'hanges i11 land use land cr:1:er result« n;njnr r'!kmge In oil scale basin condition and hydrological response. The human impact q,f watershed i011fl uxe l"md cover char.ge 11n,/ climate cht.mge on hydrologtco! charaaensucs of watershed processes ofstream flcni• have ~J? 1via'c~I}' studied. Geograpnic !tifim11<llio11 System f(ilSJ-based spatial mOdeling has play a <-.UCiai U)()/ in land use/and col'er and ciimas« chaug« lmpac! rm runoff {111(-f .Si!.ili»U?.11it:lion studies . The .Soil and Water .4s .... iessment T ool (SWATJ model is one ofm<lhy frydm!np,ir tools thcJ is efji!di•'e wol jor assessing vxuer resource wi;h the netp qf' G'IS anct iJ is also importons for sfJtdj.:ing nonpoint-source poliurion problems ~/;_;r a wide range of basin sct!le.i and · wcuershed degradanon over a long time serie ... s. The mc:rfor objective (~f tin's study lVas to evaluate the impoct of historic hmd use innd cove.- chai·ige on lr;.'tlrolo:..Tical dynamics of· Koga watershed hy IJ;,'it;g sn:1T modei. The SWAT model used WTi()US inpias: l)J:.'M. Land Use Land Cover. Soil. Climate an(/ Mt'feQro/ogictli data: 1-imr !19?3. !9!16, 2001t111d1013) land 1m• land cowr oia5s(.flca1io11s fur the watershed w.,-e developed ftam Landso: M~S, TM, F.TM using supervised ciussificatton system. Six land uxe land cover classes were identified for 1973, 1986 and 2001 but 2Ui J land use land cowr fotO'hi one addit!m:al land 11se system. because (l{ artificial dam. Lam! us" land C<"'«r results have shown duuforests land had declined between 197.1 and 2(!()il~)· 63.6% while agncuimroi land and: built-up area exp<md2d /:!y 27.7% and i0.7% respeaively. Instead c Forest area has shown increment betwee» 2001 und 2(Jl J /i·om 3.6 percent 10 8 . ./ percem in cos! qf other ICPt..i use land COL'er classes. To nal:u<1te the impac! of laJJd U5e land cover cr.ange OJ1 Koga watershed hydmlogy the SWAT model •1'((.~ ccdibrated, validared <md n+'ei\-e stream flow parameters H:ere selecied t() he sensitive for the strewn flow <~l the .. vatershed ond used for automaic and manuai model calibration. The mode! <:fticienc.v is e~aku.ll1:<I in terms 11{ ft' and ~lS objeaiwfartCtirJ1 and re .. o;,,'Uf_f for cahbration O. 76 anci 0. 7 5 respectivet» and for volidation i< result (1. 70and1).66 respecavety. Em/1101io11 o/'the impact q( land use' land cxwer ~/Ja11r;es revealed thatther« war an increase of streamfiow front 3.XX to 4 . ./.9 m'!s in 1973 10 i986 and 211i)l streamftow increase 2lJ.-!8% wttereas 2013 streom fiow decreased b.,v 28.84 percent because (Jj high forest ~~overc1,ge than 2001 I ami tt'!<: lwld ro\.l:'T, Three scenarios developed and impact on stream flow evaluated. If Kof?a watershed .:o•-e1~ on/_y b:.· i'i;res~ Built Ups Wet land and C11iti1,a!io11 la>ui th.; stre{un Jl<rH: ~i:ill tlL~a-e,rs..! by 30 4 pel'Cf!nl frum 1{}13 :<trearn .ll<.n~r recortf. Se.cone/ scenario ct>nsidt~r fl. -~oga ii·,tJersh?d ctr1:ered bJ: OJtiy Cu/1iYUJW11 luffl b&uu."'t! t!f this cht21l$!e rhe ;rret111i./l.0111 1 .. :fll fncrea)'e by .56.89 perc<?llt from 20! 3 stream j!aw record. bui if the Y'<llers!J?d only C<Jt·er by jbre.ic i<n1. the siream jlot11 i,i1/l de.<.~r<Jased b>' 51. 43 percent. 1he .st1td)• record.ul u cl1tl1tge ~t j't:·e1.un_floi..· u:hen 1he /.and 11se lar.1.i eo1:11-"r f!/~tlie 1vt1iet·shE?d cha11grJs. lkued on this re.tult d'll.Btl.~ in land use land co\'er ha\'e i1nl)(1ci 011 h)1ci1·ology o.f· Ko~(' 'tt'U/(,'1'shed Thi! rc:w!t.> "•ill f11rlkr help it1 1nakh1g dt:cisions relaletl fu 'tf'a!ersheci nu.Vt:JgftJtleJtl Oj)liOr~· (}1:1: wi/J firOJnofe sustait1abf<? 'Ulifi::utWn wJ 1nanagemenl Q{ Kof[a /and and \1'£llet resources.