Data assimilation package in Python for experimental research (DAPPER)

DAPPER is released.

DAPPER diagnostics illustration

DAPPER is a set of templates for benchmarking the performance of data assimilation (DA) methods. The tests provide experimental support and guidance for new developments in DA.

It successfully reproduces the numerical results reported in the literature (for several methods and model test cases). This safeguards the quality of its benchmarks, making it suited to fundamental research.

It is open source and written in Python. Thus, while it already comes with a range of features, it is very easy to extend to further needs.

It also focuses on code readability, striking a beneficial compromise with efficiency, and is therefore apt for teaching and academic purposes. However, its current documentation does not provide a tutorial on DA, which must be found elsewhere (e.g. here).

DAPPER has been developed at the Nansen center, by Patrick N. Raanes, using internal, basic funding. It has also seen contributions by Maxime Tondeur, who has used it to conduct data assimilation experiments on the interesting MAOOAM model.

It inherits from multiple predecessors, including the enkf-matlab package developed by Pavel Sakov.

More info at project development website:

Add comment

Login to post comments