Ground before water
A depth map is the difference of two surfaces, so the vertical accuracy of the terrain decides what depth can be claimed at all. We state the resolvable depth before the first model run and refuse to publish depths below it.
Ghana flood and shoreline intelligence / Capability 08
HydroGhana™
Model the water before it writes the outcome.
Flood, shoreline and coastal resilience intelligence that turns exposure into investable adaptation evidence.
DecisionProtect · warn · relocate
ControlGround truth before depth
CoverageAll 16 Ghana regions
Flood, shoreline and exposure layer / Ghana Live
Rain · capacity · exposure
Peak flow / delivered capacity / resolvable depth / lead time
Operating model
Water does not negotiate
with an optimistic map.
Most flood maps in this market are drawn on a free 30-metre global surface with metres of vertical error, then coloured to a tenth of a metre. HydroGhana™ computes what the ground model can actually resolve, what the drain actually carries once the refuse is in it, and how many minutes of warning the catchment physically allows — and says plainly where the evidence stops.
The record
Hydrological Services and Ghana Meteorological Agency gauge series, dated flood marks, NADMO incident records and radar extents of events that actually happened. The model is tested against the past before it is asked about the future.
Capacity as delivered
Design capacity is in the drawing. Delivered capacity is what the channel carries half full of silt and refuse, and in Accra the gap between the two is the flood.
Scenario and shoreline
Storms by return period, upstream spill, surge on a spring tide, and the combined case of catchment rainfall meeting a locked tidal outlet — with shoreline change measured as a rate against its own uncertainty.
Exposure and value
People, buildings, critical facilities and value at risk, counted rather than estimated, so priorities can be ranked and the works compared against the loss they prevent.
Adaptation and MRV
Measures that attenuate rather than relocate the peak, a maintenance regime that is funded, warning with real lead time, and a baseline fixed before the intervention so adaptation finance can be reported against measured change.
Three Ghanaian water problems
The flood, the coast
and the delta that lost its sand.
Accra floods in an hour
The Odaw catchment concentrates fast, its channels were sized for a smaller city, and solid waste removes a share of the capacity that was paid for in concrete. Every one of those is a number in this desk: time of concentration, peak flow, blockage, and the storm the network still passes.
The eastern shoreline is moving
From Ada to Keta the coast is among the fastest-retreating in West Africa. A recession rate is only evidence when it exceeds the uncertainty of locating the shoreline, so this desk computes the end-point rate and its error together, and refuses to call the difference erosion until it passes.
Hard defence moves the problem
Akosombo holds back the sediment that once fed the Volta delta, and a sea wall holds one frontage by starving the next. Any protection proposal here is assessed with its downdrift consequence attached, because the community that inherits it did not choose it.
HydroGhana™ water desk / Ghana
Model the water.
Then say what you can prove.
Model confidence13/100
Screening grade
Peak flow24.65 m³/s
Surcharge—
Exposure index—
Resilience0%
01 / The water
What is being modelled, where, and against which event
02 / The ground and the record
What the terrain can resolve, and what the model has been tested against
Urban drainage basin (Odaw, Kaneshie, Korle, Chemu): Accra floods in under an hour. Impervious cover, undersized channels and solid waste combine so that the return period on paper and the flood on the ground stopped agreeing years ago.
25-year storm — primary drainage and culvert design: Primary channels, culverts and road crossings are normally sized here.
Copernicus GLO-30 DEM (free): Better edited than SRTM and the sensible free default for catchment work, but still a surface model at metre-scale vertical error.
What this terrain can resolve: ±4 m one sigma gives a 95% resolvable depth of 11.09 m. The output is required to show 0.3 m, so anything shallower than 11.09 m on this map is drawing, not measurement.
Runoff: coefficient 0.87 at 55% impervious cover, giving 24.65 m³/s by the rational method Q = C·i·A/360.
03 / The coast
Dated shoreline positions, and the sea the outfall discharges into
This setting is inland, so the coastal inputs below do not affect the result. Complete them only if the reach discharges to a tidal outlet.
No shoreline fix entered. A change rate needs at least two dated positions measured from the same landward baseline, each with the uncertainty of the source it was digitised from.
04 / Who is exposed
The count behind the priority, and the value behind the case
05 / Adaptation measures
What is in place or committed on this catchment
06 / Evidence controls
What has to be true before anyone designs, permits or funds on this
Saved to your workspace and carried into the brief below.
07 / Evidence pack
What a designer, a regulator or a climate financier will ask to see
Terrain model with stated vertical accuracy and resolvable depthOutstanding
Surveyed cross-sections, drain inverts and structure soffitsOutstanding
Rainfall and river-level record for the catchmentOutstanding
Calibration against an observed flood eventOutstanding
Independent validation on withheld observationsOutstanding
Drainage asset inventory with condition and blockage stateOutstanding
Receiving-water bathymetry or bed surveyIn place
Tide, surge and wave boundary recordIn place
Dated shoreline series with stated positional uncertaintyIn place
EPA permit position and wetland designation checkOutstanding
NADMO and Assembly escalation route agreedOutstanding
Safeguards framework where occupants are displacedIn place
MRV baseline and repeat method fixedIn place
08 / Water action register
Close these before the next rains, not after them
Priority 01
Replace the terrain model — it resolves 11.09 m against a requirement of 0.3 m
Copernicus GLO-30 DEM (free) carries about ±4 m one-sigma vertically. A depth grid differences two surfaces, so at 95% confidence nothing shallower than 11.09 m can be separated from the terrain's own error. Every depth below that is drawn, not measured. Fly LiDAR or photogrammetry over the modelled extent, and level the drain inverts and thresholds by ground survey.
Priority 02
The drainage network is smaller than the terrain grid — 30 m cells cannot contain it
A 2 m channel does not exist in a 30 m surface, so the model routes water across ground it would never cross. Burn surveyed channel sections into the terrain, or capture the corridor at sub-metre resolution.
Priority 03
Calibrate against at least one observed flood before anyone acts on this
Surveyed high-water marks, dated photographs, Sentinel-1 radar extents from a known event, or NADMO incident records. Until the model reproduces a flood that actually happened, its outputs are an untested hypothesis, and the areas it excludes are as unproven as the areas it floods.
Priority 04
Obtain the local rainfall and river-level record
Hydrological Services Department and Ghana Meteorological Agency hold gauge series for most major catchments. A design intensity transferred from another catchment carries an assumption nobody downstream ever agreed to.
Priority 05
Survey the delivered capacity of the drain
Peak flow for this catchment at the 25-year storm is 24.65 m³/s. Without a surveyed section, invert and gradient, there is nothing to compare it against and no reach can be prioritised over another.
Priority 06
State the warning latency — the catchment concentrates in 40 minutes
Time of concentration sets the absolute ceiling on warning. Until the minutes between detection and a household being told are measured, nobody can say whether an early-warning arrangement here is protective or decorative.
Priority 07
State the value at risk so the investment case can be closed
Exposure without a value produces a priority list and no business case. Replacement values for the exposed buildings, infrastructure and stock turn this into a comparison between the cost of the works and the loss they prevent.
Design stage / the chain
Nothing here
stands alone.
HydroGhana™ answers “What does the water do here?”. It produces the flood depth, shoreline change and adaptation case — and that record is exactly what the next desk refuses to work without.
Feeds in
TerraGhana™Terrain accuracy, catchment and outfall levelsAssetGhana™Replacement values that turn exposure into a loss figureMineGhana™Tailings, discharge and downstream exposure to be modelledInfraGhana™Delivered drainage capacity as actually built and certified08 / now
HydroGhana™
Flood, shoreline and coastal resilience intelligence that turns exposure into investable adaptation evidence.
FloodShoreline MRVResilience
Connected intelligence
Evidence, research
and access.
The public evidence library and licensing catalogue stay connected to this capability. New published material appears here automatically.
From analysis to site
Brief one decision.
We resolve the ground.
Submit the catchment or frontage for terrain capture, gauge retrieval, channel survey, model calibration against an observed flood and an exposure count. We return what the ground model can resolve, what the drain actually carries, how many minutes of warning the catchment allows, and the evidence a designer, the EPA or a climate financier will accept.