Author Topic: What's the philosophy and method to catch gusts?  (Read 305 times)

0 Members and 1 Guest are viewing this topic.

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1661
    • Weather at Eau Claire, WI
What's the philosophy and method to catch gusts?
« on: August 16, 2020, 02:54:25 PM »
Meteo Mark and I have had a discussion of using a Campbell Scientific datalogger to be a weather station, and while writing tables to internal memory, also interface with Weather Display.

One question he raised (as he's further along and thought more about this than I have) is how to pull data, either from the live table or the static type, with writing to it say once a minute.

For those who have been around thunderstorms, at least, there is a tendency to have, with first gusts, a high speed, which is far from sustained over time to be captured.  This is important since down trees, towers and roof damage may be caused by a gust, with sustained winds reading much lower.

My several stations (Peet Bros, Texas Weather Instruments, and Rainwise) all seem to send data fast enough that most gusts are caught, and indeed Weather Display for instance, will log the maximum gust of the day and last hour.

If one is rolling their own, with say Arduino or Rasp Pi or even programming a datalogger to grab readings once a minute or every 5 minutes or whatever (good enough for pressure, temps and humidity) what do you do to catch the gusts?

I'm thinking that for the data logger, one would read the critical instruments (wind speed) much more often in a loop, and then every minute or whatever the chosen interval, drop into a inner loop to record and log the more sedentary values like pressure, temp, humidity, solar, etc.

Is this what folks are doing with their home brew stuff?

Thanks. Dale
ECWx.info
&
ECWx.info/t/index.php

Offline Clarkey

  • Member
  • *
  • Posts: 7
Re: What's the philosophy and method to catch gusts?
« Reply #1 on: August 17, 2020, 08:37:24 PM »
Hi Dale,

You can run the logger measurement scan at 1 second or faster and poll the 'Public' table every second (fixed at 1 sec with WD). I used this method for some time but prefer the logger to do the work and record/timestamp these values for me with a later/slower retrieval. A CR1000 is capable of scanning it measurements in the mSec range ...

Clarkey

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1661
    • Weather at Eau Claire, WI
Re: What's the philosophy and method to catch gusts?
« Reply #2 on: August 17, 2020, 08:49:38 PM »
Thanks for the real-world experience report.
There is a lot about these dataloggers that I haven't grasped yet, and as I fiddle around I'm sure I'll pick a method.  Your experience makes sense and I'll look at doing that from the get go.  Thx.
ECWx.info
&
ECWx.info/t/index.php

Offline Aussie Susan

  • Senior Member
  • **
  • Posts: 62
Re: What's the philosophy and method to catch gusts?
« Reply #3 on: August 17, 2020, 10:47:17 PM »
I suspect that this is pretty much the same as any sort of sampling frequency - I'm most familiar with it in the electronics domain.
Basically, to reproduce a waveform (in this case the speed of the wind over time) you need to sample at twice the maximum frequency you want to measure. (This goes by several names by Nyquist Frequency and Shannon Sampling Theorem are common.)
If you want to measure a wind burst that can last for 5 seconds, then you need to measure at least once every 2.5 seconds.
Therefore the first thing to work out is 'what do you call a wind gust' and 'what is the shortest gust' you want to measure.
Of course if the device you use can't provide measurements that fast, then you need to turn this on its head: how fast can it measure and that will tell you the minimum wind gust that you can reliably measure.
Susan

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1661
    • Weather at Eau Claire, WI
Re: What's the philosophy and method to catch gusts?
« Reply #4 on: August 17, 2020, 11:05:51 PM »
Aussie Susan:  Thanks, as seems always there is already much more about this than most people would even want to know.

I'll look up some of those names you gave.

When one is looking at a high speed gust, there will be plenty of switch closures from a pulsed device.  It is the very slow rotation, and not a gust situation, that makes slow speed measurement a problem of its own.

Thanks again.  Dale
ECWx.info
&
ECWx.info/t/index.php

Offline RobertVA

  • Member
  • *
  • Posts: 2
Re: What's the philosophy and method to catch gusts?
« Reply #5 on: August 27, 2020, 09:49:22 AM »
Only relatively recently have we had the technology to really profile wind gusts.  This link provides a good overview of the topic.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5948875/

It is important to decide what you want to measure (peak wind speed, gust profile, gust energy, etc.) and what you want to do with the data (analyze for damage potential, statistical wind data, etc.) since both drive your sampling rate, processing requirements, and final data volumes.

Robert

Offline DaleReid

  • Forecaster
  • *****
  • Posts: 1661
    • Weather at Eau Claire, WI
Re: What's the philosophy and method to catch gusts?
« Reply #6 on: August 27, 2020, 12:41:04 PM »
Very nice.  A sort of history-of along with various degrees of scientific analysis.

I appreciate you forwarding this link.  I skimmed it and already know I'll be taking a closer look when I can think as I read and not just glance at it.

Now, if I could just find the source of electrical noise causing my one anemometer to suddenly show 115 mph for a few seconds when it is still out side.  Grrrr.
Dale
ECWx.info
&
ECWx.info/t/index.php