Read the Amazon mosaic

The preserved image is a radar mosaic of the Amazon region from Fuyo 1 / JERS-1, assembled from more than 1,500 scenes. JAXA describes processing by NASDA’s Earth Observation Research Center and collaborating teams at ASF and JPL. Combining the scenes created a broad regional record from many individual observations rather than one instant exposure.

Read the original heading and axes before interpreting the terrain. The image itself labels September–December 1995. Its accompanying JAXA gallery text describes September–November; the later NASA archive guide gives September–December for the low-flood-season product. We retain that discrepancy and describe this image as a 1995 seasonal mosaic. Neither label supports assigning every pixel to one exact acquisition date.

Notice the broad river network and irregular coverage boundary. The complete image retains its return scale, geographic ticks, distance scale and institutional credits. These details are part of interpreting the product. Cropping them away could make a visually cleaner picture while removing information needed to understand it.

The mosaic is a processed radar product, not natural-color photography. Its bright and dark patterns must be read with the coverage and processing description. A separate JAXA page explicitly identifies black no-data regions in its 1996 mosaic. That warning cannot be turned into a rule that every black pixel in every radar image means missing data; each product’s conventions matter.

The Global Rainforest Mapping work also assembled a high-flood-season record in 1996. Comparing seasonal observations can support studies of inundation. A different season is also a potential confounding factor when investigating permanent land-cover change: a changed signal is not automatically evidence of deforestation.

In an original study design, choose a small floodplain area, identify valid observations in both seasons, and compare their signals with river-stage or field information. Record the acquisition interval for each part of the mosaic. If observations within a composite span several months, avoid presenting the result as the state of the entire basin on one day. A regional composite supports a regional question only at its actual temporal detail.