6.4                                   Synoptic scale systems and fronts

Accurate prognoses of how synoptic scale systems evolve will allow the forecaster to maximise the chances of correctly predicting the individual weather elements such as wind velocity. Fraedrich and Leslie (1991) suggest that, based on the time taken for errors in the 500–hPa field to double, in areas of high baroclinicity predictability time scales might be as short as about one day. On the other hand, as mentioned in Section 6.3.2.5 Baba (1993) suggests that at least a general trend in the local weather might be predicted well in advance in some circumstances. This section presents a brief overview of techniques that may be suitable for the prediction of synoptic scale systems and fronts in Antarctica.

6.4.1                                Forecasting strategies

There are several strategies that are useful for the forecasting of synoptic scale systems and fronts. These include: climatology; monitoring the long–wave pattern; persistence; analogues; conceptual models; and numerical models. The influence of the broad–scale on the movement of systems is discussed in Section 6.3 and is alluded to in Section 5.2.1 and is not developed further here. On the other hand, each of the remaining techniques will be briefly examined.

6.4.1.1       Climatology

Section 2.4 provided a comprehensive overview of work determining, among other things, cyclone tracks and centres of persistence of synoptic features. The only other comment to be made here is that, anecdotally, no two seasons are exactly alike and so climatology provides at best, basic background information.

6.4.1.2       Persistence forecasting

In the context of forecasting synoptic scale systems persistence might be viewed as the assumption that a system will:

·                         remain stationary in space or in a particular stage of development;

·                         maintain its current rate of progress;

·                         or in the case of a series of systems, the semi–cyclic repetition system behaviour, due perhaps to the disposition of the long wave pattern.

Persistence forecasting certainly has a place for short term forecasting in the Antarctic where the systems are evolving slowly. Satellite information will play a large role in assisting with the movement of systems under such circumstances. It is interesting to note that Fraedrich and Leslie (1991, p. 8) suggest that persistence may outperform analogue models for the forecasting of the movement of systems in the Antarctic, due primarily to the absence of good analogues.

6.4.1.3       Analogue models

An example of the analogue technique is given in Fraedrich and Leslie (1991, p. 5): their Figure 2 shows a pattern of 500–hPa fields for 18 to 20 July 1985 and uses these to predict what might happen for the period 6 to 8 August 1986. The limitation of this approach is mentioned above.

6.4.1.4       Conceptual models

In Section 5.2.1 two cyclone–frontal conceptual models were discussed briefly. Section 5.3.1 discussed the work of Junker (1977), and referred to Streten and Troup (1973), for manual methods for constructing mean sea level analyses using satellite data. Similarly, Section 5.3.2 discusses the Guymer (1978) technique for constructing 1000–500–hPa thickness fields (and as a result, 850, 700 and 500–hPa contour fields). Section 4.2.4 should be referred to for a suggestion as to how to get the best from a conceptual model.

6.4.1.5       Numerical models

For most forecasting purposes the Antarctic forecaster will want to access a range of numerical models. Pendlebury and Reader (1992), and Leonard et al. (1997), give some encouragement that, respectively, the operational numerical weather prediction systems of the ECMWF and the UK Meteorological Office perform adequately in the Antarctic context. Adams (1997), on the other hand is less convinced. However, in a personal communication Adams has more recently advised that models such as the USA NCEP Aviation Model have performed exceptionally well.

Adams' (1997) study was based on the model output, for two summer seasons, from the ECMWF model and from the Australian Bureau of Meteorology's GASP model. Adams (1997) attributes the poor performance to poor initialisation analyses. Perhaps the periods studied by Adams (1997) had particular problems as Cullather et al. (1997) note a marked improvement in ECWMF (and NCEP) model analyses for the period 1985–94.

Section 4.2 gives an overview on how to use numerical model output and, combined with the general optimism noted above, the Antarctic forecaster should be able to use numerical guidance with some confidence (though not blindly) in the Antarctic context.