OPS K,pgs1pagedoc2nearest-ileafileafUU . UDpage(if (not (fboundp 'ileaf::r)) (defun ileaf::r (data) (tell *load-object* mid:put-data 'ileaf::saved-data data)))(ileaf::r '(:pup-cmpn-names ("2columns" "refer" "para" "sectionhead" "micro:ftnote" "centered" "justified" "title" "authors" "micro:caption")) )pga1<yf3fM fon1wst:helvps9fon1wst:courps10fon1wst:helvps11fon1wst:timsps12|fon1sym:clas12|fon1wst:timsps12fon1sym:clas12fon1wst:helvps12fon1 wst:helvps14bfon1 wst:timsps10fon1 wst:timsps12bfon1 wst:helvps10fon1 wst:helvps11bfon1wst:helvps10bfon1wst:helvps11fon1wst:helvps6fon1wst:helvps18fon1wst:helvps14b fon1wst:helvps14bi fon1wst:helvps14fon1wst:helvps14 fon1wst:helvps14i fon1wst:helvps14 fon1wst:helvps14bifon1wst:helvps9ifon1wst:helvps12|fon1wst:helvps10|fon1wst:helvps10|fon1wst:helvps14b|fon1wst:helvps12|fon1wst:helvps9bclr1 @ a 9 1 IHH N0<d|&f&fLLpat1 ((nn((((nn((((nn((nn  xx xx``00 ``00  00`` 00``  xx<<xx<< <uUW@3com3, PPNormal,com30tauthors0com3z Pbulletsbb1 |:bulletuUW@S sbe1zcom3.pcentered.com3+ UUWhead  +com31 tileaf:anumtoken  1com30 tileaf:reftoken  0com3/ pjustified/com3 listsbb1 |:listuUW@ aut1 sbe1com33 Jmicro:caption  3com32 Jmicro:ftnote  2com3* para*com31 PPʤparagraph  1com31 _ refer1com33 `sectionhead3com3.  subhead  .com32 5"summary  2com3.Httitle.com3? |:bulletuUW@S ?com3J |:listuUW@ aut1 Jcom3- t~htmltarget  -pro1frame mastersmtr1nes1Sefr1+ y +dia2 g Snes1efr1"KK#ʫAuto"dia2x g E0.26179938779914941110@@@xnes1efr1/ %floatingPR/dia2x g E0.26179938779914941110@@@xnes1eefr18yfootnote:numbered8dia2 g T  Dt@-nes1mcom3a Jmicro:ftnote HTML_()(:ignore) ref1 EiXYY2e4vr . amE0.26179938779914941110@@@enes1Jefr1:ćv8footnote:unnumbered):dia2 g T  Gt@-nes1Pcom3D Jmicro:ftnote HTML_()(:ignore) DPE0.26179938779914941110@@@Jnes1efr1/%fullpage/dia2x g E0.26179938779914941110@@@xnes1efr1! `inline!dia2x g E0.26179938779914941110@@@xnes1efr1+@%pageP+dia2x g E0.26179938779914941110@@@xpro1mtr1pro1mtr1mtr1pro1header/footershft1nes1=efr1dia2 g =nes1=efr1dia2 g =nes1efr1dia2x g E0.26179938779914941110@@@xnes1efr1dia2x g E0.26179938779914941110@@@xpro1Page 1 (a component)com3UQHttitleHTML_ALIGN"CENTER" 1 nes1pro1Page 1 (a frame)efr1@" `ypageHTML_ALT"optarg3_page0"HTML_SRC"optarg3_page0.gif"(ileaf::r '(:web-frame-sig (4015276520 75 0)) )1dia2 g T  ژnes1Kcom3? ttitle HTML_ALIGN"LEFT" 15.3?KE0.26179938779914941110@@@ TARGETED OBSERVATIONS AT NCEP: TOWARD AN OPERATIONAL IPLMENTTIONUpro1Page 1 (a component)com3tauthorsHTML_ALIGN"CENTER" Zota Tot1nes1.pro1Page 1 (a frame)efr1gyfootnote:numbered(ileaf::r '(:web-frame-sig (286000348 75 0)) )1gdia2 g T  Dxnes1com3ޗ ,micro:ftnoteHTML_ALIGN"LEFT" 11 GS (Beltvill, MD) at NCE. Corsponing athor adress: Z. Tot, NCEЯEM, 5200 Aut Rd., Room 207, Camp Springs, MD 20746. Zotan.Tot@noaa.goE0.26179938779914941110@@@., Isva Sznog2nes1*efr1kyfootnote:numbered(ileaf::r '(:web-frame-sig (1836273980 75 0)) )1kdia2 g T  DxInes1com3 ,micro:ftnoteHTML_ALIGN"LEFT" 12 UCA Viiing Scetist at EMïNCE; 3.Pensyvnia Sate Unvesty, State Colege, PA; 4. MI, Boton, MA; 5.NOAER, Bouder, CO; 6. USR at NASAGSF, Greebelt, MDE0.26179938779914941110@@@*, Shra Mjuda3, Rbec Mors4, Brian Etheto3, Crai Bisop3, Stphen Lord, Mart Ralph5, Ol Perso5, and ZhaXi Pu6 Evrometal Moeing Ceter, NCE, NWӯNOA Wasinton DC 20233 pro1Page 1 (a component)com3T `sectionheadHTML_ALIGN"JUSTIFY"1. INTRODUTIONTcom3_ `2columnsHTML_ALIGN"JUSTIFY"Duing the past three witers NCE patipated in four field prgrams cocering tageted oservtions, with the aim of iproing shortrange weater forcasts by adativly taing drosond oservtions in sestive areas. In the first field prgram (FATE, JaaryFerary 1997, Jol et al., 1999; Sznog et al., 1999a) the goal was to test difeent tageing metods in the field. Based on the FATE rsults NCE chose to futher test the Esemble Tranform (ET) tecnique (Bisop and Tot, 1996; 1999) in ucoing field prgrams. In JaaryFerary 1998 (NOPE, Lanlan et al., 1999; Sznog et al., 1999b) and March 1998 (CAJE, Ralph et al., 1999) NCEP' goal was to evaate the pefomance of an oertioaly fesble tageing metoogy based on the use of the ET tecnique, both on the syotic (NOPE) and on the messcales (CAJE). com39 `2columnsHTML_ALIGN"JUSTIFY"In JaaryFerary 1999 took place the 1999 Witer Storm Rconaisance prgram (WS99; Tot et al., 1999; Sznog et al., 1999c). This was a qusioertioal prgram dsigned to iprove 24-96 hour forcasts of sinicant weater events over the cotnetal US, icluing Alaka. The weater events (time and loction) for which the forcasts were to be iproved were slected by oertioal forcasers in field ofices, coodnated by HP. Sestiity cacultions were then caried out to find the area over the norteast Pciic from where etra data could most iflence and iprove the prslected forcast events. All adative data were used in the oertioal anasisforcast cycle. The ipact of the data was evluated in near real time by uing a paalel anasisforcast cycle from where all adative data were ecluded. As a new feture, the ET sentiit cacultions were mdfied to drecly idetfy the best flight track, out of a set of prdsigne choices. With this dveoment the tageted oservtion dcsion maing agrithm can be fuly atmated. wait `2columnsHTML_ALIGN"JUSTIFY"In this pper we rview ipotant apects of the tageing metoogy: case seletion, the pefomance of the tageing tecnique, and forcast iprovment, based on yet upulished rsults.nes1Mpro1Page 1 (a frame)efr1 D@@y HTML_ALT"optarg3_ 1"HTML_SRC"optarg3__1.gif"(ileaf::r '(:web-frame-sig (1070064198 75 0)) )1dia2 g T  x0 nes1Ccom37J paraHTML_ALIGN"LEFT" 1 PRGRAM SINAL ( 0 8 @VEIFCATION X   sp  pr ( 0 sp 8 @ wn H P pr X SUM FATE 7+1-0 ( 0 8 @3-2 H P X 3-2 NOPE 8+0-2 7+1-2 07-2 8 @7-1 H P7-0 X 7-0 CAJE 3+1-1 4+1-0  05-0 8 @3-1 H P X 4-0 WSR-99 21+4-0 24+1-0 016-5 8 @15-8 H P13-2 X 18-5  TTAL 39+6-3 35+3-2 028-9 8 @28-12 P20-2 X 32-7 % SUC.:  81-94 87-95 0 76 8 @ 71 H P 91 X 85 STAT.SIG.:  ( 0 0.1 8 @ 1 H P * X *7Cpg g vx vxx6 vx66 v6 v5tU5t5T  *7Fwdnes1com3l  paraHTML_ALIGN"JUSTIFY" 1TABLE 1: Tageing and vefction sumry for the ETbased tageing stragy used at NCE for four field prgrams (left coumn). Suface presure (sp) and acmlated prcpittion (pr) sinal rsults are shown in 2n and 3rd coumns. First+scond (-third) nubers in each coumn idcate nuber of cases when the aslutelcal mamum was (was not) witin the vefction rgion. Last four coumns on the right show suface presure (sp), 1000-250 hP winds, 12-hour acmlated prcpittion, and oveall forcast vefction rsults. First (-scond) nuber in each coumn rersent nuber of cases where the forcast was iproved (dgraded). The botom three lines show the ttal nubers for the four field prgrams, the oveall sucess rate, and the corsponing sttitcal sinicance leel (where aproprate). Star stands for sinicance leels blow 0.01%.E0.261799387799149411@10@@@M waitpro1Page 2 (a frame)efr1%W@yAutoPHTML_ALT"optarg3_Auto7"HTML_SRC"optarg3_Auto7.gif"(ileaf::r '(:web-frame-sig (1186714716 75 0)) )2wait g T  @h@nes1com3  paraHTML_ALIGN"JUSTIFY" 1FIURE 1: 1-8 days lead time esemblebased Prbbiiti Quatttive Prcpittion Forcasts for 24-hour acmuate amounts eceeing one inch (25.4 mm), vaid around Jaary 19 0000UT 1998.iD%%DƬ=qKKr)˺wfUD3"˘ffffffffffffffffffffffffffffffffff3333333333333333333333333333333333˺wfUD3"˺wfUD3"˘ffffff333333˘ffffff333333˘ffffff333333˘fffff333333˘ffffff33333˘˺wfUD3"˺wfUD3"˘f3˘f3˘f3˘f3˘f3˘f3˘f3f3˘f3˘f3˘f3˘f3˘f3˘f3f3˘f3˘f3˘f3˘f3˘f3˘f3˘3˘f3˘f3˘f3˘f3˘f3˘f3˘f˘f3˘f3˘f3˘f