“II 00 ....0...»- ... 0. .. I. 00 0 . . V . . V. . V. .... 0 .n...... .V ..0-“£00.05“... . I 00.00 3 0. . 000 A . . 0 0 AA : - V. . .V V A AS. I 00.. . A0 0.0 0 .0000. 0V .0 00000.: .0. 00.. 000—..0000AOA-V .0 ~ 0 Vol—A94“ .-0-1 «J 40“.. VV0 5 0#003 . w .. 0.0.0... V" 0 .V .. V0. .. V . A 30.0900... 00.- . defifnlv... ..u.“ $000.00 000. ”0?..0-V "0:300 «0000. 0030?: 5 L00..J0000.0. 0.0.: ......0... ...- . _ . _ V . . V .V 2... .V . . . .. ... V «... ..0. 6103.. ’0 .0. .).0 ...-00.030. ‘M:I A. .A.. . 0 0 0 00” 0.0 0. 000“ 0’31... 0 0 015?.3330 050 . V. . V 10. .000. ..0 . ..0.- 0 . 50.0.1.0. V .020 . o. 00. . VV .00 0. 0 0 A6 V. 0.0.... 00. .V .0.... .0 ”0.000”! .. .0.-0% ..0 00,000. ..u WV. )0 0.. ... MAO-”0J0? .“V‘V V0 0 >090:- yV’J . .V . . V . . . .... . . 00 0. .0. i... V 0.0 001.130.000- V“: 000.300. .00 . .- 0‘00230000 0 0. V .0 0‘0 01mm“ 0..00. .00 0H0... I0. 0 .0... A 00 V 00 .V.0‘0:0.0.000 $.00.- \. V ..0 ......V... ..0-.0130... 0. . A. 2.0.. .0232. .0 ...AV 10. 00 0. 03.0 ‘00 000‘". “03.". ..V 00000 .0.-00‘ 01.903‘: “Am-0 , 1.0030 3 .0 0.306 V V . 0.0.0.0.... a 000.? 0.0 .. 0A... A .. 0.00 V ...... . V0001 V .0 :00 A.) 0.. .000 . . .00". 0A0 0.0! ...“.Mv... (0.000.... ..00...1..-.0M.V..Vn..0v.._. .... 0.0.0.0.... V V 00.. A06” .0 . ...000 ......SI .0. 3.00. ..0... 7r {..0-.0003“. £000. A00. .00: .‘0.:’0” 3.31.;I-r'04h0VJnP 1‘0 .0. 0. . 0- . 0.0. . «A. ..1. .0 ..0 0 V . V 0. 0 0. 0 . 0. ... ..AIAQV . .0. 00.0200 I. ..i. 10‘s.? 0. .0. . “...: A u. ..A 0 0.0.0 “.0”. .. . ..0» ‘. 1.3.0.4030 0.... . 0... V V . ......0 .-.00r.0»0wo _ .H.V.0.. ... «V :00. .00..JVV0:.#’..’.00V00§ 0....0. .05. . ...... 00.0.2 001- 0 . 0.. .0... : .0000” V 00%‘000..A0ww1| ..0. 0 ”’00.“. ..0...” 300000 . .0....“A .....LVMI» V V V V V ... . .. . 0 o . V0 .. . 0. . 0. ... 0.... .13....0. .. u. 00. .00) 00 .00.) 0.30 .0 0’. 0.]... 0. 0. V‘:.0.. 0.2500011”. 00;... 00 “barf. 0 .00.. A 0.50 .V V . . . . . . . . V.A0 .0. 0 .. .0 . V . . V .700: .s... , 0.30. 003.00 00.0.00 IVs ' V ‘rhuwfm 0‘320’0010 00"{I‘w ‘wmmnvu: 0.0 0.0.0 "0030-- 0: . V V I0 .V00 . A ”00”....“ 0' 0.0. .V . (.300... .3 0. - .0 U. .0 0:. V 00. 0 .0 10:5 . . .V . u n ..--0.? 0 0'0".” 0300. 03c? .... 000. «02%. . $00.3“..0000ct003000091Than“. . V V V . . ”0.0 0 .. 2.10. 0000.0. . . ..0. .. 6 0.0.3.”, n.40una‘fo00' 303.0139 V00AVA00. .. “000 '9 0Vu0V. v. V}... .... ...-‘00... . . in... V. f a...” .V.....V..V. . m. V.“M.VV.VVVVMV..V.VVVV.VVVV.VVVVV .. a. 0. ....0 0.0.. 00.0 3000‘... 0.. 00.0000 000—0 006 AV ..0 . V . . 0 . . ...1 V. V. .30....33...0:... ... «003.030? 0 . 0V0")0H . 0.0.0 . .0.. - 3.3.00.0. A 0 ..0 V10 ..0 V. . 0- 0.0 :03....Vw0"0.h.9~..13 .0... J .H:0:~0. . . . ... ... . .. ......33. 00 . foul-.0020”) .’ .50....»an V. V. l . ‘00. V V 22.02.08. ... 00%. . . ’0. . . 0020 0 0 A 1300?”.01 .03 0.0. . 0W1”)... 100- .09; 1JU»J».2”.00".«000‘ N 0 0 00.00 00 0 0. Mf.‘ .0. 0.. 2.00%...0090751000: ...usWS .00. 00.0. V 0 .1:0.. . 0 0... V. 0.. . A. 0. . .0.... L0 . 006. .0000: 0V .000. 0 —. 30““. {0. . .0A0.00V00-.A.. .....00..000.0:0C00. ”.0.”...20 V . ....V .0... 00.0. 0.0 01*. ..0. 0. ..V-03 s. V . V. ...A 3 ..0 V 0000 00 .000... ..00.00... . A00 0“ 00.0 ...0. . 3.1. V .0". V. 3.5.. . ......z 2.23.. . V. .tfinuB—fi . .00. 0.00%” agif A A. 0.00.. :V .30". .0V00 .- .0.“...0. 6.“.00203100. .0. 000:... 0.500.. .. 0“.)2 ..V... I V. ... A 0:1. 000 ..V . 0. .0. .0 3 . 1 ..0. V ... 0!... :Vns. . .. . ... ...a . 3.”! 0.0 0- ‘200‘0: A ..0 r 0 A 0 0 00.. 0.0 . 0 Ix V 0 A .100. 000.0... 0.00. . . .. - . . V . 00 . 60.0. . . A ...0 00.00. ...... 0.,- 93000 0 00.00002 3.... :f 0.. 0 0.. 20.12.31... V. .VAV . ... V V. . J.- . A. 0.. 00.1 V .0. V V ... .0. :10... V ..V ..V? . ..00 .V ....A. . .V ..I. ... 01...?! f... tat}... . I, ...-0... 00..0.0—.’ 0 0 0. Q 0 V ..w. .0.. 0| 3r. 0 A 0... .A0 000 0.00.0 V .. .A .V ....0..00A. ..V 00. 3.5.0003}: :A .V . ...-00.0“ ......V 01.0.. ..."...L .0000’: V0... . .12.!) V ...-01A kn... A... $3.599}... 0.0.”. .A .. V ..V ...... ..VV. ... 03:83.... . .. ..V 8.. .. ..VV ... A ....- ..2...... ..:.:......v 0...... nuuuau .0 .V I V 0.- .. .3... V $0600.00 30:.— .l?.A.0.0. 0‘. .. my .. V (0.0.... {...-00000 13007. ..1. V 1:... .V - .0. 0. u .. .00.. .r... 0 4:00.0- 0.:.00. 0.10"”."46’00500 00 A . . V .V. 1...... 0.. a; .0.... .0.. 0.31.3313”: 1:... V .IV V V. 7 ..V . V . ....V:A0....:.. ..0}... . 21.3.3005 .. ... ...: 5.00... :0. A A ..V.- .0: .1650“...le V0 00». a... E5. V . V 0.00.0IVV .6“.~00 _ “10,000 08’: .AJ. .. 03013100400000. . ..0 0030.0 0 . 00030.00... «.0.. 0 0 0.30.th :00 06.. ..0. A0... ..00. .0... . 10...: . .V - w: .33. 4‘03... 00.0.1.0»? 023 60 0:090...300A.0n« :0: 0 .000 000000. V. xii-.01} VV. 0.. ..V 0 .00.. .. 0 20.. : V fiMJ‘. . V . .0703 . 0.. 60.000 0 . . ..l. 0’... 026 ... 00.1.. -.0. ...-01:6. . .u-slia 0.5.0.0.! 0. 000.00? 00 V .3 0+. VV. . A-VA... v. 0.. m0. :1. 50.0.0032]. V .0)... 0.... 020.330.. V 0.0.0.. .3 . 1 I. 0. A0. .. 0.0.... .0. 0.... .0. ... :9... ..V... ...... I0... «00... IA”: {.01.}! 0! ..V V . . . ~ 2v...) .. . 0 ‘4 0V. .00.. 00‘“ 0. ..Ac. . 010001... 0:00 00.~A.4.0u00V000000. .... 0. . A .1 0.x .0. 0 .0 00.00 0fi?.00 3‘. 0 V 0&519. . 0‘0. I 100.000..0V 000 0’1. 2.00 04.0. 00.0.0 .000 «“20 3V0”.- 0 0.. ..V ..V- ..10.I 60.000099. 0 00. .0.... .0....300'08' ..V V. 0. 0:. 00.0. 0A....00V‘ . . V .IV. ."0.30. .40" .).00 A. 0 . 0 .0, ‘00 ‘A‘. .060?” . . I. .00. 0‘30. 0 00-00(200000 .00A03-0-00006VV: :14 .0 0. 0 ,0. 0-. V. ...“..0’01u0MQ0900000500Ht0H-0'“ .VV-0.0. A0 A .V n 0 .- 0.00.0120: 0.. 00 .V . ... A... 0.00.00 ......V.... 00 0 .01. . .00 .063 0.0 V1,}: 0 .00 0‘I‘A0 00!. .4 In 05. 0 0 1... . 5 A. 70.00 ..6 . .V 0 110A. 0 .0 0 00,001.19J..00V.A0 A 9.100. 9:051VI. 0 A .A . 8.0.01.0: .. 00 A A .‘ . V”. VIA. V.V .... . . . . . .0....0..0.0 .0. :07. 0000!. 0 3A. 000 . V . I. I .0....0‘0 V h .021... .7... 0... . 30.1.... 0.. 0:1... . 03.2.0300. .. 0 ‘30.. .0. 3.00:. V .. A... ...0. .0. 0. [Elna .. 0.. .0... ..V .!:0.1.A V. ..0 3.42.0. .0.....V .. 3:0. ... ..V. 0.. V. 20.0... . . V V1.0. .....0. .113 0.00.0... ......A.‘Ii.u.arltl0000010 00.0.00 . 0.70 0 00.00. :4 011 6! 3 V0000 .0 .0. t’.’ 0.301.000}: 0 0. .0. .300‘1r 00 V1.1 0.00! 0. V V . 00.00.033.00 .. Hr.‘ 000.000. 000.00.... 0.. 1...?- .0 V . 0. A.. V . V. V . . .91.... V . A 0. 0. V00 .030 ..0 00%.“.‘ ’W. 00‘. « .dflr VA 0.. .V r 0. 00.. I... 0... .0105! 0 00.. 0.0..” .0 .... ..0. 30. .0 .. V ..0 AV. .. .. ... V0. 139.. .0 3.0.0.... V V V. ...V VI... V . 0-0... . .. :0 I0..- . .. .000. "rd a 0L9 \. :A L V V ..0. .0.-.006000.’ .0 «00 u. .0 00.0000201b0000 02.0. ‘V 3. Q. 00.001000. .0 I0:.0..0.0.’0»0f. V0. 30.0.0.0. A0A0 0.30... :00 . .0.... 07... . 00 0:00.. o ..0... ... .0. 000. ..0-A0? 0... A00} 0 0. 0000.00... 000010 0‘ t. 0“: [06.60“ 0.00 V 0.0.0 .. 000000 00. .00.!000 000.000.! V. A..V.V... 0.0 0.. . .0. 0.. V 00 00000.0”. $020-... .0.: 10000130.. :30)? ...I 000...“: A 000 . . . V. ... . . .30, ... .330... . ”an? 000 3...}... .30.. .0 .0 . 4%. 00V 0. AA» 1.0.1. 51. ”1.0.3.0.. V0 20.0 ”.00. ..V. . 0-0...».‘.u. 0. a 000. ... .Smwui ...... 0 V V L030... V... . 0. 1.000 0 ..0. 0 000... V: 0.0.0... .00..000A.VV.V.. .. V _ V . V . . . V V .0... (aim: 00.0.0300 “000...? 3.3 .00 0. V... 1. IL ”Jud/.0! . 00.. “”01““... 0 ...00 . I. (00:; A .V 00:0. .40 0”. .010. ’0 00 . .... V 0.02.000 00.000.21.120. 30‘ 6.070; 0 ...0: .A 0 I. .0.. . . .0.. V 000 0.. . . . . . V V V .... ... AV: . . . .V A .0.. . 01000.0... Sf...“ 600.03. goné 0 0000. 000050 . . 00‘... A 0A.. .1 ...-V . ..A. .6... 1 50.5.0. A ... ._ .... ...“U...._0.VAVV.10$.A. . .041. . 0.0.....0.00I"0.A.00V.1...V0.0wu...1.4.10.0:... .03... .0 ... . .... .0... I22... ..V. I..A A 0... 1.0.0. ....u...:...A. .V . .V . . 9 19.20.10. ...... 0 .0..) I..:3.I:>.f!.8 010.... I 0'. .. - ..0 .00. .0 an 1.. . 0 . y 0 In .34 0.9.0.4 .0 0: .. . 06... . ...: ..V 0.. 0. .. 003.000.. 1.00 . ..0! V2.2...A00- A 0:. 0A... .’ 1.6.2 0. :0v0VA .20: . r2.... ..0. .. 10.30....VV...A..0.V. I . .0.. 2 .0. 0: :0 .0... ..0... .: .V . .0... .. 09.4.1.3...230000Vtfilvon00300’ («V 0 V . u“ .0. 0 A} . .V 0. 0 ”6“.“ 00 .. ..vV. . 0 0 0... ~00- .0. 9.1V 0.“ V V v .0.. 20.0000 S 0. ...006. 000.0210 I. 0 000. 0000.03. 00.00 ..0 A 0 .70 V ....0 {0.3. .A A . ..6 3.0.00 .V ...: 210.. ... . .. .... ‘00 00A :00 ”.30" :1 (AV. . .Q- .0 1“ .AA. «.0.. 0 ..0"; .. ..0. .100“. ... .3. .3. ‘0. 03 . JV. .. In". 0.’ ”MW? 0.000....0u‘. .3 ..0-0 . :an .0x 0 00.0..... . “7. 2.0.0. .0.-P... .0. . ...-....V .0303} .3006 AVPI 015.0 0000 rV0Vt3s 36:... 0” . V . V V .000 0.00 ’ 0. 0.0 0 0000 . 00 .. . 00 I .. 0. 0 0. 00.. 0-0 0.3.. .0... .0 ’00 00. 0 0A 0 . 0A 0 O 0AA . . .0 ’ 0. ’00 V .VaVVVifi. A...“ 9.30;... V . ..V a 5.1. .V . .55. ......V... .VVFAaVun. .. .. ..V. .. and“ 2.. . .V... v . . . .... .-.... . V ...an .3939"... 30.0.2... V taxis“... mm. 3-5- A... 0.3.03.1... . 9.5.0.000. 0t:- Jlt .65.»..2. 6.0411334. .00!!- 5.000.100... ...5...0.0......0.. 0.....2. . 40$.3!3..‘.0. .0.... 80.01.135.231: ......u. . 00.00. ..3000..YV0..A0 80.... . . 000...... 0.00”.nt 0 .35.)... 0. ..0. V ..V V ...V 0... ... I 0 ..V : . . “I V. V 0 0‘ 7001.00. ..0. 00. Juno-.... ..0 V. ... A... .00. 0.000.390... .0 . 0.0. .0V. . A . 00- V . . . .0.. . 000 - .V . V .V . 0A.. 0A.V0V..A. 0.:1 $.00... . 5.1.010.“ ..V-.0 ‘uogel V V . 0m A... 00. 0 3 03...: .00-» «flunk-V 300040.00. ... ..0... 0... . 40:00: . .0 .20 J. 6. . AV]..V.0003.0...0.0..0V. . . 0.0 ..0; 0.2.3000 VI: £0.03... .0. . 10. A. . n .0... . 3600. . V . V V V V V V V . ......0... 4.0.3000 «0 A.0:32.._VA30V ‘ .0 .0000. 0.0.0.3100§ 00 0000);; .0..- .0. ..0-10:50.10 0,900. 00.000.050.06. .6. 0:. 0 0.103.300.2971 0 07:00 1.00.. 003000 00...... .0..00.00.0.. .V0.....0~0’....00100. 00.. .. 0. 00 . V . 00- .000000. 10.0..- ..0'J.. AVA... .60 0..“ 3'01): 0.00.. “0‘ .n.“00 .0... ..V ‘00 A. . V000 0 690. 2:... 00.0.11? 0h0000‘V.AA0:.0.0.3 .00 .....00..fl6. 0. ...01‘0001010'00000'06”... V .: 0.0 70! 06:. 0.1.0.7. 0 .0.. 3.! 0.1.32.0. .0 A0 .0 0.00.. V .V . 00'2”.- ...“.0: ..‘0..’VJA..A 0’...— AA.‘ .001...0 A400 .0.-0.0000000... AHL.'00.0.0 ... .0....0.‘ 0 '10}: $000. 0 0. 0. 00’. . .0... 0.0.0.000 . ..‘ara’WD’g. 0 . 00. ..‘A .0 . 0000:! 0 0 )00 .0. :0 .0.-.00 3000 VV. 00000 09. .0 . . . ‘.00.00 01%”. .. 3.0.0.0000... 000.3 .000. 0033.00 .02.. .000... 00 0.0... 0H0. .. 0 . .0 .‘. ..0 0. .3...‘ A. . V. "000 00- 000.0 0 .006 A 0- .0.. . 0. 00. 10.00 . .0. .0. .A 00’ 0.00.06. . .‘ ...0. ..0." . .00.. 1. .0.? .017... 0-. A. AA. 0... $050003: 0 3.0...D0AQ ..0“... V 3.02000310213703'3110.’ .....200 ...-0. 0.30. ..0... ..A.... .....p... A0. : . 0.0.. 30030.... 1.0.“... .3200!!! .... ..V .V “V0.7”... . 2 0 0.0.0 0030 $0 . a ..0 . .... A 0.0.0.10 .00 0....00. 0-0... . - 0:: 0,0 .... .:.0.00!0.V .00 00. .A- .0.... 0. .0. 0.0 . . V ..V . .0... 0...: A. A 30.0. ..V .V A .0- . ...000 0 10.0 0 00 ..0 0 .V . v. 0 H. .0.! . 0. 0. 0. .. 000:3. A 0.0.0.70. 0 .IQ.000:A¢00 00000.0“ 0 0 €0.00... 0... 0 A .00.... . 0 .0-0: 0..- 0! ...V . u 1 ._ 0. 0 .00. U...- .0 ... 00 - ..0. 0 . ..00.. . 000.0. ..AI... ’0.- z A...’ V0.01 I 0030-.- s... V00. VA 0 0.0. ‘00.: ‘0‘ Fl 00.001 0.. ..0! O'lI 0., .00 .| .0.. 66!!0.l..“00‘00 0 .0 .0 00.36.00 .00lnlo..0.000..0..01 00.00. .)%:9 .00. :00... A. V 0000 . . A: 00.0. 0. f0. '50 ..0 '00.. .. Adv-0.0. III! .... 2. 0 .0 . 3.0.0.0003... 0000 . . . .010 0.0. 20.70 0. . .0. 0 0...... . Ar. 0.00 .0... ‘0 .... V 1V.» . .0. . A0. .0. . 0.0.. . ..V..- 0 . 36.1.9320. 002....0 . 000...... . .0. 0...”.0 1 0.‘ 0. 0.00 03 ’0.00000.00...000A.. 0000... 00 ’ 00+. ~90... ...! . 060:: 030. ..V 030.... ‘0 '- AI. 0.A ..0. 00.00 00 «0.00. 0". ”“000 .0. A30 00 .00 00 .AJI .0000. .000. . V. .0 . 0.30.0 M0 000.0.V .0 33.00. ,0:0 .0. ‘1000... .0..V.ol‘...i.0l.000,.0.0000 0:.0.0.000...‘0II0 020000 00 :0000 .00.... . 0 0.0 00000 V0 0 0.0.000-000.0V000.0!00§- ..0 A ‘0 i Y .0.- V0 0000.. .000 0.0. «Aer-0000 . . V .0000. ...‘000 0A . [A 1.00.. .00Q000 (0 ..0 00.0. 00.00 A ...-A V 00 . .‘.v .0. ...-00 00.3.0..~300 . :00; ..V .000...000‘.. 000000.3‘000.§0 0 ... . ‘0‘“ 01.00. 003:3 ; 00 00 V 00010.2... 20:00.0... g .. 06... ..0... 00.30.00.000. .0 0 . ....V .V 3 ... V. .0 0... A ..0. VA «.0609 .000- V . 0 .0.: : .0 .. 0.. 0 .V . 0. 00V- H..0.VA 070,0: .00 . o 0 .0. 0 . .V...... . .2 .00 0 .00 .V .100: . .V V. . . 0.00 00 . 0.030 .00...u.00..o .0.‘ 31300013000003-‘0 .0010: .06. 0300.. . 01... 0 0.07.0. ...... A. .-0 . ”VIVX... A. 70.1.1.0: 3.20.1.6? :7 0.0.0.0. .0 20.0... . .. 0. ..V 0.000. 0... .0... 00.0023». 031,20..- .0... 2.0. . . .65. . 0 . A . . A ..A. 0...... .- . . 00 A. 3.0... . .0 V 10.78 ..0... .00.! A. V......... .._00 0. 010...... 0... . . 302.504.00.103301. a: 0000» .0... V V r20'0 0.0.000 0 ..0 '00. .0- 0 0. . 0 0.0 ...-.1. .6...\I 1.10.10..- A 00 0... 0 A0 .0. A 30%.... 70.0... A 00 .0A. .9 .0 . 00. V0 0 0 :001. 0... . .00A0 ...-00. 00 0V’000VO0A0 . V V. 0 . ... .10. 0 ....0! 03.30 .. 0.0.0 . .0.-0.0.0.00 0. .0.-0000’. 0. . ..0!“ V V. .. 0 0090 . V0.00. 00501.00 .... .0.-.0000“:- «00. 000 000.. .V .00 00. 0 .A .V 0 .0-0. . V... .- V..0A’0000..0V.0.0 -. A... 0040. a 0.001. 000.90 000. 0.00.00 ... ...: 00. A . V. ..l -.00 A. V): . A. . 0... ....V V . . 0 .0 0 3.3.00. ...:‘0.0V 0000.2”. Ah” .0 ...-VA. . .00 .0030001I0V 07“”...- 00““: .‘0..00VV.A0u. 00.. .0. .0.-.3110. . .‘0.130.A0.. .. . 0 0000.00.00. 06.. 00.0.0 00810000.... 00.00? . 000.... I .0... .V .00 fo‘o ..0. 0. A .’..!V 0.0 000.... . 61...... V... . ..0... . ..V. V 0.... ..0 0 .00.. 3.90030. «31.3JJfi%AQ.A300& 0.00. .00 .0. .0. . 0 .. .0071'10 V... ... ...0V 0V. .00... V 0.001091} 0.00. A...0 A0001....V.A000 :0.- .. .0. V . . V. .000: .V... .1. . 3.0. V . . . . . . . 0. V. . V .. V 3 0. ..0 .6730. .13... .0 .0... .00 .V 00.0. .V 3,0,. . V2.00... ’I?‘ 0 ..0. .0. 6‘00003. :. 02...)! ...iyzlJIaflf 13.0.5.2...‘410 ..V. .03“. 00:0. 003...... 01. V.» A I $.00... 10...: ......0 33003.00 3. 3 00. ...-...A. ..00013: . 2:20... V . . ...A 0. . .0... . ..V . . . .0. - 01.00. 0 . .... V . V .V .V 0 V . 0.. 1:0 0102...... ..A’. .0. ..0... AIDA-VA! 00.0.0.0. A...0:0 ...-00“.. . V. - ... V 0 0 .. ...0 V .- .0! .. 0. a . 1. 0.. .V. ...JV . 0 0.00600...:0 . . ... AV0 00.0.0.0 .4 .0. 0... 3‘ .0.. u A 0.. . ... 0000.. 3.010.. I! 00. V 200.60 0. .0 .00! 00.. V. ... .0- . V. . . .. 00.23. .... .0. 0 0 0 .0. .00 0... .00 .00 .- I ‘10 v“... 0‘.th ..0“ .06 ..0 0' ‘11008 0.30. 0AA. . .V.. 0 00... 000..0w~.0 S0... ... V 0 . 0 000- : A - 0. 0.004. 003000.000... 000.0 .0.-0.000.:0. .0. .0 0:00.30 9 a. .3 . .0 3A0 0. .0. b . 0... 0.0200“ .... .000. .. 3.6.0.0.. .. A... 0 «. V00. . . 0f A ..L VIA-‘00, 0.000 V. V 000‘. . 00F. 0..” a .. “.0 A . 0 0 03 ... .0 0 I ...00. 0 00 0:0. 00. V 0.00.05 .0. I V. 0.... v0 . V. V... 041.00 .1000 .0...:... 00.. ..V ‘00- .30.: ..00‘ .00 .90. 030.00 .. 50.0.... V -:0 ..V . 00.. .30 ... 0 A. 0’. . -- 000‘91000000 ’00! 000 ..0..0... 903.13....P00u0 .01... 02V. .. 7. I. I... 11:000.} Angb'? . 00. ..- . ‘0“! "00“ 0 .6010. .200. ..0-:20 .... 000.0109. 00.... o I I1... ...0. .0.. 0-1200010... V 0. 3.2 .106: VA: 0 . . ... ..‘V V. A! . A0. . 0 0 .... . . 0 r0 . .04 . ...... . ..0rV 00: V00 . .01. .00 .V ..0. . 60!. 2.0.0: .0...V00.0:0._“..0.. :90 0 V. . .V .0. 0.... 0 0.. V 000 0.0 03“. 0000 .320... 00. $0010 {00.0.0.0 0.: 0. 0:0.00lv 2.016". ...-.0. ... 0.0. .0 00.9 ..3:.....0.0.~A.3 . 0 0.100%... 3.200.533! 0. V3.1... .6: 0.0. 060. 0000.0. “gating... 6010‘..va : 00-1001! ..0 0 $0.0V ’0 010.0. 000.0. A . . .0.-0 n.‘03VV .Y0AA ‘07 . 01.10.. 0A 0000. 0-00 A0. .A 00 .A 00.0. - .0.. 0 A V ... 000 V... . ..0 .00 . 0... .0.. 00 00 . 0 0 .. V. .. 000.. .0 ..V. 7.00:0...0. 000.00 90 0.0... 0.05.0000. 00.0 . .: 0. 0.0.0 0 0.. 0 0. 0 0.0.I:A00-A . A .00 00009 0 I... .0.!10. .‘0. 0 ...0 .0....00. 0 0. V V 0 0:0 .. ..V 0. .VV 0. 0C. 0100.00.00 . f. ”00: .50. .0.} 010.. .01“... A,0._.AA«3‘.0A 0:...0000.00.‘0 .- VI 0....0 A VIA-0.10 4 A A .D 4 .00.... 00.31.90. 0300.00... .TIA. . .0.0A:0 . . .V H.0....0.0...>..... .10. V“! ... . .. 0.00. V V «0 .00.. 0. I. .- A.. 00. . ... ’3 0.0:. V. .00. 00.. 5‘. VI ..0: 0.. . 00.0000... . u 01: V . .10. ...-0.0000 0.. . 0 . . 00 . V . V .000 .0. v. [I . 0.20.... 0.0.00.4: D ”0.00“!” ’0 ...-0.00! V3.6 00...." 0 21.00130...“ 0 2.0000333 lili‘é‘ 0.30 ...-000010304 6300 0 ..00. 0I0.0A.V0..00... 0.. .0 3:010; .0.. 0 .0 00. 0.0. .V0 V14.- 00.0 0-: V. . . A0. ...-...! A10. 0... 0:00.0'...V...~ (000. 0:€-L0r0_0 0 V V «V V0. V... .0000 0. V . _ .000 . .00.: .V.. 0.. “A. 30.0.0.2 0’ .0.... 0:00 0... ..V-... I 0.0.”... . .0.-....- 00.30. .0306». i. V ‘. u. ...0. 0.10.. ... . . . 000V0 0.0. 0.1.2.3.. 0 . V. 0. 00 0 ..000A9 l“‘0v!. ... 0.0.0 V. .0.-0.00.“..- .0 . V utrr. ..04 0'10! 000 .AV..V . . V . . V . . .0 V 06.000 0. ... 00.0. V .70. 61.0.66 .. V0...’i000.00.>0a000.. ..0 30.0100. 60...... . 0.10 .0...0.0.‘00010.. V... 0000 000.. 0.. .. .. t 00... . . .006 A 0. ..0 0-0 u. . 0.0001! . .V 000-00 00 t 00 .0..th .< 0... .:0 0 J. 0 .00A. .00 0’00AA .0 *0. : 70...... .. . V . V. 6.. .0. .‘V.0.‘ ..0-3.0.00.3... 1200. «...-0. 0 000. .V 34.00.. .00. 00 . V ..V V . A... .30.. 0.500.}. 1.003.! 6. 2 . . . V 0.. A... 0. ’l. 0. 3? ..A. 0’ . 0 . 0 .... 00.000 . V V ...001 0 . 0. 0. V. -. V 30000:...1. 1.000... .0.. ‘60 0 .... 00 01A 1 00.0..00"..0-000 0. 0. 0.0.. .0. 30.0.2.0! .: 0 .. n 0.0.. . .0....9 .00.... 0003‘. . 09 ....0. O I . II... 0 V... .0 0‘. 31.006004.001 .. 1000.. 0. A .A 00 ..0000000 000. A of. ._ .. .0V 0..” {-000.00} 900000-00, 00A. 0>.I.l. .0. 0 ...A. 0. A. . v .0... A . I- 000101. 600.009 0.. ...-.12 6.. . VI...- .0. V VA ..V.0...0 0-. ..0. V ..0. .. .05? 0. .V.. ..0. .. A. ..A. 0 ... . 100...... . .V 0 0 0.. . . 0000000. d ..fiAfosz ..0 0w...“ A000! . . kVIJCLA09000000VV0.".A.1009, 000\ 0 0V. 3.00 60.00 .0. 0. 0-.‘0 anflno. _ 30.00... - A010A_"0..00 ...I... 0.00000. A V. 0...: . .00....1. . 0.... 0 00.7.0.... . 3 0100002000... V .000”...V09.I..0.001...0 . .gh‘VIVu 000000‘.‘Cfl0fl0”0-0 I: 0000.. ......000: n9..:00.000A—.. 0‘ . ....I ..9.000“:..2......0)..A000m00.0l. . ..00A 001 ..3. . V0... - V .- A200.. . .0 ..0. V.A.. ..0-01.060... V . .00...0...0.0.. V. V .0 .0. 0.040.000 0.. Z:- 0 3 .A..\ 00.0.A0 ..A .. 0...0100310 ‘1... V .Q 00,030.... . .0. CV 100A... 00...}... .0! v o A“ V.0A.. 0. o 0n.- 000 0L0 “.00 .0 Vfior..-v_0. .00.‘..A 00.0 p :0...0. _. . .V:10.000u9:0...... Q00 0. .0... .HA 0w 0. 0 0 ‘0! .VV. 0V6 0..0.. ..0. .00 A 0:00 .00 . A0A..V 0:00... . . .0.. . . o 0 ... . 0.00:. 0. . .6 A . V. Wino-WAX} §..~0Nu......v...‘bu._t:~\fl0rV 00.2.3031... . 0Hb30 0. .0... 0r. ““0“”- A 30000) 0.0.0:...V: . ‘00,.00A0nn00‘0‘. 00. AV. “ .0 00 0. 000 0 ... . 0.0...12 V0.0..I:00. 000.100.. .04 ..V: ‘0 .0.. -. ... . .. .0 ..0... . . V . V0.00... 0.10.0003! 0P4 xii-00% é. "00000...“ ..AA. 0 10.. V ....0 0n A20... 0.0.0000rfu.r.00:_:0. ... 000-11... 0.0.10.0 I0AI._0._0..A0..10 .5316: 6.;01000‘1 003.! ’ ‘0 . .V . . .V 0: “0.6.. ..000 . V0V 000.. “0.0... 0.. 0. :- :t. o: . A. A V ..0. . V . .. V6.3... .. . 00.. .0. ......o . 7 . .033: 0.. .. .. 0 . :00 0 $0090.. 1.0.0.. .7000.- _..'00A 0000—” Jrgsnfi‘k‘ 0010‘. “l ‘00... ...6.00.00.0.. . . 03 . . 2.03.39.00.30031. 60000-00... V A . V ..V 00.000 .00‘..‘0.J.00.fi‘00.\0.0. 00" ...-.0000. 0- 200$”..‘1H03 ’0 .0.. 0000: : 0- 60.600003309000010. .6; : 0 6. VA. 00. . 0 .A.. p .0 . ..0. 0V. .0 . V V V ..0 ..0 . .A. VV. .0 .A V ...“!!000. .... 00V .000100‘0“\‘0.060I. 0.00 00000300.?! 0000’ fi #1 0. V000 VII-1.300902003000000 0'00... 0 A 1:: A00. 00.00 . . ..0-0' ‘A 0 . 0 - 000.6. 0. 000 .:‘0 9 0L0... v 0. V0.A0 0‘: .. ......‘U. .0}... .0300. :06 V000, .. 30.... .. . V00 01.. 000002000. A0. V 0000’... 0.0.0 4 A ..00‘ 6000. 0.§’l.‘o{s¢.0‘. ‘00 ziti ’ 0 . 0000.30. .00. 0 3:00.000“: .. ... 000.. V V . 0.3.. 0...? .V. .0.-0. ...-..0. .0‘0 .1‘. .3000 6.0.60.0. .0: .0.) A .V: ..00 .00... V . .0 .10 A 0 30.000.30.00"! 0. 0.0.. 0.0000 ..0.rV"110. 00000. 0000‘s... '0’ .00 :0... 0. s0 2.000.‘ A... ... b.0011" V V V V . .V A. .00: .6ch 0.20..- - . . 000.00...) 0.. . . .000 .1-0510007000. 0.30.36.31.10. .0. . . a .3030}! V . 0 ..0... .0.... A0 .. 00A . 0:02 A 00. 0 0.0000. 0000.. “2.000005... 00.0.003‘20 0;..11.’ 0 V‘. V 0 .0 o .0. 00.0 0V . V V :.. . .....L...00-DA :A: A ... .0 .. .0.. .0. 0 A. 0.0.. : ..0 . 00. ...VV... . .0.... .. V3 0. 0.0.0 .0 . 0 _. 0:. . 0. .. A .6. 00.; I. A... 0.. 3100.00,. 00-000? 0.0.0.00. ~:00. 0000.00: V. .0402-A0 0 . 000..va . .V V - V. _ V . ‘0 10.06..’ .0 6.00-. V . . V. V. V .I z... o IA-00T.0..I.....0‘0V0-’0.0Atvc...01.0-03.6!71 0.00.0.0. 00.0 17‘10‘0 . .V 0. 0.....0V.00-.00...’V.000% 0... .II. 00.0..... 0. .120. A 0 . .00... V ...-1.1.0... 0... V.. 0. .. .. 000. 010,010! .. 0‘ .§I.A.A V 0:14 . V . ..0. ..0... ...... V5.0 ... 0. V. .0» 0..V0::Ao.000‘ "120-0 .AA. . 0.. - '..I r V. .1. .... . . . 1 . T). .v V. . . . . V...” .0. I V 0 0 AA. ... 60.0... 0 V0 0. 00A... . . .' 20.1.00. 00. .0.. 0. 00"«0. A. .0300 0:.000.A.00JUA‘0~0”.0 0.00" 0m!) 0. J . . V . 0. . .00A V-J 100.60... .0.. r A0 _ . . 10.0.0.6 .. .V . .0. {3.0.1. ' ... V00 ...0- ...A 0. 0.. .010 ......0Vn. ..0. . -. .1... A 2.. .V ... 0V... V. 0. .003... ...! . ...0..P 00.00... .700 03.0 3:. 0 .. AAAV‘80V300. ..0.:AAA.‘.V 00° 340.100.100.04.” . . V V . . «01040.0. 0. 0 A3000...“ 0 . V .0 6. .0. 00 AC... .0 0000. 000 000.0:3 0.1.06 0. 0 00-0400. 000 09.00-000.3300300'0003 0 .A AA A. .0.-...: 0 0.0.1.. 0‘ . 0A: ..000. :1100300 ’A... 30., 03‘. 6.3 .00’00’00 00000...-000 . 03x10: 00 .V0 . V. V . . V. V 0 ..0”?- 0 0V '0. .0 0 v. ..0 v... . :o 00 ‘0 . ‘ . . . VV0 .0006 l 0...... . A V00 0 0 0.: O . I A 0. . V A .- .0.\!. 00-0 . .- 00.0.1.0 .000 .0.—.0. .. ..0. . 0 ..0 1.0:... 000.000.000.21... 0. A»: "00.0., A... v 03600 . . V. V .0 .00 0 u 0.00.. ..Vdo..?00~03..0AV.00V:..0A 0. 0.0.3.02002‘0. 010...... 4000 0. .0: 00A .. I... ... V. 0 A. 30.00.066110"... .A. .0)”. 00.0.V 00s.. 0. ... .. 0.. . 30.0.03. 6‘ .’0.000000‘0A000 0:663“.— 0". 0"". 0 0 0. .V V V ..0. ...AC 0.3 . A. 010.0: 0... .. 61.0014 {v.0 V..V_.. . .3 V V. ..- I02 .2. . . A 02.0.... A. ... .322... . V . . V...A. . .... . .V V . . )..3....0.I.00.. A ..0: I... ””3300: .113“ ~.A‘00.00V. ... . 00.00.... 600 Q. A... ’00 0- A .00. ..V 000 0 .. .... ......100AV. 0s 0.)..Y0... ..0-.09. . V V .010: .503: 3h. 3.00% "”000-0303‘0t00000-V‘0 . ‘ 0 0 V V ...0:00.. V 0 V 9 . : ..A 0 .‘0. 0. 0V .0 .0 000 .0-0 V 0. ..0 00000. .10 0.: .31.... .0.. . 01.03.0300... .A‘~.0 0. ‘ 00010;; 0| VN 0... .3.“ n. .23‘" .0....00'0000101. V V 3.3200 260.000.9008 0 V. IVV’Ad... 2...... ... . £710... . ... IV...- 00...0.A V 0.3.0 .0: 0.. 0.0000000... . - 006.0 00.00 2.30.0000‘00‘0000 V . :13. n V 02‘. .0 O .0. . 30.0.0 .0 . V... V V. 0.. .V ..0 .6 ..V .0. .119... . A. . V . 09.07.. 030....000A k0. V..’.n......00.0'u0)00 6 00,0».01500000... ... :08 0.0 0.. . V . V. ..0... 30.01.00 : A...‘0..00 .0. .0.}00...) . .0000VIA0OV 6.0m’0'0. 000. ..0 .. ... 01...). 0 V. . .’. 0.0.0.0- . .0.-00.0 . .V . 00.070.00.00. 0. .0. 0 . .30. ..V 00.000 . .0. ...A. :00 :0 . 000.00 :7 .0.3..‘0. I... I. .‘.’i’!."‘0. ... r {00000020 :2... 0A . . ... 0:00.; ...-V .V “2.1.0.... .l.. _. .. . V . . 30.4.0.0 .0.. 0.6:. 0.30... ...-.0.. 0.0.0}. 600V .0... 3000‘...- . 03‘ V03...- 0 .V 0. 0. 0.0.. 0.. ... 0 .0. . ..VV0. 090...... 1.00?! A .0121 A A .0. 0 1:. . .V... . . V .000: 000 .2000 ..0-000 0!). A0-l0.0 i0!!‘000 .0. 0.0 . 0.0.00. .00 0.... V0. V 0. 0 . J.‘0:.r . . .. ...0. A 0.30 .00 . 0.. ... 0.! V0001. ..0V ..000A0V .. ..00 0 00.00. ..0. V . V .- : . .0. 0 0 .0 . .0000 ‘3‘:- .10 0 0 .0. . V.JVIA’.0AV .0.. .000.- . «.0000.‘ 00006 ’0 000900 “0‘0500V‘Jn '.0.w.l ... .. 0..0 .00 n .006 .0)... . .... ... 0220.00! 3. ... .3 :01: .....H .000 .0 000.0 I A... .10. 0’ A AA!A00.AI.0 . 31.00 I 0.... J00. :. . V . V ....000 :V00'0.000:0.0A00I0 .1.- A. .00.... .n‘ 0.0 .V. . . . . .0. 00. .0 .. 0. ..0: . ..0-000002.030 .VSJAASi... .0. . 0.3.00.0..0‘0 .lg’00‘0100 00 .0 .V 0. A ..0 “0:040. 0 00..0 0000:. : 00 4 3.00? ..0 0 0 .0 I11... 0:: .0..0..0C.I0 I.- 0‘0007. . \0 0O- 000 A o. 0 0V 0 . V 3000. u 0 10.00 ..00000000’00000. I000. V . .A . V . V .0 a 00.0..1. 0 .0... 00:30} ..IQ I... 430.00.. 0... V0.10. . 0:0 00 . 00' g 0,0,: .zAAOVOSg‘Vzo’ . - ’00. . .0 0.. 3A l0.‘-.~01¢0.'4 0.. .100 H .I‘AoA' . .0. .-:00. 0. 0 .02.... 0.0-. 0.00. 001L..I..000 00 000.0 I 0.000. .0: . A . 00 003.3 0.00.0:0. 3.000 A0 ..0. .0 .0100. . ..V. 0:00. 0 ..‘.70.0. $0.. .00. 0A. .00 0... 0 6'. 00000.. 00.00.! .01.-. AA? A: 0.? 40A .0 V . V . V. .00... A.. “01.0. 0. ..0 .0 . 0.0.0... ‘0 70335010000300. A. .000“ V 66:0 0. 0 0 0 I... . 00 V 0.: 210:.0V‘00. 0. 0A ..0 0V .0.. 0.00 0300..., . ... 0 .00- .. 0 0.0.00. A .V .. A .3 .0 .V 00 .0 . . A 0 0.0-... .V . 0’! O 0 A'I.0 : . .0... V . .V 00 ...:0. .0091: . L. .04 3.00 .- 0 .Op‘g‘l0 ...0‘0. . 0.0.06“ ...-1.0.0 or“! .I 0AA V. ..V. ... .00. V0.000VWu00A .4 .1300 A0. . A. . 100....0: 0 0 0 9A :000VV0 .... ’00,. 10.000.44.00: .0; ‘0... 000033.00 000.....:..0..... .6101“ 00- . J... ..3100... ....“1‘000015 :0. .0 031.020. ..-.0.00..0.. .. ..0. Av. 00.. '7. .2.’.:.00.. . . VA 0 7-01.23? . V V V. . . V . ..V‘I0. .. 5.1.2“. .‘IA..00000§V0‘ .00‘ 010000,. ’43.. 0.01. . A 0 0 0 V. .000. .0. 00.0.0 ..0 .0 I. A 0;. 30 0000.40 . 0 ... 0 0. (0003‘...A01A.0A .3’I.00V0000:.‘:V -. A I ..0‘0V0: ’7... 3:00; .0000VV000. . 00 00.0. A .. ..-.01 . ‘20.. 0.7 00.0 00.000. .... VA A 0 . 10:100. V . . .0. 0 00.0 . V. V . . V V . 00:00.3 “M06 ..0-0‘0 £0300010i0‘00 .0.‘ 1:000. .’0 " JAOV~V0 0. .7 N... V A .... '1... A . “A . ... .0 0.0.6 ...0 . V A. .... . .... V. 0.1.1. 001. 3.6 0 0. . V .V .. .61 .A V. ...V- .00.. 60. 0. I. .. .. 5.001.003...) ..0... . . ... . .0....00 4A0”.- .04....0. 30033000 90.00... .00000A0II:.V.0AAA. U“. 0%. . V 6!. 0‘ '64 .0- .: 0.. 0.0-. . A... A... ..6. . .. I. . ...I00 ..0-VA 1.0030. .. 62.2.. .00 ..‘.1000 .0... 0!: ..V ..00000. 0160:0000... . 0 . V 0. 00:0. .0 .0. ...).V .030 0 L. V. .. .9 . V..... ... 0 V A .0. :0 Z 0 .0- u . 00A .. 01:0. V . . V ..V ...-A. ..V‘ .l V .AAA. ..0-3000.000 00003.01.- I’D-0A?!” ‘0‘060. ... 30.37001... . N V 203.10. ... . .....V 2.038.310 .02...“ . 0. .1401.-. ..V...00:.. ........V¢. .01....V..IAV.3....A0V: . ...... 0A. .011... :- . 0 0 . .0 0.0.0.0100!}00030.0V.0.300.7:0.:0.. V A: :10... ..0 ..0: . . . .. 03090... V . V . ...! 4A.... . Al’h. ..0-A00101 00.00.03.001101‘I93303I‘00 .0..)53955‘0 0 . V 06 100.0% 00.9.7... . .....0 V. V . . 2 .0. ..V. l . ..A ..0. V 0 {39.1.0902 V A AIAA . :.._:0 .01. 0.. .0 .0 00.0.0606..- .!00. .. 00.0.8 V 1.00.- 000:. ... 0.. l A0 . .0. 0.. VII-0.0. 0.. AVA: VA A. A. 0A.. 20.00.. .. . . V V V . -30.... A. 9:00.. (01:02:30.5). ..V .‘000 0004-000‘ ’0’ 0‘30; V J}: .0 .AA. .0. ..0 ... .... . 0 . A ..V I... .. .10.... _ (I .0.... .V.. 0.0.7.... .... 1:0. 02.01... 0.2.. V . 'V .. V . V V. . . .0 ’2'. . .A A 0020. . .0: A...V7 0 .I..... .0.... V . . V . 0. 01‘ 1700.00.02. 30.0 0100 .00.”? {00.00.0101“ V . 09... 000.10.0‘000. 0.00.000'00 V..... ..0 '00 00- v: 3‘ - ..0-:38... .0. 3.8.00: 600.0 60020 ....0 .00003A0.000..- .. . . ...?000000000... 30.000.30.20; . ....0V ’0 0.0. A .0.-... 1.4.0.0... 0. 0 0 .o 0 000 . 0. .0A V . . . . . .V..V1..V 00.00.). .9.’ 21,0000- 0.. ’V’..00LH0 (0’... ..W0. 0 0A.; 0000. 0 0:61 . .0.. I. . 0. O0. 0.. 0 V . It! .6 0‘0. . .000: .0. 03... 0 .000 0 . . . :00 V0 A. A 0 .AA . . . 1 0 ... .A 000.303. 0 V 00.. 00 . . V. . :V . A 0 .0. 0 A0... .00 .000 V... 0 0 39.00.: . ~000V0v0. 0,0 0... 0. 0.0.. 71.0.. . ..00 00‘: .0 0.0 50! A00. .000 .0 0- .AV. 0. 3.0000... 0- V000AQA00 0 :- A V . .0 .00: 0..., . V V . . 00V00_ O 06.0.0305. 0.0.00. 9.000 0 .V 0— .0004“; 0 V, n. . .IIOA .0. ..3. 0 n 0-090 V .A .V t... . 6 A 1.0.0. . A. 0 ..AA .0."? 00 : 0000.. . . . .21.“.03 .. 0.0 0 II 0‘30. V . . . V .00300 ...00.00.0 0.03""00000 . 0.3.0 00’ ..0 0-0. 00 000. 0. V 0.0.00.3. 0&0~.0.r.0.. 01003401023000.00. . . .:0..0 .000. ...0A0000 00:... .u 00.00.0330. 0 .VV. 0 .0 A. 1.000... 0:. 0300.3. \0.I 00.0.. ..0. 0000.. ,0. ‘0 00.0.0”; 0. ... 030000.}..1 00AV 04 .0 00. . ..V. .0.. ..100.30§00.0\A0.. l- 0. ......i 33.0 60‘00500.V.:.0..l1040. . .f. o .0 . 0.01:0 a; .0... .A. . . . 0......1' I0 0.0 . .00.... 0 1000—! 0.0... 00.00 .. 1:0... 0 00. ..0 0.0 V V .100. Al . 09. 0 00 .I-0.0.0.0 ‘ V0 0A . c . 0.. . 03‘) .3 00.000 01.0, 00‘ V) 00000,... A01 0.» .0 0:000 :00 ON ..00A000J0 A I. K3003... A0 ...00. ”‘0 :0! .0 0. .00 00000V .00. I00. 00 . 0 V .0 00",. .0000. . . 0.0. .’.000..'00. 01:00 0A .0000 000 .V 0 A0 'I 0AVVr000II0.’.V‘ 0v . . . 0,000.30: .‘0. 0 V0000. 0 0.0.0.1. 0 ... A 0 . .0.... . ..A0 .70 . -0..0-0.0 . .0. ..0 '0 . I V. . 0 3. y: .000. 00 0.. 0 .0. .000.’v‘.u .0. ..0-6:6 06,00... . . 0. 00. 00000 0 . 0.‘. 00' 000...}. . m 00) 0.10.. . . . V0.0. I. 4. . 320000.... 61403.1...1 .3\LA§30 firm-3.0.00... V. . ..000.....-..:00:_ .0 . V'IJO.A.. V .. 83002021010... ..."..31. V V . V . V .0.... 0.? .0...0-Y0.:...0AA.0.. 0 0... 2-... IV. . .0 V 0.. .AA.0A ..0... V. .4000... ...!01 . 0-..u‘t0. ’. .0 .0. .6300. IV‘ V .6 A‘A...0 ..0 A6. 0 0.. V .A. 0000 A A. ..) 0 A 0 r. . . . .A. 401.: 000 ... ..A. .000- 0.00-3.03. . A .0... .0....030009 V V . . .V . V .s 0.00.00! 0. .0. AKA A ..V . 1.. .0. V0.00 .0 V0.33. . . .JA00.!0 ..0-00 : 0‘. V ..0- .0. 10.... .030 AV A... ..0-:10 0. «0. 3.0..500‘01 . . 0 ...-.0 0900.01.- W. 0.6!. - . ...: VA. .0. 0. . V .060.- .0. 00.. A: 0000 . 00 .A- 0 .. V. V . 2000.03... 09.0: . .9: .. .... O : . ......L .V V L o .. . A... . .A 0 . 30. 0.1.0.00 000. .0.. 0.70.0..00-3110. 00V 1.00.1 ..0. 0 0.50 0 003 . 00. 0JA00 00. A 0. I”00.0...I..0....0V 00. . V .... 30000.0... .- ‘0; ..0 ‘ . I 04-. . 0 . 000:0 000.!0u0....0:1. V V 0 00V . ... .0.-u... A 002...... ..0. I ....‘00 A. . o, 0 0' . A ..0. .1000 A3... ..A’ 0.00020. .0 .00.. ... 0A. . 2 0000.. 0 . 00.... .00... 0.‘ V .:0 030 1.20.... 0 a 0.0” 0xVI ..V A. 00 .0010. . . . V . V .0.. 0...... . A : 0 . 0. ..V30 0 0‘:. A... .0.... . . £00000 . 0.00 0.0 . 3-00.. A . 0Av . v 0 V. ...VA. ...A ..... .. v 0. . ...A, 0 A ..A .0. . 00. 00on- .. ..00..'A‘V0. 6.0 .1000: 0 A. 0 .V ..0- . A ...: .00 0.. ..V-00. . . V. . . 0.“. 3.0 016 KAJ-o. . 003.0... .A ...A- (3. .0. ...AAA..0:.0.. . A. ...A. 0.... 0. 00.0.V .0. :..0 $3.00.... ... ..' .108)?! 0..V..0.VVL0HA.. 90.0.4 ...-00 0 .‘ A .... V 0.00041 02600 A . .0. 0. 0n . . . . .. . .. ..00. .‘0.. 6.0.0.004. .. 0. 00.20:...o00‘3 ...”!.A00.0.A.0000000\.0.I.01.00.70 0.00500 ... 0!; .7 0. A . [..V-0.00. A . V .V 6.. ..0 I . .VV. 0. .I.0\V0A. V... A V. .0. 0. . 01.. 0-00. 0 0 :..A V00... ..V.. . .0 V . 00.0 V. 0.: .0....0. 20.63....IVV 0.. . . . .0. . .0.. .. 00. 990.. 0...; 00 V A0 0. .V . 60. 00.) . .0.: 0.3:... ..0. 1010—0 . .‘00‘00‘0. I . 0.01% 0.0... 03* ..0. women ‘00.‘0 . .VV’V000V0V 000 ...-.0: ..VP00.0.000.0000V 00V ...VV00 0 00V 90.0!003000- 2.”.0600 00.n0wt0l00.0.0000.00000 00’. 0 000-0.I‘000300 0.00.0100 .0 0A. 0. I0.00 A000 .0. .03.. 00.0‘..0§.000|: ‘,..A. .000 . V. .90o’n0. ..00 .0.-0100.....0. O. 0 0-. I ..0-”..0 .. 0 0 AA-000V0 A .0 ... 0 ... V - .3 00.0.1? ..0-0" ‘03..0‘\..)$. fii’00-0" 0 . At. t a. .0: 100.. 0 V '2 . V. 0‘ . 6V 00 ...00000?! V 0.. 30V. A V.. 06‘ 00-... 0 9.0 ...-‘6. . 0 . . .. ...A 0 0 0 0 I 0000.0 .00 0AV 0 0V . 0. .0.. ..0 . 0 .0. -0000. 500'... .I-V. I‘ V 00.01.. 0‘.. 00 30:0. 01"... 00010. 0’00.00.".03000‘V 0‘1”. ....0. 019.5 0.01J0V0; 2‘. 0 ..00 .....00.. I V V.V.0000J. 0.!00000.0A... A .:\.0A.0 3003...)... 0. . ’3 . ..0. 00.. 6.. All? 0 ....0 .A A . 1: 00 0.00 0.A..0 . V, 0 ...-V 0. ..0 0.00030...0 . .0 V. .39 0200.3 4 0w . 0" o. 0.0... 0:00.... . I A... ..90. I00 .8: V0 .udtfl...‘,\‘0i (00. .. :0. .0 . u. 1 1 0. . a. .0. 0'0 0 A. . A .0.... ..V. ...V.: .0. . ..V 0 ...-0 .0 . V ..0 9. V. 0 .. . A .0... .00 .. A0 .0.... ..0 A 0 . .0 A! ... V - .. . .....0. V! A0 V . ...0 0 ..I .06. 3:... .0..‘.0. I! An... 0.. V.10V .3012}... .0.".0'... ....000. .1‘0 0009000010 . .0.. A 0-00 V .100: 0:00.. . V 0””. ..V-00.0.0.0. 0...“. 0V0 . ..0A. .. 0.: 001] ..V ..0 V 0. .VV 0.00000 . V V V . A :I 0 AA V. .0 . 000.0. . ..0 00.A00’I0‘0. . 3A.... 0.30.0. 0. 0.0.00... 70.0 A... ..V-00.0 . .0. .A .A S. I 0. 0.0.0.0: V0 6.. A. .0. ‘- 00.. ...A 00 0 0.00. 0.00.0.0. I 00.90.5‘0. .00.]..0. ..t00000n. 00 015.000.. .0003 .0060. V V . . . CV 00. 0000 6 A. a . A ...}... A 0.60 A. . 0:! ...0'000000000000...‘0 . 0. . ..0-...”... 00: 0. ...A .. ..0 0.00... .0:A00 .0. ... 0 ... .A 0.0.00, 00.? . .0. A. I .0... ...00” ..0 .0.... . 0 ..A.. .. . V‘v‘0- ....0-1 .0. 0........‘00100.0¢|.2.00X.0I.00I ..0-.0021. . ... ... .V. 0 0N0- Av v0. ...-0 V. 00. .... ’ .0 00.0.00!‘ 0 ) ...0. . ... : .l... 0... 0.0000030. .0. .J :o .. .A .0.-10.0.0! 5.00160‘51 ...... .... 0.0.09 “0.0.0 VI.) V. A A .0. ..0. .V 000-0... 0. 0.1.0. 0....00A0.0.. 9000-0. .. . A .001 II. 00‘. 0.0.... .00.! A .. 0.: 0. 3.1.4.. V “#0: V. ..00 .A . 0 - . 0. . . 0 w .0 0A.... A A .0 A... : .0.-AA..0.00 ...-0000 .000’0‘. ...-.0. ..000. - ..‘000 0.0 ~ ’0 . .. . I. 0... . ... A. .00I0‘A00 0 0 V . . V .A. . V 30.0.0 00 .... A .0. 0 .000. 0‘0000?.00.0 0. .0....00. 3.03“ n V .106. 0000.30.03; ..0 ... . A8“. A. I 0.2.30 ...Iw A006,..‘. 0.0. . .00 . . . .0. 060.00 V... .V . . 00.0.0 I. V .130... .3 00.... .0. . . A 6 . A . 0.6 0!. 0 0. . V. .0 AA. 0V . I! . .3 ..0 V V. .0... J . V .I..V0IA!0A‘VVI.:. ...-0 00.0000. V . V V . V.. 0. 0V .00. 00.000. ....ooV 00 .3 :00. 0 0 6.06:0...01 0 00.... .A . \I 013.1“: .0.. . . .V .V 0...... .V :0... ... 0...A000.0..0I0-.00A- .... 1000. 0 ..0. I. 0.0... :90; V ... 00"». 0.0.00.0 .A i 0.. .500? 003‘ _. 0. .. .0. .-0 V. . 0A... . 1 . 00 .0 .0.?000 0 -.. .0... 0.0.00.- 0.:000.... .A ..V 300.1100 .00. V t. V ivocn .130 00.01 0.0.0 ..V 00.0 .300 0 . .. .. 0000...... .l6i-..A0:VI 00.00610... 016100.. . .0 .0000 . 0:0. .0: . 0V0 0 0 0 0 V .6 ..0 ..0-.300 0.0:..0.‘ 3.0 2.00: 3-0 0.0. V 0.. V .07} . ‘0 :0... ..V . . V 00.00 .6 ...A. 0 . .A 00000000; .0 OT! .0 . .V0 . 00h: . .A .0 A Z . A .0600! 0 A. A 0 ..0—... 2.9 ...0uI0-0o0: J0"... f..’?..&0000: A0. A... .03 .30. 1. 0.0.. ... 2.0 ...-200.05 V :3 .3‘6000. . . A V ... .0... 0.000.000. . . V V V ..0. .0.. 0.. V) . ...0. 00.00 00 . A ..00 .. ... . . V 00.. ..0. 0 .0000...0V..A0..0V0 0. .. 0... .0-A. . 0 AA. .0000 ..V . 0 30. .0. 0.. 00 . .. . V 0. . 4 . .?00. . I0..- . 0. 0‘ o0.1.0.0.-000. .0”.0u0..‘300 4.0.}00.I ,0 V 0| : ..0. -.06V6.V‘.0A00A .0.. ..‘6001. V _ 0.0u0. .00- 0 3.0.0. 0000 V 0 A... 0: .0 ‘00000000000. A0... a- ..0“? 0A 6A0 01 ... . .03 0.0 .0. . .0 . 0: 0- 000 0 00001‘1 000,60... .000... A -J :l. .. 00 . VV .. 00. .0 0 6 A v I 1.01 ... :00. .. . V . .0. . 0 .00 .0 . V ..V. . . V. . ...00 ..A ..00 ...... I! . 0.0.1. ..0 .A?.:V.J0 0.0. . . L 0V0 000 0.00 06.10A 5.0300 0.. .06 ”AV ....00 0000' ..V V6V’ .V.‘ 60000.. I.... ..0! ) 0.00530 0 : (I. . V00. I00. n3:..0.0.. .00.:0 .0. $V0AV0~0000V).0AAAIUA 01:900. .3609 0A . V .0.. V. V 010.- 9‘0. 2000A .10310-60 . . . ...: 00‘2”. 0.00.0 0. : AV .090. V VV0V.:.!04\ 0.: . . . VIA 01.00 V. . .VvV. . ..I0. I. .0 0 .. .l.0|0 .0.... ..V 0 V... . .0.. 0 0..., . .0 V. . V V 0... ...... 000 .00.. .00 ... 0. . A ..0. .‘. . .0 .. . 0.00. V“. 0 .0 .5306... A 0..“..00)\‘. ‘61! .V V .0.V.'0 .. V6 6.00.. V. ..0. I .0000 IIA A V . . ... I 0.. r3. . A 0. V.. V‘ 30.0 .00.... . .00....0 060.0 -.0...- ..0 . £00002?! A. . ' . . 000. A . ...VA . V 00. 0V .: 0 V . . . 0.. ... V. V 00.0. ..00600 .0 0. ..0. V00. 0...}. 9.03 V A . . . ‘0. V. . . .1 . V; .0. . 0’ 0......... V. V .6. X...‘ 0: ..V . V0A I.'|00 0A0 .. ..0... 0 3:6. 0. 0. .. .0: .0.... 00.4. .0. 3.. V ..0. V .000...V§0.r0\u9.: 0. .0...0A!V..: . 0: 0‘31... 0:0... ...-0 ...... A .. A0. 1 0- A . . 0"? V .....0 .. ‘0‘. 33..0.§t"¢‘9:. .4 .66 ..0: - .... we. ...0. ....V 0.: 33.0060. ...... t 0.. a 0... I... ...... .A AAA-0.1 . . 00.1: A30... .00 .0100. 0A.AA.V 0. 0 . .. .0.. V ... 0. .01 0 50.. I 0. A 0 .V . 0. 0. ... ....0 A .0..- . .A .9... . .3. V.. .1005... 2.6.0.00! “30.00-30.00. {#3103003 ..0. 6. V .0 V . 1. 0.10.0... A: V .. .v 0. n 0 0. 0 0 a 0 0 - u 00“.. A. .A 0. I. 0 .. A000. .. . .o . .00- . ..0”- 0...... ..0. ..Z:0....0.r~.!i’| V V . 06-90.030.310...- ‘V. 0A0. ’ieV’. 0V. 000. V _ :30)- A . :‘0000 0650.00.53 1.... . .0... V . .LL‘ .A:.- 00-- .00. ... 0... 0V. .V I. v .030. . . AI. 0 026.000 .30.... ..0A0000MV0.0..V A. “2.00 . AA. 0 .00 ..0 .0 . x A 0 0:. O . 0... AI .V Pl‘). ..0000.0000000v03.000000%0".n 000.0030, li— ..V 6.106.... V :0 r .. V. .IV0V:V0 7 A ...0. 0.0.6.... 0010. 90:30.... 0. .‘VVI. .. V. ‘30.. 0.0 .. 6. ...-0.20... n . 0 .VA .0 90.. 0 ..IAHV...0000.0.A.VA 0.0. . .0.. .V I. PVT-00.. .11 0 ..‘n0..:0 ...: I...A. .0 .0....0.0..00._.. ..0”..- .1000. A 0..V\VV L... 1...... - V ...-2 ..0-0.0".0rd! .0000...Ql010‘t¥.00 0 0 Al A00 .. A90. . :00. AA’IVS: 0000.0.6' 13 .0. 0 0 V VV ...0 u A. 10.00.. ..VA ItsV .. .. .0 8:0 V . . . . V. :90. .. 00’ 40- 0.. 803.12 0 A 000“. “flu-0r ‘9’ :00..“3I( . 00.4! ..T . V. I». . A. V2300...- 000 :2 0 . . ..0. 0010;... 0 V'00. ... 101...}... . . V V 0 .00 .....00’000.’.A..: ....V 00 V . ..0. . 10.... V .... ...-.0.. ...?! ..V 1.00.. .... 10.... . . 0‘ V2.00 .... 04.00:. .V V .00. 00.063015“? .0.-00‘0”“:0‘00Il "us-0300 .00 .0.... . I. 0 A V 0..-. V. .01. 0.6.60. 0...... V 0.A .0 V . V . . ..00. . (1000.21 .0.. ..V. . .70., .. ... . 0...... . . 0a.. .A. .0 3.00.... 0.0.! ..0-0.... I)... 00.. .~.0.A0 ...! 00.0. .‘AIVVI’HVAI’AV ......‘41000000 2.. _ 0.0. .05 V V .0. V006. . . .V ..0. V. .60 . ...0 .0.-A . V. 0490.0. 00. .0. ... ...V P .0.I. 000.1200...- ..0-7'13212100: .3. 0.. A 0. V V . .0 0. . V I. 0 0 V 0. . I0 I. 3...- 0.0 ...I. .. A. 000 0.: .0 . V A .A. A... . ..A. 0. . :00... 0 0 ..V- 6 ..0 ...V\0 .. . 0... ... .0. .V 0100 0...‘ V V ’0. 0.0 .00 v. 0. 0 0.7.0. 5‘... 0. I10 .000 '0‘ .60 A0 I 63 .. 0.0 . n0..V V 6. 0. ..0A .:00=0. .. 00000 00000 0.... '01.. A A v . . V V V . .. V0AA.A.0' A 18 A... A. . .0.. V V V 9: .A . J 0.1.0.111... ... . 0.: V ...-.00. - ....V 1 ... . V . V A, ... 0. 0. 610. 1 V .. A . .00. .02: . .. V 0.0 6.. . . . . . .' Ail . . . A V. 000. . . A .. ... V .... 0:. .0.0A 0.. 520.... .0...‘ 0 - 0. 0. ... V . .0). 0 0.1). IF“..PA.'JQ.0.... 0“ 0 ..0-0. V t V V ....0 . ..0. .V . . V \V .0. .. V . A0... . . V V . I0 .1901...” .."ovV ... . . . A .. 4 . . V! ..0. VV.A. .0. .0.:51‘ ..V! V... ..0 .0.-0.0:}.‘0. .0.-.0.. .V ’10.. V"...A:AI ...AA . SI ”6.! ... 0 V. 0... 0. . . 0 . .0. A I 6 A0 0- 170.0300. V00V00 .00 0. 0 .0. . ..0.V0q. .. 10.. .6 0.. . 06.. ..6 . 0:0... . .0: A0.0.... .V . . . . .0 ,0... 0 . ..0 ,0 ‘0 ... .A. 0VV..0V..>0A..0'0 .40000‘.-1 ..0 2,9300“ 0 04.00000}.- .L0.V v... . V . V . I I. ..v .0- v: A. A . . V... 11.060. V0 0. .0. 0| ... .760. I :0 A y .0”! 0.. 10.0.0... ..0 630.. 0-. ..0. V V 0 V .V .. V. 0A.. A .... .0...) .6 A (.1. 0. 001000-0073: ‘00. ...0. 6A.... 3.0.60.0 0.00 00: .0 .A . 33:! V . ..A- ..0 ..V I 0011-. 0.... : ..0 . 3. .V 0- .00..0VV 3...: .A. .0 ..0.. .V 1...! 06.0.12 A 0.0.0. ......000 6A. 3. V A V. . ..V . 0.!.. A3. 10...‘ a. ....5 ..0 01...... f 0-0015. 00.000 0. 0000000030000. .0fVJ-J :02: V V . ....I........0.... .A. . .0 V V .V . ...A.. ....I. .10. ..IV . AV... .. .0.. . . . ...Cnl..0;.0.000 AVA... . . . 00.0.. SVHI..- . 00.0. VA AI: 0.. .060. .0 A I -. V . 8.0:”..001 ‘30 ..0A: 0': V . 0 ‘06- 000‘0.00 0A .0 0. ...-0 . u . .0 0 0. 0. . 00-500 .0 .0 .0. 0.. 0 . 0 3:00 . V r A ..hVuizpitV? 6..» $0.10 5.0.0.1... A... . . . . o . . 00A... A . 0 . V... . {IT ..V ..0. AA“? ..0. 1.... V 1......60. .... 00 0..0000010.IVA..V.V.0..0A.\V.1 A V.: A0. ..0... .. . 0 . .V .. V. 0 . I. 1C0... . ..0; .. (.0 A 500.06 A. A. . . A0010.-. ... . . 0. .0... . .30.. V V ...... . ... .... 06. 0.00‘00 0.—.' .0.. ..‘Tn: 000” r V 0... I 0600. V . V: O V 0 .0 .. ...0 0 0.01.0: 0. A0. A0 ...? 0000000A00- A. 0A.- A‘. V 00 V. V. 0 0.00.0.7 .. V 0.0 . A .0.. \- ..0-.10 3A.. .AV 0V0» ' 3.051.. 0A. 1 . Q... 2... .046A. 60.0000 . .. 0.: V. 0... . . Iii-‘00.. 0.0000. 0.9 - .0.. 000.. 000V..700.0. 00 .0.. I .10 0. .A . A 0A. 0‘..V0V 0 ...-A. V. ... ..0.."0.. .0.: .V. .0 :::30..v . V . ..V... ..0 ..:l . V. . 26:... .1. ..I. ......10. . A. .0 ..0.... ..V 0.. .0000. ...:‘00... - :AA A 0 . 07.0.0.6 0 o . A .100. |.0I00..A0AAV .0.-“0.0.10IV’.N V0.0. . A 0 . VA . 0:. :00 . . ..0». A0 0 000 . ..0-0 .0.. 6 .0 00 0 . 001.90. L 3.0.700 AAVV..0 0.000. A .A . V... .\. 0 .... . . 0. . 0..V . ...... . .3 0... ...".00005 : .0304.A|A0A’ {N-mltwv‘0 : V0“. A V... Caz-0 . .AV 6. J .....V. . . I! A.V.V V . 1:...0’0... :.. ......V .... ... -.0: 0. .. ... ......0....0w. . . V V . .. .V .V V . . . . ..- ...: 000.060 313* 8.0.0.: .60. 0000...”? V ’f . I T... ..0 ...-..V. 0.00. ”0.000" .. A .n ..0 0 2.0000. .ADVAAA : V V . : ..‘.:‘A0. 4 001V... . N... ...v .. .0V:..0.V A... . .0.. 0. V .. .. . 0 . V... . . .V . ... . o!0(.:.\.. .... A....00 000V 0 0.... “)VAQI00.‘.‘0 ..‘I . ,VJV - Al 2. A0 ... ...... i.0AVVV nixAv.‘ . Jul .000 30..V .. 0. 0000...... .0. I . V V .016 . ..0 0 .A. ... 0.0! ...: 00. . 0. A: V. . ”00:10.31... . . 0 .... .0. . .6 . 0010...... .0 0 .0.: . 0... .. 31.000 . .V .. -0.... .0 . . . . . . ’wauW... 0. ..0.1.. .. A..,)0A0.VV 00V VVh0. ..0-(.007 .. 00.018006 Alt-VA. V. ...... ‘. 06...... 1 -. ‘0‘“ 0 .3000. 0.2.2. ...-V'; I........ V JVoAAVO-V .0.: 0. ..Al ..0 A... .A. (..V. V V ..0...A.<0A.VV0.. . . A V . A. .A V -...V 0. A0 0 V V .AA.....0. .. 0 V ... V ‘J.\0.!.0.0A’>. ..0. '0 V .0VA.. . A0: . A....0 V.» .V . ..0 . 0V .. 1 0.0 A ..0. 0 00 V. V ..1. V 03. 3.630.. :0 V ..V V . ...-00.... . 310.0 0. . V. V.. .0 . A .030. .005.) o .0“. .0. 00.0.. .0 0.0.100. 0...... V ‘00 0.0.00. A 6.0. ...-00V :6! .00.: 0W6.:.\.J€.0..V 000.0000 «.10 4:0 ..0V006V.0..0.)0.V.AI.00.. . . .V V... A.0 0. 6V 0.2.0.010}... V. . .. 3A . . .. : VAVV .. 60V .01. ._ A .V .0. 0. ...: .‘AIA._~..:V06 ..V \A .0 0.. .V 0 A0: 0. 0 '00.. 00 00......0.’ 200.0 ...0. ..uo...b_20n 0 .. - . . V. . V . . .V V V 0 . .. .VVVIV. . V 0 . ... .. V ’10., 6... 01¢ ...}..A ..V. 0 0-10:0 0 . ‘. ...: . - V. 020000000. .’0.00... . . V. . . 0.. .00... . ..000 ..0“... -A.. 0 .00 ...0. .. A9. A .0. V. .0‘ 11.2.3.0}... 1.00.1... . .03 V 300'!.. V . . .0 AV‘. .0. V0160....00.00...I V V . . V0..|.V.:I V ..0 . 0.0.00 0. ..V A 0.. .. 0 ... V V. .0 A... 0.0. 36.. 33...... V60. 60:. .. . .6A. : .0 A 00 AA.. 0.0.. .‘ .05.... 0 : A.. .V. ’36:? A V. .. . .. V I.... .-...0 . ..0. A: .. 0 V: .130 000.03.! 0.:1.’ 4-.z 1"?“ ..0 10.6..0VJ ‘0’..A‘A .‘Iév. “V - {..001 I . '.0 0.6: 0 2.0.000... ‘3. . . .. A 0 . . .0. . .V. 0. . . ......0-0 . ..0.AAV:.0 A000. .3006'61; ..Ar_00A‘I.‘V0000. 0". \0.00.V0000AI..'00.0060A0‘0 0 Qt ’ \ 0... 0000 0.! . . V A .00....- 00 00003.3. .00.. 0.. ...A0A‘, A 0:30... 00:30.60 0000 . 0. . 0 0. .0 A ..0 0 0 0n. 0 I . 0.000 A . . . .... 0.. A I 00... v. 7V0- .0 0V .V.:0 V0 ..00. u .. 0. - .-SAAI-QII. - Ar .3, . 10.01 00 300,06 V . . 00‘... . 6.0 0 V. 0.00.. .A00. 0 . 0 I... V . .0 . 00 ...0 .0 ‘ ...‘I. 7:. .AA. : .003. V . . I 0. 0. 0 . 100:0 ALI-200‘ A £00001. 0’. .0... )l‘:i.‘£‘000000§£ .I....A.. . 0. .9 1.0:! .V V . V 0 . A A .0A. .....0 .600 V. 0. ..0 I V A I... A :A: ... 4032...}? ..V. V........ 0 .V .00. 0 .V . ..000 . I. ..0 ._ ...0VV: .0... . , . L2A0";0‘.’A .V 00.0.... AA6VI..V’0.0.O. 6". #0. ..V . V V V- V......00VVIV..00.\I .30... .0 ...:TAVI: V. . 0.V $.AVA..V. .0. A.-. . ...VA A. .. 00 .. 0 . . .8 6-0-0. . 6) (A... ..0 I.:...V06A.600I6..:0 3.: 0.00.0... .I’. 0 .V . . V 0 A0. 0 .600. ...0 00.0.3l . .00.. .06. I300 VV00. :V C V. ..0. 0. l. ..00. V 1. 0V 0 ..0. .00. ..0. .l 0.0.0’..:A:0_ . .. :6. 0. . . A. . A..0Ao:0\ 0.00.10. 3.0..0\ 00000. V000. ...-IA \6..‘ I1 .0630. 10.3. 000 0.6: 00-00.; V :00, . 30:21:00: I. .0 A I ...v ..0 .00. A . 0V 0. . .v V A 0 . 0 0 «.V 0 . 0552!. AV“,. .000‘00‘00. 000000,; f-6t0lp‘0‘.’ V . .A. A 0.3.: V . ..0. V .. ._ V . V A .61 ..0... V I . A... .: _ 0. _ . r V. . 0 . . .V . .0 ...-0. ,6 V§ 0.- ‘0’2A00. IIAA Alf-001000.50. {A ..0. .: .. ..OAIIV—u 0.. 6.1.13... 0. ....00. . ..V 60. 9.. 30:00.. 0.1.. ..0.0. . Ii 0... .. .. .0 . .. 0 V .... . .... .005... I ..0: . .. 1‘ . 05:0:001013000‘0 A.A 1 0.00 .00\10.A 660 130.06 00.00000. 0000.. . . . V . . .0“ I. .0. .0 0.. . ..0. . .A .0....000 0 . _. .. . . . 0.0 c: :0 0.001., 00\ 0A 0.20 A. ..‘ ".60‘.iAll-JOV0‘I"~$ .03 .§....Al.-.V0.' C.0.o. .... . . V .A 00:. ..0 V... 0.. .30 .0. ......0. .0.-0000. VVVVV .:0 V0 0 ..0. 0 .0 00.0 A . .. .... . V . 0. .10 V. . _. .0. .. A V . ..0—“WA. .‘(fi‘00010fir'0'10zi'0t‘uvr‘ 0. 1-010... . 0- ..V V V . .0 V. . VA... .0. 0 ... .V v A... O. 0.10 A V. 0 0 . _ . V. ..0 ... A ..V . :A. V I 0. 3. D 0. ...:00 ......00.60 :06 0.00%310003-(2600109‘000‘AVI‘. : A A...‘ . 0.. 0.4.0 ..0. 0.3... .0. I. 0 ' ...IV .0 .05 . .0. .. ... AVA00 v 006 0 ... . . . A. .... 0. .00.... .- .Aluw. V500 .0.... 0 0 . 0 .0. .110; ... . . V . V .0 ‘.6.‘0_.0V30‘I0'A:70 0 ..IA AI.IA| V: ..0!“ A’.0‘A..0-"Ad-O. V 6.... .V V . . .V . . . .0.- ... r 1 .. . .0. .0.. . ..0 A. . . ..0. V. 0:10.. V. . 6" .617‘. 153.033.". . ‘0 660.50 05:6... 0.000.. 1. I I 300.10.000.00. .60" 6‘ . ... 0 .. .V . . AA. . 3.00.0..- 91A... ‘ .Aa . . . 0. 6 .VA ’13..“ . .A .o A. A0...!... A 0.30. ... . . ..0 ..0... V . V . . r0 0 0000000: 0 64!: .0.: ,A ‘03... of 0 .I‘ at. Alt-AA 0.00'.‘.#0 0 A .0 V 0 .A. 0 .:l ... .33... 1.000. . .A 0A. . 0. 0 ..0 - .0.-0.0 0.00 A . V .0 I . 0. 0 0 ..0 V .: V.... V 0 0 . 0 . . 0- .0 A00.|00 0.00 000-.000"} .. :A- ‘ Ag‘z’? III-Aztal.§ ..V 00000 .....000 V ..V A. 0 00. . ..00.:00.0.... :30 6.0. : 0 ...0 0.: . ..0 6 .00 .. A V 0. . 0 . .0 A 0 .V 00-. A. .V ..0 ....A. 0 .0 0.. “0.0,; AA. 2 010.700 AV010|0-.0 ‘5!¥\AO.:AO0I’100 0 00010.1 0.. 5.0 .10. V0: 2 V 7.0.1....6 . V. ..0 .0. . . . V . .. ...VAV V. .0 ..0. .J? 0 V . 0A.. ... 01003.00. 0. _ .0. . . V . 5.1.00 0%.... 0:. .0.. l0.§600i‘b:6000‘. .60-t‘oll. 3030...“..0 .00 . l . .31.?) A0 :0 . V 0‘...- 0 3 0... 00 .. ... . ..0 : 3'0. .. 609...... ..0 A I .l. .0 . 0 V .0 VV ..0.. . V. ...V .. V .6 1‘ . .V 0. . . ....V .. .0.... 0 . . . .. A .AA.V.A..-0V‘V ... \ "6 00.. ‘30:..0 :‘ISAt‘AJA I0 600000050’0: V 0. r0 . V V . V .. All“. .. . . . o A.. .3 V. 00 . a... . .V 00. v . V 1 -.. . V V . o A. V. . ..V . 9.. . . 0 . A V . . V V V V V . . . V . .V V ...}... ‘06 0‘ .A‘IV-t.’)0:vll..2t I} :00. 11.6,?st 5: $0.". . 09... w‘...‘ . V0. .. 0 00 01V. 03‘039. ..0. 0. .0... A. A00 2... .0 V1.5. 000 .. .00. $00.06. AV0.0 ,. . V. 0 . .1 .30.. ... ..0. . .0A I. ... . V00- ..0 V .. V .0 0 v . 0 1- V V . . V V .. . V A V A. 0 0.001” 0.0.1 .0 .V0. 60.0! :0 00-3 AQJI\\0.A..IV AI 06.0.? ‘00: II A 6. I - .. 0. .0 a I.“ .... 79.10.00...030.:ud..0.00...0w01...AVA 0!..3 .00:.A 0000 ..... . ..0. 00. . vA.0A:00. 0. 00. . 0V .V V . . 2 .0 . . 0.. 000 :o. . V 0 . . 0. . . . 0. .. - u .. . .o _\ ..V-Vs .. .V . -'..60A.0.:.( ‘0‘"...z- 0612.]?! I b— .90 .00... ... .0.. V .0.... A . 0 . A. .0.. . 0; .V ... A: V. . A .. 0. . .0. A... V ... . A 0.0 . 0 . .. 0...: 0..:A .V V VV D‘A.0\-‘. .- .0.:A 6.0.A‘: IIO.\0..A . 0 .500‘0110 .0.. .... 010-10. A0... .V‘ I2 :A.:.|..00.0 A .0. V. 0V 0 .V 31.... 0‘. ..0 A; . .10... .V .0. V 90.... . . 2... . .. ... V. 0.. ... . .0 ..0 . . 0 2-,. .. V. V 10.0- . VV V.. . .V .V [‘90) {ma-Alzl :.I.VI-.I...00 : VA 0-1- A. .0... 00V. ..0. . V : 00010.... 000 : 0.. .00: A00 . 006 0. . 0. V 0 I J' 0 V . ~ 0 .. . . v V V V. .V 0 6 V .. A. ..0A . A ..0-0. . . .. )0.- V . V. 0 A 3‘036 f0 .00‘A . 6.000,.6 31010.0. .0. 030 .0... .0002 .I 0. 00.0 .I IA. 0. .0. .. . 0 ..0‘0 ... 00V 0 60 ..l ...;0 . 0A. ._ ..0. . .0 00 . . 0-. A: . 0. .000 - . .0 . n A .0 .. . . ...: . 0.0.000. 0. 0 v . V :0 .A 0 ..0: \\0.A. .00. t. 00.: ’AI . . J; V 0A ...-As? I: 11."; . .VV 0 . .. A V A .I 0. . . 6. A V .. . A .—V V. . . [A v. .A . . ..A. .V A ... 4.50 . ‘ V V... .6- 0 AII 'o: ..0-nil .IVIIIIV .0.. 0 ._ 6 V0. . V ..0. '00 0.. .- A O .A. . . .A. .1010... .V .0 0. V :0 AI .. A . .0V_ .0 .00.. 0 . . . I .. . . V . A V . A 0 V 0 V . ..QI .- A I“ I A 6 V: I ’II 0 3.1.0000. ..V... ‘ .VV. .0 ’ A. .‘A0 . .00.. 0 .I IV 0.. ... 0 V 0 A ‘ . . . . .0 A O . V _ A 0 i 0 . .. A2000: 7 A . - .V00 .. A. A . 0 . ... . A ’6 ._ A v - 0’0 0.060‘690AV I V 3 0 A. 6 . V . VV Isrll. I. .0 . VI. 60 1V . V 50.. V. A. ‘6. ' A U.- . 6A‘A A0. A A V. . . 0 0...: 0.0 . A :- A .0 ..0 0 . u - 0 v A 0 I I.“ . 0...-.. :‘V 1 . .0 I .... .6) ...-A0 0.01:1 : .0 .V A .. . I . V. 0. A . : . 0 IV. :7 . I. .- ... 0 a. . 10.07 .AA 0 0 .n . ’- . A V 0.0, A6 . A v .0 .A ... 0 0 n AA.AA .0: . 0 A - 0 . .0.. 60000 00 v' . :0 Q 0090 0..0 A . A .. V . V' I I0 0.. 0 . ‘9' 0 .0.. . :0. .0: A A A 0 . A . A . ...)’.V.I.A ..A ..0-0 0. A .— I00 0 . 0. A 3 A. 0 A A:6.0V00 0' A 3 : .J 06 ... A v . .. 000 . 0 v0 . 3. 0 0 0 : . ... . I I 0 - 0... . ._ . .A .0 A . . . - . u. I- }? Hg 2-. - 000.0 rm... 4.0....VbIVflV {mgmé an“. -11 V J rim aria... V a. 4.. Fatwa? 4.... Rafi-damn, . .0 34.0%.. 0.03m: 0% xiv-1L. a” 43 W... ......0... 0.,W0lm4..30.. . J5.0.005.“ +62%.me It) I”0 AWV¢VMVhYAZIA ”.0000 §. .00 3m»: K“ .: Au 3‘0”. 0 ;000#00Hwn.3. ”04.9.0 4.0030910. 0, # Almdywwl‘ .Hu 0'1»): 3.100.001. 33.00:} a: 0. 0000...).AH0 10uu0lfi0hfi’unmmi . .00V000...m00.. 5.0.0.6.“! ..uVV.VV.:\V.J.SWuVJum.~..1 :55»; ...-04 ”0...“...W «..0. . V 0 a .V. .....Rmuwphsfiw/ ..H..~..3.JVAAJ.13 .- .AVV .. . ....- V: (“hp—00:10 50:0":00 gNtQ-OQ” 00.‘ Luca .._.0h. 0... 0. .0‘7A0 066 60T- . 0 V. 0.. LI:6....0.. ..:0 $0.02.. 1 LIBRARY Michigan State University This is to certify that the dissertation entitled SOURCE CHARACTERIZATION, EVALUATION, AND TREATMENT POTENTIAL OF AGRICULTURAL FILTER STRIPS presented by Rebecca Anne Larson has been accepted towards fulfillment of the requirements for the Doctoral degree in Biosystems Engineering Jami. M/w- Major Professor's Signature Amy/52‘ 261 20/0 Date MSU is an Affirmative Action/Equal Opportunity Employer A -4_l-.A__A ’ ..-._._._.-.-.-._.-.-.-.-.-.-.-.-.-.-..- — - I PLACE IN RETURN BOX to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 5/08 K:lProjIAoc&Pres/CIRC/DateDue.indd SOURCE CHARACTERIZATION, EVALUATION, AND TREATMENT POTENTIAL OF AGRICULTURAL FILTER STRIPS By Rebecca Anne Larson A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Biosystems Engineering 2010 ABSTRACT SOURCE CHARACTERIZATION, EVALUATION, AND TREATMENT POTENTIAL OF AGRICULTURAL FILTER STRIPS By Rebecca Anne Larson Diffuse source pollution produced by runoff from animal feeding operations contains high concentrations of pollutants that pose serious risks to surface and ground water. Agricultural filter strips are an economical treatment option for farmstead runoff but have not been investigated as to the water quality after infiltration into soil subsurfaces. A runoff source Characterization provided water quality data for four farmstead runoff pollutant sources: animal manure, general use impervious areas, upright feed storage, and bunker feed storage. Results from these sources, including two composite samples, indicated that animal manure produced the greatest pollutant concentrations, but due to the small footprint, the feed sources were of more concern for maintenance practices. It was concluded that quantity and dilution of runoff were the most significant factors in determining the impact from pollutant sources. Three field-scale agricultural filter strips were investigated to determine pollutant removal percentages for typical operation at a small and medium sized dairy. Ten sampling events at the MSU dairy, a 160 cow dairy with a 2.42 acre drainage area, were analyzed for surface and subsurface runoff quality on two adjacent filter strips. A third filter strip was located on a small 40 cow Michigan dairy which had a drainage area of approximately 0.5 acres. The small Michigan dairy had greater removal percentages for the majority of the 17 water quality parameters, resulting from the addition of a bioretention basin, decreased loadings, and sand soils. Thirty soil columns were investigated for treatment depth, soil type, and submergence. Columns received synthetic wastewater applications two times per week, and effluent was analyzed for 11 water quality parameters. Columns with a depth of 30 inches or greater produced effluent concentrations that did not pose groundwater concerns for most water quality parameters. Concentrations of BOD5 were typically below 6 mglL for columns greater than 12 inches. Nitrate concentrations were greater than 10 mglL for all columns and posed potential to impede implementation of this technology if they cannot be reduced. Sand soils provided soil characteristics that increased pollutant removal as compared to soil columns with a sandy loam soil. Soil type and depth of treatment were determined to be Significant factors in column performance. Treatment for the field-scale and laboratory soil columns followed the same trends, although field treatment percentages were reduced. The small Michigan dairy filter strip had similar removal to the laboratory study as compared to the sandy loam columns and the MSU dairy filter strips, showing greater continuity in performance of sand soil subsurfaces due to increased porosity and a decrease in soil moisture holding capacity. ACKNOWLEDGEMENTS I would like to thank everyone for their continued support throughout this process, and in particular: Dr. Steve Safferman, Advisor Dr. Timothy Ham'gan, Dr. Brian Teppen, and Dr. Dawn Reinhold, Committee Members The MSU Biosystems Engineering Department Faculty and Staff Barb Delong Steve Marquie and Jacob Koch Phil Hill and Lou Faivor The MSU Dairy Teaching and Research Facility Bob Kreft and Rob West MSU Land Management Ben Darting The MDNRE State of Michigan Environmental Laboratory Joe Rathbun NCR-SARE Undergraduate Research Assistants Michael Holly and Adrienne Vamey And of course, my family and friends -Becky TABLE OF CONTENTS LIST OF TABLES ............................................................................................... viii LIST OF FIGURES ............................................................................................. xiv CHAPTER 1: INTRODUCTION ............................................................................. 1 1.1 Objectives ................................................................................................ 4 1.1.1 Runoff Characterization ........................................................................ 5 1.1.2 Analysis of Field Treatment Systems .................................................... 5 1.1.3 Laboratory Evaluation of Treatment System Design Components ........ 6 CHAPTER 2: LITERATURE REVIEW ................................................................... 7 2.1 Runoff Characterization ............................................................................... 7 2.2 Analysis of Field Treatment Systems .......................................................... 7 2.3 Laboratory Evaluation of Treatment System Design Components ............ 13 CHAPTER 3: METHODS AND MATERIALS ...................................................... 18 3.1 Runoff Characterization ............................................................................. 18 3.1.1 Sample Collection Methods ................................................................ 19 3.1.2 Laboratory Analysis ............................................................................. 19 3.1.3 Precipitation Data ................................................................................ 21 3.1.4 Data Analysis ...................................................................................... 21 3.2 Analysis of Field Treatment System .......................................................... 22 3.2.1 Sample Collection Methods ................................................................ 25 3.2.2 Laboratory Analysis ............................................................................. 27 3.2.3 Precipitation Data ................................................................................ 28 3.2.4 Data Analysis ...................................................................................... 28 3.3 Laboratory Evaluation of Treatment System Design Components ............ 28 3.3.1 Soil Column Experimental Design ....................................................... 29 3.3.2 Soil Column Structural Design ............................................................ 32 3.3.3 Waste Water Composition .................................................................. 33 3.3.4 Soil Column Operation ........................................................................ 35 3.3.5 Water Quality Parameters ................................................................... 36 3.3.6 Soil Column Deconstruction ................................................................ 37 V 3.3.7 Data Analysis ...................................................................................... 38 CHAPTER 4: RESULTS AND DISCUSSION ...................................................... 39 4.1 Runoff Characterization ............................................................................. 39 4.1.1. Summary and Management Strategy ................................................ 49 4.2 Analysis of Field Treatment Systems ........................................................ 50 4.2.1 BOD .................................................................................................... 51 4.2.2 COD .................................................................................................... 52 4.2.3 Nitrogen .............................................................................................. 55 4.2.4 Phosphorus ......................................................................................... 60 4.2.5 Solids .................................................................................................. 61 4.2.6 Alkalinity/pH ........................................................................................ 65 4.2.7 Metals ................................................................................................. 67 4.2.8 TOC .................................................................................................... 70 4.2.9 Chloride ............................................................................................... 71 4.2.10 Conductivity ...................................................................................... 71 4.2.11 Cold Weather Performance ............................................................... 72 4.3 Laboratory Evaluation of Treatment System Design Components ............ 73 4.3.1 BOD .................................................................................................... 78 4.3.2 COD .................................................................................................... 82 4.3.3 Nitrogen .............................................................................................. 89 4.3.4 Phosphorus ....................................................................................... 104 4.3.5 pH/Alkalinity ...................................................................................... 105 4.3.6 Metals ............................................................................................... 109 4.3.7 Plant Tissue ...................................................................................... 116 4.4 Comparison of Treatment from Field and Laboratory Studies ................. 118 CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS ........................... 123 APPENDIX A ..................................................................................................... 133 APPENDIX B ..................................................................................................... 136 APPENDIX C .................................................................................................... 137 APPENDIX D .................................................................................................... 143 vi APPENDIX E ..................................................................................................... 168 REFERENCES .................................................................................................. 171 vii LIST OF TABLES Table 1: Runoff water quality parameters for dairy and beef feedlots ................... 1 Table 2: Water Quality Parameters for Source Characterization and Field Treatment Systems ............................................................................................. 20 Table 3: Soil Column Treatment Assignment ..................................................... 32 Table 4: Nutrient Solution Constituent Concentrations for Synthetic Wastewater ............................................................................................................................ 35 Table 5: Water Quality Analysis Parameters and Methods For Soil Columns ..... 37 Table 6: Feedlot Runoff Water Quality Parameter Average Concentrations ....... 40 Table 7: BODs Percent Removal - MSU Dairy Filter Strip .................................. 51 Table 8: BOD5 Percent Removal - Small MI Dairy Filter Strip ........................... 51 Table 9: COD Percent Removal - MSU Dairy Filter Strip ................................... 53 Table 10: COD Percent Removal - Small MI Dairy Filter Strip .......................... 53 Table 11: Soluble COD Percent Removal - MSU Dairy Filter Strip .................... 55 Table 12: Soluble COD Percent Removal — Small Ml Dairy Filter Strip ............. 55 Table 13: TKN Percent Removal - MSU Dairy Filter Strip .................................. 56 Table 14: TKN Percent Removal — Small MI Dairy Filter Strip ........................... 56 Table 15: Ammonia Percent Removal - MSU Dairy Filter Strip .......................... 58 Table 16: Ammonia Percent Removal - Small Ml Dairy Filter Strip ................... 58 Table 17: Nitrite Percent Removal - MSU Dairy Filter Strip ................................ 59 Table 18: Nitrite Percent Removal — Small MI Dairy Filter Strip ......................... 59 viii Table 19: Nitrate Percent Removal - MSU Dairy Filter Strip .............................. 60 Table 20: Nitrate Percent Removal - Small Ml Dairy Filter Strip ........................ 60 Table 21: Phosphorus Percent Removal - MSU Dairy Filter Strip ...................... 61 Table 22: Phosphorus Percent Removal - Small MI Dairy Filter Strip ............... 61 Table 23: TS Percent Removal - MSU Dairy Filter Strip .................................... 62 Table 24: TS Percent Removal — Small Ml Dairy Filter Strip .............................. 62 Table 25: VS Percent Removal - MSU Dairy Filter Strip ..................................... 63 Table 26: VS Percent Removal — Small Ml Dairy Filter Strip .............................. 63 Table 27: T88 Percent Removal - MSU Dairy Filter Strip .................................. 64 Table 28: T88 Percent Removal - Small Ml Dairy Filter Strip ........................... 64 Table 29: V88 Percent Removal - MSU Dairy Filter Strip ................................... 64 Table 30: V88 Percent Removal - Small MI Dairy Filter Strip ............................ 64 Table 31: Alkalinity Percent Removal - MSU Dairy Filter Strip ............................ 65 Table 32: Alkalinity Percent Removal - Small Ml Dairy Filter Strip ..................... 66 Table 33: pH Percent Removal - MSU Dairy Filter Strip ..................................... 66 Table 34: pH Percent Removal — Small Ml Dairy Filter Strip .............................. 67 Table 35: Mn Percent Removal - MSU Dairy Filter Strip .................................... 67 Table 36: Mn Percent Removal — Small MI Dairy Filter Strip ............................. 68 Table 37: Fe Percent Removal - MSU Dairy Filter Strip ..................................... 68 Table 38: Fe Percent Removal - Small Ml Dairy Filter Strip .............................. 69 Table 39: AS Percent Removal - MSU Dairy Filter Strip ...................................... 69 Table 40: As Percent Removal — Small Ml Dairy Filter Strip ............................... 70 Table 41: TOC Percent Removal - MSU Dairy Filter Strip .................................. 70 Table 42: TOC Percent Removal — Small MI Dairy Filter Strip ............................ 70 Table 43: Chloride Percent Removal - MSU Dairy Filter Strip ............................. 71 Table 44: Chloride Percent Removal — Small MI Dairy Filter Strip ...................... 71 Table 45: Conductivity Percent Removal - MSU Dairy Filter Strip ...................... 72 Table 46: Conductivity Percent Removal - Small MI Dairy Filter Strip ............... 72 Table 47: Sand and Sandy Loam Soil Column Characteristics ........................... 76 Table 48: Soil Column Volume and Porosity ....................................................... 77 Table 49: Soil Column BOD Statistical Model .................................................... 80 Table 50: Soil Column BOD Differences of Least Squares Means - Comparisons of Significance ..................................................................................................... 81 Table 51: Soil Column COD Statistical Model .................................................... 84 Table 52: Soil Column COD Differences of Least Squares Means - Comparisons of Significance ..................................................................................................... 86 Table 53: Soil Column TKN Statistical Model ..................................................... 91 Table 54: Soil Column TKN Differences of Least Squares Means - Comparisons of Significance ..................................................................................................... 93 Table 55: Soil Column Ammonia Differences of Least Squares Means - Comparisons of Significance ............................................................................... 98 Table 56: Soil Column Nitrite Statistical Model ................................................ 100 Table 57: Soil Column Nitrite Differences of Least Squares Means - Comparisons of Significance ............................................................................. 101 Table 58: Soil Column Nitrate Statistical Model ............................................... 103 Table 59: Soil Column Nitrate Differences of Least Squares Means — Comparison of Significance .............................................................................. 104 Table 60: Soil Column Alkalinity Statistical Model ............................................ 108 Table 61: Soil Column Alkalinity Differences of Least Squares Means - Comparisons of Significance ............................................................................. 109 Table 62: Soil Column Mn Statistical Model ..................................................... 112 Table 63: Soil Column Mn Differences of Least Squares Means - Comparisons of Significance ................................................................................................... 112 Table 64: Soil Column Fe Statistical Model ...................................................... 115 Table 65: Soil Column Fe Differences of Least Squares Means - Comparisons of Significance ....................................................................................................... 116 Table 66: Plant Tissue Concentrations by Soil Column .................................... 116 Table 67: Removal Percentages (%) for Filter Strips in Comparison to Soil Columns ............................................................................................................ 1 19 Table 68: Filter Strip Soil Characteristics .......................................................... 121 Table 69: QA/QC ............................................................................................... 136 Table 70: MSU Dairy Alkalinity Data ................................................................. 143 Table 71: MSU Dairy Ammonia Data ................................................................ 144 Table 72: MSU Dairy COD Data ....................................................................... 145 Table 73: MSU Dairy Soluble COD Data .......................................................... 146 Table 74: MSU Dairy TS Data ........................................................................... 147 xi Table 75: Table 76: Table 77: Table 78: Table 79: Table 80: Table 81: Table 82: Table 83: Table 84: Table 85: Table 86: Table 87: Table 88: Table 89: Table 90: Table 91: Table 92: Table 93: Table 94: MSU Dairy TSS Data ........................................................................ 148 MSU Dairy VS Data .......................................................................... 149 MSU VSS Alkalinity Data .................................................................. 150 MSU Dairy Nitrite Data ...................................................................... 151 MSU Dairy Nitrate Data ..................................................................... 152 MSU Dairy pH Data ........................................................................... 153 MSU Dairy Phosphorus Data ............................................................ 154 MSU Dairy Mn Data .......................................................................... 155 MSU Dairy Fe Data ........................................................................... 156 MSU Dairy TOC Data ........................................................................ 157 MSU Dairy Conductance Data .......................................................... 158 MSU Dairy Cl Data ............................................................................ 159 MSU Dairy Arsenic Data ................................................................... 160 Small Ml Dairy TOC Data .................................................................. 160 Small Ml Dairy Mn Data .................................................................... 161 Small MI Dairy Fe Data ..................................................................... 161 Small MI Dairy Conductance Data .................................................... 161 Small Ml Dairy CI Data ...................................................................... 162 Small Ml Dairy As Data ..................................................................... 162 Small MI Dairy 8005 Data ................................................................ 162 xii Table 95: Small Ml Dairy Alkalinity Data ........................................................... 163 Table 96: Small Ml Dairy COD Data ................................................................. 163 Table 97: Small MI Dairy TKN Data .................................................................. 163 Table 98: Small Ml Dairy Ammonia Data .......................................................... 164 Table 99: Small MI Dairy Nitrate Data ............................................................... 164 Table 100: Small MI Dairy Nitrite Data .............................................................. 164 Table 101: Small Ml Dairy pH Data ................................................................... 165 Table 102: Small Ml Dairy Phosphorus Data .................................................... 165 Table 103: Small Ml Dairy Soluble COD ........................................................... 165 Table 104: Small Ml Dairy TS Data ................................................................... 166 Table 105: Small MI Dairy VS Data ................................................................... 166 Table 106: Small MI Dairy TSS Data ................................................................ 166 Table 107: Small Ml Dairy VSS Data ................................................................ 167 xiii LIST OF FIGURES Figure 1: MSU Dairy Concrete Storage and Sedimentation Basins ................... 23 Figure 2: Filter Strip lnfluent Flow Dispersion ..................................................... 24 Figure 3: MSU Dairy Filter Strip Sampling Locations ......................................... 26 Figure 4: Small Ml Dairy Filter Strip Sampling Locations ................................... 27 Figure 5: Filter Strip Power Analysis .................................................................. 31 Figure 6: Soil Column Construction ..................................................................... 33 Figure 7: Average COD Concentrations .............................................................. 42 Figure 8: Average Phosphorus Concentrations. ................................................. 43 Figure 9: Average Total Solids Concentrations. .................................................. 44 Figure 10: Average Alkalinity Concentrations. .................................................... 45 Figure 11: Average pH Concentrations. .............................................................. 46 Figure 12: Average Ammonia Concentrations. ................................................... 47 Figure 13: Average TKN Concentrations. .......................................................... 48 Figure 14: COD Effluent Concentrations as a Function of Influent COD Concentrations .................................................................................................... 54 Figure 15: Effluent TKN Concentrations as a Function of lnfluent TKN Concentrations .................................................................................................... 57 Figure 16: Soil Column Flow Rates - all 12 inch and 30 inch sandy loam columns ............................................................................................................... 74 Figure 17: Soil Columns Flow Rates — 30 inch sand and all 48 inch columns ....75 Figure 18: Soil Column B005 Concentrations ..................................................... 79 xiv Figure 19: Figure 20: Figure 21: Figure 22: Figure 23: Figure 24: Figure 25: Figure 26: Figure 27: Figure 28: Figure 29: Figure 30: Figure 31: Figure 32: Figure 33: Figure 34: Figure 35: Figure 36: Figure 37: Figure 38: Sand Soil Column COD Concentrations. ........................................... 83 Sandy Loam Soil Column COD Concentrations ................................. 84 COD Effluent Concentrations as a Function of Soil Column Depth....88 COD Effluent Concentrations as a Function of Soil Column Depth....89 Sand Soil Column TKN Concentrations. ............................................ 90 Sandy Loam Soil Column TKN Concentrations. ................................ 91 Sandy Soil Column Ammonia Concentrations. .................................. 94 Sandy Loam Soil Column Ammonia Concentrations. ........................ 95 Sand Soil Column Nitrite Concentrations ........................................... 99 Sandy Loam Soil Column Nitrite Concentrations. ............................ 100 Sand Soil Column Nitrate Concentrations ........................................ 102 Sandy Loam Soil Column Nitrate Concentrations. ........................... 103 Sand Soil Column pH Concentrations .............................................. 105 Sandy Loam Soil Column pH Concentrations. ................................. 106 Sand Soil Column Alkalinity Concentrations. ................................... 107 Sandy Loam Soil Column Alkalinity Concentrations. ....................... 108 Sand Soil Column Mn Concentrations. ............................................ 110 Sandy Loam Soil Column Mn Concentrations .................................. 111 MnSoil Concentration as a Function of Depth .................................. 113 Sand Soil Column Fe Concentrations. ............................................. 114 Figure 39: Sandy Loam Soil Column Fe Concentrations. ................................. 115 Figure 40: Mn Plant Tissue Concentrations by Soil Column ............................. 117 Figure 41: Fe plant Tissue by Soil Column ....................................................... 118 CHAPTER 1: INTRODUCTION Diffuse source pollution from animal feeding operations has the potential to contaminate ground and surface water. Animal manure, feed, and other animal farmstead additives and wastes are susceptible to transport during a precipitation or thaw event, resulting in non-point source pollution (US EPA 2003). Feedlot wastes have the potential to contaminate surface water due to runoff from impermeable surfaces or saturated soils and aquifer contamination due to leaching through permeable soils (Burkholder et al. 2007). Water quality from these types of operations has been examined as early as the 1970’s. Literature shows feedlot runoff contains high oxygen demanding wastes, elevated nutrient concentrations, organic material, sediment, salts, viruses, bacteria, and other microorganisms (US EPA 1993). Reported pollutant concentrations (Table 1) reveal that runoff from beef cattle feedlots exceed those originating from dairy operations. However, both pose potential for contamination if not properly handled, treated and disposed. Table 1: Runoff water quality parameters for dairy and beef feedlots Parameter Concentration (mglL) TKN 20-180 300 30-400 1 122 n/a n/a N Na nla nla n/a 2640 580 TS 500-7100 3700 2800-8400 12777 57800 1 1230 COD 130-14000 4220 600-5000 14288 79600 7850 P 240 64.1 20-50 n/a 770 120 Farm Type Dairy Dairy Dairy Beef Beef Beef Larson Dickey and Dickey and Edwards Clark et Reference (2009) Vanderholm Nye (1982) Vanderholm et al. al. (1981) (1981) (1983) (1975) Human and environmental health concerns result from improper handling and treatment of these waste streams (Burkholder et al. 2007). Of particular concern 1 is nitrogen and pathogens. Additionally, runoff containing solids, oxygen demanding waste, and excess nutrients contribute to anoxic conditions in waterways and impact aquatic communities and habitats (Burkholder et al. 2007). Excess nutrient concentrations have been reported as a cause of environmental concern throughout the world. Nitrate contamination from various wastewaters, which sources include animal production facilities, have been measured at elevated concentrations in numerous countries including the United States (Kirby et al. 2003). Assessment has shown that agricultural sources are a leading source of impaired waterways (US EPA 2004) and diffuse agricultural phosphorus sources are a leading contributor to this water pollution (Parry 1998). Further, barnyard and animal storage areas are the leading source of runoff containing phosphorus among agricultural operations (Hively et al. 2005). Eutrophication of waterways can occur with only small additions to phosphorus concentrations (Hart et al. 2004). Excess phosphorus concentrations result in algal blooms and decreased oxygen as It is typically the limiting nutrient for the processes producing these effects (Anderson et al. 2002). Decreased oxygen concentrations in waterways leads to fish kills and habitat destruction (Burkholder et al. 2007; Anderson et al. 2002). Metals within runoff and in the leachate from soils subject to land application have reached surface and groundwater. The release of dissolved Fe2+ into groundwater is one of the most prevalent groundwater problems worldwide (Lovley 1991). Agricultural practices have had direct effects on the concentration of N033 SO43 Cl', P, C, and As within groundwater (Bohlke 2002). Twelve percent of groundwater wells in Michigan (of a sampling of 73 wells in 1997) indicated arsenic contamination greater than the US EPA maximum contaminant level of 50 ug/L (Kim et al. 2002). Management practices for runoff can be costly and if not properly designed, installed, or operated are ineffective in reducing environmental concerns. Current treatment options include land application, conventional wastewater treatment, and runoff infiltration designs. Land application has limitations dependent upon available field area based on accepted agronomic application rates. Conventional treatment requires extensive capital and Operational costs that are not economically feasible for many animal operations. Vegetative filter strips are being investigated as an economically feasible management option for treating farmstead area runoff. Agricultural vegetative filter strips are engineered treatment systems which direct flow over a vegetated soil. The vegetation and design reduces runoff contaminant concentrations by increasing sheet flow thereby increasing sedimentation and infiltration. Biological, physical, and chemical processes within the infiltration zone are the principal mechanisms to effectively reduce the loadings and improve water quality prior to reaching groundwater (Koelsch et al. 2006). Surface water quantity issues are addressed as increased infiltration restricts runoff flow to surface water. Vegetated filter strips are proven to reduce the pollutant concentrations from feedlot runoff. However, literature has shown variability in treatment performance of the various pollutants (Koelsch et al. 2006). Detailed studies have examined the trapping effectiveness and surface outflow to determine the ability of filter strips to eliminate surface water discharge (details of these studies can be found in the literature review). However, research has not provided definitive results as to the effectiveness of filter strips to reduce contaminant loads prior to leachate reaching groundwater. To protect against groundwater contamination, continued research is critical to determine the comprehensive pollutant removal capacity of vegetated agricultural filter strips. Filter strips are currently functioning in the absence of supporting data on the fate of contaminants once they have infiltrated into the soil subsurface. Determining pollutant removal capacity of these engineered systems is critical to formulate design standards that can maintain a sustainable water cycle. 1.1 Objectives Assessment of the treatment potential and evaluation of design recommendations for agricultural filter strips was organized into three critical research elements. Initially, preliminary research focused on characterizing on- farrn runoff sources for quantity and quality concerns. Secondly, analysis of the field treatment processes to determine pollutant removal processes, and finally a laboratory evaluation designed to identify critical issues associated with design depth. Below is a detailed explanation of the objectives for each of the three research elements. 1 .1.1 Runoff Characterization Collect and analyze precipitation data to determine the source characteristics from a representative dairy farm runoff. Included is the evaluation of the quantity and quality impacts from the heat check lot, upright silos, bunker silos, and general impervious roadway areas used for transport and mixing of farmstead operational inputs and outputs. Based on the characterization, recommend on- fann management practices to reduce the pollutant quantity and increase water quality. 1.1.2 Analysis of Field Treatment Systems Determine the pollutant removal of agricultural filter strips in typical environmental and farmstead conditions. Specific objectives include the following: . Assess the surface and subsurface water quality at two field sites. 0 Assess current practice standards in regards to operation and maintenance procedures. 0 Determine if agricultural filter strips are an effective agricultural treatment/management option as designed, with a particular emphasis on metal leaching into groundwater. 5 . Determine treatment consistency throughout season and rainfall events. 1.1.3 Laboratory Evaluation of Treatment System Design Components Conduct soil column experimentation to assess the required soil depth to achieve adequate treatment of land applied agricultural runoff prior to infiltration to groundwater. o Statistically determine the pollutant removal capacity of a volume of the overall soil column system for the various water quality parameters. . Determine impact of soil depth and total soil volume to pollutant removal. 0 Examine the influence of groundwater capillary rise on the depth of soil required for treatment of agricultural runoff. . Find the degree of treatment variance between two defined soil types, sand and sandy loam, to determine if further detailed analysis for soil type is warranted. CHAPTER 2: LITERATURE REVIEW Previous work done on the three research areas provided a basis for experimental design. The following literature review sections contain the details from these studies relevant to the proposed research. 2.1 Runoff Characterization Pre-treatrnent of waste is a critical for agricultural filter strips (Koelsch et al. 2006). An effective management plan can reduce runoff concentrations, reducing the environmental effects and the loading to treatment systems through source reduction, reduction in transport mechanisms, and/or removal/degradation of pollutants prior to reaching waterways (Azevedo, 1974; Sweeten, 1998; US EPA, 2003). Practices include covering pollutant sources prior to precipitation, sweeping impervious surfaces, and/or maintaining faces on feed bunkers. Although previous research has examined water quality data of agricultural feedlots as a whole, there has been no source investigation. Identifying contaminant sources and strength is critical in developing an effective management plan. 2.2 Analysis of Field Treatment Systems Engineered filter strips have two main mechanisms for pollutant removal, sediment trapping and infiltration treatment processes. Sediment trapping is a result of vegetation and sheet flow, which reduces flow velocities and captures sediments and sediment bound pollutants. Sediment bound pollutants have greater removal rates than dissolved or soluble contaminants due to higher trapping efficiencies (Goel et al. 2004; Schmitt et al. 1999). However, infiltration is responsible for the majority of pollutant removal, in particular dissolved contaminants (Dosskey et al. 2007, Lee et al. 2003). Infiltration allows for pollutant soil assimilation, microbial degradation, and plant uptake. Removal rates by infiltration are determined by biological activity, adsorption, filtration, and oxidation, which are the primary mechanisms (Brown and Caldwell 2007). Microbial degradation rates are dependent upon environmental conditions including temperature, moisture, energy sources, and oxygen and nutrient availability (Donker et al. 1994). Temperature is typically directly related to microorganism’s cell reaction rates and the environmental conditions of the cell habitat. Microorganism decomposition rates increase with increasing temperature up to approximately 45°C, after which the rate declines (Paul and Clark 1996). Oxygen and soil moisture also have significant effects on degradation rates. Aerobic conditions result in greater degradation rates as compared to anaerobic cells (Paul and Clark 1996). Moisture has an indirect effect on degradation rates as high levels decrease the oxygen content therefore leading to the slower rates associated with anaerobic microorganisms, but also controls the solubility and availability of nutrients required by microorganisms to maintain activity (Paul and Clark 1996). Removal of specific contaminants varies with environmental conditions. Nitrogen removal, for example, is more effective in the soil subsurface than at the soil surface and is dependent upon soil type, hydrology, and biogeochemistry (Mayer et al. 2007). Nitrogen removal is accomplished primarily through various nitrification and denitrification processes in addition to plant uptake in the overland flow and soil infiltration. Although the processes are not generally understood, denitrification plays the dominant role in both (Corbitt 1998). The majority of phosphorus is fixed within the soil profile, although small amounts are removed via plant uptake (Corbitt 1998). In addition to metals within waste streams, an overload of biodegradable organic material can lead to incomplete removal within the soil profile and mobilization of iron, manganese, and other metals (McDaniel 2006), greater details in this process are discussed in the next section. Filter strip design dimensions of width, length, and slope impact pollutant removal. An increase in filter strip width increases infiltration, reducing the volume and contaminant concentration and surface outflow (Schmitt et al. 1999) as they provide more area for infiltration. Nutrient removal, such as nitrogen, is more effective in wider strips (Mayer et al. 2007). Trapping, however, is not impacted by filter strip width as it is a function of the vegetation and slope (Jin and Romkins 2001). Literature values are available for minimum and maximum filter strips widths. There is a point in which increasing the filter strip width begins to impact the flow design, as it is difficult to maintain even distribution and sheet flow over very large widths, and does not result in increased removal. 9 Increased filter strip lengths can also reduce pollutant loads. The length of the filter strip has been shown to impact the removal of inorganic compounds, such as Cu, Fe, Zn, K, Na, and Ni (Edwards et al. 1997). A longer filter strip has also been shown to increase removal due to increased trapping (Lee et al. 2003). Sediment removal of over 90% resulted with a filter strip of 10 m length (Dillaha et al. 1988; Goal et al. 2004). Magette et al. (1989) investigated 4.6 m and 9.2 m long filter strips and found an increase in pollutant removal with an increase in length, it was also found that a 4.6 m filter strip was below the threshold to achieve any removal of some pollutants. Dickey and Vaderholm (1981) found that channelized systems require greater lengths than those designed for overland flow for equivalent removal performance. Many pollutants experienced an exponential reduction with increasing length, but NOa', TKN and TOC did not undergo significant reductions after 3 m and NH3, PO4‘, and TP beyond 6 m (Srivastava et al. 1996). Increased Slopes result in reduced treatment effectiveness (Hay et al. 2006). Sediment trapping and transport is strongly dependent upon the slope of the filter strip. An increase in the slope leads to a reduction in the trapping efficiency and an increase in pollutant transport (Jin and Romkins 2001, Dillaha et al. 1988). However, the slopes must be great enough to maintain sheet flow. 10 Consequently, length, width, and slope are all critical components for sizing agricultural filter strips. Recommendations from Dickey and Vanderholm (1981) include a minimum width of 61 m and a length to accommodate runoff volumes for a 1-yr 24-hr storm calculated based on slopes and contact time. Others investigated ratios of drainage area to infiltration areas from 1:1 to 6:1 (Nienaber et al. 1974, Lorimor et al. 2003). NRCS designs are based on the infiltration of a 25-yr 24-hr storm and the length and width are based on the infiltration of these volumes using the equations provided in Appendix C. Filter strip soils must provide adequate filtration to avoid flooding during wastewater application. Minimum hydraulic conductivities have been suggested by previous researchers from 0.27-0.5 inlhr (Schueler 1987). The soil textures that fall within this range include sand, loamy sand, sandy loam, loam, and silt loam (silt loam falls within the lower range only) (Rawls et al. 1982). Previous research done by Mokma (2008) added to the validity of these assumptions in which clay loam was excluded as it did not adequately treat waste water. Komor and Hansen (2003) speculated that poor performance and a greater impact to groundwater was due to greater hydraulic conductivities at a site with Silt loam soil as compared to a second site with loamy soils. Saturated hydraulic conductivities are increased as compared to that of unsaturated conducitivties (Miyazaki 1993). This can lead in increased flow within land application of wastewater through soil profiles, decreasing the time for adsorption and increasing the transfer of pollutants within the soil. 11 Vegetation plays a role in the uptake of pollutants by impacting velocity and infiltration processes. In addition, vegetation develops dense mats of roots on the upper portions of soil profiles which can provide nutrient trapping and increases soil oxygen through respiration (Bhaskar, 2003). Various researchers have experimented with the selection of vegetation to increase pollutant removal and demonstrated that some species are contaminant specific (Schmitt et al. 1999). For example, nitrogen removal is dependent upon the vegetations’ depth of root zone and ability to provide flow paths that favor microbial denitrification (Mayer et al. 2007). Goal et al. (2004) has shown that sod grasses have the greatest effect on soluble phosphorus removal and particulate nutrients in comparison to rye grass and mixed grasses. In addition, Goal et al. (2004) found no trends with grass type and N03“ removal. Some results have shown no change in the collective removal efficiencies between entire vegetation Classes, such as forest vegetation and grasses (Dosskey et al. 2007). In terms of plant life cycles, perennial plants have been shown to trap sediments effectively and allow for greater infiltration and reduce erosion in comparison to annual plants (Lovell and Sullivan 2006). However, it has been shown that there is no difference within perennial species in terms of removal (Schmitt et al. 1999). Trapping efficiency is increased due to an increase in vegetation density (Lee et al. 2003). After a two years of growth there is no difference in infiltration due to age of vegetation (Dosskey et al. 2007). Multiple plant species allows for 12 numerous soil root sizes, and various stalk and leaf sizes to have the greatest overall impact on infiltration and sedimentation (or trapping). Although previous studies have investigated various filter strip design parameters including length, width, slope, and vegetation in an effort to determine design standards and maximize treatment efficiency, there have been no studies to date correlating loading to the depth of soil and groundwater table required to effectively treat runoff prior to reaching groundwater. Determination of this depth is critical in developing effective standards for the implementation of the practice. 2.3 Laboratory Evaluation of Treatment System Design Components Primary soil assimilation mechanisms include biological oxidation, adsorption, filtration, and oxidation (Brown and Caldwell 2007). The soil profile provides the environmental conditions to support biological and biochemical activity (Haggblom and lVlilligan 2000). Application of wastewater increases the soil pore water thereby decreasing available oxygen within the soil. Under aerobic conditions, carbon sources are the electron donors with oxygen accepting the electrons (T arradellas et al. 1997; Rittmann and McCarty 2001). After most of the oxygen is depleted, anaerobic and facultative microorganisms become dominant. The carbon source remains the same but the electron acceptor changes in order of energy potential (Haggblom and Miller 2000). AS oxygen is removed or fixed within the system, other oxidants act as electron acceptors. 13 The diagenesis model (or electron tower) ranks the oxidants in order of free energy yield per mole of organic carbon oxidized, or 02, N05, MI'IOz, Fe(OH)3, 8042', and methanogenesis (Froelich et al. 1979; Postrna and Jacoksen 1996; Matocha et al. 2005). After the oxygen is utilized within the system, oxidants will be reduced in accordance to free energy yield. A reduction of metals causes previously immobile metals to mobilize and leach into groundwater. Mn solubility is increased when converted from Mn(lV) to Mn(ll) as Mn(ll) is typically released into solution (Norvell 1988). Fe(|l) is more soluble than Fe(lll) and is governed by pH, as reduced forms are more prevalent in soils with a lower pH.(Lindsay 1985). Solubility of metals can be decreased through an increase in pH (McGowen and Basta 2001). Mn oxides are present as surface coatings arranged in octrahedra sheets or tunnel structures and are typically associated with Fe oxides (Bartlett and Ross 2005). Acidic or saturated soils in combination with soil organic material can easily lead to Mn(lV) reduction (Bartlett and Ross 2005). Reduction mechanisms can be biological or physical/Chemical in nature. Mn complexes are more available for microbial reduction than the Fe complexes (Lovley 1991). Biological iron reduction from Fe(lll) to Fe(ll) can follow numerous pathways including bacterial reduction, acting as a respiratory electron acceptor, and interactions with microbial end products (Paul and Clark 1996). Enzymatic conversion of Fe(lll) to Fe(ll) under anaerobic conditions is the main cause of iron reduction (Roden and Zachara 1996). Organisms reduce Fe(lll) and typically Mn(lV) enzymatically, in addition Fe2+ reduces Mn(lV) nonenzymatically (Paul and Clark 1995). The affinity for Mn(ll) to adsorb to manganese oxides 14 results in excess bound Mn(ll), so when Mn(lV) is reduced, soils release the excess bound Mn(ll), further increasing the release of Mn(ll) (Fendorf et al., 1993a, 1993b),. Mn(lV) and Fe(lll) reduction rates are governed by the mineral surface area (Burdige et al. 1992; Roden and Zachara 1996; Larsen et al. 1998; Matocha et al. 2005). In addition, nitrogen species within the soil can also affect metal mobilization. The accumulation of N02' results in the reduction of MnOz to Mn(ll) (Vandenabeele, 1995) while N03' typically inhibits iron reduction (Paul and Clark 1995). Although some relationships and mechanisms have been developed found, there are many exceptions that are still to be explained. For example, sulfate reduction is typically preceded by Fe oxide reduction according to the diagenesis model, there have been cases where pore water has shown a reduction in 8042’ prior to Fe oxides (Matocha et al. 2005). The relationship between the various water quality parameters is not completely known. Although Mn and Fe leachate are mainly aesthetically unpleasant in groundwater, metals further down the electron tower, such as arsenic, will leach after other oxidants have been exhausted and pose serious human health risks. Groundwater wells high in As concentrations also reported high concentrations of Mn(ll) and Fe(ll) (Kim et al. 2002). Although the processes involved have been investigated, there are still many factors to be determined, and in particular, how these processes will occur simultaneously (Holden and Fierer 2005). An increase in soil depth is predicted to provide greater pollutant removal due to the increase in available soil surface area, an increase in the soil pore area for 15 microbial activity, and an increase in available area for the vegetative root system (Bratieres 2008). Microorganisms are naturally present within the soil profile (Paul and Clark 1996), and a larger soil volume will in turn increase the soil microbial mass. However, it has been shown that microbial biomass is the highest at soil surfaces and decreases with an increasing depth (Paul and Clark 1996; Holden and Fierer 2005). The change in microbial activity may not be linear. Microbial degradation rates may also be affected due to depth. Oxygen at the soil surface is at atmospheric concentrations but drops significantly with increasing depth (Wood and Petriatis 1984), which in turn would decrease microbial degradation rates with increasing depth. Changes in microbial mass are also greater with increasing depth than the Change reported in different soil types (Holden and Fierer 2005), so it is assumed that soil depth will have a greater impact on degradation and metal mobilization than soil type. Adsorption of Mn and Fe is a minor component in Fe and Mn reactions within soil as these reactions are mainly driven by pH and oxidation reduction reactions (Shuman 2005). Therefore the increase in the soil cation exchange capacity (CEC) within the sandy loam due to increased fractions of silt, clays, and organic matter, is not predicted to have a large overall effect in metal leaching. A lack of literature exists for the effect of capillary rise on pollutant removal capacity. An unpublished study by Mokma (2008) investigating pollutant assimilation of food processing waste in soil columns revealed soil saturation had 16 risen 2-3 inches from the bottom of the column during deconstruction due to soil saturation within the bottom of the column due to poor drainage resulting in capillary rise. The increase in the soil water has a direct effect on the availability of oxygen within the soil and will lead to anaerobic soil conditions. However, it has been shown that biomass increases directly above the water table (Paul and Clark 1996). In addition, the microbial activity increases in the capillary fringe as the rising and falling of the water table redistribute necessary nutrients and microbial mass (Holden and Fierer 2005). 17 CHAPTER 3: METHODS AND MATERIALS Experimental design was based on previous research. Details of the design and experimental operation for each research section is outlined below. 3.1 Runoff Characterization Runoff for characterization of water quality was collected at the Michigan State University Dairy Teaching and Research Facility (MSU dairy). The MSU dairy is a fully operational 160 head dairy facility that was originally designed to transport runoff using a traditional urban storm water collection system. In 2008, the existing system was modified to collect and divert water from two approximately one acre in size areas into two 86,774 gal storage basins. Four source locations were sampled to investigate pollutant sources; the areas adjacent to the heat check lot, upright silos, bunker silos, and main roadway. The two storage basins were also sampled for the composite runoff water quality. The heat check lot is an outdoor cattle holding area and consequently, is typically high in animal waste. Both silo locations are feed storage areas. The upright silos are covered, but are prone to spillage and produce dry weather leachate. Bunker silos are partially uncovered feed storage and remain open to the elements making them particularly susceptible to environmental conditions, leaching, and runoff. The main roadway is used to transport and mix feed and animal waste, it provides data for multifunctional impervious feedlot areas. Storage basin 1 collects runoff from a 1.28 acre area containing the heat check lot and roadway areas. The heat Check lot accounts for 9% of the total drainage surface area for storage 18 basin 1 with the remaining area comprised of roadway surfaces. The second storage basin collects runoff from a 1.14 acre area where 23% of the area is bunker silos, 6% of the surface area is upright Silos, and the remaining surface is roadway. A comprehensive management plan, Appendix A, was given to the MSU dairy prior to sample collection. 3.1.1 Sample Collection Methods Samples were collected during precipitation events that produced runoff volumes adequate for sampling. Clean plastic sample bottles were used for each new sample and collection devices were cleaned with a dilute bleach solution and rinsed with de-ionized water a minimum of three times to avoid cross contamination of samples. Grab samples were collected above sewer grates where large quantities of runoff accumulated, then were preserved if required, and stored until analyzed following proper quality control and quality assurance (QAIQC) protocols, Appendix B. 3.1.2 Laboratory Analysis All samples were evaluated for the water quality parameters listed in Table 2. These parameters are typical water quality indicators used by environmental 19 regulatory agencies. Nutrient removal was analyzed for nitrogen species and total phosphorus as these are the major nutrients of concern associated with agricultural practices. Oxygen requirements were measured via the 5-day biochemical oxygen demand (B005) and chemical oxygen demand (COD). Manganese (Mn) and iron (Fe) concentrations were measured as indicator species for metal leaching and reduction potential. Arsenic was also included as it is currently a significant groundwater contamination concern in Michigan. Sedimentation, filtration and loading limits were investigated using solids data. Samples shaded in Table 2 were preserved and transported to the Michigan Department of Natural Resources and Environment State Environmental Laboratory for analysis. The remaining parameters were analyzed at the MSU Ecological Engineering Laboratory. All laboratory analyses were subject to detailed QNQC procedures, see Appendix B. Table 2: Water Quality Parameters for Source Characterization and Field Treatment Systems Alkalinity USEPA 310.1 (1) 24 hours USEPA 405.1 (1) Analyze Immediately vN ‘ \ USEPA 350.3 (1) 28 days USEPA 354.1 (1) 24-48 hours USEPA 353.3 (1) _ 24 hours at ' ' ‘ - 415.2 Total and Soluble COD USEPA 410.4 1 TP USEPA 365.1 1 TS USEPA160.3 1 TSS USEPA160.2 1 (1) (US EPA 2009a) (2) (US EPA 1996) 20 3.1.3 Precipitation Data In addition to water quality data, precipitation data was measured using a Campbell Scientific TE-525 rain gage produced by Texas Instruments. Precipitation data provided an estimate as to the intensity of a rainfall, identification of rainfall return period and precipitation duration and, in conjunction with the recorded sample time, an estimate of the water accumulated on the ground at the time the samples were taken for analysis of covariance when appropriate. 3.1.4 Data Analysis Statistical analysis was conducted using ANCOVA in SAS, with rainfall and season as covariates when appropriate, to determine the statistical significance of location on each measured water quality parameter. Assumptions that residuals are normally distributed and the variances are homogenous were evaluated using normal probability plots and side-by-side .box plots to ensure their validity. Covariates were selected to an attempt to reduce the experimental wide error; they were used within the statistical model when ANOVA indicated that they increased model significance. When ANCOVA or the ANOVA was significant, difference of least squares means was used to compare the treatment means and their interactions. 21 3.2 Analysis of Field Treatment System The second phase of research was a full scale implementation and analysis of vegetated filter strips at two site locations, the MSU dairy and the small Ml dairy, details for this second site are discussed on page 24. Each site was designed in compliance with the specifications of the NRCS Technical Guide titled ‘Wastewater Treatment Strip 635,” Appendix C. Sample collection and analysis of runoff water quality pre and post treatment provided data for assessment of treatment. The design at the MSU dairy is composed of two filter strips, each 400 feet long and 40 feet wide with a 4% slope. Side slopes of 12.5% along the length of the filter strip created the channel which was backfilled with the sandy loam soil native to the site. Vegetation was planted as a mixed grass species containing 37% Tuscany II Tall Fescue, 28% Smooth Bromegrass, 20% Graze N Gro Annual Ryegrass, and 12% Chiefton Reed Canarygrass. After a two year growing period, the Annual Ryegrass was the dominant species, with the three remaining species onsite but at lower densities. Five rock checks extended across the width of the filter strip (with a depth and width of two feet) at the flow entrance and every 100 feet downslope to redistribute flow. Storm drains divert runoff from two locations, one from 1.14 acres surrounding the feed sources and a second from 1.28 acres which included the heat check lot and roadways, into two separate 86,774 gallon concrete basins, Figure 1. Grab samples were 22 collected from each of the two storage and sedimentation basins for baseline data. Figure 1: MSU Dairy Concrete Storage and Sedimentation Basins From each storage and sedimentation basin, wastewater is transported to the two small concrete distribution basins at the top of each filter strip, each ~5,000 gallons using a pump system active by level sensors (flow rate ~350 gal/min). Wastewater exits the distribution basins via four vertical slots, each 1 in wide and 24 in tall, which empties into a rock Check to evenly disperse flow across the width of the filter strip, Figure 2. 23 Figure 2: Filter Strip Influent Flow Dispersion Nine collection boxes were installed in the rock check of each filter strip at the MSU dairy for surface samples. Subsurface drainage tile was installed 9 to 15 inches below the surface 25, 50, and 150 feet downslope to collect infiltrate that has passed through the soil profile. The tile drained to a sample well for collection using a sampling pole affixed with a clean sample container (washed with a 10% bleach solution and tripe rinsed with deionized water between samples). Ten sampling events were collected over a 2 year period. At the second site, the small Ml dairy was designed to treat runoff from a 40 cow dairy from an approximately ‘/4 acre drainage area. Dairy feedlot and manure 24 storage runoff were diverted via overland flow to a small concrete basin. Effluent from the concrete basin flowed over a weir to a bioretention basin for storage of runoff volumes up to a 25-yr 24-hr storm for the % acre area. The subsurface of the bioretention basin was lined with an impermeable geomembrane with a subsurface collection tile located above the membrane to transport effluent that leached through the soil to the filter strip via gravity. The filter strip was 110 ft long, 40 ft wide, with a 0.5% slope and sandy soil present at the site prior to installation. Rock checks were located at the top of the filter strip and 50 feet. Six surface collection boxes were installed in the two rock checks for surface water collection. Subsurface samples were collected using 1.5 ft and 2.5 ft collection wells made from corrugated pipe buried 3 ft and 13 ft downslope of the first rock check. This subsurface collection method was different from the MSU dairy site as this site had a sandy soil which increased hydraulic conductivity and decreased the length the runoff traveled downslope before infiltrating. 3.2.1 Sample Collection Methods Grab samples were collected within 24 hours after a rainfall event for all sample locations. Influent data was collected from the two concrete storage basins for the MSU dairy site and the concrete sedimentation basin and bioretention basin at the small Ml dairy. After baseline samples were collected at the MSU dairy the pumps were manually activated and one location from each rock check was sampled for surface water quality and all subsurface sampling locations were 25 sampled if effluent was present. Figure 3 provides a diagram of the sampling locations. .. Sample Well If C) O C) Rock Check Figure 3: MSU Dairy Filter Strip Sampling Locations Typical operation at the MSU dairy site relies on pump activation due to level floats at the top of the basin to activate the pumps and 1 foot from the bottom of the basin to turn the pumps off. If a storm event is not large enough to initiate pumping on the high float level switch, pumps must then be manually operated within 72 hours of the storm event. At the small Ml dairy site one sample from each rock check was again sampled and effluent was collected from the 4 samples wells if effluent was present, Figure 4 is a diagram of the sampling locations. 26 8::8:>— Sample Well Rock Check Figure 4: Small lllll Dairy Filter Strip Sampling Locations Rainfall on the filter strip surface area was assumed to be insignificant in dilution of samples. At the MSU filter strip site, water was applied after rainstorm events and rainfall on the filter strip was therefore not a factor in dilution of the applied runoff. Although the small MI dairy filter strip was a gravity fed system, there was a delay from the transport of water from the source location to the filter strip, and the area of the filter strip to which the water was actually applied (within the first 15 ft) was negligible compared to the drainage area (< 1%). All samples were transported immediately to the MSU laboratories and preserved if necessary. Samples for the State Environmental Laboratory were preserved and transported in a cooler within a two week period. Ten sampling events were obtained from the MSU dairy and five sampling events from the small MI dairy. 3.2.2 Laboratory Analysis Water quality evaluation was determined by the identical parameters to those of the source characterization listed in Table 2. Analysis procedures are also identical to those of the source characterization, Section 3.1.2. 27 3.2.3 Precipitation Data Precipitation data was collected at the MSU dairy using a rain gage. Precipitation for the small Ml dairy was obtained from a rain gage in Charlotte, Ml, approximately 10 miles from the fann. This data was critical for comparison of runoff volumes and filter strip performance. 3.2.4 Data Analysis Data was evaluated for general trends including removal percentages for each water quality parameter. Because conditions varied for each storm, reliable replication was not possible which prevented statistically significant results for analysis. 3.3 Laboratory Evaluation of Treatment System Design Components Soil columns with surface vegetation were designed, constructed, and operated to evaluate the objectives for the laboratory research. The columns provided data to correlate pollutant removal to soil depth, soil type, and simulated groundwater effects. Wastewater was applied to column surfaces and allowed to leach, producing effluent that could then be analyzed to evaluate pollutant removal. 28 3.3.1 Soil Column Experimental Design Soil treatment columns were evaluated for three treatment depths, two soil types, and submerged or not submerged conditions. The three column lengths were 12 inches, 30 inches, and 48 inches. The 12 in column was selected for direct comparison to field data obtained at this depth. An increase in soil depth for the remaining two columns allowed for the investigation of soil depth to pollutant removal. The two soil types were sand and sandy loam, selected from the list of l soils in Section 2.2. Sand soil provided data for a soil with the greatest hydraulic conductivity and sandy loam a lesser hydraulic conductivity as a comparison of soils within those accepted for the technology. These soil types also corresponded with the soil types at each field site. Groundwater simulation, or capillary rise effects, were investigated by submerging the bottom end of a set of identical sets of columns for each design depth to mimic the interface between the soil and groundwater, commonly termed the vadose zone. When submerged, the design prevented air from entering the bottom of the column and allowed for capillary rise within the soil column system. Vegetation, hydraulic load and organic load were held constant throughout testing. Research has shown that vegetation pollutant removal performance varies by individual pollutant and vegetation type, so a mixed grass species was selected for use in all columns to maximize overall pollutant removal. A combination of 37% Tuscany ll Tall Fescue, 28% Smooth Bromegrass, 20% 29 Graze N Gro Annual Ryegrass, and 12% Chiefton Reed Canarygrass was selected, identical to the selection for the MSU dairy. The mixed species provided the necessary variation required for pollutant removal mechanisms associated with vegetation (trapping, uptake, and root size). This vegetation will also provide adequate food for grazing in typical farmstead operations. A constant hydraulic load for wastewater application was determined using a BOD concentration of 225 mg/L. This BOD concentration was selected from preliminary source characterization data for typical dairy runoff loadings with adequate management (Larson 2009). Source characterization research has shown that this concentration is achievable, although concentrations have been found that exceeded this number by an order of magnitude. An organic load of 75 lbs/acre/day was used in the simulated wastewater, as was determined in evaluating data from a prior study conducted by Mokma (2008), which is currently in review for publication. The study results showed metal leaching from a column depth of 36 in from a loading of 75 lbs BOD/acre/day but not from a loading of 50 lbs BOD/acre/day, so the higher loading was selected to produce leachate. A power analysis in SAS, a statistical computing program, was completed to determine the required number of replications to predict a statistical difference within the treatment effects. As the analysis is relatively variable due to interpretation, only depth and soil type were included to determine power. Unpublished data from Mokma (2008) provided the necessary variance for soil 30 type and depth required for the analysis. Statistical power analysis indicated that 3 replications are required to produce a power of 0.94 for column depth, Figure 5. An increase to 4 replications did not increase the power significantly for column depth, so 3 replications were deemed appropriate. Soil type did not establish a significant power even after 6 replications. As more replications did not produce a significantly greater power for soil type, and the amount of resources necessary to predict a highly significant result are not feasible, replications were limited to 3. 1.2 0 8 I + SOII I 0.6 ,’ / +Depth I 0.4 ‘ / a. -Soil*Depth Power 0.2 Replications Figure 5: Filter Strip Power Analysis The three treatments each had the following levels, depth - 3, soil - 2, groundwater — 2, and would require 12 columns for each replication, or 36 columns total. The interaction of capillary rise and soil type were not outlined in the objectives, so direct comparison of the two was not required. Therefore, a 31 groundwater simulation was conducted for sandy loam soils only at each treatment depth. This reduction allowed for determination of the study objectives with a total of 30 soil columns, Table 3. Columns were assigned randomly to experimental conditions to minimize experimental wide error. Table 3: Soil Column Treatment Assignment Application Soil Type Length (in) Submergence Column #3 WW Sand 12 Air 12, 25, 26 WW Sand 30 Air 1, 7, 13, 20 WW Sand 48 Air 3, 19, 24 WW Sandy Loam 12 Air 10, 18, 23 WW Sandy Loam 30 Air 4, 5 VWV Sandy Loam 48 Air 14, 15, 17 Water Sand 30 Air 30 Water Sandy Loam 30 Air 22, 29 WW Sandy Loam 12 Water 2, 11, 21 WW Sandy Loam 30 Water 6, 16, 28 WW Sandy Loam 48 Water 8, 9, 27 3.3.2 Soil Column Structural Design Soil columns, Figure 6, were constructed from 6 inch drainage pipe. The lengths of the columns correspond to the actual soil depth plus two additional inches on the top of the columns for application of wastewater and another two inches on the bottom that was packed with washed pea gravel to prevent soil from settling and leaching from the column. The pea gravel rests on a fine fiberglass screen attached to the bottom of the column to allow for free flow of leachate. 32 I Figure 6: Soil Column Construction Buckets located directly beneath the columns collected effluent. Wastewater was fed by hand in two doses one followed directly by the second, as the 2 inch free board excess could not hold the entire 1.4 L in one application (discussed further in section 3.3.4). A single batch of simulated wastewater was prepared at the time of application to ensure uniform loading among columns. During feeding the wastewater was mixed prior to application on each column to maintain even dispersion of pollutants. 3.3. 3 Waste Water Composition A Synthetic wastewater was used to provide the carbon and nutrients required by the microbial biomass. BOD concentrations were achieved by adding D- g/UCOse (dextrose) to dechlorinated tap water. The estimated oxygen demand of 33 glucose for BOD is 70% of the theoretical 02 (Gray 2005) as calculated using Equation 1 and 2. Seventy percent of this oxygen demand is then used to estimate the BODu for glucose, Equation 3. C6H1206 + 602 —) 6C02 + 6H20 Eqn.1 1925—0—2- 1 CH 0 "'0 =1.07 0 En.2 g 6 12 6 1808C6H1206 g 2 4 mol (1.07g02X0.70) = 0.75 g 301),, Eqn. 3 Consequently, 1 g of glucose produces 0.75 g of BOD. To achieve the desired organic loading of 75 lbs BOD/acre 2 times per week, the simulated waste water was designed with an average BOD concentration of 225 mglL. A hydraulic load of 1.4 L/day with a glucose concentration of 300 mg/L will be applied 2 days a week to achieve a loading of 75 lbs BOD5/acre twice a week, Equation 5. 1gC6H1206 (225mgBOD5) = 300C6H1206y£ Eqn.4 0.75 g 3005 L L 751b BOD . ——7Scr—e——5— (27r(3m)2) day = 1.4L Eqn. 5 300C6H1206TL§ day The synthetic wastewater was prepared according to Trulear and Characklis ( 1982) to provide the essential micro and macro nutrients for the microbial biomass. Table 4 details the nutrient solution composition which has been 34 proportionally adjusted according to BOD concentrations. One substitution from the nutrient solution is MnOz to replace MnCIz. MnOz adds manganese to the soil columns as Mn(lV), an immobile form of manganese, as an objective of the research is to determine reduction from the immobile form to the mobile form, Mn(ll). Constant agitation during application is required to distribute this chemical as in this form it is insoluble. Table 4: Nutrient Solution Constituent Concentrations for Synthetic Wastewater comment Twist—2.31823”) ‘gflciftatztiag't"? 06H1206 10 300 FeCI3 0.045 1.35 Mn02 0.005 0.15 20304 ' 7 H20 0.008 0.24 CuCIz ' 2 H20 0.005 0.15 CoCl2 - 6 H20 0.007 0.21 (NH4)6 M07 024 ' 4 H20 0.005 0.15 Na28407 - 10 H20 0.003 0.09 Na3C6H507 ' 2 H20 0.408 12.24 NaH2P04 - H20 0.575 17.25 (NH4)2 804 0.367 11.01 NH4 CI 3.417 102.51 CaClz 0.308 9.24 M90|2 - 6 H20 0.565 16.95 3.3.4 Soil Column Operation Columns were fed simulated wastewater for 7 months twice per week, on day 1 (Monday) and day 4 (Thursday) to allow for drying between applications. Control columns were fed 1.4 L of dechlorinated tap water coinciding with wastewater application. Declorination was achieved using a chemical chelating agent 35 commonly used for fish tanks. The dechlorinating agent was added then mixed, and then chemicals added as described in the synthetic wastewater section above. Column influent and effluent was collected bi-weekly following wastewater application on day 4. Twice a week 10-12 hours after feeding (as was determined to be the time required for columns to leach the entire wastewater volume) the effluent was measured for volume and ambient air temperatures recorded. Air temperature was assumed to be the soil temperature as size permitted columns to equilibrate. Prior to wastewater application, water was removed from all submerged columns for effluent collection. After effluent collection and the leaching of all the wastewater volume, the soil columns were then re-submerged. Samples were prepared for laboratory analysis according to the QAIQC procedures outlined section 3.1.2 and in Appendix B. 3.3.5 Water Quality Parameters Nutrient removal evaluation required lab analysis for nitrogen and phosphorus. Nitrogen measurements included TKN, ammonia, nitrate, and nitrite to assess the full nitrogen cycling as well as the impact on the other various soil-water biogeochemical processes. Oxygen requirements were measured via the 8005 and COD. Metals were analyzed to evaluate the reduction potential and metal leaching. Manganese was present within the system as Mn(lV) and Iron as Fe(lll). The reducing conditions result in conversion from these insoluble forms to soluble forms, Mn(ll) and Fe(ll). Measurement of the influent and effluent allowed for determination of the redox conditions within the columns and the 36 loading conditions that result in leaching of metals. For a detailed list of the parameters measured and the methods for their collection and analysis see Table 5. Table 5: Water Quality Analysis Parameters and Methods For Soil Columns - . .~ .: .arameterf' .. .. . . other!“ " i " ‘.Deteciion'leit".. II". If, .dfl'tmg:;;.‘,_j,. Alkalinity (mg/L CACO3) USEPA 310.1 (1) 10 mg/L CaC03 24 hours 3005 (mg/L) USEPA 405.1 (1) 2 mg/L Analyze Immediately Iron SW-846 method 6010B (2) 0.02 mg/L 6 months Magqanese SW-846 6020 (2) 5 pg/L 6 months NH3 (mg/L) USEPA 350.3 (1) 0.02 mg/L NH3-N 28 days N02 (mg/L) USEPA 354.1 (1) 0.002 mg/L NOZ-N 24-48 hours N03 (mg/L) USEPA 353.3 (1) 0.3 mglL NOS-N 28 days pH Analyze immediately TKN (mg/L) USEPA 351.1 (1) 1 mg/L 28 days COD (mg/Q USEPA 410.4 (1) 1 mg/L 28 days TP (mg/L P) USEPA 365.1 (1) 0.02 mg/L P 28 days (1) (us EPA 2009) (2) (us EPA 1996) 3.3.6 Soil Column Deconstruction Before deconstructing, a soil column flow rates study was conducted over two feeding periods. Volumes were recorded every 3 minutes for short columns and every 15 min for the longer columns increasing to every half an hour after an hour. The volumes were then divided by the time interval to obtain an average flow rate for each time segment. After effluent sampling, one of each replicate column was deconstructed and the soil was sampled every Six inches to determine the fate of metals within the columns. Soils samples were digested and analyzed for Mn, Fe, and COD at the State of Michigan Environmental Laboratory. 37 3.3.7 Data Analysis Statistical analysis was conducted using ANOVA in SAS, with time as a repeated measure, to determine the statistical significance of depth, soil, and submergence on each measured water quality parameter. Assumptions that residuals are normally distributed and the variances are homogenous were evaluated using normal probability plots and side-by—side box plots to ensure their validity, and adjusted using grouping and data transformations when necessary. When the ANOVA was significant, difference of least squares means was used to compare the treatment means and their interactions. Statistical results in addition to treatment averages and percent reduction allowed for evaluation of research objectives. 38 CHAPTER 4: RESULTS AND DISCUSSION Studies were carried out based on the designs outlined in the previous section. Results for each of the three sections, comparisons between field and column performance, and design implications are reported in this chapter. 4.1 Runoff Characterization Runoff results for 9 storm events from July 2008 through May 2009 were analyzed at six sampling locations. Average concentrations at each sampling location for 17 water quality parameters are in Table 6. Storage basin 1 primarily receive wastewaters from the heat check lot and roadway, while storage basin 2 collects runoff diverted from the bunker and upright silos. 39 Table 6: Feedlot Runoff Water Quality Parameter Average Concentrations g. ,. ,. :r a 3 :.~ 3 3 5 g i, g, E) g 3, a 2 a, a I b g g v .E 5 5 P E g Q. 'E I V O 3 8 I N I (0 \E/ I V —' : 8 8 8 E 2 9 § 8 2 E '- Bunker Silo 5.87 277 17 2320 900 18 0.14 14 115 12 C .s N .L 00 00 0| 0| .3 .3 N Heat Check Lot 8.35 940 1 7 3180 930 Roadway 6.37 168 1 1 1380 410 Upright Silo 6.61 193 16 1910 730 Storage Basin 1 7.06 445 8 790 240 .L O .0 . O m if N .L 101 20 N on P S N 0) O O on RQ‘JOI .s .I. 8 P . N N Storage Basin 2 5.21 136 20 2510 1 180 74 7 8 9 9 g g. 2‘ 3 ’5‘. fig 9 g ‘8 £3 E g 5; 3 s g 32 g a O C a) — co 4 1’3 ‘9 I2 s 2 L .9 55. 0 < Bunker Silo 2490 1310 410 190 418 3950 990 1342 39 4.9 Heat Check Lot 4910 2060 1030 800 491 2530 1250 6730 526 4.3 Roadway 1880 1340 370 250 216 1180 440 836 39 1.7 Upright Silo 1540 1210 780 540 221 3650 570 1124 24 11.8 Storage Basin 1 2490 900 450 270 216 3200 210 1226 151 2.3 Storage Basin 2 2970 910 310 250 41 1 5560 790 1384 26 3.0 Statistical analysis of water quality parameters (with a minimum of 8 complete data sets from the 9 total sampling events) was generated using SAS software (SAS 2008) to determine statistical differences in the mean source concentrations. Statistical models were fit using ANCOVA to determine if covariates of season and rainfall reduced the overall error within the model. If covariates did not reduce the experimental wide error for each water quality parameter assessed, the covariates were eliminated. If ANCOVA or ANOVA was determined to be statistically significant for each parameter then the model was evaluated for comparison of the treatment means for main effects for 40 location using differences of least squares means. Analysis results for each parameter are outlined below. Runoff from animal waste and feed contain high concentrations of COD, as can be seen by the values for the heat check lot and the silo locations, Figure 7. Average values for animal waste were near 3,000 mg/L while feed average concentrations were approximately 2,000 mglL. Although there were large differences within the averages of these sources, there was not a statistical significance between the two COD sources over the length of the study. However, there was a statistical difference between COD concentrations from the heat check lot to that of the roadway and storage basin 1. This indicates that although the concentrations from the heat check lot were high, the dilution from the roadway runoff was significant enough to impact the composite storage basin concentrations. The second storage basin has a greater COD concentration as the source locations, bunker and upright silos, have significant COD contributions. Analysis of 8005 produced similar results although had only 5 complete data sets. 41 3500 3000 - 2500- 2000- COD (mglL) 1500 ‘ 1000- 500- Heat Check Roadway Storage Basin Upright Silo Bunker Silo Storage Basin Lot 1 2 Sampling Location Figure 7: Average COD Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. Phosphorus mean values for runoff sources are similar in concentration and produced statistically significant differences for the mean values for storage basin 1 and 2, Figure 8. Again, storage basin 1 is statistically different from the heat check lot taking on the phosphorus characteristics of the roadway, indicating phosphorus is also dependent upon dilution and runoff quantity. 42 25 20 — A 3 v ' AB 3 15 «L A3 AB —— —— —— In 2 o S a 10+- __ __ __ _ 2 BC a. C S l— -— —— —— —— ——- —— O 4‘ I I I I r ‘ I Heat Check Lot Roadway Storage Basin Upright Silo Bunker Silo Storage Basin 1 2 Sampling Location Figure 8: Average Phosphorus Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. The heat check lot produces significantly higher total solids concentrations than all other sources, Figure 9. Notched grooves for animal footing and safety within the heat check lot concrete reduces the impact of scraping and allows build-up of animal waste, the likely source of the solids concentrations. A more effective cleaning technique or diversion of rainwater from manure is required to reduce solids concentrations from this source. However, in this case the farm may not benefit from additional maintenance as the composite sample again takes on the characteristics of the roadway runoff in regards to solids concentrations. 43 6000 5000 1 . A“ II a 4000 -— s c» . I 1.5. t ,. 3 3000 i = i t. «r ‘2 9 . 5 4- ‘B. B a . 19 2000 —— 9 ., *‘——— :49— ’ :1 t“ 5‘" a“ f" 1000 ~— . ‘*“——— ':'—-— .' ' :3 fl, ”—1" '.,. 5t" “at!— 4‘— Heat Check Roadway Storage Basin Upn’ght Silo Bunker Silo Storage Basin Lot 1 2 Sampling Location Figure 9: Average Total Solids Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. Alkalinity concentrations in the heat Check lot are significantly greater than all other source locations, Figure 10. This supports the findings that the heat check lot is largely affected by the manure concentration as liquid dairy manure has an alkalinity of over 4,000 mg CaC03/L (Debusk et al. 2007). All other source locations and basins did not produce statistically different mean concentrations. 44 1000 800 -— Alkalinity (mglL as CaC03) B— 0 I T “ I I ._ Heat Check Roadway Storage Basin Upright Silo Bunker Silo Storage Basin Lot 1 2 Sampling Location Figure 10: Average Alkalinity Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. Heat check lot values produced an alkaline average pH value of 8.35 due to manure concentrations. Acidic pH concentrations were produced from feed sources, Figure 11. Unlike previous parameters, storage basin 1 was significantly different from both the heat check lot and the roadway. The two parameters combined to produce a composite concentration that was between the two mean values from the source locations, and was not impacted to the degree other parameters were from runoff quantity and dilution. Storage basin 2 is significantly different from the upright silos but not from the bunker silos, 45 revealing the composite pH is affected directly by the low acidity from the bunker silos. This low pH poses potential problems to biological treatment, however eliminates E. Coli at a pH below 5 (which was common in storage basin 2) but did not fully eliminate all Coli forms. 9 a Ii... 7 _. t , “ is“ W ” B 6 | w B ..,,,. (i. x . b ”2““ F 033‘ pH l . :- a. 1.4,, __4_ __m__ _ 0 *. a}; it“! we? Heat Check Lot Roadway Storage Basin Upn'ght Silo Bunker Silo Storage Basin 1 2 Sampling Location Figure 11: Average pH Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. Ammonia and TKN concentrations were greatest in the heat check lot, with organically bound nitrogen and ammonium as the majority of the total nitrogen in the system. Statistically, the heat check lot mean for ammonia and TKN is significantly different from the roadway and storage basin 1, with two times the 46 concentration of ammonia than any other source, but again not contributing significantly to the composite sample, Figure 12. Storage basin 2 in this case is governed by the upright silo as the concentrations within the basin are statistically similar to those from the upright silo. The bunker silos do not produce as high of concentrations of ammonia in runoff as the upright silos, but do not reduce the composite concentrations within the basin. 70 60 B 50 i—Ic "‘ z ,. ‘I 8 4 3 -L_ B 40 at ”I E ...- .2 T. A‘ g 30 J—‘i. '4 it ‘7 _ 20 --——"‘ ., BC — - B 10 r— - —— —— —— — B 0 r T I ' ‘ Heat Check Roadway Storage Basin Upright Silo Bunker Silo Storage Basin Lot 1 2 Sampling Location Figure 12: Average Ammonia Concentrations. Sampling locations with the same letter are not statistically significant at an alpha value = 0.05. Storage basin 2 concentrations for TKN are governed by the bunker silo concentrations as they are statistically similar, Figure 13. 47 200 180 140 -— 120 ~— 100 -— TKN (mglL) 80 -— —— 40-——————————————— 20 ——————————————————— o I I I I I I Heat Check Roadway Storage Basin Upright Silo Bunker Silo Storage Basin Lot 1 2 Sampling Location Figure 13: Average TKN Concentrations. Sampling locations with the same letter are not statistically significant at an- alpha value = 0.05. Average metal concentrations from each source can be found in Table 6. Manganese concentrations were two times as great from the heat check lot and bunker silo than other locations. Arsenic concentrations are below water standards for all sources except the upright silos which have an average concentration of 11.8 ug/L, above the 10 ugll. US EPA drinking water maximum (US EPA 2009b). 48 4.1.1. Summary and Management Strategy In summary, the heat check lot produces the largest concentrations for nearly all water quality parameters (Table 6) suggesting that animal waste on farmstead operations is the leading source for water quality issues. However, when examining composite samples in the storage and sedimentation basins that take into account water quantity in addition to water quality, feed sources are a greater concern at this farm. The footprint of the heat Check lot is 5000 sq ft, bunker silos are 3100 sq ft, and upright silos 11,350 sq ft. The remaining farmstead area is impervious roadways. The heat check lot is 9% of the storage basin 1 drainage area. In regard to basin 2, the bunker silo is 23% and the upright silos 6% of the drainage area. If farmstead manure locations are at or below 9% of the drainage area and the remaining area has low concentrations similar to that of the roadway area, then management requirements for animal waste sources are low. The greater footprint of the silos in addition to the high pollutant concentrations make the feed sources a greater focus for farmstead maintenance TBSOUTCGS. On-famt management practices should focus on feed sources to limit the impact of runoff on water quality. In addition, runoff from upright silos contains higher pollutant loads in the fall in comparison to the spring due to filling practices. The fall months produced concentrations that are 10 times greater than those measured in the spring (data not shown). Note that the concentrations listed in Table 6 are the averages for all seasons. Properly loading silage, including 49 harvesting at the correct temperature and moisture, rapid filling, and proper compaction, can improve silage quality and reduce leachate production (Saxe, 2007). Bunker silo runoff in general produces greater pollutant loads than upright silos (if the loading of the upright silos is managed properly). To minimize this impact, it is important to cover bunker silage prior to precipitation events, sweep impervious areas around feed sources, and maintain feed faces. On typical farm operations when manure may be a larger source of concern, additional steps should be taken in combination to those listed above for feed sources. If possible, manure should be covered and/or ben'ns or curbs provided to limit transport. Increasing vegetation in drainage areas with overland flow can also decrease transport of pollutants to treatment systems. Installation of gutters for diversion of Clean water will reduce the volume required for treatment, therefore reducing the size and cost of implementation. Lastly, care should be taken to eliminate dry weather leachate and ensure farmstead operations other than runoff are not increasing the load to treatment systems. 4.2 Analysis of Field Treatment Systems Application of farmstead runoff and treatment evaluation of 3 full-scale agricultural filter strips was conducted from 9/08/2009 to 6l24/2010. Ten sampling events were completed at 2 filter strips located at the MSU dairy ranging from 005-124 inches of rainfall, an additional 6 sampling events were investigated at the small Ml dairy ranging from 0.04-1.71 inches of rainfall. 50 Surface water and subsurface effluent were measured to determine percent removal for 17 water quality parameters. All raw filter strip data can be found in Appendix D. 4.2.1 BOD BOD removal for all three filter strips was relatively equal for surface water and subsurface effluent, Table 7 & 8. The greater differences within the removal percentages at the MSU site were negligible as variation between samplings was large as indicated by the standard deviation. lnfluent loading to filter strip 1 ranged between 35-270 lbs BOD5/acre/rain event and loading to filter strip 2 between 140-910 lbs BOD5/acrelrain event. Note the surface removal percentages for the small MI dairy represent reduction percentages from the bioretention basin. Table 7: BOD5 Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averagj Std Dev Max Min Average Std Dev Max Min Surface 37% 15% 54% 24% -4% 14% 5% -20% 1 ft 28% 19% 45% 2% —6% 36% 20% -31% Table 8: BOD; Percent Removal - Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 72% 4% 76% 68% 1 .5 ft 89% 1 % 91 % 88% 2.5 ft 79% 5% 87% 76% 51 Influent concentrations were similar for the small Ml dairy site and for filter strip 2 at the MSU site, both with average concentrations of 1300 mglL BOD5, whereas filter strip 1 at the MSU site had a much lower influent average of 230 mg/L BOD5. This resulted in loadings of The difference in infiuent concentrations is due to dilution of wastewater at filter strip 1 and build-up of contaminants on impervious surfaces at the other 2 locations. Final 8005 subsurface effluent concentrations were similar for filter strip 1 and the small Ml dairy filter strip at ~150 mglL. Filter strip 2 had an average subsurface effluent of almost 2500 mglL BOD5, and increase in concentration. The sandy soil at the small Ml dairy site which results in a reduction in the soil water holding capacity and an increase in oxygen diffusion is hypothesized as the cause for the increase in oxygen content and the greater pollutant reduction. 4.2.2 COD COD removal at the MSU dairy site was even less than that for 3005, Table 9. Average influent concentrations were again similar for the MSU filter Strip 2 and the small Ml dairy filter strip at 4700 mglL COD and 4400 mglL respectively. The first MSU filter strip had reduced influent COD concentrations at ~450 mglL. Higher COD concentrations as compared to BOD5 concentrations indicate the presence of recalcitrant carbon, most likely cellulose or lignin materials (Nielsen 2003) 52 Table 9: COD Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Max Min Average Std Max Min Dev Dev Surface 8% 30% 70% -19% -38% 54% 29% -109% 1 ft 18% 17% 49% 1% 4% 33% 57% -30% The COD subsurface effluent at the small Ml dairy Site sustained removal percentages above 70% for all sampling events, and performed much more consistently with reduced standard deviations, Table 10. Again, the greater removal rates are thought to be due to the sandy soil and increased porosity which increases oxygenation and diffusion rates, the decrease in soil moisture holding capacity, and the reduced infiuent flow rates. Table 10: COD Percent Removal — Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 59% 31% 87% —1 % 1.5 ft 85% 9% 96% 70% 2.5 ft 86% 8% 97% 78% Effluent COD concentrations at the MSU dairy site have a linear trend when plotted as a function of influent COD concentrations, resulting in increased effluent concentrations when influent concentrations increase. The small Ml dairy data does not follow this same linear trend. In the case of the sand soil, the effluent COD concentrations remain relatively constant for all COD influent concentrations, Figure 14. 53 10000 9000 P 8000 7000 S 6000 E o MSU Dairy F51 O 8 5000 I MSU Dairy r52 —- ‘0 C . g 4000 ASmaIl Ml DaIry 1.5 ft 5 a X Small Ml Dairy 2.5 ft 3000 2000 A I mm 0' J! * x o p r T T I I I ‘l o 2000 4000 6000 8000 10000 12000 14000 Influent COD (mg/ L) Figure 14: COD Effluent Concentrations as a Function of lnfluent COD Concentrations The trends within this data indicate that there is a something rate limiting at the MSU site resulting in the increased effluent concentrations. The small Ml dairy had greater removal rates for soluble COD, Table 11 & 12, as compared to the MSU filter strips, consistent with the BOD5 and COD removal. Soluble pollutant concentrations are more difficult to remove as indicated by the literature review in previous sections and the small Ml dairy removal percentages for COD and soluble COD. The MSU dairy filter strips performed poorly and 54 inconsistently for both COD and soluble COD with removal percentages below 30% for all locations. Table 11: Soluble COD Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -2% 35% 48% -70% -21% 24% -2% -52% 1 ft 26% 15% 46% 7% 15% 35% 63% -29% Table 12: Soluble COD Percent Removal — Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 23% 24% 45% -2% 1.5 ft 62% 25% 76% 34% 2.5 ft 59% 7% 65% 51% 4.2.3 Nitrogen TKN concentrations are the sum of organically bound nitrogen, ammonium, and ammonia. Nitrogen that originates as organically bound nitrogen must undergo ammonification to convert organic nitrogen to inorganic forms. These inorganic forms can then undergo the process of nitrification in aerobic conditions followed by denitrification under anaerobic conditions to exit the treatment system as nitrogen gas. Reductions in TKN were not constant for the MSU dairy locations, but were consistently above 80% at the site, Table 13 & 14. 55 Table 13: TKN Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface 18% 14% 35% 5% -7% 36% 23% -54% 1 ft 32% 16% 52% 10% 20% 30% 61% -19% Table 14: TKN Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 51% 11% 62% 34% 1.5 ft 87% 2% 89% 85% 2.5 ft 83% 10% 94% 67% Again the effluent TKN concentrations at the MSU dairy farm are dependent upon the influent concentrations, as represented by a general linear trend. The small Ml dairy maintains similar effluent concentrations regardless of infiuent concentrations, Figure 15. These characteristics hold true for COD and TKN as shown, but also hold true for Ammonia, TOC, arsenic, and solids (data not shown). 56 250 I 200 O MSU Dairy F51 —— I MSU Dairy r52 2. A Small MI Dairy 1.5 ft " 150 —— E X Small Ml Dairy 2.5 ft ‘5 i5 E g 100 m E a \1 X A O A e. A Q X w 0 T T r 1 0 50 100 150 200 250 300 350 400 lnfluent TKN (mg/L - N) Figure 15: Effluent TKN Concentrations as a Function of lnfluent TKN Concentrations Initially, 40% of the TKN within the MSU treatment systems is in the form of ammonia, this fraction increases over the surface of the soil then reduces again as the runoff infiltrates through the soil profile. This indicates ammonification is not occurring at the same rate as conversion of ammonia to other nitrogen forms. The small Ml dairy also has 40% of the total TKN as ammonia, and increases steadily as the runoff moves over the surface and infiltrates. Reductions within the first foot of the soil indicate there may be sufficient oxygen for at least a portion of the ammonia to go through the nitrification process. Limited reduction in filter strip 2 in comparison to filter strip 1 can be a source of the decreased pH 57 in filter strip 2 and the high levels of ammonia over 20 mglL, which are reported as inhibitory for NOz-N oxidation, all reducing nitrification rates and leading to nitrite build-up (T chobanoglous et al. 2003). Table 15: Ammonia Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -2% 32% 33% -56% -52% 55% 1 1 % -139% 1 ft 30% 37% 78% -29% 22% 22% 49% -17% The small Ml dairy site saw a greater removal of ammonia within the system, on the surface and within the soil, Table 16. This supports the hypothesis that the sandy soil (and the associated mechanisms which decrease water content) and reduction in influent flow in comparison to the MSU site increases the available oxygen therefore increasing nitrification. At a depth of 2.5 ft however, there is a slight decrease in the removal as average ammonia concentrations climb from 14 mglL-N at 1 ft to 26 mglL-N at 2.5 ft indicating oxygen may become limiting at an increasing depth. Table 16: Ammonia Percent Removal - Small Ml Dairy Filter Strip Percent Removal Averagg Std Dev Max Min Surface 41% 13% 62% 31% 1.5 ft 84% 6% 94% 76% 2.5 ft 73% 13% 93% 57% Nitrite concentrations increased for all filter strips examined within the study resulting in negative removal percentages, Table 17 & 18. Increases in nitrite 58 concentrations indicate conversion of ammonia to nitrite, but the increases confirm that the conversion from nitrite to nitrate is not occurring at the same rate. The high level of accumulation indicate oxygen limiting conditions at the MSU dairy site as Nitrobacter is more effected by low dissolved oxygen concentrations, resulting in increases nitrite concentrations (Tchobanoglous et al. 2003). The greater increase at the second filter strip site is likely due to the low pH levels which inhibit denitrification (Sahrawat 2008; Yue—Mei et al. 2008; Tchobanoglous et al. 2003 Table 17: Nitrite Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Max Min Average Std Max Min Dev Dev Surface -950% 2301% 93% -5645% -1494% 2661% 80% -4567% 1 ft -649% 1530% 59% -4107% -1203% 1347% -63% -2689% Table 18: Nitrite Percent Removal — Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 89% 19% 100% 67% 1 .5 ft -20% 19% -6% -33% 2.5 ft -16% 118% 68% -100% Nitrate concentrations Show a large variability, Table 19 & 20. Final average nitrate concentrations after infiltration are 11 mglL-N for filter strip 1, 45 mglL-N for filter strip 2 at the MSU site and 25 mglL-N at the small Ml dairy. These pose human health concerns and exceed the US EPA drinking water standard of 10 mglL (US EPA 2009b). 59 Table 19: Nitrate Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface 1% 73% 78% -86% -208% 446% 53% -1 108% 1 ft 26% 37% 62% -17% -164% 178% 63% -413% Table 20: Nitrate Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface -105% 212% 29% -350% 1.5 ft 4% 83% 75% -131% 2.5 ft 36% 43% 78% -32% Nitrification and denitrification are occurring within the soil as removal of the sum of all nitrogen species in all forms from runoff after infiltrating the soil profile varies from 25% to 80% for the full-scale system. Nitrate concentrations pose potential problems for groundwater contamination and may impact the implementation of this practice if improvements on removal cannot be made. 4.2.4 Phosphorus Average influent phosphorus concentrations are 11 mglL for filter strip 1, 24 mglL for filter strip 2, and 94 mglL for the small Ml dairy. The greater concentrations at the small Ml dairy site are much larger than reported literature values (Table 1) and are due to the large sources of animal waste and lack of containment or maintenance to control runoff from these sources. Removal rates for phosphorus are low for the MSU dairy, Table 21. The average concentrations for the subsurface effluent are the same as the influent concentrations at the MSU dairy 60 site. As assimilation in soil is the main mechanism for phosphorus removal, it was theorized that removal for this parameter would be negligible over time. Table 21: Phosphorus Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -89% 169% 1 1% -422% -75% 76% -16% -185% 1 ft -5% 52% 35% -126% -25% 71 % 48% ~158% The small Ml dairy has significantly greater removal with average subsurface runoff concentrations of 19 mglL, Table 22. The assimilative capacity of this location will become exhausted over time as was seen with the increasing phosphorus concentration trend in the infiltrate of the MSU dairy filter strips over only a year of sampling (Appendix D). Table 22: Phosphorus Percent Removal — Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 62% 27% 88% 1 1% 1.5 ft 88% 12% 100% 66% 2.5 ft 80% 21% 100% 55% 4.2.5 Solids Removal of total solids at the MSU site did not occur. Solids within the surface runoff and the subsurface samples were typically greater than the sampled basins, Table 23. Solids in the basin were allowed to settle prior to sampling which reduced the solids concentration within the basin samples. But, when the pumps were activated, this stirred the sediment located within the basin. There 61 was however, solids settling within the storage basins as the solids had to be removed numerous times throughout the research period. Surface samples also collected sediment from solids which have settled from previous applications and within the rock checks where the surface samples were collected. Table 23: TS Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -88% 88% 4% -209% -26% 8% -18% -36% 1 ft -27% 31 % 3% -62% -38% 29% -1 % -79% The small Ml dairy had increased removal for total solids within the soil profile as the soil acted like a filter, Table 24. This may be attributed to the small settling basin that was not mixed and a bioretention basin where the runoff is forced to flow through the soil profile in this unit, 41% of the total solids are removed prior to filter strip application. Table 24: TS Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 41 % 42% 83% -21% 1.5 ft 67% 14% 84% 50% 2.5 ft 67% 19% 88% 40% Average influent concentrations for VS are 320 mg/L for filter strip 1, 2420 mglL for filter strip 2, and 2490 mg/L for the small Ml dairy site, indicating nearly 50% of the total solids are volatile indicating a large volume of organic material. 62 Removal for volatile solids follows the same trends as that for total solids, Table 25 & 26. Table 25: VS Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface 0% 19% 25% -19% -27% 21% -3% -58% 1 ft -183% 237% 7% -521% -13% 21% 15% -31% Again greater removal is realized in the small MI dairy as 60% of the VS are removed after the bioretention basin. Table 26: VS Percent Removal — Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 60% 20% 81% 36% 1.5 ft 78% 14% 92% 61% 2.5 ft 82% 7% 92% 72% Total suspended solids account for less than 20% of the total solids for all filter strips. TSS and VSS behave in a similar manner as TS and VS, Tables 27 — 30. However, TSS and VSS removal increased with increasing depth. Due to the great increase in the surface TSS and VSS, even the first MSU filter strip subsurface samples (which have negative removals) represent a decrease in the TSS and VSS concentrators. 63 Table 27: TSS Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -219% 313% 40% -763% -31% 74% 74% -129% 1 ft -22% 41 % 22% -89% 26% 17% 49% 13% Table 28: TSS Percent Removal — Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 78% 20% 96% 57% 1 .5 ft 89% 10% 98% 75% 2.5 ft 92% 6% 98% 84% Table 29: VSS Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -61% 1 11% 63% -200% -26% 85% 77% -122% 1 ft -17% 47% 27% -83% 10% 42% 40% -50% Table 30: VSS Percent Removal - Small Ml Dairy Filter Strip Percent Removal Averagg Std Dev Max Min Surface 72% 24% 94% 48% 1.5 ft 92% 1 1% 98% 78% 2.5 ft 92% 8% 98% 82% Solids removal was significantly improved by the addition of the bioretention area at the small Ml dairy. This is also reflected in phosphorus removal as phosphorus is commonly sediment bound, and the greater sediment removal results in a significant reduction in phosphorus concentrations. Forty percent of 54 the TS and 78% of the TSS were removed by the bioretention basin which is reflected in the 62% decrease in phosphorus after the bioretention basin. 4.2.6 Alkalinity/pH The average alkalinity values for the influent and effluent are 200 mg/L and 250 mglL, respectively for filter strip 1. Although the alkalinity for filter strip 1 increases (as shown by the negative removal), the standard deviation and average values indicate that there was only a slight increase in alkalinity, Table 31. The second filter strip at the MSU site has larger influent concentrations at 225 mg/L and a more significant increase indicated by the negative removal. Table 31: Alkalinity Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averagi Std Dev Max Min Averag: Std Dev Max Min Surface -1 1% 23% 14% -46% -136% 196% -1% -513% 1 ft -32% 52% 8% -152% -277% 403% 0% -962% A decrease in alkalinity for the small Ml dairy site, Table 32, is indicative of the increased nitrification and denitrification processes. Nitrification decreases alkalinity while denitrification increases alkalinity by half of the nitrification process resulting in a net decrease (T chobanoglous et al. 2003). 65 Table 32: Alkalinity Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 30% 21% 51% 2% 1.5 ft 54% 23% 74% 1 1% 2.5 ft 60% 12% 71% 39% Filter strip 1 at the MSU site had very little change in pH between sampling locations as both the average influent and effluent pH values were 6.7, Table 33. Filter strip 2 had a slight average increase from 5.5 to 6.0. However, the second filter strip commonly had acidic pH values between 4 and 5. These low pH values pose problems to biological treatment and caused burning of vegetation at the top of the filter strip. Limiting the dry weather leachate in the spring by effectively managing the upright silage filling processes can reduce the problem of low pH values. Table 33: pH Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averag: Std Dev Max Min Average Std Dev Max Min Surface -2% 2% 0% -5% -6% 7% 2% -20% 1 ft 1% 2% 5% -3% -9% 12% 0% -37% A slight decrease from an average pH of 8.1 to 7.3 occurred at the small MI dairy. These concentrations pose no issues for treatment practices and require no management or treatment operational changes. 66 Table 34: pH Percent Removal — Small Ml Dairy Filter Strip Percent Removal Averagi Std Dev Max Min Surface 9% 3% 15% 5% 1 .5 ft 10% 5% 14% 4% 2.5 ft 10% 7% 22% 5% 4.2.7 Metals Manganese concentrations from the influent concentrations are increased drastically in the subsurface effluent as the Mn within the soil profile leached into the subsurface samples, Table 35. The average influent and subsurface effluent values for filter strip 1 at the MSU site are 200 ug/L and 665 ug/L over a 3x increase, and 555 uglL and 2550 uglL for filter strip 2, nearly a 5x increase. Table 35: Mn Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Max Min Average Std Max Min Dev Dev Surface -118% 155% 78% -372% -158% 190% 56% -417% 1 ft -343% 814% 81% -2311% -375% 304% -12% -783% The small Ml dairy follows the same trend, Table 36, but to a lesser degree as sand soils have lower initial Mn concentrations and are theorized to have increased oxygen availability (due to decreased moisture holding capacity) as discussed previously. The average Mn influent concentration is 1370 uglL and the effluent is 3650 ug/L, a 2.5x increase in Mn concentrations in the subsurface 67 effluent. Greater average effluent concentrations are theorized due to the higher average infiuent of soluble Mn concentrations in comparison to the MSU Dairy. Table 36: Mn Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 13% 33% 50% -35% 1.5 ft -27% 67% 42% -147% 2.5 ft -8% 76% 81 % -104% Average Fe concentrations did not increase in the subsurface samples, Table 37. Fe influent concentrations at the MSU dairy are 1670 ug/L and 6720 uglL for filter strip 1 and 2 respectively. These averages decrease to 1430 uglL and 4830 ug/L for subsurface samples. This is in accordance with the electron tower as Mn is continuing to serve as the electron donor and Fe will not leach within the soil profile until Mn is exhausted. Table 37: Fe Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -347% 594% 54% -1534% -1 14% 230% 49% -510% 1 ft -2% 68% 65% -139% 27% 22% 64% 2% The general decrease in Fe at the small Ml dairy follows the same trend as the MSU site, Table 38. 68 Table 38: Fe Percent Removal — Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 30% 53% 72% -62% 1.5 ft 43% 27% 81 % 22% 2.5 ft 58% 30% 86% 10% Arsenic concentrations for filter strip 1 and 2 have influent values 1.5 uglL and 3.7 ug/L and subsurface effluent values of 2.8 uglL and 8.4 ug/L. Although there are slight increases, the averages are still below the US EPA drinking water standards of 10 ug/L (US EPA 2009b). The increase in the subsurface samples is due to the increase in the surface samples prior to infiltrating the soil profile, average As concentrations at the surface are 6.2 uglL and 10.2 uglL, which indicates that the soil profile reduces the As concentrations within the soil profile. Table 39: As Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averagg Std Dev Max Min Averag: Std Dev Max Min Surface -297% 298% -1 1% -737% -153% 83% -57% -253% 1 ft -76% 80% -1 1% -225% -1 19% 53% -19% -185% lnfluent, surface and subsurface concentrations for the small Ml dairy are 26.7 ug/L, 35.3 uglL, and 16.3 ug/L, respectively. The increase in As concentrations at the surface level and the decrease in the subsurface is in agreement with the MSU site, as the soil profile is assimilating a portion of the As. However, unlike the MSU site the final concentration of As is of concern as it is above the EPA 69 drinking water standards. The increase is thought to be due to the addition of excess plate cooler water to the settling basin as high concentrations in groundwater have been reported in Michigan and in this area (MDEQ 2006; Myoung-Jin 2002). Table 40: As Percent Removal — Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 15% 23% 48% -13% 1.5 ft 52% 8% 62% 41% 2.5 ft 59% 16% 88% 45% 4.2.8 TOC Decreases in organic carbon occurred for all filter strip subsurface samples, Table 41 and 42. Table 41: TOC Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averafige Std Dev Max Min Average Std Dev Max Min Surface 7% 24% 39% -27% -21% 24% 15% -42% 1 ft 16% 20% 41 % -21 % 8% 30% 60% -21% Total removal is greater within the small Ml dairy filter strip, again attributed mainly to the soil type and increased oxygen concentrations. Table 42: TOC Percent Removal - Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 58% 13% 71 % 35% 1.5 ft 84% 4% 87% 76% 2.5 ft 82% 11% 94% 62% 70 4.2.9 Chloride Removal rates for chloride were negative, resulting in a net increase in average concentrations. Table 43: Chloride Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Average Std Dev Max Min Average Std Dev Max Min Surface -35% 71% 16% -176% -28% 37% 16% -71% 1 ft -135% 300% 20% -856% -1 1 % 28% 26% -41% At the small Ml dairy the final effluent concentrations for CI' decrease as the effluent moves through the soil. Removal is achieved mostly within the bioretention basin. Chloride concentrations remained below the US EPA secondary maximum contaminant limit of 250 mg/L for all subsurface filter strip samples. Table 44: Chloride Percent Removal - Small MI Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 47% 18% 64% 20% 1.5 ft 57% 17% 73% 28% 2.5ft 63% 9% 70% 49% 4.2.10 Conductivity Conductivity can be representative of salts and soluble nutrients. This follows the same trend for chloride indicating an increase in salts within the soil profile and an increase in soluble nutrients as wastewater enters the soil. For filter strip 2 there is a general increase in conductivity over time even with generally 71 consistent influent concentrations, indicating a build-up of salts and soluble nutrients within the soil treatment system. However, data from filter strip 1 and the small Ml dairy do not follow this same trend. Table 45: Conductivity Percent Removal - MSU Dairy Filter Strip Filter Strip 1 Filter Strip 2 Percent Removal Percent Removal Averagg Std Dev Max Min Averagg Std Dev Max Min Surface -17% 39% 9% -97% -26% 26% 3% -53% 1 ft -28% 43% 10% -112% -15% 33% 23% -61% The small Ml dairy showed a distinct reduction in conductivity indicating a decrease in salts as is supported by chloride removal. Table 48: Conductivity Percent Removal — Small Ml Dairy Filter Strip Percent Removal Average Std Dev Max Min Surface 41% 14% 51% 17% 1.5ft 59% 15% 69% 31% 2.5 ft 61% 7% 67% 49% 4.2.11 Cold Weather Performance Cold weather performance was evaluated at the MSU Dairy site only, and was found to have no effect on performance. However, performance at the MSU Site was poor in warm weather, so temperature may have a greater effect on a more efficient site. The main issues associated with cold weather performance are due to frozen ground leading to surface runoff commonly associated with land application of waste in the winter. However, this application has a number of differences that would limit the issues commonly faced in land application. The 72 settling basin within these systems can collect the runoff as it melts (as temperatures rise as compared with land application during cold winter months), and application will not occur until the gravity or pump system has thawed, requiring warmer temperatures that will inevitably increase the thaw in the filter strip subsurface. The gravity fed system has a pipe feed above the freeze point within the soil subsurface, requiring the ground to thaw to that depth prior to any runoff application in addition to the large bioretention subsurface that the water must flow through prior to reaching the treatment strip which would also require the subsurface soil to thaw. The MSU pump based system has a much larger capacity and the ability to manually operate pumps to determine the optimal time for application. Although there is a more limited concern for cold weather surface discharge, investigation at a more efficient site will be required to assess the microbial performance at reduced temperatures as their activity is known to decrease with decreasing temperatures. 4.3 Laboratory Evaluation of Treatment System Design Components Soil column effluent was assessed to determine the pollutant removal capabilities of the column experimental treatments. It should be noted that soil columns were investigated for 7 months, but did not reach a steady-state or equilibrium. This is important as removal and processes involved in removal are dynamic, and Changes in one water quality parameter concentration may invoke Changes in other parameters or mechanisms associated with removal within the soil column. 73 A number of soil and soil column characteristics were measured to evaluate the processes associated with removal of the water quality parameters discussed within this section. Physical characteristics include soil mechanical properties, bulk density and porosity, flow rates, and soil constituent concentrations. Flow rates for the 12 inch columns were initially more than 3 times the 30 inch columns and more than 10 times the 48 inch columns, Figure 16 and 17, not the differences in scale on the y-axis. 450 400 I 350 I, --I- 12 SL Sub “ - 0' ° 12 S g 300 e, +12 SL 2 250 I -o—30 SL Sub 3' ‘, —l—30 SL 1- \ g 200 1‘ g i g 150 * ‘r t\ ‘ \ 100 ‘ \ ‘. \ | 50 - 0 _ 0 5 10 15 20 25 30 35 40 Time (min) Figure 16: Soil Column Flow Rates - all 12 inch and 30 inch sandy loam columns. 74 - I- -30 S g unr- 48 S E g —o—48 SL g —o—48 SL Sub a: in IL 600 Time (min) Figure 17: Soil Columns Flow Rates — 30 inch sand and all 48 inch columns The 12 inch soil columns leached the entire wastewater volume within 20 minutes of application. Greater depths show a lag time in the increase in flow rates, and take over 10 hours to leach the entire volume of wastewater. An increase in depth decreases the flow rate from the column, increasing residence time resulting in greater contact time with the soil surface for increased adsorption and increased time for microbial metabolization. This increase is not linear with an increase in depth, indicating that the water not only has to travel through a greater depth of soil, but it also travels as a reduced rate of flow through the column. Soil characteristics dominate not only the hydraulic conductivities and flow rates, but also control the oxygen diffusion rates. Soil characteristics for the sand and sandy loam are in Table 47 below. 75 Table 47: Sand and Sandy Loam Soil Column Characteristics Parameter Sand Sandy Loam pH 8.8 6.9 P (ppm) 3 98 K (ppm) 8 133 Ca (ppm) 632 966 Mg (ppm) 198 198 Zn (ppm) 3.1 4.9 Mn (ppm) 4.4 13.9 Cu (ppm) 0.8 13.9 Fe (ppm) 8.1 44.7 Organic Matter (%) 0.3 2.3 Chloride (ppm) 61 59 Total N (ppm) n.d. 0.10 Nitrate-N (ppm) 0.6 1 1.0 Ammonium—N (ppm) 0.5 1.4 Sand (%) 93.5 69.8 Silt (%) 2.8 25.9 Clay (%) 3.7 4.3 Bulk densities for the sand and sandy loam soils were determined experimentally to be 1.55 and 1.65, respectively. Bulk density can then be used to calculate soil porosities, which were 42% and 38%. The decrease in soil porosity in the sandy loam is due to the reduced size of the clay and silt particles filling the void space of the sand particles. Sandy loam soils typically have a greater porosity than sand soils, but in this case the compaction of the sandy loam clays has a significant effect on the bulk density reducing overall porosity and pore size. It has been shown that oxygen diffusion rates are increased due to increases in air- filled porosity and decreases in soil water content and bulk density (Feng et al. 2002). This relationship indicates that the oxygen diffusion rates are less in the sandy loam soils compared to that of the sand soils as they have greater bulk 76 densities. In addition, as water fills the air voids within the soil, this also reduces the oxygen diffusion rates. This has implications for the columns of lesser depths as the ratio of wastewater volume to porosity volume is increased, Table 48, therefore increasing the overall soil pore column water content and reducing oxygen diffusion rates. Soils with decreased porosity require greater suction to remove soil pore water (Rose 2004), therefore sandy loam soils will retain more water when exposed to the same conditions as sand soils, a main factor in determining soil oxygen concentration. Soil moisture was measured every six inches in depth prior to deconstruction. Although the data was not precise enough to show the progression of the wetting front, it did show that sand soils had significantly decreased soil moisture after only a short time period as compared to the sandy loam columns. This again would have a positive effect on oxygen diffusion for sand columns. Average effluent volumes from the columns were also measured after each wastewater application, for columns of the same depth the sand soils had 15-25% more final effluent volume than the equivalent sandy loam soil columns. This is further evidence that the sandy loam soils retain more moisture than their sand counterparts. Table 48: Soil Column Volume and Porosity Depth Soil Volume Porosity Volume (mL) Raf: 82313322362; fright-3 (in) (in3) Sand Sandy Loam Sand SandyLoam 12 339 2335 2113 0.60 0.66 30 848 5838 5282 0.24 0.27 48 1357 9341 8451 0.15 0.17 77 Submerged columns also showed an increase in flow rates as compared to their non-submerged counterparts. When the columns are submerged, soil water rises within the capillary fringe due to pressure differences (Rose 2004), again increasing the soil water content and decreasing the soil oxygen content. The physical characteristics indicate that the columns of greater depth, those with sand soils, and those that are not submerged will have increased oxygen diffusion rates. In addition, columns of shorter depths have increased flow rates decreasing residence time and the associated soil mechanisms dependent upon contact time. 4.3.1 BOD The synthetic wastewater was designed to have a BOD concentration of 225 mg/L. Actual 3005 influent averaged 172 mg/L with a standard deviation of 19 mglL, but it should be noted that during analysis the results were commonly over the test range and therefore could not be included in the average resulting in a lower average than was actually realized. In addition, the design of the experiment used ultimate BOD concentrations, and 5-day BOD concentrations are always only a portion of the ultimate BOD. Column leachate was analyzed for B005, and as mentioned prior, multiple set-ups are required to cover a large range for each sample. The low end of the test detection limit was designed to be 6 mglL BOD5, of which many of the samples were commonly below. Column replicates were used for analysis of statistical design for all parameters, but graphs were made based on the averages of the three replicate columns for 78 each treatment combination. A graph of BOD5 concentrations can be found in Figure 18, all columns that did not produce any readings over 6 mg/L BOD were not included. +12 Sand 250‘ +12 Sandy Loam Sub ' . . ' -I-30 Sandy Loam Sub -O-12 Sandy Loam 200- +30 Sandy Loam - +lnfluent \/ BOD (mg/L) at 9 '8‘ Figure 18: Soil Column BOD5 Concentrations. Columns 48 inches long, regardless of soil type or submergence criteria did not produce effluent BOD5 concentrations over the detection limit of 6 mglL. The 30 inch depth columns with sandy soil and the control columns of both soil types also did not produce effluent concentrations over 6 mglL. Removal percentages for the 12 inch sandy loam columns, regardless of submergence, are in the mid- 50% range. This increases for the 30 inch sandy loam submerged columns to an average of 70% removal. The remaining columns average removal percentages are 90-99%. Consequently, the statistical model had significance for depth and soil but not for submergence because there was not a significant difference in 79 BOD5 concentrations for those columns that were and were not submerged, Table 49. Table 49: Soil Column BOD Statistical Model Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F Soil 1 13.4 11.79 0.0043 Depth 2 12.8 6.71 0.0101 Soil*Depth 2 12.8 6.31 0.0124 A'significant difference in effluent concentrations resulted for the 12 inch columns and 48 inch columns, Table 50, confirming the greatly reduced performance for the 12 inch columns. The shorter columns had decreased oxygen availability during the 20 min the wastewater traveled through the columns due to the greater ratio of wastewater volume to porosity. Statistical analysis also identified a significant difference between 3005 effluent concentrations in columns with sand and sandy loam. Those with sand had greater removal than the equivalent column with a sandy loam soil for all columns, Table 50. This is most likely due to the greater porosity and decreased soil moisture within the sand which increases oxygen diffusion rates contributing to the removal of oxygen demand in these columns. The soil*depth interaction effects revealed that columns with sandy loam soils had significant differences at all depths for BOD5 concentrations. The 12 inch 80 sandy loams columns (submerged and non-submerged) performed the poorest with an average removal percentage of 56%, increasing to 70% for the 30 inch sandy loam columns, and finally 94% for the 48 inch sandy loam columns. Again, the greater depths have a more favorable wastewater to porosity ration, increasing the available oxygen within the greater depths. Table 50: Soil Column BOD Differences of Least Squares Means - Comparisons of Significance Effect Soil 033;" Soil 035;“ Estimate Staff)?” DF tValue if Pr > In Depth I 12 48 35.1619 9.7862 A 8.25 3.59 0.0067 Soil Sand 32:: -28.2909 8.2397 13.4 -3.43 0.0043 Soil*Depth Sand 12 $2an 12 84.376 17.9825 7.78 -3.58 0.0075 Soil‘Depth Sand 30 $823 12 86.6267 16.0061 20.6 -4.16 0.0005 Soil’Depth Sand 48 fig: 12 65.8154 10.0546 8.8 -6.55 0.0001 Soil*Depth Sand 48 $223} 30 22.7544 8.5357 14.7 -2.67 0.0178 Soil*Depth €223 12 i223: 30 43.0611 10.3102 11.1 4.18 0.0015 Soil‘Depth $2an 12 fig: 48 68.8844 9.6512 7.91 7.14 0.0001 Soil‘Depth :2an 30 :32"? 48 25.8233 8.0567 14.5 3.21 0.0061 Other statistical comparisons of significance in the sand columns and the sandy loam columns can be found above, but the remaining effects all follow the same general trend, sand soils have greater reduction in BOD5 than sandy loam soils. Impact on treatment system design indicates that a depth of 48 inches, regardless of soil type, is sufficient for treatment of 8005 at these loadings or below. Sites with sandy soils require only 30 inches of depth for treatment to concentrations below 6 mglL. Both of these conditions results in B005 reading of 6 mg/L or less and do not pose a danger to groundwater resources. Surface discharge concentrations as determined by the Clean Water Act are 30 mglL 81 BOD (US EPA 2007), which were met by all columns greater than 12 inches in depth. 4.3.2 COD The synthetic wastewater applied to the soil columns had an average influent COD concentration of 325 mg/L and the water applied to the control columns had an influent concentration of COD 30 mg/L. Treatment performance for COD for sand and sandy loam is presented in Figure 19 & 20, respectively. The sand control columns performed similarly to the 30 and 48 inch sand columns, indicating that these sand columns did not produce a greater COD effluent with wastewater application than with the application of water. The 12 inch sand column, although follows the same trends as the other sand columns, was determined to be statistically significantly different from the 48 inch column only. 82 350 M 300 ~ 250~ a + Control 3 + 12 2, 200— -e- 30 ‘ E "3'48 8 150- + lnfiuent a 100 a 50 _ 00 Figure 19: Sand Soil Column COD Concentrations. The spike in the 30 inch sand column at week 8 was the result of one of the column readings used to calculate the average being significantly greater, 176 mglL, than all of the other readings. This is most likely due to an error in laboratory analysis as at no other time did the effluent for this column exceed 28 mg/L. Although here readings became slightly more erratic as experimentation continued, effluent concentration remained within the same general range. Examining the graph for sandy loam columns (Figure 16), the 12 inch columns, regardless of submergence, performed more poorly than all other columns. The control column responded similarly to the 30 and 48 inch sandy loam columns that were not submerged and the 48 inch column that was submerged. 83 350 r r r 1 W + Control 300* -a-1ZSub -l- 3OSub 250 - +48$ub A 4‘ -e- 12 3'200— 3, , ~. ,M -e-3o g i x ” I \O .& ...-'48“ t x / ~ + n uen 8150- ~. \V’ \‘x - o t ........ at. u—‘l O O I I, I \\ I I’I \ I I I ' r ‘\ I ‘\ U I \\ \\ \\ \ r. (I l ’l I, I, 4- I) I ’I I ‘1 I, «I f ' '4 f l j K I \ ‘7 \ L ._ F Soil 1 30.7 83.94 <.0001 Depth 2 37.5 55.69 <.0001 Soil*Depth 2 17.5 6.77 0.0067 Depth*Sub 2 19.4 13.98 0.0002 Time 9 73.9 3.95 0.0004 Soil*Time 9 74.1 4.4 0.0001 Sub*Time 9 71.6 3.53 0.0012 91 A statistically significant reduction in TKN concentrations exist for sand columns as compared to sandy loam columns. Depth was significant for TKN effluent concentrations as well, as each increasing depth has a greater statistically significant removal, Table 54. In comparing the interactions, the sandy loam soils had a significant difference between all depths for the submerged and non- submerged columns whereas there was no Significant difference for sand soils at 30 and 48 inch depths. The increased contact time of the longer columns is thought to play a distinct part in the microbial removal of organically bound nitrogen. Mineralization rates, or ammonification rates, are the greatest when aerobic microorganisms are dominant (Vymazal 2007). Sand columns are also theorized to increase oxygen diffusion which would have a significant effect on increasing the microbial activity. 92 Table 54: Soil Column TKN Differences of Least Squares Means - Comparisons of Significance Effect Soil Sub Depth Soil Sub Depth Estimate 5;; DF Vatlue Pr > M Soil Sand SL223 48741 0.532 30.7 -9.16 <.0001 Depth 12 30 5.1498 0.7739 46.8 6.65 <.0001 Depth 12 48 6.8569 0.7272 50.1 9.43 <.0001 Depth 30 48 1.7071 0.3164 26.9 5.39 <.0001 Séfi Sand 12 Sand 48 4.1987 1.2895 35.3 3.26 0.0025 5:; Sand 12 ”:22? 12 -8.5225 1.4478 36.2 -5.89 <.0001 5gb Sand 30 Sand 48 1.8633 0.5379 20.2 3.46 0.0024 Sgt“ Sand 30 :22? 12 40.858 0.8384 58.7 42.95 <.0001 5231.. Sand 30 $.22nt 30 -2.8938 0.5132 34.2 -5.64 <.0001 1385,18 Sand 30 $323: 48 4.3428 0.5103 17.2 -2.63 0.0174 531,1 Sand 48 SL231 12 42.7213 0.7524 50.5 46.91 <.0001 52:; Sand 48 :22? 30 47571 0.4067 14.4 41.7 <.0001 03:3,; Sand 48 :22? 48 -3.2061 0.432 4.12 -742 0.0016 3:8 $2233! 12 $223 30 7.9642 0.7602 58.7 10.48 <.0001 03:51" :22? 12 :22? 48 9.5151 0.7322 47.6 13 <.0001 548:; $322! 30 1:223 48 1.5509 0.368 10.5 4.21 0.0016 0:33" No 12 No 30 6.3796 0.768 37.5 8.31 <.0001 ”$3 No 12 $93 30 3.5603 0.8974 32.3 3.97 0.0004 0&3" No 12 No 48 6.127 0.7555 29.4 8.11 <.0001 0:33" No 12 2:3 48 7.2271 0.7758 31.3 9.32 <.0001 0:33“ 3:3 12 No 30 6.7394 1.2443 63 5.42 <.0001 033:1" 3&2: 12 31:3 30 3.9201 1.3281 58.6 2.95 0.0045 9:33" 3:13 12 No 48 6.4867 1.2366 57.4 5.25 <.0001 0:33" :33 12 2;: 48 7.5869 1.2486 58.6 6.08 <.0001 0:33" No 30 Egg 30 -2.8193 0.595 21.7 4.74 0.0001 Dag" No 30 fig: 48 0.8475 0.3786 16.1 2.24 0.0397 0:33“ 2:19er 30 No 48 2.5667 0.5725 14.3 4.48 0.0005 0:33" 22:3 30 3:323 48 3.6668 0.5991 16.6 6.12 <.0001 0:33" No 48 fig: 48 1.1002 0.4334 4.17 2.54 0.0615 93 Ammonia concentrations followed similar trends to that of TKN, Figure 25. lnfluent concentrations for ammonia were 25.8 mglL-N for the synthetic wastewater and 2.6 mglL-N for the water applied to the control columns. Sandy soils produced removal percentages of 85% for the 12 inch columns and 97% for the 30 and 48 inch columns, with final average ammonia concentrations of 3.8 mglL-N for the 12 inch columns and under 1 mglL-N for the sand columns of greater depth. Control sand columns produced similar effluent concentrations to the 30 and 48 inch sand columns which received wastewater, below 1 mglL-N. 35 30— - ,‘25— — "‘7 ..l 320— 4 m 'E +12 ‘5’ 15 -o—-30 _ E 548 10— +|nfluent — Figure 25: Sandy Soil Column Ammonia Concentrations. The sand columns had lower concentrations than sandy loam columns of equal depths as removal of ammonia is an aerobic process and the sand columns are thought to have increased oxygen diffusion rates and remain saturated for a reduced amount of time increasing nitrification rates, Figure 26. 94 30— a A255 - 2. 5’, £20“ ‘ .g +3OSub .4 015- +48Sub - _ = “ g ...12 _ r " - 10r +30 “I V ‘, _ +48 4’ a - *Influent / - at 5r \.-_ Wk- ' 5 \m / mm \-:_:'—:"‘—\—1 00 5 15 20 25 Week Figure 26: Sandy Loam Soil Column Ammonia Concentrations. The increase in ammonia concentrations within the 12 inch sandy loam columns could be caused by the increase in TKN therefore increasing ammonia, a decrease in the conversion of ammonia to nitrite, a reduction in binding sites within the 12 inch sandy loam soil columns or a combination of these mechanisms. It is most likely a combination of these mechanisms as a decrease in the available oxygen (due to decreased oxygen diffusion and increased soil pore water) would retard ammonification and nitrification rates leading to an increase in TKN and ammonia, and a decrease in the conversion of ammonia to nitrite both resulting in ammonia build-up over time. In the case that sorption plays a role, it is reasonable to assume that the 12 inch columns were not sufficient in providing the sorption Sites required to prevent ammonia within the effluent. 95 The sandy loam columns had average ammonia removal percentages of 60-63% for the 12 inch submerged and non-submerged columns, 84% for the submerged 30 inch columns, and 93-94% for the remaining columns. The final effluent concentrations were 9-10 mg/L for all sandy loam 12 inch columns, 4 mg/L for the 30 inch submerged and 1-2mg/L for the remaining columns. The sandy loam control column produced an effluent concentration below 1 mglL. The larger effluent concentrations for the 12 inch sandy loam columns suggest a lack of oxygen during the time when water was leaching though the column producing the greater concentrations within the effluent. The low concentrations within the columns of greater length indicate there was not a lack of oxygen required for nitrification of ammonia. This is further supported by previous research which has shown soil nitrification rates are dependent upon temperature, pH, available carbon, and aeration which is effected by soil moisture and structure (Barker et al. 2000, Paul and Clark 1996, Yue—Mei et al. 2008; Sahrawat 2008). In this case, temperatures fall between 14.2°C — 242°C near the optimal nitrification temperature of 25°C and within the range of 5°C - 40°C outside of which nitrification if inhibited (Sahrawat 2008; Paul and Clark 1996; Yue-Mei et al. 2008). Microbial nitrification requires a carbon source, typically 002' or carbonate (Paul and Clark 1996), which alkalinity concentrations indicate are in available in excess, section 4.3.5. The optimum pH range is 6.6 to 8.0 (Paul and Clark 1996; Tchobanoglous et al. 2003; Sahrawat 2008; Yue-Mei et al. 2008), of which all columns remain within the range for the duration of the experiment, 96 section 4.3.5. This leaves aeration, which is affected by soil moisture, as the only remaining impact factor on nitrification rates, which further verifies the observations based on the experimental data. Nitrification rates are known to be inhibited at dissolved oxygen concentrations below 2 mglL (Y ue—Mei et al. 2008). Statistical analysis support these observations with statistically significant differences for the main effects of soil and depth, with sand outperforming sandy loam in ammonia removal and increasing ammonia removal with increasing treatment depth. Interaction effects Show that there is no difference between the effluent concentrations for the 12 inch sand columns and the 30 and 48 inch sandy loam columns. There is no difference in treatment for sand soils with increasing depth, but there is a significant difference at all depths for sandy loam soils indicating that aeration required for nitrification was adequate in all depths for sandy soil, but not for the 12 inch sandy loam columns. Other significant interaction effects can be found in Table 55. 97 Table 55: Soil Column Ammonia Differences of Least Squares Means - Comparisons of Significance . Depth . Depth Std t Effect Sorl Sub (in) Sorl Sub CID Est Error DF Value Pr > M Soil Sand SL323 -2.7811 0.6451 10.9 4.31 0.0013 Depth 12 30 3.8341 0.8632 19 4.44 0.0003 Depth 12 48 5.7116 0.8442 18.9 6.77 <.0001 Depth 30 48 1 .8775 0.4582 49.4 4.1 0.0002 Soil‘ Sandy Depth Sand 12 Loam 12 -6.6274 1.6626 11.9 -3.99 0.0018 Soil" Sandy Depth Sand 30 Loam 12 -7.757 0.9111 31.2 -8.51 <.0001 Soil' Sandy Depth Sand 30 Loam 30 -1.2185 0.5739 74.1 —2. 12 0.0371 SO". Sandy Depth Sand 48 Loam 12 -9.2739 0.9457 18 -9.81 <.0001 Soil‘ Sandy Depth Sand 48 Loam 30 -2.7354 0.6866 15 -3.98 0.0012 Soil" Sandy Sandy Depth Loam 12 Loam 30 6.5385 0.8259 25.8 7.92 <.0001 Soil' Sandy Sandy Depth Loam 12 Loam 48 8.7766 0.845 14.6 10.39 <.0001 Soil' Sandy Sandy Depth Loam 30 Loam 48 2.2381 0.5395 8.81 4.1 5 0.0026 0:331 No 12 No 30 5.8851 0.8794 13.6 6.69 <.0001 Depth' Submer Sub No 12 9 ed 30 2.8741 1.0537 16.7 2.73 0.0145 0:3? No 12 No 48 6.0163 0.9241 9.66 6.51 <.0001 Depth' Submer Sub No 12 9 ed 48 6.4979 1.0142 11.8 6.41 <.0001 Depth‘ Subm Sub erged 12 No 30 4.7941 1.3207 25.7 3.63 0.0012 Depth’ Subm Sub erged 12 No 48 4.9252 1.3509 20.2 3.65 0.0016 Depth' Subm Submer Sub erged 12 ged 48 5.4069 1.4141 21.8 3.82 0.0009 Depth“ Submer Sub No 30 9 ed -3.0111 0.7199 40.8 4.18 0.0001 Depth' Subrn Sub erged 30 No 48 3.1422 0.763 15.4 4.12 0.0009 Depth‘ Subm Submer Sub erged 30 9 ed 48 3.6239 0.8699 19.8 4.17 0.0005 Typical ammonia surface discharge limits are 8 mglL, which were achieved by all columns except the 12 in sandy loam submerged and non-submerged. Filter strip design requires a depth of 12 inches for sand soils and 30 inches for sandy loam soils for removal of ammonia below surface discharge levels. 98 lnfluent nitrite concentrations in the wastewater and water for the control column are negligible. Nitrite concentrations in the sand columns remain close to or below the detection limit in accordance with the control column for all columns except the 12 inch, Figure 27. 0.7 T l l l l l l l l r I +Control 06— ° +12 _ -...30 -8. 48 A 0.5 -~- lnfluent 3 R a 0.4 '- i a A s i 1. 0 I t g 0 3’- q\\ i E '— 2 \\\\ A i 'R‘ ‘E 0 2 " \\\\ I, I,” ““ \‘ .‘ \‘\‘ "l ”I! \\\ “0‘ 9 0 1 — ~,\ H" \3‘ _. x‘ ’ “ " .......... .:~ _ _ ., 0 —— :— - ~ .2 _-_ --.. --.... 4 6 8 10 12 14 16 18 20 22 24 Week Figure 27: Sand Soil Column Nitrite Concentrations The sandy loam columns follow a similar trend except the spikes at weeks 10 and 14 which are a magnification of the spikes in the TKN and ammonia, Figure 28. In this case however, the effluent concentrations are increased by one order of magnitude in comparison to the sand columns for the 12 inch columns. 99 Nitrite (mglL-N) O) U 0" " +Control +128ub -l- 30Sub +488ub -O- 12 -9- 30 w 4-48 -- lnfluent —4 Figure 28: Sandy Loam Soil Column Nitrite Concentrations. A statistical model was fit to the nitrite data which was significant for the main effects of soil, depth, and the repeated measure of time with a number of interaction effects, Table 56. Table 56: Soil Column Nitrite Statistical Model Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F Soil 1 78.9 6.36 0.0137 Depth 2 64 4.01 0.0228 Soil*Depth 2 62.3 5.74 0.0052 Time 9 75.3 4.32 0.0001 Soil*Time 9 87.5 4.12 0.0002 Depth*Time 18 162 2.34 0.0026 There is a statistically significant difference in the nitrite effluent concentrations of the sand and sandy loam columns, Table 57. In terms of depth there is 100 significant difference between the 12 inch columns and the longer columns, but not between the 30 and 48 inch columns. Table 57: Soil Column Nitrite Differences of Least Squares Means - Comparisons of Significance . Depth . Depth . Standard Effect Soul (in) Sorl (in) Estimate Error DF 1 Value Pr > N Soil Sand fem" 0.3147 0.1248 78.9 -2.52 0.0137 oam Depth 12 30 0.5443 0.1914 77.8 2.84 0.0057 Depth 12 48 0.5292 0.19 76.3 2.79 0.0067 Soil*Depth Sand 12 $223 12 -0.9835 0.3683 75.5 -2.67 0.0093 ' w Sandy Soul Depth Sand 30 Loam 12 4 .0591 0.2189 79.3 -4.84 <.0001 Soil*Depth Sand 48 SL223 12 -0.9782 0.216 76.5 4.53 <.0001 Soil*Depth Sand 48 $223 48 0.08563 0.03574 34.9 2.4 0.0221 . ,, Sandy Sandy Soul Depth Loam 12 Loam 30 1.0129 0.2176 78 4.65 <.0001 . , Sandy Sandy Soul Depth Loam 12 Loam 48 1.0638 0.2157 75.9 4.93 <.0001 Nitrate is the final step in the nitrification process and can indicate if the rates of nitrification/denitrification are not in sync. lnfluent nitrate concentrations are ~1 mglL and unlike many other parameters in the sand columns increase with depth, Figure 29. Average effluent concentrations increase from 12 mglL for the 12 inch sand column to 27 and 24 mglL for the 30 and 48 inch columns respectively. Denitrification is inhibited by the presence of oxygen and low pH values (optimal pH from 6-8) (Paul and Clark 1996). Rates for denitrification, so long as organic carbon is available, require a water-filled pore space of 60—90% (Paul and Clark 1996). Values for pH are within the optimal range for denitrification, section 4.3.5, and there is available organic carbon as seen by the 101 BOD5, COD and TOC concentrations, indicating denitrification is again controlled by oxygen and water content. 3*. 30 h .\ ’fl‘ / \\ .4 x / \ t. ’/< \ \ ’,.~~~.. / Q-.. \‘ ‘ 4’ 4 ‘ ‘ 25- 6’ ‘~._....-‘4-o—-----e, x 0 I/ 1. fl ‘\ \‘ / A ’I A q \‘\ {a'l’ ‘\ j, 5 ,r ,e a, 20 ‘ ’/ ‘B" _ E O §15- _ E 10 -5.” 5~ +lnfluent — 0 i—F—W Week Figure 29: Sand Soil Column Nitrate Concentrations. The sandy loam columns follow the same trends as the sand columns, with an increase in nitrate with depth, Figure 30. In this case the average nitrate concentrations for the submerged sandy loam columns increase with increasing depth from 10 mg/L, to 19 mg/L, to 29 mg/L for the 48 inch column. Non- submerged column nitrate effluent increases from 16 mglL, to 19 mg/l, to 40 mg/L with depth. Nitrate conversion to nitrogen gas requires anaerobic conditions which have been evidenced to occur within the short columns only. This is thought to have caused the nitrate build-up within the longer column effluent. Indicating that the soil moisture requirements for denitrification are either not met or are not sustained. 102 88888 N 9 Nitrate (mglL-N) N 01 ‘\ v ...: ”A:.n...-+QM-. ”_-‘.-" ’ 15 — .- ‘t— C t ’ \ /” Figure 30: Sandy Loam Soil Column Nitrate Concentrations. The details for the statistical model for nitrate are in Table 58. Significant main effects for soil, column, and submergence were found. Table 58: Soil Column Nitrate Statistical Model Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value E[ >_ F Soil 1 184 14.44 0.0002 Dgath 2 224 33.06 <.0001 Soil*Depth 2 223 13.7 <.0001 Sub 1 189 25.79 <.0001 Time 9 265 4 <.0001 There is statistical significance between the soil types for nitrate effluent concentrations. Long columns of 30 and 48 inches are statistically different from the 12 inch columns. Sandy loam columns Show Significance for an increase in nitrate with increasing depth, Table 59. 103 Table 59: Soil Column Nitrate Differences of Least Squares Means - Comparison of Significance . Depth . Depth . Std t Effect Soul Sub (in) Sorl Sub (in) Estimate Error DF Value Pr > M - Sandy Sorl Sand Loam -9.1463 2.4071 184 -3.8 0.0002 Depth 12 30 -1 7.8048 2.4032 224 -7.41 <.0001 Depth 12 48 -19.9701 2.6394 220 -7.57 <.0001 Soil’ Depth Sand 12 Sand 30 -24.181 1 4.1512 221 -5.83 <.0001 5:3,; Sand 12 Sand 48 46.2928 4.4818 224 -3.64 0.0003 Soil" Sandy Depth Sand 12 Loam 12 -1 0.9456 4.5699 202 -2.4 0.0175 Soil* Sandy Depth Sand 12 Loam 30 -22.374 4.1479 216 -5.39 <.0001 Soil" Sandy Depth Sand 12 L oa m 48 -34.5931 4.8526 173 -7.13 <.0001 03:3,, Sand 30 Sand 48 7.8883 2.7565 271 2.86 0.0045 Soil“ Sandy Depth Sand 30 Loam 12 13.2355 2.906 228 4.55 <.0001 Soil’ Sandy Depth Sand 30 Loam 48 -10.412 3.339 161 -3.12 0.0022 Soil“ Sandy Depth Sand 48 Loam 30 -6.0813 2.7494 267 -2.21 0.0278 8011* Sandy Depth Sand 48 Loam 48 -18.3003 3.7416 178 -4.89 <.0001 Soil’ Sandy Sandy Depth Loam 12 Loam 30 -1 1.4284 2.4181 234 -4.73 <.0001 Soil" Sandy Sandy Depth Loam 12 Loam 48 -23.6475 2.8013 205 -8.44 <.0001 Soil* Sandy Sandy Depth Loam 30 Loam 48 -12.2191 2.8364 164 -4.31 <.0001 Sub Sub No mer 11.3549 2.2361 189 5.08 <.0001 ed 4.3.4 Phosphorus Complications with phosphorus testing methods result in validity in general trends only, not in concentrations. The general trends indicate removal percentages from 25-75%, but removal has a high rate of deviation and no conclusions can be drawn with confidence. However, phosphorus removal is based on adsorption within the soil profile so will theoretically have increased 104 removal for greater depths as the residence time for assimilation is greater, and there is more surface area within the larger columns for adsorption. 4.3.5 pH/Alkalinity The pH values for sand soils can be found in Figure 31 and Figure 32 for sandy loam soils. Average pH values for all treatment columns are between 7.3 and 8.0. All columns produce a neutral pH and pose no issues for treatment. 8.4 l l T T 8.2+ 7.8— 7.6~ pH 7.4L 7.2~ 6.8~ -a—48 d +Influent 66 l l 1 Week Figure 31: Sand Soil Column pH Concentrations. 105 8.4 if I I I +C0nm +12Sub 32" -I-3OSub 8 - +483ub -O- 12 7.8V +30 +48 7.6 b ; -- lnfluent E 7.4% 1 7.2% _ 7 - .. 6.8~ _ 6.6- _ 6'40 5 10 15 20 25 Figure 32: Sandy Loam Soil Column pH Concentrations. The statistical model for pH produced no statistical significance for all effects, indicating that the treatment means are not significantly different for pH. Alkalinity influent concentrations were measured at 336 mg/L as CaC03 for the synthetic wastewater and 340 mg/L as CaC03 for water applied to the control columns. Effluent from the sand columns produced average alkalinity values of 293 mg/L as CaC03, 239 mglL as CaC03, and 251 mg/L as Ca003 for increasing depths, which was similar to the control column with an average of 253 mg/L as CaCOa, Figure 33. 106 450 l l l l -I- Control -o- 12 400” -.. 30 " 8 -a- 43 8 -- lnfluent a 350— — o t In G .\ §300~ . , _ :E {II \ ; 5 250— Elf-”ax. \ — t9 , = z o 200» a 1500 5 10 15 20 25 Week Figure 33: Sand Soil Column Alkalinity Concentrations. Alkalinity concentrations for submerged sandy loam columns averaged between 303-335 mglL as CaCOa, the non-submerged from 262-294 mglL as CaC03, Figure 34. 107 3: o .b O ‘? ‘5 Alkalinity (mg/L as CaC03) 61’ 8 9 <7 8 150 Figure 34: Sandy Loam Soil Column Alkalinity Concentrations. The statistical model for alkalinity is shown in Table 60. Table 60: Soil Column Alkalinity Statistical Model Type 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F Soil 1 20.5 1.61 0.2186 Depth 2 63.6 6.83 0.0021 Soil*Depth 2 20.3 5.09 0.0162 Sub 1 20.6 15.6 0.0008 Depth*Sub 2 21.8 4.38 0.0252 Time 11 113 16.38 <.0001 Soil*Time 1 1 1 13 4.53 <.0001 There is a statistically significant difference between all depths and those columns that are submerged and not submerged, Table 61. 108 Table 61: Soil Column Alkalinity Differences of Least Squares Means - Comparisons of Significance . Depth . Depth . Std t Effect Sorl Sub (in) Sorl Sub (in) Estimate Error DF Value Pr > |t| Depth 12 30 32.9088 13.477 51.7 2.44 0.0181 Depth 30 48 41.1901 11.7213 101 -3.51 0.0007 53th Sand 12 Sand 30 70.0459 24.6058 40.6 2.85 0.0069 52',“ Sand 30 Sand 48 -72.1576 20.3979 40.8 -3.54 0.001 Soil“ Sandy Depth Sand 30 Loam 12 -56.5002 15.6832 74.9 -3.6 0.0006 Soil‘ Sandy Depth Sand 30 Loam 30 -60.7283 13.0973 207 4.64 <.0001 Soil' Sandy Depth Sand 30 Loam 48 -70.9509 16.2391 29.2 4.37 0.0001 Sub No 3;: 44.678 11.3128 20.6 -3.95 0.0008 Depth’ Subm Sub No 12 arsed 48 -70.0605 20.7665 24.8 .337 0.0024 Depth' Subm Sub arsed 12 No 30 54.4062 17.9676 48 3.03 0.004 Depth' Subm Subm Sub 9,99,, 12 fled 30 43.9307 22.2405 58.3 1.98 0.053 Depth‘ Subm Sub fled 12 No 48 53.4981 20.0274 22.9 2.67 0.0137 Depth' Subm Sub No 30 ”led 48 -919475 17.647 29.9 -521 <.0001 Depth“ Subm Subm Sub 9,39,, 30 aged 48 -81.472 21.9823 41.2 -3.71 0.0006 Depth' Subm Sub No 48 aged 48 -91.0394 21.8758 9.87 4.16 0.002 Larger decreases in alkalinity concentrations for columns which had increased nitrification is in agreement with nitrogen columns data as nitrification uses carbonate as a carbon source for aerobic metabolism. 4.3.6 Metals lnfluent Mn concentrations for the wastewater were 550 uglL, and 436 ug/L for the water applied to the control columns. Sand columns had reduction percentages of 63% for the 12 inch columns and 98% for the 30 and 48 inch 109 columns. Final Mn effluent concentrations for the 12, 30, and 48 inch column depths were 109 uglL, 5 ug/L, and 7 uglL, Figure 35. 1800 I r I ' 1600— — +Control 1400' +12 ‘ -o-30 12005 .5- 48 — -- Influent _ Mn (ug/L) ES 3 Figure 35: Sand Soil Column Mn Concentrations. Sandy loam submerged columns produced average removal percentages of -66%, 0%, and 17% with increasing depth, with final effluent concentrations of 608 mg/L, 304 ug/L, and 256 uglL, Figure 36. Leaching of Mn within the 12 inch columns confirms that the conditions within the 12 in columns were anaerobic, and greater depths maintained more aerobic conditions. In addition, the 12 inch column has greater effluent concentrations as the column is Short enough to allow leaching metals from the rest of the column to reach the bottom without re- oxidizing to an insoluble form. The non-submerged columns had improved removal percentages with depth from 25% for 12 inch columns to 46% and 85% for the 30 and 48 inch columns, respectively. Final average effluent 110 concentrations had values of 290 uglL, 150 ug/L, and 37 uglL reducing concentrations with increasing depth. 1800 t 16“). +COfltl'Ol +128ub 1400~ +3OSub n +48$ub 1200- -O— 12 3' 1000— +48 3 R, -- Influent C 2 5533 i f I | l Figure 36: Sandy Loam Soil Column Mn Concentrations. Initial soil Mn concentrations are 3x greater within the sandy loam soil than the sand soil, Table 47. This difference in initial soil concentration may contribute to the differences in subsurface effluent concentrations. Table 62 includes the details for the statistical model for Mn. The model produced Significance for the all levels of depth, between soils, and between submerged and non-submerged columns validating the differences noted above, Table 63. 111 Table 62: Soil Column Mn Statistical Model 'Lype 3 Tests of Fixed Effects Effect Num DF Den DF F Value Pr > F Soil 1 71.1 9.7 0.0027 Depth 2 64 13.69 <.0001 Sub 1 50.2 63.83 <.0001 Time 11 153 5.96 <.0001 Soil*Time 1 1 71.1 4.36 <.0001 Depth*Time 22 89.1 3.1 <.0001 Sub*Time 11 50.2 2.49 0.0141 Table 63: Soil Column Mn Differences of Least Squares Means - Comparisons of Significance Effect Soil Sub 0:3“ Soil Sub 035‘: Est 3:333” DF value Pr > [1] Depth 12 30 192.39 45.3612 101 4.24 <.0001 Depth 12 48 235.79 45.7017 109 5.16 <.0001 Depth 30 48 43.4007 19.4957 33.9 2.23 0.0327 Soil Sand :22: -75.916 24.3724 71.1 -311 0.0027 Sub No it": -208.5 26.0969 50.2 -7.99 <.0001 Soil Mn concentrations as a function of depth Show a decrease in Mn to a depth of 12 inches where the concentrations levels off, Figure 37. This decrease again supports the conclusions from the statistical analysis and effluent trends which Show a significant difference between the 12 inch columns and those deeper, again supporting the conclusion that the 12 inch soil column is not significant for treatment of Mn. This data also suggests that sorption may be Significant within the sand columns but does not appear to be within the sandy loam columns. 112 “E ‘D 1’ —o—30 - s E 5 -l—30 - s 20 -— 48 — S 10 0 4 j 0 10 20 30 40 50 Depth (in) Figure 37: MnSoil Concentration as a Function of Depth According to the electron tower, Fe is leached following Mn leading to the mobilization of iron when it is reduced from Fe(lll) to Fe(ll). Leaching of Fe before the complete leaching of Mn is due to the high levels of soluble Fe added as compared to the insoluble form of Mn. lnfluent average Fe concentrations for the synthetic wastewater and the tap water applied to control columns were 2050 ugll and 1600 ug/L. Sand columns reached removal percentages and final effluent concentrations of 37% and 677 uglL for the 12 inch columns, 94% and 63 uglL for the 30 inch columns, and 91% and 81 ugll for the 48 inch columns. Sand column at a depth of 12 inches performed much more poorly than those at 113 greater depths, Figure 38. Control columns performed similarly to the 30 and 48 inch columns. -l- Control + 12 -O- 30 ~8- 48 -- Influent Fe (ug/L) Figure 38: Sand Soil Column Fe Concentrations. Average Fe removal percentages for the submerged sandy loam columns for depths of 12, 30, and 48 inches are -50%, 38%, and 31 %. Final effluent concentrations for the submerged columns with increasing depth are 1631 uglL, 760 ugll, and 794 uglL. The non-submerged columns had increased removal rates of 23%, 44%, and 34% for the 12, 30, and 48 inch sandy loam columns with final concentrations of 1026 ug/L, 658 ug/l, and 778 ug/L, Figure 39. The average final effluent concentration for the control columns was 722 uglL, very similar to the 30 inch and 48 inch submerged and non-submerged columns. 114 + Control -a-1ZSub -l- 30Sub +488ub -O- 12 -e- 30 +48 -- Influent Fe (uglL) Figure 39: Sandy Loam Soil Column Fe Concentrations. The statistical model for Fe is in Table 64. Statistical analysis indicated there was a significant difference for the 12 inch columns with the 30 and 48 inch columns, but there was no difference between the 30 and 48 inch Fe effluent concentrations, Table 65. Statistical analysis also indicated a significant difference for soil type. Looking at the removal percentages and the final effluent concentrations it is apparent that the sand was more effective in reducing Fe leaching. Table 64: Soil Column Fe Statistical Model Type 3 Tests of Fixed Effects 1 Effect Num DF Den DF F Value Pr > F Soil 1 16.8 107.57 <.0001 Depth 2 21.4 8.67 0.0017 Time 11 117 2.55 0.0063 Soil*Time 11 312 13.61 <.0001 Depth*Time 22 202 2.91 <.0001 115 Table 65: Soil Column Fe Differences of Least Squares Means - Comparisons of Significance . Depth . Depth . Standard Effect Sorl (i n) Soul (in) Estimate Error DF t Value Pr > |t| Depth 12 30 645.41 153.76 82.2 4.2 <.0001 Depth 12 48 593.29 157.13 70.7 3.78 0.0003 Soil Sand fiandy 545.49 62.236 16.8 40.37 <.0001 oam 4.3.7 Plant Tissue Plant tissue was analyzed after column deconstruction for nutrients and metal content. Plant tissue from the sand columns did not provide enough tissue mass for analysis. The control column had decreased nutrient content and increased metal concentrations as compared to columns with wastewater application, Figure 66. Differences in nutrient percentages within the columns applied with wastewater were not significant; however there were significant increases within some metal concentrations. No differences in nutrient percentages or metal concentrations were found with depth. Table 66: Plant Tissue Concentrations by Soil Column 12-SL 30—SL 48-SL 12-SL 30—SL 48—SL Sub Sub Sub Control-SL Nitrogen % 2.89 2.93 2.68 2.99 3.01 2.29 1.63 Phosphorus % 0.33 0.51 0.44 0.36 0.35 0.32 0.19 Potassium % 4.65 5.53 4.14 3.09 4.14 3.06 1.35 Calcium 96 0.79 0.64 0.8 1.31 0.72 0.77 0.8 Magnesium % 0.49 0.45 0.44 0.56 0.32 0.48 0.27 Sodium % 0.14 0.08 0.07 0.17 0.04 0.09 0.03 Sulfur % 0.27 0.28 0.26 0.44 0.24 0.24 0.18 Iron (ppm) 456 515 1417 2243 914 1483 6037 Zinc (ppm) 33 38 62 67 38 48 74 Manganese (ppm) 31 50 49 89 49 55 172 Copper (ppm) 12 14 22 37 21 14 51 Boron (ppm) 9 12 9 9 10 11 6 Aluminum (ppm) 134 186 655 904 341 632 2620 116 The increase in nutrient within the columns with wastewater application is expected due to the increase in available nutrients for uptake. The increase in the concentration of metals within the tissue of the 12 inch submerged columns with applied wastewater compared to the non-submerged scale indicates the availability of soluble metals with increased water content and corresponding anaerobic conditions. High metal concentrations within the control columns indicate possible competition with microbial biomass, Figure 40 & 41. 200 180 Mn (ppm) Figure 40: Mn Plant Tissue Concentrations by Soil Column 117 7000 6000 .... 5000 4000 3000 :~ .. 2000 I '1: Fe (ppm) '1 lull Column Figure 41: Fe plant Tissue by Soil Column 4.4 Comparison of Treatment from Field and Laboratory Studies Soil columns of equal depth had greater removal percentages than field studies for the majority of the measured water quality parameters, Table 66. The table includes field and soil column removal percentages (shaded) in order to compare like soils and depths. Columns follow general trends found within the field system, although there are significant differences particularly at the MSU dairy site which can be explained by the experimental conditions explained below. However, it should be noted that there are significant differences in performance, but the general trends and the conclusions drawn hold true for the data presented. Table 67: Removal Percentages (%) for Filter Strips in Comparison to Soil Columns Location BOD COD TKN Ammonia Nitrite Nitrate Alkalinity pH Mn Fe Filter Strip 1 28 18 32 30 -694 26 -32 1 -343 -2 Filter Strip 2 -6 4 20 22 -1203 -164 -227 -9 -375 27 12 SL 58 43 56 60 -1113 -10580 16 -4 25 23 12 SL sub 56 45 55 63 -5566 -6581 8 0 -66 -50 Small Ml Dairy 15 ft 89 85 87 84 -20 4 54 10 -27 43 12 S 95 93 85 85 -300 -8311 1 1 -5 63 37 Small MI Dairy 25 ft 79 86 80 73 -16 36 60 10 -8 58 30 S 95 94 98 98 -495 -17348 27 —8 98 94 MSU dairy filter strips do not perform as well as soil columns of similar characteristics, the 12 inch sandy loam submerged and non-submerged. This can be attributed in part to the influentleffluent relationship of the sandy loam soils found at the MSU site. The greater increases in the influent concentrations have a direct effect on the higher effluent concentrations leading to reduced removal percentages. Additionally, it has been shown that removal is thought to be based primarily on oxygen availability within the soil. The soil at the MSU dairy Site has an average clay content of 11.3%, whereas the soil column content has a reduced average clay content of 4.3%, reducing the soil porosity and increasing the soil moisture holding capacity at the MSU site, Table 67. An increase in clay reduces soil oxygen diffusion rates, increases the soils ability to retain moisture, and reduces overall oxygen within the soil prior to wastewater application, limiting available oxygen. The BOD5, COD, TKN and ammonia reduction percentages all support this theorized difference in oxygen availability. Increases in nitrification that result due to increased oxygen also result in the lower concentrations of alkalinity within the soil columns. Increases in nitrite 119 concentrations are theorized as a result of the stress on the nitrifying bacteria due to low dissolved oxygen required for conversion to nitrate, and the increase in the hetertrophic bacteria which can dominate and reduce nitrifying bacteria (T chobanoglous et al. 2003). Both of these theories are supported by data in which there is less oxygen availability within the shorter columns and an increase in BOD removal requiring heterotrophic bacteria (as it is shown that 12 inches is not sufficient to remove BOD). Smith et al. (2003) also indicate that an accumulation of nitrite will only occur due to increases in pH, decreases in dissolved oxygen, or inhibition by free ammonia (which plays a more significant role than pH), or high organic matter (Master et al. 2004). A build up of nitrate within the system, as indicated by the negative removal, would be expected to occur as data indicated that there was oxygen inhibiting denitrification. The 12 inch submerged columns and the field system all leached more metals than what was present in the influent in contradiction to the 12 inch sandy loam columns that provided some metal removal. The submerged columns were anaerobic at the bottom of the column which can result in the leaching of metals. Non- submerged columns were exposed to the air at the bottom, and it is theorized that this allowed increased oxygen diffusion and more aerobic conditions. It can also be theorized that the increase in the initial concentration of Mn and Fe in the filter strips had an impact on the final concentrations within the leachate. 120 Table 68: Filter Strip Soil Characteristics Parameter Filter1 Strip F ilter'2 Strip 1:223 C 3:2?“ Columns pH 7.4 7.5 6.9 8.8 P (ppm) 29 31 98 3 K (ppm) 106 1 12 133 8 Ca (ppm) 1239 1412 966 632 Mg (ppm) 223 247 198 198 Zn (ppm) 5.2 3.4 4.9 3.1 Mn (ppm) 32.7 40.5 13.9 4.4 Cu (ppm) 2.8 3.0 13.9 0.8 Fe (ppm) 60.7 71.5 44.7 8.1 Orggic Matter (%) 2.9 3.2 2.3 0.3 Carbon (%) 1.7 1.9 Chloride (ppm) 56 55 59 61 Total N (ppm) 0.14 0.17 0.10 n.d. Nitrate-N (ppm) 1.83 2.45 1 1.0 0.6 Ammonium-N (ppm) 1.45 1.99 1.4 0.5 Sand (%) 61.1 61.2 69.8 93.5 Silt (%) 26.2 28.9 25.9 2.8 Clay (%) 12.7 9.9 4.3 3.7 . Sand Sand Sand Sand 8°" Type Loan: Loan¥ Loan? The small Ml dairy site resulted in much closer removal rates for the full-scale implementation as compared to the soil columns for BOD5, COD, TKN, and ammonia. Again, significant build-up in nitrite may indicate toxicity. Increases of nitrate within the soil columns are theorized to be due to aerobic conditions from the exposure of the bottom of the column to the atmosphere. Final effluent concentrations from the columns for sandy loam soils are much lower than the effluent from the filter strips, which is highly dependent of the influent concentrations. The sand columns reflect the conclusion that the soil 121 was aerobic as there was significant reduction in nitrogen from nitrification and denitrification. However, the metal leaching and a reduction in nitrate at the small Ml dairy site indicate that reducing conditions may have occurred. 122 CHAPTER 5: CONCLUSIONS AND RECOMMENDATIONS Examining feedlot sources for runoff quality was able to identify the most problematic sources in terms of pollutant loadings to provide recommendations for on-farm management. Nine storm events were used to characterize farmstead runoff pollutant sources. The heat check lot produced the greatest concentrations of COD, BOD5, ammonia, TKN, SO43 solids, TOC, and Cl‘, 7 however, was only 9% of the drainage area. Runoff from the roadway had i substantially lower concentrations resulting in a more significant effect on the composite concentrations. Feed sources did not produce as great of average ' concentrations as the heat check lot, but due to its large surface area, 29% of the drainage area, posed greater concern to waterways and treatment. Overall composite concentrations from feed areas and associated roadways were too great for complete removal in agricultural filter strips. In particular, the low pH values between 4 and 5 from feed sources impede biological treatment and burned the vegetation. The upright silos also produce a significant amount of arsenic within runoff, posing environmental and human health concerns as the concentrations were over the US EPA 10 uglL drinking water standard. Water quantity and dilution proved to be the determining factor to pollutant loading concerns and allocation of management practices. The fall months were responsible for a large portion of the high concentrations of pollutants in the feed sources due to leaching from the upright silos. Consequently, management of feed sources requires allocation of resources to 123 ensure proper upright silage filling practices. Additionally, bunker Silos need to be covered and swept prior to precipitation and feed faces need to be maintained to limit the transport of pollutants from this large surface area. If manure is a larger component than in this study, covering the manure prior to precipitation events or providing barriers such as ben'ns or curbs can limit the transport of pollutants. F A field study was conducted to determine pollutant removal of three agricultural b filter strips within the soil profile to determine the potential impact to groundwater. The three filter strips were designed and operated according to the NRCS standard to treat farmstead runoff. Two filter strips were installed at the MSU 160 cow dairy at and the third at a 40 cow small Ml dairy. Comparison of the three filter strips revealed that the small Ml dairy site had greater overall pollutant removal suspected to be due to a bioretention area that provided a large percentage of removal before application of runoff to filter strips, and the sandy soils which provided characteristics to improve oxygen availability and reduce the moisture within the soil subsurface. However, these soils still produced nitrates and metal leaching which posed environmental ground water concerns. The MSU filter strip 1 (heat check lot and roadway sources) performed better in terms of removal than the second MSU filter strip (feed runoff sources). Poor performance of the second filter strip is suspected to be due to the greater influent concentrations, as subsurface samples had higher conductivity, indicative of salt and soluble nutrient build-up over time, which is consistent with 124 overloading. In addition, performance initially was impacted by the preferential flow to one side of the filter strip 2 resulting from improper grading, determined to be a critical design component. Based on the results from all three Sites (10 sampling events at the MSU dairy, r and 6 at the small Ml dairy site) subsurface samples indicated that a soil depth of Z. ‘ '-‘A. "-"L 1 to 1.5 feet is not capable of eliminating pollutants to a degree suitable for ‘LAA. . groundwater protection. Concentrations of BOD5 were well above the 30 mg/L discharge limit. The greatest removal still resulted in average subsurface effluent concentration averages of 150 mg/L, however this is similar to the level from a septic tank. Increases in nitrite concentrations occurred in all systems, which is unusual and may indicate toxicity. Nitrate values were consistently over the 10 mg/L standard. A build up of nitrate within the system, as indicated by the negative removal, may be a result of available oxygen inhibiting denitrification. Arsenic concentrations were also over the 10 ug/L drinking water standard, particularly at the small Ml dairy filter strip. The greater sources of arsenic at this site were theorized to be due to influences from high groundwater arsenic levels being transported through excess plate cooler water entering the storage basin. Metal leaching was also a concern for groundwater sources as those measured in subsurface effluent were in greater concentrations than were in the waste stream prior to infiltration. Metal leaching and a reduction in nitrate at the small Ml dairy site indicate that reducing conditions may have occurred. Overall, the small Ml dairy had higher removal rates in 1.5 and 2.5 feet deep subsurface 125 samples compared to the first foot in the MSU subsurface samples, but final effluent concentrations indicated potential problems for nitrate and metal leaching. Improved performance of the MSU site can be achieved through a variety of r. operations and design alterations. As was indicated by the influent to effluent L concentration correlations, an increase in management to reduce influent concentrations would have a Significant effect on effluent concentrations. ‘1 Additionally, mechanical aeration in the basin of within the field (using a lawn aerator or venting within the soil profile) can provide an increase in oxygen required to increase treatment. A further measure of backfilling sand has potential to increase aeration and decrease moisture within the soil profile to increase performance. Cold weather performance was evaluated at the MSU Site only and was found to have no effect on performance, although the poor performance of these systems year rOund may influence the difference realized with seasonal variation, as a correlation may result in a more efficient system. However several deductions could be made from analysis of daily operation. Due to the transport system design requiring increased temperatures for application of waste to the treatment systems it is likely that the filter strips will have unfrozen soil subsurfaces during a runoff application ina thaw event, reducing the runoff from impermeable soils. 126 However, there is a need to investigate the temperature effect on microbial degradation in a more efficient treatment system as microbial p0pulations are known to be affected by temperature differentials. Results from the field research indicated the need for a laboratory study which could investigate treatment depth with greater control of the environmental and experimental conditions. Further investigation over a 7 month period examined the relationship of depth of treatment, soil type, and groundwater interaction on pollutant removal. Depth was determined to be a Significant factor for all measured pollutant subsurface effluent concentrations. There was a significant difference in removal at 12 inches compared to that at greater depths. Removal percentages for many parameters increased into the 90% range for depths of 30 inches or greater. Soil was also determined to be a statistically Significant as a main effect for all measured parameters except alkalinity. Sand soil had greater pollutant removal percentages than sandy loam soils for almost all parameters. Sandy loam soils also had significant increases in Mn leaching. Soil physical properties have a significant impact on soil moisture and are largely responsible for the oxygen availability within the soil profile. Soils with a high porosity, such as sand, have a high oxygen diffusion rate and the larger macro pore size decreases the soil moisture holding capacity enabling the maintenance of high oxygen levels within 127 the system. Oxygen availability was theorized to be rate limiting in the nitrification process. Nitrification processes within the soil, and lack of denitrification, indicate that the soil in columns greater than 12 inches had oxygen present. It is unknown whether sorption mechanisms were a significant factor in pollutant removal within the soil columns. Sand metal data indicated that the sorption of metal may have been a factor in reducing the metal T” concentrations. However, the increased CEO in the sandy loam soils due to the it increased clay content and organic matter did not increase removal, therefore indicating that sorption was not the dominant factor in pollutant removal. % Column submergence was only statistically significant as a main effect for Mn, Alkalinity, Nitrate, and COD. However, many other parameters were Significant for the interaction of depth and submergence. In general column submergence resulted in decreased pollutant removal. Similar trends for the filter strip columns and the field data were observed, indicating that results from the soil column data can be extended to field scale filter strip design. Differences within the results were Significant in some areas, but for the conclusions drawn the general trends were consistent and a Significant portion can be explained by the experimental conditions. The MSU filter strips had sandy loam soils and subsurface samples were collected at an average of 1 foot from the soil surface, a similar design to the 12 inch sandy loam 128 column. Higher treatment removal percentages are a result of the greater pollutant loadings as can be seen from the increase in concentrations to the decreased removal from filter strip 2, to filter strip 1, to the column results for 12 inch sandy loam soils. The MSU filter strips also received a larger hydraulic loading and had a Significant increase in soil clay content, reducing oxygen concentrations as is indicated by the reduced BOD5, COD, TKN, and ammonia results. Increased leaching of metals also indicated anaerobic conditions within the field, as indicated by the results from the submerged columns. Increases in metal leaching from the field were due to the more than 2x higher concentrations in the initial concentrations in field soils and the decrease in available oxygen. As with sandy loam soils, the 12 inch sand columns had on average greater removal percentages than the filter strips at the small Ml dairy site, however the results were very similar. Again the differences in the concentrations can all be explained by the slight differences in the soil characteristics and the reduced oxygen concentrations. Comparison of the laboratory data to field results allowed for further interpretation of field performance and implications to the larger scale design, and include the following: 0 Sand soils have greater performance than sandy loam soils, as increase in porosity increases oxygen diffusion and decreases the moisture holding capacity, and reductions in the ratio of wastewater volume to pore volume increases available oxygen. Ratios of wastewater volume to soil porosity should be below 0.27 to provide adequate oxygen. Backfilling filter strips with higher porosity soils will increase treatment, or potentially 129 mechanically increasing soil porosity, or selecting vegetation that can increase oxygen diffusion and porosity. BOD5 removal requires a depth of 30 inches for sand and sandy loam soils to reach a concentration below 30 mglL. lf groundwater is present within the system, it is recommended that the depth be increased to 48 inches. Appropriate site selection for filter strips at increased depth to groundwater will reduce groundwater impact. COD removal requires only 12 inches of depth in sand columns, as increases in depth do not produce a significantly greater removal, and a depth of 30 inches or greater for sandy loam soils. This is verified by BOD5/COD ratios that indicate incomplete removal of biodegradable material in 12 inch columns. Removal of TKN and ammonia rely on nitrification and require 30 inches of treatment depth in sand. Sandy loam soils reach adequate treatment in 30 inch soils but realize an even greater removal in 48 inch soils. Nitrification rates are controlled by alkalinity availability, pH, temperature, and oxygen availability. All soils produce an excess of alkalinity and remain between the optimum pH, so if temperatures can be maintained between 5°C and 40°C, nitrification is based solely on oxygen. Design implications are to increase oxygen within the treatment system by increasing porosity to reduce the soil moisture holding capacity or increase oxygen diffusion as discussed above. 130 o Reductions in wastewater volumes at field locations can increase the available oxygen thereby increasing removal percentages. This can be accomplished by diverting clean water not required for dilution, or increasing the width or slope of the filter strip to increase the volume of soil used to treat runoff. Siting plays a critical role in treatment efficiency and potential for contamination of ground and surface water. Siting Should include the following: 0 Location to surface water: minimum of 150 feet to surface water is required to reduce the potential for direct runoff discharge. 0 Depth to groundwater: minimum of 30 inches is required, but may require an increase in depth depending upon soil type and loading. 0 Soil type: hydraulic conductivities to eliminate ponding and increased porosities and a decrease in soil water holding capacity to increase oxygen diffusion; use of naturally occurring soils will reduce capital costs. 0 Farm maintenance practices: minimizing open feed sources and general farm upkeep to determine potential loadings to the filter strip. 0 Slope: adequate natural slope over 1%, or locations suited for excavation (increases in costs required for excavation). - Available land area: availability of land adjacent to the farmstead area to reduce transport costs associated with pumping. 131 This research has identified that depth of soil and soil type play the most significant role in pollutant removal for land application systems, and identified application to filter strip design. Further research should focus on: 0 Further investigating nitrite build-up, including possible toxicities. - Reducing nitrate concentrations within the system prior to reaching groundwater, but without reducing the soil oxygen within the desired treatment depth outlined above. Possibilities include a defined clay layer to provide an anaerobic zone after aerobic treatment. 0 Measurement of dissolved oxygen under various treatment conditions. 0 Life cycle of soil in terms of metal adsorption. 0 Optimization for vegetation in pollutant removal within the soil subsurface, particularly in regards to an increase in soil oxygen and the impact to soil conductivity. Results provide insight to source characterization, provide runoff infiltrate data for field-scale implementation, and make design recommendations based on soil columns experimentation. Further research can expand in this area to increase the performance of land application systems. 132 APPENDIX A MSU Dairy Teaching and Research Facility Management Plan Properly store all materials to avoid transfer due to environmental processes (wind, rain, etc.). Source Area 1. Sweep Silage areas, feed areas, and traffic areas and scrape livestock areas on a minimum weekly basis, increase frequency when rain is forecast or an excessive amount of solids have accumulated Maintain cover over silage and feed, ensure the rainwater is diverted from the area Maintain a smooth and vertical feed out face Avoid vehicle travel through manure piles, feed supplies etc. to avoid tracking into paved areas Avoid spillage around silage, clean up when necessary When collecting manure, ensure the trailer is positioned correctly to avoid any spillage. Properly clean off the conveyor before moving the trailer. Maintain areas around storm drains, avoid solids build-up Keep manure, feed supplies, bedding etc. in their designated places only, avoid temporary storage especially while moving Do not overflll trailers or trucks Collection System 10. Dry weather leachate Should be transferred to a storage facility on a minimum weekly basis, increase frequency if a large volume has accumulated or rain is forecast 133 11.Scrape accumulated solids from the storage basin after removing dry weather leachate and transport to a storage area Transfer System 12. Manually operate basin pumps within 72 hours after a storm event to low alarm level 13. Do not reset flow meters, they will be monitored and maintained by researchers 14. Log pump operation, maintenance issues, mowing or any other event concerning the filter strip in the log book located in the pump operation housing Vegetative Area 15. Do not graze animals in the filter strip area until research has ended 16. Reseed vegetation to maintain the desired cover when necessary 17. Mowing: a. Harvest vegetation to promote growth and maintain health and upright growth b. Maintain a vegetation height of 6 inches c. Harvest vegetation and remove from the filter strip area (I. Mow direction should be across filter strip, not down the grade (mow north to south) e. Do NOT mow filter strips when the soil is saturated f. Do NOT use herbicides on or around the filter strip vegetation area 18. Rake rock checks bimonthly to maintain an even surface 134 19. Inspect the rock checks biannually for solids accumulation, clean/replace if necessary Contacts 20. Contact Becky Larson prior to pump operation or with any questions/concerns, phone is preferred for pump operation notification, but an email is acceptable if sufficient time is given 135 APPENDIX B Table 69: QAIQC Analysis QAIQC Descriptlon Purpose Frequency Requiring Acg'mgce Corrective Action Procedure Take one . . sample volume trelaunrgeogztr ingngnscegearztrzte Determination use of All Relative Improve handling 53mm Sm ch of precision in equipment Parameters percent and precision, Duplicate are then equipment and once (excluding difference repeat procedure to re a red and and every 10 pH and less than ensure acceptance P P . prowdures samples BOD) 20% criteria is met analyzed usrng within each identical use procedures . . Ensure proper set- . Minimum of at up and procedure, Detection of least once per Should clean equipment, R ...... “2:17“: :7“... 12:52. manners 889° "9' eq pme Parameters and chemicals, find Blanks analyzed as a procedure and once ( ex clu di neutral the error in sample contamination, every 10 H) ng reading, or procedure or or background samples p used as an equipment concentrations within each offset reanalyze until "3° criteria met Minimum of at 533:an 3:32:53:- . . least once per . ‘ Known quantities Determination use of Relative clean equipment, of sample are All check all reagents of accuracy of equipment percent . analyzed to . Parameters . and chemrcals. find Standards determine equrpment and once (excluding difference the error in and every 10 . . less than accuracy of '05 nitrite) 20% procedure or uipment prowdures samp equipment eq wrthrrgach reanalyze until " criteria met . Ensure Clean probe, . 2-pornt pH 10.05 pH . Calibrate . . . accurate and . retest, replace If pH meter cahbtfiggpsw'th precise Every use pH evin'tsbger acceptance criteria readings ry cannot be met 136 APPENDIX C USDA-NRCS-MICH Standard Wastewater Treatment Strip (Acre) 635 DEFINITION A treatment component of an agricultural waste management system consisting of a strip or area of herbaceous vegetation. PURPOSES The purpose of this practice is to improve water quality by reducing loading of nutrients, organics, pathogens, and other contaminants associated with animal manure and other wastes, and wastewater by treating agricultural wastewater and runoff from livestock holding areas. CONDITIONS WHERE PRACTICE APPLIES This practice applies where all the following conditions apply: 1. Wastewater is generated by runoff from areas where livestock are concentrated, runoff and leachate from silage storage areas, runoff and leachate from waste storage facilities for solid manures, runoff from composting areas, or runoff from feed handling areas; 2. Polluted runoff (storrnwater and snow melt) may be treated in a wastewater treatment strip; 3. Manure and/or Silage solids from the contributing drainage area can be effectively 137 trapped prior to discharge to the wastewater treatment strip; and 4. The area contributing runoff and/or leachate to the wastewater treatment strip is less than 1 acre and confines less than 200 animal units (1 animal unit = 1,000 pounds live weight). This practice does not apply to filed borders (practice standard 386) or Filter Strips (practice standard 393 A). This practice standard does NOT apply to the control or treatment of milking center wastewater or any other process washwater. CRITERIA General Criteria Applicable to All Purposes Wastewater treatment strips shall be planned, designed, and installed to meet all federal, state, local and tribal laws and regulations. The term “silage” as used in this standard include haylage, wheatlage, and any other ensiled livestock feed stored on the farm. Location and Use. To minimize the potential for contamination of streams, wastewater treatment strips, including the outlet storage area, should be located outside of floodplains. However, if Site restrictions require location within a floodplain, the wastewater treatment strip, including the outlet storage area, shall be protected from inundation or damage from a 25-year flood event, or larger if required by laws and regulations. Wastewater treatment strips shall not be constructed within an area that typically has a seasonal high water within 1 foot (0.3 m) of the surface. Subsurface drainage may be used to lower the seasonal high water table to an acceptable level provided the subsurface drain lines are at least 10 feet (3 m) away from the wastewater treatment strip. All other field tile (subsurface drains) within 10 feet (3 m) of a wastewater treatment strip shall be removed and capped. Wastewater treatment strips must limit access and control grazing, where appropriate. Do not use wastewater treatment strips as a travelway for livestock of farm equipment. Dilution of the runoff to be treated in the wastewater treatment strip is not needed if the contributing drainage area is managed to minimize pollution of the runoff by manures and/or silage. Where suitable management is not provided, the runoff shall be diluted by combining clean runoff with the polluted runoff. The clean runoff contributing area shall be at least equal in area to the polluted runoff contributing drainage area. The combined area for both shall not exceed 1 acre. Suitable management to minimize pollution of runoff includes the following actions by source area: 0 Livestock areas - scraped at least weekly. 138 o Silage storage areas - have impermeable covers over stored silage, scrape and/or sweep the storage floor and apron at least weekly to collect feed that is spilled, and the silage is kept nearly vertical where it is being removed for feeding. 0 Waste storage facilities for solid manure or composting facility - manure or compost is stacked as high as possible (based on design height) and in as small an area as possible; the area where manure or compost has not yet been stacked is scraped and/or swept at least weekly to collect manure that has been spilled. Collection system. A collection system shall be provided to settle and collect solids, collect dry weather leachate (where applicable), and control the discharge of runoff to the wastewater treatment strip. The collection system shall be designed to facilitate clean-out. Where dry weather silage leachate is anticipated, the collection system shall be designed to minimize deterioration from exposure to the leachate. Collection systems that may erode during an overflow event shall have a freeboard of 6 inches (150 mm). Collection systems that will not erode during an overflow event are not required to have a freeboard. Refer to Waste Storage Facility (313) for collection system structural design criteria. Structures Shall be designed to withstand the anticipated static and dynamic loading. Settling facilities shall be installed above the water table. When curbs are needed in conjunction with collection systems, they shall be constructed of either concrete or pressure-treated wood and Shall be adequately anchored. Curbs Shall be of sufficient height to ensure flow control up to the design discharge. Refer to the Manure Transfer (634) practice standard for safety criteria and for design criteria for pipes associated with the collection system. Livestock shall be excluded from the collection system, as appropriate, to prevent damage and to avoid harm to the animals. The minimum collection system design volume shall be the volume of runoff from the 25-year, 24-hour rainfall event on the contributing drainage area less the outflow volume at the design discharge over a 24 hour period. The design outflow discharge from the collection system to the wastewater treatment strip shall not exceed the peak discharge from a 2-year, 24-hour rainfall event on the contributing drainage area. The collection system design volume shall include a solids and dry weather leachate storage area with a minimum capacity equivalent to the volume of 0.15 inches 94 mm) of runoff from the contributing drainage area. The liquid in the solids and dry weather leachate storage area may not be directed to the wastewater treatment strip during a runoff event. To facilitate dewatering of accumulated solids, the liquid in the solids and dry weather leachate storage area may be directed to the wastewater treatment strip after the 139 runoff event has ended. Stored dry weather leachate may not be directed to the wastewater treatment strip. The solids and dry weather leachate storage area shall be emptied within 72 hours of a runoff event or weekly when dry weather leachate has accumulated. The collected solids and dry weather leachate shall be transferred to a storage facility or utilized in accordance with the Nutrient Management practice standard (590). Design Discharge and Dimensions. Minimum wastewater treatment strip dimensions shall be based on the peak inflow rate resulting from a 2-year, 24-hour rainfall on the contributing drainage area. A level Spreader, grated pipe, sprinklers, or other facilities shall be provided across the upstream and of the wastewater treatment strip to establish sheet flow. The wastewater treatment strip shall not allow discharge to surface waters for up to the 25-year, 24—hour storm event. The wastewater treatment strip shall prevent lateral discharge to surface waters as the water passes along the length of the wastewater treatment strip up to the 25-year, 24-hour storm event. This may be accomplished by natural or artificial boundaries. Use the equation below to compute the design peak discharge from the contributing drainage area. (Tabular hydrograph method maximum unit peak discharge of 1,000 csm (cfs/mizlin runoff) for Type II storms in Urban Hydrology for Small watersheds, Technical Release No. 55, NRCS, June 1986.) Peak Discharge Qp (cfs): Qp = R x A x 0.000036 R = Runoff depth (in.) Compute using a curve number of 90 for unpaved areas and 98 for paved areas or roof areas A = Contributing drainage area (sq. ft.) The wastewater treatment strip Shall be a relatively uniform grass area of grassed channel. Wastewater treatment strips shall be designed for natural or constructed slopes of 0.3 to 6 percent. The first 100 feet at the upstream end should not be flatter than 1 percent. Where constructed Slopes are required, salvage existing topsoil and spread at final grade. Grass are (overland) wastewater treatment strips shall be generally on the contour and sufficiently wide to pass the peak flow at a depth of 0.5 inches (13 mm) or less. Maximum flow width (perpendicular to the direction of flow) shall be 100 feet (30 m). Flow length parallel to the direction of flow) shall be sufficient to provide at least 15 minutes of flow- through time. Flow-through time equals the wastewater treatment strip length divided by the average velocity. Average flow velocity shall be determined using Manning’s equation with an “11” value of 0.3. To minimize the development of flow concentrations which will Short-circuit the sheet flow need to maintain the effectiveness of the grass 140 wastewater treatment channel, rock checks will be installed at 100 foot intervals along the length of the channeL A rock check is a shallow trench filled with MDOT 22A or 23A coarse aggregate. The trench should be 1 to 1.5 feet (0.3 to 0.5 m) deep, extend 2 to 4 feet (0.6 to 1.2 m) in the direction of flow, and extend the full width of the channel up to the design depth. The top of the stone in the trench should be flush with the bottom of the channel. Preventing discharge to surface water. The outlet of the wastewater treatment strip shall be designed to prevent discharge to surface water. To accomplish this, the wastewater treatment strip must outlet into an outlet storage area or must maintain a minimum outlet setback distance to surface water. Outlet storage areas shall have the capacity to contain the entire runoff volume from the 25-year, 24-hour storm from the contributing drainage area plus the wastewater treatment strip area. The outlet storage area capacity may be reduced by the volume of runoff captured by the solids and leachate collection system. The outlet storage area may be a natural depression area, or a constructed depression area. The outlet storage area shall not be a wetland. The outlet storage area Shall be able to infiltrate the collected water within 72 hours based on the permeability of the most restrictive layer in the root zone regardless of its thickness. Earth berms used for constructed depressional areas shall be less than 3 feet (1 m) in height, have a freeboard of at least 0.3 (0.1 m) feet above the design high water elevation, have a top width of at least 4 feet (1.2 m), and side Slopes of at least 3:1 or flatter. Minimum outlet setback distance is 150 feet (45 m) measured along the flow path from the outlet of the wastewater treatment strip to the surface water. Surface water may be a stream, surface drain, surface inlet, road ditch or other conveyance. The slop on any portion of the outlet setback distance may not exceed 12 percent. The flow path must be either established permanent vegetation (such as hayland, pasturelands, grassland, or vegetated buffer) or cropland. Establishment of vegetation. Runoff shall be diverted away from the wastewater treatment strip channel until the vegetation is well established. A minimum height of 4 inches (0.1 m) and 90 percent ground cover is desirable. Select one of the seed mixtures in table 1, depending on soil type and drainage conditions. Limed and fertilized in accordance with the Critical Area Planting (342) practice standard. TABLE 1 — Vegetative Mixtures for Wastewater Treatment Strips Soils - Well and moderately well drained coarse to fine textured soils. Brome Tall Fescue 20* 60 Smooth 12 60 Brome Tall Fescue* Orchardgras 70 8 Timothy Red Clover Alfalfa ODODU'IU’IA Soils — Somewhat poorly drained or poorly drained soils without artificial drainage Species or Seeding Establishe Seeding Rate d Stem Mixtures (lbs/acre Density ) (stems/sq ft) Garrison 10 70 Creeping Foxtail Reed 10 50 Canarygrass Tall Fescue* 20 60 Orchardgras 5 70 s 2 Redtop 3 Alsike 2 Clover White Dutch Clover Species or Seeding Establishe Seeding Rate d Stem Mixtures (lbs/acre Density ) (stems/sq f0 Reed 1 0 50 Canarygrass Smooth 20 50 * Do not use Endophyte fungus susceptible Tall fescue varieties if area is planned for grazing or forage. Use vegetation adapted to the site that will accomplish the desired purpose. Preference shall be given to native species in order to reduce the introduction of invasive plant species; provide management of existing invasive species; and minimize the economic, ecological, and human health impacts that invasive species may cause. If native plant materials are not adaptive or proven ineffective for the planned use, then nonOnative species may be used. Refer to the Field Office Technical Guide, Section II, Invasive Plant Species for plant materials identified as invasive species. CONSIDERATIONS Consider the potential effects of installation and operation of wastewater treatment strips on the cultural, archeological, historic and economic resources. Consider the ability of the landowner/operator to manage and operate the wastewater treatment strip in accordance with the operation and maintenance plan. 142 APPENDIX D Key: MSU Dairy: F=filter strip (1 or 2), MH=man hole (numbered 1 through 3 down the slope), RC=Rock Check (numbers following indicate the rock check numbered from top to bottom and the letters from left to right of the filter strip looking up the slope), SB=storage basin (1 or 2). Small Ml Dairy: Basin=settling basin, BlO=bioretention basin, RC=rock check, T1 =subsurface samples at 1.5ft (A=3 ft down slope, B=13 ft down Slope), T2=subsurface samples at 2.5 ft. Table 70: MSU Dairy Alkalinity Data g; 5 8 a 8 3 9 3 3 g a g» 2 S E 8 9 9. s a a 2 V 8 $2 8 8 g 3 5‘2 g S F1MH1 304 330 250 300 192 155 140 100 350 F1MH2 320 330 260 320 155 145 170 390 F1 MH3 280 315 150 325 F1 RC1A 140 320 235 315 F1RC1B 160 370 240 315 125 F1RC1C 116 330 250 260 196 135 315 F1 RCZA 196 250 290 90 F1 R028 1 52 235 295 380 F1 RC2C 188 245 305 F1 R03A 250 31 0 290 F1 R038 245 310 125 F1 RC3C 245 315 F2MH1 380 340 195 0 798 535 415 345 F2MH2 396 290 675 0 490 635 335 F2MH3 360 245 0 360 305 F2RC1 A 330 190 0 490 250 F2RC1 B 380 190 0 1 55 F2RC1 C 350 200 0 940 F 2RC2A 320 220 0 1 50 F2RCZB 1 80 0 1 35 F2RCZC 1 95 0 F2RC3A 330 245 0 F 2RCSB 340 260 O F 2RCSC 320 250 0 1 75 230 $81 124 320 285 315 193 130 140 80 240 $82 260 330 35 0 650 200 350 55 160 143 Table 71: MSU Dairy Ammonia Data S! A co co do a: a: 8 8 8 8 o E .5, a a, a a g 93 5 g e E < o o - o do do ,_ 9 v F1MH1 26.75 3.75 7.75 0 10.5 0.5 1.5 F1MH2 29.25 6.25 12.75 1.5 14.5 4 54.5 1 24.5 F1MH3 24.5 5 13.5 1 16.75 F1RC1A 10 16.75 6 21.5 F1 RC1 B 8.75 16.25 12.5 20 0.5 F1 RC1C 8.75 21 6 17.5 10.5 2.5 30.25 F1RC2A 10.5 4.5 21.5 3 F1 RCZB 10.5 6.5 21 29 F1 RCZC 11.5 6.5 20 F1 RC3A 7 19.5 19.25 F1RC3B 8.5 19.5 1.5 F1 R030 13.5 21 F2MH1 14.25 28 64 29.5 67.5 105 75.5 19.5 7.5 F2MH2 13 31.5 59.25 15.5 46.5 68 3 4.75 F2MH3 10.5 51 27.5 61 16.5 7.75 F2RC1A 47 78.5 72.5 13.5 F2RC1 B 40.5 68 40.5 F2RC1C 41.5 83 160 F2RC2A 39 81.5 34.5 F2RC28 36 81.5 12.25 F2RCZC 40 82 F2RC3A 36 72 F2RC3B 33 76.5 F2RC3C 37 77 32.5 10.25 881 23.75 10.75 10 19.5 9 5.5 O 3.5 16.75 882 18.25 58.5 31 50 179 72.5 4 15 8 144 Table 72: MSU Dairy COD Data Q E 9 z 9 a ‘- 3 ‘0 \ g 8 a 2 8 8 g g ,6. g S F1 MH1 555 190 510 544 600 246 220 222 136 272 F1 MH2 760 220 350 432 616 190 308 111 222 F1MH3 675 572 628 155 310 F1RC1A 350 160 668 780 F1RC1B 250 190 704 780 271 F1RC1C 250 280 844 672 334 200 400 F1 RCZA 175 660 792 161 322 F1RCZB 300 564 832 F1RC2C 350 588 732 F1 RC3A 532 808 F1 RC3B 544 732 124 248 F1 RC3C 616 744 F2MH1 1015 1100 1910 3716 10384 7550 6830 1440 2880 F2MH2 940 1230 1750 2556 7472 8110 F2MH3 865 1210 3452 9704 1520 3040 F2RC1A 1660 3056 10368 6930 F2RC1B 2250 2992 9432 2540 5080 F2RC1C 2060 3948 11576 13440 F2RCZA 1 180 3836 1 1 104 2410 4820 F2RCZB 3736 1 1632 F 2RC2C 3776 1 1336 F2RC3A 1440 3756 10864 F2RC38 1700 3548 11296 F2RC3C 1590 3588 11616 2190 4380 $81 705 400 700 556 788 354 248 281 136 272 S82 1 130 2700 2225 3040 7232 1 3260 4920 6440 1 140 2280 145 Table 73: MSU Dairy Soluble COD Data '5 O Q t 9 C 9 B 4: 5 g} ; w 8 8 8 2 8 8 5 g .9 :2 <7 F1MH1 650 210 190 300 346 96 67 105 71 306 F1MH2 775 200 325 272 464 66 106 60 329 F1MH3 615 312 292 76 266 F1RC1A 425 265 360 466 F1RC1B 350 170 352 460 139 F1RC1C 350 300 352 460 106 62 343 F1R02A 275 284 560 91 F1RCZB 450 332 476 342 F1 R020 325 308 464 F1RC3A 252 466 326 F1RC3B 292 484 66 F1RC3C 206 506 F2MH1 1025 970 2210 3676 10950 7260 6400 1610 417 F2MH2 900 1130 1915 2504 7640 7790 320 367 F2MH3 935 2060 3420 10030 1930 376 F2RC1A 2005 3392 10440 6620 567 F2RC1B 3565 3240 9330 3110 F2RC1C 2340 3716 11950 12630 F2RCZA 1610 3716 11640 2990 F2RCZB 3264 11660 549 F2R02C 3672 11540 F2RC3A 1910 3476 11500 F2RC3B 2120 3420 11670 F2RC3C 2560 3524 11730 2670 516 $81 605 350 475 360 464 120 69 113 66 198 $62 1215 2600 2235 2740 7450 12400 4460 6150 1660 330 146 Table 74: MSU Dairy TS Data =6". 6 6 8 § 9 E E :1 g Q‘ <3 Q a :7; a g 3 5 “6° 3 p— 0 00 CD .— .- F1MH1 704 1024 830 744 984 552 F1 MH2 852 81 36 5380 680 1 140 780 F 1 MH3 804 1048 700 F1 RC1 A 1 064 1 376 F1RC1B 1436 1268 860 F1 RC1 C 2008 1408 1225 552 F1 RCZA 1 740 1084 492 F1 RCZB 1 01 2 1200 F1RCZC 1092 1116 F1 RC3A 824 1212 F1 RC3B 1 096 1 236 488 F1 RC3C 1 356 1 21 6 F2MH1 2608 7636 4956 1704 F2MH2 1 932 5848 5668 952 F2MH3 2564 6916 1652 F2RC1A 3204 7440 5020 F2RC1 B 2536 6780 2272 F2RC1 C 2572 8128 7125 F2RC2A 2900 8272 2128 F2 RCZB 2760 8920 F2RC2C 2336 8568 F2RC3A 341 6 81 96 F2RC3B 2820 1 044 F2RC3C 2832 8600 2096 $81 816 1268 1005 588 876 316 $82 1 960 4984 7070 3072 41 28 1 21 2 147 Table 75: MSU Dairy TSS Data TIE/i) 05/14/09 06/09/09 8/1 9/2009 8/31/2009 1 0/8/2009 4/1/201 0 F1MH1 120 180 100 140 110 F1 MH2 280 100 180 120 130 F1 MH3 80 1 00 F1 RC1A 500 200 F 1 RC1 B 1 000 200 60 F1RC1C 1100 260 1380 120 F 1 RCZA 1 520 120 F1 RCZB 200 230 F1 RC2C 600 460 F1 RC3A 340 140 130 F1 RC3B 280 360 F1 RC3C 480 240 F2MH1 260 200 400 220 110 F2MH2 80 250 50 F2MH3 120 90 F2RC 1 A 420 280 400 1 70 F2RC1 B 300 300 F2RC1 C 40 280 200 F2RC2A 140 220 F2RC28 580 660 100 F2RCZC 80 620 F2RC3A 1400 220 F2RC3B 520 280 F2RCSC 440 190 $81 360 1 20 1 60 1 00 1 00 60 $82 300 160 780 100 270 140 148 Table 76: MSU Dairy vs Data VS (m g/L) 05/14/09 06/09/09 8/19/2009 8/31/2009 10/8/2009 10/23/2009 F 1 MH1 256 384 21 5 1 80 368 396 F1 MH2 404 5548 3140 1 52 488 268 F1 MH3 356 344 F1 RC 1 A 424 660 F1 RC1 B 432 532 336 F1 RC1 C 556 564 245 280 F1 RC2A 524 364 200 F1 RCZB 356 588 F 1 RCZC 344 456 F1 RC3A 396 540 F1 RC3B 400 564 232 F 1 RC3C 432 516 F2MH1 1716 5172 3168 912 F2MH2 1064 3828 3496 424 F2MH3 1620 4896 944 F2 RC1A 1804 4788 2976 F2RC1 B 1484 4440 1568 F2RC1C 1572 5272 4305 F 2RC2A 1760 5712 1380 F2RC28 2348 5620 F2RC2C 1264 5616 F2RC3A 1680 5744 F 2RC3B 1668 488 F2RCSC 1740 5740 1304 $81 364 544 270 96 448 200 $82 1 364 3544 41 95 1900 2624 896 149 Table 77: MSU VSS Alkalinity Data VSS (mg/L) 05l14/09 06109l09 8/19/2009 8/31/2009 10/8/2009 4/1/201 0 F1MH1 120 120 60 120 50 F1 MH2 220 40 80 60 60 F1 MH3 60 70 F1 RC1 A 280 140 F1 RC1 B 220 140 30 F 1 RC1 C 200 160 360 70 F 1 RC2A 360 1 00 F1 R028 80 1 10 F1 RC2C 140 140 F1 RC3A 240 80 90 F1 RC3B 220 100 F 1 RC3C 1 40 1 20 F2MH1 120 160 360 200 80 F2MH2 0 21 0 5O F2MH3 100 0 F2RC1A 140 180 230 160 F2RC1 B 160 240 F2RC1 C 0 260 140 F2RC2A 1 20 160 F2RCZB 360 300 70 F2RCZC 0 160 F2RC3A 440 180 F2RC3B 200 220 F2RC3C 320 1 30 $81 1 20 1 00 120 0 80 1 0 $82 80 140 600 60 230 60 150 Table 78: MSU Dairy Nitrite Data Nitrite I! 1 '1 -N) 09/1 7/08 05/14/09 06/09/09 8/1 9/2009 8/31/2009 1 0/8/2009 1 0/23/2009 4/1/201 0 F1 MH1 0.36 0.022 0 0.372 0.044 0.032 0.272 0.026 F1 MH2 0.229 0.044 0.02 0.052 0.07 0.034 0.018 F1 MH3 0.024 0.026 0.07 0.04 F1 RC1A 0.013 0.04 0.008 F1 RC1 B 0.008 0.022 0.045 0.028 F1RC1C 0.006 0.14 0.008 0.104 0.018 0.026 F1 RCZA 0.002 0.042 0.034 0.02 F1 RCZB 0.524 0.036 0.008 0.034 F1 RC2C 1.86 0.03 0.008 F1 RC3A 0.03 0.02 0.042 F1 RC3B 0.026 0.008 0.022 F1 RC3C 0.022 0.018 F2MH1 0.005 0.328 3.6 0.662 2.224 0.012 F2MH2 0 0.052 0.094 0.444 0.008 6.7 F2MH3 0.046 0.08 1.73 0.012 F2RC1A 0.034 0.576 0.724 0.012 F2RC1B 0.016 1.89 0.018 F2RC1 C 0.04 0.016 0 F2RC2A 0.032 0 0.05 F2RCZB 0.03 0 0.012 F2RCZC 0.002 0 F2RC3A 0.032 0 F2RC3B 0.03 0 F2RC3C 0.042 0.038 0.014 0.02 SB1 0.007 0.026 0.01 0.012 0.018 0.28 0.068 882 0 0.03 0.006 0 0 0.138 0 151 Table 79: MSU Dairy Nitrate Data Nitrate (mglL-N) 05/14/09 06I09/09 8/1 912009 8131/2009 10/8/2009 10/23/2009 4/1/201 0 F1MH1 7.6 16 3.8 23.0 13 1.4 11.8 F1MH2 10.4 8 2.7 30 2.2 27 F1 MH3 8.6 6.6 6 3.8 F1 RC1A 9.4 15.8 F1 RC1 B 7.6 17.2 12.6 F1 RC1 C 6.4 16.4 11.1 2.6 6.6 F1 RC2A 15.2 14.8 9.8 F1 RCZB 10 20.6 5.6 F1 RC2C 9.4 16.8 F1 RC3A 6.2 15.8 11.8 F1 RC3B 10.2 21.8 3 F1 RC3C 13.6 21.4 F2MH1 6 13.4 49.2 176 17.4 17.4 F2MH2 10.4 12.4 148 22 41.2 F2MH3 8.2 10.2 22.2 0 F2RC1A 4.8 23 54 2.6 F2RC1 B 15.4 24 2 F2RC1C 21.4 24.2 116.0 F2RC2A 9.6 18.2 20.8 F2RC2B 6.8 20.2 4 F2RCZC 10.8 15.2 F2RC3A 1.6 22.4 F2RC38 10.4 14.4 F2RC3C 14 44 26.4 6.4 SB1 7.6 9.6 0.0 18.0 0 8.4 36.8 832 6.6 32.8 9.6 0.0 39 7.2 9.2 152 Table 80: MSU Dairy pH Data 6 6 6 6 6 6 6 6 2 6 6 6 6 6 6 6 6 6 6 6 e 6 8 g 6 5‘2 6 6 F1 MH1 6.87 6.7 6.63 6.62 6.54 6.61 6.62 6.45 6.78 F1 MH2 6.87 6.94 6.79 6.7 6.58 6.55 6.55 6.74 F1 MH3 6.73 6.95 6.56 6.45 6.41 6.83 F1 RC1A 6.96 6.88 6.74 F1 RC1 B 6.99 6.2 6.74 6.67 F1 RC1 C 6.82 6.75 6.88 6.7 6.92 F1 RCZA 6.98 6.94 6.59 F1 RCZB 6.91 6.75 7.19 F1 RCZC 6.82 6.9 F1 RC3A 6.95 6.8 7.1 F1 RC3B 6.98 6.87 6.49 F1 RC3C 7 6.81 F2MH1 6.53 6.03 5.2 4.3 5.82 5.63 6.33 6.67 F2MH2 6.53 6.12 5.86 4.58 5.61 6.75 6.71 F2MH3 6.75 6.45 5.29 4.45 6.41 6.95 F2RC1A 6.41 5.08 4.48 5.61 6.77 F2RC1B 6.28 5.05 4.65 5.21 F2RC1 C 6.4 5.13 4.36 5.39 F2RCZA 6.66 5.12 4.28 5.41 F2RCZB 5.13 4.33 6.95 F2RCZC 5.2 4.36 F2RC3A 6.64 5.23 4.36 F2RC3B 6.54 5.38 4.36 F2RC3C 6.5 5.29 4.36 6.49 7.1 1 $81 6.8 6.78 6.65 6.74 6.73 6.92 6.66 6.3 7.07 882 6.2 6.11 4.88 4.27 5.48 5.45 5.46 4.75 6.79 153 Table 81 : MSU Dairy Phosphorus Data (D m E \ E \ Z \ a: v- 00 ; g v 6 6 2 6 6 g 6 6 6 F1MH1 12.5 9 7.5 2.2 1.2 9.3 11.2 20 20 F1MH2 12 9 1.3 1.1 5.5 22.5 22.75 F1 MH3 10.5 1.9 1.1 13.5 F1 RC1A 7.5 5 6.0 9.0 F1RC1B 0 6.5 5.5 8.5 20.5 F1RC1C 10.5 9 11.6 9.0 12.4 20.75 F1RC2A 10 4.6 8.0 F1RC28 10.5 7.5 8.5 15.5 F1RCZC 12.5 5.0 8.0 F1RC3A 3.3 7.2 24.25 F1RC3B 3.0 7.7 F1RC3C 5.0 7.8 F2MH1 12.5 29 18 32.5 23.3 46 21.5 F2MH2 10 32 15.5 13.4 14.2 57 18.25 F2MH3 9.5 12 27.9 19.5 20.5 F2RC1A 37.4 57 31.25 F2RC1B 35.6 F2RClC 34.6 F2RC2A 30.9 F2RC28 30.1 26 F2RCZC 33.3 F2RC3A 31.0 25.9 F2RCSB 27.7 28.2 F2RC3C 17.4 27.9 27.25 881 10.5 9.5 2.8 1.6 12.9 10.7 20 17.5 882 20.5 22.5 23.1 16.6 40.7 20 24.3 154 Table 82: MSU Dairy Mn Data Mn (uglL) 09/09/08 09/17/08 10/02/08 5/14/2009 06/09/09 08/19/09 08/31/09 4/1/2010 F1MH1 1900 1200 280 130 310 52 62 280 F1 MH2 980 1300 450 140 400 31 190 F1 MH3 5800 1 80 370 1 30 F1 RC1A 330 1100 580 250 F1 RC1B 300 1600 1500 190 F1 RC1 C 330 430 1 900 140 59 320 F1 RCZA 540 1 100 280 F1 RCZB 320 270 1 100 240 F 1 RC2C 420 1 90 290 F1 RC3A 620 310 79 F1 RC3B 1 100 360 F1 RC3C 81 0 230 F2MH1 1 300 3600 860 1 300 4100 270 F2MH2 1 700 4700 840 2400 6900 4500 130 F2MH3 1400 570 1 500 4200 69 F2RC1A 1500 1600 1700 130 F2RC1 B 1300 1 500 1 500 F2RC1 C 970 1500 1700 2100 F2RC2A 590 1800 2200 F2RCZB 2000 2700 31 F 2RC2C 2000 4100 F 2RC3A 870 3200 3100 F2RCSB 1 200 2600 3500 F2RC3C 970 2400 4300 25 SB1 120 220 330 190 230 270 140 120 $32 200 470 540 400 720 1400 580 140 155 Table 83: MSU Dairy Fe Data Fe (uglL) 09/09/08 09/1 7/08 1 0/02/08 5/1 4/2009 7 06/09/09 0811 9/09 08/31 [09 4/1/2010 F1MH1 2100 1300 2100 1400 1300 910 880 1200 F1 MH2 1600 1 300 2200 1 000 1 200 640 1600 F1 MH3 4200 1400 1200 860 F1 RC1A 2600 7300 1 7000 4800 F1RC1B 1800 9400 75000 2100 F1 RC1 C 2500 4800 100000 3200 1200 900 F1 RCZA 4600 48000 2000 F 1 RCZB 2800 6900 38000 820 F1 RC2C 3200 3600 1 500 F1 RC3A 16000 1400 810 F1 RCSB 34000 7300 F1 RCSC 23000 2900 F2MH1 1200 2900 2900 4200 14000 13000 560 F2MH2 1100 3900 2100 1800 7000 1700 F2MH3 720 1 200 3400 1 0000 320 F2RC1A 14000 13000 12000 590 F2RC1 B 17000 16000 6900 F2RC1 C 4700 5700 16000 13000 F2RC2A 1600 14000 19000 F2 RCZB 1 2000 1 8000 470 F2RC2C 10000 91000 F2RC3A 3600 66000 22000 F 2RC3B 4800 29000 1 7000 F2RC3C 2300 1 0000 30000 370 SB1 1 100 1500 2200 2200 1900 2600 1 100 740 $82 1 300 6600 5700 3200 12000 1 8000 6000 940 156 Table 84: MSU Dairy TOC Data TOC (mg/L) 09/09/08 09/17/08 10/02/08 05/14/09 6/9/2009 08/19/09 08/31 [09 4/1/2010 SB1 240 100 190 190 210 56 44 110 $82 420 980 780 940 2800 3600 1400 130 F1 MH1 200 55 97 160 160 46 37 140 F1 MH2 270 64 130 140 190 41 140 F1 MH3 230 170 170 120 F1 RC1 A 85 120 200 260 F1 RC1 B 73 89 190 220 F1RC1C 84 140 180 210 55 140 F1 RC2A 65 160 240 F1 RCZB 76 160 240 140 F1 RCZC 66 160 230 F1 RC3A 160 240 140 F1 RC3B 1 50 230 F1 RC3C 1 50 230 F2MH1 400 370 710 1200 3900 160 F2MH2 330 41 0 660 850 2800 130 F2MH3 320 620 1200 3500 130 F2RC1A 630 1200 3600 190 F2RC1 B 770 1 300 3300 F 2RC1 C 730 1 300 41 00 3900 F2RCZA 540 1 300 4100 F2RCZB 1300 4100 180 F2RC2C 1200 4100 F2RC3A 620 1200 4200 F2RC3B 660 4200 F2RC3C 690 1 200 41 00 1 70 157 Table 85: MSU Dairy Conductance Data 73311131? 09/09/06 09/17/08 10/02/06 05/14/09 06/09/09 08l19/09 08I31l09 4/1/2010 SB1 1032 560 1492 973 1357 1242 636 1239 $82 1030 1634 1944 1290 2700 4950 2211 536 F1MH1 1573 696 1290 996 1337 1199 942 2553 F1MH2 1509 1410 996 1316 950 2664 F1MH3 1626 1054 1129 2446 F1RC1A 561 1373 1036 1601 F1RC1B 576 1362 1022 1535 F1RC1C 563 1324 1059 1555 1224 2536 F1RC2A 562 1021 1477 F1RCZB 606 1006 1464 2434 name 610 1005 1450 F1RC3A 976 1431 2340 F1RC3B 992 1448 F1RC3C 1007 1456 F2MH1 1227 1379 1615 1995 3660 3790 901 F2MH2 1206 1391 1744 1639 3230 906 F2MH3 1127 1576 1942 3720 760 F2RC1A 1925 1972 4050 673 F2RC1B 1953 1964 3790 F2RC1C 1652 1990 4060 5070 F2RC2A 1657 1961 4000 F2RCZB 4060 671 F2RCZC 1977 4140 F2RC3A 1636 2005 4060 F2RC3B 1670 1927 4260 F2RC3C 1906 1940 4280 701 158 Table 86: MSU Dairy CI Data 7 Cl (mg/L) 09/09/08 09l1 7/08 1 0l02/08 05l1 4/09 6/9/2009 08/1 9/09 08/31/09 4/1/2010 F1 MH1 118 53 143 123 225 236 172 548 F1MH2 120 162 121 187 176 654 F1 MH3 135 134 125 548 F1RC1A 48 162 134 251 F1 R018 47 156 132 242 F1 RC1 C 45 163 136 265 253 550 F1 RC2A 31 135 216 F1 RCZB 41 134 216 549 F1 RC2C 38 136 218 F1 RC3A 141 210 518 F1 RC3B 139 213 F1 RC3C 139 215 F2MH1 20 1 8 24 39 1 1 2 83 26 F 2MH2 17 19 27 30 102 23 F2MH3 18 22 36 108 23 F2RC1A 23 42 140 23 F2RC1 B 23 41 1 17 F2RC1C 22 42 135 1 13 F2RCZA 22 42 124 F2RC2B 124 20 F2RC2C 42 133 F2RC3A 22 41 133 F2RC3B 22 41 130 F2RC3C 25 44 132 19 SB1 13 38 191 114 187 253 161 195 $82 18 20 27 26 76 1 12 44 17 159 Table 87: MSU Dairy Arsenic Data Arsenic (pg/L) 09/09/08 0911 7/08 10/02/08 05l14/09 6/9/2009 0811 9/09 08/31l09 4/1/2010 F1 MH1 5 3.1 3.4 2.8 2.5 2 1.8 1.5 F1MH2 3.4 3.4 4.2 2.8 2.4 2 2 F1 MH3 5.7 2.9 2.2 1.6 F1 RC1A 6.7 12 7.4 2.5 F1RC1B 6.2 14 20 1.2 F1RC1C 5.5 5.5 26 2.5 2 1.7 F1 RC2A 11 15 1.1 F1 RCZB 9.8 5.5 10 1.3 F1 RCZC 11 3.9 1.2 F1 RC3A 7.5 1 1.6 F1 RC3B 14 3.8 F1RC3C 7.4 2.1 F2MH1 3.1 6.4 4.5 6.8 14 20 2.6 F2MH2 3.6 7.4 4.3 7.3 13 2.9 F2MH3 3.9 5.1 7.5 14 2.6 F2RC1A 9.8 8.4 9.5 1.8 F2RC1 B 6.8 8.6 8.2 F2RC1 C 5.1 4.8 9.8 13 F2RCZA 11 9 1o F2RCZB 8.7 13 2.7 F2RCZC 1o 40 F2RC3A 9.8 24 13 F2RC3B 13 14 14 F2RC3C 7.2 11 22 2.8 SB1 1.9 1 2.4 2.5 0 1.8 1.4 1.2 $82 1.9 2.8 3.9 3.1 4.8 8.3 4 1.2 Table 88: Small MI Dairy TOC Data 700 (mg/L) 5/16/2010 5/24/2010 6/4/2010 6/10/2010 6/16/2010 6/24/2010 BASIN 1400 960 1300 1200 820 490 BIO 840 680 490 330 760 320 RC1 400 430 520 380 320 320 T1A 190 170 130 140 150 130 T18 110 210 170 110 110 TZA 86 77 110 110 210 240 T28 74 330 270 150 130 160 Table 89: Small Ml Dairy Mn Data Mn (uglL) 511612010 512412010 61412010 6110/2010 6116/2010 6124/2010 7 BASIN 2100 1500 1500 1400 1100 600 BIO 6100 4400 2600 2900 3300 1600 RC1 6200 2200 3500 3100 2400 1000 T1A 6500 4000 2100 1400 2600 990 T1B 1 100 2600 4500 7700 7900 T2A 1500 1500 4000 6800 2600 1600 T26 1600 6600 3500 2800 2900 Table 90: Small MI Dairy Fe Data Fe (uglL) ’ 5/18/2010F 512412010 61412010 611012010 6116/2010 6124/2010 " BASIN 16000 9600 6400 6600 3700 2400 BIO 130000 50000 23000 32000 42000 20000 RC1 210000 14000 24000 18000 17000 6300 T1A 66000 46000 22000 5100 6100 5300 T16 26000 12000 7100 16000 26000 T2A 16000 1000 4500 9600 9400 6600 T26 63000 37000 10000 4800 5400 Table 91: Small Ml Dairy Conductance Data (1331111521: 511 612010 5124/2010 61412010 611012010 611612010 6/24/2010 BASIN 5540 4560 5920 5330 4672 2990 BIO 4760 4420 3150 2795 2616 RC1 2696 2397 3210 3000 2479 TM 1836 1921 1667 2076 2202 2072 T16 957 1635 1602 2062 2063 T2A 1647 1766 1847 2021 2142 1646 T26 1362 2144 2296 1636 1208 161 Table 92: Small Ml Dairy Cl Data 01 (mg/L) ' 5116/2010 5124/2010 61412010 6/10/2010 6116/2010 6124/2010 BASIN 394 319 387 262 312.2 179 610 312 326 206 160 162 RC1 141 134 193 169 144 T1A 106 122 142 155 115 12 T16 62 144 140 134 137 T2A 119 117 125 131 106 102 T26 92 129 154 90 67 Table 93: Small Ml Dairy As Data As (uglL) 511612010 512412010 61412010 611012010 611612010 612412010 BASIN 30 26 22 24 24 33 BIO 7o 39 34 30 32 25 RC1 79 30 32 31 23 17 T1A 41 26 14 12 16 11 T16 16 12 15 13 15 T2A 6.3 3.1 0 11 17 20 T26 25 27 22 14 15 Table 94: Small Ml Dairy BOD5 Data 3095 511612010 512412010 6/4/2010 6110/2010 6124/2010 .4660 BASIN 1161 1542 1434 1434 961.5714266 BIO 672 314 314 RC1 277 404 455 455 T1A 146 T1B 165.426571 165 T2A 243 129 T26 344 344 162 Table 95: Small Ml Dairy Alkalinity Data (mg/£323.03) 511612010 512412010 61412010 611 012010 6/16/2010 6124/2010 BASIN 1655 2060 2500 3280 2440 1 140 610 2105 2160 3500 1720 2440 1320 RC1 1190 1260 3440 1600 1540 1240 T1A 795 1060 920 1020 1320 1220 T18 720 920 1000 1260 1140 T2A 620 940 960 1040 1240 960 T26 780 1040 1100 980 640 Table 96: Small Ml Dairy COD Data 000 (mg/L) 511 612010 512412010 6/4/2010 611 012010 611612010 612412010 BASIN 6460 4790 5340 5360 3030 3010 810 7120 13690 1670 1010 2120 2320 RC1 7160 1760 1790 1470 610 1270 T1A 2130 722 522 537 410 960 T1B 486 1006 365 450 270 T2A 361 266 398 754 930 370 T28 642 1595 1070 430 560 Table 97: Small Ml Dairy TKN Data TKN (mg/L- N) 5/18/2010 5/24/2010 6/4/2010 6/10/2010 6/16/2010 6/24/2010 BASIN 350 270 290 270 200 130 BIO 370 290 130 1 10 240 91 RC1 210 110 140 120 100 86 T1A 54 37 25 23 28 17 T1 B 30 43 39 27 23 T2A 23 19 24 24 63 61 T28 28 87 75 28 24 163 Table 98: Small Ml Dairy Ammonia Data Ammonia (mglL-N) 5/18/2010 512412010 61412010 611012010 611612010 612412010 BASIN 146 74 100 169 61.5 51 810 154 37 61 74 61.5 52 RC1 59 49 66 - 79 56.5 33.5 T1A 25 14 23 5 12.5 10 T16 12.5 4.5 15.5 15.5 15 T2A 11 33.5 7 15.5 39 30 T26 30.5 36.5 44 10.5 9.5 Table 99: Small Ml Dairy Nitrate Data Nitrate (mglL-N) 511 612010 512412010 61412010 611012010 611 612010 6/24/2010 f BASIN 25 30 90 4o BIO 105 30 10 40 7.5 RC1 75 45 20 37.5 T1A 40 33 0 1.1 35 17.5 T18 20 50 45 20 17.5 T2A 100 25 5 12.5 T26 0 3.3 1.3 35 17.5 Table 100: Small Ml Dairy Nitrite Data Nitrite (mglL-N) 511612010 512412010 614/2010 BASIN 0.3 0.425 0.06 BIO 0.1 0.003 0 RC1 0.1 o 0 T1A 0.1 0.12 T16 0.8 0.04 T2A 0.275 0.16 T26 0.001 0.06 164 Table 101: Small Ml Dairy pH Data pH 5/18/2010 5/24/2010 6/4/2010 6/10/2010 6/16/2010 6/24/2010 BASIN 8.07 8.15 7.91 8.05 7.66 8.96 BIO 7.82 7.85 7.06 6.88 7.69 7.19 RC1 7.45 7.37 7.27 7.35 7.33 7.64 T1A 7.2 7.51 7.44 7.37 7.94 T18 6.84 7.63 6.79 7.44 7.4 T2A 7.64 7.5 6.94 7.4 7.17 T28 7.46 7.55 7.38 7.1 6.87 Table 102: Small Ml Dairy Phosphorus Data Fhosphorus (mg/L) 511612010 512412010 614/2010 6/1 012010 611612010 612412010 BASIN 96 121 74 74 128 73 BIO 154.5 317 9 16 143 44 RC1 45 95 9 19 59 65 T1A 20 43.5 0 0 20 21 T18 37 0 0 7.5 28.5 T2A 69 17.5 0 0 29.5 38.5 T28 45.5 0 0 20.5 27.5 Table 103: Small Ml Dairy Soluble COD Soluble 000 (mg/L) 6/10/2010 6116/2010 6/24/2010 BASIN 1686 1420 BIO 716 1550 930 RC1 729 923 1053 T1A 360 488 355 T18 590 320 322 T2A 397 745 768 T28 303 440 356 165 Table 104: Small MI Dairy TS Data TS (mg/L) 511812010 512412010 61412010 611 01201 0 611612010 612412010 BASIN 5788 6204 9940 4584 3028 BIO 10514 14568 7892 8936 2212 RC1 12714 2476 4712 2460 2416 T1A 3756 3120 4224 2124 1532 T18 1604 2192 4376 1584 1504 T2A 6330 1 136 2756 3312 1900 1600 T28 2444 1336 2812 1288 912 Table 105: Small Ml Dairy VS Data VS (mg/L) 7 511812010 512412010 61412010 611012010 611612010 612412010 BASIN 2360 3372 3552 1856 1312 BIO 4812 6252 5640 4032 816 RC1 2652 1160 1732 1080 840 T1A 362 588 1744 824 524 T1 8 420 1068 2656 796 340 T2A 916 272 892 724 888 516 T28 740 320 926 604 226 Table 106: Small Ml Dairy TSS Data TSS (mg/L) 512412010 61412010 611012010 611 612010 612412010 BASIN 2520 1880 540 720 BIO 16967 440 7440 220 RC1 910 800 300 260 T1 A 2580 440 127 T18 880 5020 467 187 40 T2A 84 3700 1 87 1 33 1 07 T28 2180 148 407 113 13 166 Table 107: Small Ml Dairy VSS Data VSS (mg/L) 512412010 61412010 611 012010 61161201 0 612412010 BASIN 1920 1240 520 400 BIO 7633 180 3680 0 RC1 570 640 220 180 T1 A 293 1 87 60 T1 B 20 1060 347 73 T2A 24 780 93 1 07 67 T28 253 48 360 73 1 3 167 APPENDIX E Statistical analysis was conducted by first defining the statistical models, then validating assumptions of normality of the residuals and the homogeneity of the variances, looking at time, soil, and depth. Groupings were examined when necessary, no transformations of data were necessary, and the models were then evaluated for Significance using ANOVA and the covariance structures examined to determine the best fit models. The final soil column SAS statistical analysis models for each parameter follow. proc mixed data=FSBOD; class Soil Depth Sub Time; model BOD=Soil Depth Soil*Depth/ddfm=kr outp=BODoutput; random Column(Soil Depth Sub); repeated time/group=Depth subject=Column(Soil Depth Sub) type=cs; Ismeans Depth Soil Soil*Depth/pdiff; run; proc mixed data=FSCOD; class Soil Depth Sub Time; model COD=Soil Depth Soil*Depth Sub Depth*Sub Time Sub*Time Soil*Time Sub*Time/ddfm=kr outp=CODoutput; random Column(Soi| Depth Sub); repeated time/group=Depth subject=Column(SoiI Depth Sub) type=arh(1); Ismeans Soil Depth Soil*Depth Sub Depth*Sub/pdiff; run; proc mixed data=FSTKN; class Soil Depth Sub Time; model TKN=SoiI Depth Soil*Depth Sub Depth*Sub Time Soil*Time Sub*Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time/group =Depth subje0t=Column(Soil Depth Sub) type=cs; Ismeans Soil Depth Soil*Depth Sub Depth*Sub/pdiff; run; 168 proc mixed data=FSAmmonia; class Soil Depth Sub Time; model Ammonia=Soil Depth Soil*Depth Sub Depth*Sub Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time/group =Depth subject=Column(Soil Depth Sub) type=cs; Ismeans Soil Depth Soil*Depth Sub Depth*Sub/pdiff; run; proc mixed data=FSNitrite; class Soil Depth Sub Time; model Nitrite=Soil Depth Soil*Depth Time Soil*Time Depth*Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time/group =Depth subje0t=Column(Soil Depth Sub) type=cs; Ismeans Soil Depth Soil*Depth/pdiff; run; proc mixed data=FSNitrate; class Soil Depth Sub Time; model Nitrate=Soil Depth Soil*Depth Sub Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time1subject=Column(Soil Depth Sub) type=cs; Ismeans Soil Depth Soil*Depth Sub/pdiff; run; proc mixed data=FSpH; class Soil Depth Sub Time; model pH=Soil Depth Sub Time Soil*Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time1subje0t=Column(Soil Depth Sub) type=csh; Ismeans Soil Depth Sub Timelpdiff; run; proc mixed data=Alk; class Soil Depth Sub Time; model AIkalinity=Soil Depth Soil*Depth Sub Depth*Sub Time Soil*Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated Time/subject=Column(Soil Depth Sub) type=arh(1); Ismeans Soil Depth Soil*Depth Sub Depth*Sub/pdiff; run; 169 proc mixed data=Mn; class Soil Depth Sub Time; model Mn=SoiI Depth Sub Time Soil*Time Depth*Time Sub*Time lddfm=kr outp=Mnoutput; random Column(Soil Depth Sub); repeated time/group=depth subject=Column(SoiI Depth Sub) type=cs; Ismeans Depth Soil Sublpd'rff; run; proc mixed data=FSFe; class Soil Depth Sub Time; model Fe=Soi| Depth Time Soil*Time Depth*Time/ddfm=kr outp=output; random Column(Soil Depth Sub); repeated time/group =Depth subject=Cqumn(Soil Depth Sub) type=cs; Ismeans Depth Soil/pdiff; run; 170 REFERENCES Anderson, D.M., GM. Glibert, and J.M. Burkholder. (2002). “Harmful Algal Blooms and Eutrophication: Nutrient Sources, Composition, and Consequences.” Estuaries, 25(4b), 704-726. Azevedo, J., and PR. Stout. (1974). “Farm Animal Manures: An Overview of Their Role in the Agricultural Environment.” California Agricultural Experiment Station and Extension Service Manual 44. Berkeley: California Agricultural Experiment Station and Extension Service. Barker, A.V., M.L. Stratton, and J.E. Rechcigl. (2000). “Soil and By—Product Characteristics that Impact the Beneficial Use of By-Products.” In: Land Application of Agricultural, Industrial, and Municipal By-Pmducts, No. 6, WA. Dick, edS.,Soil Science Society of America, Inc., Madison, Wisconsin. Bartlett, R.J., and Ross, OS. (2005). “Chemistry of Redox Processes in Soils.” In: Chemical Processes in Soils, No. 8, Tabatabai, M.A., Sparks, D.L., AI- Amoodi, L. and Dick, W.A., eds., Soil Science Society of America, Inc., Madison, Wisconsin. Bhaskar, V. (2003). Root hairs: the ‘gills’ of roots, development, structure, and functions. Science Publishers, Enfield, NH. Bohlke, J.H. (2002). “Groundwater Recharge and Agricultural Contamination.” Hydrogeology Joumal, 10(1), 153-179. Bratieres, K., Fletcher, T. D., Deletic, A., and Zinger, Y. (2008). "Nutrient and sediment removal by stomiwater biofilters: A large scale design optimization study." Water Research, 42, 3930-3940. Brown and Caldwell. (2007). Manual of Good Practice for Land Application of Food Processing/Rinse Water. Davis, California. Burdige, D.J., Dhakar, SP, and Nealson, K.H. (1992). “Effects of manganese oxide mineralogy on microbial and chemical manganese reduction.” Geomicmbiol. J. 10, 27-48. Burkholder, J., B. Libra, P. Weyer, S. Heathcote, D, Kolpin, P.S. Thorne, and M. \Mchman. (2007). “Impacts of Waste from Concentrated Animal Feeding Operations on Water Quality.” Environmental Health Perspectives, 115(2), 308-31 2. 171 Clark, R.N., C.B. Gilbertson, and HR. Duke. (1975). “Quantity and Quality of Beef Feedyard Runoff in the Great Plains.” ln Managing Livestock Wastes. St. Joseph, Mich.: ASAE. Corbitt, RA. (1998). Standard Handbook of Environmental Engineering, 2"” Edition, Ed. by RA. Corbitt, McGraw Hill, Mew York, NY. DeBusk, J., J.A Ogejo, N.G. Love, and KP. Knowlton. (2007). “Adjusting nitrogen to phosphorus ratios in liquid dairy manure through nitrification and chemical phosphorus removal to match crop fertilizer requirements.” ASAE Paper No. 074048. St. Joseph, Mich.: ASAE. Dickey, EC, and DH. Vanderholm. (1981). “Vegetative Treatment of Livestock Feedlot Runoff.” J. Environ. Qual., 10(3), 279-284. Dillaha, T.A., J.H. Sherrard, D. Lee, S. Mostaghimi, and V0. Shanholtz. (1988). “Evaluation of Vegetative Filter Strips as a Best Management Practice for Feed Lots.” Journal of Water Pollution Control Federation, 60(7), 1231 - 1238. Donker, M. H., Eijsacker, H., and Heimbach, F. (1994). Ecotoxicology of Soil Organisms, Lewis Publishers, Boca Raton. Dosskey, M. G., Hoagland, K. D., and Brandle, J. R. (2007). "Change in filter strip performance over ten years." Journal of soil and water conservation, 62(1), 21-32. Edwards, D. R., Moore, P. A., Jr., Daniel, T. C., Srivastava, P., and Nichols, D. J. (1997). "Vegetative filter strip removal of metals in runoff from poultry litter- amended fescuegrass plots." Transactions of the ASAE, 40(1), 121-127. Edwards, W.M., L.B. Owens, and R.K. White. (1983). “Managing Runoff from a Small, Paved Beef Feedlot.” J. Environ. Qual. 12(2): 281-286. Fendorf, S.E., D.L. Sparks, J.A. Franz, and 0M. Camaioni. (1993a). “Electron paramagnetic resonance stopped-flow kinetic study of manganese(ll) sorption-desorption on birness'rte.” Soil Sci. Soc. Am. J. 57, 57—62. Fendorf, S.E., R.J. Zasoski, and R.G. Burau. (1993b). “Competing metal of soil analysis ion influences on chromium (Ill) oxidation by bimessite.” Soil Sci. Soc. Am. J. 57, 1508—1515. 172 Feng, G., L. Wu, and J. Letey. (2002). “Evaluating Aeration Criteria by Simultaneous Measurement of Oxygen Diffusion Rate and Soil-Water Regime.” Soil Science, 167(8), 495-503. Froelich, P.N., Klinkhammer, G.P., Bender, M.L., Luedtke, N.A., Heath, D.C., Dauphin, P., Hammond, D., Hartman, 8., and Maynard, V. (1979). "Early oxidation of organic matter in pelagic sediments of the eastern equatorial Atlantic: suboxic diagenesis." Geochim. Cosmochim. Acta 43, 1075-1090. Goel, P. K., Rudra , R. P., Gharabaghi, 8., Das, S., and Gupta, N. (2004). "Pollutants Removal by Vegetative Filter Strips Planted with Different Grasses." In: 2004 ASAE/CSAE Annual International Meeting, American Society of Agricultural and Biological Engineers, Ottawa, Ontario, Canada. Gray, N. F. (2005). Water technology: an introduction for environmental scientists and engineers, Elsevier Butterworth-Heinemann, Oxford. Haggblom, M. M. and Milligan, P. W. (2000). “Anaerobic Biodegradation of Halogenated Pesticides: Influence of Alternative Electron Acceptors.” Soil Biochemistry, Jean-Marc Bollag and G. Stotzky, eds., Marcel Dekker Inc., New York. Hart, M.R., B.F. Quin, and ML. Nguyen. (2004). “Phosphorus Runoff from Agricultural Land and Direct Fertilizer Effects: A Review.” J. Environ. Qual., 33, 1954-1972. Hay, V., Pittroff, W., Tooman, E. E., and Meyer, D. (2006). "Effectiveness of vegetative filter strips in attenuating nutrient and sediment runoff from irrigated pastures." Journal of agricultural science, 144(4), 349-360. Hively, W.D., R.B. Bryant, and T.J. Fahey. (2005). “Phosphorus Concentrations in Overland Flow from Diverse Locations on a New York Dairy Farm.” J. Environ. Qual., 34, 1224-1233. Holden, PA. and Fierer, N. (2005). "Microbial Processes in the Vadose Zone." Vadose Zone Joumal, 4, 1-21. Jin, C.-X., and Romkins, M. J. M. (2001). "Experiemntal Studies of Factors in Determining Sediment Trapping in Vegetative Filter Strips." Transactions of the ASAE, 44(2), 277-288. 173 Kim, M.-J., Nriagu, J., and Haack, S. (2002). "Arsenic species and chemistry in groundwater of southeast Michigan." Environmental Pollution, 120, 379- 390. Kirby, R.M., Bartram, J. and R. Carr. (2003). “Water in food production and processing: quantity and quality concerns.” Food Control, 14(5), 283-299. Koelsch, R. K., Lorimor, J. C., and Mankin, K. R. (2006). "Vegetative Treatment Systems for Management of Open Lot Runoff: Review of Literature." Applied Engineering In Agriculture, 22(1), 141-153. Komor, SC. and 0.8. Hansen. (2003). “Attenutation of Runoff and Chemical Loads in Grass Filter Strips at Two Cattle Feedlots, Minnesota, 1995-98." Water-Resources Investigations Report 03-4036, Minnesota Pollution Control Agency. Larsen, |., Little, 8., Nealson, K.H., Ray, R., Stone, AT. and Tian, J. (1998). “Manganite reduction by Shewanella putrefaciens MR-4.” Am. Mineral, 83, 1564-1572. Larson, R.L. (2009). “Storm Water Runoff Characterization for Animal Feeding Operation.” ASABE 2009 International Conference, Reno Nevada, June 24- 29, 2009. Lee, K. H., lsenhart, T. M., and Schultz, R. C. (2003). "Sediment and nutrient removal in an established multi-species riparian buffer." Journal of Soil and Water Conservation, 58(1), 1. Lindsay, W.L.. (1985). “Solubility and redox equilibria of iron compounds in soils.” In: Iron in Soils and Clay Minerals, ed by J.W. Stucki, B.A. Goodman, and U. Schwertmann, Reidel Publishing Company, NorweIl, MA. Lorimor, J.C., L. Wulf, and P. Jaranilla. (2003). “An Infiltration-Wetland System for Treating Open Feedlot Runoff.” In: Proc. of the 9th Intemational Symposium, Animal, Agricultural, and Food Processing Wastes, St. Joseph, MI, ASAE, 405—410. Lovell, S. T., and Sullivan, W. C. (2006). "Environmental benefits of conservation buffers in the United States: Evidence, promise, and open questions." Agriculture, Ecosystems & Environment, 112(4), 249-260. Lovley, Derek R. June 1991. “Fe(lll) and Mn(lV) Reduction.” Microbiological Reviews, 55(2), 259-287. 174 Magette, W.L., R.B. Brinsfield, R.E. Palmer, J.D. Wood. (1989). “Nutrient and Sediment Removal by Vegetated Filter Strips.” Transactions of the ASAE, 32(2), 663-667. Master, Y., R.J. Laughlin, R.J. Stevens, A. Shaviv. (2004). “Nitrite formation and nitrous oxide emissions as affected by reclaimed effluent application.” J. of Env. Qual., 33(3), 852-860. Matocha, C.J., Scheckel, K.G., and Sparks, D.L. (2005). “Kinetics and Mechanisms of Soil Biogeochemical Processes .” In: Chemical Processes in Soils, No. 8, Tabatabai, M.A., Sparks, D.L., Al-Amoodi, L. and Dick, W.A., eds., Soil Science Society of America, Inc., Madison, Wisconsin. Mayer, P.M., Reynolds, 8. K., Jr., and McCutchen, M. D. (2007). "Meta-Analysis of Nitrogen Removal in Riparian Buffers." Journal of Environmental Quality, 36(4), 1172-1 180. McDaniel, P. (2006). "Anaerobic Processes." In: Encyclopedia of Soil Science, L. Rattan, ed., Taylor 8 Francis, New York. McGowen S.L. and NT. Basta. “Havey Metal Solubility and Transport in Soil Contaminated by Mining and Smelting.” In: Heavy Metal Release in Soils, ed by HM. Selim and D.L. Sparks, Lewis Publishers, Boca Raton. FA. MDEQ. (2006). “Arsenic in Well Water.” Michigan Department of Environmental Health, Michigan Department of Environmental Quality, Lansing, MI. http://www.michigan.gov/documentslarsenicbroch_41426_7.pdf Miyazaki, Tsuyoshi. (1993). Water Flow in Soils. Marcel Dekker, Inc, New York, NY. Mokma, D. (2008). Data Unpublished. Myoung-Jin, K., J. Nriagu, and S. Haack. (2002). “Arsenic species and chemistry in groundwater of southeast Michigan." Environmental Pollution, 120, 379-390. Nielsen, S. (2003). Food Analysis, 3’" Edition. Springer Science + Business Media, Inc., New York, NY. Nienaber, J.A., and CB. Gilbertson, T.M. McCalla, and FM. Kestner. (1974). “Disposal of effluent from a beef cattle feedlot runoff control holding pond.” Transactions of the ASAE, 17(2), 375-378. 175 Norvell, WA. (1988). “Inorganic Reactions of Manganese in Soils.” In: Manganese in Soils and Plants, ed by RB. Graham, R.J. Hannam, and NC Uren, Kluwer Academic Publishers, Boston, MA. Nye, J. C. (1982). “Runoff control from livestock feed lots. Research results in manure digestion, runoff, refeeding, odors.” North Central Regional Research Publication No. 284. Midwest Plan Service. p 10-17. Parry, R.. (1998). “agricultural Phosphorus and Water Quality: A US. Environmental Protection Agency Perspective.” J. Environ. Qual., 27(2), 258—261. Paul, EA. and FE. Clark. (1996). Soil Microbiology and Biochemistry, Second Edition. Academic Press, Inc., San Diego, California. Postrna, D., and Jakobsen, R. (1996). "Redox zonation: Equilibrium constraints on the Fe(III)/SO4 reduction interface." Geochim. Cosmochim. Acta 60, 3169-3175. Rawls, W. J., Brakensiek, D. L., and Saxton, K. E. (1982). "Estimation of Soil Water Properties." Transactions of the ASAE, 25(5), 1316-1320. Rittmann, BE. and PL. McCarty. (2001). Environmental Biotechnology: Principles and Applications. McGraw Hill, Boston, MA. Roden, EB and Zachara, J. (1996). “Microbial reduction of crystalline Fe3+ oxides: Influence of oxide surface area and potential for cell growth.” Environ. Sci. Tech. 30, 1618-1628. Rose, C. (2004). An Introduction to the Environmental Physics of Soil, Water and Watersheds, Cambridge University Press, Cambridge, United Kingdom. SAS. (2004). Version 9.1. Cary, N.G.: SAS Institute, Inc. Sahrawat. (2008). “Factors affecting nitrification in soils.” Communications in Soil Sci. and Plant Analysis, 39, 1436-1446. Saxe, C. (January 2007). “Managing Forage in Bunker Silos.” University of VVIsconsin Extension, Team Forage, Focus on Forage Factsheet, Vol. 3, No.6. http://www.uwex.edulces/cropsluwforage/ManageBunker.pdf. 176 Schmitt, T. J., Dosskey, M. G., and Hoagland, K. D. (1999). "Filter strip performance and processes for different vegetation, widths, and contaminants." Journal of Environmental Quality, 28(5), 1479-1489. Schueler, T. R. (1987). "Controlling Urban Runoff: A Practical Manual for Planning and Designing Urban BMPS." Metropolitan Washington Council of Governments, Washington, DC. Shuman, L. M. (2005). “Chemistry of Micronutrients in Soils.” In: Chemical Processes in Soils, No. 8, Tabatabai, M.A., Sparks, D.L., AI-Amoodi, L. and Dick, W.A., eds., Soil Science Society of America, Inc., Madison, Wisconsin. Smith, R.V., L.C. Burns, R.M. Doyle, S.D. Lennox, 8.H.L. Kelso, R.H. Fox, and , R.J. Stevens. (1997). “Free Ammonia inhibition of nitrification in river I sediments leading to nitrite accumulation.” J. of Env. Qual., 26(4), 1049- i 1055. Srivastava, P., D.R. Edwards, T.C. Daniel, P.A. Moore Jr., and TA Costello. (1996). “Performance of Vegetative Filter Strips with Varying Pollutant Source and Filter Strip Lengths.” Transactions of the ASAE, 39(6), 2231- 2239. Sweeten, J.M. (1998). “Cattle Feedlot Manure and Wastewater Management Practices.” In Animal Waste Utilization: Effective Use of Manure as a Soil Resource. Ed by J.L. Hatfield and BA. Stewart. Chelsea, MI: Ann Arbor Press. Tarradellas, J., Bitton G., and D. Rossel. (1997). Soil Ecotoxicology, CRC Press Lewis Publishers, Boca Raton, FL. Tchobanoglous, 6., Burton, FL, and Stensel, HD (2003). Wastewater Engineering: Treatment and Reuse, 4th ed. Metcalf and Eddy Inc. New York, NY: McGraw-Hill. Trulear, MG. and W.G. Characklis. (1982). “Dynamics of Biofllm Processes.” Journal of the Water Pollution Control Federation, 54(9), 1288-1301. US EPA. (1993). “Guidance Specifying Management Measures for Sources of Nonpoint Pollution in Coastal Waters.” Paper No. EPA 840-8-92-002. US EPA. (1996). SW-896 On-Iine. Washington, DC: US. Environmental Protection Agency's Office of Solid Waste. http://www.epa.gov/osw/hazardltestmethods/sw846/onlinefindex.htm 177 US EPA. (2003). “National Management Measures for the Control of Nonpoint Pollution from Agriculture.” EPA 841-8-03-004. US EPA. (2004). National Water Quality Inventory: Report to Congress. US EPA. (2007). “Secondary Treatment Regulation.” Code fo Federal Regulations, 40 CFR Part 133. US EPA. (2009a). “National Primary Drinking Water Regulations.” EPA 816-f- 09—004. US EPA. (2009b). “Water Science Analytical Methods: Title 40 Part 136: Protection of Environment.” Washington, DC: Electronic Code of Federal Regulations. http:l/ecfr.gpoaccess.gov/cgiltltextltext- idx?c=ecfr&rgn=div5&view=text&node=40:22.0. 1 .1 .1 &id no=40 Vandenabeele, J.M., Woestyne, M.V., Houwen, F., Germonpre, R., Vandesande, D., and Verstraete, W. (1995). “Role of autotrophic nitrifiers in biological manganese removal from groundwater containing manganese and ammonium.” Microb. Ecol., 29, 83-98. Vymazal, J. (2007). “Removal of nutrients in various types of constructed wetlands.” Science of the Total Environment, 380, 48-65. Wood, W.W., and Petraitis, J.M. (1984). “Origin and distribution of carbon dioxide in the unsaturated zone of the southern high plains of Texas.” Water Resour. Res, 20, 1193—1208. Yue—Mei, L., J. Tay, Y. Liu, and Y. Hung. (2008). “Biological nitrification and denitrification processes.” In: Handbook of Environmental Engineering Volume 8: Biological Treatment Processes, ed by: L.K. Wang, N.C. Pereira, and Y. Hung, Humana Press, New York, NY. 178 IES “11116111111111“iiiiiiiiiiifliiili