Elevation Measurement

Part of Cartography & Surveying

Techniques for determining height differences between points, establishing elevation datums, and running precise leveling surveys.

Why This Matters

Elevation is the dimension of maps most often neglected and most often consequential. A road that appears direct on a flat map may climb 200 meters and descend 300, making it far more costly than an alternative route that looks longer but stays level. An irrigation channel that seems to point in the right direction may slope the wrong way, sending water away from the intended field instead of toward it.

Elevation measurement makes these hidden facts visible. It reveals where water will naturally flow, where frost will pool, where flooding will reach, and where the best building sites are. In terms of return on survey effort, few investments pay off more consistently than establishing a good local elevation network early in a community’s development.

The methods range from improvised water-level tubes accurate to centimeters, to systematic leveling runs that can carry precise elevation data across an entire valley.

Datum and Benchmark

Before measuring any elevations, establish a datum — an arbitrary reference elevation from which all others will be measured.

The datum might be sea level (if you have a way to relate to it), the floor of a key building, or a prominent rock surface chosen simply because it will not move. What matters is permanence and documentation. The datum point is called a benchmark.

Constructing a benchmark:

• Chisel a standard mark (a horizontal line with a notch below, indicating the reference surface) into a stable rock outcrop or large buried stone.
• Build a reinforced concrete pillar driven below frost depth if stone is not available.
• Record the benchmark’s location in writing, in multiple places, with enough description to find it independently.

Assigning datum values: Assign the benchmark an arbitrary elevation (often 100.000 m or 1000.000 m) so that nearby points are unlikely to have negative elevations. All surveyed heights are described relative to this datum.

Multiple benchmarks: Establish a network of secondary benchmarks throughout your region, each with a precisely measured elevation relative to the primary. If the primary is ever destroyed, the network survives. Mark the elevation of each on the benchmark itself as well as in records.

Differential Leveling

Differential leveling is the standard method for accurate elevation transfer. It measures the height difference between two points by sighting along a horizontal line from an instrument to graduated rods held at each point.

Equipment needed:

• A leveling instrument: a telescope or sighting tube with a bubble level, mounted so it can rotate horizontally while remaining exactly level. This can be improvised from a bubble level attached to a sighting tube on a tripod.
• Two leveling rods: straight poles graduated in centimeters (or your local unit), white with black markings for visibility.

Procedure (one setup):

1. Set up the instrument between two points A (known elevation) and B (unknown).
2. Level the instrument carefully.
3. Sight to rod held at A; read the rod. This is the backsight (BS).
4. Sight to rod held at B; read the rod. This is the foresight (FS).
5. Height of instrument (HI) = elevation of A + backsight reading.
6. Elevation of B = HI − foresight reading.

Example: A is at 100.000 m. Backsight to A reads 1.450 m. HI = 101.450 m. Foresight to B reads 0.780 m. Elevation of B = 101.450 − 0.780 = 100.670 m. B is 0.670 m higher than A.

Extending the survey: When B and the next point C are too far apart for one setup, use B as a turning point — record its elevation precisely, move the instrument forward, and take a new backsight to B. Chain as many setups as needed to reach the destination.

Water Level Methods

A long flexible hose filled with water provides a simple, accurate level over distances up to 20–30 meters (longer with careful handling and a wider hose).

Principle: Water seeks its own level. Both ends of a water-filled hose will show the same water surface elevation regardless of how the hose bends between them.

Procedure:

1. Fill the hose with water, eliminating all air bubbles. Air bubbles cause errors.
2. Hold one end at point A at a measured height above the ground. Hold the other end at point B.
3. The water surface at both ends is at the same elevation. If you hold end A so the water is 1.2 m above the ground at A, and at B the water surface is 0.9 m above the ground, then B is 0.3 m higher than A.
4. Read both ends simultaneously if possible — have a second person hold and read the far end while you read the near end.

Sources of error: Air bubbles (eliminate before measuring), wind pressure on open water surfaces (use sealed ends or narrow tubes), temperature differences causing slight density differences in long hoses. For spans over 10 m, the water level method is often more practical than a sighting level for improvised equipment.

Slope Measurement and Gradient Calculations

For planning roads, channels, and drainage, you need to know not just elevation difference but the gradient — elevation change per unit of horizontal distance.

Gradient = (elevation change) / (horizontal distance) × 100%, expressed as percent, or as a ratio (1 in 20 = 5%).

Practical gradient limits:

• Wheeled cart on unpaved road: 8–10% maximum for comfortable travel, 15% maximum for short stretches
• Animal-drawn vehicle: 6–8% sustained, 12% maximum
• Canal or gravity irrigation channel: 0.05–0.5% (too steep and water scours the channel; too flat and it stagnates)
• Building floor drainage: 1–2%
• Road drainage ditch: 0.5–3%

Setting a gradient in the field:

1. Decide the required gradient (e.g., 1%).
2. Calculate the elevation change per unit of distance: for 1% over 50 m, you need 0.50 m of fall.
3. Starting at the known high point, run a leveling survey setting stakes every 10–20 m at the required elevation below each previous stake.
4. The line connecting these stakes at the correct gradient can then be excavated or graded.

Trigonometric Leveling

For longer distances or when leveling is impractical (steep cliffs, wide rivers), elevation differences can be calculated from measured distances and vertical angles.

Formula: Elevation difference = horizontal distance × tan(vertical angle)

If the horizontal distance from A to B is 200 m and you measure an elevation angle of +3.5° from A to B, the height difference = 200 × tan(3.5°) = 200 × 0.0612 = 12.24 m. B is 12.24 m higher than A.

Accuracy: Trigonometric leveling accumulates error from both the distance measurement and the angle measurement. Each measurement contributes independently; over long distances, accuracy degrades. For heights of large natural features (cliffs, hills), this method is often the only practical option and is accurate to within a few percent if instruments are well-calibrated.

Closing the Loop

Any serious elevation survey should close on itself — either returning to the starting benchmark or connecting to a second independently known benchmark.

Closure check: After running a level loop back to the starting benchmark, the final calculated elevation should match the starting elevation. The difference is the misclosure. Acceptable misclosure depends on distance and method — a rough rule is ±12 mm × √(distance in km) for good leveling work. Greater misclosure means an error occurred; find and correct it before trusting any of the intermediate elevations.

Distributing misclosure: small acceptable errors can be distributed proportionally across the survey legs rather than corrected in the field. Apply a correction at each turning point equal to its proportional share of the total misclosure.

A reliable elevation network, built once and maintained as infrastructure, shapes every subsequent planning decision a community makes.