For the outlook forecast interval, the following question is on the forecaster's mind: Is today's environment favorable for microburst development? Two indices based upon sounding information that help answer our question. These indices are the Microburst-Day Potential Index (MDPI) and the Wind Index (WINDEX).

MDPI was developed at the 45WS and NASA/Applied Meteorology Unit (AMU) as a result of a microburst event on 16 August 1994 (Wheeler 1996, 1997). This index utilizes vertical profiles of environmental equivalent potential temperature, . The resulting equation, based in part on the results from the Microburst and Severe Thunderstorm (MIST) project (Atkins and Wakimoto, 1991), is
 

,
where the first term in the numerator represents the warm, moist low-level air and the second term in the numerator represents the dry, cool midlevel air. The greater the spread between these values, the greater the effect that relatively cool, dry environmental air will have on a developing downburst. The 30 K in the denominator is a local tuning constant (~20 K was the threshold from the MIST results) such that when MDPI is greater than or equal to one, microbursts are likely. Since the MDPI does account for the probability of convection, use of MDPI as a probabilistic forecast tool requires the assumption that convection will form on a given day.

MDPI was designed to determine likely and unlikely environments for downbursts (30 kts, not severe downbursts 50 kts). The intensity of downbursts that may form is not predicted by the index. When computing MDPI, it is best to use complete vertical profiles of . It is particularly important to have complete data at midlevels where the minimum in is located so that it is measured accurately. It is also important to accurately predict how the atmosphere will change during the day. Forecasters want to know what the MDPI will be at the time of convection, not at the time of the sounding. Fortunately, future values of MDPI are not difficult to predict (Fig. 2).

The 45WS tested MDPI over two convective seasons and achieved nearly similar results each year (Wheeler, 1996; Wheeler, 1997). During the 1995 season, 84 events were verified; 92 events were verified during the 1997 season. The results of this verification study are shown in Table 2. The statistics for each year were very similar (the greatest difference was 2%), showing great promise for the MDPI as a predictive index.

Another tool that is useful is the WINDEX. WINDEX, which was introduced by McCann (1994) and based upon research by Proctor (1989) and Wolfson (1990), predicts the maximum expected gust possible for a given environmental sounding. The WINDEX equation is
 

,
where
 
WI = Maximum wind gusts (kts)
H_M = Height of melting level (km)
R_Q = Q_L/12 but not >1
= Lapse rate (°C km^-1) from melting level to surface
Q_L = Mixing ratio in lowest 1 km
Q_M = Mixing ratio at melting level.
 
A value for WINDEX can be computed from a morning sounding and then updated later in the day (see

This problem does limit the effectiveness of WINDEX for wet microburst forecasting. Anecdotal evidence at the 45WS suggests that WINDEX may have some usefulness for wet microbursts, though. For instance, if the environment has the capability to support strong, wet microbursts and has a WINDEX of 49 at the time of convection, then the forecaster can use WINDEX as a lower (instead of upper) bound. Such a situation may call for a forecaster to issue a convective wind warning with wording suggesting outflow winds may gust to severe levels (>=50 kts). Although this method is subjective, it provides the best currently known use of WINDEX for wet microburst forecasting.