Are models too simple? Arguments for increased parameterization

Area of Science:

• Environmental modeling
• Hydrology
• Geophysics

Background:

• Model parsimony is often prioritized, potentially limiting utility.
• Traditional model calibration focuses on simplicity, sometimes at the expense of predictive accuracy.
• Calibration parameterization is often constrained by data limitations and computational capacity.

Purpose of the Study:

• To challenge the assumption that simpler models are always better.
• To advocate for advanced calibration methods that enhance model utility.
• To highlight the benefits of regularized inversion over traditional model calibration.

Main Methods:

• Contrasting traditional model calibration with regularized inversion.
• Discussing parameterization strategies and their impact on predictions.
• Illustrating differences using mathematical concepts (in appendix).

Main Results:

• Excessive model simplification can degrade predictive performance.
• Regularized inversion allows for more parameters, potentially improving calibration.
• Advanced methods and computing capabilities enable more robust model calibration.

Conclusions:

• Model calibration should prioritize predictive utility over simplicity.
• Regularized inversion provides greater insight from calibration data.
• Adoption of advanced calibration techniques is recommended with improved computing power.