Hydrologic model predictability improves with spatially explicit calibration using remotely sensed evapotranspiration...

Area of Science:

• Hydrology
• Remote Sensing
• Environmental Modeling

Background:

• Traditional hydrologic models calibrated solely on streamflow data often yield inaccurate landscape water balances.
• Multi-objective calibration using evapotranspiration (ET) and streamflow data shows promise for improving spatial water balance representation.
• Methodological clarity is needed on integrating ET data and model parameters for optimal multi-objective calibration.

Purpose of the Study:

• To assess the impact of a spatially explicit, distributed calibration approach using remotely sensed ET and biophysical parameters on watershed water balance components.
• To improve the predictability of streamflow and ET at various watershed locations.
• To provide methodological insights for integrating remote sensing data into hydrologic modeling.

Main Methods:

• Utilized the Soil and Water Assessment Tool (SWAT) in a large watershed within the Prairie Pothole Region.
• Employed a stepwise calibration approach integrating Moderate Resolution Imaging Spectroradiometer (MODIS) ET data and biophysical parameters.
• Compared lumped versus spatially explicit calibration strategies for ET data integration.

Main Results:

• Inclusion of biophysical parameters related to vegetation dynamics and energy improved model accuracy.
• Spatially explicit calibration of ET yielded better model states than lumped approaches.
• Sub-basin-level spatially explicit calibration reduced water yield uncertainty and improved ET and streamflow accuracy.

Conclusions:

• Spatially explicit calibration is crucial for accurate watershed-scale water balance simulation.
• Remotely sensed ET data, when integrated effectively, enhance hydrologic model performance.
• The proposed approach offers a generalized method for integrating big data into hydrologic modeling for improved watershed management.