5.1                                   Analysis and diagnosis fundamentals  

5.1.1                                Overview

A primary objective of meteorological analysis is to develop a coherent picture of existing atmospheric patterns, building up a model of the atmosphere. There are many components to the analysis phase. The basic observations as plotted on the weather chart may be interpreted to form initial impressions of atmospheric patterns. The two–dimensional quasi–horizontal representations of the atmosphere (at mean sea level and at upper levels), vertical representations and, as appropriate, other forms of analysis that will be discussed later, complement the observations.

Traditionally observations and analyses have been done at fixed time intervals, particularly for hand drawn analysis; however the increasing availability of asynoptic data (e.g. satellite soundings) is tending to modify the fixed–time analysis approach. The forecaster must assimilate the new data as it comes to hand to confirm his/her expectations or modify his/her picture of the atmosphere.

Several texts deal exclusively or in part with meteorological analysis and prognosis. For example, a comprehensive summary of manual analysis methods is given by Saucier (1955) and this is recommended as reference study for all forecasters. Many other texts have sections on analysis and prognosis, perhaps the most convenient being in Berry et al. (1945), Malone (1951) and Wickham (1970). Oliver and Oliver (1945) give useful guidance on analysis of single station data, while Fujita (1963) and Williams (1963) deal with mesoscale analysis techniques. Miller (1972), Crisp (1979) and Matley (1981) contain some good practical hints on secondary analysis of externally prepared products, and of course the literature is full of synoptic case–studies that demonstrate the advantage of various analysis techniques.

In this handbook basic competence in isobaric, streamline and contour analysis is assumed; also, the interpretation of the present weather identifications and standard clouds types, in terms of their value for analysis and short–term prognosis, is not discussed. For mid–latitude meteorology these matters can be found in texts such as Petterssen (1956) and Wickham (1970) or often in–house training notes (e.g. Bureau of Meteorology, 1983). This chapter is concerned with broad principles that may be applicable to a particular situation. It should be noted that analysis in the Antarctic is not fundamentally different from that carried out for other parts of the world. However, a number of factors need to be taken into account. These include:

·                         the lack of data;

·                         the variations in orography from near seal level to above 4 km requiring differing analysis techniques to be followed;

·                         the lack of weather systems with the standard frontal characteristics.

5.1.2                                Basic observations

The basic observations plotted on a mean sea level chart usually provide strong clues to the physical processes taking place in the atmosphere and form an integral part of the forecaster's analysis and prognosis. Some examples of the value of individual observations in the short–term forecasting context are given below.

·                         An observation of strong winds at a station must be carefully assessed from a geographic and climatological viewpoint, as well as in the overall meteorological context. The strong winds may be, for example, a normal katabatic that will ease as insolation increases; or may be due to a passing low–pressure system. Winds in katabatic prone areas may not be representative of the synoptic scale flow.

·                         An AVHRR channel 3 cloud observation of supercooled water cloud near the coast might indicate a risk of airframe icing should aircraft unwisely venture into the cloud.

·                         Relatively large corrected pressure tendency falls often indicate the approach of a large–scale low–pressure system and the possible eventual onset of strong winds. Again though, the overall meteorological context of the pressure falls is important: it may be that there is a general pressure fall over an area not necessarily linked to a discrete low.

Competent visual weather observations taken at more frequent intervals than the normal three–hourly time space may reveal aspects of the mesostructure of the atmosphere. Certain changes in visibility, cloud growth rates, cloud structure and movement provide critical indications of ongoing change.

5.1.3                                Mean sea level pressure analysis and satellite data

The MSLP field provides the basic integrated representation of wind flow near the ground. As has been discussed (Section 2.4.6 and Section 2.6.1; see also Section 5.4) MSLP analyses have limited usefulness over the Antarctic continent itself above a height of 1 km, but remain a core resource for coastal Antarctic forecasting. Isobaric analysis enables weather systems such as highs, lows and fronts to be identified and tracked over time.

A primary concern of the short–term forecaster is to determine, where possible, small–scale or mesoscale perturbations in the larger–scale flow. At most Antarctic stations weather watch radar is not available and sequences (or loops) of hourly and three–hourly satellite imagery from geostationary satellites, although inadequate for mesoscale forecasting, do provide a useful temporal perspective of such development. While the less frequent AVHRR data from the polar–orbiting satellites often gives information that is very useful but in the context of short term forecasting, has a limited span of useful currency. On the other hand, the AVHRR and APT data have a role in the more medium–term forecast time frame in identifying the main cloud features in the absence of geostationary satellite data.

The mesoscale features must be integrated with the larger–scale features on the MSLP charts to identify the various components that are causing the current weather.

5.1.4                                Upper–air analysis

The most practical way of displaying the patterns of circulation in the free atmosphere is by the construction of contour charts for selected pressure levels (850, 700 hPa etc.). Contour analysis enables systems such as upper–air lows, ridges and troughs, to be identified. The degree to which a mean sea level pressure system is reflected throughout the troposphere can be readily discerned. In certain circumstances the drawing of streamlines may be preferable to contours.

The thickness of a layer refers to the difference between the contour heights of two pressure levels, and it depends on the mean virtual temperature of the layer – the greater the mean virtual temperature, the greater the thickness. Thickness analysis then represents the integrated temperature pattern between two constant pressure analyses and features such as warm or cold pools of air can be identified.

Charts of atmospheric parameters, such as humidity, vorticity and vertical motion complement the standard layer charts and enhance the forecaster's three–dimensional picture of the atmosphere and assist in diagnosis of phenomena such as snowfall. Similarly, wet–bulb potential temperature charts on isobaric surfaces (e.g. 850 hPa, 500 hPa) may be a valuable aid in analysis and forecasting frontal movement and precipitation (see Bradbury, 1977).

5.1.5                                Meteograms or time series

Meteograms effectively display the magnitude of a number of elements on a common time base (timeline) so that their interaction can be seen. Figure 5.1.5.1 is, for example, a time series of observed surface data from Mawson Station for 1 March to 14 April 2000 and at about 21 March the surface wind has peaked at about the time of minimum surface pressure, presumably as the apex of a low–pressure trough moved past the station.

NWP may be used to produce time series for prediction or case study diagnosis Figure 5.1.5.2 shows a time cross section of upper winds, potential temperature and height for a GASP model grid point near Mawson for a period similar to that shown in Figure 5.1.5.1. It may be seen that shortly after 21 March mid–tropospheric winds veered in a manner consistent with a low–pressure system passing close to Mawson.

5.1.6                                Quantitative values

Analysis involves calculations of quantitative values of appropriate parameters that are useful for short–term forecasting: usually these values would be assessed through examination of NWP output. The parameters calculated clearly depend on the particular meteorological situation and could include:

·                         directions and speeds of movement of key features, such as fronts, or lows;

·                         gradient and geostrophic wind;

·                         rates of advection of temperature, moisture, vorticity, etc.;

·                         changes in cloud top temperature or areal extent of cloud features;

·                         vertical and/or horizontal wind shear;

·                         vertical motion.