Seasonal Climate Watch
June to October 2016
Date: May 20, 2016
1. Advisory
Current observations still show the fast decay of El-Niño. As most ENSO (El-Niño
Southern Oscillation) prediction models indicate, however, there is a possibility for the
development of a weak La-Niña toward late spring through to the coming summer season.
Despite the fact that current climate conditions and most of the forecast models are
indicating the tendency of a warmer and drier winter season, the confidence in the
forecasting systems is marginal.
2. Recommendation
There is insufficient evidence that supports the continuation of the current cooler and wetter
conditions over the country to persist through the winter season. However, the current
forecast is still clouded by the growing uncertainty. It is therefore highly recommended that
medium- and shorter-range weather forecasts be monitored for the development of
conditions that may alter or strengthen the expectation of the current forecast.
3. State of Climate Drivers
Observations show that the state of El Niño is steadily decaying and most of the forecast
models’ predictions indicate the possible development of a moderate (weak) La-Niña state
toward late spring and summer 2016/17. The Indian Ocean Dipole (IOD) still shows a
tendency of a neutral state towards the winter season but with a tendency of negative phase
development through spring. The neutral phase of IOD is due to the continual basin-wide
warming of the Indian Ocean which may suppress the pressure or temperature gradient
between the west and the east section of the ocean.
Despite the Southern Annular Mode (SAM) has been showing a tendency for a positive
phase over recent weeks, it is now fluctuation around the mean with a possibility of
remaining in a positive phase. Should the positive phase be persisting, the transport of
moisture and colder temperatures from the southern Atlantic Ocean to the subcontinent
may be suppressed. The negative phase of the SAM and the weakening of the polar vortex
are usually associated with a colder and wetter winter season over the region, as most of
the cold fronts tend to reach the sub-continent. Therefore, the monitoring of this climate
system is very important since its behaviour may change in a matter of few weeks.
Generally, it is known that ENSO has a noticeable impact on the climate of our region
during the austral summer season while the IOD is also found to influence rainfall activity,
particularly during spring. Furthermore, the SAM is found to affect South African climate
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008
conditions by regulating the south/northward positioning of the mid-latitude jet stream and
transport of associated air masses from the southern Atlantic Ocean. Its impact is
pronounced in winter, although its predictability is marginal at the seasonal timescale.
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008
4. Climate forecast Details
4.1 Rainfall
Despite the large uncertainty and marginal confidence (Figure A1) in the forecasting
system, there are chances for above-normal rainfall conditions over the western and northeastern parts of the country for the winter season.
For improved confidence in a probabilistic prediction use is made of skill scores most
notably the Relative Operating Characteristic (ROC) which indicates the relative
performance of the prediction system. Areas of ROC scores above 0.5 may be considered
as areas of added confidence for the prediction (Figure A1).
Figure 1: Rainfall forecasts for the three overlapping seasons valid for the period of June
to October 2016 and extreme forecasts for June to August 2016 season (right panel).
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008
4.2 Minimum and Maximum Temperatures
Minimum and maximum temperature forecasts show a tendency of warmer than usual
temperatures over the country particularly over the north-eastern part during mid-winter
season (Figure 2).
For improved confidence in a probabilistic prediction use is made of skill scores most
notably the Relative Operating Characteristic (ROC) which indicates the ability of the
forecasting system to distinguish events from non-events. As noted earlier, areas of ROC
scores above 0.5 may be considered as areas of added confidence for the prediction (Figure
A2).
Figure 2: Probabilistic minimum (left panel) and maximum (right panel) temperature
forecasts for the three overlapping seasons valid for the period of June to October 2016.
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008
Contributing institutions
All the forecasts are a result of an objective multi-model prediction system developed at the
South African Weather Service. This system comprises of long-range forecasts produced by
the following institutions:
5. Appendix
Figure A1: The skill of the forecasting system in discriminating wet or dry events during the
forecasting period as shown in the caption of each plot. Those regions with no shades imply
that the forecasts are not better than chance.
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008
Figure A2: The skill of the forecasting system in discriminating hot or cold events during the
forecasting period as shown in the caption of each plot. Those regions with no shades imply
that the forecasts are not better than chance.
For further inquiries contact cobus.olivier@weathersa.co.za
Tele: +27 12 367 6008