Ahmed Nasr 1*, Michael Bruen 1, Richard Moles 2, Paul Byrne 2 & Bernadette O'Regan 2

Transcription

1 The significance of the differences in soil phosphorus representation and transport procedures in the SWAT and HSPF models and a comparison of their performance in estimating phosphorus loss from an agriculture catchment in Ireland Ahmed Nasr 1*, Michael Bruen 1, Richard Moles 2, Paul Byrne 2 & Bernadette O'Regan 2 1 Centre for Water Resources Research, Civil Engineering Department, University College Dublin, Earlsfort Terrace, Dublin 2, Ireland. ahmed.nasr@ucd.ie 2 Department of Chemical and Environmental Science, University of Limerick, Limerick, Ireland.

2 ORGANIZATION OF THE PRESENATION Background of the water quality situation in Ireland The Clarianna catchment Objectives Outlines of the SWAT and HSPF models (flow and phosphorus components) Results Conclusions

3 SITUATION IN IRELAND - (1) Percentage of channel length in four biological quality classes Class A: Unpolluted Class B: Slightly Polluted Class C: Moderately Polluted Class D: Seriously Polluted There is a continual increase in slight and moderate pollution (Classes A and B) in Irish rivers at the expense of previously unpolluted (Class A) watercourses. Overloaded sewage treatment works (point source pollution) are likely to be responsible for seriously polluted rivers.

4 SITUATION IN IRELAND - (2) Inputs of nutrients (particularly phosphorus) from diffuse sources associated with agriculture are the primary causes of the increased levels in slight to moderate pollution of Irish rivers. Inputs of P to the Irish soils : Trend of fertiliser phosphorus application over the Irish soils. Fertiliser P (Tonnes) Tonnes of P come from animal wastes.

5 Point Source Pollution SITUATION IN IRELAND - (3) Tacking this type of pollution has been addressed by upgrading the existing sewage treatment plants (including tertiary process). Diffuse or Nonpoint Source Pollution This type of pollution still remains to be tackled however a Catchment-Based Strategy has been set to mitigate its effects. (One attempt is to employ existing physically-based models to quantify phosphorus losses from a number of Irish agriculture catchments).

6 LOCATION OF THE CLARIANNA CATCHMENT $ $ $

7 DEM DEM, SOIL AND LAND USE MAPS OF THE CLARIANNA CATCHMENT Soil Map Land Use Map

8 OBJECTIVES OF THE STUDY (1) To test the significance of phosphorus loss modelling in the SWAT and HSPF models in simulating the flow discharge and the total phosphorus (TP) at the outlet of the Clarianna catchment for the 2.00 period 1/12/ /7/2002 Flow Hydrograph Total Phosphorus Graph (concentration) Total Phosphorus Graph (Load) Qobs (m 3 /sec) TPobs (mg/l P) TPobs (kg P) Dredging

9 OBJECTIVES OF THE STUDY (2) To test the significance of the phosphorus loss modelling in the SWAT and HSPF models in simulating the flow discharge and the dissolved reactive phosphorus (DRP) at the outlet of the Clarianna 2.00 catchment for the period /12/ /7/2002 Flow Hydrograph Dissolved Reactive Phosphorus Graph (concentration) Dissolved Reactive Phosphorus Graph (Load) Qobs (m 3 /sec) DRPobs (mg/l P) DRPobs (kg P)

10 CONCEPTUAL REPRESENTATION OF THE WATER DYNAMIC MODELLING IN THE SWAT MODEL Precipitation Vegetation Storage Fraction of Precipitation to soil surface Direct Runoff Surface Storage Infiltration Lateral Flow Total Flow Subsurface Storage Percolation Base Flow Groundwater Storage Interception, Surface Runoff, Infiltration, Evapotranspiration, Lateral flow, Percolation, Baseflow

11 CONCEPTUAL REPRESENTATION OF THE WATER DYNAMIC MODELLING IN THE HSPF MODEL SURFACE ZONE UPPER ZONE INTER FLOW ZONE LOWER ZONE SURFACE FLOW INTER FLOW TOTAL FLOW GROUND WATER ZONE BASE FLOW The infiltration distribution is focused around the two lines which separate the moisture available to the land surface into what infiltrates and what goes to interflow.

12 PHOSPHORUS MODELLING IN THE SWAT MODEL (A) Soil Phosphorus State Variables as described by SWAT

13 PHOSPHORUS MODELLING IN THE SWAT MODEL (B) Mineralization/immobilization of active organic phosphorus :- P = 1. 4 β γ γ P transport in runoff water :- ( ) P sol Q = ρ b P D k d Q P Q P sed P transport attached to sediment :- ( ) P min sed tmp SY = Patt ε A sw P act Pact : active organic P β : rate coefficient of mineralization γtmp : nutrient cycle temperature factor γsw : nutrient cycle water factor Mineralization/immobilization of fresh organic phosphorus :- P = δ P P fresh : fresh organic P dec ntr fresh ( ) Adsorption/desorption :- P ads/des pai Pads/des = Psol Pads 1 pai δ ntr : residue decay constant Psol : solublemineralinorganicp Pads : adsorbedinorganicp pai: phosphorusavailabililty index Psol : soluble P in the top layer ρb : bulk density of the soil D : depth of the top soil layer kd : soil P partitioning coefficent ( ) P min ( ) P dec P att : amount of P in the soil parent material SY : sediment yield A : Area of the land ε : P enrichment ratio

14 PHOSPHORUS MODELLING IN THE HSPF MODEL (A) Soil Phosphorus Cycle

15 PHOSPHORUS MODELLING IN THE HSPF MODEL (B) Adsorption/desorption, mineralization, immobilization, plant uptake using first order kinetics :- P flux : phosphorus flux P ( T 35 P = P K θ ) stor : phosphorus storage flux stor K :first order rate parameter for the process θ : temperatu re correction factor for the process T :soil temperatu re P transport in runoff water :- ( ) P Q P P Q sol Q = sw P transport attached to sediment :- ( ) P sed P = Ratio sed P surf P sol : soluble P in the Q : surface runoff sw : soil water P surf : Ratio storage of P in the top surface layer : ratio of sediment eroded to exist in the surface layer layer that

16 RESULTS OF THE FLOW AND TP SMULATIONS (1) 0.3 Simulation with SWAT model 0 TPobs, TPest (mg/l P) TPobs TPest Qobs Qest Qobs, Qest m 3 /sec) Simulation with HSPF model 0 TPobs, TPest (mg/l P) TPobs TPest Qobs Qest Qobs, Qest m 3 /sec) 3

17 RESULTS OF THE FLOW AND TP SMULATIONS (2) TPobs, TPest (kg P) TPobs, TPest (kg P) Simulation with SWAT model TPobs TPest Simulation with HSPF model TPobs TPest 0

18 RESULTS OF THE FLOW AND TP SMULATIONS (3) Cummulative TP - Observed Cummulative TP - Simulated (SWAT) Cummulative TP - Simulated (HSPF) 250 TP (kg P)

19 RESULTS OF THE FLOW AND DRP SMULATIONS (1) Simulation with SWAT model DRPobs, DRPest (mg/l P) DRP DRP Qobs Qest Qobs, Qest (m 3 /sec) Simulation with HSPF model 7 0 DRPobs, DRPest (mg/l P) DRP DRP Qobs Qest 1 Qobs, Qest (m 3 /sec)

20 RESULTS OF THE FLOW AND DRP SMULATIONS (2) Simulation with SWAT model DRPobs, DRPest (kg P) DRPobs DRPest DRPobs, DRPest (kg P) Simulation with HSPF model DRPobs DRPest

21 RESULTS OF THE FLOW AND DRP SMULATIONS (3) Cummulative DRP - Observed Cummulative DRP - Simulated (SWAT) Cummulative DRP - Simulated (HSPF) 200 DRP (kg P)

22 CONCLUSIONS Flow simulation with the HSPF model was better than the SWAT model in the prediction of peak events. Total Phosphorus and Dissolved Reactive Phosphorus simulations with the SWAT and HSPF models were generally acceptable. Both models failed to simulate high values of Total Phosphorus and Dissolved Reactive Phosphorus due to the underestimation of removable soil phosphorus. Soil phosphorus modelling in the SWAT model includes parameters to account for the effect of soil moisture and soil temperature while the HSPF model parameters take into account the effect of soil temperature only.