Working Together to Forecast Weather Better

Friday, April 12, 2013
2:57 PM

If any of you ever watched a certain t.v. show that’s about more than just card games, you are familiar with the picture above.

We need to work together more as a society in general. I mean, look at Congress. The rifts between the Democratic and Republican Parties are huge, and progress is being stifled by people who are unwilling to make compromises. I disagree strongly with a lot of the actions that are being perpetrated both domestically and abroad by our government, and I know that a lot of my priorities don’t match up with those being debated in our nation’s capital. However, even if I disagree, it is still better to try and work out a solution that works for the majority of people rather than just debate, pass, filibuster, veto, etc. One of the sectors that is in the most dire need of cooperation, though, is the meteorological sector.

There is plenty of international cooperation when it comes to climate change. There is a conference called the IPCC (International Panel on Climate Change) that has produced four major scientific reports drawing upon scientists all over the world with the first one occurring in 1990 and the last one occurring in 2007, with the next one expected to be completed in 2014. The Kyoto Protocol (1997) to the United Nations Framework Convention on Climate Change (UNFCCC) is an international treaty that restricts the amount of greenhouse gases that an industrialized country can release. Greenpeace, a non-governmental environmental organization that has offices in over 40 countries, focuses on essentially everything necessary to keeping our earth green and peaceful, and climate change is one of their chief priorities. There are countless other organizations that work together to research paleoclimate, assess our current climate and GHG emissions, and predict future scenarios based on the magnitude of GHGs emitted into the atmosphere, with particular emphasis on predicting both the direct warming from the greenhouse effect due to these gasses and the additional warming that will be experienced due to positive feedbacks, as there are far more positive feedbacks than negative feedbacks at this time. One of the strongest positive feedbacks right now is the relationship between declining sea ice in the arctic and an increased absorption of solar insolation due to the much lower albedo of water. Just to clarify, albedo represents the amount of solar radiation radiated back into space. Snow has a high albedo, so it reflects a bunch of solar radiation to space. Water has a low albedo and absorbs most of this radiation, thus warming the planet at a higher rate.

Unfortunately, the meteorological sector of atmospheric sciences is dominated the competition and not cooperation. While some competition is good (basketball games would be really boring if the teams both tried to help each other win), competition results in inconsistencies in data collection and forecasting. This was brought into the forefront of American media when Superstorm Sandy hit the United States, as the European model consistently forecast a U.S. landfall while the American model took the storm harmlessly out to sea. Around a month ago, the two main American models, the NAM and the GFS, predicted a monster snowstorm to hit D.C.. The storm never came. To be fair, this was an extremely difficult forecast and made forecasting Sandy look like a piece of cake, and even the European model wasn’t perfect. However, it produced far less of a snowstorm than the American models. The bottom line is that there are countless examples of the European model clearly outperforming the U.S. models, and there are very few examples of the U.S. models triumphing over the European model.

Now, at first glance, it may seem like I’m trying to advocate for the U.S. to simply try to hitch a ride with the European model to improve their weather forecasting. Well, it’d be nice if we could run UW’s super high resolution mm5/WRF models using output from the European model, but I believe we could even improve upon the European model if we lumped our resources together. The European model has ten times the computing power of the GFS, yet we have some extremely powerful computers that we would be using that are currently being used for other services, particularly climate prediction. If we could work together with the Europeans to “spread the responsibilities” of meteorological and climatological prediction, we would certainly end up with better forecasts for the short term. Much of climate research is applicable worldwide (after all, when was the last time there was an international conference on Puget Sound Convergence Zones?), but meteorological research pertains to certain areas. We share our expertise on climate with the world, but we have limited access to the extensive numerical prediction programs that are occurring in Europe. If we could further propagate our research throughout the world and create agencies that work across countries (ex: not NOAA or NASA but something that all countries are involved in), I feel like we would have better success. I’m sure there are some agencies like this, but I cannot name any off the top of my head. Regardless, I still think they are a good idea.

Additionally, there is research that has been done by the U.S., particularly for mesoscale modeling, that has not been done on the same scale in Europe. A prime example is our high-resolution models that have a resolution of 4/3 km. This is extremely useful in predicting certain types of weather in certain regions, but it is useless if the models they are initialized off of (like the GFS and NAM) are poor forecasters in the first place. If we could share our knowledge, research, and resources dedicated to mesoscale meteorology with the European’s clear advantage in synoptic scale meteorology, amazing things could be accomplished.

But wait, there’s more! In the U.S., we have a very cool program associated with the National Weather Service (NWS) called “CoCoRaHS” (Community Collaborate Rain, Hail, & Snow Network) where volunteers send in daily precipitation reports. On average, 18,000 – 20,000 precipitation reports from average citizens all across the U.S. and Manitoba are reported each day. That’s a lot of precipitation reports. Because of the higher computer power that the European model has, they can take in more data and therefore obtain better forecasts by being more specific with their numerical inputs and avoiding the need to “approximate” what the weather would be in certain locations between the data points. I do not know if the ECMWF (European Center for Medium-Range Weather Forecasts) has a system like this, and I also do not know if any of the models use any of this data. Still, it just goes to show that there is plenty of data available, and by combining resources with those Europeans, we could achieve even better weather forecasting.

I am technically a CoCoRaHS member but I have had problems with my weather station and haven’t sent in reports. It’s something I definitely want to do in the future; it’s a great way to help education and research applications. You definitely learn a lot about simple ways to accurately measure precipitation and report it to the NWS. Anybody can become a CoCoRaHS member, but you are required to attend a training session (either in-person or online) to be certified and have your reports actually be recognized by the NWS. It’s something I’d highly recommend; you can apply here.

I should point out that I believe it is important to have different weather models, however, as the equations that each weather model uses to create forecasts from numerical data are slightly different. Often, the weather that ends up occurring is a “blend” of what was forecast by individual models, so it is definitely important to have different models. Another thing that is becoming more and more common in weather forecasting is the use of “ensembles” of output from models, with each ensemble using a slightly different set of initial conditions. This statistical and probabilistic approach to weather forecasting has been proven to have greater accuracy than the “operational” approach, which just uses all the data from one set of initial conditions (at least to my knowledge, somebody please correct me if I’m wrong). The UW runs their own probability forecasting tool for the Pacific Northwest, and the few times I have used it, it has done an exceptional job, particularly with short-range temperature forecasts. You can learn more about it here.

I know I’ve written a lot… I’m not the best writer by any means and I’m definitely not as knowledgeable about this subject as I would like to be, but to sum things up, the Americans and Europeans could share the resources they have between them and obtain better forecasts as a result. And as we saw with Hurricane Sandy, the benefits of doing so would be crucial to the health of our economy and the safety of individuals.

On another note, the competition between private sectors can also have negative consequences for the accuracy of forecasts. Most people do not get their weather information from the NWS; they get it from guys like Al Roker, Steve Pool, Jim Cantore, and large commercial organizations like The Weather Channel or Accuweather. Now, my understanding of the economy is extremely poor, as evidenced by this meme I just made.

However, I do know that some weather sites have resources that others do not, and that weather forecasts would be better communicated to the public if these private companies helped each other out and built upon the resources that each of them have.

Am I a socialist? I don’t think so. I’m just very ignorant to how the economy works. Perhaps somebody could take these ideas that I have laid forth on this blog and think of some way to practically apply them. But the benefits of increasing weather forecasting accuracy would be huge, and I believe that international, domestic, and commercial cooperation is a very efficient way to do this.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s