Skip to content
Loading the beaches...
Draw on the map:
water, shaded by surface slopefoamdry sandlandreefbreaking linemean currenttransect (click the map with the transect tool)
y = 300 m from the top

What is simulated live: a 2D Boussinesq-type model (nonlinear shallow water with Peregrine dispersion) on a map-aligned grid, 480 m east-west by 600 m north-south at 2 m cells, 0.07 s step, periodic north-south, with a random-phase directional spectrum relaxed in along the eastern edge. The coastline runs through the grid at the beach's own orientation; land fills the west. Refraction, shoaling, headland shadowing and diffraction, wrapping, sets and groups, breaking position (front-steepness and depth-limit criteria with the D21 wind correction), bores, wave setup, undertow, longshore current and rips come out of the equations. Foam is advected by the resolved flow. The bank, reef, headlands, bay curvature and water level are built from the inputs as sections B and C describe; the presets are fictional Sydney-like layouts.

Drawing: pick a tool above the map and drag. Land paints headland or rock platform, Reef raises the bed to the paint depth with reef friction, Sand bank does the same with sand friction, Channel lowers the bed to the paint depth, Erase restores the parametric bed. Drawn edits sit on top of the slider-built beach and survive slider changes; choosing a preset clears them. The transect tool sets the east-west profile line.

What is not: the barrel (the A.4 clip library would be stitched on at the plunging segments; here plunging cells are only tagged and foamed), wind texture physics (chop is a render-only overlay), headland wind shelter, sediment movement (the bank is a parameter, not a state), the shelf (the eastern edge sits in 5 to 13 m of water depending on the slope), and dispersion above kd of about 1.7, so periods under 6 s are approximate. Because the north-south boundary is periodic, each spectral component's direction snaps to the nearest periodic value (a few degrees); the stats panel shows the effective angle. Swell from the western half of the compass cannot enter and is treated as outside the window. The breaking-rate threshold is calibrated to the 2 m grid (0.35, where SWASH uses 0.6 at half-metre cells) and D21 scales it in proportion.

Offshore source and clean zone: every beach now has an extension to seaward. The outer strip is the swell source, drawn as stylised crests: it is the incident spectrum, the swell as it would be infinitely far out, relaxed into the model, so the crests there tilt with the swell direction and pulse with the groups. Between the source and the beach is about 100 m of flat deep water where the simulated swell travels undisturbed, so nothing the beach does reaches back into the source.

Map beaches: the seven real beaches are drawn from real survey data, not estimated from the photos. The ground (land, sand and seabed) is the NSW Marine LiDAR Topo-Bathy survey of 2018, regridded north-up with smooth interpolation; lagoons and anything walled off from the sea are land to the solver and are the photo; the sea is always the simulation, coloured by its own depth. Reef is the mapped reef extent. Heights above 3 m, and wherever the laser did not reach (white water and murky water are filled smoothly from the depths around them), are not measured, and the open sea added to the right is a flat 12 m. The satellite photos are placed by fitting their sea to the surveyed coastline, so they line up with the bed to about a photo pixel; where the surveyed waterline and the photo's differ (the photo has its own tide, and a sandbar moves), the water level is the thing to tune. Cells are 4 to 8.5 m on these beaches, so the surf zone is coarser than on the parametric beaches. Edges that show open sea are source edges.

Checks: the "model vs formula" table compares what the simulation measures at its own breaking line with the section D formulas computed from the inputs. Expect agreement within about 20% on Hb and db when the inputs sit inside the typical bands; outside them the formulas extrapolate as much as the model does.

Data credits: seabed and ground, NSW Marine LiDAR Topo-Bathy 2018, NSW Department of Planning and Environment, CC BY 4.0, via the Australian Ocean Data Network. Coastline, NSW Spatial Services, NSW Hydrography, CC BY 4.0. Reef extent, NSW Seabed Reef Extent, Department of Climate Change, Energy, the Environment and Water (2023), CC BY.