import historic cumulus data into cumulus

[skmorgan1944]

For several reasons including new weather stations, I have 2 historic cumulus data sets. Naively I had hoped that by simply copying the old data sets into the new Cumulus data folder (eg copy Apr11log.txt). Unfortunately Cumulus doesn't seem to recognise these additional data sets.

I would appreciate some guidance on how to merger these old data sets as I want to get monthly average temperatures for the past few years.

[steve]

I copy other people's data files into my test installation all the time, and they work fine. What exactly are you doing after copying the files that makes you think they're not being picked up? Cumulus does very little with old monthly logs anyway, apart from Select-A-Graph.

[skmorgan1944]

I am trying to obtain Average Temperature for each month using View/This Period then selecting 1st day then last day of the month in question.

I have added the month log files from my historical data to the Cumulus/Data folder. However the View/This period option doesn't show any data for the imported log files. Clearly I am doing something wrong.

I do note that when I try Select-a-graph, the data are shown as you suggest so Cumulus is recognising the month log in this case.

Your advice would be most welcome.

[skmorgan1944]

I have just noted that a more appropriate way of obtaining monthly Av Temp would be to use the View/This Month option however this also doesn't show any data.

[steve]

View This Period etc use data from dayfile.txt for almost everything. You can use the dayfile.txt editor to create missing entries in that file, from the monthly log data.

[skmorgan1944]

Thanks for this info. I have appended old dayfile.txt files (backed up original first!!) using Notepad++ and this seems to work. I now know why I changed from La Crosse WS2300 to Fine Offset (aka Maplin's special) as there are numerous spurious readings in the Temperature data.

I will probably have to copy the whole file into Excel then write a routine to look for Temperature values that deviate by more than a certain amount from adjacent values and replace with a mean value. I can foresee difficulties when there are several consecutive spurious readings. I will need to think about this a bit. It would be great if I could manually adjust a graphical view as the spurious readings are very easy to spot. Even more thought necessary

Anyway thank you for putting me straight about location of data.

[steve]

Yes, lots of old bad data is tedious to correct. I even had that problem with my Davis when the temperature sensor went faulty (again) during the summer. I still have some bad data lying around. It's better to try to stop the values being recorded in the first place using the spike removal settings; but it's not perfect.