Go to Home Page
   Limitations of the weather station

As with many things in life, they aren't always as simple as they appear to be. I have outlined below some of the limitations of the information displayed on this website.

The forecast displayed at the top of the Home Page takes into account only those conditions in the immediate vicinity of the weather station. The situation would probably be different in another area.

Measuring snowfall
You hear TV announcers stating things like, ‘4” of snow fell today’, but how do they know? Measuring snowfall accurately can be a challenging task as there are many things to take into account, e.g. wind, slopes, warm buildings etc.

A good deal has been written about measuring snowfall and there are a number of electronic devices that claim to measure snowfall accurately but whether they do I don’t know.

The surface you use as a ‘table’ must be flat and level and 20-30 feet away from buildings. Grass is not a good surface to use as snow lays on the top but the ruler will slide to the ground below.

You have to consider the frequency of the measurement especially if it is windy otherwise all your snow could be blown off the ‘table’ before you get a chance to measure it. If the air temperature is fluctuating the snow may start to melt and flatten down, over the course of a day it is possible for the snow to melt a bit or even completely and snow again several times so if you only take one daily measurement the chances are high you will understated the amount of snow that has fallen.

If you use a snowboard (flat board about 16” x 16”) it should be cleaned off after each measurement. Ideally you would use 2 snowboards, one you clean off after each measurement to give the amount of snow fallen over a period of time, the other you would leave to give the current snow depth. Snow depth is not the accumulation of each new snowfall measurement.

It is possible to melt the snow and then measure the amount of water produced. The rule of thumb: 10 centimetres of snow will produce one centimetre of water. You would, however, have to be careful that your container didn’t contain any other water, for example, it may snow for a while and then turn into sleet or rain which would distort your measurement.

The data for graphs are gathered at specific time intervals (sampling time) and so may not necessarily show extremes, e.g. a windspeed graph may indicate max. gust of 41 m.p.h. at 05:00 whereas the max. gust may have been 46 m.p.h. but this was at 05:02 which isn’t at a graph data gathering point. To display all the collected data would make the graphs too large. Even the sensors themselves have a sampling time, this varies depending upon the sensor, e.g. anemometer 2.5-3 secs., thermometer 10-12 secs.

Rainfall measurement isn’t straight forward either
Is it raining? You look out of the window and you see that it has just started, you then look on the weather website and it says dry. Why?

Rain will fall into a rain collector to be measured, the measuring is often done electronically and in my automatic weather station (AWS) it uses a tipping bucket system, see: Inside the ISS. Until there has been sufficient rain to fill a bucket to tipping point there would be no electronic indication of rain. It takes 2 bucket tips within 15 minutes to signal rain so it is possible for it to rain briefly and then stop without ever tipping the buckets even though the ground may be wet the AWS is still reporting no rain.

When there is an indication of rain there comes a point when it stops but the weather station could still be reporting rain. Once the rain has stopped the water is still trickling through so ‘tipping’ events can still occur thus registering ‘rain’.

Snow also has a side-effect on 'rainfall'. Snow will accumulate in the rain collector. If it is a heated rain collector (which mine isn't) the snow will begin to melt, this might be at a rate whereby it causes 2 tipping events within 15 minutes thus indicating rain. In an unheated rain collector, as the temperature rises and the snow inside begins to melt it could cause an indication of rain over a period of several hours when in fact there is none.

Windspeed - erroneous value
Very occasionally, for reasons unknown, a data spike may cause a windspeed reading to be shown as more than 79 m.p.h. This usually shows in the records as around 79.4 and should be ignored.


Any errors, comments or ideas for inclusion can be e-mailed to Peter

Copyright 2007 - Menstrie Weather
IMPORTANT - Do not use the data on this website for the protection of life and/or property