The simulation is an educational model of scent dispersion. Real scent behaviour is complex and can vary with many local factors.
An interactive SAR (search and rescue) laboratory for Mantrailing handlers, instructors, students and search and rescue teams. You build the environment and draw the missing person's path. Then you watch how wind, buildings, sun, rain, surfaces and the passing of time move the available scent away from the physical path.
Who built it and how. The simulator is developed by the Mantrailing Romania team of Asociația Drumeții Montane, starting from what we see in mantrailing training and in search and rescue exercises with dogs. The working rules of the simulated dog (orientation at the start, casting in circles from the last good point, direction chosen by the freshness of the trail) are the rules we work by in training. Trail persistence and the effect of weather are calibrated on the specialist literature. Everything runs in your browser; your scenarios stay with you.
About the accuracy of the simulation
The simulator is an educational tool. The dispersion model uses simplified approximations: a global wind vector, local deflection around obstacles, wake and wind shadow, channelling between buildings, stochastic turbulence, relative deposition and re-emission. It does not solve the equations of air flow. It is not a forensic prediction engine and it is not computational fluid dynamics.
All scent values are relative and normalised into four categories: Very low, Low, Medium, High. No real concentrations are shown and there are no fixed biological persistence times. The simulation profiles (Educational, Conservative realistic, Enhanced visualisation) are presentation and sensitivity presets, not "scientifically exact" modes.
The results must not be used to establish with certainty the location of a missing person or the exact behaviour of a dog in a real operation. The purpose is to illustrate why the scent available to the dog may not coincide with the person's physical path. It also shows how weather, surfaces, buildings and time shape that distribution.
What you can do in the simulator
Build the scenario
buildings, houses, apartment blocks, warehouses, fences, streets, alleys, pavements, car parks, tunnels, natural terrain (grass, soil, sand, forest, vegetation, water, snow), vehicles, street furniture. Everything in world coordinates, measured in metres.
In the guide: Editing, environment and route →The missing person
a path drawn with points, with walking, running, standing still, sitting or getting in and out of a vehicle; the path is timestamped automatically and scent is emitted continuously along it.
In the guide: A new scenario, step by step →Wind and weather
direction 0–359°, speed 0–80 km/h (also shown in m/s), variability, gusts, turbulence; temperature, humidity, weather (clear → blizzard), sun and shade, rain, snow, events on the timeline.
In the guide: The Weather panel →Scent visualisation
particles, scent cloud, heat map, ground deposition, airborne, combined, isolines; scent probe and cross-section profile.
In the guide: The Scent panel →The dog's perspective
only the local olfactory information from the area investigated by the nose, with the "Olfactory signal" chart and simulation states (searching, following, checking, casting, weak signal, scent lost, reacquisition, approaching the source).
In the guide: The Dog panel and the dog's behaviour →Timeline, replay and analysis
play/pause, speeds 0.25×–60×, weather events, saved moments, replay with scrubbing, exercise analysis, intersection analysis, instructor notes, comparisons of conditions and of moments.
In the guide: Timeline and events →
How the simulation works
The conceptual chain is:
- the person's physical path
- scent emission
- interaction with the environment and the weather
- scent field
- local sampling by the dog
- simulated dog behaviour
- replay and analysis
The physical path and the scent field are two different things, drawn differently, precisely to show why the dog does not necessarily walk "in the footsteps".
Scent has two components tracked separately. Airborne scent is carried by the wind, deflected by buildings, channelled along streets and lifted by convection from hot surfaces. Surface-bound scent is deposited on the ground, retained differently by grass, asphalt, soil, leaves or snow, washed and redistributed by rain, and re-emitted when the surface warms up.
Trail age is not a simple fade. The simulator actually evolves the scent field through transport, diffusion, deposition, re-emission, rain and decay. During fast-forward it uses large time steps, so that a 12-hour-old trail is computed in a few seconds.
Who it is for
- Mantrailing handlerscasting, reacquisitions, aged trails
- Instructorsworkshop scenarios, presentation mode with split screen
- Studentswhy the dog does not necessarily walk "in the footsteps"
- Search and rescue teamssectorisation by wind and terrain
For Mantrailing handlers who want to understand casting and reacquisitions. For instructors designing scenarios in workshops: there is a presentation mode with large labels and a split instructor / dog screen. For students and for search and rescue teams discussing sectorisation by wind and terrain. See also how a search with dogs works, Mantrailing training and the search and rescue section.
What comes next
- preparedGPX importthe tracks of the person, the dog and the handler
- preparedHistorical weatherthe real conditions of the training day
- preparedReal mapsOpenStreetMap and georeferencing
- preparedTraining reportthe real session replayed over the simulated scent, with the instructor's notes
The architecture is prepared for importing the GPS tracks (GPX) of the person, the dog and the handler, for historical weather and for real maps (OpenStreetMap). A real training session will then be replayed over the simulated scent and analysed together with the instructor's notes. These functions are marked in the application as "Feature prepared for a future version".