(8) When a leveling line is complete to a point which is to hold the elevation of the line
overnight or longer, establish at least two points with at least one instrument setup apart. When the
leveling is continued from or to such a pair of points, set up the instrument midway between them, and
take rod readings on each point to determine if either of them has been disturbed. If the new difference
does not agree with the first determination, leveling must be extended to other points to verify the
stability of the point used in carrying the elevation ahead. When establishing a pair of points, decide
which one is to be the TBM and which one is to be the checkpoint. To avoid confusion, carry the
elevation forward over the point selected as the TBM. The difference in elevation between the points of
the pair should be at least 0.3 meter. Then, should confusion occur in carrying the elevation ahead
through the line, any blunder would be so large that it would stand little or no chance of remaining
undetected when the line is tied out or a circuit is closed.
(9) Permanent benchmarks should be established at about 2-kilometer intervals throughout the
line. When needed, TBMs should be placed between PBMs. Depending upon the terrain and the
control to be extended from the line, it may be necessary to break up the line into shorter sections so that
it will only be necessary to rerun a small section of the line if it busts.
b. Recording Precise (Spirit) Leveling Data. All observations of either first- or second-order
leveling are recorded in the field book, as shown in Figure 4-7, page 4-17. The circled numbers in the
figure correspond to the numbers shown in the following recording procedures:
(1) Enter the date.
(2) Enter the direction of the sun, with respect to the direction of the running of the line. If
conditions are cloudy, the abbreviation should be 0 (refer to Figure 4-8).
(3) Enter the designation of the level-line section.
(4) Enter the strength of the wind (refer to Figure 4-8).
(5) Enter the time.
(6) Number the instrument stations (not turning points) consecutively throughout the day.
(7) Record the backsight readings to the nearest 0.001 meter.
(8) Carry the mean reading to one more place than the thread reading. The mean is the average
of the three readings for each rod reading.
(9) Enter the reading of the middle thread on the back of the rod to the nearest 0.01 foot.