I'd like to look into the possibility of using a Raspberry Pi and the PiCam module to detect rain, and to try to make some measurements of droplet size and rate. This could be done using an LED flash or a (modified/pulsed) laser pointer equipped with a fan-out element, synchronizing the flash/pulse with the Pi camera's electronic shutter for a "freeze frame" effect, followed by some Python image analysis (e.g. PIL).

This is for fun, and not meant to be a quantitative rain gauge necessarily.

In order to better estimate the challenge, I'd like to get some feel for the distribution of sizes, speeds, and either areal rates (drops per sec per square meter) or number densities (drops per cubic meter). I can convert between various units and histograms, but I don't know where to find a good survey to understand over what ranges these can vary.

See also Will long-term viewing of a sunny sky hurt the Pi Camera? for pics of a cool Pi sky camera by @ThomasJacquin as described here.

Edit: per this comment I should point out my Raspberry Pi will be within a few meters of the Earth's surface, where I can keep an eye on it.

  • $\begingroup$ check out the work done under the TRMM project. There are some great references on this. $\endgroup$
    – gansub
    Mar 26 '18 at 8:00
  • 1
    $\begingroup$ in your question you should mention surface rain drops if that is the case. $\endgroup$
    – gansub
    Mar 26 '18 at 8:29
  • 1
    $\begingroup$ references on surface radars should be a good way to start on this. $\endgroup$
    – gansub
    Mar 26 '18 at 9:22
  • 2
    $\begingroup$ I highly suggest Chapter 2.6 and 2.7 of "Physics and Dynamics of Clouds and Precipitation" - Pao K. Wang. If I find some time I may do a write up. While this does not answer all of your questions, you will find typical dropsize distributions and will be introduced to some of the major problems arising when finding the quantities you specified. $\endgroup$
    – J. Fregin
    Jul 26 at 11:51
  • 1
    $\begingroup$ Some frozen raindrops get... REALLY big (ok, that's likely helped by some collision\agglomerating... of course so is rain) Let's see how the Pi sensor handles that :-p $\endgroup$ Jul 26 at 12:55

I listed Harnessing of Kinetic Energy of Raindrops as a reference when I answered one of your recent questions, Ways to make a “How hard is it raining?” detector for personal use?.

Serendipitously, that reference has information relevant to this question.

The size and velocities of rain drops varies according to the type of rainfall event.

For Light Stratiform Rain, light rain had drops 0.5 mm in size (diameter) and a terminal velocity of 2.06 m/s. The size of large drops is 2.0 mm and their terminal velocity is 6.49 m/s.

For Moderate Stratiform Rain, light rain had drops 1.0 mm in size and a terminal velocity of 4.03 m/s. The size of large drops is 2.6 mm and their terminal velocity is 7.57 m/s.

For Heavy Thundeshowers, light rain had drops 1.2 mm in size and a terminal velocity of 4.64 m/s. The size of large drops is 4.0 mm and their terminal velocity is 8.83 m/s.

When rain drops start to get to be larger than 4.5 mm in size they split in two, as illustrated by this diagram, which relates rain drop shape to its size. Below 1 mm in diameter, rain drops "are almost spherical".

enter image description here


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.