-S o u th w e s t/aasa FAUl. 5.< YSo^
River Wylye
Analysis and Modelling of Phosphates
Celia Bartlett & Neil Murdoch
May 1998 1 IN T R O D U C T IO N ...... 1 1.1 Background ...... 1 1.2 Objectives ...... 1
2 UPPER HAM PSHIRE AVON RIVER SYSTEM ...... 2 2.1 Location ...... 2 2.2 Point Source Discharges ...... 2 2.3 Routine Monitoring Sites ...... 3
3 DATA A N A L Y S IS ...... 4 3.1 Selected Data ...... 4 3.1.1 D isch arges...... 4 3.1.2 Tributaries ...... 4 3.1.3 D eterm inand...... 4 3.1.4 W ater Quality Monitoring ...... 5 3.1.5 River F lo w s...... 5 3.2 Seasonal and Longitudinal Variations in Phosphate Concentrations ... 6 3.3 Concentrations and Loads - Annual and S u m m er...... 8 3.3.1 C oncentrations...... 8 3.3.2 L o a d s ...... 14
4 ' M ODELLING ...... 20 4.1 A im s ...... 20 4.2 M odel Components ...... 20 4.2.1 D isch arges...... 20 4.2.2 Tributaries ...... 20 4.2.3 Gauging Stations ...... 21 4.2.4 Calibration and V a lid ation ...... 22 4.2.5 M odelling Scenarios...... 24
5 SUM M ARY and CONCLUSIONS ...... 26
© Environment Agency
NATIONAL LIBRARY & INFORMATION SERVICE
SOUTH WEST REGION
Manley House. Kestrel Way. Exeter EX2 7LQ River Wylye______Draft Model Report
1 INTRODUCTION
1.1 Background
This report was written as part of an ongoing study of the Upper Hampshire Avon catchment initiated in response to concerns raised by English Nature about the levels of phosphates in this area.
Statutory requirements which may enable the Agency to control phosphate levels are the Urban Waste Water Treatment Directive and the Habitats Directive. The UWWTD can be used to force nutrient stripping at PLC STW discharges if appropriate qualifying conditions are met. The Habitats Directive has been recently introduced and is designed to protect flora and fauna in designated areas.
So far, in the Hampshire Avon catchment, Salisbury STW is to have nutrient stripping applied so as to ameliorate the phosphate levels in the lower Avon. This was done under the UWWTD regulations. Other discharges in the catchment which come under the scope of the UWWTD are Christchurch STW and Warminster STW, the latter being situated on the River Wylye.
This report is concerned with the River Wylye from its source near Kingston Deverill south of Warminster to its confluence with the Nadder at Wilton. It is known, in general terms, that diffuse and point sources contribute to phosphate levels in the river. Diffuse sources arise from runoff from agricultural land* A rule of thumb appears to be that about 1 % of phosphates put onto the land will find its way into the river. Point sources may be large or small; the large ones being the PLC STWs.
The purpose of this report is to examine in some detail the point source contributions to phosphate levels in the Wylye. The intention is to provide information on which decisions about management of the catchment may be based, but not to make those decisions.
1.2 Objectives
The objectives are to:
A determine the spatial and temporal trends of phosphate concentrations along the River Wylye B quantify the seasonal and annual in-river and discharge phosphate loads C perform discharge modelling to quantify the impact of nutrient stripping
20th May 1998 1 C Bartlett & N Murdoch EA S W River Wylye Draft Model Report
2 UPPER HAMPSHIRE AVON RIVER SYSTEM
2.1 Location
The major rivers of the Upper Hampshire Avon catchment are shown below:
2.2 Point Source Discharges
The map below shows the monitored discharges in the upper Avon catchment.
20th May 1998 2 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report
2.3 Routine Monitoring Sites
The map below shows the river sites routinely monitored in the upper Avon catchment.
20th May 1998 3 C Bartlett & N Murdoch EA SW River Wylye Draft "Model Report
3 DATA ANALYSIS
3.1 Selected Data The present analysis is restricted to the River Wylye, which is shown schematically in Figure 4 from Kingston Deverill to its confluence with the River Nadder.
3.1.1 Discharges The main discharge to the River Wylye is from Warminster STW. Other larger discharges include those from Hurds (Hill Deverill) Fish Farm, Bishopstrow Fish Farm and Great Wishford STW - details of all these are given below.
Discharge URN Flow (m3/s) 0) Hurds Fish Farm 30117000 0.11 (MAX) Warminster STW 41406900 0.03 (DWF) Bishopstrow Fish Farm 30135000 0.16 (MAX) Gt Wishford STW 41415900 0.01 (DWF)
Table 1
(1) These values are taken from the associated discharge consents
3.1.2 Tributaries The tributaries considered are shown below. In the. case of the Were tributary the monitoring point 30131000 - Were Calveswater - has not been sampled since October 1995 and there is some uncertainty about its location; the concentration levels of phosphate at this URN are relatively high, perhaps indicating some influence from Warminster STW.
Tributary URN MeanFlow (m3/s)(2) Shear Water 30179000 0.136 Were 30131000 0.19 Chitteme Brook 30138000 0.668 Till River 30146000 0.896
Table 2
(2) These values are derived from the program Micro Low Flows.
3.1.3 Determinand The determinand studied is ortho-phosphate, which has determinand code number 0180.
20th May 1998 4 C Bartlett & N Murdoch EA SW Figure 4: River Wylye Diagrammatic representation of Point Discharges and Routine Monitoring Points KEY WARMINSTER 38.9km Distance downstream from Kingston Dewrtfl 10km y ' B e ra tn nia B1S9680Q (DWP"ft.4) N — ► ^jlril discharge JScrahvnBr 30133000 (MF-1.57B) • Routine Monitoring F^lnt (tarton S1476000 <«000) QUASAR_TRIB1 MF ■rmcnLkXa 81900800 (<1Q),eo*mny 1 (&JoaorFVw= 257) MIcrolowFlow* Mean Flow (ml/s) amrtbro<* 30100000 (MF-1.S7B) WERE_TRIB MFGS MIcroLowFlows Mean Flow for Gouging Station (ml/s) BbtKDrtrew Mtl 50134000 (MF-1.58*) ADF Gauged Awrege Deify Flow for GS (m3/s) (MF=\19: Qua«ariTcMP.399) FFerm 30115000 (<13«0 ButonVmr 91026900 (*8.4) M M I 30120000 CNtorne, AM mOo 31441900 TILL RIVER htayMury, BUcftnrtte* 31630900 (*13,8) <<«>. (MFi.896; Onwu*fk*«= 1.568) Tleheetf 41420000 KhookCnv 31476900 (<90) V.^ LmoB0130123000 BSOOSBr 3012 10 0 0 (M F»1 .062) f 4 Han C od M , HriAHo. 91474000 (4.S), (toavaPH 31643900 ( Hunj» C m s Bede and Pith Farm M C m M i 3012Q200 (MF-1.092) (Ufl Inn WO*** 31P76900 <<8.0 HQIDvCraMtMdt 97TMCT1/2000 HufdeGnMbede SO1402MMOO CM_Q« m TM i 37043000 (*14790) Ptou^i.Stodrten 81076M0 (<10^, M tknny • (<&79*<4779* HWEMTafm 30117D00 (<0100) »127200(UP-4 077) BemtckBUamee 41420900 (tW F -j02s» Soun lAwiifl rxnraxtt »l«W00 (MF=S J!5. WFG&s:j i'f.«r=-1 OW MFGt=.7, ADF-.DS3I O S f Ot WWifcrt flTW 41415900 (DWF=7fl1) MatfaiDev 316215000 < 41428800 (EW «30) KBmkxjfen, (M BMid Oidhampton 00245000 (MF»5.33> ' S 91491000 (*16) Kfcnkiglon The M m I WILTON sisa4000( 3.1.4 Water Quality Monitoring Figure 4 is a schematic representation of the River Wylye showing all monitored discharge points and routinely sampled river sites. Essentially, these are the sites which exist on the W1MS archive and have been sampled at some time during the period 1994- 97 Many of them are very small discharges and have been included solely for the purposes of completeness. A number of them do not discharge directly into the river - where it is known that the discharge is into a soakaway, this has been indicated on the figure. In retrospect, the inclusion of 1997 in the data analysis was superfluous, as monitoring of discharges for ortho-phosphate concentrations was only carried out at UWWTD designated sites during 1994-6. The year 1997 has been excluded from some of the analysis and from the modelling due to the lack of ortho-phosphate discharge data. Shown in Figure 4 is a diagram of point discharges and routine monitoring points. The text in red indicates those points which have been incorporated into the model - see Section 4. 3.1.5 River Flows The program MicroLowFlows (MLF) was used to derive the mean flows along the length of the Wylye and the mean flows of the main tributaries at their point of confluence. MLF is essentially a database of natural flows (no account is taken of abstractions and discharges), based on geology, topography, precipitation, etc/ Gauging Stations: There are three gauging stations on the Wylye. Station Name NGR Ref. No. Q95(1995) ADF(1995) (m3/s) (m3/s) Brixton Deverill ST 8580 3810 432310 0.065 0.353 Norton Bavant ST 9090 4280 432320 0.441 1.089 South Newton SU 0860 3430 432330 1.170 4.084 The associated MLF mean flows for these points give mean flow values of 1.043, 1.833 and 5.245 m3/s at these gauging stations, which are significantly higher than the ADF's given above, particularly near the head of the river at Brixton Deverill. This highlights the problem of abstractions and subsequent low flows in the river. Average monthly flows at each of the gauging stations for 1994-97 were downloaded from the archives and used both for calculating the loads and for the modelling. This was done by scaling the monthly flows for the nearest gauging station by the ratio of the MLF mean flow at the point in question and the MLF mean flow at the gauging station. 20th May 1998 5 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 3.2 Seasonal and Longitudinal Variations in Phosphate Concentrations Shown in Figures 5 and 6 are yearly time series (1994-7) for ortho-phosphate concentrations upstream and downstream respectively of Warminster STW (URN’s 30129000 and 30130000). Seasonal Variations The main point to note at both sites is that there is a strong seasonal cycle. The concentration generally peaks during late summer and autumn in all years, when flows are low, and is frequently higher in 1997, a year of low rainfall, than in earlier years. When flows are low there is less available water for dilution, and it is at these times that the impact of the constant point sources predominates over that of diffuse sources resulting from land runoff. Longitudinal Variations Due to exigencies of time, a detailed analysis on variations in concentration along the length of the river has not been done. Figures 5 and 6 show the impact of Warminster STW - concentrations downstream of the STW are generally an order of magnitude higher than those upstream. Variations in average annual and summer river concentrations along the length of the River Wylye are shown in "bar chart" form in Figures 10 to 12 and 16 to 18 respectively. 20th May 1998 6 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report r,.wr.»n.Miu tiivnuiMiv 30123000 04-1T-1990 cneo Figure 5 Ortho-Phosphate Concentrations in mg/1 upstream of Warminster STW ^ a b is O O O O 0 4 -1 T - 13 39 owo Figure 6 Ortho-Phosphate Concentrations in mg/I downstream of Warminster STW 20th May 1998 1 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 3.3 Concentrations and Loads - Annual and Summer The figures on the following pages show the following: average annual ortho-phosphate concentrations and loads for discharges and for in-stream monitoring points 1994-96 average summer ortho-phosphate concentrations and loads for discharges and for in-stream monitoring points 1994-96 Also shown in these figures are the number of samples at each point discharge or river monitoring point. As already mentioned on p.5 very little monitoring of ortho-phosphate * in discharges was done during 1997, and this year has been excluded from the analysis in this section. Summer was deemed to be the months of June, July and August. 3.3.1 Concentrations Discharges Average annual point source concentrations vary considerably, ranging from about 5 to 20 mg/1. I should be borne in mind that the flows of some of these are negligible, and a more realistic idea of their impact can be derived from consideration of the loads figures. The discharge of most concern - Warminster STW has an average annual concentration of about 6mg/l. Average summer point source concentrations are about the same as the annual values, as would be expected. In-stream river monitoring points Average annual in-stream concentrations at the upstream end of the Wylye are of the order of 0.1 mg/1, and about double that at the downstream end where the river joins the Nadder. The highest concentrations are immediately downstream of Warminster STW, where they range from approximately 0.6 to 0.8 mg/1. What is interesting to note is that the high concentrations downstream of Warminster STW persist for a distance of some 4 km or more. The average summer concentrations in the river are generally about .2 mg/1 higher than the average annual concentrations. 20th May 1998 8 C Bartlett & N Murdoch EA SW River Wylye Draft ModeI Report KEY TO ACCOMPANY FIGURES 7 T030 The following list gives the full name and URN for the points plotted along the x-axis. Discharges In-Stream Monitoring Abbreviation/Full Name/URN Abbreviation/Full Name/URN Klml/Kilmington The Street/31524900 BxDv/BrixtonDeverill/3018800 Klm2/Kilmington Old Sawmills/31491900 usHdff/upstreamHillDeverillFishF arm/3 0116000 MdBr/MaidenBradley/41429900 inHdff/InletHillDeverillFishFarm/30143000 MkDv/MonktonDeverill/31526900 ??? /Old sampling point-DoubtfiilPos/30145000 Hdff/HillDeverillFishFarm/30117000 us Hcr/upstream Hurd sCressB eds/30121000 Hcrl to 6/Hurds CressBeds/37040/1/2000 inHcus)lnlet to Hurds Cress Beds and 301452/4/600 )upstream Settlement Tank/30144100 GeoP/George PH/3153900 inHcds)downstream Settlement Tank/30144200 LbD v/L'bridgeDe v'UHoSTW/31591900 dsHcr/downstreamHurdsCressBeds/30120200 JbMl/JobsMill/31546900 B3095/B3095Bridge/30121000 Wstw/WarminsterST W/41406900 LbDv/LongbridgeDeverill/30122000 HuMn/HuntersMoon/31979900 ShrCr/ShearWaterCrockerton/30178000 BhMl/BorehamMill/31535900 ShrCf/ShearWater at Confluence/30179000 WmGr/WarminsterGarr/31475900 BlMl/BullMill/30126000 HmLn/HarmanLines/31980900 JbMl/JobsMill/30127000. B wff/B i sh op stro wF i shFarm/30135000 usWST/upstreamWarminsterSTW/30129000 Nbav/NortonBavantFishFarm/30125200 HfMa/HenfordMarsh/30130000 SuVy/WWDCSuttonVeny/31528900 usWer/upstreamWereConfluence/3 0149000 Wlnd/WWDC Westlands/31523901 WerCf/WereCal veswaterPS/30131000 SUH/SUohnsHosp/31538900 SmBk/SmaIIbrook/3 0150000 KnCm/KnookCamp/31476900 BoBr/BorehamBridge/30133000 HpEt/Happy Eater/31579900 BwMI/BishopstrowMiII/30134000 CdfdH/CodfordHall&Ho/31474900 inB wff/InletFishFarm/30135100 ChAb/ChittemeAbdonClo/31481900 PtMd/PitMeadsFB/30136000 ChS W/Chitteme/41434900 inNbff/InletFishF arm/30125100 ChNR/CodfordNewRoad/31510900 NBav/NortonBavant/3 0125000 ChOr/Cod ford/Cherry Orchard/31501900 Chcd/CodfordChitterne/3013 8000 Plow/Plough Stockton/31976900 Cdfd/CodfordStMary/30127100 TlTlhd/Tilshead/41428900 StLd/SteepleLangford/30127200 TlShr/Shrewton/41412900 TlSf/Stapleford/30146000 TIBl/Bell lnn/31978900 . SNew/SouthNewton/30128000 TIMd/SDCMeadow View/31529900 Qdhn/Quidhampton/30245000 TlBSU/BerwickStJ ames/41420900 GtWf/GtWishfordSTW/41415900 20th May 1998 9 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report River Vtylye Discharges (Annual) Average annual Ortho-Phosphate Concentrations - 1994 f - y e ten mi hcm ho* uov vmw mi uti* tav vm ktodC^qgw ck> Tm to tbslj rrt? l*Ov Hoi H o ) He* J M HM> WttO M SuVy SJH QM CM) Bov TOr TM GfW River Wylye Discharges (Annual) Average annual Ortho-Phosphate Concentrations - 1995 4 . 4 - 4 - .JL lii- L lL I I Q i i i - i . A. 1 nnl lie *OI Hcr2 Hot U£* Wktw M l l+Ttn tte VM ttCmCcfiMOeW QOr TTTW TH TB9U nrQ M(Ov Hai Hcrt KfcrS G«cP JtM hU^WrTO M Stf* SUH H|£l CMt> CM3 Fte TOt TttAJ COM River Vtylye Dischaiges (Annual) Average annual Ortho-Phosphate Concentrations - 1996 U i .i * i . i . A . ill ml U f i KCn Hcr2 k m Ho6 ibOv B M Mrln i e * M H ttOnCdcMOSW O O f TTThd TB TBSu KM Md>> HcM Hert ftoS GtcP JtM HMMMrCi M SA/y SUH OXt, a t f t n » Toy HUD GfM Figures 7 to 9 Average Annual Discharge Ortho-Phosphate Concentrations 1994-6 20th May 1998 10 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report Figures 10 to 12: Average Annual In-stream Ortho-Phosphate Concentrations 1994-6 20th May 1998 11 C Bartlett & N Murdoch EA SW River Wylye Drafi Model Report River Wylye Discharges (Summer) Average summer Ortho-Phosphate Concentrations - 1994 i t a i l U f i Hon hfcrt HcH LbO* VW*» M l tM ji « M NCm CdW CWW O O t TTht TB TB Nnfl M d » Hen Hen Hoi GmP JUU KfcMiWitt Bnfl GiMy S JH t*EI CH»6 Plow TT9f TMt 1 River Wylye Discharges (Summer) Average summer Ortho-Phosphate Concentrations - 1996 f » S t i n 1 YU HOT Ho2 Hert Hot UOv W tt* M i HrOi WH IKniCd^OfiMQOf TTDd TB TBSU Art? hfcDv K*1 Hcrt M G « ? l i l M SJH h^Q C M C t M TOr T M GMV Figures 13 to 15: Average Summer Discharge Ortho-Phosphate Concentrations 1994-6 20th May 1998 12 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report Figures 16 to 18: Average Summer In-stream Ortho-Phosphate Concentrations 1994-6 20th May 1998 13 C Bartlett &.N Murdoch EA SW River Wylye Draft Model Report 3.3.2 Loads Discharges Constant flow rates were used for the entire year - these were either the DWF or the Maximum value shown on the consent. Warminster STW has the highest average annual discharge load at approximately 0.3 g/s. Gt Wishford and Tillshead both have loads of about 0.1 g/s, but the latter is a soakaway so does not have a direct effect on the river loads. Bishopstrow fish farm contributes a load of approximately 0.05 g/s. The average summer discharge loads are much the same as the annual loads, as would be expected, given that the same discharge flow rates were used for both annual and summer loads. Figures 31 and 32 show a comparison of in-stream and discharge loads for the years 1994-6 both as annual averages and as summer averages. Also given on these figures are the average annual and summer flows at the downstream end of the system. These figures clearly show that on an annual basis, the discharge load is less than the downstream load during the wetter years, but is equal to it in 1996. In the case of summer loads, the discharge load is consistently higher than that downstream. In-stream River Monitoring Points Monthly loads were calculated by multiplying the mean monthly flow (evaluated as described on p.5) and mean monthly sample concentration values. Annual mean loads were found by taking the mean of the monthly loads. There is a buildup of the average annual in-stream load on proceeding downstream, reflecting the increasing flows. The load exported from the system is given by the load at the downstream end, which is approximately 0.8 g/s in 1994 and 1995 and 0.6 g/s in 1996, a drier year. The sum of the discharge loads is roughly 0.6 g/s, so that during the • two wetter years there is a net export of phosphate from the system, whereas in the drier year, 1996, there is no export, and, possibly, even an accumulation of phosphate. In summer, the load at the downstream end (approximately 0.4 g/s) is always less than the sum of discharge loads, indicating that during these summer months there is a buildup of phosphates; probably bound up within the river sediments. Export of the accumulated phosphate relies on their being sufficient winter rain to flush out the phosphate bearing sediment. 20th May 1998 14 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report River Wytye Discharges (Annual) Average annual Ortho-Phosphate Loads - 1994 FI n * • • • •n i II i Itrtt i a m M & hCTI Ha7 Ktafl UOi Wfctw M l ttiUi l » r VMnd KnCmaf* River Wylye Discharges (Annual) Average annual Ortho-Phosphate Loads * 1995 ♦ ♦ KM tt& ten Hoi M LH> Wu» M l Mrln ten Vrtnd NOnCdtM QfiW O O r TTThd 7 B TBSU ‘ me 1*0 Hoi No] Hoi G«P JtM HJ*iWnOr Bufl SNy 3JH Hfi CDM> OMR Moo TOr TMd GfW River Wylye Discharges (Annual) Average annual Ortho-Phosphate Loads - 1996 Ao- -Ri QjA l Uni t t f i HJB H a l hk>4 M LfcO. MWif. » M rttLii h t e M nt KiCmCd Figures 19 to 21: Average Annual Discharge Ortho-Phosphate Loads 1994-6 20th May 1998 15 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report River Wylye In-stream Monitoring (Annual) Average annua) Ortho-Phosphate Loads -1994 " - 7 & ♦ ♦ ♦ ♦ ► m \ » i ! if £ K- y i i 2 s ■? i?- f,’;; 1 5 $ & & | -dOoOU .mm i i i *- i H BiDv u»Ha rH aii KXB5 SiCr BM isWST S n A M l RUd NOiv Q fd T13 isttff 7T? M u 6Hcr LbO fiatO JtM HMa VVmO &G MOT Chcd aid 9 River Wylye In-stream Monitoring (Annual) Average annual Ortho-Phosphate Loads -1995 IHE3ElE].i--k ■* .+.E3EEL .. 1 Ha BiDv taHdl ibHd iN a h KJCBS SiCr BM t*WST i*VV» Srfli BuM FV« H Bn CUF6 TtST i i W 7T7 kH aa cbHcr LbO Srrt3 MS m b W n a EM> ■£* iMM CUcd StLd 9 River Wylye In-stream Monitoring (Annual) Average annual Ortho-Phosphate Loads -1996 - -nj' * y 15 'B *- ♦ I I & 5 1-j L- i s i *1 n m 3 % &£> rt-WT i*Her ritaa ffXBS ShO BM i*W5T isW* M RMj C Figures 22 to 24: Average Annual In-stream Ortho-Phosphate Loads 1994-6 20th May 1998 16 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report River Wylye Discharges (Summer) Average summer Ortho-Phosphate Loads - 1994 j . . ELI m l M B HER Hert Ha* H a t UOv Wfciw 9 M Kkrtn » » Wnd KK^nCdcHCxSiV OCK TTTKf TO TB3J K M M Or t o l t o ] t o ] G«*> MA KU*i VttiCr M l &Ny SJM ►*£! ChM O N Fto. R 9r T M G f« River Wyiye Discharges (Summer) Average summer Ortho-Phosphate Loads - 1995 4 "B j _ k X 4 4 I ml M B hEH Ho2 Ha* m e LbOv VMw 9 M Kkrtn i a « Mud ttCm C A M O SW OOt TTThd T » T fflJ K M M D r Hoi Her] HaS C cP JIM KtMiW rfir M BWf HJH H£I CMb C M Pkw TISv TUd CfM River Wylye Discharges (Summer) Average summer Ortho-Phosphate Loads - 1996 [Ua£eiiaUt^*jn^fl!MF^UAXv£^^ i-Eli mi MB HOT Ma2 He*» KtoS UOv Wu» 9M ►Mji lb« MM KiCm ca<« aiSW CW> TTTM|L TO T8aj KM UOr KfeM to ] HaS GmOP JbU W ti WhO B*l SuVy SJH 0 * b OMR Ro» TB » TMH GIW Figures 25 to 27: Average Summer Discharge Ortho-Phosphate Loads 1994-6 20th May! 998 17 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report River Wylye In-stream Monitoring (Summer) Average summer Ortho-Phosphate Loads -1994 I on £ i; £ p ♦ & 1 i I l $ k I s i 1 3 Jl|i Brf* tnHifl lalta rttah SX85 S iO BM 0*W5T w W v Sirfli B«M PIVU N b* O f 6 TtSI Oitn laHfT 777 M u t* H a LtO* S iO JtM I M Wtaa Bc& i M rM/I Q n t SUf SN>w River Wylye In-stream Monitoring (Summer) Average summer Ortho-Phosphate Loads -1995 _ » ♦ ♦ # • & * ! 1 I t 1 $ j: if * ff* £ & J3@ ii m BiDv irtIT uiHct rttah B3QB6 S i O BM wWST itVlta Srr& M l FU4J N ta Ctfd T O 0 River Wylye In-stream Monitoring (Summer) Average summer Ortho-Phosphate Loads -1996 - i — i . & Dv hrHcff tahkx fttcdi B3QB5 9iC r M i*W5T v M r d T B M l RMd N to C Figures 28 to 30: Average Summer In-stream Ortho-Phosphate Loads 1994-6 20th May 1998 18 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report Figure 31: Comparison of Annual Discharge and Downstream Loads River Wyle Summer Loads Comparison of Instream and Discharge Loads Figure 32: Comparison of Summer Discharge and Downstream Loads 20th May 1998 19 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 4 MODELLING o 4.1 Aims The aims are: a To improve our understanding of the behaviour of phosphates in rivers, so as to improve the modelling b Estimate the effect of nutrient stripping for the Warminster STW discharge. 4.2 Model Components Whilst load analysis provides a succinct summary of the catchment features, modelling gives a better insight into the impact of individual point sources, and the consequences of modifying a discharge, through nutrient stripping. 4.2.1 Discharges All the smaller discharges were excluded from the model as these have a very small impact on the system. The discharges which were included are as follows: Hill Deverill Fish Farm Warminster Bishopstrow Fish Farm Gt Wishford 4.2.2 Tributaries One of the main problems encountered in the modelling was that of water accretion along the river. The Wylye runs through a chalk landscape and there is considerable seepage of water from the ground via springs and channels. This seepage was allowed for in the QUASAR model by the introduction of artificial tributaries at points along the river. Actual tributaries were as follows: Shear Water - Were Chitteme Brook Till River The flows for these tributaries were derived using MLF and the monthly flow records for the nearest gauging station in the same way as already described on p. 5 Two additional tributaries were introduced near Norton Bavant and near Knook Camp to allow for water accretion. 20th May 1998 20 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 4.2.3 Gauging Stations There were three fairly equally spaced gauging stations along the river, at Brixton Deverill, Norton Bavant and South Newton, and data from these was used in conjunction with MLF to derive monthly flows at the upstream limit of the model and for the tributaries. A schematic figure of the model structure is shown below. Figure 33: Schematic View of Components of River Wylye Model 20th May 1998 21 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 4.2.4 Calibration and Validation Only the years 1994 and 1995 were modelled as only for these two years is there a full set of sample ortho-phosphate data. Outputs for the model reaches were compared against observed concentrations at routine monitoring points within these reaches. The results are shown in figures 34 to 39. These figures show monthly values of modelled and observed ortho-phosphate for each reach throughout the year. The data for 1994 was used to calibrate the model and 1995 data used for validation. Phosphate treated as Conservative The initial modelling run was done treating ortho-phosphate as conservative. Suprisingly, this gave quite good agreement with observations for reaches as far downstream as Knook, but modelled concentrations further downstream of this were persistently higher than those observed, as would be expected, given that phosphates are generally perceived to be subject to decay processes. Phosphate treated as subject to Decay The phosphate decay process was modelled by an equation of the form: d— =-k(C-Cn) d t 0 Here 'C0 ' is some equilibrium/saturation concentration value. This equation has been used in previous modelling work on the Upper Hampshire Avon. The value chosen for k was .08 and for C0 0.02. These same decay parameters were used for all reaches. The results obtained from the modelling runs using these parameters were rather disappointing. Although there was a reasonable fit with observations at the downstream end, the fit was worse for the middle reaches of the river. General Comments The 'spikiness' of the 1995 results can be attributed to the rather low sample values recorded for Warminster STW phosphate discharges during July and August of that year. The resulting drop in modelled concentrations downstream of Warminster persists to at least Norton Bavant, a distance of 4km, again indicating the strong influence of the STW discharge. The sudden increase in concentrations in December 1995 downstream of the Chitteme tributary is a result of the very high (1.5 mg/1) sample value recorded in the tributary at URN 30138000 for that month. The consistent under prediction of phosphate levels during the winter may. result from the fact that the model has no mechanism to represent flushing effects on sediments. The fact that it under predicts implies that there is an unmodelled source of phosphates, which could be due to winter flushing. 20th May 1998 22 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report Concentrations of Ortho-Phosphate along the River Wylye -1994 Monthly values for entire year for det code 0180 shown at each stretch Figure 34: 1994 Ortho-Phosphate Concentrations Modelled (no decay) and Observed Concentrations of Ortho-Phosphate along the River Wylye - 1994 Monthly values for entire year for det code 0180 shown at each stretch Figure 35: 1994 Ortho-Phosphate Concentrations Modelled with decay and Observed 20th May 1998 23 C Bartlett & N Murdoch EA SfV River Wylye Draft Model Report Concentrations of Ortho-Phosphate along the River Wylye -1995 Monthly values for entire year for det code 0180 shown at each stretch Figure 36: 1995 Ortho-Phosphate Concentrations Modelled (d o decay) and Observed Concentrations of Ortho-Phosphate along the River Wylye - 1995 Monthly values for entire year for det code 0180 shown at each stretch 1/sHaDvFf cVsHflCVFf 1/sVWnST. tfsWnST i/sBstroFf tfsBstroFf tfsNortB* tfsKnxk tfsCMltem i/sGWST tfsGWST Figure 37: 1995 Ortho-Phosphate Concentrations Modelled with decay and Observed 4.2.5 Modelling Scenarios One scenario run was done to assess the effect of reducing STW ortho-phosphates to a level which represents the realistic minimum which could be imposed - this was taken as 2 mg/1. The phosphate concentrations were modelled using the decay parameters mentioned above, although as stated already, these did not result in a very good model fit, which limits the reliance which can be placed on the modelled predictions. 20th May 1998 24 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report As can be seen from the figure, reducing the Warminster STW ortho-phosphate discharge levels to 2 mg/1 certainly effects a significant reduction in concentrations for a considerable distance (approximately 10km) downstream of Warminster, but concentrations at the downstream end are only slightly reduced. 1994 Concentrations of Ortho-Phosphate along the River W ytye- S T W discharge of 2mg/l Monthly values for entire year for del code 0160 shown at each stretch 1.2 — Modelled with decay +■ Observed o.s 0.6 ' **■ 0.4 02 i/sHODtfff tfsHflDvfl i/sVttnST dteVWnST i/s&trcff tfsfctroFf tfsNortBw tfsKna* tfsOittem tfsGWST tfsGWST Figure 38: 1994 Ortho-Phosphate Cones Modelled with decay (STW 2 mg/1) and Observed 1995 Concentrations of Ortho-Phosphate along the Rver Wylye - S T W discharge of 2mgfl Monthly values for entire year for det code 0180 shown at each stretch______ 1.5 -t- — Modeled with decay + Observed +■ as i- iVsHflDvff tfsKIDvfl i/sVMnST tfsWknST t/sEGtrcfl (ftBstrcFf cVsNortBaw tysKro* tfsCKltem i/sGWST tfsGWST Figure 39: 1995 Ortho-Phosphate Cones Modelled with decay (STW 2mg/l) and Observed. 20th May 1998 25 C Bartlett & N Murdoch EA SW River Wylye Draft Model Report 5 SUMMARY and CONCLUSIONS 1 Phosphate concentrations in the River Wylye exhibit a strong seasonal cycle, with peak values of over 1.0 mg/1 in August downstream of Warminster STW and minimum values of around 0.08 mg/1 in May in reaches upstream of Warminster. 2 Warminster STW has a major impact on the longitudinal downstream phosphate profiles. The small, non PLC discharges have negligible impact on in-situ concentrations. 3 In wetter years the discharge loads are approximately three quarters of the load carried by the river and phosphate is exported from the system. During dry years the discharge loads may exceed the load carrying capacity of the river and phosphate accumulates in the system. 4 In the summer months, the discharge loads always exceed the downstream load and there is a buildup of phosphate in the system. 5 The attempt to model the concentration of ortho-phosphate was unsatisfactory, although the results generally supported the conclusions reached by the analysis of loads and concentrations - see 1 to 4 above. 6 Trying to simulate the accretion of water along the river when undertaking the modelling caused some problems, and 'lumping' this accretion together into artificial tributaries is not a very satisfactory solution. 7 The main problem in trying to calibrate the model was that high in-stream concentrations in the middle reaches could only be adequately simulated by assuming that phosphate behaved in a conservative manner, whereas lower downstream concentrations could only be simulated by imposing some decay on the phosphate. More modelling work needs to be undertaken to resolve this. 8 The problems outlined above may be explained by their being other sources of ortho-phosphate in the reaches of the Wylye between Bishopstrow and Codford - these could be either point discharges, groundwater infiltration or runoff from the land - no attempt has been made to model the latter. 20th May 1998 26 C Bartlett & N Murdoch EA SW