All that remains for this section is to test the unit. The data returned is the date and time on every occasion that the bucket tips in one direction or the other. (Apologies to the American who read this, but the date format is day-month-year. To change this you will have to edit the code). Therefore using the analysis method set out previously we know the depth of rain that fell in a given time interval and hence the rainfall intensity over time. An example calculation and a graph of the data are given here.
When you open the csv file (comma separated file) in a spreadsheet or text editor so you view the content you will get rows of data like this,
Reedcount, 1, 02/05/2014, 08:43:24
Reedcount, 2, 02/05/2014, 10:36:12
Reedcount, 3, 02/05/2014, 10:46:43
Reedcount, 4, 02/05/2014, 10:50:02
Reedcount, 5, 02/05/2014, 10:56:12
Reedcount, 6, 02/05/2014, 11:02:17
Reedcount, 7, 02/05/2014, 11:17:34
Reedcount, 8, 02/05/2014, 11:27:23
Reedcount, 9, 02/05/2014, 13:06:59
The first and last readings happened when I knocked the bucket over putting the rain gauge out and taking it in. We can see that the second tip of bucket happened after 1hr 52 minutes and 48 seconds (1.88 hours), the third after 10 minutes and 31 seconds (0.175 hours) and the fourth after 3 minutes and 19 seconds (0.055 hours).
We know that the each tip of the bucket corresponds to a depth of rainfall of 0.5mm (diameter of rain gauge 11cm and volume of bucket = 4.75ml).
Therefore the rainfall intensity varied from 0.3 (0.5/1.88) to 2.9 (0.5/0.175) to 9.0 (0.5/0.055) mm/h between 08:43 and 10:50 am. (Unfortunately we don’t know when the rain actually started falling so the rainfall intensity estimate for the first period is a bit dodgy. It assume it was raining when we first put the rain gauge out - which it wasn’t).
If we complete this calculation for the entire period we get a graph of the rainfall intensity shown.
This data is an interesting data set but the steps in data are a little large for my liking. The size of the data steps is due to the size of funnel and the bucket. Secondly we don’t really know when the rainfall event actually starts and stops. This is because we don’t know when first started collecting in the buckets (we only know when it first tipped). Also we don’t know how much rain is left in the bucket when the rain stops.
At the moment we are just going to have to live with these limitations. We can improve the accuracy though.