Leveling
Part of Surveying
The techniques for measuring elevation differences between points to establish heights above a datum.
Why This Matters
Leveling is the branch of surveying concerned with measuring vertical distances — how much higher or lower one point is compared to another. This is distinct from the horizontal measurements of traversing and triangulation. Leveling answers the question: “How high is this?” and “Will water flow from here to there?”
Every drainage system, every irrigation canal, every foundation, every floor in every building, and every road that sheds water properly depends on correct leveling work. The consequences of poor leveling are sometimes catastrophic: a canal that slopes the wrong way, a building foundation that collects water, a cellar that floods every spring. The consequences are also sometimes subtle: a house whose floors visibly tilt, a field irrigation system that over-waters one end and under-waters the other.
Leveling technique is relatively simple compared to triangulation or traverse surveying. A modest investment in learning the method and making a basic instrument gives you the ability to control vertical dimensions across any project, regardless of its scale.
Fundamental Concepts
Datum: An arbitrary reference elevation from which all heights are measured. Sea level is the global datum, but for local work you can designate any convenient point as zero. Assign it a positive number (like 100.000 m) large enough that no point in your survey area will come out with a negative elevation.
Benchmark (BM): A permanent physical mark whose elevation above the datum is known and recorded. Benchmarks are the starting and ending points for all leveling runs.
Height of Instrument (HI): The elevation of the surveyor’s horizontal line of sight. HI = elevation of the point you are reading from + the rod reading on that point (backsight).
Backsight (BS): A rod reading taken to a point of known elevation. Used to calculate HI. The instrument does not move between a backsight and the subsequent foresights.
Foresight (FS): A rod reading taken to a point whose elevation you want to determine. Elevation of point = HI − FS.
Turning Point (TP): A temporary point used to carry the elevation forward when the instrument must be moved. You take both a foresight (determining the TP’s elevation) and a backsight (using the TP to establish the new HI after moving) at each TP.
Basic Differential Leveling
Differential leveling determines the elevation difference between two points that may be too far apart, or separated by too much terrain, to reach with a single instrument setup.
Procedure:
- Set up the instrument approximately midway between the starting benchmark and a convenient turning point.
- Take a backsight (BS) on the benchmark rod. Compute HI = BM elevation + BS.
- Sight to the turning point and take a foresight (FS). Compute TP elevation = HI − FS.
- Move the instrument forward, past the turning point. The rodman stays at the turning point.
- Take a backsight on the TP rod. Compute new HI = TP elevation + BS.
- Sight forward to the next point and take a foresight.
- Continue until you reach the destination point.
Why midway? Setting up midway between backsight and foresight points cancels out the instrument’s collimation error (any slight tilt in the line of sight). An instrument set closer to one point than the other introduces a systematic error that accumulates through the run. This equal-distance setup is the single most important technique for accurate leveling.
Sample computation:
| STA | BS | HI | FS | ELEV |
|---|---|---|---|---|
| BM-1 | 1.823 | 51.823 | 50.000 | |
| TP-1 | 2.145 | 52.682 | 1.286 | 50.537 |
| TP-2 | 1.034 | 52.443 | 1.273 | 51.409 |
| BM-2 | 2.107 | 50.336 |
Check: Sum BS (1.823 + 2.145 + 1.034) = 5.002 Sum FS (1.286 + 1.273 + 2.107) = 4.666 Difference = 0.336 Starting elev + difference = 50.000 + 0.336 = 50.336 ✓
This arithmetic check confirms no calculation errors were made. Always perform it before leaving the field.
Profile Leveling
Profile leveling measures the ground elevation at regular intervals along a line (a road centerline, a drainage channel, or any linear route). The result is a vertical cross-section of the terrain along that line.
Procedure:
- Establish the center line with stakes at regular intervals (typically every 10, 20, or 50 meters depending on terrain variability).
- Level along the line, taking foresights to each intermediate stake (called “intermediate foresights” or “IFS”). These points do not serve as turning points — the instrument is not moved after sighting them.
- At convenient intervals, establish true turning points to advance the instrument.
Recording: The profile level book has columns for STA (station number along the route), BS, HI, IFS (intermediate foresight, not used in arithmetic check), FS (turning point foresight), and ELEV.
Plotting the profile: On graph paper, plot distance horizontally and elevation vertically. Use a horizontal scale matched to your survey (e.g., 1 cm = 10 m) and a vertical exaggeration of 5× to 10× (e.g., 1 cm = 1 m) to make grade changes visible. Draw a smooth line through the plotted points.
The resulting profile shows you exactly how the ground rises and falls along your route, enabling you to design the grades for the finished road, canal, or drain.
Cross-Section Leveling
Cross-section leveling measures the ground elevation along lines perpendicular to the main profile, at each station. This gives a three-dimensional picture of the terrain and is essential for calculating earthwork volumes.
At each station, measure elevations at set distances left and right of the center line — typically at 2, 5, and 10 meters out, plus at any significant terrain break in between. Record the ground elevation at each offset distance.
Plot each cross-section on paper, showing the ground profile, the design template (the shape of the finished road, canal, or drain), and the resulting cut or fill area. Calculate the area of cut or fill at each cross-section geometrically or by counting squares on graph paper.
Volume calculation: Volume of earthwork between two stations = average of the two cross-section areas × the distance between stations. Sum all the volumes along the route for the total earthwork quantity.
Reciprocal Leveling
When you must level across a wide obstacle — a river, a lake, a deep valley — where the instrument cannot be set midway, use reciprocal leveling to cancel systematic errors.
Method:
- Set up on side A. Take a backsight on A’s benchmark and a foresight on a rod at B (across the obstacle). Compute elevation of B.
- Move everything to side B. Take a backsight on B’s rod and a foresight on a rod at A. Compute elevation of A from this setup.
- The two computations will give slightly different results because of collimation error, curvature, and refraction. Average the two computed height differences for the best estimate.
The accuracy of reciprocal leveling depends on the measurements being made in quick succession, before atmospheric conditions change. Temperature differences across a wide gap cause refraction that shifts the line of sight unpredictably.
Leveling Errors and Corrections
Parallax: When the image of the rod and the crosshair of the instrument are not in the same focal plane, moving your eye slightly gives different readings. Focus carefully to eliminate parallax before each reading.
Rod not vertical: A tilted rod reads longer than the actual vertical distance. The rodman should use a hand bubble level or have the instrument observer watch for tilt. The minimum reading while the rod is being waved slightly forward and back is the closest to vertical.
Refraction and curvature: Over long distances, the Earth’s curvature causes the horizontal line of sight to diverge from the level surface, and atmospheric refraction bends the line of sight downward. These effects nearly cancel each other and are negligible for setups under 100 m. They are significant for setups over 200 m and must be computed and applied as corrections.
Systematic errors accumulate: A small consistent error (instrument slightly out of level, or slightly too close to one end) adds up over many setups. Use the midpoint setup rule religiously, and check leveling runs by running loops that should close back to the starting elevation.
The Peg Test
Before any leveling job, perform the peg test: set two stakes 50 m apart. Set up midway and read both rods. Then set up 1 m behind one stake and read both rods. If the instrument is perfectly adjusted, the difference in readings should be the same in both setups. If it is not, the instrument has a collimation error. Correct it by adjusting the crosshair or noting the error and applying a correction to all subsequent readings.