My question is about Sentinel 5P + Level-2-Product + Nitrogen-Dioxide + pixel boundaries(pixel corners)
In particular, given a measurement determined by i,j (in scanline X ground_pixel), I would like to know whether the area(on the surface of the earth) that corresponds to this measurement (and is delimited by the corresponding pixel corners) is a rectangle.
I am not sure I understand the question, but the rectangles are generally due to sensor (on the satellite) geometry. Surface on the earth is obviously not directly aligned with measurement representation.
Resolution of the pixel is 7x7km, which makes pixelization obvious on the map.
In you answer you assert that the pixel corners define a rectangle (are you sure about that?..this is exactly my question…). I did not find such a sentence in the reference of the s5p - a single measurement is made during a second while the satellite moves for around 7 kmts(p.14 of reference link bellow ), and because of that, it is not clear to me that the boundary-points define a rectangle(unless the satellite moves in a line w.r.t earth)
Pixel is just a graphical representation of the sensor. In raster data, all pixels are presented as squares (this can be distorted a bit due to reprojection, but it is still more or less square).
Perhaps this link might be of use: https://crisp.nus.edu.sg/~research/tutorial/image.htm