Combine SAC-C observations

SAC-C operated from 2000 to 2013 and participated in the morning constellation with Landsat 7, EO-1 and Terra. Cooperation made complementary observations useful together; it did not make every instrument or pixel interchangeable. SAC-C’s own cameras illustrate the same distinction on a smaller scale.

CONAE’s historical image-integration description lists the HRTC panchromatic camera at 35-metre sampling and MMRS at 175 metres with five spectral bands. A panchromatic channel combines a broad wavelength range; several spectral channels retain separate responses. Finer spatial detail and richer spectral information answer different questions. A sharp edge in one channel does not supply every wavelength measurement at that same scale.

In an ideal aligned square-grid example, one 175-metre cell spans five 35-metre cells along each side: 175 ÷ 35 = five. Its area therefore corresponds to 25 finer cells. This is a scale comparison, not a statement that the instruments always align exactly. Merely resizing a multispectral image onto the finer grid cannot create 25 independently measured spectra.

CONAE developed a merged product using simultaneous camera observations, aiming to retain spectral characteristics while improving spatial detail. Such image fusion uses a method to combine information; it does not rewrite the original instrument sampling. In a fictional mixed cell containing bright soil and dark water, a fusion method must decide how to distribute the coarse spectral information across finer structures. Validation should test whether those choices distort the quantities needed for the analysis.

A finer grid does not create independent spectra

The historical SAC-C catalogue preserves a 2008 notice about correcting image georeferencing and reprocessing affected earlier scenes. That notice is a practical reminder to retain product versions. An apparent movement of a boundary might arise from changed geometric placement rather than a physical change on the ground. Old operational notices are historical evidence, not statements that SAC-C is observing today.

Consider an original comparison in which a fixed shoreline is displaced by one 35-metre cell between two products. Before calling it erosion, compare registration, acquisition conditions and processing versions against independent reference features. Combining multiple missions adds further checks on wavelength response and observation time. A useful combined record documents those differences so increased coverage does not conceal a change in what was actually measured.