Landscape Evolution

A landscape is not a static sculpture. Rain falls, water is routed, and moving water removes material where drainage area and slope combine into erosive power. Soft rock yields; hard rock resists; uplift slowly raises the stage again. As channels deepen, they steal water from neighbours, and the map of who drains where becomes a battlefield. The sphere lets visitors watch that feedback: hydrology writes topography, topography rewrites hydrology.

Geomorphology’s stream-power family relates erosion rate to drainage area (A) and slope (S):

Compact form: (∂ z)/(∂ t) = U - K A^m S^n

(with uplift (U), erodibility (K), and exponents (m,n)). The pack exposes defaults near (K=0.012), (m=0.5), (n=1) (advanced coefficients), while the ecospheres simple surface compresses experience into rainfall, rock hardness, uplift, soil depth and storminess.

Coupled modules in the implementation

• Virtual-pipe hydrology: shallow-water style exchange among neighbours moves surface water.
• Rainfall / storms: water is added; storminess intensifies storm-front behaviour.
• Stream-power erosion: geological macro-steps lower bedrock/soil according to drainage and slope.
• Priority-Flood: depressions are resolved so drainage networks remain coherent.
• Layered heightfield: bedrock, soil, surface, water and suspended sediment participate in the stacked state.
• Hillslope / sediment processes (advanced): creep, capacity and erodibility refine the macro story.

Model time is not deep time

Telemetry or copy that speaks in “years” refers to model time scaled for interaction. It is not a claim of calibrated Cenozoic chronology. Teach rates as relative: more rain, softer rock, faster uplift—relative to the running model.

Distinct from evolution

	Landscape Evolution (landscape)	Evolution Landscape (evolution, uncatalogued)
Object	Heightfield, water, rock	Genotypes on a fitness landscape
Process	Erosion, hydrology, uplift	Selection, mutation, population genetics
Public atlas	Yes	No

Never cross-link them as the same sphere.

Quantitative geomorphology connected channel incision to drainage area and gradient through stream-power and related transport laws (work associated with Howard and later Whipple, Tucker and many LE models). Landscape Evolution Models (LEMs) couple those laws to hydrologic routing, tectonics and sometimes sediment flux. Priority-Flood and related depression-filling algorithms (e.g. Barnes and colleagues) made robust drainage extraction practical on digital elevation models. Virtual-pipe and shallow-water methods provide interactive hydrology that can run beside slower geological macro-steps.

ecospheres.ai joins that tradition as an interactive teaching instrument, not as a site-specific Earth forecast model.

Aspect	Current implementation
Domain	Layered heightfield (bedrock, soil, surface, water, suspended).
Hydrology	Virtual-pipe shallow-water outflow; rainfall and storm fronts.
Erosion	Stream-power macro-steps with (K,m,n) coefficients.
Drainage	Priority-Flood depression resolution and drainage network logic.
Uplift	Tectonic-like rise opposing erosion.
Public simple controls	Rainfall, Rock hardness, Uplift rate, Soil depth, Storminess.
Public simple tools	Brush, Erase.
Brush targets	Rain, Raise land, Harden rock, Soften rock, Dam.
Field actions	Storm, Drop sea, Reset terrain, Checkpoint.
Showcase	Desktop Watershed War (0); mobile First Rain Mobile (2).
Hero framing	Host observer camera enabled (Reset view); presentation-only.
Not included	The uncatalogued evolution fitness-landscape engine.

The most truthful one-line description is: a coupled hydrology–stream-power heightfield where catchments and relief coevolve.

Rivers that invent themselves

Water should find and deepen paths rather than follow a painted river texture. Early sheets become threads; threads become valleys.

Divides and capture

In Watershed War, watch competing catchments. Capture appears when one channel steals upstream area and the drainage divide jumps.

Hard rock as a spine

Raise rock hardness or paint Harden: resistant material becomes a ridge while softer neighbours incise.

Uplift versus cut

Increase uplift and ask whether relief rebuilds faster than channels can bevel it. Decrease uplift and watch progressive wearing-down.

Storms as experiments

Storm and high storminess pulse water. Look for sudden incision, flooding of low relief and temporary rewiring of flow.

Deltas and coasts

Braided Delta and sea-level actions show depositional and base-level stories beyond pure mountain incision.

First 30 seconds

1. Load Watershed War on desktop (or First Rain Mobile on a phone).
2. Watch water organise before touching sliders.
3. Trigger Storm once; follow new incision.
4. Brush Rain on one flank of a divide; see whether capture begins.
5. Paint Harden across a ridge; compare later valleys.
6. Nudge Uplift rate and look for renewed relief.

Three experiments

Experiment	Question	Do	Watch for	Why it matters
Capture	Can a channel steal a neighbour’s water?	Watershed War → Rain brush on one headwater.	Divide migration and abrupt rerouting.	Core LE feedback: area begets power begets area.
Resistance	Does lithology organise relief?	Hard Spine or Harden brush vs Soften.	Persistent highs vs rapid gullies.	Rock hardness is causal, not cosmetic shading.
Rise vs cut	Who wins, uplift or erosion?	Rising Range → vary uplift vs rainfall.	Juvenile steepness vs worn low relief.	Competing rates, not a single destiny.

Parameter	What it really controls	Increase it	Decrease it	What to watch
Rainfall	Water added per storm substep.	More discharge and erosional work.	Quieter hydrology.	Channel initiation and flood sheets.
Rock hardness	Global bedrock resistance multiplier.	Slower incision, stronger spines.	Faster wearing.	Ridge persistence.
Uplift rate	Tectonic-like rise in macro-steps.	Renewed relief.	Gradual bevel.	Mountain youth vs peneplain mood.
Soil depth	Soil mantle thickness scale.	More mobile regolith story.	Bedrock-near behaviour.	How quickly soft cover is stripped.
Storminess	Storm-front intensity.	Flashier pulses.	Gentler rain.	Event-driven incision.
Stream-power K/m/n (advanced)	Erosion law coefficients.	Stronger or retuned incision.	Weaker incision.	Scientific LE levers behind the simple surface.
View mode (advanced, visual)	Relief shading / analytic view.	Categorical.	Categorical.	Reading aid.

The best first pair is Rainfall and Rock hardness: water supplies work; rock meters how costly that work is.

Watershed War

Desktop showcase. Competing catchments and capture—best signature phenomenon.

First Rain

Quieter onset of organisation; good literacy before war.

First Rain Mobile

Mobile showcase with reduced surfaceResolution for small plates.

Braided Delta

Sediment and base-level story; pair with Drop sea carefully.

Rising Range

Uplift-forward regime; orographic rain themes may appear in presentation of process, not as a climate GCM.

Hard Spine

Lithology lesson: resistance structures the map.

Monsoon Scar

Extreme storminess/rainfall—stress test, not a weather forecast.

Recommended learning order: First Rain → Watershed War → Hard Spine → Rising Range → Braided Delta → Monsoon Scar (use First Rain Mobile on small viewports).

• Brush → Rain: local water input.
• Brush → Raise: constructive topography edit.
• Brush → Harden / Soften: local erodibility edits.
• Brush → Dam: blockage that rewires flow.
• Erase: removes or cuts according to tool semantics (pack CUT).
• Storm: pulses storm hydrology.
• Drop sea: base-level intervention.
• Reset terrain: returns to preset/seeded land.
• Checkpoint: marks a restore point for comparison experiments.

This is not a forecast model of a real catchment, climate or tectonic province.

“Years” are model time, not calibrated geologic calendar time.

It is not biological evolution and not the uncatalogued evolution fitness-landscape engine.

Brush dams and raised land are interactive experiments, not engineering certifications.

Relief shading is not a second physics.

Sediment and sea-level behaviours are stylised; do not claim sequence-stratigraphic completeness.

The defensible conclusion is:

When water routes across an erodible heightfield under uplift, drainage area and relief rewrite one another through stream power and capture.

Landscape Evolution extends the atlas from life-like agents and fields into Earth-surface process feedback. It shows a slow, spatial contest of rates—rain, rock, rise—that produces rivers and divides as emergent infrastructure. Beside Flow and Materials, it is the clearest geomorphic world in the catalogue, and its name teaches visitors to keep genetic “evolution landscapes” conceptually separate.

Foundational and primary sources

1. Stream-power / detachment-limited incision tradition in quantitative geomorphology (Howard and subsequent LE modelling literature; Whipple–Tucker stream-power frameworks). Use for the (\partial z/\partial t \propto -K A^m S^n) lineage without over-claiming one paper as the engine’s sole equation source.
2. Priority-Flood and related depression-filling / routing algorithms for digital elevation models (Barnes et al. and successors) as the drainage-resolution lineage behind robust catchment extraction.
3. Virtual-pipe / interactive shallow-water hydrology methods as the family for neighbour water exchange used in coupled visual LE systems.