application of the swat model in a decisional framework ... · application of the swat model in a...

ApplicationApplication of the SWAT model in a of the SWAT model in a decisionaldecisional frameworkframework forfor the the CaiaCaiacatchment, catchment, PortugalPortugal

Mauro SulisMauro Sulis11

Luis RodriguesLuis Rodrigues22

Claudio PaniconiClaudio Paniconi33

Nelson LourencoNelson Lourenco4 4

(1)(1) CRS4 (Center CRS4 (Center forfor AdvancedAdvanced StudiesStudies, , ResearchResearch and and DevelopmentDevelopment in in SardiniaSardinia), Cagliari, ), Cagliari, ItalyItaly

(2)(2) UATLA (UATLA (UniversidadeUniversidade Atlantica), Barcamena, Atlantica), Barcamena, PortugalPortugal(3)(3) INRSINRS--ETE (ETE (InstitutInstitut NationalNational de la de la RechercheRecherche ScientiqueScientique -- Eau, Terre Eau, Terre

etet EnvironnementEnvironnement), ), UniversitèUniversitè dudu QuébecQuébec, Canada, Canada(4)(4) UniversidadeUniversidade Nova de Nova de LisboaLisboa, , LisbonLisbon, , PortugalPortugal

MotivationMotivation and and objectivesobjectivesCaiaCaia catchment (catchment (southeasternsoutheastern PortugalPortugal) water management ) water management problemsproblemsinclude: include: equitableequitable distributiondistribution of of scarcescarce water water toto differentdifferent regionsregions and and forforvariousvarious usesuses; ; irrigationirrigation planning; planning; contaminationcontamination fromfrom nutrientsnutrients..

ThisThis work:work:

§§ImplementImplement a water a water balancebalance model model asas a a supportsupport tooltool forfor a DSS;a DSS;§§ExploreExplore and and definedefine the modeling scenarios and I/O the modeling scenarios and I/O transformationstransformations thatthat willwillconstituteconstitute the interface the interface betweenbetween model and DSS.model and DSS.

LongLong--termterm work:work:

§§Implementation and application of the DSS for the Caia study areImplementation and application of the DSS for the Caia study area, making a, making use of model simulation results where appropriate (e.g., as compuse of model simulation results where appropriate (e.g., as components of onents of the DSS analysis matrix);the DSS analysis matrix);§§Tighter Tighter integrationintegration betweenbetween the DSS and the the DSS and the hydrologicalhydrological model (SWAT model (SWAT and/or and/or simplersimpler ““metameta--modelsmodels”);”);§§ExtensionExtension of the water of the water balancebalance model model toto considerationconsideration of water of water qualityqualityaspectsaspects asas wellwell..

OverviewOverview of the of the watershedwatershed

n DEM (digital elevation model)=100m resolution

n Min. elevation=180m

n Max. elevation=998m

n Area=780km2

n Subbasins=32 (threshold area=11km2)

Land Land useuse characterizationcharacterization

n Land use map with a resolution of 100 m

n Prevalent land use classes:· 56% agriculture (AGRL)· 31% deciduous forest (FRST)· 5% mixed forest (FRSD)· 3% surface water (WATR)

SoilSoil characterizationcharacterization

n Absence of a map convertible to the classificationused by SWAT (USDA)

n One type used for the entire watershed

n Main soil characteristics: C (hydr. group); Clay (10%), Silt (30%),Sand (60%);100mm/hr(hydr.conductivity)

ClimatologyClimatology of the of the watershedwatershed

n Monthly data set (1960-1990) of precipitation, temperature given for 17 meteo stations in the Caia catchment area

n Daily data set (1960-1990) obtained bymeteorological analysis (NCEP-NCAR) forprecipitation, temperature and solar radiation

0

200

400

600

800

1000

1200

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974

Precip_observed Precip_generated Precip_meteo

SWAT SWAT modulesmodules selectedselected

n Surface runoff: SCS-CN method

n Potential evapotranspiration: Hargreaves

n Routing phase: Variable storage

CalibrationCalibration and and validationvalidation proceduresprocedures

n Monthly streamflow data given for one gagelocated within the watershed (1960-1990)

n Average annual conditions

n Model parameters investigated: available soil water capacity, soil depth, curve number

0

200

400

600

800

1000

1200

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974

Precip Streamflow_measured Streamflow_simulated

0

100

200

300

400

500

600

700

800

900

1000

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990

Precip Streamflow_measured Streamflow_simulated

Scenario generation Scenario generation forfor the DSSthe DSS

n Non operational reservoir (rainfed farming)

n Operational reservoir combined with the auto-irrigation module (irrigated agriculturalproduction)

3030--year year averagedaveraged soilsoil water water distributiondistribution

withoutwithout reservoirreservoir withwith reservoirreservoir + irrigationirrigation

SWAT and DSS interactionSWAT and DSS interaction

n Processing of model-generated spatio-temporal data to extract indicators for the DSS analysis matrix

n The DSS is not spatially or temporally distributed, so for any given indicator a single value must be obtainedfrom any (combination of) model outputs (min, max, weighted average, threshold probability, etc.)

ConcludingConcluding remarksremarks

n Acquire additional data (soil, irrigation, etc) and assessthe quality of the available data (e.g., anomalies in the climate and streamflow time series);

n Improve the hydrological modeling (entire catchment, daily observations)

n Include new scenarios (actual rather than model-generated irrigation program, additionaldams/reservoirs?, …)

n Future work as described earlier