There is an important resource that members of the AMS Weather Band may not have on their radar, but should: Statements of the AMS. These are relatively short (a few pages) documents that provide the authoritative position of the AMS on an issue or topic. AMS Statements are created following a very tightly proscribed and rigorous process that involves experts in the subject are being covered and multiple levels of careful review.
Statements can cover “the state of the science” on things like climate change, provide “best practices” for professionals in the AMS community (or others), or offer policy recommendations to lawmakers or government agencies. In each case, because a statement has gone through very careful review during its creation, and because it represents the official position of the AMS, the statement can be trusted as relying on the very best science and understanding available at the time.
A terrific example of an AMS Statement providing the “state of the science” on a topic is the
AMS Statement on Weather Analysis and Forecasting, available here. This statement offers a summary of the kinds of observations and modeling that go into creating the forecasts we are used to seeing every day, including reviewing the very important role that human forecasters play in the process despite the ever-increasing skill of computer-based forecasts. The statement then provides information on the reliability of various types of forecasts based on current capabilities.
This last major section of the statement is especially important. When we see an individual or
website claiming to be able to forecast the day-to-day weather for a month from now, we might wonder if that is possible. The AMS statement makes it clear that the science shows that it is not possible, and we should therefore dismiss any source of forecasts that make such a claim. As noted in the statement, there is good skill in regular weather forecasts out to about eight days.
Beyond about 10 days, and especially when looking at timeframes beyond two weeks, the forecasts can, at best, point to expected trends (such as, warmer/colder than normal, or wetter/dryer than normal). Those forecasts can still be terrifically useful in many situations (such as agriculture), but you cannot rely on them to plan an outdoor event like a family picnic a month from now. The skill of these sorts of trend forecasts has been improving steadily, and there is some skill in them at the seasonal level (that is, forecasts warmer/colder and wetter/dryer over the next several months).
Our ability to forecast the weather has improved steadily with new observational capabilities (such as ever more sophisticated weather satellites), increased computer power (allowing great improvements to the computer simulation models), improved understanding of atmospheric processes, and new forecasting techniques (such as machine learning and artificial intelligence). An example of that improvement is that 20 years ago, you might have been reluctant to plan that family picnic on Saturday based on Monday’s forecast, but now we all do that fully expecting the forecast to be correct. That said, there are inherent limits to the predictability of the atmosphere on different scales and for different types of phenomena. Understanding our current capabilities, as well as some sense of those limits, allow us all to use forecasts more effectively. The AMS Statement on Weather Analysis and Forecasting provides an authoritative assessment on those capabilities that we can trust.
You can see the entire list of AMS statements here. This list covers an impressive array of topics, and it is worth taking a few minutes to browse it. I suspect many of you will find several statement that will be of interest.