IX. What are our most important significant research findings? What do we believe caused the Kalamazoo tornado? What can we learn from this? How can we use this information to aid in anticipating tornadogenesis in the GRR CWA?

Boundaries are absolutely critical in assessing the storm environment and tornado potential. It is believed that one of the key factors in the development of significant tornadoes is the altered low-level wind field in their vicinity. Backed low-level wind fields as a result of outflow boundaries, near warm fronts, etc. contribute to greater 0-1 km storm relative helicity values. Around the time of the 13 May 1980 Kalamazoo tornado, the quasi-stationary front was sagging south to near (or just a few miles north of) the city of Kalamazoo. In this environment, as you would expect, southerly surface winds were observed south of the boundary (as reflected by the 4 p.m. Kalamazoo observation, actually taken at 3:50 p.m. at the airport in Portage, seven miles south of the “downtown” area of Kalamazoo). This is important because seven miles to the north in “downtown” Kalamazoo, we suspect the boundary was right over, or just a few miles north of, this area. This implies that surface wind fields in downtown Kalamazoo may have been out of the east or southeast thereby contributing to greater 0-1 km SRH.

The latest tornado research from the Warning Decision Training Branch, Storm Prediction Center, and the National Severe Storms Laboratory is putting much greater emphasis on the importance of 0-1 km SRH, and this value as a percentage of 0-3 km SRH. Markowski (1998) found that 0-1 km SRH is likely more important for tornado development than 0-3 km SRH. In environments associated with significant tornadoes (> F1) the 0-1 km SRH was typically greater than 50% of the 0-3km SRH.

This was your classic “Type One” high shear, relatively low instability environment. It is my belief that being near or just south of the boundary, and with surface temperatures rising to 73-74 degrees, that there was enough surface based instability to generate surface based thunderstorms. With sufficient instability in place the deep layer shear was very supportive of supercell development. In addition, the low-level helicity was certainly enhanced in the vicinity of the surface boundary where the low level flow was backed to the east-southeast. It is likely that a supercell in close proximity to the boundary was able to ingest a substantial amount of low-level vorticity into it’s updraft aiding in the development of a substantial tornado. In addition, a low LCL height is indicative of environments supportive of strong tornadoes in that supercells that form in environments with low LCL are less likely to have their mesocyclones undercut by a substantial downdraft generated cold pool. Recent tornado research by WDTB (2002 tornado warning guidance) also alludes to these “warm RFD” characteristics that aid in tornadogenesis. The environment south of the warm front on 13 May 1980 was very conducive to low LCL heights.

It is important to dispel the notion that we need high temperature and dew point values and tremendous instability in addition to speed and directional shear in order to generate a significant tornado. Although this would be the ideal supercell and tornado environment, in reality, in Michigan, this does not happen all that often. In fact, our Significant Tornado Climatology for Lower Michigan indicated that the majority of F3 or stronger tornadoes in southern Lower Michigan developed in environments with surface temperatures only in the 50’s and 60’s.

As we have known for a long time, time of day is also critical. Almost every significant tornado case we examined in the Significant Tornado Climatology for Lower Michigan occurred during the afternoon or evening (one occurred in the morning). As with this Kalamazoo case, the majority occurred in the afternoon. At night, exorbitant values of speed and directional wind shear do not necessarily imply that the tornado risk is high. Without sufficient instability, this rotational potential will never be realized and ingested into the updraft of a developing thunderstorm. Often times tremendous amounts of 0-1 km and 0-3 km SRH are mitigated by the fact that we have a stable layer of air near the surface or one in which elevated convection well north of a warm front will not be affected by this low level shear.

The following are a few extremely important excerpts from the 2002 Tornado Warning Guidance document from WDTB, which re-emphasizes the importance of near-ground SRH and boundaries:

“In terms of the helicity ingredient, supercells tend to produce significant tornadoes in regions with enhanced near-ground SRH. In many situations, enhanced low-altitude SRH will be associated with locally backed and strengthened surface winds. This SRH enhancement can be on the order of from 200 to 300 m2/s2 initially to near 1000 m2/s2 over a very short distance prior to tornadogenesis. The magnitude of this enhancement was observed in some VORTEX cases. All available low-altitude wind data should be monitored, including routine surface observations, mesonet data, lowest-tilt radial velocity, and wind profiler data (where available).”

“In many situations where significant tornadoes (>F1) occur, meso-beta scale enhancement of low-level SRH, or the combination of low-level SRH and CAPE, as quantified by parameters such as Vorticity Generation Parameter (VGP) or Energy Helicity Index (EHI), develops in conjunction with baroclinic boundaries. This augmentation occurs above what is normally observed in synoptic scale environments associated with tornado outbreaks. Thus, closely monitor storms moving into areas of enhanced low-level SRH and CAPE based on integrated sensor analysis. Even lacking wind and temperature data, the mere presence of a boundary should lead to heightened awareness, and storms crossing or interacting with boundaries merit special scrutiny for rapid increases in rotation in their lower altitudes. This implies that forecasters need to remain aware of the locations of radar fine lines, satellite-indicated cloud lines, and mesonet-detected surface temperature gradients and wind-shift lines.”

“Exercise heightened awareness on storms interacting with boundaries - these should be closely monitored because the likelihood for tornadogenesis is greater for storms interacting with boundaries. Although many storm-boundary interactions do not result in tornadoes, if rapid mesocyclogenesis is observed in the radar data after boundary interaction, a tornado warning most likely should be issued.”

I think one more really important point was addressed by Markowski, Straka, Rasmussen and Blanchard in their paper titled “Variability of Storm-Relative Helicity during VORTEX” (Monthly Weather Review, November 1998, pp. 2959-2971). They note that “important SRH enhancement can be anticipated near boundaries using conventional data. If SRH is to be a useful forecast parameter, forecasters must be sensitive to the potential for highly variable SRH in regions where deep convection is occurring and boundaries are being generated. The application of a single value of SRH in a forecast in many cases may be inappropriate. Boundaries may supply additional buoyancy-generated streamwise vorticity in environments where only marginal values of SRH for severe storms are indicated.”

This definitely could have applied in the Kalamazoo tornado case. Though the value of 0-1 km SRH derived from the 18 UTC modified hodograph for surface winds in Kalamazoo was modest (35 m 2/s 2), I think this is where forecaster instincts really come into play. We do not have enough wind data observations from ASOS, mesonets, wind profilers, etc. to account for very small scale (but potentially large) degrees of variability in SRH. This is where pattern recognition, anticipation, and simply having the situational awareness to realize that we are in the heart of severe weather season in lower Michigan, with a very strong boundary over our area, and other factors discovered in our Significant Tornado Climatology for Lower Michigan all come together to heighten awareness of tornado potential. Please keep in mind that locally, near these boundaries, SRH values can vary greatly. It is more important to understand the storm environment and that there is a high degree of localized variability of SRH in the vicinity of boundaries.

  1. Introduction
  2. Methodology
  3. Large Scale Synoptic Pattern over the United States on May 13, 1980
  4. Hourly Surface Weather Maps Focused on Great Lakes Region from 12 UTC May 13 to 00 UTC May 14
  5. Observed 12 UTC Soundings for Flint, MI and Peoria, IL and Data Derived from them
  6. Modified Flint Sounding
  7. Flint and Peoria hodographs from 12Z observed soundings
  8. Modified Peoria Hodograph (using 18 UTC surface winds in AZO)
  9. What are our most important significant research findings? What do we believe caused the Kalamazoo tornado? What can we learn from this? How can we use this information to aid in anticipating tornadogenesis in the Grand Rapids CWA?
  10. So what exactly happened? Chronology of events occurring between 3:30 and 4:25 p.m. EDT across Van Buren and Kalamazoo Counties.
  11. Tornado Victims
  12. Dr. T. Theodore Fujita’s Kalamazoo Tornado Findings
  13. A Personal Account of the Kalamazoo Tragedy
  14. Bronson Park Devastated
  15. Acknowledgments
Return to The May 13, 1980 Kalamazoo Tornado Case Study Main Page

USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services.