PARMESAN: Meteorological Timeseries and Turbulence Analysis Backed by Symbolic Mathematics

TL;DR

PARMESAN is a Python package for atmospheric data analysis that uses symbolic mathematics to derive, verify, and automate calculations of meteorological quantities and turbulence parameters, enhancing transparency and reducing errors.

Contribution

It introduces a symbolic mathematics-based approach for deriving and automating atmospheric data analysis routines, improving accuracy and flexibility over existing tools.

Findings

Provides a symbolic derivation of meteorological quantities

Enables automatic generation of analysis routines

Ensures physical consistency and error propagation analysis

Abstract

PARMESAN (the Python Atmospheric Research Package for MEteorological TimeSeries and Turbulence ANalysis) is a Python package providing common functionality for atmospheric scientists doing time series or turbulence analysis. Several meteorological quantities such as potential temperature, various humidity measures, gas concentrations, wind speed and direction, turbulence and stability parameters can be calculated. Furthermore, signal processing functionality such as properly normed variance spectra for frequency analysis is available. In contrast to existing packages with similar goals, its routines for physical quantities are derived from symbolic mathematical expressions, enabling inspection, automatic rearrangement, reuse and recombination of the underlying equations. Building on this, PARMESAN's functions as well as their comprehensive parameter documentation are mostly auto-generated, minimizing human error and effort. In addition, sensitivity/error propagation analysis is possible as mathematical operations like derivations can be applied to the underlying equations. Physical consistency in terms of units and value domains are transparently ensured for PARMESAN functions. PARMESAN's approach can be reused to simplify implementation of robust routines in other fields of physics.