La Niña and Tucson Winter Rain

Soon media attention will focus on the monthly and weekly outlooks from the Climate Prediction Center, which are consistently expecting a possibly moderate to strong La Niña this winter. Much will be said, some of it accurate, about La Niña's effects on Tucson rainfall.

Below is a plot of winter rainfall at the Tucson airport over the last 60 years¹ as a function of sea surface temperature anomalies in the Niño 3.4 region.² The SST anomalies are averaged over five-month periods of November through March, while the precipitation is totaled for the five-month periods.³

The winters are color coded by decade, as labeled. The decade is for the beginning of the 5-month period, so the most recent winter (2009-2010) is included in the 2000's. The two lines are manual fits to the 90th percentiles and the 10th percentiles of the total precip. On each line the abrupt change in slope is intentional, emphasizing aspects of the data distribution that may be overlooked when attention is focused on the means. Noting first the right side of the figure, El Niño is not so much about well-above normal winter rainfall. Very wet winters occur even without El Niño. More striking, especially as you move into the moderate and strong El Niño territory, beyond +1 °C, is the sparsity of winters with rainfall below average (60-year mean 3.89 inches). Last winter's moderate to strong El Niño, at 4.94 inches easily satisfied the “not below average” expectation. Similarly, La Niña's are not the only times to be concerned about the possibility of well-below normal rainfall. Several neutral winters have been well below normal. Attention instead is focused on the complete absence of above normal rainfall for moderate and strong La Niñas. There are two weak La Niña winters that had well-above normal rainfall. Both of these followed moderate to strong El Niños, but by two years, not one. The intervening year was very dry in both cases.

In summary, there will be times this winter, as there are in every winter, when the weather pattern shifts to one that favors periods of rain in Tucson. With La Niña these wet patterns will probably be relatively infrequent and/or short-lived. It's very unlikely that the total precipitation for the winter will be above average, but it's plausible that it could be close to average. It's also possible that it could be well below average.

1. Source is the Western Regional Climate Center.


2. Source is the CPC link to the Extended Reconstructed SST (ERSST) anomalies.
   As explained in the CPC updates, the Oceanic Niño Index (ONI) is defined as the three-month running-mean of SST departures in the Niño 3.4 region. Since the ONI is already a 3-month average, the 5-month average for NDJFM, plotted in the figure above, is approximately, but not quite the same as an average of the ONI for NDJ, DJF and JFM, but weighted by (2/5, 1/5, 2/5).


3. Why a 5-month season, instead of 3 months? Because there's not much difference between weather systems that manage to produce rain in Tucson during the “border” months of November and March vs. Tucson rains during the standard meteorological winter of December, January and February. Winter rains in Tucson are often separated by extended dry periods, sometimes a month or more, and so I think the longer period gives a better picture. Years that manage to produce significant rain in November and March deserve the extra credit.