Times when Dorothea Mackellar’s “droughts and flooding rains”* affected Manilla in the years from 1997 are shown by the wavy line on this graph. The climate swings in and out of times of high and low rainfall.

Peaks or troughs were often a year or two apart, but most of them were not very far from the normal rainfall value. Only two of the troughs were so far below normal that they were severe droughts: August 2002, and December 2013 (or maybe later). Milder droughts came in October 2006 and September 2009.
The rainfall in these 17 years was not below the long-term average, but slightly above it. As well as droughts there were two peaks of extremely high rainfall: in July 1998 (when the new Split-Rock reservoir suddenly filled) and in November 2011. These “deluges” had rainfall that was further from normal than the low rainfall in the droughts. Other rainfall peaks came in November 2005, October 2008, and October 2010.
In total, there were nine peaks and troughs with rainfall outside the normal range. Six of them came in the spring months of September, October or November.
Peaks and troughs in rainfall at Manilla quite often come near times of La Niña and El Niño. These are events in the record of Pacific Ocean temperatures called ENSO (El Niño – Southern Oscillation). The ENSO record for the last 17 years is shown in the second graph.

In the middle years, from 2006 to 2011, the match between Manilla rainfall and ENSO was close, but most peaks and troughs in rainfall came one or two months earlier than those in ENSO.

The extreme drought trough of August 2002 came three months before the El Niño peak. Similarly, the deluge of July 1998 came five months before a La Niña peak.

When there was a “Super El Niño” in November 1997, this ENSO event brought no rainfall shortage at all to Manilla. On the other hand (as at April 2014), the current drought has no El Niño event to match it.

I discussed the relation of Manilla rainfall to ENSO in the shorter period 2000-2010 in an earlier post. For the same period I have also discussed relations of Manilla  temperature and humidity to ENSO.

Notes.
Monthly rainfall totals, in mm, are smoothed using a Gaussian smoothing function with a width (at half height) of six months.
ENSO values for the area NINO3.4 use the monthly OISST.v2 data, taken from the website shown on the graph. They are averaged (1:1:1)/3, as had been done in the ONI data set.

* By arrangement with the Licensor, The Dorothea Mackellar Estate, c/- Curtis Brown (Aust) Pty Ltd.