Worked example: reported growth versus a revised baseline

Two reported acreage values do not, by themselves, tell you how much land burned between two observations. To see why, separate the original estimate, any later revision of that same observation, and the estimate for a later observation.

All values below are invented for teaching. Assume the later mapping process revises the earlier snapshot as well as producing a new snapshot. The incident identity, area definition and units stay the same.

Calculation 1: change between the first and latest reported values

Subtracting A from C gives 1,200 − 1,000 = 200 hectares. Relative to A, that is 200 ÷ 1,000 × 100 = 20%. This describes the difference between those two published values.

Calculation 2: change using the revised earlier snapshot

Subtracting B from C gives 1,200 − 900 = 300 hectares. Relative to B, that is 300 ÷ 900 × 100 = 33.3%, rounded to one decimal place. The result differs because the earlier baseline changed.

The arithmetic can be checked directly: C − A = (C − B) + (B − A). Here, 200 = 300 + (−100). This is a bookkeeping identity for the three records, not a general method for recovering true fire growth from incomplete reports.

What if the revised earlier estimate is unavailable?

Then B is unknown. Do not invent it, apply a generic correction percentage or label the A-to-C difference as verified new burned area. Report the observed change in published acreage and explain that the available records do not isolate the contribution of mapping revisions.

Even B and C describe mapped area under the stated assumptions. Interpreting their difference as newly burned surface requires checking the underlying geometry, classification, exclusions and observation methods. See fire perimeter versus burned area for the distinction.

Do not confuse hectares per hour with forward spread speed

The illustrative B-to-C interval is 12 hours. Dividing 300 hectares by 12 gives 25 hectares per hour, an average mapped-area change per unit time under this example’s assumptions. It is not a forward spread speed in metres per minute and does not predict when fire will reach a location. The rate-of-spread guide explains distance-based calculations.

Suggested wording for a research note

In this illustrative dataset, published area increased from 1,000 to 1,200 hectares. A later revision placed the earlier snapshot at 900 hectares, so the difference between the revised earlier snapshot and the later snapshot is 300 hectares. These figures describe mapped-area estimates; they do not independently establish the area newly burned.

Keep the original record, its revision, observation timestamps, publication timestamps and source identifiers together. A later publication time does not necessarily mean a later observation. If you cannot establish what time a value represents, record that limitation explicitly.