APPLICATION OF SWAT MODEL ON THREE WATERSHEDS WITHIN THE VENICE LAGOON WATERSHED (ITALY): SOURCE APPORTIONMENT AND SCENARIO ANALYSIS

Transcription

1 APPLICATION OF SWAT MODEL ON THREE WATERSHEDS WITHIN THE VENICE LAGOON WATERSHED (ITALY): SOURCE APPORTIONMENT AND SCENARIO ANALYSIS R. Salvetti*, A. Azzellino*, D. Gardoni*, R. Vismara*, M. Carpani**, C. Giupponi**, M. Acutis**, M. Vale***, P. Parati*** * DIIAR - Environmental Engineering Department, Technical University of Milan ** Department of Crop Science (Di ProVe), University of Milan *** ARPAV (Regional( Agency for Environmental Prevention and Protection in Veneto), Osservatorio Regionale Acque Interne 4th International SWAT Conference July 3 rd rd -5 th th 27

2 Aims of the study APPLICATION OF SWAT MODEL TO THREE BASINS OF THE VENICE LAGOON WATERSHED TO Assess the apportionment of point and non point sources Quantify the non point sources in terms of rain-driven and not-rain rain-driven diffuse sources Simulate a scenario analysis to assess the effect of a reduction of agricultural loads

3 Study area Dese Zero Watershed 29 km 2 Vela Watershed 11 km 2 Naviglio Brenta Bondante Watershed 37 km 2 About 35% of the VLW area ~ 5. tn y tn y -1 Venice Lagoon Watershed 238 km 2

4 Watershed characteristics NBB DZ VL Urban Area 24% 26% 13% Agricultural Area 69% 72% 86% Dominant crops Corn, soy, wheat, sugar beet Corn, soy, wheat, sugar beet Corn, soy, wheat, sugar beet Groundwater Recharge Area Contribution from external basins Very complex network of drainage/irrigation irrigation channels

5 Measures available S S S S WWTP Streamflow discharge data Industrial Water quality discharge data S S S

6 Sources apportionment DRY WEATHER LOADS WET WEATHER LOADS Point loads Diffuse loads Diffuse loads WWTPs discharge Industrial discharge Direct sewer discharge Direct discharge/instream measurements BOD/nitrogen mass balance Groundwater Surface recharge runoff loads Channel network USEPA QUAL2E BASINS-SWAT SWAT loads from WWTPs/Industries and from channel network loads from direct sewer discharge and from groundwater recharge

7 BASINS-SWAT input 2 soil classes 18 Subbasins 15 HRU 1 stations - 13 years daily meteorological data 33 Subbasins 169 HRU 42 Subbasins 13 landuse212 classes HRU years crop rotations 9 - four

8 SWAT calibration 8 Groundwater flow (m3/s) Groundwater estimates Groundwater simulated GW from external watersheds jan feb mar apr may jun jul aug sep oct nov dec Parameter calibration Additional inlet contribution BLAI (corn, wheat) HVSTI (corn, wheat) USLE_P ERORGN ERORGP NPERCO FRT_LY1 CN OV_N

9 SWAT model: hydrological calibration DDE station 1999 Nash 1 Sutcliffe coefficient of efficiency E ~ Measured 6 Predicted Predicted Measured Flowrate (m 3 s -1 ) m 3 s -1 January February m 3 smarch -1 April May June July August September October November December January February March April May 15/1/ 14/2/ June15/3/ July August Daily flowrate - DDE station Predicted Measured Nash Sutcliffe coefficient of efficiency E =.65 21/12/98 2/1/99 19/2/99 21/3/99 2/4/99 2/5/99 19/6/99 19/7/99 18/8/99 17/9/99 17/1/99 16/11/99 16/12/99 14/4/ 14/5/ 13/6/ 13/7/ 12/8/

10 SWAT model: RESULTS kg N during rainstorm events Predicted Observed Error rainstorm event ±15% 4 2 kg N 3May2 4May3 5May4 6May5 May(Rain_Event) 6June5 7June6 June(Rain_Event) 1Oct2 11Oct3 12Oct4 13Oct5 14Oct6 15Oct7 Oct(Rain_Event) Daily event Rainstorm event 5/5/96 7/5/96 9/5/96 11/5/96 13/5/96 15/5/96 17/5/96 19/5/96 Flowrate (m 3 s -1 )

11 5% SWAT model: RESULTS mean of 1 years simulation SWAT output WWTP and industrial discharge Tributary channels or irrigation systems Atmospheric deposition runoff Urban runoff 3% 13% 34% Total N annual load ~ 22 t N y -1 Total P annual load ~ 14 t P y -1 Present State - Ntot Direct sewer discharge Groundwater recharge Agricultural runoff 6% 8% 31% 2% runoff loads 15% point loads 65% dry weather diffuse loads WWTP and industrial discharge Tributary channels or irrigation systems Atmospheric deposition runoff Urban runoff 9% Present State - Ptot Direct sewer discharge Groundwater recharge Agricultural runoff 13% 24% 19% 35%

12 Implementation of an agricultural scenario kg N ha -1 y -1 kg P ha -1 y -1 Corn - 37% -17% Corn + manure -53% -27% Agricultural nitrogen runoff load 3 Present state 25 Agricultural scenario 2 tn/y DZ NBB VL Sum N = 139 tn y -1 (~ 5% agricultural nitrogen runoff load) N = 7 tp y -1 (~ 15% phopshorus runoff load)

13 Loads at the basin closure N N = 139 tn y Present state Agricultural scenario Nitrogen load at the basin closure (- 6 %) tn/y DZ NBB VL Sum Phosphorus load at the basin closure P P = 7 tp y Present state Agricultural scenario (- 5 %) tp/y DZ NBB VL Sum

14 Conclusions The application of SWAT model allowed to quantify the total annual nutrient load and to assess the source apportionment The dry weather diffuse sources (i.e. groundwater/spring recharge and tributary/irrigation channels coming from bordering watersheds) constitute the most important source (65% N and 35% P); Runoff loads cover about 2% of the total N load and about 3% of the total P load. Agricultural runoff constitute about 2/3 of the runoff load; Better-business agricultural scenario: reduction in agricultural runoff loads of about 5% for N and of about 15% for P decrease in the total annual load of about 5-6%. Most significant model outputs implemented in a decision support system software (mdss)