<br />."1'
<br />.~ {,~::
<br />
<br />OOJ47J.
<br />
<br />:..-~
<br />
<br />.<~'7.....:,
<br />
<br />"!~
<br />
<br />,,'.-,:
<br />
<br />"
<br />
<br />DEVELDPMENT'OF A FLASH FLOOD DETECTION. AND WARNING SYSTEM
<br />FOR BOULDER COUNTY, COLORADO
<br />
<br />REAL TIME DATA ACQUISITION AND MANAGEMENT
<br />
<br />r
<br />II
<br />
<br />George X. Hefnrichs
<br />
<br />Boulder County Sheriff's Dep.rtment
<br />Technic.l Systens Group'
<br />Boulder, Color.do
<br />
<br />1. INTRODUCTION
<br />
<br />The Boulder'County Sheriff's Dep.rtment;
<br />the' Boulder County Publfc Works Dep.rtment; the>
<br />City of Boulder, .nd the' Urban ,Drain.ge'.nd.Flood
<br />'. Control'Dtstrict.beg.n..acooperative effort in.
<br />1977 to develop. system to.providereal~time .
<br />'monitoring of fl.sh flood.eventpotenti.l. The
<br />systen was initlally. Implemented. using renote .'.
<br />.radio telemetry .tipping.bucket. rain g.liges .nd
<br />..cont.ct closure. stream,. lI!veL g.uges. ,. All; of.. .
<br />the remote sensing equipment inst.lled reported,
<br />to .' central ,processing .site viO',. systenot;:
<br />r.dio repe.ters., The "Centr.Lsite, was.1i ,Sierr.'
<br />. Misco We.therD.ta Te\"lllin.l.' .,~':.' :.,' ,
<br />. . The We.ther Olta'. Termin.l reported .the
<br />time .nd"accumulator value'for..e.ch gauge in.the"
<br />systen; Thts systen w.s also c.pable of logging
<br />.11 incomtng ,d.t. tr.nsmissions on an ,..ddtng,
<br />tape. printer.' This was .ari.:extremely ,import.nt
<br />option since. .stngle,.power .h1t~ .wou'ld;result in
<br />the loss of .1lhistoricll iI.ta. . .' '. '
<br />In the. c.se,.of .. 'severe '.storm event, per-
<br />sonne.l recorded ~heacclljWl.tor,v.lues. for e.ch
<br />. remoteg.uge.. and..then used the. data to comp 1 ete
<br />'manual.Thiessenpolygonlime~based flow projec-
<br />tions 'flir, themajl!r ,basins.'" "These projection~.
<br />...redes,igti~il"~p"be ~omplei,ed '!!nce ever,-,1o,:': c'
<br />. lIinutes"'and'were; li.lli ted;,to..thec.nalysis'of:~ ,',',
<br />threehour event; 'As.t~enumber. of.r.~ngauges
<br />grew from tile "initial'.8to"40;:'"lRanli.lilbu-''' .
<br />l.ti9n,o(data required at '1e~st20minutes_per,
<br />.10 minute cycle; leavlng'techtiical .dvisors and
<br />other 'support peTS!!nne 1, overwhe lnie'dwi th : the, data
<br />'reduction"task; As a result, .tllere w.s very.",
<br />little tillll! for e~aluatirj~ of d~ta;' ,", '
<br />. "The.system:.lso'provi~ed:~o.direct.method
<br />of ,c!,",pensati9~ ,far.'.!lalfunctioning or,.de.d". .
<br />gauges. 'This proved to be . serious problen
<br />since the probability that a lightning strike
<br />will c.use gauge failure is high during a storm
<br />event. .. \
<br />
<br />,
<br />'.'~:.'
<br />
<br />2. ACQUISTIDN AND ANALYSIS OF ALERT SOFTWARE
<br />
<br />In 1979 we acquired a Radio Sh.ck TRS BO
<br />Model II computer systen with ALERT (Autom.ted
<br />Loc.l Ev.lu.tton in Re.l Time) softw.re from the
<br />N.tton.l We.ther Service. This processor .nd
<br />softw.re provided addition.l data reduCtion
<br />c.p.bilities, but sttll did not meei'our needs in
<br />. the fa !lowi ng .re.s:
<br />
<br />A) . Datavalid.tion for incoming gauge
<br />reports was jn.dequ.te to 'fUter .the' 'interference
<br />level we were.experiencing. Single'gauge'.ma.l-,
<br />functions .nd.lnvalid d.ta transmissions. caUSed':'
<br />by routlne g.uge malnten.nce resulted' in f.lse
<br />alarm conditions..nd erroneousd.ta displ.ys.
<br />8)' Systen..display groups werelnflexible'.
<br />and did not provide,.n....~ ,identific.tionot; . ".
<br />reporting g.uge.-:sites. '. :';. . '. '. '. ;..
<br />C).. Database' m.intenance,utUtties were. .
<br />non-existent. and. as a result .rchived d.ta' ' .
<br />. contained signifi'c.nt errors. ' .' ,',' ....
<br />. . 0) ,,'Alarm .functions were .inflextble. .rid
<br />the ..rchitectu..e.nd';packaging :,ot ,the .TRS;,BD. .
<br />series'of ;computers,.....de..extern.l',interf.cing".
<br />difficult;,.... .,'.. , ; '.' .'. ",., ,...> ;
<br />. .E).The:TRSBO MOdelll'was not as.., ....
<br />rellable ,as desired'.and required'use .of'complex',.:
<br />procedures.tolnitiallze the data collection
<br />process. ..... : _ .. .
<br />. ..F) .The software avaUable did not".. .
<br />support time-based project tons of storm.events..
<br />.G) ,The.d.t. collection procedure ..did not:
<br />accept reports from cert.in m.nuf.cturer's g.uges
<br />due to ..bit st.tus.dnconsistencies.',..,'''' .
<br />'" '~).' The.most 'signific.nt:.deficiency of:: .
<br />this..system;,w.s,.iis..tnJhOity;;to.provideJmultt. ...'.
<br />tasking(",~Hl~4~@r ~~ppont.,.,TI1e:'need for'.-;',;',,;c";',
<br />. cilncurr:e~t~ SUPPil~! ....liF~wh't:,;tf. ':eyent:jll'ilJect i 9ns'
<br />.was !s"Slm~i.l .to,p'roride,ad!qui~e!te~hnl~.L'~:"", .
<br />,support ,for real-ti.,. warning' decisions; " :0:'-
<br />. . Since.,sou~ce.:.code was ,.nOt,av.llable .for
<br />this version 'of the', AlERT .sySten; ,it was .'
<br /><:.' imposs.ible,to correct these'deflcie~cies.,:'.,
<br />", Project. personnel )'e~e, ~eluctant.tQ.develop...
<br />.dditional software..whi.ch wouJd...be, dependent upon,
<br />thts h.rdware :,.nd 'oper..ting. syst~. J',: ,",,:
<br />
<br />3. .-DEVELOPMENT 'OF.THE' STORIlMSYSTEM
<br />
<br />, ", ~,L'. " ;' ",-' , ',,':
<br />The.STORRM (Sheriff's Telemetry Operated
<br />R.infall ;and:RtYer'-.Monitor.) systen. was 'developed.
<br />by the Boulder County Shertff!.s Dep.rtment . ..'
<br />Technic.l Systems. Group'.nd Emergency Services
<br />o tv is i on to correct these def i c i enc t es .nd to
<br />meet .ddition.ldesign criteria identified
<br />through p.st ,experience.
<br />The crttic.lity of thts appltc.tion w.s a
<br />major. factor. tn h.rdw.re and softw.re configu-
<br />r.tion'decisions. Potentt.lly hundreds of l1ves
<br />m.y depend on the proper .nd reliable oper.tton
<br />of the system. The b.se processor chosen w.s the
<br />
|