Cooper Gauge Question
Printed From: ProfessorPaddle.com
Category: Site Support
Forum Name: Run Submissions & Gauge Discussions
Forum Discription: Submit a Run Description or Run Beta. Discuss Gauges or post messages for the Rio Banditos
URL: http://www.professorpaddle.com/forum/forum_posts.asp?TID=9844
Printed Date: 27 Apr 2024 at 7:11am
Topic: Cooper Gauge Question
Posted By: arnobarno
Subject: Cooper Gauge Question
Date Posted: 09 Jul 2010 at 1:16pm
|
Why do we report the Cooper from this report on the Cle Elum:
http://www.usbr.gov/pn/hydromet/yakima/yakstats.html instead of from NWRFC which is updated more frequently (4x daily) and also has a forecast: http://www.nwrfc.noaa.gov/river/station/flowplot/flowplot.cgi?CLUW1 ------------- arn9schaeffer@gmail.com (remove 9 for my real email address) |
Replies:
Posted By: James
Date Posted: 09 Jul 2010 at 1:46pm
|
If you are offering to write an image to data parsing tool, I will jump on your offer kindly!!!! The data is still inaccurate as the NWRFC mostly uses USGS and other locality gauges and most of the time does not operate gauges on their own. In many cases they will post past data based on forcasted trends until a gauge updates in which case you can see the big graph chiggers and resets. But still I would love to get the Image to Txt parser if you can help me there  |
Posted By: arnobarno
Date Posted: 09 Jul 2010 at 1:56pm
|
James,
Can't you just use the data from here - http://www.nwrfc.noaa.gov/river/station/flowplot/text/CLUW1.QI.0.0.1.0.txt So, you just have to parse the text, not any image. And it looks accurate if you compare the 4AM and 10AM data to the 6:40 AM data from BuRec. BTW - Dreamflows seems to use the NWRFC (or USGS) data, not the BuRec data. Maybe they are just better programmers :-) :-) :-) ------------- arn9schaeffer@gmail.com (remove 9 for my real email address) |
Posted By: James
Date Posted: 09 Jul 2010 at 2:32pm
|
I have never seen that txt plot... Me likey!!! USGS does not have a gauge for the Cle Elum so they are probably using the NWRFC. I will add that into the todo list but you won't see it replace the gauge until I can confirm that they either have direct access to the gauge stream for more updated data which is unlikely or a separate gauge on the river. Sure they might be accurate right now but there has not been any weather changes from what is forcasted. You can always access the NWRFC with a simple click but I don't think it is in our best interest to post theoretical data as observed data which is what the NWRFC does until a source is available to revise or make a correction. What I am saying is that when a gauge only posts every 4 hours NWRFC will add the forcasted amounts and report it as ovbserved until a new true observation rebuilds the forcast and graphs, that is why often they show big changes or drops in the graph and sometimes complete blanks. I think it is better to air on the side of caution and allow the paddler to obtain his forcasted data seperate and ensure what he gets from the USGS, PUD etc. which we mirror is indeed true gauge readings. Posting up the NWRFC when they are incorrect could cause people to plan on rivers that are not going to be runnable or could be way to high. This happens to the NWRFC all the time not during heat spells that go exactly to plan, more during rain events, fall storms and the occasional hot winter day). I have seen the NWRFC report thunder creek holding at 1000 which is runnable when over the period of 2 hours it jumped into the 1900 range, the same is true with Robe to. Again this is just how we have done it... if folks want it done differently it can always change. So the above is just our history and a bit of my opinion. As for Dreamflows, they are not just better programmers and paddlers, they also have a more refined social skill set and greater diversity of cleavage... in their mountain ranges  |
Posted By: arnobarno
Date Posted: 09 Jul 2010 at 2:36pm
|
Well, actually this is much more interesting than I thought (at least to me). In researching this a bit further, there is no gauge on the inflow - it is a computed value. BuRec knows the amount they are releasing per hour and they know what the current amount of storage in the lake is. Based on this information, they compute a daily average "inflow" and that computed value is what is published in the morning.
So, if you want to get an hourly flow, it is actually relatively easy (at least for a computer program). You can look at how much water they released in the last hour and compare it to what happened to the reservoir storage. So, for example, in the last 3 hours (I picked 3 hours to smooth the data a little), the reservoir dropped 284 acre feet. An outflow at an average flow the past three hours of 2550 cfs tells us the reservoir should have dropped 632 acre feet if there was no inflow (1000 cfs = 82.65 acre feet in an hour). But, the reservoir only dropped 284 acre feet. Therefore, the inflow is 348 acre feet. Converting back to cfs yields 1404 cfs as an hourly flow the past three hours. Here is where the hourly data is on flow: http://www.usbr.gov/pn-bin/yak/load_file.pl/cle__q__100709142408.html?src=cle__q__100709142408.html And, here is reservoir storage amount: http://www.usbr.gov/pn-bin/yak/load_file.pl/cle__af_100709142220.html?src=cle__af_100709142220.html You'll notice that the reservoir storage amount looks a little sticky so I suspect that you could use this calculation over a rolling 3-6 hour period and get reasonable data but if you just looked at the past hour it may be suspect (btw, if you use a 6 hour average, it was 1444 cfs over the past 6 hours to 2PM). Now then, none of this tells you the amount of water in the Cooper, just the Cle Elum. ------------- arn9schaeffer@gmail.com (remove 9 for my real email address) |
Posted By: James
Date Posted: 09 Jul 2010 at 2:47pm
|
I am digging the method of extrapolation your using which seems more accurate than even the NWRFC model. That is however only so accurate because it is posting every 15 minutes If it were hourly or every 4 hours which most gauges are it would not be so accurate. I am pretty sure they use the inflow data posted every 24 hours which is why your calculation is much more accurate than theirs. Again tomorrow morning you will see more accurate details as they post up observed amounts and revisions. What your suggesting now is not simply using the NWRFC but a new method, which again we can do, and it might be more accurate for the cooper than just the inflow, but I would personally prefer to either have your method or a 24 hour inflow reading than a hippity hoppity graph from NWRFC |
Posted By: James
Date Posted: 09 Jul 2010 at 2:53pm
|
You can ask JP, Chris or Oliver about that run and the NWRFC, We sat out there a few weeks back and watched it slowly climb while they showed it tanking until the next morning, so I am pretty sure that it is a 24 hour update they base the data on along with weather models not the method your using. More fickle weather will answer that question in terms of time before revisions are made to forcasting and observed flows. Now then, please don't sent this thread to Dreamflows or they will use your new method and I'll have no hope of ever catching them!!!! |
Posted By: H2Ohta
Date Posted: 09 Jul 2010 at 2:59pm
|
Just go paddle the Cooper!!! If there is enough water to get down, it's fun... ------------- H2Ohta |
Posted By: James
Date Posted: 09 Jul 2010 at 3:06pm
|
But what happens when there is too much water.... Would that make it too much fun? |
Posted By: H2Ohta
Date Posted: 09 Jul 2010 at 4:10pm
|
Sometimes there is just too much fun to be had... ------------- H2Ohta |
Posted By: arnobarno
Date Posted: 09 Jul 2010 at 4:23pm
|
Hahaha. Just paddle, it all goes. Well, it is going to be my first time down when I do it, so getting on the low side is where I'm at.
But, back to our math problem... I have no idea what the NWRFC is doing. There must not be a gauge on the inflow because if there was, surely the BuRec would be using that rather than doing some funky calculation. Now then, I can tell you with certainty that what BuRec publishes at 6:30 AM is the average computed inflow from the previous day - not the previous 24 hours even. I just used my methodology and checked it against their number, and, sure enough, I got the same answer as them if I begin at midnight Wednesday night through midnight Thursday night to produce a number for a 6:30 AM report Friday morning. So, when you go paddle on Friday at 5PM, for example, the flow you see on the website is for events that happened between 17 and 41 hours ago. PRETTY USELESS. You can use my method and compute an average inflow for any period you want. I'd use the last 3-4 hours at a minimum though as the calculation for acre feet in the reservoir is based on the "forebay" elevation height and that is measured (only) to the hundredths of a foot. So, that is what is producing the apparent stickiness in the number of acre feet. Just as an FYI, for the 15.5 hours of data today, we've averaged 1430 cfs an hour (though that is still dropping somewhat before it will probably rise again in its 24 hour cycle). ------------- arn9schaeffer@gmail.com (remove 9 for my real email address) |
Posted By: arnobarno
Date Posted: 10 Jul 2010 at 9:02am
|
Hey - for anyone interested - the actual inflow the past three hours has been 1480 cfs. Not sure why BuRec doesn't even have a report this morning... ------------- arn9schaeffer@gmail.com (remove 9 for my real email address) |
Posted By: water wacko
Date Posted: 11 Jul 2010 at 7:12pm
| Arn, that run holds water really well. I was surprised as i ran it higher and higher each year. It just keeps getting better!! Up to a point. Above 1700 or so the river starts changing a little... |
Posted By: H2Ohta
Date Posted: 11 Jul 2010 at 7:20pm
|
Big fun today!!! Fluffy and boofalicious... ------------- H2Ohta |