Database Access and Analysis
On this page, you can upload detector data to the MusIC database using your user credentials provided during the outreach session. Or, without needing credentials, you can download anyone’s detector data to use to create your own analysis. You can also select a detector and quickly view a snapshot of the hits it has recorded with our pre-made figures.
Upload your Cosmic Watch data here:
Enter your username and password for database access and then click "Upload". Upload the data files exactly as your detector wrote them; the detector's name is read from the file's "Detector:" line. A file that overlaps data already uploaded for the same detector is refused, so nothing is counted twice. For large files (e.g. multiple days' worth of data) this may take a moment.
Download Cosmic Watch data here:
Download all the hits recorded by any detector as one file, in time order and in the same format the detector writes. It is meant for your own analysis rather than for uploading back: once a detector has sent in more than one file, the merged download cannot be uploaded again, because the detector starts numbering its packets afresh in every file it writes. Columns are: Date and Time (when the hit happened, in UTC), Flag (1 if both detectors saw it), ADC (SiPM signal in raw 12-bit ADC counts), SiPM (SiPM signal in mV), Deadtime (total time the detector has spent busy since it was switched on, in seconds), Temp (temperature in Celsius), Press (air pressure in pascals) and Packet (the number of the packet the hit belongs to — see below).
Plot Detector Data
Three graphs are provided: hit rate vs. time, pulse strength distribution, and hit rate vs. temperature.
The detector records hits in numbered packets of ten. A packet tells us the detector was definitely switched on and recording from its first hit to its last, so rates are worked out only from those stretches — the detector’s live time. Between packets, and between runs, we can’t tell a quiet sky from a detector that was switched off, so those times are left out, and the rate plot shows a break wherever there is no data. The detector also throws away packets whose rate is implausibly high, such as when it is being moved, so that time is left out too. Time spent processing each hit (the dead time) is also subtracted, since the detector can’t see a new muon while it is busy.
In the hit rate vs. time plot, notice the difference between coincident frequency and single detector frequency. What does this tell you about how the detector is affected by incoming particles that may not actually be atmospheric muons? Also notice any distinct increases or decreases in frequency from the mean. What could cause this?
In the pulse strength plot, notice where there is a bump in the overall pattern of exponential decrease. What do you think this tells us about where muons hide in the radiation background?
In the hit rate vs. temperature plot, do you see a correlation between hit frequency and temperature? If no, why do you think that is? If yes, can you explain why that might make sense? Think about how particles from space scatter off of particles in the atmosphere during a cosmic shower, and think about how hot and cold air change density.
You are encouraged to make your own plots! These graphs can only tell you so much and are made to be one-size-fits-all, but your data will be unique and tell its own story. Libraries like matplotlib in Python are excellent for making scientific figures, and are very well-documented for users and learners.
