<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p><font face="DejaVu Sans">Hi Ria</font></p>
<p><font face="DejaVu Sans">Thanks for the feedback. A model is only
useful if there is a "closed loop" to verify that the
predictions match actual conditions experienced on the day.<br>
</font></p>
<p><font face="DejaVu Sans">First thing I need to check/fix is the
timestamps on the Beta forecasts. I have not had a chance to
look at this yet. These are labelled "1100 SAST (1100Z). There
is a two hour discrepancy between those two times as 1100 local
time is 0900 GMT. I suspect the 1100Z is correct and that the
beta forecast is actually for 1300 local. </font></p>
<p><font face="DejaVu Sans">The archiving of the beta models is
still not properly setup but I managed to find ZS2+1 (ie one day
earlier) forecasts for 0900, yesterday 5th July, which, if the
above is correct, match the times of your actual readings. I
have attached 2m and 10m surface wind forecasts together with
the head and side graphics which have the header and scale
information. I think these are closer to the legacy RASP and the
actuals.</font></p>
<p><font face="DejaVu Sans">When you look at a forecast like RASP,
the forecaster needs to:</font></p>
<ul>
<li>Interpret from the forecast data into actually flying
decisions. (Ie can I fly? etc)<br>
</li>
<li><font face="DejaVu Sans">Determine a confidence level in the
forecast. (Do I believe this forecast?)</font></li>
</ul>
<p><font face="DejaVu Sans">Disclaimer - I am not familiar with wind
conditions required for foot launching at Signal Hill. so the
following is a general comment.</font><br>
</p>
<p><font face="DejaVu Sans">If the conditions are "unusual" then
confidence level drops. You can see from the charts a
significant wind speed difference between Cape Point,
Scarborough and Table Bay. Somewhere between those points the
wind becomes "too strong" to foot launch. The only difference
between the models is where that line falls. A few km either way
becomes is significant to your flying decisions. You have to
watch the forecasts and compare them to actual conditions and
develop interpretation skills and confidence levels to help make
that call. In other weather conditions, the decisions might be
much easier.</font></p>
<p><font face="DejaVu Sans">Note there are technical differences
between the RASP models. Both are based off the "GFS" weather
model:</font></p>
<ul>
<li><font face="DejaVu Sans">The legacy forecasts extrapolate off
standard resolution GFS data, that was available when the code
was written. It uses version 2 of the WRF forecast program to
extrapolate the high resolution forecasts. The extrapolation
requires a two iteration "window" forecast to reach the 1.3km
resolution.<br>
</font></li>
<li><font face="DejaVu Sans">The beta forecasts use new high
resolution GFS source data. It extrapolates with version 3 of
the WRF program and it does it in a single run to a 2km
resolution.</font></li>
</ul>
<p><font face="DejaVu Sans">I would like to try a window run of the
Beta model to get a higher resolution than 2km. But there is
more basic housekeeping that has to be fixed first.</font><br>
</p>
<p><font face="DejaVu Sans">Please keep us updated with your
observations.</font></p>
<p><font face="DejaVu Sans">Thanks!</font></p>
<p><font face="DejaVu Sans">Ian</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<p><font face="DejaVu Sans">I am not sure if that would explain your
discrepancies?</font></p>
<p><font face="DejaVu Sans">Regards</font></p>
<p><font face="DejaVu Sans">Ian</font></p>
<p><font face="DejaVu Sans"><br>
</font></p>
<div class="moz-cite-prefix">On 7/5/25 11:31, Ria Moothilal wrote:<br>
</div>
<blockquote type="cite"
cite="mid:CAPKYDunovyFJi31Q5UyKpgRVBjEC3U7FifnGhKW=aorAXkrytw@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">
<div>Hi <br>
<br>
I'm giving feedback as requested. Today, the RASP Beta was
wrong again, and interestingly, the old RASP 4km model was
more accurate. Below are the forecasts for today, the 5th
July at 11 am, as well as the actuals for Signal Hill taken by
3 live weather stations. Both the 1.3km and 4km old RASP
models show that the north-west wind slows down before it
reaches Signal Hill / Table Mountain. The RASP Beta 2km does
not. This suggests that it's not just a resolution difference
on the RASP Beta, but also either a source data or model
parameter change that has taken place. The old RASP was
accurate when compared to actuals, but the RASP Beta was
completely wrong <br>
<br>
The old RASP was the most accurate weather forecast for the
Western Cape. Nothing came close. I think it would be sensible
to merely recreate these same models on the new user interface<br>
<br>
1. RASP Beta 2km </div>
<div><img src="cid:part1.7DZLfS2a.wwSEeedr@zomerlust.org"
alt="image.png" class="" width="472" height="374"></div>
<div><br>
2. Old RASP 4km<br>
<img src="cid:part2.xYbrfnwJ.n4Pz5Muv@zomerlust.org"
alt="image.png" class="" width="472" height="385"><br>
3. Old RASP 1.3km<br>
<img src="cid:part3.AhME0PKA.19w2joGw@zomerlust.org"
alt="image.png" class="" width="472" height="463"><br>
4. Actuals from weather stations at Signal Hill and Sea Point:<br>
<img src="cid:part4.dN7MRcN0.qJlNljAP@zomerlust.org"
alt="image.png" class="" width="472" height="254"><br>
<br>
</div>
<div>
<div dir="ltr" class="gmail_signature"
data-smartmail="gmail_signature">
<div dir="ltr">Ria Moothilal<br>
Chief Flight Instructor<br>
+27 78 095 8136<br>
<a href="http://www.airschool.co.za" target="_blank"
moz-do-not-send="true">www.airschool.co.za</a></div>
</div>
</div>
</div>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
</blockquote>
</body>
</html>