You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been playing with the BmiGeoTiff and have come across a couple issues.
The first is that the data are reported to be located at faces, however the data array for each band is of size number-of-nodes, not number-of-faces. I think either the x and y arrays should be two elements longer, or the data should be reported to be located at nodes. In addition, because the grid is 3D rectilinear, the term face refers to the sides of each of the cubes that make up the grid, which is definitely not what is intended here and leads to the second issue...
The second is that the grid is represented as a 3D cube but the third dimension is not a spatial dimension but, rather, the band number. I think it might make more sense to represent this data as three separate variables (maybe gis__band_1_raster_data, gis__band_2_raster_data,gis__band_3_raster_data?) on a 2D rectilinear grid.
@mdpiper Do you have any thoughts on this? I think we would only need to modify the BMI and not touch the GeoTiff api.
The text was updated successfully, but these errors were encountered:
I've been playing with the BmiGeoTiff and have come across a couple issues.
The first is that the data are reported to be located at faces, however the data array for each band is of size number-of-nodes, not number-of-faces. I think either the x and y arrays should be two elements longer, or the data should be reported to be located at nodes. In addition, because the grid is 3D rectilinear, the term face refers to the sides of each of the cubes that make up the grid, which is definitely not what is intended here and leads to the second issue...
The second is that the grid is represented as a 3D cube but the third dimension is not a spatial dimension but, rather, the band number. I think it might make more sense to represent this data as three separate variables (maybe gis__band_1_raster_data, gis__band_2_raster_data,gis__band_3_raster_data?) on a 2D rectilinear grid.
@mdpiper Do you have any thoughts on this? I think we would only need to modify the BMI and not touch the GeoTiff api.
The text was updated successfully, but these errors were encountered: