The first step in time series analysis is deciding which time interval to use when exploring the data. You can choose several, and it is good practice to do so when getting to know your IoT data. You can start with monthly trends, then weekly, then daily, and even to the minute.
For our analysis, we will continue with the daily interval. Start with the Qpcp measure and view it by station to graph a daily summed Qpcp value. Filter out the extreme values that are almost certainly not actual measurements (you can use a 0 to 900 range in the filter). We have already identified an unusually high record count for the Boulder 2 CO US station that occurred in September 2013.
You also learned that there is not a data ...
