Services

Modeling, systems, software.

Engagements run from a two-week scoping study to a forecast system in continuous operation.

01

Flood risk modeling

ADCIRCSWANHEC-RASSFINCSDelft3D

Coastal surge, riverine, pluvial, and the compound cases where more than one arrives at once. How deep, and how often — answered with the evidence still attached.

What you get
  • Depth and elevation rasters
  • Return-period surfaces
  • Model archive: mesh, forcing, control files
  • Technical report with the limitations stated
02

Forecast and early warning systems

GFSECMWFERA5Parametric TCHPC / cloudThreshold design

Operational chains that pull forcing, run the model, and publish on a schedule without anyone watching — and the threshold logic that turns the result into a warning somebody acts on.

What you get
  • Deployed chain with scheduling and monitoring
  • Threshold catalogue and escalation logic
  • Runbook covering every known failure mode
  • Back-test: skill, hits, false alarms, lead time
03

Machine learning and computer vision

PyTorchSentinel-1 SARSentinel-2SegmentationSurrogate models

Flood extent from satellite imagery, surrogate models that put a screening answer in seconds instead of hours, and structure inventories built from aerial and street-level photography. Trained against physics, scored against ground truth.

What you get
  • Trained model, weights, and inference pipeline
  • Labelled training set, yours to keep
  • Accuracy assessment against withheld ground truth
  • A stated boundary on where the model should not be used
04

Modeling tools and software

PythonFortranOceanMesh2DOWI / NWS12GAHM / NWS20

Mesh generators, forcing builders, validation harnesses — the tools that take preparation off the critical path. Tools commissioned for you are delivered with source and tests; the platform we build them on is licensed, not handed over.

What you get
  • Source repository with tests and CI
  • Command line and GUI entry points as the work warrants
  • Documentation written for the next engineer
  • Worked examples using your own data
05

Validation and independent review

NOAA CO-OPSNDBCUSGSHigh water marks

Skill assessment against gauges, buoys, and high water marks — per station and in aggregate, using metrics chosen up front. Also a second set of eyes on somebody else’s study.

What you get
  • Station-by-station report with maps and time series
  • Aggregate skill tables and error distributions
  • Reproduction package: scripts and observations as pulled
  • A clear position on fitness for purpose

Not sure which of these it is?

Most projects start that way. Describe the decision you are trying to make and we will tell you what the modeling needs to be.

Start a conversationSee how we work