Welcome to Thursday.
Today is going to be a hot one for the area-- let's just get that one in right off the bat. I suppose two questions are: "How hot are we talking?" and "Do we need to worry about showers?" One of these questions can be tackled by using yesterday's weather as an example.
I showed three maps to you on yesterday's blog post to identify possible threats for storms. Each one had a different resolution. Re-visiting that blog post and comparing it to what actually happened would crown a winner...
Goes to the 4km NAM Simulated Radar for it's excellent work depicting the sunshine and the shower positions during much of the day. Now, in truth we did have a few poppers up there just like we talked about, but it's important to know where "most" of the weather is generally going to happen, because in a way it's easier to be able to simply say that "isolated showers and storms are going to be out there somewhere today". We cover a very large area (Pikeville, KY to Sutton, WV or Athens, OH ... Richwood, WV on past Portsmouth, OH...etc. take your pick-- it's a long drive) :-)
Here's what that model depicted, and what happened...
|NAM 4km Simulated Radar from Wednesday||Radar from Wednesday|
Not a bad job... and a good idea to have a look at what it's saying about today's weather as well...
---Weather Geeky Tangent About Weather Models (feel free to skip)---
As I briefly described before, computer models with simply just a better resolution of the terrain can often yield a better forecast when looking at the same data. We saw it here in this example yesterday, but it happens all the time when looking at the complex Appalachian terrain. Having more data points on the ground gives us more dots to connect when trying to get how the weather is resolved for the next day.
Here's a relevant image from a paper by Kwang-Deuk et al (2005):
This shows wind streamlines and intensities, with the terrain resolution increasing more and more from (a) to (d). Notice how much extra goodies jump out at the forecaster from a better terrain resolution.
Closer to home, here's a look at how a model with a higher terrain resolution resolves the same data as one with a coarser resolution (from MetED):
So that settles it then, right? Let's have the finest resolution of the terrain, and the most gridpoints, and run that model in 5 minute increments all the way out through the next 4 weeks, please... Well, unfortunately it's not that easy yet. With all the leaps and bounds we've been making in computing these days, you would have thought the sky is the limit. Well, this is one computer that is cranking out one of the models we look at each day:
Apparently, billions and billions of calculations per second are simply not enough. We have to limit some element of what we want to do with these models, or else it would take 6-hours (or longer) to crank out a 6-hour forecast (which would obviously be useless). So until computers are stronger (and exponentially so)...Meteorologists would use some hi-resolution models for near-term forecasts, or ones that confine their area of resolution to just something really close (like the Great Lakes region), or both. For other products, coarser resolutions are simply unavoidable. But hey, it's certainly better than the days of the 70s and 80s. The first forecast model, the LFM, had grid spaces around 50km (that's one grid point to represent all weather between Charleston and Huntington, with a single number for 'elevation'). Getting a single value for terrain across a vast area like the Great Plains or the Ohio Valley isn't a big problem. But throw the Appalachians into the mix, and your 'bust' potential goes way up :-)
Okay, so anyway, here's what we're looking at on the hi-res simulated radar for today:
|NAM 4km Sim Radar - 8am||NAM 4km Sim Radar - 2pm||NAM 4km Sim Radar - 8pm|
Looking at the maps, most of our day will be spent under hazy, hot, and humid sunshine, but bookended by nocturnal showers and storms. We might as well have them that way, because we can get things done outside and then be in when it rains. But given the heat around here (temperatures around 90-degrees), any storm that does form will definitely be a big rainer.
The rest of the week still retains the opportunity for showers and storms, something we can talk about tomorrow, but here's a graphic that explains the rainfall potential going forward:
|HPC - Total Rainfall - By Saturday PM|
In the grand scheme of things... this is not a whole lot of rain over three days in a turbulent weather pattern. To me this indicates that individual areas may only get one thunderstorm overhead during the whole period (which also comes across as unlikely considering the bad luck some of you have been having). The good news is that we'll have most of the day to see how things are going before most of the storms pop. The bulk of the storms in the wider region will be in the overnight, fizzling during the morning hours, then picking back up late in the next day.
A final point: Respect the heat out there today... If you're going to be out there for a while, please take along the water bottle and put on sunscreen. Your body is not yet used to this kind of heat, so those kinds of excuses don't apply yet ;-)
Onward to the maps!
|Regional Radar/Satellite with Warnings Tracking||
From the Storm Prediction Center (below): Click For a Larger Image
|Activity Overview||Storm Outlook||Watches||Potential Watches||Storm Reports|
|Temperatures||HD Doppler Radar||Estimated Rainfall||Active Warnings|
|Click For Larger||Click For Interactive Radar||Click For Larger||Click For Larger|
Have a great day everyone!