SECTION 1. INFLUENCES ON TROPICAL CYCLONE MOTION

Tropical cyclone (TC) motion is the result of a complex interaction between a number of internal and external influences. Environmental steering is typically the most prominent external influence on a tropical cyclone, accounting for as much as 70 to 90 percent of the motion (Neumann, 1992). Theoretical studies have shown that in the absence of environmental steering, tropical cyclones move poleward and westward due to internal influences (Elsberry et. al., 1987).

The dominant influence on movement is the large-scale steering or large-scale environmental steering. Determination of the large-scale environmental steering for a given TC is not unique. Many different levels and layers have been proposed for use and there is some difficulty in deciding which level or layer is best for a given storm.

Large-scale environmental steering is usually computed by separating the TC wind fields from the large-scale environmental wind fields. The separation of flow is anywhere from 1-7° radially from the TC center (Fig. 4.1). The winds on the outside of the separation are used for determining environmental steering and are commonly named the steering flow. In this section, several useful concepts for determining steering are discussed. One of the complicating factors in determining steering is analysis uncertainty near a TC vortex. Because of this, some models such as JTWC92 (Neumann, 1992) make use of what is referred to as implied steering. Here, more conservative predictors (geopotential heights at rather large distances from the storm center (up to 15 degrees of latitude) are used to infer geostrophic steering flow. For example, for TCs located in the deep tropics, the intensity of the poleward subtropical ridge line, which is typically located about 600 nm poleward of the TC, is used by these models to estimate the steering flow near the storm itself. This procedure might be considered to be a satisfactory operational trade-off between the desire to measure actual steering and the uncertainty of the analysis near the TC.

The first step in computing environmental steering is determining a level to use. Generally, more intense storms extend higher in the troposphere and have higher steering levels (Dong and Neumann, 1986). Figure 4.2 demonstrates how the steering level can change through intensification.

a. Deep Layer Mean - Although early studies concentrated on different pressure levels for determining environmental steering (George and Gray, 1976, Brand et. al., 1981), recent studies show that layer averages appear to be better suited to the task (Sanders et al., 1980; Chan and Gray, 1982; Dong and Neumann, 1986). Neumann (1979) tested a number of functions as to their ability to explain the variance of Atlantic TC motion. He found that the Sanders and Burpee (1968) pressure weighted wind function also provided for the maximum variance reduction when applied to the height fields. The deep layer mean value (DLM), specified by Neumann, is

D = (75*D1 + 150*D2 + 175*D3 + 150*D4 + 100*D5 + 75*D6 + 50*D7 + 50*D8 + 50*D9 + 25*D10)/900

where D1 through D10 are the heights (or winds) at the 1000, 850, 700, 500, 300, 250, 200, 150, and 100 hPa levels, respectively. The DLM is obtainable from Fleet Numerical Meteorological and Oceanography Center (FNMOC), Monterey, CA.

b. Shallow Layer or Single Level Steering Flow - There are times when shallow layer or single level steering flow is superior to the DLM steering. The following is general guidance for which layer or levels to use for steering.

(1) When in doubt, use the 10-level deep-layer-mean. Not having a DLM or other information on the storm, the 500-hPa level is the best single level from which to estimate future motion.

(2) Be aware, particularly for large storms, that inflow at 1000 hPa and outflow at 100 hPa might distort the steering flow. However, these layers are not weighted very heavily in the DLM computation.

(3) Weak storms are steered by a shallow layer that can be estimated by using the 700 hPa alone.

(4) For storms embedded in a sheared environment, the lower portion of the storm tends to follow the low-level flow and the upper portion of the storm tends to follow the upper circulation and storm weakening is typical. Thus, since the "eye" is associated with the lower level circulation, it is better to use 850 or 700 hPa.

(5) For very large storms, interactions between the storm and the environment make it difficult to define a steering flow even with the availability of a DLM.

Large-scale environmental forcing (such as steering) typically explains most of the TC motion variance and smaller scale internal influences are sometimes masked by the larger scale pattern. However, when the steering flow is small, these internal influences might be the dominant factors moving the storm.

Beta, a mathematical notation, denotes the latitudinal variation of the Coriolis parameter or the latitudinal gradient of earth's angular speed. The Coriolis parameter, twice the component of the earth's angular velocity about the local vertical, has zero value at the equator, and becomes extreme at the pole (i.e., |1.4584E-4| radian per second). On the other hand, the beta parameter has a maximum value at the equator (i.e., 2.289E-11 per meter per second) and becomes zero at the pole.

The beta effect can be interpreted as the local change of the vertical component of relative vorticity due to the product of the beta parameter and meridional wind speed, i.e., the beta term in the absolute vorticity equation. The equatorial beta effect causes a northern hemispheric TC to move toward the northwest with a speed about several degrees per day when the large-scale steering flow is absent (Rossby, 1948 and 1949; Kuo,1950; Kitade, 1980; Anthes,1982; DeMaria, 1985; Chan and Williams, 1987 and 1994; Fiorino and Elsberry, 1989; Smith, 1993; Wang and Li, 1994; Jones, 1995). In the Northern Hemisphere, the beta effect causes tropical cyclones embedded in the easterlies south of the subtropical ridge to move faster and slightly to the right of the steering flow. The beta effect also causes Northern Hemisphere tropical cyclones moving northwest tend to move faster and to the left of the steering flow, and tropical cyclones moving northeastward tend to move slower and to the left of the steering flow (Elsberry et. al., 1987).

The beta effect on a TC can be a function of the TC size, but not necessarily the TC intensity (DeMaria, 1985). When the TC size is small and the steering flow is moderate to strong (e.g., about 15 knots or 7.7 m/s), the direction of motion reflects the direction of the steering flow. When the TC size is large, the beta effect may have a major impact on the motion.

Large TC s have strong winds out to 5-7° radially that could affect steering flow computations. Super Typhoon Abby (1983) was large enough (a 30 knot wind radius of over 350 nm or 650 km) that TC winds were included in steering flow computations (Chan, 1986). There is evidence that the steering flow concept may not be as applicable in large size TCs as it is for small size TCs.

Since the advent of looped geostationary satellite imagery it has been noted that the eye of a TC sometimes moves relative to the apparent center of circulation of the TC. This wobble is usually less than the eye diameter, but can result in erroneous determinations of the current TC motion if the TC is tracked from center fix to center fix.

Sheets (1985) suggests tracking the mass field envelope of the TC rather than the eye. The mass field envelope is defined as an area of high winds surrounding the center of the TC and is best determined by reconnaissance aircraft flying through the TC. Tracking the mass field envelope eliminates the effect of eye wobble on the track.

Another technique is the best track method (Fig. 4.3). This method uses subjective weighting of various types of fixes to determine motion trends over a 24-hour period.

Some forecast agencies (e.g., Joint Typhoon Warning Center and National Hurricane Center) typically try to eliminate eye wobble and suspected fix errors from official tracks. This is why the official working best-track position is frequently in conflict with fix positions.

***** End of SECTION ONE *****

Main

Chapter 4

Section 1