Levelling loops

Differential level circuit reduced from field notes

Reduce a four-setup differential level run from BM-1 to BM-2 through three turning points, then prove the page with the sum of backsights against the sum of foresights.

Recall· about 15 minutes by hand· 6 steps

Given

  • A differential level run from BM-1 to BM-2 over four instrument setups with three turning points. All distances are in US survey feet; rod readings are recorded to 0.001 ft and elevations are carried to 0.001 ft.
  • Published elevation of BM-1 = 412.386 ft on the local vertical datum.
  • Setup 1: backsight 4.271 on BM-1, foresight 6.842 on TP-1.
  • Setup 2: backsight 8.113 on TP-1, foresight 2.096 on TP-2.
  • Setup 3: backsight 3.774 on TP-2, foresight 9.428 on TP-3.
  • Setup 4: backsight 5.902 on TP-3, foresight 7.335 on BM-2.
  • Backsight and foresight lengths were paced equal to within about 10 ft at every setup.

Required

  • The reduced note page with backsight, height of instrument, foresight and elevation columns.
  • The elevation of each turning point and of BM-2, to 0.001 ft.
  • The arithmetic page check.

Work it through yourself before reading on — the solution below shows every step, so there is no way to skim it without giving the answer away.

Worked solution

The two equations that run the whole page

Differential levelling has exactly two arithmetic operations, applied alternately. A backsight is a rod reading taken on a point of known elevation, so adding it to that elevation gives the elevation of the line of sight itself. That value is the height of instrument, and it is not the height of the telescope above the ground or above the tripod plate; it is an elevation on the same datum as the benchmarks.

A foresight is a rod reading taken on a point whose elevation is wanted, so subtracting it from the height of instrument gives that elevation. Every line of a level page is one of those two operations, and the columns are laid out so that a backsight is always added and a foresight is always subtracted.

HI = elevation of the point sighted + BS elevation = HI - FS

Setups 1 and 2

Start on BM-1 at its published elevation and work the first two setups. The turning points are the only points that carry the elevation forward, which is why they are read to 0.001 ft and why they must be firm, well-defined objects that will not settle while the instrument is moved.

Notice that TP-2 comes out higher than TP-1 by 6.017 ft. That is a normal figure for a single setup on rolling ground and is worth a glance in the field: a turning point that jumps more than about 8 ft usually means the rod was read on the wrong foot mark.

HI(1) = 412.386 + 4.271 = 416.657 TP-1 = 416.657 - 6.842 = 409.815 HI(2) = 409.815 + 8.113 = 417.928 TP-2 = 417.928 - 2.096 = 415.832

Setups 3 and 4

The same alternation carries the line to BM-2. Nothing in the arithmetic remembers where the line has been, so an error made at setup 2 propagates unchanged into every elevation after it. That is the reason the run has to be closed on a second benchmark or brought back to BM-1 before the elevations are trusted.

HI(3) = 415.832 + 3.774 = 419.606 TP-3 = 419.606 - 9.428 = 410.178 HI(4) = 410.178 + 5.902 = 416.080 BM-2 = 416.080 - 7.335 = 408.745

The reduced page

Written out in the standard five columns, the page reads as follows. Blank cells are shown as a dash; a station has either a backsight or a foresight in a given row, and a turning point has both.

Differential level notes, BM-1 to BM-2, elevations in feet
StationBS (+)HIFS (-)Elevation
BM-14.271416.657-412.386
TP-18.113417.9286.842409.815
TP-23.774419.6062.096415.832
TP-35.902416.0809.428410.178
BM-2--7.335408.745
Sums22.060-25.701-

The page check

Every backsight was added once and every foresight was subtracted once, so the net change in elevation from the first station to the last must equal the sum of the backsights less the sum of the foresights. Any addition or subtraction error inside the page destroys that identity, which is why the check is worked before the crew leaves the site.

The check is purely arithmetic. It proves the page was added correctly; it says nothing about whether the rod was plumb, whether the instrument was level, or whether a turning point settled between the foresight and the backsight taken on it.

sum BS = 4.271 + 8.113 + 3.774 + 5.902 = 22.060 sum FS = 6.842 + 2.096 + 9.428 + 7.335 = 25.701 sum BS - sum FS = 22.060 - 25.701 = -3.641 last elevation - first elevation = 408.745 - 412.386 = -3.641

What still has to be checked in the field

The page check passes, so the arithmetic is sound and the elevations follow from the readings as recorded. The observations themselves are checked by closing the work: either run the line back from BM-2 to BM-1 and test the loop misclosure against a tolerance, or continue to a third benchmark of known elevation and compare.

Balancing the backsight and foresight lengths at each setup, as this crew did, removes the effect of a collimation error in the instrument and of curvature and refraction, because both errors are proportional to sight length and enter the two readings with opposite sign. Balanced sights are the cheapest error control available on a level line and cost nothing but a little pacing.

Answer

  • TP-1 = 409.815 ft
  • TP-2 = 415.832 ft
  • TP-3 = 410.178 ft
  • BM-2 = 408.745 ft
  • Net fall from BM-1 to BM-2 = 3.641 ft

Check

Page check: sum BS 22.060 - sum FS 25.701 = -3.641 ft, and 408.745 - 412.386 = -3.641 ft. The two agree exactly, so the page is added correctly.

Independent check: the four setup differences are +4.271 - 6.842 = -2.571, +8.113 - 2.096 = +6.017, +3.774 - 9.428 = -5.654 and +5.902 - 7.335 = -1.433. Their sum is -3.641 ft, matching the net fall computed from the elevations.

More levelling loops

All in this category