1
An Evaluation of Flash Flood Warnings and Flash Flood Reports for South Central Texas Kevin Barrett Texas State University-San Marcos Abstract Flash flood warnings issued by the Austin/San Antonio forecast office of the National Weather Service from 1986 to 2006 were spatially evaluated to find possible trends related to population growth and change in government severe weather warning policy. It was found that almost three times the number of warnings were issued in the five year time period of 2001 to 2006 compared to the decadal time period of 1986 to 1995, with the highest number of warnings trending to center on urban areas and the location of the National Weather Service office. When the warnings are compared to actual flash flood reports, a strong statistical correlation exists suggesting a low false alarm rate for the area. Results indicate that National Weather Service modernization, technological upgrades in flood monitoring, public awareness of flood dangers and population growth may be factors contributing to the spatial distribution of warnings. AP Photo/ Waco Tribune Herald, Jerry Larson Comparison P-Value Warnings 86-95 & Population 0.39 0.025 Warnings 96-05 & Population 0.17 0.344 Reports 86-95 & Population 0.63 < 0.001 Reports 96-05 & Population 0.33 0.061 Warnings 95-00 & Population 0.16 0.374 Warnings 01-06 & Population 0.19 0.290 Reports 95-00 & Population 0.17 0.344 Reports 01-06 & Population 0.43 0.013 Warnings 86-95 & Reports 86-95 0.82 < 0.001 Warnings 96-05 & Reports 96-05 0.83 < 0.001 Warnings 95-00 & Reports 95-00 0.85 < 0.001 Warnings 01-06 & Reports 01-06 0.73 < 0.001 Comparison Bexar County P-Value Warnings 93-06 & Reports 93-06 0.87 < 0.001 0 10 20 30 40 50 60 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Warnings 1986-2006 Bexar County Single County Analysis National Weather Service Modernization The above map comparison illustrates the drastic increase in the number of flash flood warnings issued after the National Weather Service modernization effort. The most populated county in the study area is Bexar. The number of warnings for this county increased drastically after the National Weather Service modernization effort, starting in 1997. In 2002 and 2004, significant rain events caused major flooding for the county. The county shows a high statistical correlation for issued warnings and flash flood reports for the time period of 1993-2006 (1993 is the first year that accurate flash flood reports are available). Statistical Analysis Statistical comparisons show a very high correlation between warnings and reports. Graph Comparison 0 500 1000 1500 2000 2500 3000 3500 4000 1995-2000 2001-2006 Warnings Reports 0 500 1000 1500 2000 2500 3000 3500 4000 1986-1995 1996-2005 Warnings More than four times the number of warnings were issued from 1996-2005 compared to 1986-1995. There is also an increase in the number of warnings issued between 2001-2006 compared to 1995-2000. Post National Weather Service Modernization The above maps compare warnings and reports for time periods after the National Weather Service modernization. The majority of warnings and reports are from the more populated counties in the study area. Findings National Weather Service modernization corresponds to an overall increase in the number of flash flood warnings issued. The strong statistical correlation between flash flood warnings issued and flash flood reports suggest a low false alarm rate. The spatial pattern of flash flood warnings suggests the warning forecast favor counties with higher populations. Extreme rainfall events in 2002 and 2004 exaggerate the findings for the later time period of this analysis. Further Study Compare flash flood warnings and reports to physiographic features known to have an influence on flash flooding. Compare the spatial distribution of flood monitoring stations to flash flood warnings and reports. Determine exact false alarm rates for each county in the study area. Test the theory that the population of certain counties can tolerate a higher number of false alarms. Compare correlations of population demographics to flash flood warnings and reports. Build a warning systems computer model that can test various influences on flash flood warning forecast decisions. Purpose The purpose of this study is to spatially analyze flash flood warnings issued by the National Weather Service office in New Braunfels, Texas to determine if there are patterns related to population and government severe weather warning policy. Study Area The study area is the 33 county warning area (CWA) covered by the National Weather Service Office located at New Braunfels, Texas. The most populated counties are Bexar (1,392,931) and Travis (812,280). Population data are from the United States Census Bureau 2000 census. San Antonio Austin Dallas Houston All flash flood warning data from the National Weather Service. All flash flood report data from the National Climatic Data Center. Contact: [email protected] © 2009

An Evaluation of Flash Flood Warnings and Flash Flood

  • Upload
    others

  • View
    6

  • Download
    0

Embed Size (px)

Citation preview

Page 1: An Evaluation of Flash Flood Warnings and Flash Flood

An Evaluation of Flash Flood Warnings and Flash Flood Reports for South Central TexasKevin Barrett

Texas State University-San MarcosAbstract

Flash flood warnings issued by the Austin/San Antonio forecast office of the National Weather Service from 1986 to 2006 were spatially evaluated to find possible trends related to population growth and change in government severe weather warning policy. It was found that almost three times the number of warnings were issued in the five year time period of 2001 to 2006 compared to the decadal time period of 1986 to 1995, with the highest number of warnings trending to center on urban areas and the location of the National Weather Service office. When the warnings are compared to actual flash flood reports, a strong statistical correlation exists suggesting a low false alarm rate for the area. Results indicate that National Weather Service modernization, technological upgrades in flood monitoring, public awareness of flood dangers and population growth may be factors contributing to the spatial distribution of warnings.

AP Photo/ Waco Tribune Herald, Jerry Larson

Comparison R² P-ValueWarnings 86-95 & Population 0.39 0.025

Warnings 96-05 & Population 0.17 0.344

Reports 86-95 & Population 0.63 < 0.001

Reports 96-05 & Population 0.33 0.061

Warnings 95-00 & Population 0.16 0.374

Warnings 01-06 & Population 0.19 0.290

Reports 95-00 & Population 0.17 0.344

Reports 01-06 & Population 0.43 0.013

Warnings 86-95 & Reports 86-95 0.82 < 0.001

Warnings 96-05 & Reports 96-05 0.83 < 0.001

Warnings 95-00 & Reports 95-00 0.85 < 0.001

Warnings 01-06 & Reports 01-06 0.73 < 0.001

Comparison Bexar County R² P-Value

Warnings 93-06 & Reports 93-06 0.87 < 0.001

0

10

20

30

40

50

60

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Warnings 1986-2006 Bexar County

Single County Analysis

National Weather Service Modernization

The above map comparison illustrates the drastic increase in the number of flash flood warnings issued after the National Weather Service modernization effort.

The most populated county in the study area is Bexar. The number of warnings for this county increased drastically after the National Weather Service modernization effort, starting in 1997. In 2002 and 2004, significant rain events caused major flooding for the county. The county shows a high statistical correlation for issued warnings and flash flood reports for the time period of 1993-2006 (1993 is the first year that accurate flash flood reports are available).

Statistical Analysis

Statistical comparisons show a very high correlation between warnings and reports.

Graph Comparison

0

500

1000

1500

2000

2500

3000

3500

4000

1995-2000 2001-2006

Warnings

Reports

0

500

1000

1500

2000

2500

3000

3500

4000

1986-1995 1996-2005

Warnings

0

500

1000

1500

2000

2500

3000

3500

4000

1986-1995 1996-2005

Warnings

More than four times the number of warnings were issued from 1996-2005 compared to 1986-1995. There is also an increase in the number of warnings issued between 2001-2006 compared to 1995-2000.

Post National Weather Service Modernization

The above maps compare warnings and reports for time periods after the National Weather Service modernization. The majority of warnings and reports are from the more populated counties in the study area.

Findings•National Weather Service modernization corresponds to an overall increase in the

number of flash flood warnings issued.

•The strong statistical correlation between flash flood warnings issued and flash

flood reports suggest a low false alarm rate.

•The spatial pattern of flash flood warnings suggests the warning forecast favor

counties with higher populations.

•Extreme rainfall events in 2002 and 2004 exaggerate the findings for the later

time period of this analysis.

Further Study•Compare flash flood warnings and reports to physiographic features known to have an

influence on flash flooding.

•Compare the spatial distribution of flood monitoring stations to flash flood warnings and

reports.

•Determine exact false alarm rates for each county in the study area.

•Test the theory that the population of certain counties can tolerate a higher number of

false alarms.

•Compare correlations of population demographics to flash flood warnings and reports.

•Build a warning systems computer model that can test various influences on flash flood

warning forecast decisions.

PurposeThe purpose of this study is to spatially analyze flash flood warnings issued

by the National Weather Service office in New Braunfels, Texas to determine if there are patterns related to population and government severe weather warning policy.

Study Area

The study area is the 33 county warning area (CWA) covered by the National Weather Service Office located at New Braunfels, Texas. The most populated counties are Bexar (1,392,931) and Travis (812,280). Population data are from the United States Census Bureau 2000 census.

San Antonio

Austin

Dallas

Houston

All flash flood warning data from the National Weather Service.All flash flood report data from the National Climatic Data Center. Contact: [email protected]

© 2009