Traverse Adjustment Calculator

Begin with final directions and horizontal lengths

This traverse adjustment calculator accepts an ordered sequence of straight survey legs. Each leg has a starting point, ending point, final direction and horizontal distance. It computes coordinate increments and unadjusted coordinates, checks endpoint agreement when independent control exists, and applies the Bowditch rule only when you select it. Keeping those stages visible helps separate a calculation from the decision to accept or adjust observations.

The direction must already be expressed relative to the intended coordinate north. This page does not reduce raw interior angles, apply an angular closure correction or determine a starting orientation from a backsight observation. It also does not reduce slope distances, apply a grid factor or transform a datum. Complete any required earlier processing with an appropriate method and document it in the reference field before using these results.

Arrange the file as a connected route

The first four columns are From, To, Direction and HorizontalDistance. Choose comma, semicolon or tab, and confirm whether the first row is a header. Column positions are fixed even if the header wording differs. Additional source columns remain in JSON. Each row's From point must match the previous row's To point, and a leg cannot start and end at the same named point. This makes a broken sequence visible instead of silently joining unrelated observations.

Invalid directions, missing names, nonpositive distances, malformed quotes and inconsistent row widths block the entire route. All original rows remain available in the diagnostic report. The program does not remove a bad leg and continue with a shorter traverse. If a point is revisited, the sequence remains the basis of the calculation; the page does not average repeated coordinates or perform a network adjustment.

Choose an angle format without guessing

Four direction modes are available: decimal azimuth, explicit degrees-minutes-seconds azimuth, decimal quadrant bearing and explicit DMS quadrant bearing. An azimuth is measured clockwise from coordinate north, so north is zero, east ninety, south one hundred eighty and west two hundred seventy degrees. For a DMS azimuth, enter a value such as 45 30 0. A decimal value of 45.30 means 45.30 degrees, not forty-five degrees and thirty minutes.

A quadrant direction combines N or S, an angle from zero to ninety degrees, and E or W. For example, N 45 30 0 E belongs in the DMS bearing mode. This mode also recognizes the cardinal letters for due north, east, south or west. Use a single selected format throughout the file. Mixed notation should be normalized before calculation so that the interpretation of every line remains clear.

Relate coordinate increments to the chosen axes

For a direction alpha and horizontal length L, the east increment is L times sine alpha and the north increment is L times cosine alpha. These increments are accumulated from the supplied starting E and N. The display deliberately uses Easting and Northing labels because some field documents use X for north and Y for east. Map coordinates by their physical meaning, rather than copying values based only on the letter names.

The same length unit must be used for observations and control coordinates. Selecting metres, feet or US survey feet labels the result; it does not convert the input numbers. Exact cardinal directions avoid artificial sine or cosine residues on the perpendicular axis. Other directions involve floating-point trigonometry. Decimal accumulation preserves small changes against a large coordinate origin, but the number of output digits should never be interpreted as achieved field precision.

Distinguish a loop, an attached traverse and an open route

A loop ends on the starting named control point and compares the computed endpoint with the supplied starting coordinates. An attached traverse ends on another point whose known E and N you enter. For that case, closure is measured against the known endpoint, not against zero sums of the increments. A traverse with an unknown endpoint may produce useful provisional coordinates, but it has no independent positional closure check.

Choose the no-known-endpoint mode when that is your actual situation. The closure fields then remain unavailable and adjustment is disabled. Do not invent an endpoint solely to obtain a ratio or an adjusted diagram. Conversely, having positional endpoint control does not mean the raw angle observations have been checked; this page works with the final directions you provide and makes no angular adjustment claim.

Read the original closure before adjusting it

The east and north closure components are defined as the unadjusted computed endpoint minus the known endpoint. Their vector length is the linear closure. Relative error is that length divided by the total observed traverse length. The familiar representation one in N uses the reciprocal: total length divided by linear closure. A larger N expresses a smaller relative endpoint discrepancy under this specific calculation.

A zero closure at the retained precision has no finite one-in-N denominator, so the page says that directly. A very small nonzero closure remains nonzero in the report, even if the plot cannot visually separate the endpoints. No pass or fail tolerance is supplied automatically. A suitable acceptance criterion depends on the work, observation process and applicable project requirements. Endpoint closure alone cannot reveal every compensating error along the route.

Apply Bowditch as an explicit allocation model

Bowditch is also called the compass rule. It allocates the negative east closure and negative north closure to the legs in proportion to their observed lengths. For each component, a leg's correction equals the negative total closure multiplied by its length divided by total length. The corrected increments are then accumulated from the fixed start point. This is a simple allocation method, not a least-squares solution with separately supplied observation uncertainties.

The audit keeps original increments, individual corrections, corrected increments, original coordinates and corrected coordinates. Cumulative correction is evaluated from the total fraction reached, so the final point receives the complete required correction without losing it to rounded intermediate values. The original closure is never overwritten by the adjusted endpoint. If the initial discrepancy is large or unexpected, inspect the observations and control before choosing whether an adjustment is justified.

Use the public bearing example carefully

The example button loads a four-leg teaching traverse published by Jerry Mahun. It begins at Easting 2000 and Northing 500 feet, uses explicit quadrant bearings, and has a total observed length of 1347.57 feet. Its original endpoint does not coincide exactly with the starting control. The published table rounds each coordinate increment to three decimal places; the calculator retains the trigonometric values instead of using the printed rounded increments as new observations.

For that reason, compare each increment at the published precision and do not force the full-precision closure to equal a sum assembled from rounded values. Selecting Bowditch should make the adjusted endpoint agree with the supplied start while leaving the original discrepancy available. This example checks implementation behavior, not the suitability of a particular adjustment for your project. The source and the selected method remain separate parts of the report.

Review the plot and export the intended coordinates

The plan view uses relative coordinates, equal scale, north up and east right. Blue points show the unadjusted route and amber points show the adjusted route when available. Long routes use a bounded overview; selected points can be inspected without changing the full output. Small closure differences may overlap at overview scale, so use the numeric closure and row tables for the actual check. The plot has no basemap and cannot verify that the coordinate system was selected correctly.

After reviewing the current result, approve it to unlock the point CSV. No adjustment exports original computed coordinates; Bowditch exports adjusted coordinates. Elevations stay blank because this is a horizontal calculation. The closing control row is preserved even when it repeats the starting name. The audit CSV and JSON retain the input, method, control and original closure so a receiver does not mistake an adjusted file for an unmodified observation record.

Keep the calculation reproducible

Processing takes place locally in a browser worker. Editing input invalidates the previous result and export approval. You can cancel a run, recover the latest completed input scenario or clear the session. Pagination affects only the visible rows; reports retain the complete route. The maximum input is one hundred thousand observations and ten mebibytes. CSV reports also have a ten-mebibyte limit, so a heavily expanded audit may require a smaller batch or JSON.

Decimal inputs support up to twelve places and magnitude one trillion; computed coordinates and total length are bounded to the supported range. Preserve the original file and the chosen settings whenever results are handed off. A coordinate table is most useful when another person can reconstruct its observations, control, direction convention and adjustment method without guessing.

Point CSV keeps identifiers exactly as supplied after the documented surrounding-space cleanup. Audit CSV prefixes formula-like text with an apostrophe for spreadsheet review; JSON retains the original fields. Use the point file for coordinate import and the audit for review, rather than treating their text fields as interchangeable.