SymbiOS™ Platform Climate Adaptation Coastal Resilience Inflorescence Intelligence™

"Light is the left hand of darkness, and darkness the right hand of light. Two are one, life and death, lying together like lovers in kemmer, like hands joined together, like the end and the way." — Ursula K. Le Guin · The Left Hand of Darkness · 1969

वन · วน · 森林 · Forest

VANA

"The Word for World is Forest."

SymbiOS™ is a coastal resilience platform — a computational design and ecological intelligence system synthesizing eco-sensing mesh networks, MOF-integrated phytoremediation, adaptive zoning algorithms, and an insurance overlay layer. What BIM + ArcGIS would be for urban ecologists — or ArcGIS meets IoT meets predictive analytics, built for living systems.

VANA is Module 01 — a field-deployable plant intelligence database for climate-vulnerable coastal contexts. The full stack reveals itself in the tabs that follow: ARCHITECTURE (eco-sensing mesh), PHYTOREMEDIATION (MOF chemistry), AUXEN (predictive policy + insurance), and PILOT (field sites).

To the Climate Circle review committee: This site is the technical companion to our 2026–27 application. The platform substance referenced in Q17 (problem), Q18 (solution), Q22 (technical edge), and Q32 (why CC) lives in the tabs above. ARCHITECTURE and PHYTOREMEDIATION show the methodology. AUXEN shows the insurance/zoning surface. PILOT shows Brooklyn Inlet Park (BIP), Bangkok, and the Axolotl Rhino/Grasshopper MVP.
i. Etymology · A name that contains its method
वन
SANSKRIT · VANA
forest · grove · the wild. The root from which the project takes its name.
วน
THAI · WON
to return · to circle back · cyclical. The hydrology, the season, the monsoon.
森林
CHINESE / JAPANESE · SĒNLÍN / SHINRIN
forest as compound — many trees. Shared ideographs across writing systems.
Forest
ENGLISH · LATIN ROOT
the outer world. From foris — outside. The system beyond the door.
ii. The platform · Module ecosystem
VANA PHASE 1 · ACTIVE · MODULE 01
Plant Intelligence Database · Field Tool · Dashboard
A field-deployable, climate-aware species library and analytical dashboard for landscape architects working in coastal contexts. The reference layer underneath every downstream module.
XĒN PHASE 1.5 · IN DEVELOPMENT
Generative Pattern Library · M/V/A-System Grammar
A custom growth grammar — 89 botanical specimens encoded as Morphology / Venation / Architecture parameters. Generative companion to VANA's typological data.
AUXLOGRAM PHASE 2 · DESIGN
Eco-Sensing Mesh · Bayesian Inference Layer
Drones + ground sensors + satellite + citizen-IoT phones, fused into a probabilistic decision network. Higher resolution than current satellite products. See ARCHITECTURE.
PHYTOFLEET PHASE 2 · DESIGN
MOF-Integrated Phytoremediation Hardware
Metal-organic frameworks bonded into bio-substrate — phyto-filtration units, tidal blocks, micro-site stormwater berms. Sensing-and-remediation as one object. See PHYTOREMEDIATION.
AUXEN PHASE 3 · LONG-TERM MOAT
Predictive Analytics · Policy Simulation · Insurance Overlay
DAPP decision trees, PPPP scenario engine, exceedance probability maps, RLP insurance correlation, adaptive zoning. The decision surface for insurers, planners, and property owners. See AUXEN.
AXOLOTL MVP · FAST PATH
Rhino / Grasshopper Plugin · Food4Rhino
The thin-slice MVP — VANA + XĒN inside the working tool of landscape architects worldwide. Distribution path through GSD studios + the Food4Rhino ecosystem. See PILOT.
iii. The thesis · in one paragraph

Climate adaptation runs on backward-looking instruments. FEMA flood maps model present conditions and the 100-year storm — they are baked into building codes that cannot respond to a changing climate. Satellite products give continental coverage at site-irrelevant resolution. Insurance pricing arrives after loss, not before adaptation. SymbiOS™ inverts the stack — ecological sensing at site resolution, fused into a probabilistic model that produces forward-looking risk surfaces. Coupled to phyto-remediation hardware that is both treatment and sensor. Coupled to an insurance/zoning overlay that prices adaptation, not loss. The reference modules are open enough for the design profession to extend; the analytics surface is proprietary enough to defend.

AUXLOGRAM · Sensing Mesh Phase 2 · Eco-Intelligence Layer Bayesian Inference

02 · ARCHITECTURE

The Sensing Mesh.

"The landscape is already sensing. AUXLOGRAM makes it legible."

A federated, multi-modal sensing topology. Four ingestion classes — satellite, aerial drone, ground sensor, citizen-IoT — fused into a Bayesian decision network that infers ecological state and projects risk. Hardware-light: the citizen-IoT layer reuses billions of decommissioned smartphones as distributed acoustic, thermal, and barometric nodes. The result is site-resolution intelligence at continental scale, calibrated against ground truth and self-updating against drift.

SENSING MODALITIES
fused per node
~10m
TARGET SPATIAL
RESOLUTION
≤15min
TARGET TEMPORAL
RESOLUTION
$0.08
PER NODE / DAY
(citizen IoT layer)
A. The topology · ingestion → inference → output
FIG. A.1 — ECO-SENSING MESH TOPOLOGY
INGESTION INFERENCE OUTPUT ◉ Satellite Sentinel-2 · Landsat 9 10–30m · 5d revisit ◈ Aerial Drone LiDAR · multispectral · thermal 10cm · campaign-based ◇ Ground Sensors soil · salinity · piezometric site-bound · 15min cadence ○ Citizen IoT repurposed phones — mic, cam, IMU distributed · continuous ◐ PhytoFleet MOF saturation · contaminant flux on-treatment in situ Bayesian Decision Network probabilistic graph · 60+ variables hydro soil veg risk zone EVIDENCE · LATENT · OUTPUT ↻ Predictive Forecast flood · drought · ecosystem state ↻ DAPP / PPPP Sim scenario branching · policy triggers ↻ Adaptive Zoning parcel-resolution recommendations ↻ Insurance Pricing RLP correlation · rate buffer signal ↻ Carbon / ESG Ledger blue-carbon attribution per node
A federated mesh: heterogeneous ingestion, probabilistic fusion, multi-surface output. Bayesian structure makes the model auditable — every output is conditional on stated evidence, and uncertainty is propagated, not hidden.
B. The citizen-IoT layer · billions of phones, already manufactured
THE INSIGHT
A decommissioned smartphone is a 6-sensor instrument.
Every retired phone carries a microphone array, a 12MP camera, an IMU (accelerometer + gyro + magnetometer), a barometer, GPS, and a temperature die. Globally, ~5.3 billion units are decommissioned per year — manufactured electronics whose embodied carbon has already been paid. The citizen-IoT layer reframes these as distributed acoustic + visual + barometric nodes, deployed in mesh density across coastal sites at marginal cost-per-node approaching free.
WHAT IT MEASURES
Phenology, hydrology, acoustics, weather.
Audio: bird call density (biodiversity proxy), insect chorus (pollinator activity), water flow, wind events.
Vision: phenological state, canopy density change, water level (waterline detection).
Barometer + IMU: storm-front passage, vibration from infrastructure stress, soil settlement.
GPS + ambient: air temperature, humidity, position fix for mesh self-localization.
DEPLOYMENT MODEL
A citizen app, not a hardware product.
A municipal partner, NGO, or property owner sponsors a node site. Volunteers or pilot residents install the SymbiOS sensor app on a retired phone, connect it to power, and place it under weatherproof housing (~$8 in materials). The phone joins the mesh, contributes data, and earns the household carbon credits, insurance rate buffers, or property-level resilience scores through the AUXEN overlay. The model is structurally cheap — the hardware exists, the user base is active, and incentives align across the stakeholder map.
C. The Bayesian inference layer · why probabilistic, not deterministic

A Bayesian decision network encodes the conditional dependencies between sensed evidence (the ingestion layer), latent ecological variables (hydrology, soil chemistry, vegetation viability), and downstream outputs (risk, zoning recommendation, insurance signal). Two properties matter for coastal resilience: uncertainty propagates — every output carries calibrated confidence bounds, so insurers and planners can price risk honestly — and missing evidence degrades gracefully — when a sensor goes offline, the network falls back on priors rather than failing. This is the inverse of black-box ML; this is auditable, scientifically defensible inference.

EVIDENCE NODE
DEPENDS ON
INFORMS
Local hydrology
satellite SAR, ground piezometric, drone DEM, climate priors
flood risk, vegetation viability, MOF deployment density
Soil contamination
historic land use, citizen-IoT vision (industrial proximity), PhytoFleet MOF saturation
remediation priority, plant-species selection, insurance overlay
Vegetation viability
satellite NDVI, drone multispectral, citizen audio (biodiversity), VANA plant database
carbon attribution, biodiversity score, blue-carbon ESG ledger
Composite risk
all of the above + DAPP triggers + scenario branches (PPPP)
zoning recommendation, insurance rate buffer, retrofit ROI
PhytoFleet · Phase 2 MOF-Integrated Phytoremediation Sensing-as-Treatment

03 · PHYTOREMEDIATION

Treatment as Sensor.

"A barrier that knows what it has captured."

Conventional phytoremediation works — slowly, opaquely, and at the mercy of which species happen to be planted. PhytoFleet couples plant biology to metal-organic frameworks (MOFs) — crystalline nanoporous materials with tunable affinity for specific contaminants — and bonds the composite into a bio-substrate that serves three functions at once: physical barrier, contaminant capture, and chemical sensor. Saturation state is read off the MOF, so the unit reports its own performance in real time and feeds back into the AUXLOGRAM mesh.

A. The chemistry · why MOFs change the curve
WHAT THEY ARE
Metal-organic frameworks.
Crystalline materials formed by metal ion clusters connected by organic linker molecules — the result is a highly porous, tunable lattice with internal surface areas exceeding 7,000 m²/g. By choosing the metal cluster and the linker, the lattice can be engineered to selectively bind specific molecules: PFAS, heavy metals (Cd, Pb, Hg), nitrates, hydrocarbons, even atmospheric CO₂.
WHY BIO-COUPLED
Living substrate · MOF substrate.
MOFs alone are powdery and require engineered housings. Bonded into chitosan / cellulose / mycelium scaffolds, they become biocompatible matrices that plant roots colonize. Roots maintain hydration, deliver microbial communities, and (critically) translocate captured ions into vacuoles for periodic above-ground harvest. The plant becomes a renewable extraction pump; the MOF becomes a high-affinity buffer.
FIG. P.1 — MOF / PLANT / SUBSTRATE INTEGRATION (CROSS-SECTION)
SURFACE ↑ harvest cycle translocated ions in tissue MOF-Integrated Bio-Substrate chitosan + cellulose scaffold · embedded MOF crystals · porosity ≈ 7,000 m²/g Pb · Cd · heavy metals PFAS / nitrates hydrocarbons contaminated influent tidal flux · stormwater · groundwater treated effluent contaminant-stripped water back to flux ↻ MOF SATURATION 62% · Pb-affine replace cycle: 18d
Plant + MOF + bio-substrate as one integrated unit. The MOF lattice captures specific ion classes; plant roots translocate ions for periodic harvest; saturation is read electrochemically and reported to the mesh — the unit knows when it is full and signals its own replacement.
B. The hardware family · three deployment geometries
TYPOLOGY · ONE
Tidal blocks.
Cast-form MOF-bonded substrate blocks placed along the high-tide interface. Function as coastal surge barriers + contaminant capture. Self-stabilizing keyed geometry, biocompatible for oyster + mangrove colonization. Pilot site: Brooklyn Inlet Park (BIP), Bangkok delta.
TYPOLOGY · TWO
Phyto-cisterns.
Buried micro-site stormwater treatment vessels — modular, parcel-scale. Stormwater enters, passes through a MOF + plant root mat, exits remediated. Saturation monitored; replacement scheduled per AUXLOGRAM readout. Insurance-qualifying retrofit.
TYPOLOGY · THREE
Phyto-fleet eco-machines.
Mobile / barge-mounted MOF-plant arrays for active deployment in contamination events — oil spills, algal blooms, sewage discharges, agricultural runoff plumes. Sensor-equipped; report capture rate and remediation footprint live.
C. Target contaminants · what the system removes
CONTAMINANT CLASS
MOF FAMILY
PLANT PARTNER (SE ASIA / NEW ENGLAND)
SOURCE / RELEVANCE
Heavy metals · Pb, Cd, Hg, As
UiO-66-NH₂ · ZIF-8 (carboxyl, amine functional)
Vetiver · Sunflower · Spartina · Mangrove
Industrial legacy · coastal sediment · brownfield
PFAS · "forever chemicals"
MOF-808 · MIL-101(Cr) (fluorophilic pore)
Phragmites · Typha (with rhizosphere microbes)
Firefighting foam · landfill leachate · airport sites
Nitrates · phosphates · agricultural runoff
HKUST-1 · UiO-66-SO₃H
Cyperus papyrus · Eichhornia · Typha
Coastal eutrophication · algal bloom precursor
Petroleum hydrocarbons · BTEX
MIL-53(Al) · IRMOF-1 (hydrophobic pore)
Casuarina · Salix · Populus
Port-area runoff · vehicle infiltration · spill response
Atmospheric CO₂ · point-source
Mg-MOF-74 · CALF-20
Coupled to mangrove blue-carbon stock
ESG ledger attribution · carbon credit instrument

Note: MOF selections drawn from published literature; specific compositional engineering and biocomposite integration are part of the provisional patent scope.

AUXEN · Phase 3 Adaptive Policy + Insurance Layer Long-Term Moat

04 · AUXEN

Pricing Adaptation.

"Don't design yourself into a box."

AUXEN is the decision surface. It takes the sensing mesh's probabilistic forecasts and the PhytoFleet deployment record and produces the three outputs that planners, insurers, and property owners actually transact on: scenario pathways (DAPP), futures (PPPP), and prices (RLP-correlated insurance buffers + adaptive zoning recommendations). The technical edge is not the data — it's the model that converts uncertain ecological signal into defensible financial and regulatory instruments.

A. DAPP · Dynamic Adaptive Policy Pathways

Borrowed from Dutch flood-management literature (Haasnoot et al.), DAPP encodes adaptation as a branching decision tree triggered by threshold crossings — flood elevation, repetitive-loss rate, sea-level rise, vegetation collapse, MOF saturation. Each branch is a policy pathway with its own cost, time-to-implement, and reversibility profile. AUXEN simulates the tree against the sensing mesh's forward forecasts and surfaces which pathway you are on, which trigger is closest, and what the next branch costs.

FIG. AX.1 — DAPP DECISION TREE — BROOKLYN INLET PARK · BIP (BROOKLYN) · 2025–2080
2025 2040 2055 2070 2080 Pathway A · Status Quo FEMA flood maps · 100-yr storm only trigger missed · 2045 SLR > 0.4m Pathway B · Managed Retreat strategic buyout · zoning rollback trigger: insurance > 4× baseline Pathway C · Hybrid · PhytoFleet + Soft Defense tidal blocks · MOF cisterns · mangrove restoration trigger: vegetation viability index < 0.7 trigger: MOF saturation 80% Pathway D · Hard Engineering seawall · pumps · grey-grey infrastructure cost-prohibitive after 2060 AUXEN RECOMMENDED Pathway C · 28% cost · 71% efficacy
Four pathways under 2050–2080 sea-level-rise scenarios. AUXEN simulates each against the live mesh forecast and surfaces both the current pathway and the next-trigger horizon. The hybrid (Pathway C) is recommended for Brooklyn Inlet Park (BIP) — same site logic transposes to Bangkok with different species and tidal regimes.
B. PPPP · Present / Probable / Plausible / Possible / Preferable

A speculative-futures framework — originally Dunne & Raby (Speculative Everything, 2013), adapted for climate adaptation by Kira Clingen — that refuses the false certainty of a single forecast. Each site is rendered under five concurrent futures, each design proposition serving as a foil for the others. The user sees not "the answer" but the cone of consequences their site is moving through.

FIG. AX.2 — PPPP SCENARIO CONE · BANGKOK COASTAL ZONE · 2070
Present 2026 Possible · catastrophic Plausible · unmitigated Probable · adapted Preferable · regenerative Speculative · post-coastal 2m+ SLR · displacement 1.2m SLR · partial loss 0.8m SLR · phyto + retreat 0.6m SLR · restored deltas post-grid floating typologies ↓ AUXEN-RECOMMENDED CORRIDOR
Five futures held in simultaneous view. AUXEN biases recommendations toward the Probable–Preferable corridor, which is achievable with current PhytoFleet hardware, and uses the Possible slice as a stress-test against design propositions that fail catastrophically under that branch.
C. The insurance overlay · RLP correlation as primary signal

Insurers price coastal risk on backward-looking Repetitive Loss Property (RLP) data — properties that have filed multiple flood claims. The signal arrives late and is geographically coarse. AUXEN correlates the forward Bayesian risk surface with current and projected RLP densities parcel-by-parcel, generating a forward-looking actuarial layer that insurers can underwrite against. The same layer feeds a rate-buffer mechanism — property owners who deploy PhytoFleet hardware or sponsor citizen-IoT nodes earn measurable premium reductions, indexed to verified site improvement.

INSURER VALUE
Price coastal risk like credit risk.
Forward-looking loss curves, parcel resolution, conditional confidence bounds, auditable evidence trail per node. The carrier underwrites adaptation, not loss. Same instrument logic that transformed credit pricing in the 1990s — applied to climate.
PROPERTY OWNER VALUE
Adaptation as a financial instrument.
Install PhytoFleet → measurable improvement on the verified risk surface → indexed premium reduction. Sponsor a citizen-IoT node → carbon credit + insurance buffer + neighborhood-level resilience score. Hardware investment recovers through pricing, not aesthetics.
D. Adaptive zoning · the regulatory output

Zoning codes today are static documents written against historical hazard maps. AUXEN's adaptive zoning algorithm produces parcel-resolution recommendations under each PPPP scenario, with DAPP triggers that schedule when zoning should automatically tighten or relax. Output formats target the actual instruments planners use: FEMA Letter of Map Revision (LOMR), municipal overlay districts, transfer-of-development-rights schedules, and buyout-eligibility maps. The model does not replace the planner — it gives them a model-defensible baseline to negotiate from.

Field Sites · MVP Path Brooklyn Inlet Park · BIP · Bangkok · Axolotl Phase 1 → Phase 2

05 · PILOT

Three Tracks · One Year.

"Strategically coupled to where landscape architects already work."

Three pilots, parallel-tracked. Two field sites at opposite ends of the climate-vulnerability spectrum — a New England salt marsh facing repetitive-loss insurance collapse, and a Bangkok coastal development facing subsidence + storm-surge. Plus a distribution channel pilot: the Axolotl plugin, which embeds VANA + XĒN inside Rhino/Grasshopper — the tool every landscape architect already opens at 9am.

i. Field Site 01 · Brooklyn Inlet Park (BIP), Brooklyn
Brooklyn Inlet Park · BIP BUSHWICK INLET · BROOKLYN · NEW YORK · USA
FEMA Special Flood Hazard Area on a ~160-year petroleum-contamination legacy. Adjacent to dense residential parcels with rising NFIP premium pressure. The test case for AUXEN's North American insurance overlay.
Why here: a remediated post-industrial waterfront inside a designated flood-hazard zone, in a major coastal city with active managed-retreat and resilience policy — the right complexity-to-access ratio for a Phase-1 pilot.

What gets deployed: (1) ground-sensor array along the inlet-upland transition; (2) citizen-IoT app pilot with an adjacent neighborhood; (3) a 12-month DAPP scenario simulation against NYC climate-resiliency inundation projections; (4) a single PhytoFleet phyto-cistern installation at one cooperating property as proof-of-concept.

Partnerships: Resilience-track advisors are the right early channel. NYC DEP, the Mayor's Office of Climate & Environmental Justice, and Brooklyn Community Board planning are the public-sector targets. DCP for regional zoning. NFIP / FEMA Region II for the insurance overlay test.
ii. Field Site 02 · Bangkok / P Landscape
Bangkok Coastal Zone SAMUT PRAKAN / BANG KHUN THIAN · THAILAND
2.5cm/year subsidence. Sea-level rise compounding. Monsoon-storm-surge interaction. The test case for tropical PhytoFleet typologies and the SE Asia coastal market.
Why here: Bangkok is the most-cited near-term canonical case for coastal urban climate collapse. The vulnerability is severe, the political will is real, and the design-and-construction ecosystem is sophisticated. The luxury-resort layer in particular operates at a price-per-square-meter where adaptation hardware is financially trivial relative to project budgets.

What gets deployed: tropical species library expansion within VANA (mangrove, nipa palm, casuarina, sea hibiscus, coastal fig); MOF-bonded tidal-block prototype against monsoon surge; PPPP scenario set for 2050 / 2070 SLR under three subsidence trajectories.

Anchor partner: P Landscape (PLA) — Bangkok-based luxury-resort landscape architecture firm. Founded by Wannaporn "Pui" Phornprapha, Harvard GSD MLA alumna. PLA has had an open Digital Solution Specialist role since late 2025; the VANA module set is effectively a portfolio for it. Engagement model: written agreement for a pilot site partnership, regardless of formal employment outcome.
iii. Distribution Pilot · Axolotl plugin
Axolotl RHINOCEROS / GRASSHOPPER · FOOD4RHINO ECOSYSTEM
VANA's plant intelligence layer + XĒN's generative grammar — inside the tool 80% of landscape architects already use daily. The fast-path MVP for distribution + traction.
Why this matters: the field doesn't need another platform to learn. It needs its existing toolchain to become climate-literate. Axolotl ships VANA's plant database as queryable Grasshopper components (filter by salt tolerance, climate zone, ecosystem service), XĒN's grammar as procedural growth nodes, and AUXEN's risk surface as a context-overlay layer for site model imports.

Why "Axolotl": regenerative species; the GSD's mascot of resilience; reads cleanly in the Rhino plugin taxonomy alongside other animal-named tools (Lunchbox, Anemone, Pufferfish, Kangaroo). USPTO TESS clearance pending — fallback names ready if not available.

Distribution: Food4Rhino (free tier, ~250K user base); GSD studio integration (direct channel); McNeel community channels (Discourse, Discord). A working v0.1 with five exemplary components is the Phase-1 ship target.
iv. 18-month milestone view
QUARTER
BROOKLYN INLET PARK · BIP
BANGKOK / PLA
AXOLOTL
Q3 2026
Site MOU · sensor array spec
PLA written agreement · species library expansion
v0.1 ship · 5 GH components · Food4Rhino listing
Q4 2026
Sensor install · citizen-IoT app beta · 1 PhytoFleet unit
Tidal block prototype · monsoon-season data
v0.2 · AUXEN context overlay · 200+ installs target
Q1 2027
DAPP scenario report v1 · insurance overlay test
PPPP scenario set · resort pilot design integration
v0.3 · paid tier · pro features
Q2 2027
RLP rate-buffer pilot with NFIP · publication
PhytoFleet field performance data · ESG ledger
v1.0 · documented API · 1K+ installs
XĒN · Generative Grammar M/V/A-System · 89 specimens Companion Module

06 · XĒN

XĒN

A generative pattern library — 89 botanical specimens encoded as Morphology / Venation / Architecture parameters. Companion to VANA's typological data; substrate for generative computational design.

OPEN XĒN COMPENDIUM
89 SPECIMENS · MORPHOLOGY · VENATION · HALLÉ–OLDEMAN ARCHITECTURE
WHY XĒN
The grammar underneath every site decision.
VANA tells you which plant. XĒN tells you how it grows — the morphological grammar that generative design tools (Rhino, Grasshopper, Houdini) need to render plants as parametric objects, not as static reference images. Every specimen encodes leaf morphology, venation topology (after Hallé), and architectural model (after the Hallé–Oldeman tropical-tree typology). The compendium is the substrate the Axolotl plugin draws from.