Rainfall data - gauge or AWS

Times of day to observe, manual and electronic recording, display of real-time or recent data on your website

Rainfall data - gauge or AWS

Postby Scott Whitehead » 2013 May 07, 16:07

This may sound a stupid question but how many observers use rainfall data collected using a manual gauge when submitting monthly data to COL - as opposed to their AWS figure. For the record I find my AWS under-reads the manual gauge by an average 14%. It is also unreliable in snowfall, very heavy rain and rain falling in very windy conditions
Scott Whitehead
Wanstead, East London - 18m N 51° 33' 52" E 00° 02' 25"
Scott Whitehead
Posts: 21
Joined: 2013 May 03, 15:02
Location: Wanstead, East London - N 51° 33' 52" E 00° 02' 25"

Re: Rainfall data - gauge or AWS

Postby Len Wood » 2013 May 08, 22:41

You are correct Scott on all your suspicions of AWS rain gauges measuring less than the standard 5 inch manual gauge.

Quite a few people have done comparisons.
http://groups.google.com/group/uk.sci.w ... f54ec258f9

I only use a 5 inch standard gauge.

Wembury, SW Devon coast
N50.33 W4.13
Altitude 83 m asl
Len Wood
Posts: 349
Joined: 2012 Jan 28, 16:12
Location: Wembury, coastal SW Devon, 83 m asl

Re: Rainfall data - gauge or AWS

Postby Richard Hunt » 2013 May 09, 08:08

Same here, I only use data from a manual gauge and never from my AWS tipping bucket.
I use the tipping gauge to indicate when it rained, and the rain rate, but thats all.
COL Station: 25020.

Watson W-8681 AWS. Standard Stevenson Screen. Maximum and Minimum screen thermometers. Grass minimum thermometer.
5" Copper standard rain gauge. Roof mounted annemometer.
Richard Hunt
Posts: 374
Joined: 2012 Jan 27, 19:23
Location: Hornsea, East Yorkshire. 10m AMSL

Re: Rainfall data - gauge or AWS

Postby Darren Rogers » 2013 May 09, 19:15

What type of AWS?

If a Davis VPro2 (read Stephen Burts appraisal/study of this) then obviously the mouth of the gauge is approx' 5ft of the ground to start, so it should always under record anyway.

In the VPro2 there are x2 'adjustment bolts' (take the rain gauge cone off and you will see them) which you can change, but obviously make sure that you tur it the right way.

Philip Eden once told me that 10-15% was not uncommon in the difference but over the last few years I have done a little 'study' to compare te two - that is using a standard copper gauge - VPro 2 Mn-Mn readings - but also the VPro 2 0900-0900 readings (this I have to calculate myself as weatherlink does not allow for it) and I have found the following, this is using a Aug-July calendar:

Year 1 = 7.98 %
Year 2 = 7.91%
Year 3 = is the current year and stands at 7.89%

Yet at some points the difference as been as high as 15% before falling back.

You would think that wind would effect the readings yet in year one the windiest month of the year saw the VPro2 match the gauge.
Also sometimes on a fall of say one inch the difference might just be 0.2 mm yet on a fall of 3-4 mm the diff' could be 1.0mm - !!! ???

The main difference though is in the number of rain days - this differs widely, upto 25-30 extra days per year on the VPro2.

basically if you can get a standard gauge - have both, but quote the standard.
Darren Rogers
Maulds Meaburn
Cumbria (Half way between Shap and Appleby)

Darren Rogers
Posts: 145
Joined: 2011 Nov 27, 20:22
Location: Cumbria - half way between Shap and Appleby in the Eden valley

Re: Rainfall data - gauge or AWS

Postby Stephen Burt » 2013 May 11, 18:42

You can find the results of my year-long comparison trial between a Davis Vantage Pro2 and conventional instruments - including a standard 'five-inch' raingauge - on this link:


There's also a more detailed explanation of some of the differences between gauge types in my recent book.

Bottom line is, as the others have rightly said, that quoted rainfall amounts should always be from a standard manual raingauge at the same site where one is available, and never from an AWS alone. If no manual gauge is available, the site description should be shown as 'C2' in the bulletin to make this clear, because (as others have already pointed out) the differences can be both substantial and variable.

On a similar note, over the last few months I have been conducting comparison trials on the (much cheaper) plastic CoCoRaHS raingauge as widely used in the USA and now available in Europe (see, for example, http://www.weatherstations.co.uk/cocorahs.htm ) and the Davis Vantage Vue2 AWS, and expect to publish preliminary results shortly. The full analysis of the Davis Vantage Vue AWS over 12 months will be presented at the Amateur Meteorologists Conference in September.

Stratfield Mortimer, Berkshire - central southern England
51.4°N, 1.0°W, 60 m AMSL, station grade A - AAAA47R
Records commenced here 1987 - local records available back to 1862
The Weather Observer's Handbook: www.measuringtheweather.com
Stephen Burt
Posts: 247
Joined: 2011 Dec 02, 19:36
Location: Stratfield Mortimer, Berkshire; a well-exposed rural site, 10 km SW of Reading

Re: Rainfall data - gauge or AWS

Postby Nick Gardner » 2014 May 27, 18:55

Darren Rogers wrote:If a Davis VPro2 (read Stephen Burts appraisal/study of this) then obviously the mouth of the gauge is approx' 5ft of the ground to start, so it should always under record anyway.

I've separated the various 'units' of the VP2 Pro Plus. The rainfall TBR sits near to my standard copper gauge with the rim at the same height (30 cm); yet it still under-records by around 10%!

The other components:
Anemometer is at 10 metres height
The UV & solar on the same mast but about 1 metre below the anemometer.
The thermometer/humidity sensor is in a Stevenson Screen.
Near Newton Poppleford, Sidmouth - Devon
50:41N 3:17 W; 20 metres AMSL; Station Grade B-BAAA37
Nick Gardner
Posts: 429
Joined: 2012 Jan 27, 17:25
Location: Rural riverside valley location near the village of Newton Poppleford and 2 km from the sea.

Return to Procedures

Who is online

Users browsing this forum: No registered users and 1 guest