Land Surface Model Calibration Using Satellite Remote Sensing Data

Area of Science:

• Hydrology
• Remote Sensing
• Geophysics

Background:

• Satellite remote sensing offers extensive spatial and temporal data for hydrological variable monitoring.
• Land surface models require accurate calibration for reliable simulations.
• Terrestrial water storage (TWS) and soil moisture are key hydrological variables.

Purpose of the Study:

• To calibrate a land surface model using satellite-derived TWS and soil moisture data.
• To enhance model parameter estimation using multi-objective evolutionary algorithms.
• To evaluate the model's forecasting skill after calibration.

Main Methods:

• Utilized TWS data from the Gravity Recovery and Climate Experiment (GRACE) mission.
• Incorporated soil moisture products from the Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E).
• Employed the Non-dominated Sorting Genetic Algorithm (NSGA) for multi-objective model calibration.
• Developed a novel combined objective function accounting for observation uncertainty.

Main Results:

• The calibrated land surface model showed improved simulation accuracy.
• Enhanced model performance was observed during both the calibration (2003, 2010) and forecasting (2011) periods.
• The new objective function effectively improved model parameters.

Conclusions:

• Satellite data (GRACE TWS, AMSR-E soil moisture) are valuable for land surface model calibration.
• Multi-objective evolutionary algorithms, like NSGA, improve model parameterization.
• The calibrated model provides more reliable hydrological simulations and forecasting capabilities.