Reclaiming First Principles: A Differentiable Framework for Conceptual Hydrologic Models

Jasper A. Vrugt, Jonathan M. Frame, Ethan Bollman · 2026

Read Reclaiming First Principles: A Differentiable Framework for Conceptual Hydrologic Models by Jasper A. Vrugt, Jonathan M. Frame, Ethan Bollman (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-06429.txt.

Description

Rainfall-runoff models guide flood and water planning, but fitting their settings is often slow and fragile in numbers. Jasper A Vrugt and 2 co-authors present a framework that returns to first principles and gives exact gradients. Standard gradient methods lean on finite differences or heavy auto-differentiation tools, which add error and overhead, especially for river basin models.

Their method adds response rules to the model, evolving states and setting changes together. The resulting matrix gives exact gradients for any smooth scoring function, from classic error sums to water efficiency scores and flow curves. Exact math removes step-size guessing and solver wobble from rough estimates, making tuning faster and steadier.

Catalog record

CREATORJasper A. Vrugt, Jonathan M. Frame, Ethan Bollman
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE238 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-06429

Browse the full library · About the Tower of Babel