FLOOD01375 - Laserfiche WebLink

Home Browse Search

meteorological conditions associated with a flash flood over many spatial and temporal scales, In the following sections we present a meteorologi- cal overview of the FCL flood, First, a discussion of the climatology of extreme rainfall events in Colorado is presented in section 2, Section 3 discusses rain gauge data and presents accumulated rainfall totals for the entire 2-day event (e.g" 27-28 July 1997), as well as the 6-h time period encompassing the flash flood. Section 4 presents an overview of the synoptic envi- ronment, followed in section 5 by a presentation of surface mesoanalyses, Section 6 summarizes the preconditioning of the large-scale environment and subsequent triggering of convection over FCL. Multiparameter and dual-Doppler radar observations including radar estimates of accumulated rainfall are described in section 7. 2. Climatology of extreme rain events in Colorado The Fort Collins flood occurred in a synoptic en- vironment that is commonly associated with severe weather in northeastern Colorado (Doswell 1980), and during a time of year when extreme weather events are expected in northeastern Colorado (yY eaver and Doesken 1990), Indeed, even in the semiarid cli- '~-of Colorado, extreme rainfall events with local lliIlIIIitrlds exceeding 4 in, in less than 24 h occur several ~ each year somewhere in the state, However, Mms producing greater than 10 in, of rain are rare, A recent climatological investigation of extreme pre- cipitation events in Colorado (McKee and Doesken 1997) showed a distinct tendency for the largest events recorded during the twentieth century to occur at or near the eastern base of the Rocky Mountains, ex- tending from the lower foothills eastward onto the western fringe of the Great Plains, Prior to July of 1997, there had been seven documented storms this century that produced at least 10 in, of rain within 24 h, Several other Front Range storms have produced excessive rainfall but at lower rates and over longer periods of time, For example, an early May storm in 1969 produced close to 15 in, of precipitation over a 4-day period in the mountains and foothills imme- diately northwest of Denver (Fig. 2), The most pub- licized of Colorado's extreme rainfall events is the Big Thompson storm (e.g" Maddox et al. 1978; Caracena et ai, 1979) of 31 July 1976 (Fig, 2), a can- yon flash flood that resulted in 139 confirmed fatali- 194 ties and a list of seven additional missing persons (Gruntfest 1996), Based on an examination of more than 300 of Colorado's heaviest precipitation events since the late 1800s, the vast majority of these events were found to occur between mid-April and mid-October but with a double peaked distribution. The first peak occurs late in May and early June, associated with synoptic-scale quasi-stationary late spring storms. These storms carry moisture from the Gulf of Mexico and the Mississippi Valley westward to the Front Range of the Rockies (Fig, 2), where it typically falls as widespread pre- cipitation of moderate intensity with localized convec- tive and orographic enhancement The maximum precipitation from these storms is typically found along the eastern foothills of the Rockies, A distinct lull in the frequency of extreme rainfall events occurs from late June into Jnly, This is followed by the sec- ond and greater peak in storm frequency from late July extending into early September with a pronounced maximum frequency from the last week of July into the first few days of August These summer storms are highly convective, often small in areal extent, and have occurred in nearly all parts of Colorado. However, the greatest of these, in terms of maximum rainfall, have occurred east of the mountains and of- ten near the eastern foothills of the Rockies, The FCL storm of 28 July 1997 fits very neatly within this cli- matological description. 3. Rain gauge measurements of accumulated rainfall for the FCL flood Immediately following the 27-28 July FCL events, the Colorado Climate Center began an extensive ef- fort to document the rainfall (e,g" Figs, 3a,b), Using radio, newspaper, and broadly distributed electronic mail requests in combination with traditional door-to- door and phone surveys, rainfall reports were solicited from all possible sources. In all, more than 300 rain- fall reports were gathered, 90% of which were based on rain gauge measurements, As much information as possible was gathered for each rain report including gauge type, exposure, observing procedures, latitude and longitude of the gauge location, and any additional human observations to supplement the gauge measure- ments, Upon completion of data gathering activities, gauge locations and amounts deemed highly reliable (110 reports for the night of 28 July; 209 reports for the two-day event, 27-28 July) were plotted on a map Vol, 80, No.2, February 1999