This also bothers me for your inherent faith in first order climate stations. I want to leave this here for you to read about the history of the Harrisburg station, but needless to say I would take any and all readings before 1941 as only reliable within an error range of probably 5-10x the acceptable range today if not more. Especially in the 1920s, since they decided for decades to take readings on top of buildings and construction from buildings around it had out sized effect on rain and wind readings. As well as it being on top of a building. At some point data was taken from two locations without it being always clear. But from its inception in 1888 till it's move to capital city airport in 1939 all reading were taken from atop buildings in various parts of Harrisburg. From 1939-1981 all readings were taken at what is now capital city airport. A couple years ago I did a deep dive regarding the persistent differences in rain, snow, temperature between capital city and KMDT, which shocked me at first due to them being so close by straight line distant. But you have far different microclimates and even far different extreme local effects from the positioning of the weather instruments. To say that Harrisburg weather data is superior just because it is a first order climate site after knowing it's history is laughable. I assume cheaters readings have been taken in a rural area for a long time consistently. As long as the record keepers were diligent in there record keeping and instrument calibration they are a valid and important climate data point. After actually investigating Harrisburg climate data and locations I feel there is no reliable way to properly compare any specific data from before 1941 to any data after. In general you can take a anything with an error of at least 25% +/- , and weather it rained, snowed, was sunny or cloudy. But specific amounts are a fools errand. How can you compare weather stations on top of buildings to one on the river in a sea of black asphalt with lots of planes aroundAt least after capital city has continued to keep data after a many year hiatus so that some sort of comparison to KMDT can be calibrated. Read for yourself https://www.google.com/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.weather.gov/media/ctp/ClimateStationHistory/Harrisburg%2520LCD%2520Site%2520History.pdf&ved=2ahUKEwj95JLmurCDAxVGfzABHULLClkQFnoECB0QAQ&usg=AOvVaw3o9ZqjV8uN7wryJgySUuZ2 Sent from my SM-G970U using Tapatalk