Overview

pySVA is a Python-based open-source toolbox for computing terms in a salinity variance budget on unstructured mesh model output. It applies to model output from models with unstructured, staggered C-grid meshes, particularly from models like D-FLOW Flexible Mesh.

The toolbox implements the salinity variance budget framework of Burchard et al. (2008) and Li et al. (2018), which decomposes tracer variance changes into contributions from advection, straining/mixing, and dissipation. This approach enables quantification of both physical and numerical mixing processes in hydrodynamic models.

It is available as open source on Github under the MIT license.

Purpose

The pySVA toolbox enables quantification of physical and numerical mixing processes in hydrodynamic models by decomposing changes in tracer variance into contributions from:

  • Advection: Transport of variance by mean flow

  • Straining: Variance production/destruction due to vertical mixing and shear

  • Dissipation: Variance reduction due to diffusion (turbulent mixing)

This approach provides insight into the mechanisms responsible for spatial patterns of stratification in coastal and estuarine systems.

Key Features

  • Handles multiple input formats: NetCDF files, xarray Datasets, and xugrid UgridDatasets

  • Automatically extracts grid topology and coordinate systems

  • Provides cached geometric properties for efficient computation

  • Implements the variance budget framework of Burchard et al. (2008) and Li et al. (2018)

Status

⚠️ Note: This package is currently in limited maintenance mode. Issues can still be opened, but fixes might take longer.