PDAF - the Parallel Data Assimilation Framework

PDAF is developed, hosted and maintained by the Data Assimilation Team in the Computing Center of the Alfred Wegener Institute.

ISDA-Online We co-organize the International Symposium on Data Assimilation - Online. For more information see the ISDA-Online website at University of Vienna

News August 21, 2024 Webinar on HGS-PDAF We will showcase the new assimilation system HGS-PDAF which add data assimilation with PDAF to the model Hydrogeosphere in a webinar hosted by Aquanty Inc. April 29, 2024 Release of Version 2.2.1 - with performance improvements for local filter algorimths; adding a 2D+1D factorized localization; some bug-fixes in OMI release notes April 15-19, 2024 PDAF at the EGU General Assembly 2024 There will be several presentations of research in which PDAF is used. In addition, there will be a short course on data assimilaation on Friday. March 8, 2024 Release of Version 2.2 - adding e.g. a non-isotropic localization, exclusion of potential observation outliers, and a new way to set the offline mode release notes July 11, 2023 New model binding to run PDAF with the NEMO ocean model. Available on Github: github.com/PDAF/NEMO-PDAF.

The Parallel Data Assimilation Framework - PDAF - is a software environment for data assimilation. PDAF simplifies the implementation of the data assimilation system with existing numerical models. With this, users can obtain a data assimilation system with less work and can focus on applying data assimilation.

PDAF provides fully implemented and optimized data assimilation algorithms, in particular ensemble-based Kalman filters like LETKF and LESTKF and nonlinear filters as well of variational methods (3D-Var and 3D ensemble Var). It allows users to easily test different assimilation algorithms and observations. PDAF is optimized for the application with large-scale models that usually run on big parallel computers and is applicable for operational applications. However, it is also well suited for smaller models and even toy models and can be used to teach data assimilation.

PDAF provides a standardized interface that separates the numerical model from the assimilation routines. This allows to perform the further development of the assimilation methods and the model independently. PDAF makes new algorithmic developments readily available through the interface such that they can be immediately applied with existing implementations. The PDAF release package provides small models for easy testing of algorithmic developments and for teaching data assimilation.

PDAF is a community open-source project. Its functionality will be further extended by input from research projects. In addition, users are welcome to contribute to the further enhancement of PDAF, e.g. by contributing additional assimilation methods or interface routines for different numerical models.

Content:

• Features and requirements of PDAF
• List of models connected to PDAF
• Implementation Concept of PDAF
• Software download
• Basic Documentation
  • First Steps with PDAF
  • Assimilation with the Lorenz-96 model and PDAF
  • Assimilation with the Lorenz-63 model and PDAF
  • Description of the software package
  • Implementation tutorials
  • Which filter should one use?
  • Generating ensembles
• Full Documentation
  • Compiling the PDAF library
  • Implementation Guide for the online assimilation system
  • Implementation Guide for the offline assimilation program
  • Diagnostic functions
  • Observation handling
  • Generate synthetic observations for twin experiments
  • Overview of filter-specific options
  • Adding a filter algorithm to PDAF
  • Lists for direct access to documentation on particular routines
  • Auxiliary and advanced functionality
  • FAQ

• Publications and Presentations involving and about PDAF