From Penn State University and the “but we guarantee you there’s no predictability limit in climate science” department comes this interesting study.
UNIVERSITY PARK, Pa. — In the future, weather forecasts that provide storm warnings and help us plan our daily lives could come up to five days sooner before reaching the limits of numerical weather prediction, scientists said.
“The obvious question that has been raised from the very beginning of our whole field is, what’s the ultimate limit at which we can predict day-to-day weather in the future,” said Fuqing Zhang, distinguished professor of meteorology and atmospheric science and director of the Center for Advanced Data Assimilation and Predictability Techniques at Penn State. “We believe we have found that limit and on average, that it’s about two weeks.”
Reliable forecasts are now possible nine to 10 days out for daily weather in the mid-latitudes, where most of Earth’s population lives. New technology could add another four to five days over the coming decades, according to research published online in the Journal of the Atmospheric Sciences.
The research confirms a long-hypothesized predictability limit for weather prediction, first proposed in the 1960s by Edward Lorenz, a Massachusetts Institute of Technology mathematician, meteorologist and pioneer of the chaos theory, scientists said.
“Edward Lorenz proved that one cannot predict the weather beyond some time horizon, even in principle,” said Kerry Emanuel, professor of atmospheric science at MIT and coauthor of the study. “Our research shows that this weather predictability horizon is around two weeks, remarkable close to Lorenz’s estimate.”
Unpredictability in how weather develops means that even with perfect models and understanding of initial conditions, there is a limit to how far in advance accurate forecasts are possible, scientists said.
“We used state-of-the-art models to answer this most fundamental question,” said Zhang, lead author on the study. “I think in the future we’ll refine this answer, but our study demonstrates conclusively there is a limit, though we still have considerable room to improve forecast before reaching the limit.”
To test the limit, Zhang and his team used the world’s two most advanced numerical weather prediction modeling systems — The European Center for Medium Range Weather Forecasting and the U.S. next generation global prediction system.
They provided a near-perfect picture of initial conditions and tested how the models could recreate two real-world weather events, a cold surge in northern Europe and flood-inducing rains in China. The simulations were able to predict the weather patterns with reasonable accuracy up to about two weeks, the scientists said.
Improvements in day-to-day weather forecasting have implications for things like storm evacuations, energy supply, agriculture and wild fires.
“We have made significant advances in weather forecasting for the past few decades, and we’re able to predict weather five days in advance with high confidence now,” Zhang said. “If in the future we can predict additional days with high confidence, that would have a huge economic and social benefit.”
Researchers said better data collection, algorithms to integrate data into models and improved computing power to run experiments are all needed to further improve our understanding of initial conditions.
“Achieving this additional predictability limit will require coordinated efforts by the entire community to design better numerical weather models, to improve observations, and to make better use of observations with advanced data assimilation and computing techniques,” Zhang said.
###
The paper: (open access) “What Is the Predictability Limit of Midlatitude Weather?”
https://journals.ametsoc.org/doi/10.1175/JAS-D-18-0269.1
Abstract
Understanding the predictability limit of day-to-day weather phenomena such as midlatitude winter storms and summer monsoonal rainstorms is crucial to numerical weather prediction (NWP). This predictability limit is studied using unprecedented high-resolution global models with ensemble experiments of the European Centre for Medium-Range Weather Forecasts (ECMWF; 9-km operational model) and identical-twin experiments of the U.S. Next-Generation Global Prediction System (NGGPS; 3 km). Results suggest that the predictability limit for midlatitude weather may indeed exist and is intrinsic to the underlying dynamical system and instabilities even if the forecast model and the initial conditions are nearly perfect. Currently, a skillful forecast lead time of midlatitude instantaneous weather is around 10 days, which serves as the practical predictability limit. Reducing the current-day initial-condition uncertainty by an order of magnitude extends the deterministic forecast lead times of day-to-day weather by up to 5 days, with much less scope for improving prediction of small-scale phenomena like thunderstorms. Achieving this additional predictability limit can have enormous socioeconomic benefits but requires coordinated efforts by the entire community to design better numerical weather models, to improve observations, and to make better use of observations with advanced data assimilation and computing techniques.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Planning on weather beyond 15 hours is really risky business. As the saying goes, fuel weights less than weather.
Weather is the sum of what is happening at a given place at a given time. Climate, at its most basic, is the sum of all weather.
If weather will never be able to be predicted more than 2 weeks in advance what chance is there that climate can EVER be any better than that. Sure, it is always possible to say that temperature will go up (or down) but how much? and when? and where? is impossible.
Interesting. I have been tracking the forecast for April 22 (start of golf season) since April 1. The range through April 16 is 49º to 68º for the high temperature. This range was within the last week. I’d say the greatest accuracy is probably within 3-4 days. However, I have complete faith that our climate will increase by an average of 5º in 30 years because 97% of social scientists say so.
I think it depends on what exactly is being forecast. The temperature in a given location? Whether it will rain or not?Her in Qld there was a guy many years ago, “Cyclone Bill” Devonshire who posted long range forecasts in the annual tide times book. If he said there would be a major storm in Brisbane on the 10th of December there would be a storm within 48 hours either side of that date.
Scary reliable.
Weather does not change that fast, so the upper limit of the weather in the future being the same as today is one to two weeks. Probably closer to one. That is not prediction, that is dependence on stasis.
2 weeks – yeah nah as they say here in NZ. I rely on the weather forecasts to do parts of my job and our local MetService regularly can’t even get their predictions right a couple of hours before or even after. The number of times I’ve told my clients we’ll pencil it and we’ll see what happens on the day when we wake up, in fact I say that every single time now.
A bit late to the discussion, but I’ll add this FWIW.
I spent Saturday on the Ouachita Trail tracking runners in a 50 mile ultra running event. All week the forecast had been for showers to begin about 5AM Saturday morning. And sure enough, they arrived right on schedule. Normally one can watch fronts move in from the plains NW of us. Not hard to predict that kind of weather. This weather came up from the Gulf, and wasn’t apparent on any radar map until materialized overnight. But the forecasters knew what was going on and nailed it.
Two weeks may be the limit, but there is enough variability that I usually wait until 2-3 days before an event to start relying on a forecast. At that point, the forecasts are right more often than wrong.
But none of this has anything to do whatsoever with forecasting “climate change.” Just because WEATHER cannot be predicted accurately more than a few days (to two weeks) out does not mean that CLIMATE cannot be predicted beyond that horizon. Large scale weather patterns — on shorter time frames, ENSO; on longer time frames, PDO, AMO — do provide a basis for longer term forecasts. The accuracy window widens — months, for ENSO, to years for PDO or AMO — but changes within such natural cycles are still predictable.
I agree wholeheartedly that predicting long term climate change from CO2 increases is folly. But this isn’t proven just because we cannot predict weather more than a week or two into the future. This article was talking about predicting weather, not climate. Do not read anything else into it.
The dreaded word ‘models’ appears which means that no matter how much data they chuck in, if the model is wrong so is the outcome. The human experience seems to be no longer part of a forecast which perhaps explains why I think they have got worse. A Met Office forecaster on BBC actually admitted that with the jetstream flow meridional instead of zonal, their models don’t work. The constant changing of forecasts allow the MetO to claim they are far more accurate than they truly are.
Never say never at state universities with grant writers. They will continue the moonshot spending request for supercomputer clusters and supporting power plants to get to 2.5 weeks prediction power while across campus the quest for fusion power soldiers on with other grant funding pots.
Sounds like an open-ended, non-circular, chain-of-events: get a set of conditions, get two weeks chain of events. Climate science is not a c-o-e, but a set of forces that act upon an initial set of conditions to necessitate a series of outcomes that become a continuous set of conditions.
The weather low predictability is not in conflict with climate science. Weather forecasting is trying to predict within the noise of natural variability. Climate science is trying to predict the change in maximum, minimum and modal elements of the natural variability. Two different things.
They say that the limits to accurate daily forecasts are two weeks, yet in the same breath, climate “scientists” claim that climate models can accurately predict what will happen decades into the future??? The hubris of these “scientists” continue to astound me.
Consider the following statements I believe to be true:-
1. Modern climate models now apparently use the same differencing schemes as the weather models, (this wasn’t the case in the early days when I worked at the UK Met Office. Weather models lost energy so as to be almost useless after a few days. Climate models used differencing schemes based on the Arakawa Jacobian which preserved energy. They have now converged allowing weather models to be run out for more than a few days).
2. In spite of this, weather models are OK to 1-3 days, rarely longer.
3. With a co-author, we proved in 1976 that cyclonic patterns effectively randomise atmospheric trajectories within a few days or less even in 2-dimensions. “Computation of Horizontal trajectories based on the surface geostrophic wind”, Sykes and Hatton, Atmospheric Environment, (10) p.925-934. (1976)
4. The models are remarkably resilient to major changes. In one experiment, I dropped all the non-linear terms every other time-step and there was barely any difference after 7 days with a 6 minute timestep. This was described to me as “the effects of smoothing”. I have always doubted this.
5. Parameterisation of the physics in these models is enlightened guesswork.
6. Software is full of unquantifiable error, “The case for open computer programs”, Ince, Hatton and Graham-Cumming, Nature, doi:10.1038/nature10836 (2012).
It follows (at least to me), that predicting long-term climate behaviour and climate policy on the basis of these models alone, even when there are multiple models and multiple scenarios, is little more than guesswork.