During the last several years, there has been an increase in the monitoring of meteorological phenomena. For
accurate analysis of the spatial and temporal variability of precipitation as well as air temperature, recorded
observations must be trustworthy. Therefore, one must implement effective methods for assuring data quality.
This paper implements meteorological data quality controls using open-source Python code.We implemented
six types of automatic quality-control processes for climatic data series. This work also introduces a quality
index for meteorological time series data and uses three-dimensional kriging to fill in missing data. We apply
these methods to daily historical data from 1980 to 2019 from the Institute of Hydrology, Meteorology, and
Environmental Studies of Colombia. Applying the quality-control process to the available meteorological
stations allows us to validate more than 90% of the time series. We find that approximately 6.4%, 4.4 %,
5.4%, and 5.5% of the data are flagged as atypical values in the time series of minimum temperature,
mean temperature, maximum temperature, and precipitation, respectively. We verified the accuracy of the
quality-control procedures by introducing multiplicative random errors and computing the probabilities of
false positive and false negative errors. This procedure emphasizes the relationships between neighboring
meteorological stations to detect errors in the data, even if they belong to different climatic regions. This
analysis also implements a method of reconstructing missing data in the absence of a trustworthy reference
time series.