Loading…
All of it is public data. If you want the underlying observations, please take them from the sources above rather than from this site — they are authoritative and this is not.
Daily highs, lows and precipitation come from the NOAA co-op station only. That station reads a true min/max thermometer, so it catches the actual daily extreme whenever it occurs. The CSU station samples hourly and structurally misses peaks that fall between readings. Blending the two would apply that bias to recent decades only, putting a step change in the middle of the record the normals are built from.
Where NOAA recorded nothing at all, a CSU estimate is used instead. It never overrides or averages with a NOAA reading.
Every statistic is computed per calendar day across all years in the record. The record range is the coldest and warmest value ever observed on that date; ties go to the earlier year. The shaded band is the 10th to 90th percentile, representing the middle 80% of years, slightly wider than one standard deviation. February 29 is excluded, so every series is 365 days long. Days with no observation are drawn as gaps, never interpolated.
Thresholds come from a gamma distribution fitted to cumulative precipitation for each calendar day, which is the transform the Standardized Precipitation Index (SPI) specifies. Years missing more than 5% of their readings are excluded from these calculations.
see:
Climate Data Guide - SPI
US Drought Monitor - Classification
I couldn't find a good presentation of weather data for Fort Collins. 2026, the year this site was originally deployed, was such a hot and dry year, I wanted to know how it compared to normal/previous years but could not find the data I was looking for. So I built it myself.