I hope you're not going to deny having told me that daily would suffice. You are looking for CLIMATIC data are you not? How fast do you think the climate changes?
I asked for raw data. I said raw data in a daily format as long as the time is noted is acceptable. The best we have so far is a daily max and daily min. While that's better, it's still data that has been processed and the methodology used to process that data is not open for verification.
I took part in a hydrology study, defining certain areas as wetlands as part of a thesis for someone else. In order for the study to be accepted and published, ALL data had to be submitted. We could not leave out data that we deemed irrelevant and we could not use averages, maximums, minimums, or summaries. The specifics (in this case air temperature, water temperature, soil composition, and many others I don't remember) were important to check for biases. Without a complete picture, regression analyses and cross-checks were not possible.
The reason it's important to have all the raw data for stations is that one cannot check for equipment errors, calculation errors, site errors, or environment biases without it. Actually a better way to put it is that no set of data summarization can be perfect, but more raw data increases the accuracy.