Get started
Work

Assessing How Snow Data Assimilation And Model Complexity Influence Estimation Of Snowpack Liquid Water Content

Year: 2025
Type: preprint
Abstract: Liquid water content (LWC) in seasonal snowpacks is a critical state variable influencing snowmelt runoff, flood forecasting, avalanche prediction, and radar remote sensing. However, LWC observations are not commonly measured, leading to a dependence on model estimation. The representation of LWC in process-based snow models varies widely, with differences in how meltwater retention, percolation, and refreezing are handled, ranging from simple bucket schemes to physically-based formulations using Darcy’s law or Richards’ equation. Despite its significance, the simulation of LWC across physics-based models remains understudied, particularly within data assimilation (DA) systems. Leveraging the physically-consistency of a model, data assimilation of one type of snow observation (snow depth) has been shown to improve other unobserved states (like snow density), but the improvements to LWC have not been assessed. In this study, we test the impact of assimilating snow depth and snow disappearance date (SDD) on LWC estimates using Flexible now Model (FSM2). The model is forced with local meteorological observations and snow characteristics are compared before (reference) and after assimilation.We validate the results using in situ LWC measurements and also explore how changes in FSM2’s process complexity affect performance. We hypothesize that data assimilation will improve the LWC estimation and hence, this research will address the key gaps related to LWC uncertainty and the value of model complexity, offering insights for improving LWC retrievals in the absence of remotely sensed LWC data.
DOI: 10.22541/essoar.176036962.26304741/v1
Language: English
Open Access status: gold
OpenAlex ID: W4415106922
Cites:
Cited by:
Related to: