FLOOD01375 - Laserfiche WebLink

Home Browse Search

the passage over the Denver area of the long-lived convective feature traceable in infrared satellite im- agery all the way back to central New Mexico (Fig. 5), The development of the bow echo in the weakly sheared environment illustrated in Figs, 7c-d is par- tially consistent with modeling simulations discussed by Weisman and Davis (1998), The Weissman and Davis study indicates that bow echoes can develop in environments of weak vertical wind shear (e.g., 10 m s-I/2,5 Ian) where there is large CAPE (2500- 4500 J kg-I), However, such values of CAPE well exceed those associated with the environment of the FCL flood (868 J kg-I). The bow echo in this case did not produce the strong surface winds that are often as- sociated with such phenomena (Przbylinski 1995), It did, however, exhibit a rear-inflow jet and weak cir- culation features that resembled bookend vortices (a lower-tropospheric cyclonic circulation on the north end and an anticyclonic circulation on the south; con- firmed by dual-Doppler observations and discussed in section 7b) consistent with previous studies of bow echoes (Fujita 1978; Weisman 1993). During the 4- h period represented in Figs, 7 a-d there is a remarkable contrast between the convection to the south of FCL, characterized by the mobile bow echo, and that over FCL, characterized by quasi- stationary convection, What accounted for this differ- ence is unclear, However, convergence associated with the Denver cyclone and the convective feature visible on infrared satellite imagery (Fig, 5) crossing the Denver area from the southwest may have helped to initiate the transitory convective phenomenon in the Denver area, The flow in the FCL area was not influ- enced by the Denver cyclone as it arrived unimpeded by this topographic circulation throughout the period of the flood, However, it does appear to have been in- fluenced somewhat by the bow echo that passed to the south, Comparison of analyses at 2000 and 2100 MDT (Figs, 7c-d) indicates an acceleration of the east- southeasterly flow between these two times, appar- ently in response to the surface downdraft outflow behind the bow echo, Further, winds in the immedi- ate vicinity of FCL became more easterly during this period, implying an increase in the component of the flow normal to the north-south-oriented foothills just west of FCL. These changes increased the moisture flux into the FCL area at approximately the same time that the third, and heaviest, rainfall episode developed (Fig,I), Bulletin of the American Meteorological Society 6. The roles of preconditioning and triggering in the storm environment Surface upslope flow was the primary precondi- tioner for the FCL storm because the high humidity associated with the easterly winds meant that relatively little lifting was necessary to raise boundary layer air to its LFC at 690 hPa (compared to an LFC of 620 hPa for the BT storm; Caracena et al, 1979), Unlike the BT storm, where a front combined with orographic lift provided the primary trigger for convection (Caracena et al, 1979), the ouly apparent' trigger for the initial area of convection associated with the FCL storm was the foothills, For example, considering only the east- erly component of the wind profile (approximately or- thogonal to the foothills) presented in Fig. 6, it can be inferred that the lowest 1-1.5 Ian of the troposphere would have been lifted upon encountering the foothills west ofFCL (Fig, 2), This situation contrasts somewhat with the BT flood; namely, the lower LFC in the FCL case would have permitted generation of heavy rain- fall as the flow reached its first pronounced orographic lift on the west side of FCL, Of course, it is too sim- plistic to think of forcing throughout the storm ouly in terms of the foothills orography, It is likely that other factors such as outflow boundaries (which played a role in cell initiation near FCL later in the event; cf, section 7), latent and sensible heating gradients, and pres- sure perturbations also played roles in the distribution of convection and heavy rainfall along the Front Range, In the few hours before the flood, the wind profiler in Platteville, Colorado (about 45 Ian southeast of FCL), registered a 5..,8 m S-I wind that veered from southeasterly to southerly over the lowest 2 Ian of the troposphere. Satellite-derived cloud-drift wind vector fields (at approximately I Ian AGL) were calculated from GOES-9 visible imagery for two periods on 28 July (1400-1423 and 1700-1723 MDT; Figs, 8a,b) using three images per period, Consistent with the wind profiler data, at 1416 MDT (Fig. 8a) the c1oud- drift winds also indicate the presence of a 5 m S-1 low- level southeasterly wind, However, at 1715 MDT (Fig, 8b) the cloud-drift winds suggest that the winds backed slightly, becoming more east-southeasterly and increasing in speed by 5 m S-I, The increase in upslope flow over the plains of northeastern Colorado (also 3The nearest resolvable front was almost 700 kIn south when the Fort Collins storm developed. It is possible that a proximate boundary did exist but was not apparent in the conventional observations. 199