# FLOOD Scientific Atlas — research and implementation notes

This collection contains **72 original, asset-free patches**, arranged as six collections of twelve. Every patch has three geometry looks, six named scene controls, projection controls, line material controls and its own serialized shader. The shared FLOOD media, mixing, audio routing and export facilities are supplied by the parent engine.

The graphics are deliberately instruments for visual performance. They do not supply a plasma calculation, quantum execution environment, disease forecast or surveyed terrain. Some patches directly encode a named mathematical construction; others use the visual vocabulary of a scientific subject. That distinction is recorded below and in patch descriptions.

## Primary source grounding

The following sources informed the subject matter and terminology. No diagrams, photographs, source code, laboratory datasets or private source material were copied into the patches.

- **Princeton Plasma Physics Laboratory:** [Model C Stellarator (1961–1969)](https://www.pppl.gov/posts/2026/model-c-stellarator-1961-1969) describes the historical progression of stellarator magnetic confinement and associated research. [IPPEX](https://ippex.pppl.gov/) provides an institutional educational context for fusion and tokamak ideas. These sources ground the toroidal, stellarator, coil and confinement vocabulary; FU geometry is original and uncalibrated.
- **MIT OpenCourseWare:** [8.03SC, Lecture 9 — Wave Equation, Standing Waves, Fourier Series](https://www.ocw.mit.edu/courses/8-03sc-physics-iii-vibrations-and-waves-fall-2016/pages/part-i-mechanical-vibrations-and-waves/lecture-9/) grounds the use of separable standing waves and superposition. PH02, PH03 and PH06 draw explicit finite formulas. They do not solve an arbitrary wave boundary-value problem.
- **Union College, Davide Cervone:** [Geometry of the Fourth Dimension](https://www.math.union.edu/~dpvc/math/4D/) is a primary university course on higher-dimensional geometry, rotations, projections and sections. HD01 constructs a tesseract rather than relabeling a three-dimensional cube; HD02 and HD11 explicitly construct four-coordinate curves. The other HD instruments are distinct three-dimensional mechanisms or immersed surfaces.
- **IBM Quantum Learning:** [Quantum information — single systems](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/single-systems/quantum-information) grounds the distinction between complex amplitudes and normalized probabilities. [Entanglement in action](https://quantum.cloud.ibm.com/learning/en/courses/basics-of-quantum-information/entanglement-in-action/introduction) grounds the terminology around Bell states and correlations. [VQE learning module](https://quantum.cloud.ibm.com/learning/en/modules/computer-science/vqe) provides additional Bloch-sphere context. QC10 encodes a normalized two-outcome pair; decorative matrices, gates and linked curves are not quantum simulations.
- **CDC Center for Forecasting and Outbreak Analytics:** [Explainer: What is a transmission model?](https://www.cdc.gov/cfa-modeling-and-forecasting/about/explainer-transmission-models.html) distinguishes approaches to describing spread. [Field Epidemiology Manual — Describing epidemiologic data](https://www.cdc.gov/field-epi-manual/php/chapters/describing-epi-data.html) supplies institutional context for contact diagrams. DV patches borrow only abstract network, compartment and mobility vocabulary. They contain no pathogen identity, biological parameters, route recommendations, real contacts or public-health predictions.
- **USGS:** [Topographic maps](https://www.usgs.gov/programs/national-geospatial-program/topographic-maps) and [What is a Digital Elevation Model?](https://www.usgs.gov/faqs/what-a-digital-elevation-model-dem) ground the distinction between height fields and measured geographical data. TF patches are invented analytic height fields sampled as projected profiles; no USGS data is downloaded or reproduced.

Sources were consulted on 9 October 2026. The formulas and GLSL described below are the implementation's own construction, not quotations from those sources.

## Fusion — FU01–FU12

All twelve are conceptual geometric studies. The controls use graphical distances and angles, not engineering units. There is no electromagnetic field solver, thermodynamic model, plant design or weapon-related content.

| ID | Composition | Distinct construction and six scene controls |
| --- | --- | --- |
| FU01 | Toroidal winding atlas | Nested toroidal helices: filament count, major radius, tube radius, winding turns, vertical elongation, breathing excursion. |
| FU02 | Stellarator braid | A rippled toroidal axis with independent filament roll: filament count, axis radius, ripple depth, axis lobes, roll, vertical excursion. |
| FU03 | Nested plasma cross sections | Open toroidal cross-section bank: section count, center radius, tube radius, opening, elongation, tilt. |
| FU04 | Mirror throat coils | An axial ring chain contracted at the center: station count, axial span, end radius, contraction, axial ripple, cant. |
| FU05 | Cusp lattice vessel | Cusped loop planes with alternating folds: plane count, radius, height, fold count, sharpness, torsion. |
| FU06 | Reconnection saddle | Opposing hyperbolic-like banks: line count, approach span, gap, curvature, sheet thickness, separation pulse. |
| FU07 | Flux surface foliation | D-shaped contour layers: layer count, radius, triangularity, stretch, separation, axis wobble. |
| FU08 | Divertor fan study | Curved traces opening from a throat to a fan: trace count, landing width, height, branch split, bow, landing ripple. |
| FU09 | Current sheet corrugation | Sinusoidal sheet profiles with shear and curled edges: profile count, span, ripple height, cells, shear, curl. |
| FU10 | Poloidal coil constellation | Rounded rectangular coil assemblies: count, chamber clearance, width, height, squareness, swivel. |
| FU11 | Alfven-inspired bridge | Bowed bridges with standing transverse perturbations: strands, anchor separation, rise, nodes, displacement, spread. The name is an analogy, not an Alfvén-wave solution. |
| FU12 | Interlocked helical cage | Counter-wound axial helices: strands, length, radius, turns, end flare, bank separation. |

## Physics equations — PH01–PH12

The equations here define the displayed geometry. They are not general numerical solvers. Time is a bounded periodic phase, with a default twelve-second visual cycle.

| ID | Composition | Encoded mathematics and scene controls |
| --- | --- | --- |
| PH01 | Oscillator phase portraits | `x=A cos(theta)+e cos(3theta+t)`, `y=B sin(theta+lag)` plus a small layer motion. Controls: layers, position amplitude, momentum amplitude, lag, perturbation, spacing. The unperturbed core is an ellipse. |
| PH02 | Standing wave laboratory | `A sin(n pi u) cos(t)+B sin(2n pi u) sin(t)`. Controls: strings, span, fundamental amplitude, nodes, second harmonic, separation. |
| PH03 | Counterpropagating ribbons | Sum of oppositely traveling sine terms. Controls: ribbons, span, forward amplitude, wave count, reverse amplitude, spread. |
| PH04 | Lissajous observatory | Three sine coordinates with integer spatial ratios. Controls: layers, horizontal frequency, vertical frequency, vertical extent, depth excursion, phase separation. |
| PH05 | Elliptic orbit folio | `x=a(cos(theta)-e)`, `y=a sqrt(1-e²) sin(theta)` on inclined planes. Controls: planes, semi-major axis, eccentricity, inclination, apsidal spread, center excursion. No gravitational integration. |
| PH06 | Fourier ribbon synthesizer | Finite first-, third- and fifth-harmonic sum. Controls: ribbons, span, three harmonic weights, layer separation. |
| PH07 | Two-source interference sections | Sine waves evaluated at the distances from two points and added. Controls: sections, source separation, wave number, amplitude, relative phase, width. No diffraction boundary solver. |
| PH08 | Dipole stream folio | Revolved `r=R sin²(theta)`-style curves with a cut polar opening. Controls: planes, reach, opening, axial stretch, plane spread, sway. No field integration. |
| PH09 | Helical trajectory chamber | Circular coordinates plus axial travel, with an optional radius perturbation. Controls: trajectories, radius, turns, displacement, guiding-center spread, radius modulation. |
| PH10 | Gaussian wave packet | `exp(-x²/sigma²)` envelope multiplied by a chirped cosine. Controls: sections, envelope width, carrier count, height, travel, chirp. No dispersive PDE evolution. |
| PH11 | Potential well contours | Quadratic and quartic radial height terms on elliptical contours. Controls: levels, radius, quadratic stiffness, quartic stiffness, anisotropy, equilibrium excursion. |
| PH12 | Light cone sections | Radius proportional to absolute axial coordinate, plus optional waist and shear. Controls: slices, span, opening, waist, shear, oscillation. A geometrical diagram, not a relativistic simulation. |

## Higher-dimensional machinery — HD01–HD12

“Higher-dimensional” identifies the collection's theme. It does not imply that every instrument is a four-dimensional object. The actual four-coordinate projections are explicitly identified.

| ID | Composition | Construction and controls |
| --- | --- | --- |
| HD01 | Tesseract differential | All 32 hypercube edges; rotations in XW and YW, perspective through W, then a 3D camera. Controls: shells, scale, W extent, XW excursion, YW excursion, W perspective. |
| HD02 | Clifford torus transmission | Product-of-circles curve in four coordinates, XW rotation, W perspective. Controls: fibers, two radii, second-circle turns, rotation extent, projection strength. |
| HD03 | Mobius strip drive | Möbius parameterization with an odd half-twist order and optional perturbation. Controls: rails, loop radius, width, half-twist order, corrugation, hub tilt. |
| HD04 | Klein immersion manifold | Figure-eight Klein-bottle-style immersion, sampled as longitudinal rails. Controls: rails, loop radius, section width, eight separation, neck skew, breathing. Self-intersections are intentional. |
| HD05 | Nested hinge frames | Independently hinged nested squares. Controls: frames, size, taper, opening, separation, hinge phase spread. A 3D mechanism. |
| HD06 | Orthogonal gimbal engine | Rings rotating about two ordinary spatial axes. Controls: rings, radius, stagger, two swing amplitudes, bearing separation. A 3D mechanism. |
| HD07 | Screw extrusion press | Polygonal sections rotated along an extrusion axis. Controls: frames, sides, radius, length, twist, taper. A 3D mechanism. |
| HD08 | Prismatic torsion scaffold | Twisting longitudinal scaffold rails. Controls: uprights, radius, height, torsion, waist contraction, breathing. A 3D mechanism. |
| HD09 | Moving section engine | Tilted ellipse sections shifted through depth. Controls: sections, width, height, spacing, tilt fan, travel. A 3D section instrument. |
| HD10 | Linked ring differential | Alternating perpendicular linked-ring geometry. Controls: rings, radius, spacing, tilt, ellipticity, flex. Linkage is visual; contact is not simulated. |
| HD11 | Four-axis ladder | A four-coordinate helical ribbon, XW rotation, W perspective. Controls: rails, radius, axial travel, turns, W amplitude, rotation. |
| HD12 | Octahedral rack | Alternating diamond cage sections. Controls: cages, width, height, spacing, rotation, pole precession. A 3D rack, not a regular 4D polytope. |

HD01 and HD02 clamp their projection denominator when a control combination moves geometry too close to the projection eye. This avoids singularities; it is a graphical near-plane treatment, not a different dimensional formula.

## Quantum computations — QC01–QC12

None of these patches executes a quantum circuit. QC10 uses an exactly normalized two-outcome formula; the other compositions mix geometric representations and conceptual interface graphics. Ordinary linked loops must not be interpreted as physical entanglement.

| ID | Composition | Construction and controls |
| --- | --- | --- |
| QC01 | Bloch meridian instrument | Spherical coordinates as the geometric starting point, with deliberate display offsets and precession. Controls: meridians, radius, polar opening, azimuth spread, axis bias, precession. Nonzero axis bias makes this an illustrative display rather than an exact unit sphere. |
| QC02 | Gate rail sequencer | Raised windows on circuit rails. Controls: rails, span, gate height, gate width, separation, depth. Conceptual gates; no unitary matrices. |
| QC03 | Bell pair bridge | Coupled loop/bridge geometry. Controls: connections, separation, radius, twist, bow, breathing. An entanglement diagram metaphor. |
| QC04 | Density matrix terraces | Rows of pulsed heights with diagonal emphasis. Controls: rows, columns, footprint, diagonal height, off-diagonal height, phase motion. The heights are not constrained to form a valid density matrix. |
| QC05 | Fourier phase loom | Integer-frequency sinusoidal phase rails. Controls: rails, base frequency, span, amplitude, separation, offset fan. Conceptual register view. |
| QC06 | Discrete walk fan | Piecewise branching/zigzag paths. Controls: paths, steps, travel, spread, alternating displacement, lift. No walk amplitudes are propagated. |
| QC07 | Complex amplitude wheels | Complex-plane circles and rotating phasors. Controls: wheels, radius, falloff, spacing, tilt, wobble. Radii are amplitudes, not probabilities. |
| QC08 | Phase estimation dial | Nested arcs with radial marker pulses. Controls: arcs, radius, spacing, opening, markers, excursion. No estimation algorithm is run. |
| QC09 | Stabilizer lattice instrument | Alternating raised check loops. Controls: checks, radius, spacing, elevation, rotation, pulse. No syndrome or error-correction calculation. |
| QC10 | Born weight comb | `p0=cos²(theta/2)`, `p1=1-p0`. Controls: registers, span, state angle, angle excursion, separation, height scale. The two coefficients sum to one before visual height scaling. |
| QC11 | Two-arm interferometer paths | Bowed arms with a marker ripple. Controls: paths, arm length, separation, bow, plane spacing, ripple. Schematic path geometry. |
| QC12 | Tensor contraction orchard | Hierarchical branching curves. Controls: leaves, height, fan width, contraction pinch, depth, flex. No tensors are contracted. |

## Disease vectors — DV01–DV12

The title is thematic. Every scene is an invented, abstract contact or flow graphic. No pathogen is represented, no transmission coefficient is estimated, and no real location, person or biological dataset is included.

| ID | Composition | Distinct construction and controls |
| --- | --- | --- |
| DV01 | Contact constellation | Spatial curved contact rings: count, reach, hub separation, bow, rotation, pulse. |
| DV02 | Bipartite bridge map | Routes between two node banks: connections, bank separation, spread, crossing reversal, bridge rise, depth drift. |
| DV03 | Mobility hub wheel | Spiraling hub-to-destination routes: routes, radius, bend, elevation, central clearance, lift pulse. |
| DV04 | Abstract transport field | Sheared and lifted streamlines: count, span, shear, eddies, elevation, bank separation. No aerosol model. |
| DV05 | Compartment exchange channels | Repeated expanded channel chambers: channels, span, chambers, expansion, separation, excursion. No epidemic compartment equations. |
| DV06 | Bottleneck islands | Cluster banks joined by narrow bridges: strands, separation, radius, narrowing, depth, flex. |
| DV07 | Contact wavefront rings | Asymmetric pulsing rings: count, radius, spacing, asymmetry, lift, pulse. Not an infection-radius display. |
| DV08 | Delayed relay ladder | Staggered loop stages: stages, width, height, separation, phase lag, stagger. |
| DV09 | Neighborhood contact grid | Warped rectangular lattice loops: cells, width, height, spacing, stagger, lift. |
| DV10 | Interrupted route gates | Routes diverted around an exclusion window: routes, span, window width, diversion height, layer spacing, opening pulse. |
| DV11 | Multiplex contact decks | Separate polygonal network layers: decks, sides, radius, separation, rotation, flex. |
| DV12 | Relay attenuation garden | Shrinking relay loops: loops, initial radius, attenuation, separation, plane fan, excursion. Attenuation is a visual scale rule. |

## Topographical flyovers — TF01–TF12

These use explicit height functions and a closed orbital sampling path. They are projected profile sweeps rather than opaque triangulated terrain or measured contour maps. The unoccluded line style intentionally shows the underlying sections.

| ID | Composition | Height field and controls |
| --- | --- | --- |
| TF01 | Folded mountain flyover | Folded absolute-sine ridges. Controls: profiles, relief, ridge frequency, cross folding, orbit reach, valley lift. |
| TF02 | Impact basin flyover | Gaussian annular rim minus a central depression. Controls: profiles, rim height, radius, depression, orbit reach, rim thickness. |
| TF03 | Terraced plateau survey | Softened steps of a Gaussian mound. Controls: profiles, height, footprint, terrace count, orbit reach, asymmetry. |
| TF04 | Meandering canyon flight | A curved Gaussian channel removed from a plain. Controls: profiles, wall height, width, meander, orbit reach, meander frequency. |
| TF05 | Archipelago elevation sweep | Repeated Gaussian hills over an undulating base. Controls: profiles, summit height, spacing, footprint, orbit reach, floor undulation. |
| TF06 | Dune sheet traversal | A bent sine ridge plus second-harmonic crest asymmetry. Controls: profiles, height, frequency, bend, orbit reach, asymmetry. |
| TF07 | Glacial valley corridor | Squared U-shaped flanks with shoulder ridges. Controls: profiles, shoulder height, floor width, bend, orbit reach, ridging. |
| TF08 | Braided delta scan | Repeated curving depressions. Controls: profiles, channel depth, braid count, width, orbit reach, braiding excursion. No water-flow simulation. |
| TF09 | Seamount transect | Main and secondary Gaussian peaks over rippled relief. Controls: profiles, elevation, footprint, secondary shoulder, orbit reach, basin ripple. |
| TF10 | Caldera rim traverse | Broad Gaussian cone minus a narrow central Gaussian. Controls: profiles, elevation, cone footprint, caldera radius, orbit reach, depression. |
| TF11 | Fault escarpment flight | Smooth stepped diagonal fault with rolling background relief. Controls: profiles, step height, diagonal shear, transition width, orbit reach, background relief. |
| TF12 | Mesa corridor flyover | Repeated flat-topped softened box hills. Controls: profiles, height, spacing, top width, orbit reach, saddle relief. |

## Implementation contract

- Module: `MPU_Revision_Studio/scientific-atlas.cjs`, directly exported factory `makeScientificAtlas({N,C,B})`.
- Returns FU01–FU12, PH01–PH12, HD01–HD12, QC01–QC12, DV01–DV12 and TF01–TF12. Each has `id`, `title`, `file`, exact collection name, functional category, tags, description, license, fields, looks, defaults, serialized source and builder.
- Each patch has six descriptive scene parameters under stable keys `saA` through `saF`. Their values are passed through live `MPU_MODULATED` getters. The same field names do not imply the same meanings between scenes; the schema labels and metadata carry the actual meanings.
- Additional controls: loop duration, yaw, pitch, framing, trace width, halo, primary color, alternate color and background color. The alternate color requires at least two traces. Controls based on layer spread naturally have less or no effect on a single central trace; their settings remain available when more traces are restored.
- Looks have IDs `original`, `study`, `accent`, named per collection. Every look resets to defaults and sets an explicit six-parameter geometry vector. Pairwise looks differ in at least three scene parameters, as well as the study/accent camera. They are not palette swaps.
- `extraCode` declares `MPU_STUDY_GLSL`; `build` is closure-free and installs a Hydra source before calling `MPU_FINISH` once. There are no runtime dependencies, textures, samplers or custom globals beyond the normal serialized patch contract.
- Defaults request **720p at 30 fps** through `defaultOverrides`; shared resolution and playback controls remain available. The factory adds no feedback chain. User-added shared effects, feedback, cameras or audio may alter motion, repeatability and GPU cost.
- Rendering uses a bounded maximum of six traces. Normal curves use 24 segments; selected simple frames use 4 or 8, and more complex curves use up to 32. HD01 uses 32 distinct edges per shell, with at most three authored shells. The renderer evaluates each ordinary curve vertex once, projects it, then computes screen-space segment distance. It does not raymarch.
- Integer spatial frequencies do not multiply the live time by uncontrolled factors. All scene time terms use sine/cosine of the normalized loop phase. The base geometry's default period is twelve seconds. Shared feedback and external media are outside that mathematical guarantee.
- 3D and 4D perspective denominators are clamped near the projection eye. Squared terms use explicit multiplication; the shader contains no `pow` calls, avoiding undefined negative-base cases on WebGL1/mediump implementations.

## Verification status

The focused Node test suite passed **5/5 tests** after the complete 72-scene factory and first browser-discovered correction were written. Those tests check the exact six-by-twelve catalog; unique shader source per scene; six bounded, named, shader-bound controls; all 216 described and bounded geometry looks; pairwise geometry setting differences; standalone serialization with live modulated getters; phase wrap; and the absence of nested GLSL helper declarations, unbounded loops, samplers, unsafe power calls and invalid numeric literals. The additional regression prevents an authored scalar from shadowing the shared segment renderer's vector locals.

The first browser batch found a shader compilation failure in HD05: its square-frame snippet used the scalar name `edge`, which collided with the segment renderer's vector `edge`. The snippet now uses `frameSide`. The correction is confined to HD05; a fresh GPU receipt is still required before declaring that patch visually accepted.

These are structural/export checks, not a claim of visual acceptance or GPU performance. The parent task owns actual WebGL compilation, rendered control-difference checks, screenshots and integration. Browser findings should be appended here after their reports arrive. No public deployment or package receipt is asserted by this document.