-7- 
Some experiments have been performed to determine the utility of 
ERTS imagery for improving the rendition of terrain features on aero 
nautical charts and for refining coastal and shallow water depiction 
on nautical charts, but the experiments have not yet been carried to 
the stage of complete products. 
In all of the products described above, the imagery has been used 
without any change in geometry other than simple enlargement. The map 
reference system defined by the UTM grid has been warped to fit the 
imagery. A more cartographically appealing approach would be to warp 
the imagery to fit the map projection and reference grid. This has 
been accomplished in a completely digital system developed by IBM (9). 
Other investigators are also pursuing completely digital approaches to 
interpretation as well as geometric correction. 
The ERTS image data can be supplied by NASA as Computer Compatible 
Tapes (CCT) containing the intensity level in an xy array of picture 
elements (pixels). Both the radiometric intensity and the position of 
the pixels can be manipulated in the computer. By comparison of input 
image coordinates to ground-control-point UTM coordinates, the geometric 
correction parameters can be computed. The output image is then mapped 
to the input image, and the intensity of the nearest neighboring pixel 
is assigned. The corrected images are then printed on a drum recorder. 
Excellent image quality has been obtained, the output grid is on a true 
UTM projection, and the accuracy is compatible with NMAS for scale 
1:250,000. 
The observed high geometric integrity of the MSS images has prompted 
an investigation into the actual geometry of the records. By making changes 
in the geometrical corrections applied when the images are printed by the 
EBR, Colvocoresses has suggested that they can be produced on a Space Oblique 
Mercator projection (10). This is a continuous cylindrical conformal 
projection with a scale distortion of less than 1 part in 10,000. 
Certainly not all the cartographic problems of using ERTS data have 
been solved. Maintenance of resolution through the many photographic steps 
is a major problem. However, it has been demonstrated that 1:500,000 is 
about the optimum scale for presentation as opposed to the 1:1,000,000 
scale predicted before launch. Obtaining proper color balance is another 
problem; the same scene processed by different laboratories may look 
totally different. It has been found that good color quality for mapping 
can be obtained from only bands 5 and 7. This does not deny additional 
information for multispectral analysis in bands 4 and 6, but simply notes 
that for pictorial purposes bands 5 and 7 are sufficient. Some map users 
object to the overall red impression obtained from the conventional pre 
sentation of the infrared response in color. Attempts have been made to 
assign other color; to obtain a more subjectively natural appearance. And 
other attempts have been made to use the multispectral characteristics to 
produce color separations for water and infrared-reflective vegetation, but 
these have not yet been carried to printed maps. Considerably more work is 
needed to make maximum use of the parameter of multicolor presentation 
available from the ERTS records.