The Effects of Data Gaps on the Calculated Monthly Mean Maximum and Minimum Temperatures in the Continental United States: A Spatial and Temporal Study

Abstract

Gaps in otherwise regularly scheduled observations are often referred to as missing data. This paper explores the spatial and temporal impacts that data gaps in the recorded daily maximum and minimum temperatures have on the calculated monthly mean maximum and minimum temperatures. For this analysis 138 climate stations from the United States Historical Climatology Network Daily Temperature and Precipitation Data set were selected. The selected stations had no missing maximum or minimum temperature values during the period 1951?80. The monthly mean maximum and minimum temperatures were calculated for each station for each month. For each month 1?10 consecutive days of data from each station were randomly removed. This was performed 30 times for each simulated gap period. The spatial and temporal impact of the 1?10-day data gaps were compared. The influence of data gaps is most pronounced in the continental regions during the winter and least pronounced in the southeast during the summer. In the north central plains, 10-day data gaps during January produce a standard deviation value greater than 2°C about the ?true? mean. In the southeast, 10-day data gaps in July produce a standard deviation value less than 0.5°C about the mean. The results of this study will be of value in climate variability and climate trend research as well as climate assessment and impact studies.