Calibration for any instrument means transforming the raw output of the instrument into meaningful physical quantities. For the Cluster FGM instruments, the calibration allows for the transformation of the measured magnetic field delivered by the instrument from engineering units in a rotating non-orthogonal coordinate system to the magnetic field expressed in physical units (nT) represented in an orthogonal physical coordinate system (such as GSE or GSM).

There are three fundamental groups of calibration parameters for the FGM:

- Offsets: O_i, one for each axis
- Gains: G_i, one for each axis
- Alignment angles: azimuth f_i, elevation t_i, two for each axis

In total each instrument needs 12 independent calibration parameters. For reasons such as temperature changes of the sensor or of the associated electronics, variable magnetic fields generated by currents inside the spacecraft, magnetic permeability of parts of the spacecraft, ageing of the electronic parts, exposure to radiation, and other factors, these parameters do not remain constant in time. Therefore they need to be determined on a regular basis by in-flight calibration.

Since the Cluster spacecraft are spin-stabilized, 8 of the 12 independent parameters can be determined using Fourier analysis. Errors in these parameters are visible either as a fundamental spin tone or as the second harmonic of the spin frequency in the Fourier dynamic spectra of the calibrated data. Unlike the natural geophysical signals, the spin tone and its second harmonic due to errors in the calibration parameters are coherent narrow banded signals. Therefore they are easily separated from the external field to be measured and a minimization procedure can be applied to recover the calibration parameters responsible for producing these signatures.

Because each of the six in-flight used ranges has its own set of calibration parameters, data intervals containing range changes can be used for further refinement. Departures of the calibration parameters from their correct values are evidenced either as discontinuities of the measured component, discontinuities of its first derivative, or as a change in phase of the spin tone if one is present (or introduced for validation purposes).

The study of nonstationary events taking place at the Earth bowshock is an area of active international research. Large amount of data measured during decades of space explorations is now available for testing theoretical models and for confronting with numerical simulation results. Measurements from missions such as VEKA, ISEE, WIND, CLUSTER, GEOTAIL, etc, show the impact of the Earth bowshock on the electron distribution functions, plasma heating and other related processes.

Besides the question of nonstationarity and reformation of the Earth bow shock, another interesting and yet not completely understood problem is the electron acceleration and the heating processes at the fast mode quasiperpendicular collisionless shocks. An electrostatic field must exist in a collisionless magnetosonic shock in order to explain the deceleration of the incoming supersonic plasma to the subsonic velocities from the downstream region. The strength of the potential shift for a normal ion trajectory across a quasi perpendicular shock can be defined as the net exchange of the incoming ion energy per charge unit. This cros-sshock potential is reference dependent and the deHoffmannTeller (dHT) frame is the proper system for the study of the broadening of the distribution functions and of the electron heating. Since the motional electric component vanishes in the deHoffmann-Teller frame, the electric field is directed only along the shock normal of the shock layer. Consequently, it decelerates the incoming ions and accelerates the electrons coming from the lower density regions towards higher density regions.

The dHT cros-sshock potential is the linear integral of a very small parallel electric field along the shock normal and its value is difficult to be measured by standard measurement techniques. The dependence on the measurement system (e.g. errors in estimating the shock normal direction and the relative shock-spacecraft velocity), the effect of the satellite/satellites electric noise, and the availability of direct measurements of only two components of the electric field vector as it is the case for Cluster mission, raise serious problems in direct measurement of the dHT crossshock potential. Therefore obtaining the cross-shock potential using numerical simulations and comparing this with theoretical predictions and with the potential from measured data can significantly advance our understanding in this area.