pdf