Schultz, D. M., 1998: Does it rain more often on weekends? Annals
of Improbable Research, 4(2), 29.
Does It Rain More Often on Weekends?
by Dr. David M. Schultz
National Severe Storms Laboratory, Norman, Oklahoma
Ever since God declared, "remember the Sabbath day by keeping it holy" (this was reported by Moses in 1200 BC), and especially since the five-day work week was invented in the early 20th Century, society has desired to make the most of its free time-especially Saturdays and Sundays. People have generally favored clear, sunny weekend days to those with cloudy, rainy weather. Our investigation focuses on two key questions:
- Does precipitation in the United States occur more often on weekends than weekdays?
- If so, what locations in the United States are most susceptible to wet weekends?
We used 24-hour precipitation observations from nearly 200 stations across the United States. The data were derived from the Daily Weather Observations dataset (TD-3220), archived by the National Climatic Data Center and the National Center for Atmospheric Research, and obtained on the World WeatherDisc (WeatherDisc Associates, Inc. 1994). The day of the week was noted for each precipitation event (trace, 0.01 inch, and greater) for each station during the 40-year period1
1951-1990, summing the number of events for each of the seven days of the week at each station. The total number of precipitation events at each station was also determined, so that the percentage of precipitation occurrence on each day of the week (Pi, where i= Monday, Tuesday,..., Sunday) could be calculated. We compared the deviation of the observed probability of weekend precipitation occurrence for each location from the expected value DSS = PSaturday + PSunday - 28.57%. The interested reader can show that
Results, Discussion, Conclusion, and Dismissal
Based upon the computations described above, Figure 1 presents the values of DSS in hundredths of a percent for the observing stations across the United States. The areas with an excess of wet weekend days (unhatched) include much of the East Coast, the Southeast, the Great Plains extending from northern Texas to North Dakota, the Intermountain West, and the Pacific Northwest. In contrast, areas with an excess of dry weekend days (hatched) include the Midwest, the Texas Gulf Coast, the Continental Divide and the High Plains, most of California, Alaska, and Hawaii.
(Our results do not explain the apparent misnomer of locations found in dry-weekend areas such as: Rainy, Michigan; Rainsboro, Ohio; Wetmore, Texas; and the Wet Mountains, Colorado.)
That the pattern of DSS exhibits coherent regions of positive and negative values indicates that there is perhaps some validity to the hypothesis that certain locations favor weekday over weekend precipitation, and vice versa.
We have been asked if our findings are statistically significant. That question is beyond the scope of this investigation, and far beyond the scope of interest of the reader.
Figure 1. The areas that are unmarked on this map (positive
values) have an excess of wet weekends. The cross-hatched areas
(negative values) have an excess of dry weekends.
- All investigations that are of biblical scope must, by definition, involve 40-year periods.
Copyright © 1998 The Annals of Improbable Research (AIR). All rights reserved.