Notes:

• The weather station, such as it is, is a raspberry pi connected to a variety of sensors on a breadboard all inside a tupperware on my balcony facing the hof behind our apartment.
• The "internal sensors" in this setup are therefore in the tupperware, the "external sensors" are the HTU21D breakdout board attached via an old phone cord outside the tupperware near the edge of the balcony.
• To make the numbers on the particle graphs more meaningful to more people, we've started displaying our data in nGy/h instead of cpm. The datasheet for the LND 712 (linked above) specifies "Gamma sensitivity Co60 (CPS/MR/HR): 18". We take this to suggest that 18 counts/s = 1 mR/h (for gamma radiation). I collect data by counting per minute, therefore we divide the collected number of counts counted every minute by 1080 (60s/m * 18 c/s/mR/h). We then multiply by 10 to go from mR/h to uGy/h. Thus dividing our collected cpm data by 108 should yield uGy/h. Because 1uGy/h is a large does rate relative to our background radiation, we then multiply again by 1000 to put everything in nanograys/hour (nGy/h), which is why in all our code you see our collected data multiplied by 1000/108. If anyone finds this to be in error, please let me know. While the graphs are displayed in nGy/h the raw collected data remains unconverted in counts per minute.
• Gy are however a unit of dose, and I can't actually determine dose w.o. knowing the energy of the incoming particles. Since i've calibrated the dose measurements to Co60 (cuz that's the only data in the datasheet) we expect the dose rate to read kinda high, and that's exactly what we see when we compare my data to the german gov measurements.
• This geiger tube is also somewhat unique in that it's an alpha, beta, and gamma tube, and my calibration is only against a gamma source.