End of Summer Downbursts in Oklahoma and Validation Results

On the last full day of the Summer season, 21 September 2009, strong convective storms developed along a cold front as it was tracking eastward over Oklahoma. In general, the pre-convective environment downstream of the cold front over eastern Oklahoma was moist and very unstable with a shallow, well-mixed boundary layer. These conditions favored convective storms that produced strong outflow winds (downbursts) associated with heavy rain. Elevated Geostationary Operational Environmental Satellite (GOES) microburst index values, indicated in the 1800 UTC product images in the vicinity of downburst occurrence over eastern Oklahoma, served as evidence of the presence of this favorable environment. Strong downbursts that were recorded by Oklahoma Mesonet stations between 2100 and 0000 UTC 22 September resulted from a combination of heavy precipitation and sub-cloud evaporation of precipitation.

Figure 1. Geostationary Operational Environmental Satellite (GOES) imager microburst product at 1800 UTC 21 September 2009, with the location of Oklahoma mesonet observations of downburst wind gusts plotted on the image.

Figure 2. Geostationary Operational Environmental Satellite (GOES) Microburst Windspeed Potential Index (MWPI) product at 1800 UTC 21 September 2009, with the location of Oklahoma mesonet observations of downburst wind gusts plotted on the image.

Figures 1 and 2 display elevated microburst risk (yellow color) ahead of a cold front tracking eastward over Oklahoma. The cold front served as a trigger for convective storms over eastern Oklahoma during the afternoon and evening of 21 September 2009. Elevated Geostationary Operational Environmental Satellite (GOES) microburst index values displayed in the 1800 UTC product images in the vicinity of downburst occurrence over eastern Oklahoma, indicated wind gust potential of 35 to 50 knots. Downbursts wind gusts between 39 and 50 knots were recorded by the Oklahoma Mesonet stations plotted in Figures 1 and 2 between 2040 and 0000 UTC 22 September.

Validation results for the 2007 to 2009 convective seasons have been completed for the MWPI and imager microburst products. GOES sounder-derived MWPI values have been compared to mesonet observations of downburst winds over Oklahoma and Texas for 168 events between June 2007 and September 2009. The correlation between MWPI values and measured wind gusts was determined to be .62 and was found to be statistically significant above the 99% confidence level, indicating that the correlation represents a physical relationship between MWPI values and downburst magnitude and is not an artifact of the sampling process. Comparison of GOES-11 imager microburst risk values (output brightness temperature difference (BTD) in degrees K) to measured downburst wind gusts for 61 events in Oklahoma between June and September 2009 yielded a correlation of .40. This correlation was higher than the correlation computed between MWPI values and downburst wind gusts (.27) for the same time period. The correlation between output BTD and measured wind gusts was determined to be statistically significant at the 82% confidence level, indicating a high confidence that the correlation represented a physical relationship between output BTD values and downburst magnitude.

Figure 3. Scatterplot of Geostationary Operational Environmental Satellite (GOES) Microburst Windspeed Potential Index (MWPI) versus observed downburst wind gust speed as recorded by mesonet stations in Oklahoma and Texas between June 2007 and September 2009.

Figure 4. Scatterplot of Geostationary Operational Environmental Satellite (GOES) -11 imager output BTD values (K) versus observed downburst wind gust speed as recorded by mesonet stations in Oklahoma between June and September 2009.

Figures 3 and 4 are scatterplots of MWPI values and GOES-11 imager output BTD values (K) versus observed downburst wind gust speed as recorded by mesonet stations in Oklahoma and Texas. The MWPI scatterplot identifies two clusters of values: MWPI values less than 50 that correspond to observed wind gusts of 35 to 50 knots, and MWPI values greater than 50 that correspond to observed wind gusts of greater than 50 knots. Similarly, the GOES-11 imager microburst risk scatterplot identifies two clusters. The dominant cluster contains output brightness temperature difference (BTD) values less than 50K that correspond to observed wind gusts between 35 and 50 knots. The scatterplots illustrate that both microburst products demonstrate effectiveness in distinguishing between severe and non-severe convective wind gust potential.

Oklahoma Spearhead Echoes and Downbursts

During the afternoon of 9 September 2009, scattered convective storms developed along a weak cold front as it was tracking southeastward over Oklahoma. Although there was a weak temperature contrast across the front, the front acted as a convergence zone and a trigger for deep, moist convection. The pre-convective environment downstream of the cold front over western Oklahoma was dominated by vertical mixing that fostered the development and evolution of a convective boundary layer. Elevated Geostationary Operational Environmental Satellite (GOES) imager brightness temperature difference (BTD) values in the vicinity of downburst occurrence over western Oklahoma served as evidence of the presence of a well-developed mixed layer. Strong downbursts that were recorded by Oklahoma Mesonet stations in southwestern Oklahoma between 2145 and 2250 UTC resulted from a combination of precipitation loading and sub-cloud evaporation of precipitation. These downbursts occurred in proximity to high microburst risk values as indicated in the 2000 UTC imager microburst product.


The images above are a Geostationary Operational Environmental Satellite (GOES) imager microburst product with overlying radar reflectivity from Frederick, Oklahoma NEXRAD (KFDR) at 2145 UTC (top) and 2249 UTC (bottom), times of downburst occurrence at Medicine Park and Apache Oklahoma Mesonet stations, respectively. The product image displayed scattered convective storms developing along the cold front over western Oklahoma. Downburst wind gusts of 54 and 41 knots were recorded at Medicine Park and Apache stations at 2145 and 2250 UTC, respectively. Apparent in both images is the spearhead echo associated with both downbursts, occurring in a region of high microburst risk as indicated by the orange shading. Output BTD near 50K corresponded to wind gust potential near 50 knots. Fujita and Byers (1977) related the spearhead echo signature in radar reflectivity imagery to the occurrence of downbursts as illustrated below.


The Rapid Update Cycle model (RUC) analysis sounding above at 2000 UTC at Medicine Park shows an "inverted v" profile, and indicates the presence of a deep and dry mixed layer that favored the development of intense downdrafts due to the evaporation of precipitation in the sub-cloud layer. This profile corresponded well with high output BTD associated with downbursts and resultant spearhead echoes over southwestern Oklahoma.

References

Fujita, T. T., and H.R. Byers, 1977: Spearhead echo and downbursts in the crash of an airliner. Mon. Wea. Rev., 105, 1292-146.

2009 Preliminary Microburst Product Validation Results

Preliminary validation results for August 2009 have been completed for the Geostationary Operational Environmental Satellite (GOES) Microburst Windspeed Potential Index (MWPI) and imager microburst products. GOES sounder-derived MWPI values have been compared to mesonet observations of downburst winds over Oklahoma and Texas for 156 events between June 2007 and August 2009. The correlation between MWPI values and measured wind gusts was determined to be .64 and was found to be statistically significant above the 99% confidence level, indicating that the correlation did represent a physical relationship between MWPI values and downburst magnitude and was not an artifact of the sampling process. Comparison of GOES-11 imager microburst risk values (output brightness temperature difference (BTD) in degrees K) to measured downburst wind gusts for 49 events in Oklahoma between June and August 2009, yielded a correlation of .42. This correlation was higher than the correlation computed between MWPI values and downburst wind gusts (.32) for the same time period. The correlation between output BTD and measured wind gusts was determined to be statistically significant at the 85% confidence level. The results of this t-test also indicated a high likelihood (85%) that the correlation represented a physical relationship between output BTD values and downburst magnitude.

The images above are scatterplots of Geostationary Operational Environmental Satellite (GOES) Microburst Windspeed Potential Index (MWPI) values (top) and GOES-11 imager output BTD values (K) (bottom) versus observed downburst wind gust speed as recorded by mesonet stations in Oklahoma and Texas between June 2007 and August 2009 (top) and mesonet stations in Oklahoma between June and August 2009 (bottom). The MWPI scatterplot identifies two clusters of values: MWPI values less than 50 that correspond to observed wind gusts of 35 to 50 knots, and MWPI values greater than 50 that correspond to observed wind gusts of greater than 50 knots. Similarly, the GOES-11 imager microburst risk scatterplot identifies two clusters. The dominant cluster contains output BTDs less than 50K that correspond to observed wind gusts between 35 and 50 knots. The scatterplots illustrate that both microburst products demonstrate effectiveness in distinguishing between severe and non-severe convective wind gust potential.