Forward Modelling of Space Plasma Detectors: An Open Source Toolkit

Abstract:

The analysis of data from charged particle detectors is greatly enhanced by the availability of forward models that can produce synthetic datasets from known plasma parameters (e.g., density, pressure, kappa index/other parameters specific to particular distribution functions). Such forward models have a wide range of uses including estimating bulk plasma parameters by fitting distributions to raw data; investigating random and systematic errors on estimated bulk parameters; evaluating instrument/mission requirements for future space missions, and observation strategies for current missions; and the integration of multiple heterogeneous data sets. They are also an enabling technology for the use of advanced statistical methods and an important link in the chain of scientific reproducibility from raw data to higher level data products.

In this presentation we discuss the design, implementation and testing of an open source forward modelling toolkit written in C++/Python (with full bindings for the C++ elements into Python). The toolkit features the inclusion of various non-ideal effects including sensor obscuration; resolution and point spread; efficiency as a function of look direction, species and incoming energy; non-uniform spacecraft potentials; and non-stationary and inhomogeneous noise sources. We demonstrate a number of example applications, including an evaluation of systematic errors in bulk parameters, testing the resolution of narrow beams and loss cones whose dimensions are close to resolution limits, and the detection of charged dust by electrostatic analysers. We also comment on the use of the toolkit for advanced statistical inference and discuss future development directions in order to encourage collaboration in the use and development of the toolkit.