Useful Weather Sites for HARBOR Flights
Weather Prediction Introduction and Background:
I use a lot of websites. But here are two that are especially easy to use
but have a LOT of information. Here is a bit of an explanation of
some atmospheric physics to put the links in context.
Do not read the rest of this page if you don't have at least 15
minutes (an hour would be better) to think through it. Still, this
is just the very start for learning about atmospheric physics. So,
let's start the crash course.
There are many weather forecasting models.
The three most important ones for us are:
- GFS = Global Forecasting System. This is often called the US
model.
- ECMWF = European Centre for Medium-range
Weather Forecasts. This is often called the European model.
- HRRR = High-Resolution Rapid Refresh model. This is a short term
forecast based on the US model that does a finer grid scale and more
accurate terrain modeling. It is very computer intensive and hard to
do. It has 3-km resolution and is really great for the next day or
so. The max forecast is 48 hours. I never, ever, do high altitude
balloon flights without referencing this data set just before
launch. (Brandon: this data set should be especially useful for
wildfire spot forecasts and is probably what the NWS uses when they
make specific forecasts for your wildland fire teams in the summer
months.)
The GFS and ECMWF are often making different predictions. There is
no preference or advantage of one over the other. The simple fact is
that sometimes one of them does a better job then the other one. No
one knows why certain conditions work in one case but not the other,
there are just too many variables. For the sake of ease, all the
discussion and links below are based on the GFS system.
First, the Tropical Tidbits blog provides an interface to the
GFS forecast models that is very intuitive and much easier to
navigate than the raw NOAA data. There is a massive amount of
information on this website. For the main map of interest you will
need to understand how to read it. Most of it is fairly
straightforward, but the details are interesting.
To read the maps that I find most useful for storm forecasting:
The black lines are isobars showing pressure in mb (same thing as
hPa) where sea level is 1013 mb and Powder Mountain is about 725 mb
on a "perfect average" day (see the PS below). These pressures on
the maps are all adjusted to sea level values as if there were no
mountains, just a smooth surface of Earth. In general, you can
assume that air will flow from higher pressure to lower pressure.
The closer together the lines are, the more serious the
weather changes. Think of it as a topo map
with elevation contour lines mapping out high pressure ridges and
low pressure troughs.
You probably already know that, in general, high pressure regions
are usually fair weather and low pressure
regions are associated with storms. In the Northern Hemisphere winds
move clockwise around high pressure systems and counter clockwise
around low pressure systems.
More complex: The dashed red and blue lines with numbers in the 500
range are geopotential heights. That starts to get wicked
complicated. It is an integral of gravity, Tv (average virtual
atmospheric temperature), and atmospheric pressure where the
integral's limits are two pressures. In the case of this map, that
would be 1000 mb to 500 mb which is the weight (potential energy) of
the lower and middle troposphere. It is in units of decameters (tens
of meters).
Whew. What that really means is the average temperature of the
atmosphere at that location but in very funny units. As a general
rule, if that dashed line is red you will probably get rain, if it
is blue, you'll probably get snow. That will change with surface
altitude and surface temperature. If you are at sea level and just
into the blue dashes, you will probably get rain. If you are at high
ground level altitude and just into the blue dashed lines, you'll
probably get snow.
Why is that important for this chart? If you look at the right side
you will see precipitation forecasts as a "heat map". Unfortunately,
many of the colors overlap. That is where understanding the
geopotential height (remember that's really an average atmospheric
temperature) comes in.
Start here with one of these two links, I suggest the first one for
details, the second one for an overview:
1. For the NW Continental USA.
2. For the entire continental USA (CONUS):
On these you can use the various controls to move the forecast
forward and backwards with the arrow controls just above the map. It
is mostly pretty intuitive. One item that is not obvious (at least
to me at first) is the plus/minus controls. Those speed up and slow
down the animation speed when you hit the run arrow.
I usually prefer to just step through with the right arrow.
Advanced Topic: Click on some spot on the map to get a "skew-T" plot
and a few other plots such as a "hodogram." (A hodograph is a plot
of wind speed and direction with altitude.) The DGZ levels listed on
the left edge of the skew-T plot are the dendritic growth zone which
is a fancy way to say that ice crystals can easily form in this
region. That will only happen if there is moisture. For snow for
skiing, ideally, this will be wide and close to the ground.
The more general access to the data can be found here:
Then select the "Forecast Zones" along the top menu bar.
You want the GFS model. If that isn't what shows up in the upper
left corner, then click on "Global" and select GFS.
At the bottom left of the map is a tab called "Regions." Click on
that and select either "Northwest U.S." or CONUS.
Second NOAA Data Set:
Okay, my next most favorite NOAA data is directly from the
satellites. This is "live" data from the satellites. I like
selecting the "animation loop". Note that many of these
wavelengths are in the visible (0.4 um to 0.7 um) and you won't
see anything at night. That's why I like the infrared bands,
which also shows moisture.
Key things to do:
- Select your satellite and image filters carefully. One of my
favorites is "Band 10" That band shows moisture in the bottom
2/3 or so of the troposphere.
- Select the "Loop Size" which is the number of images at
(typically) 5-minute intervals. 72 images is 6 hours.
The more general starting point is here:
P.S. Atmospheric scientists prefer to refer to altitude in pressure
units because atmospheric pressure is what drives air flow and can
determine temperature, humidity, etc. Ballpark values on a perfect
average day:
Sea level = 1013 mb (This one is at 25C, standard room temperature.)
Ogden (my office on campus, 4,685') = 845 mb
Pico de Orizaba (second highest peak in North America, an 18,941'
stratovolcano = 466 mb (about half the air at sea level)
Mt. Everest summit = 293 mb (about a quarter the air at sea level)
You can calculate your pressure altitude using the NASA/NOAA
standard atmosphere very easily at this website: