Laserfiche WebLink
<br />.~ <br /> <br />r <br /> <br />f <br /> <br />T:-:E ~;:..;S J;":'~3:;:K ~.'~i::::": <br /> <br />( <br /> <br />;...:I:K :.JS::~S :.;;~J:: <br /> <br />Aos:~act <br /> <br />( <br /> <br />A dam-break flood forecasti~g ~odel ()A~SRK) is described a~d applied to <br />two act'Jal dam-break flood waves, ,he mocel consists of a breach component <br />which utilizes simple parameters to provide a temporal and geometrical de- <br />scription of t:1e breacn. The moa..: :omp'Jtes the reservoir O'Jtt:ow hyarograph <br />res'Jl:i~g from the breac:1 via a oroad-crested weir flow approximatlon, which <br />i~clJdes ef~ects ~f s~omergen~e from downst~eam tai:.ater depths and co~rec- <br />tions for approach velocities, Also. the effects of storage depletion and <br />upstream inflows on tne compute a o'ltf,ow hyCrograph are accounted for through <br />storage rO'Jting within the reservoir, The basic component of the DAMBRK model <br />~a dynamic routlng techni.ue for determlning the mOdifications to the dam- <br />b'Mlak flood wave as it a::lvances thro'Jgh the downstream valley, incl'Jdins its <br />travel time and resulting water s'Jrface elevations. The dynamic routing,com- <br />ponent is based On a weighted four-poi~t, nonlinear finite-ditference solution <br />of the one-dimensional equations of 'lhsteady flow (Saint-Venant equations) <br />which allows variable time and distance steps to be used in the solution <br />proced'Jre. Provlsions are incbded for routing supercritical flows, subcrit1- <br />cal flows, or a spontaneous mixt'Jre of each, and incorporating the ettects ot <br />downstream obstructions such as road-bridge embankments and/or other dams, <br />routing mud/deOris flows, <br /> <br />[ <br />( <br /> <br />[ <br />[ <br />I <br />I <br /> <br />Pressurized flow, landslide-generated reservoir .aves, accounting tor <br />vol'JlDe and flow losses d'Jring the rO'Jting of the dan:- :reak wave, conllidering <br />the effects ot Off-channel (dead storage), floodplaic3. and floodplain <br />compartments. Model input/output may be in either English or metric units. <br />Moaeling difficulties and parameter 'lhcertainties are described and methods ot <br />treating them are discussed. Model data requirements are flexible, allOWing <br />minimal data input when it is not available while permltting extensive data to <br />be used when appropriate. The model was tested On the 7eton Dam failure and <br />the B'Jtfalo Creek coal-wallte dam collapse. Computed o'Jtf,ow vol'.lllles through <br />the breaches coincided with the obllerved val'Jell in magn:t'lCe and timing. <br />Obllerved peak dischargell atons the downstream valleys were satisfactorily <br />reproduced by the model .v.n though th. tlood waves were severely attenuated <br />all they advanced downstr.... The computed peak flood e,:vations were within <br />an average ot 1.9 tt and 2.1 tt of the observed maximum el~v~tions for Teton <br />and Buttalo Creek, rellpectively. Both the Teton and &,f~a.o Cree~ simulationll <br />indicated an important lack ot sensitivity or downstrea~ ~i5charge to errors <br />in the rorecast or the breach size and timing. Such errQrs produced <br />signitioant ditterences in the peak discharge in the vici~itf of the dams; <br />howev.r. the di tferencell were rapidly red'lced as the wav:s advanced <br />downstream. Computational req'Jirements of the model are q'H te reasible for <br />mainframe, mini~ or microcomputers, Sugge~ted ways for ~s;~g the DAMBRK model <br />in preparation or pre-computed flood information and in real-time forecallting <br />are presented along with several examples illustrating tne ~se or the DAMBRK <br />model. <br /> <br />I <br />I <br /> <br />I <br /> <br />( <br /> <br />l <br />i <br /> <br />l <br /> <br />i <br />l <br /> <br />l <br />