You open your favorite weather app. Oh no — the National Weather Service just issued a Flash Flood Emergency for your neighborhood. A wall of water 10 to 15 feet deep is expected to surge through your neighborhood within the next 30 minutes. Do you grab your cats and run for higher ground? Most of us would. Modern weather forecasts are impressively accurate. A 5-day forecast correctly predicts the overall weather about 90% of the time. Temperature highs and lows usually land within 1–2 degrees Fahrenheit of what actually happens.
How do they know?
Reading about meteorology, it’s almost enough to restore your faith in humanity. People from all over the world have set up weather instruments; they pretty much blanket our planet. Tens of thousands of sources of data stream in around the clock. Surface stations report temperature, humidity, air pressure, wind speed, and precipitation. Weather balloons climb up to miles high, measuring conditions as they go. Doppler radar tracks storm rotation in real time. Satellites scan cloud patterns, sea temperatures, and moisture from space. Every day in the U.S. alone, more than 210 million observations pour in.
Computers turn this flood of numbers into weather forecasts. We know the atmosphere follows strict physical laws. Scientists have translated these physical laws, like conservation of energy, into a small set of mathematical equations. Weather models divide the entire planet’s atmosphere into millions of three-dimensional cubes. Each one represents a small box of air. The models fills every box with exact measurements: temperature, wind, air pressure, and water vapor. Then, the models calculate how these quantities change as air, heat, and moisture flow between neighboring cubes. It advances the entire atmosphere in tiny increments, like flipping through a flipbook one frame at a time. Then it runs the model again and again (imagine dozens of flipbooks) with slightly different starting conditions each time. When most versions agree, confidence is high. When they disagree, it’s low.
OK, now imagine how you would feel if Oregon passed a law that barred meteorologists from using any observations collected after 1998?
Class Action Lawsuit: Exposing the human cost of relying on decades old data to predict the future
Read here about our recent class action lawsuit challenging the Board of Parole and Post-Prison Supervision’s sex offender risk assessment methodology. We argue that Oregon’s Board of Parole and Post-Prison Supervision is breaking the law by systematically misclassifying long-term sex offenders as Level 3 (highest risk), relying on outdated risk assessments instead of looking at their current, present-day risk to the public. One of our main goals is to get the court to declare the practice unlawful.