Weekly rainfall cycle

E. Linacre and B. Geerts


Rainfall in Sydney appears to be more frequent during the 24 hours to 9 am on Sunday, than during for any other day of the week, according to the Australian Bureau of Meteorology (1) (Table 1).

Table 1: Number of days that at least 0.2 mm of rain was recorded in Sydney, Australia, between 1992-1997. The day refers to a 24h period starting at 9 am (local time) on the previous day.















Similar results emerge from data from Melbourne, where the average winter weekday rainfall is 0.2 mm less than at the weekends, i.e. 10% more. Also, the daily maximum temperature is 0.29 K higher and the minima 0.24 K higher, on winter weekdays than at the weekends.

These effects must be due to human activity, since Nature knows no seven-day cycle. Vehicle and industrial emissions are less on weekends. These emissions include aerosol, or aerosols are formed through a reaction with the emissions. Therefore the aerosol concentration (and that of cloud condensation nuclei CCN) tends to be smaller on weekends. Therefore on weekends there are fewer droplets, which grow larger, hence droplet growth by collection becomes more efficient. Hence more rainfall on weekends.

This argument appears to apply only for cloud systems that only marginally produce (warm) rain. Very wet precipitation systems are so depleted of CCN, and more importantly of ice nuclei, that an anthropogenic source of aerosols will enhance the total rainfall. In Mexico City, for instance, rainfall is relatively enhanced downwind of the main industrial area for heavy rain events (at least 30 mm/24h), whereas for light rain events it is suppressed in that area, relative to the surrounding areas (2). If rainfall is marginal than the abundance of CCN may reduce the average drop size and therefore the chances of rain. The same has been observed for the city of St. Louis on the Mississippi river (3).

A weekly temperature cycle has also been noted: weekends may be warmer than weekdays (4). It is possible that such findings (for rain or temperature) arise from a statistical quirk (5). If you consider the week ending on Saturday and there is a longterm warming, the warmest day of eack week will be Saturday. Likewise, Sunday is the warmest if there is a longterm cooling. Either way the weekend is the warmest. However, quite different conclusions are reached if you regard the week as starting on Wednesday. Then Tuesday is the warmest day when there is longterm temperature rise, and Wednesday if overall cooling.



  1. Macey, R. 1998. It is official: itís most likely to rain on Saturday. Sydney Morning Herald 15th August. 
  2. Klaus, D., E. Jaurequi, E., and W. Lauer, 1983. Urban-induced precipitation effects in the Mexico City region. Archives for Meteorology, Geophysics, and Bioclimatology, Ser. B, 33, 275-88.
  3. Changnon, S. A., Jr., R.G. Semonin, R. G., and F.A. Huff, 1976. Hypothesis for urban rainfall anomalies. J. Appl. Meteor., 15, 544-60.
  4. Coakley, K. 1999. Report of the Nat. Inst. Standards & Technol., Boulder, USA.
  5. Seife, C. 1999. Lies, damned lies... New Scientist 25th September, p. 22.