Home | About Us | Contacts | Help | Feedback | SEARCH Global | Australia | NSW | Vic. | Qld | WA | SA | Tas. | ACT | NT | Ant. | Weather & Warnings | Hydrology | Climate | Numerical Prediction | About Services | Learn About Meteorology | Registered User Services |

Global Guide
Main Menu

Global Guide to Tropical Cyclone Forecasting: CHAPTER 1: GLOBAL OVERVIEW

1.4 FORECAST PROBLEMS

1.4.1 The Concept of Forecast Difficulty

Neumann (1981, 1985) introduced the concept of forecast difficulty to assess forecast improvements over the North Atlantic basin. The concept is based on the use of residual errors of the statistical CLIPER (CLImatology and PERsistence) forecast model (Neumann, 1972) to provide a threshold skill level and a basis for determining forecast difficulty. These errors are directly proportional to operational forecast errors, for example, both have lowest errors when tropical cyclones move slowly and on persistent tracks. Although the index of motion steadiness (Section 1.3.4.5) addresses persistency, it is not sensitive to translational speed.

Thus, CLIPER is an excellent measure of forecast difficulty and normalising all forecasts by the CLIPER errors provides a homogeneous indication of forecast skill. IWTC-II recommended that all cyclone motion forecasts be normalised by concurrent CLIPER forecasts to provide a universal standard for comparison.

Pike and Neumann (1987) developed CLIPER models based on the same criteria for each of the tropical cyclone basins⁽¹⁾ and compared residual errors (Fig. 1.25). Their Forecast Difficulty Level (FDL) varies widely, from a maximum in the Australian/southeast Indian and southwest Pacific regions to a minimum in the north Indian basin. This implies that, given similar information on the current and past positions of a cyclone, the most accurate forecasts can be expected in regions with low FDL. For example, mean 72-h forecast errors are 665 km for the North Atlantic, but only 460 km for the eastern North Pacific. After accounting for the FDL, forecast skill is comparable in the two basins.

Note that the FDL relates only to the intrinsic nature of the cyclone tracks. For example, no account is taken of operational factors such as the accuracy of centre fixes, which will affect the persistence forecasts. This qualification is particularly applicable to the north Indian basin, where cyclones are relatively weak and centre positions difficult to locate accurately. Moreover, the FDL does not address inherent warning problems, such as storm surge potential, communications to coastal populations and evacuation.

Figure 1.25: Forecast difficulty level vs. forecast period for specified basins (Pike and Neumann, 1987).

A further problem is that not all CLIPER techniques are equally skilful. When small data sets are used, or historical data have considerable uncertainty, for example, the CLIPER equations may be unstable.

Nevertheless, provided its limitations are known, the FDL provides an excellent means of comparing ocean basins, different forecast techniques, and seasonal changes in forecast errors. We strongly recommend the use of FDL in all comparisons of forecast errors from different data sets.

1.4.2 Special Forecast Problems

Surveys conducted for this guide and by McBride and Holland (1987) indicate a universal need for improved observations leading to better assessment of the environment in which tropical cyclones are embedded. Other parochial forecast difficulties, such as binary cyclones and erratic motion, are discussed in this section.

1.4.2.1 Binary Tropical Cyclones

The presence of more than one tropical cyclones adds to the workload of a forecast office. Additionally, if these tropical cyclones are in close proximity they may interact in a process known as the Fujiwhara effect after Fujiwhara (1921, 1923, 1931) and introduce significant forecast difficulties. Brand (1970), Bao et al. (1979), Dong and Neumann (1983) and Elsberry (1988) discuss operational aspects of binary tropical cyclones and recent papers by Lander and Holland (1993) and Holland and Dietachmayer (1993) provide an observational and theoretical discourse. The appropriate analysis methods are discussed in section 3.3.5.

In general, noticeable interaction does not take place until separation distances are less than 15^o lat and the Fujiwhara effect dominates at separation of less than 8^o lat. Lander and Holland indicate that initial capture and final escape of the binary pair is typically rapid and extremely difficult to forecast.

A global reassessment of binary tropical cyclone interaction is provided in Fig. 1.26, which tabulates the number of tropical cyclones simultaneously existing within various separation distances of less than 1850 km (1000 nm, 16.5^o lat). This figure confirms earlier studies that binary tropical cyclones are most common in the western North Pacific, where about four cyclones per year occur within 15^o lat of each other. Binary interaction is also quite common in the eastern North Pacific basin and Australian/southwest Pacific regions. Indeed, if the number of interactions is normalised by tropical cyclone frequency⁽²⁾ (inset to Fig. 1.26), the rate of binary interaction is about the same for all three basins. However, as pointed out by Dong and Neumann (1983), binary tropical cyclones over the western North Pacific basin typically occur in the monsoon trough where large-scale steering currents are weak and binary interaction dominates. In the eastern North Pacific, and to a lesser extend the Australia/southwest Pacific basin, binary interaction often is masked by the stronger environmental steering.

Figure 1.26: Frequency of binary interaction over each basin. The lower inset contains the average number of tropical cyclones per season for the specified time period.

1.4.2.2 Unusual Motion

Dealing with tropical cyclones having unusual motion has always been a forecast problem (Chen, 1988) and this was cited as a major problem in virtually all basins. Neumann (1981) ranked one thousand 48-h forecast errors over the North Atlantic basin according to the amount of error. He found that the 10 cyclones with lowest error followed smooth tracks, but that the 10 cyclones with the highest error were all quite erratic. Average error in the latter group was about twice that of the former.

A classic example of a tropical cyclone with erratic motion characteristics is provided by Typhoon Wayne (1986, Fig. 1.27). A combination of frequent changes in heading, speed and intensity contributed to large forecast errors (Chan and Lam, 1988). Such erratic movement occurs in all basins with varying degrees of frequency, with the Australian region being the worst (Fig. 1.14). We note, however, that smooth, steady tracks are in the minority in all ocean basins and that "unusual motion" is more usual than implied by its title.

Rapid acceleration and weakening of tropical cyclones interacting with the baroclinic westerly flow and recurving has been identified at both IWTCs as a major forecast problem. The responsible mechanisms are discussed in Elsberry (1987).

Figure 1.27: Track of the western North Pacific Typhoon Wayne (1986). Circled positions and associated maximum wind (km h^-1) are at 0600 UTC. Maximum intensity of 170 km h^-1 occurred on September 4, 0000 UTC.

1.4.2.3 Storm Surge

Storm surges from tropical cyclones produce far more damage and human casualties than wind in some areas if the world⁽³⁾. As pointed out by Frank (1987), "the shallow waters of the Bay of Bengal, the low flat coastal terrain and the funnelling shape of the coastline can lead to devastating losses of life and property due to the surge from a tropical cyclone of even moderate intensity." Since storm surge prediction is much dependent on track forecasts, this conflicts, in a sense, with the Pike and Neumann (1987) study, which found that the north Indian basin is the least difficult forecast region. However, the latter study considers only forecast error and not related factors such as storm surge potential.

1. Pike and Neumann combined basins 6 and 7 of Table 1.2 into a single basin.

2. These annual frequencies may differ from those in Table 1.3 due to the different periods of record.

3. Notably, the north Indian and portions of the North Atlantic basins; an exception in the North Atlantic was the massive wind damage from Hurricane Andrew in 1992.

Contents Chapter 2