Fall+Banter

[Baroclinic Zone]

Looks like 1.65".

[moneypitmike]

Man, the Pats are really stinking up the joint today.  At least it's not drizzling so I can make better use of my time and stack wood.

 

53*

[beavis1729]

Interesting article and subsequent comments on ASOS and "human adjustments".  As the author mentioned, it's a bit unsettling as it relates to the integrity of the long-term climate record.  This article is focused on Alaska...but I wonder how common this issue is around the country.

 

http://ak-wx.blogspot.com/2016/10/climate-observations-human-element.html#comment-form

 

[ORH_wxman]

46 minutes ago, beavis1729 said:

Interesting article and subsequent comments on ASOS and "human adjustments".  As the author mentioned, it's a bit unsettling as it relates to the integrity of the long-term climate record.  This article is focused on Alaska...but I wonder how common this issue is around the country.

 

http://ak-wx.blogspot.com/2016/10/climate-observations-human-element.html#comment-form

 

 

The more one looks at climate records for an individual station, the more one will uncover that our accuracy is far less than is often stated or claimed (hell, just look at the all the ASOS failures we uncover in our own region on an annual basis). This will tend to matter most just for individual records and such (we also have threaded records at many first order sites which is another mess). For the larger scale climate monitoring, it will tend to have less impact since those temperatures go through rigorous QC in the USHCN network before they are used for such research. We also have a less known network of temperatures called "USCRN" which started in 2004 and consists only of stations with pristine siting...with redundant instrumentation in place as a protection against instrumentation failure. These won't need any adjustments at all, but of course the only problem with them is that with such a short record, they aren't all that useful yet for longer term climate monitoring. But another couple decades and they will be a fantastic dataset.

[snowman21]

44 minutes ago, ORH_wxman said:

 

The more one looks at climate records for an individual station, the more one will uncover that our accuracy is far less than is often stated or claimed (hell, just look at the all the ASOS failures we uncover in our own region on an annual basis). This will tend to matter most just for individual records and such (we also have threaded records at many first order sites which is another mess). For the larger scale climate monitoring, it will tend to have less impact since those temperatures go through rigorous QC in the USHCN network before they are used for such research. We also have a less known network of temperatures called "USCRN" which started in 2004 and consists only of stations with pristine siting...with redundant instrumentation in place as a protection against instrumentation failure. These won't need any adjustments at all, but of course the only problem with them is that with such a short record, they aren't all that useful yet for longer term climate monitoring. But another couple decades and they will be a fantastic dataset.

When you say accurate, do you mean representative? I'd imagine the instrumentation of the ASOS is the most accurate thing we have next to the CRN. Maybe it's not entirely representative of the local climate, but accuracy in the tenths of degrees should be good enough.

[dendrite]

You need a station with 100 yrs of data with the same siting and same instrumentation to make any confident climatological hypotheses from them. Unfortunately most of our stations have gone from Hg to digital, passive to aspirated shielding, and changes in the siting/geography. It's really apples and oranges comparing today to 100 years ago. There's plenty of other ways to know that the planet is warming...don't get me wrong because I'm not a denier. It's just that I don't think you can make a lot out of the "A" climo station trends.

[CoastalWx]

You are almost better off going by trend analysis vs actual raw values..unless you know the site is in good standing. I know some will scoff at the values of stations sometimes adjusted and the actual number changes from whatever time you were measuring until now....but the overall trend is there. I wonder if we are just better off taking the more conservative approach from the change in temperatures from year A to year B. At least we know we probably have more confidence with that, and you still have the underlying trend....which is likely warmer. You get on a slippery slope if you make policies based on armageddon scenarios.

[dendrite]

I just read through that blog post and left him an anonymous comment. The ASOS max/min is not the official 24hr high and low for that station. For instance, CON reports their METARs at :51 past the hour including their 24hr max/min ob. Our "midnight" ob occurs at 0451Z, however the climo day is really 5z-5z. On days with a late midnight low temp it's possible to drop another 1-2F between 0451Z and 0500Z.

The 0451Z-0451Z max/mins are usually the same as 0500Z-0500Z, but occasionally it makes a difference of 1F here and there.

[snowman21]

3 minutes ago, dendrite said:

I just read through that blog post and left him an anonymous comment. The ASOS max/min is not the official 24hr high and low for that station. For instance, CON reports their METARs at :51 past the hour including their 24hr max/min ob. Our "midnight" ob occurs at 0451Z, however the climo day is really 5z-5z. On days with a late midnight low temp it's possible to drop another 1-2F between 0451Z and 0500Z.

The 0451Z-0451Z max/mins are usually the same as 0500Z-0500Z, but occasionally it makes a difference of 1F here and there.

I was going to say the same thing, that the official numbers come from the SOD message which is transmitted at midnight local time, but his bigger point about Bettles is just weird because I can't imagine the contract observers messing with the temps like that. We can't even get them to measure snow, but apparently in Alaska they are concerned about the low temperatures at an airport with a few dozen flights per day?

[ORH_wxman]

53 minutes ago, snowman21 said:

When you say accurate, do you mean representative? I'd imagine the instrumentation of the ASOS is the most accurate thing we have next to the CRN. Maybe it's not entirely representative of the local climate, but accuracy in the tenths of degrees should be good enough.

 

When ASOS stations are working correctly, they are definitely the cream of the crop, but we see failures on them too often to pretend that we're getting constant "nearest tenth degree" accuracy. Hell, all the ones we identify as drifting are usually by 1-2F before we notice it....how many have drifted by a few tenths of a degree without us noticing? Probably not many. The errors will probably tend to even out over time, so taking a collection of 30 or 40 ASOS stations in a region will give you very good data for climate analysis, but if you are looking at a specific station, you might have more incorrect data than you think.

 

As an example, I wonder how many people a few years from now who look at BGR's summer temps this year will know that it had drifted too warm by 2-3F?

[snowman21]

2 minutes ago, ORH_wxman said:

 

When ASOS stations are working correctly, they are definitely the cream of the crop, but we see failures on them too often to pretend that we're getting constant "nearest tenth degree" accuracy. Hell, all the ones we identify as drifting are usually by 1-2F before we notice it....how many have drifted by a few tenths of a degree without us noticing? Probably not many. The errors will probably tend to even out over time, so taking a collection of 30 or 40 ASOS stations in a region will give you very good data for climate analysis, but if you are looking at a specific station, you might have more incorrect data than you think.

 

As an example, I wonder how many people a few years from now who look at BGR's summer temps this year will know that it had drifted too warm by 2-3F?

I will admit I wouldn't have given BGR's temps a second thought if not for the folks on here pointing it out. I'm more curious as to how that happens. Poor maintenance? Broken aspirator fan (the fan can signal it's broken, so I'd be surprised if this doesn't get addressed quickly)? Defective sensor? Laying pavement beneath the equipment?

[CoastalWx]

Just now, snowman21 said:

I will admit I wouldn't have given BGR's temps a second thought if not for the folks on here pointing it out. I'm more curious as to how that happens. Poor maintenance? Broken aspirator fan (the fan can signal it's broken, so I'd be surprised if this doesn't get addressed quickly)? Defective sensor? Laying pavement beneath the equipment?

It's definiely good that people notice. If Will is right...hopefully things like that are corrected when dealing with the long term regional record. If you think temps are bad...look at snowfall..lol. And that gets a lot of play too. 

[ORH_wxman]

1 minute ago, CoastalWx said:

It's definiely good that people notice. If Will is right...hopefully things like that are corrected when dealing with the long term regional record. If you think temps are bad...look at snowfall..lol. And that gets a lot of play too. 

 

Something like BGR would get diluted in the USHCN record because it is an outlier...but if BGR was not surrounded by many reliable coops or other USHCN stations, then it could actually cause other nearby stations to get adjusted in the wrong direction during the QC process because they would weight BGR pretty highly given it is a first order site. So there are scenarios where it could be an issue...but on the whole, I wouldn't think it would be a big deal. Usually it's in an area with sparse data like Greenland or something where it could be a big problem.

 

31 minutes ago, snowman21 said:

I will admit I wouldn't have given BGR's temps a second thought if not for the folks on here pointing it out. I'm more curious as to how that happens. Poor maintenance? Broken aspirator fan (the fan can signal it's broken, so I'd be surprised if this doesn't get addressed quickly)? Defective sensor? Laying pavement beneath the equipment?

 

I'm sure all of these can be factors. Dendrite would probably know more than me about the mechanics of the weather stations...I'm more an expert on how they are integrated into the climate record and what types of QC are done.

[HIPPYVALLEY]

Nice day out here, mid-60's partly sunny skies so far. 

[HIPPYVALLEY]

.10"  so far in tonight's showers.  

[Ginx snewx]

4 minutes ago, HIPPYVALLEY said:

.10"  so far in tonight's showers.  

.42 with thunder

[HIPPYVALLEY]

Just now, Ginx snewx said:

.42 with thunder

You are well ahead of the curve on moisture compared to parts of NE.  No complaints here though.  Rivers quite low but we have not had a bad 8 weeks here.  Farmers markets have been absolutely flush with gorgeous produce! 

[OSUmetstud]

first flakes tonight.  Pretty early for here, actually. 

[Hoth]

26 minutes ago, OSUmetstud said:

first flakes tonight.  Pretty early for here, actually. 

Congrats!

[tamarack]

Had 0.05" overnight.  It's now rained each of the 4 days this month, and the total precip is 0.17", less than 1/4 of my avg.  (Oct. avg total is 5.49", or about 0.18" per day.)  Drip by drip, we get drier.

[weatherwiz]

Does this seem correct?  

[dendrite]

You need the V component of the SE wind. The U component provides no advection.

[weatherwiz]

31 minutes ago, dendrite said:

You need the V component of the SE wind. The U component provides no advection.

How would I find that?  Do I have to use trig or something? 

[CapturedNature]

Just got back from my cross country trek selling syrup at some western farmers markets.  We got to see some interesting weather along the way but mostly had great weather.  The first interesting event was the heavy rain we got in the high desert country of Utah/Nevada and Idaho from a remnant tropical system.  Some places received their first rains in months and got over 3".  My son looks at me and says "I thought it didn't rain in the desert?"...lol.  As that system lifted north it dropped several feet of snow over portions of Wyoming and we got to see that as we came home.  It was awesome driving across a snow covered landscape with banks along the road.

Now, if we could just bring some of the rain and snow back with us....

It's nice to see the foliage coming along so well.  I saw that I got down to 33° last week so I'm sure that helped.

[ORH_wxman]

2 hours ago, weatherwiz said:

How would I find that?  Do I have to use trig or something? 

 

Yes. Though if the wind is straight out of the southeast, then the U and V component will be the same...so it's like a 45/45/90 triangle...so just divide the hypotenuse by the square root of 2 to get the U and V values....so 20 divided by the square root of 2, which is about 14.14 knots.

[weatherwiz]

Does this seem 

2 hours ago, ORH_wxman said:

 

Yes. Though if the wind is straight out of the southeast, then the U and V component will be the same...so it's like a 45/45/90 triangle...so just divide the hypotenuse by the square root of 2 to get the U and V values....so 20 divided by the square root of 2, which is about 14.14 knots.

Thank you!  We never really went over how to do this in class...basically just asked if he knew how to do trig and that was it 

[dendrite]

Will's right, but if you have something like a 160 degree wind it's important to break it down with trig. Make a triangle out of it. You have your hypotenuse (total wind of both u/v components) and your angle via the direction. The total wind vector points from the origin into the 2nd quadrant (SE wind is -u and +v). Use SOHCAHTOA to find u and v. 

[Baroclinic Zone]

Man, I haven't done that math in 20+ years.  Don't miss that one bit.  Good luck with that Paul.

[dendrite]

Eek...I found you a new greenhouse.

http://nh.craigslist.org/grd/5813401011.html

[eekuasepinniW]

1 hour ago, dendrite said:

Eek...I found you a new greenhouse.

http://nh.craigslist.org/grd/5813401011.html

I crunched some numbers to estimate what sort of heating capacity would be needed on a worst case scenario night and got 438,000btu.

I'll pass.