www.srh.noaa.gov

Site Map

News

Organization

Search

A record-breaking flood occurred in the streams in and near Fort Worth, Tex., on May 17, 1949, as a result of heavy rains falling during the previous night. Ten lives were lost and great damage was done in Fort Worth largely by flood in [and along the] Clear Fork [of the] Trinity River. This was undoubtedly the greatest flood in Clear Fork at Fort Worth since 1908 or possibly longer ¹. Smaller streams tributary to West Fork Trinity River between Fort Worth and Dallas had very large floods and contributed materially to the runoff at Grand Prairie and Dallas. The Elm Fork [of the] Trinity River, although not adding greatly to the peak flow at Dallas, did contribute materially to the runoff of the Trinity River at Dallas.

The failure of levees in Fort Worth along Clear Fork caused great damage to industrial and residential property. The Fort Worth water plant, located in the flood plain of Clear Fork and protected by inadequate levees, was flooded. Backwater along West Fork [of the] Trinity upstream from mouth of Clear Fork caused considerable damage to property in unprotected areas. Failure of a levee along right bank of West Fork Trinity a short distance above 12th Street bridge in Fort Worth caused considerable damage in that area. . . .

Considerable flood damage was done downstream from Fort Worth but was confined largely to roads, bridges, and crop and soil losses. Total direct damage from the flood was estimated to be in excess of fifteen million dollars, of which eleven million was in the city of Fort Worth.

The purpose of this report is to present certain rainfall and runoff data in the flood area in greater detail than is usually found in regular Water-Supply Papers.

This report was prepared in the Water Resources Division of the Geological Survey, S. G. Paulsen, chief hydraulic engineer. Field data were obtained and computations made by the Geological Survey Fort Worth area office, L. N. Jorgensen, engineer-in-charge. Special field data were obtained and analyses made under the immediate supervision of Hollister Johnson of the Washington office of the Geological Survey.

Base data used in this report were obtained in cooperation with the Texas State Board of Water Engineers and the Corps of Engineers, Department of the Army. The Corps of Engineers furnished special help in obtaining field data for the computation of maximum discharge by slope-area method for Clear Fork Trinity River near Benbrook, and a statement of damages resulting from the flood. Rainfall data, including records from official stations and unofficial records obtained by field surveys, were furnished by the Weather Bureau [now known as the National Weather Service].

Records show that the heavy rainfall producing the flood in Clear Fork Trinity River fell mostly between 6 p.m. May 16 and 3 a.m. May 17 with the maximum measured being 11 inches [28 centimeters]. A maximum of 12 inches [31 centimeters] was measured on Village Creek in West Fork Trinity River basin southeast of Fort Worth. Rainfall at Dallas and vicinity fell mostly during the fist 6 hours of May 17. Records show that the storm moved slowly from west to east and generally downstream through the Clear Fork and West Fork Basins from west of Fort Worth to Dallas. Isohyetal lines [(lines of equal rainfall depth)] shown on map (figure 1) were based on all available rainfall records. No record from recording gage[s] was available for [the] area of heaviest rainfall. Graphs of cumulative rainfall from recording gages where the rainfall was comparatively heavy, at Cresson and at Tarrant County Courthouse, are shown in figure 2. Cresson is southwest of Fort Worth at the headwaters of Mustang Creek.

The following discussion of the meteorological situation in north-central Texas on May 16, 17, 1949 was prepared by A. J. Knarr, Meteorologist, Weather Bureau Airport Station, Fort Worth, Tex., June 20, 1949:

Fort Worth's worst flood of history was caused by very heavy rains that began in the late afternoon of May 16 and ended early in the morning of May 17.

At 0030C[DT] [(0530 UTC)], May 16, preceding the thunderstorms, the southeastern United States was dominated by a stable anticyclonic cell [high pressure system]. At the same time a weak pressure trough extended from West Texas along the eastern slope of the Rocky Mountains to Montana. This pressure distribution resulted in a flow of air from south to north over all of Texas except that section west of the Guadalupe Mountains. The southerly flow extended from the surface to at least 10,000 feet [(3,000 meters)] and was characterized by high moisture content (mixing ratio approximately 10g [of water] / kg [of air]) from the surface to about 7,000 feet [(2,100 meters)]. Above this moist layer the air was comparatively dry and the lapse rate [(the decrease of temperature with height)] relatively steep. Constant pressure charts for 2200E[DT], May 15 [0200 UTC May 16], showed a discontinuity in wind, temperature and moisture near, and parallel to, the New Mexico-Texas border.

In the succeeding six hours the front moved to a position just east of Dodge City, Kansas, and just west of Abilene and Ozona, Texas. By this time it was causing thunderstorms rather generally throughout its length. By 1230C [(1730 UTC)] it had progressed to a position from Wichita, Kansas, to just east of Wichita Falls, then to just west of Junction and Del Rio, Texas. Thunderstorms still were observed along and practically its entire length. The upper air picture at 1000E (1400 UTC) May 15, showed the south to north flow of moist air continuing over east and central Texas but the discontinuity aloft had moved to ... eastward of Dodge City, Kansas, and Big Spring, Texas. The winds at 700[mb] at these stations had shifted to southwesterly, temperatures had dropped slightly, and moisture had decreased sharply.

Throughout the afternoon the upper cold front moved steadily eastward to a position at 1830C [(2330 UTC)] extending from Kansas City to just east of McAlester, Oklahoma, and Dallas, Texas to San Antonio. At that time thunderstorms were in progress at Mineral Wells, Fort Worth, and Dallas. After 1830C [(2330 UTC)], May 16, the northern end of the front continued moving eastward but the southern end remained nearly stationary. This resulted from deepening low pressure aloft over the northern Rocky Mountains which caused a backing in winds aloft over Texas. Therefore, as the upper front moved to east Texas, the upper winds behind it were becoming more nearly parallel to the front. This resulted in a much slower movement.

Since the moisture content of the air was great enough to produce heavy showers in the frontal zone due to steepening of the lapse rate aloft and to convergence of southwest and south winds, the duration factor became all-important. Throughout the night of May 16-17 the front moved so slowly that thunderstorms continued in the Fort Worth area, ending only during the morning of the 17th when the converging wind-flow diminished and the greatly weakened front moved eastward.

Further development of the northern Rocky Mountain disturbance was such as to restore the south to north wind-flow over all of Texas. Absence of excessive instability and convergence resulted in no rain of consequence in the state during the next few days.

Stage and discharge records obtained at eight stream-gaging stations in the Trinity River Basin during the flood of May 1949 are given on the following pages. The stations are listed in table 1 and the locations of the stations are shown in figure 1. Graphs of discharge plotted from station records are shown in figure 3 and figure 4. Complete records for these stations are published regularly in the annual series of reports on Surface Water supply of the United States, Part 8, Western Gulf of Mexico Basins, except that publications of West Fork Trinity River at Lake Worth dam above Fort Worth was discontinued in 1934. Daily Records of stage and discharge are available for these stations as follows:

The gaging-station records on the following pages consist of a station description, a table of daily mean discharges and daily and total runoff for the flood period, and a table of discharges at indicated times during the flood in sufficient detail for a reasonably delineation of the hydrograph. Under the heading "Maxima", the first paragraph gives the maximum discharge and gage height that occurred during the flood of May 1949; the second paragraph gives the period of know floods and the maximum discharge and gage height where known, that occurred prior to the flood of May 1949. It may be assumed that no reliable information was available concerning floods prior to the earliest date shown in the second paragraph.

A system of levees for the Trinity River at Dallas was completed in 1930 and in June of that year the old river channel was abandoned and flow was diverted through the new channel. A pilot channel was dug through the levee system to carry low flow. Records of flow were obtained from gage at Commerce Street bridge from July 1903 to June 1930 and from October 1932 to date. Records were obtained at site 6 miles [9.6 kilometers] downstream July 1930 to September 1932. First discharge measurements at Commerce Street gage site after completion of levee system were made in 1932. Measurements made in 1932 when compared with subsequent measurements show that the pilot channel had not obtained its maximum scour in 1932. Subsequent to 1932, substantial flood ratings were developed by discharge measurements in 1935, 1942, 1945, and 1949. These ratings, show in figure 5, show a continuing decrease in conveyance of the leveed channel.

1. Landis, D. S., "Rainfall and flood at Fort Worth, Tex. Apr 24-25, 1922": U.S. Weather Bureau Mo. Weather Rev., Vol. 50, pp. 188-189, April 1922.