A new look at Spitzer primary transit observations of the exoplanet HD189733b

Blind source separation techniques are used to reanalyse two exoplanetary transit lightcurves of the exoplanet HD189733b recorded with the IR camera IRAC on board the Spitzer Space Telescope at 3.6$\mu$m during the "cold" era. These observations, together with observations at other IR wavelengths, are crucial to characterise the atmosphere of the planet HD189733b. Previous analyses of the same datasets reported discrepant results, hence the necessity of the reanalyses. The method we used here is based on the Independent Component Analysis (ICA) statistical technique, which ensures a high degree of objectivity. The use of ICA to detrend single photometric observations in a self-consistent way is novel in the literature. The advantage of our reanalyses over previous work is that we do not have to make any assumptions on the structure of the unknown instrumental systematics. Such "admission of ignorance" may result in larger error bars than reported in the literature, up to a factor $1.6$. This is a worthwhile trade-off for much higher objectivity, necessary for trustworthy claims. Our main results are (1) improved and robust values of orbital and stellar parameters, (2) new measurements of the transit depths at 3.6$\mu$m, (3) consistency between the parameters estimated from the two observations, (4) repeatability of the measurement within the photometric level of $\sim 2 \times 10^{-4}$ in the IR, (5) no evidence of stellar variability at the same photometric level within 1 year.