For astronomical observations - which can transfer to surveying techniques - I suggest the channels on the north and east sides of the pyramid. This ability must firstly come from understanding astronomy in the first place, and knowing what the stars are doing in regards to Earth's rotation, orbit and moving between the summer and winter solstices.
We mostly assume that the ancients only used the curvature of the sky to enable any specifics to do with associating constructions with celestial alignments. What we little consider is the reflection of the cosmos upon water. So, imagine these channels filled with water, and standing upon one end, being able to see the spread of the sky reduced to a certain width and even with just one string line running its length, being able to perfectly see what is where in regards to that line.
Stand anywhere over that channel upon its length and finer observations can be made of specific stars rather than a whole stretch, and certainly the width of the channel is associated to a length of time that measures when a star goes from one side to the other (this is for the east channel, stretching north/south). Used in conjunction with the north channel (stretching east/west) and one can stand anywhere above it and see exactly when a star crosses a specific point along its length...section this channel off into parts and we have specifics of time and distance broken down into the finer points of degrees, minutes, arc seconds, time and distance, all related to ground-based surveying from astronomical observations.
All that matters in doing such as this is how high one is above the water line so that a consistent application is made and all observations are exacting. When one considers the use of the 'palm stem' and its sections, math would take care of any variation in the heights above the water line...and certainly be able to cross reference results so that certainty of observations and results is made. Considering the orientation of the Great Pyramid and its nigh-exactness in observing the cardinal points (using polar north/south rather than magnetic), we should accept that they didn't do everything by looking up, but rather down.