600B (top) and 600RI (bottom) retrieved transmission spectra comparing original analysis, 1D GP re-analysis and analogous 2D GP analysis. The 2D GP best-fit atmospheric retrieval is shown as a red dotted line and random draws taken from the 2D GP spectrum are shown in grey. Top: all three analyses show largely consistent results. Bottom: the error bars of each method may appear consistent but the random draws from the 2D GP spectrum demonstrates a large uncertainty in the slope of this spectrum after accounting for the covariance between transit depths.