The Infrastructure Problems of the National Weather Service

 This is a lot of talk about the new infrastructure bill that is being considered in Congress.

Supporters of U.S. weather prediction and the National Weather Service need to get organized since the National Weather Service has profound infrastructure problems that need attention.  Serious enough that it is holding back the quality and reliability of weather information that is being provided to the American people.

Let me tell you about a few of them.


U.S. weather radars

 The current U.S. radar network was installed in the late 1980s and early 1990s and applies the radar technology of forty years ago, with some minor upgrades (dual polarization).  Furthermore, the radar network has major gaps in coverage including the Oregon coast, eastern Oregon, and large portions of eastern Washington (see map).   

Amazingly, the National Weather Service has no process in place to replace that current radar network (called NEXRAD or WSR-88D) and has given no priority to filling the radar gaps.  A major infrastructure upgrade is acutely required.

National Weather Service Computer Infrastructure

This is an area of profound need in many ways.   

• Insufficient computing resources on which to run numerical weather prediction models.   At this point, the NOAA/NWS Environmental Modeling Center has inadequate computer resources to apply the physics, data assimilation, ensemble approaches, and resolution needed for state-of-the-art forecasting.  To do so, would require at least 100 times what they have now.  
• They lack the computer resources for "hot" backups that would immediately take over when the primary system fails.    
• Their current computer infrastructure lacks the bandwidth and capabilities to distribute even the current observational and modeling products, 
• The NWS computer systems lack sufficient storage capabilities.
• The current computer infrastructure has single failure modes, which results in the loss of important products.   This happened recently with the several-day loss of key marine products.
• There is inadequate bandwidth (ability to communicate) between and to National Weather Service offices around the nation.
• There are insufficient communication capabilities to derive maximum benefits from the current observing network.

• 

The solution to this problem is a hybrid one:  more centralized computing capabilities for key model runs and far more extensive use of cloud computing for needs that are variable in time and for distributing data sets.

U.S Coastal and Ocean Buoys

NOAA/NWS maintains hundreds of ocean and coastal buoys that provide essential information for the safety of mariners.  Unfortunately, roughly half of the buoys (red colors in the map below) are not working or have major sensor failures.  NOAA/NWS has not upgraded and hardened the buoys and lack the resources to rapidly repair/replace broken or lost buoys. 

I could provide several other examples, but you get the point.  

The infrastructure of the National Weather Service needs a profound upgrading and improvement if the American people will be able to enjoy the skillful weather prediction required to protect lives and property, as well as promoting the productivity of the nation.

Hopefully, some of these needs will get into the infrastructure bill.

____________________________________________

Announcement:  The Northwest Weather Workshop on May 1

The agenda for the Northwest Weather Workshop is up....please check it out here.  This meeting is the major local gathering to discuss the weather of the past year and the latest developments in understanding Northwest Weather. The meeting will be a half day (morning of May 1) and will be online.  Anyone interested can attend and we recommend you register if you want to be on our mailing list.