Skip to content
Cranley

Flood rehearsal

Shows what a flood does inside this house: how deep in each room, which doors and windows go under, and the order rooms flood, with the floor-height error carried through.

What it answers

Section titled “What it answers”
  • What happens inside if 60 cm of water comes up the street?
  • Rehearse Harvey in this house

For the agent, the tool is described as: Simulate still water at a scenario depth above the ground outside one scanned house. Measures the floor height above ground (drone photos of the door, the scan’s outdoor panoramas, Vexcel elevations) with a range, then returns per-room depth, wetted floor and wall area, flood order, overtopped openings, a water overlay and a depth lens. Scenario harvey_2017 reads USGS marks.

Who it is for

Section titled “Who it is for”

Emergency manager or caseworker before a storm, Houston homeowner before a storm, Insurance underwriter.

Inputs

Section titled “Inputs”
Input Type Required Default Notes
scan string yes Scan id. Filled from the selected building.
depth_m number (m) yes Water above the ground outside. Asked as: “How deep could the water get above the ground outside your house, in metres? Use a storm-surge forecast (given as height above ground) or a street-flooding depth.” Saved preference flood.depth_m. Range: at least 0, at most 3.
scenario string "depth" Where the depth comes from: typed depth, storm surge, rainfall, or Harvey 2017 from USGS high-water marks (then depth_m is not used). Saved preference flood.scenario. One of: depth, surge, rainfall, harvey_2017.
floor_height_m number (m) Floor height above the ground at the door, if you measured it. Saved preference flood.floor_height_m.
use_drone boolean true Measure the door threshold in the drone photos (vision model, cached).

An input that nothing fills (the selected building, a saved preference, a persona’s preset) becomes a question to the person. The agent asks only for those.

Data used

Section titled “Data used”
Layer Id Needed If it is missing
Rooms matterport.rooms required The verdict is unknown, with the reason.
Doors and windows matterport.openings optional depths only, no flood order or openings
Drone photos drone.photos optional The tool runs without it.
Ground heights vexcel.elevation optional The tool runs without it.

What it returns

Section titled “What it returns”

A tool result: a verdict (pass, warn, fail or unknown), metrics, checks and findings, each with its source.

  • Draws on the building: a water level; a colour per room; doors, walls or rooms coloured by status; pins on what it found.
  • Panel blocks: VerdictHeader, MetricRow, Controls, FloorPlanMini, CheckList, EvidenceStrip, MiniChart, AssumptionChips, StepLog, ActionBar.
  • Speed: slow. Read-only.

How it works

Section titled “How it works”

Flood rehearsal: what a water level does inside one scanned house, with the floor-height error carried through.

Reliability

Section titled “Reliability”

Sources this tool gave on the test home:

Where it came from Layer Method Reliability Note
Simulated from the scan matterport simulation medium Still water filling the rooms through openings by sill height; no flow, pressure or drainage
Measured from a model model geometry low Floor height from a typical Houston slab (assumed); range 0.10 to 0.50 m
Looked up in a model model lookup high Your number

The tool sets the label for each value. A model never does. The tools give a scan’s wall graph the layer matterport whichever device made the scan, so a phone scan’s geometry carries that label too.

Limits

Section titled “Limits”
  • Needs: rooms. On a scan without it, the verdict is unknown and says why.
  • It is slow: progress steps reach the workspace while it runs.
  • A result is a planning aid, not an inspection.

On the made-up test home

Section titled “On the made-up test home”

This section was written by running the tool when the docs were built, on a hand-made RoomPlan export (no real place was scanned), with no sponsor keys and the network switched off.

Inputs used: depth_m = 0.6 (an answer given for this example).

Fail With water 0.60 m above the ground outside, water stands 0.30 m on the entry floor (0.10 to 0.50 m); 6 of 6 rooms take water. The floor height is assumed, not measured.

Value Result Where it came from Reliability
Water on the entry floor 0.3 m Simulated from the scan medium
Rooms taking water 6 Simulated from the scan medium
Floor above the ground 0.3 m Measured from a model low
Wall below the water line 43.4 m2 Simulated from the scan medium
Water on the entry floor, low end 0.1 m Simulated from the scan medium
Water on the entry floor, high end 0.5 m Simulated from the scan medium
Rooms taking water, low end 6 Simulated from the scan medium
Rooms taking water, high end 6 Simulated from the scan medium
Water above the ground outside 0.6 m Looked up in a model high
Floor above the ground, low end 0.1 m Measured from a model low

7 more values are in the full result.

Check Measured Against Status
Water on the floor, Living Room 0.3 0 m fail
Water on the floor, Bedroom 0.3 0 m fail
Water on the floor, Bedroom 0.3 0 m fail
Water on the floor, Kitchen 0.3 0 m fail
Water on the floor, Bathroom 0.3 0 m fail
Water on the floor, Hallway 0.3 0 m fail

Findings: 1 unknown, 7 fail.

What it reported while it ran:

  1. The scan has no location, so the aerial and drone methods are off
  2. Scan outdoor panoramas: none near an exterior door
  3. Water levels in the scan: lo +0.10 m from the entry floor, mid +0.30 m from the entry floor, hi +0.50 m from the entry floor

Run it

Section titled “Run it”
threshold run flood_rehearsal --scan <scan-id>
curl -X POST http://127.0.0.1:8787/api/tools/flood_rehearsal/run -H 'content-type: application/json' -d '{...inputs}'

Over MCP the tool is named flood_rehearsal. See MCP server and CLI.