The Compatibility of Flare Temperatures Observed with AIA,GOES, and RHESSI

We test the compatibility and biases of multi-thermal flare DEM (differential emission measure) peak temperatures determined with AIA with those determined by GOES and RHESSI using the isothermal assumption. In a set of 149 M- and X-class flares observed during the first two years of the SDO mission, AIA finds DEM peak temperatures at the time of the peak GOES 1?–?8 Å flux to have an average of T _p=12.0±2.9 MK and Gaussian DEM widths of log₁₀(σ _T )=0.50±0.13. From GOES observations of the same 149 events, a mean temperature of T _p=15.6±2.4 MK is inferred, which is systematically higher by a factor of T _GOES/T _AIA=1.4±0.4. We demonstrate that this discrepancy results from the isothermal assumption in the inversion of the GOES filter ratio. From isothermal fits to photon spectra at energies of ?≈6?–?12 keV of 61 of these events, RHESSI finds the temperature to be higher still by a factor of T _RHESSI/T _AIA=1.9±1.0. We find that this is partly a consequence of the isothermal assumption. However, RHESSI is not sensitive to the low-temperature range of the DEM peak, and thus RHESSI samples only the high-temperature tail of the DEM function. This can also contribute to the discrepancy between AIA and RHESSI temperatures. The higher flare temperatures found by GOES and RHESSI imply correspondingly lower emission measures. We conclude that self-consistent flare DEM temperatures and emission measures require simultaneous fitting of EUV (AIA) and soft X-ray (GOES and RHESSI) fluxes.