Our full technical support staff does not monitor this forum. If you need assistance from a member of our staff, please submit your question from the Ask a Question page.


Log in or register to post/reply in the forum.

Numerical integration over rain data over time, gives different results when changing time window


MM81 May 12, 2017 05:19 PM

Hello.

I have SBS 500 rain gauge and a CR1000 to measure its reponse.

I am getting aquainted with the rain sensor, its working principle, and the data it outputs. To this aim, I am testing and overlapping reults obtained with different time windows set for variable Total_precipitation_mm. Inother words, the same variable is output at different intervals.

So far, my predecesor had set the timewindo at 10 minutes, but I think that for sporadic, intemrittent and intense showers (crucial for my study case) that was no good choice.

I have output 5 different timewindows. Then I have proceeded with numerical integration for each case, by doing:
Amount of water = sum (millimeters * timestep).

I expected the value be roughly the same, or at least, fluctuate no more than a 10%. To my greatest disappointment divergence spans over one whole order of magnitude. Also result obtained for dt=5s, equals almost results for dt=10min (approx same value). Then for dt=60 s (which is what I was expecting to be the best option) the result is, indeed, smaller by a factor 10.

Anyone already encountered this problem? And solved it?
Any feedback is greatly appreciated.

Kind regards
Mark

Log in or register to post/reply in the forum.