NOAA's National Weather Service Weather Forecast Office

Introduction

The MARC velocity signature is a Doppler radar-based precursor towards forecasting the initial onset of damaging straight-line winds in a linear MCS or bowing convective system. One of the challenges in the severe storm warning process is forecasting the initial onset of damaging winds. Since the late 1970's, the bow echo reflectivity signature has been used extensively as a severe weather signature for forecasting damaging winds. However, storm surveys and observations have shown that damaging winds often precede the initial bowing of the convective line segment. Thus we looked at other reflectivity and velocity precursors in forecasting the initial onset of damaging winds.

Our initial thoughts in tackling this problem keyed upon studies associated with the isolated pulse microbursts storms. Eilts et al. (1996) from the National Severe Storms Laboratory examined Doppler data from 85 microburst / downburst cases across Oklahoma. They discovered that one of the three most important precursors in forecasting damaging winds included strong and deep convergence at the storm's mid-altitudes. We fully understood that the dynamics between a pulse 'isolated' storm and a linear convective line segment comprised of a collection of storms are radically different. However, we wanted to observe how effective this velocity signature might be with a linear convective line.  Thus, we tested Mid-Altitude Radial Convergence for the first time along with Vertical Integrated Liquid (VIL) on the June 8, 1994 linear MCS which produced damaging winds over parts of the St. Louis metro area and adjacent counties.  

After this initial case, we surveyed MARC on 15 linear MCSs which evolved across the Mid-Mississippi Valley Region during the period from 1992 through 2000.  In each of these events, the linear convective line eventually evolved into bowing structures. Swaths of damaging winds occurred with 10 of the 11 events. The overall length of the linear MCSs ranged from 30 to as long as 140 km.  We initially surveyed the reflectivity and storm-relative velocity fields of each MCS which produced damaging winds, beginning with the 2 July 1992 derecho over Missouri and Illinois.  We often discovered that strong Mid-Altitude Radial Convergence sometimes extending as long as 120 km was present along the forward or downshear side of a linear convective system and preceded bowing of the line segment.

During the MCS's 'intensifying stage' the radar detects a component of two opposing currents along the MCS's forward flank. When viewing a storm approaching from the west or northwest, the region of strong outbound velocities signifies a component of the MCS's steep sloping updraft current or front-to-rear flow. The region of strong inbound velocities, depicts a component of the storm's 'steep sloping convective-scale downdrafts and origins of the mesoscale rear inflow jet (RIJ) (see Rasmussen and Rutledge 1993 AMS JAS).  Strong convective cells along the leading convective line will create a stronger warm pool aloft which in turns enhance horizontal vorticity thus intensifying the mesoscale rear inflow jet.  Remember, the mesoscale RIJ oftens forms near the rear of the leading convective towers and develops rearward with time. Thus during the 'intensifying stage of an MCS' we often observe tall erect convective towers suggesting a balance between the environmental shear and the MCS's cold pool.   The 'close-coupling' of the inbound and outbound velocity maximums reflects where 'radial convergence is strongest' and signifies the strength of the mid-level, hydrostatic Le'mone Low' in that part of the linear convective MCS.  These enhanced velocity differentials or areas of strong radial convergence are usually located in or just downwind from the high reflectivity cores embedded within the linear convective line segment.

MARC Characteristics:  (Latest Information - July 2001)

These characteristics are based on 15 linear MCSs, 13 which evolved during the afternoon and early
evening hours and one which occurred during the early morning hours.

1) Horizontal extent of the MARC velocity signature varied from 30 to as large as 120 km in length. Local enhanced velocity maximums were embedded within the sheet of inbound (outbound) velocities.  We are testing the relationship between these enhanced velocity maximums and the degree of surface damage.

2) Average depth of MARC was 6.2 km. Strongest magnitudes of radial convergence were located
between 4 and 7 km.

3) Width of the zone of radial convergence varied from 2 to 6 km. Be alert for strong velocity gradients
between the inbound and outbound velocities.

4) Once radial velocity differentials reach 25 m s-1 or greater, the potential for severe straight-line winds increased. Average lead time from the initial detection of this magnitude (25 m s-1) and first reported
severe wind event was 20 minutes.

5) Recent observations revealed that the MARC velocity signature has been observed more frequently
with nearly a solid linear convective line echo compared to discrete convective cells along the southern
flank of an asymmetric MCS.  

6) The viewing angle is extremely important when viewing the MARC velocity signature. The convective
line must be nearly orthogonal to the radial which you are calculating MARC -  Personal discussions
(Jim Wilson - NCAR). 

The example below are plan views of reflectivity and storm-relative velocity images taken from the paper by Schmocker et al. (1996) (AMS - 15th Conf. WAF) showing MARC. Note the strongest MARC is located just downwind from a relatively large 50 dBZ core.  On the SRM image, negative (positive) values are inbound (outbound) velocities respectively. 

The persistent area of locally enhanced velocity differentials within the larger zone of convergence along the forward flank of the convective line appears to linked to the greatest degree of damaging winds. Convective-scale vortices (tornadic as well as non-tornadic) often form in the zone or interface between the two drafts (mainly on the updraft side) where cyclonic or negative horizontal vorticity is strong. For more information on vortex evolution with bow echoes see (convective-scale vortex evolution). 

Another example of identifying MARC was taken from the 22 September 1993 MCS event over the Mid-Mississippi Valley region (the year of the Great Floods across our area).  A nearly solid convective line formed between 2100 - 2130 UTC from northeast through central Missouri (north of Columbia). The linear MCS evolved into a bowing structure after 2240 UTC northwest of KLSX producing several swaths of damaging winds and spawning a number of non-supercell tornadoes. The reflectivity / storm-relative velocity image at 2231 UTC from KLSX (shown below)  is taken at the 2.4° elevation slice.  We are viewing base reflectivity and storm-relative velocity images at approximately 6.0 km. MARC is observed at two locations along the leading edge of the convective line at this time: a) 295° @ 66 nm and b) 304° @ 64 nm.   Magnitudes of MARC at these locations are 23 and 24 m s-1 respectively.
the greatest degree of damaging winds occurred after 2240 UTC, 5 to 20 km southeast of Mexico Missouri across southeast Audrain and into parts of northern Montgomery counties in east-central Missouri.

Reflectivity Image (2.4 degree slice) (left); Storm-Relative Velocity (SRM) image (right) from KLSX for 2231 UTC 22 September 1993.  On the SRM image, outbounds (warm colors - red) represent the MCS's updraft current while inbounds (cool colors - green) signify convective-scale downdrafts.  MARC is detected along the leading edge of the Quasi-Linear Convective System (QLCS)at approximately 6.1 km (20 Kft).  Magnitudes of MARC exceeded 25 m s^-1 over each area.

Table 1:  Pre-convective environments,  MARC characteristics and lead times from the initial identification of MARC to first reports of damaging winds are summarized.