||
Rebel's Daedalus
A Contemporary Interpretation of the Victorian Meteorological Pillar
SINGAPORE · EQUATORIAL CLIMATE ZONE
An active art conservation field experiment in one of the world’s most climatically aggressive environments.
Most fine art conservation is studied in climate-controlled laboratories. We built a live field test instead: a Rebel artwork print on museum-grade cotton-bamboo paper using pigment inks rated for up to 200 years of lightfastness, sealed inside a custom-engineered, passively buffered microclimate frame, and exposed to the ambient tropical conditions of Singapore.
Internal sensors continuously log the four primary drivers of paper and pigment degradation (temperature, relative humidity, UV radiation, and visible light), while external telemetry tracks local air quality (PM2.5, CO2, TVOCs, and NOx).
Two postcard-sized control prints serve as reference baselines. By streaming internal and external data in real time, this project maps material deterioration profiles under equatorial stress to test the limits of passive preservation and quantify long-term color shift.
The Science of the Permanent
A Work In Progress
Humans are wired to collect as epitomised by the greatest institution of the ancient world: the Library of Alexandria, the embodiment of ultimate learning and a "repository of everything".
Yet only in relatively recent history have we had the means to measure and protect what we select to preserve. For nearly 150 years, museums and archives have monitored their environments to safeguard cultural memory — from manual readings in the 19th century to today's continuous digital systems. Rebel Archive brings that same institutional discipline and audacious aims inside the frame, in its bespoke Meteorological Curiosity.
I. NASA-Grade Digital Fidelity
At the core of this archive is a commitment to data density. Traditional digital art often suffers from resolution loss when translated to large-scale physical mediums. To combat this, every work in the Rebel Archive is processed using interpolation algorithms similar to those utilized by NASA for enhancing satellite imagery. This produces a master file of 24,000 × 24,000 pixels, ensuring that even under a microscope, the generative complexity remains sharp and the transition from virtual to physical is limited only by the printer technology.
II. Paper As The Substrate
The choice of Hahnemühle Bamboo (290gsm) is a deliberate intersection of environmental ethics, artistic choice and material science. Made from 90% bamboo fibre and 10% cotton, it offers a unique structural integrity. It is acid-free, lignin-free, calcium carbonate buffered, and meets the ISO 9706 museum standard for age resistance. It contains no Optical Brightening Agents (OBAs) that turn yellow or brittle over time, ensuring the base white of the art remains constant for centuries.
Hahnemühle's own technical specifications test this paper at 23°C and 50% relative humidity — the manufacturer's calibration point for the substrate. The paper's recommended storage and display range is 35–65% RH and 10–30°C. Singapore's open-air ambient conditions exceed both thresholds routinely.
III. Ink As The Medium
We utilize the Canon PRO-566 system, featuring the newly engineered Lucia Pro II ink set. These are 100% pigment-based inks, designed for molecular stability and rated for up to 200 years of lightfastness under UV-filtered display. This project seeks to verify these laboratory claims against real-world, localized "Rebel" conditions.
It is worth noting what the research literature makes clear: in tropical conditions, humidity accelerates fading and it changes the degradation pathway entirely. At ambient RH levels above 80%, which Singapore regularly reaches, inkjet dyes can bleed visibly within 24 hours on unprotected prints. The pigment-based Lucia Pro II formulation is significantly more resistant to this hydrothermal attack than dye-based alternatives, but it is not immune.
IV. The UV Shield
The frame is glazed with Tru Vue Optium Museum Acrylic, the conservation standard trusted by the world's leading museums. It blocks 99% of UV rays across the 300–380nm range. It carries anti-reflective coating on both faces for optimal light transmission, anti-static protection to repel dust and particulates in an open-air environment, and abrasion-resistant hard coat for long-term durability.
Critically for this installation, it passes humidity resistance testing, salt fog corrosion resistance, and has undergone accelerated aging equivalent to approximately 100 years of exposure at 100,000 lux with properties remaining unchanged.
The acrylic is sealed into the frame rebate using a permanently flexible, solvent-free sealant. Rigid sealants crack under the thermal expansion and contraction driven by Singapore's day-night temperature cycling. A designed air gap between the acrylic face and the print surface ensures that even during rapid temperature drops in heavy rain — when condensation risk is highest — the glazing surface never contacts the print.
V. The Humidity Buffer
Behind the print sits a layer of Art Sorb conservation buffering sheet, preconditioned to 50% relative humidity. This target is derived directly from two converging sources: Hahnemühle's own test conditions for the Bamboo substrate, and the Art Sorb specification for papers, woods, and organic materials, which recommends 50–60% RH. The lower end of that range — 50% — is chosen deliberately, because in an environment where ambient humidity regularly exceeds 70–80%, starting closer to ambient would exhaust the buffer faster and provide less working margin.
Art Sorb functions by absorbing moisture when internal RH rises above its set point, and releasing it when RH drops — maintaining a stable microclimate regardless of external conditions. It has over five times the moisture buffering capacity of standard silica gel, responds with minimal hysteresis, and is completely inert in contact with artwork.
The back panel is 3mm Dibond — an aluminium composite panel manufactured by 3A Composites, comprising two 0.3mm aluminium skins bonded to a solid polyethylene core. It is chemically inert, contains no plasticisers or off-gassing compounds, and the aluminium skins provide a near-impermeable vapour barrier across the rear face of the sealed cavity.
The frame was assembled in a climate-controlled environment at approximately 50% RH and let to equalize for days prior to being sealed. The Testo 160 THE sensor monitors internal RH continuously. When the internal RH begins tracking ambient conditions more closely, losing its stability around 50%, the buffer is weakening.
VI. The Research Methodology
To quantify the "Environmental Cost of Memory," we deploy a three-tier observation model with the delta between them, tracked over monsoon seasons and years:
Protected by museum-grade framing composed of Optium Museum Acrylic, conservation mat width calculated using our Golden Ratio Formula (framing guide), additional conservation mat float with Art Sorb 50% sheet, Dibond aluminium composite back panel, sealed with permanently flexible solvent-free alloy tape throughout.
A "Control Postcard"—an unframed print of the same image exposed directly to the local atmosphere as an early-warning system for material degradation.
A dark-stored "Reference Postcard" print in a climate-stabilised environment to measure Delta E (colour shift) over time against both the Subject and the Variable.
VII. Toward a Research Paper
All sensor data is cloud-logged, API-accessible, and owned by Rebel Archive. The intention is perpetual operation for the lifetime of the installation.
The three-tier methodology, fixed location, continuous sensor logging, and documented material stack combine to produce something the existing literature does not have: a measured deterioration profile from an equatorial field station, with full provenance of every variable. It may become one of the longest continuously monitored conservation field records in the tropics. The data spans paper cockling, pigment oxidation, photochemical degradation via cumulative UV dosage and illuminance, and temperature-humidity cycling across monsoon seasons.
Conservation science has well-established standards for controlled environments. It has far less to say about passive microclimate engineering in an external tropical environment — whether a sealed, instrumented frame can substitute for active climate control where humidity regularly exceeds 80% and UV loads are among the highest on earth. That question matters for galleries, collectors, and archives across Southeast Asia, West Africa, and the wider equatorial belt that cannot afford museum-grade infrastructure.
Should the findings prove significant, Rebel Archive intends to publish. It is our attempt to find out, with some rigour, whether beautiful things can be made to last while being enjoyed.
Art Frame Environment
Live Readings from Testo 160 THE
Outdoor Environment
Live DataConditions surrounding the art frame
Regional Climate Drivers
Southeast Asia
Awaiting dataFire detections: NASA FIRMS VIIRS, by country and region, nominal + high confidence, refreshed 6-hourly. FRP = summed fire radiative power · ENSO: NOAA CPC weekly Niño 3.4 · IOD: HadISST DMI, ~1 month lag · Monsoon phase per MSS convention
Conservation Log
Cumulative environmental data since installation
Installation
—
Today (°C)
(mW·h/m²)
(mW·h/m²)
(visible light dose)
(visible light dose)
(1-hour PM2.5 > 55 µg/m³, Central)
(four counters added; load, not exposure)
Installed: 2026-06-20 · UV dose = irradiance (mW/m²) × time (h) · Lux dose = illuminance (lx) × time (h) · Outdoor UV Index: NEA data.gov.sg · Full dataset written to Google Sheets every 30 min (83 columns)
Singapore Climate Comparison
NEA Climate Stations · Historical Data · Live Readings
Temperature
Relative Humidity
UV Radiation
Illuminance
Dew Point
Pollution Standards Index
Historical Origins & Inspiration
Weather forecasting began when the world's mariners first agreed to speak the same technical language, standardising their reports from sea and sky into a single, decipherable script.
It seemed oddly fitting, then, that after a pint in a centuries-old riverside tavern in Salzburg, Austria, one of our number should stumble upon a meteorological pillar, standing there with the quiet assurance of an old friend who has seen too much and forgotten nothing. A legendary place that Faust himself is said to have visited, with the tavern taking the following as their motto: "Art is long, and life is short" (Goethe, Faust I, Vers 558 f.).
They found another, some years later, in Zagreb, Croatia; and these two mute sentries of air and pressure marked their minds in a way no cathedral ever quite does, paying, as they did, unconscious tribute to those who dabbled in and advanced the craft. Much like FitzRoy and his first storm warnings for British sailors in 1860 and, later, the earliest public forecasts, they stood in the in-between: a show of technological prowess to the public and a subtle nod to the heavens that we humans could track her, if not yet tame her.
It was over another pint that the question arose: how, in this open-air furnace of Singapore, with its alternating spells of wet heavy air and dry lite metallic sunshine, could we possibly show a REBEL work and keep it alive? This sealed microclimate frame is our answer. Designed to maintain a stable internal environment while monitoring and logging the conditions within. We compare that with ambient environmental data and have set thresholds to warn us in real time in the event the conditions deteriorate outside of acceptable parameters.
It is quiet a coincidence that this new pillar should end up on a quay in Singapore, tucked inside a bar with a nautical signal (... --- ...), and beside a room called Fitz, under the unobtrusive care of one E. Fitzpatrick.
This is propaganda. Not art.
Satellite Observations
Indonesia, 24 h
complexes
detections
About NOAA-20
Monsoons, Fires, and Singapore
I. The monsoon year
Singapore sits 137 km north of the equator, so there is no summer and winter. The year is shaped instead by two monsoons and the two quieter transitions between them. Temperatures run 24 to 34 °C all year, relative humidity sits between 70 and 95 percent, and about 2,100 mm of rain falls, most of it in short afternoon downpours. What changes through the year is the wind, and with it where the rain, the heat and the smoke come from.
| Season | Months | Wind | Weather at the bar |
|---|---|---|---|
| Northeast monsoon, wet phase | December to January | Steady northeast | Monsoon surges bring long spells of continuous rain and the coolest days of the year |
| Northeast monsoon, dry phase | February to early March | Northeast, brisker | Drier and windier, the driest weeks of the year |
| Inter-monsoon | Late March to May | Light and variable | Hottest period, afternoon thunderstorms, lightning |
| Southwest monsoon | June to September | Southeast to southwest | Drier, pre-dawn Sumatra squalls, and the haze season when the Sumatran peatlands burn |
| Inter-monsoon | October to November | Light and variable | Afternoon thunderstorms, smoke can linger when winds slacken |
The map below showed the then current monsoon phase and pointed its arrows along the seasonal wind, or along the live wind at Singapore when it blew at more than 2 m/s. Phase dates follow the Meteorological Service Singapore convention; the actual onset moves by a few weeks from year to year.
II. Why the fires happen
Much of eastern Sumatra and southern Borneo sits on tropical peat: metres-deep layers of waterlogged, partly decayed vegetation. When drained for agriculture and plantation over recent decades, peat dries and becomes fuel. Fire is a cheap method used in the region to clear land, and in a normal wet year it stays relatively small. In a dry year, especially an El Niño year when the rains fail, fires spread into the peat itself and burn underground for weeks, producing far more smoke per hectare than a forest fire and resisting rain and firefighting alike. Governments across ASEAN, including Singapore and Indonesia, have worked on this together since the 2002 Agreement on Transboundary Haze Pollution.
We are not apolitical. We do have a view, which can be inferred from some of our obsessions. This page, however, is about the art, the science and the education.
III. Why Singapore
Riau's peatlands are about 300 km from the frame, across the Strait of Malacca; South Sumatra is about 500 km. From June to September the southwest monsoon carries air from Sumatra over Singapore, and in the September to October transition the winds slacken and smoke pools. The worst episodes on record, in 1997, 2013, 2015 and 2019, were all dry-season events in El Niño or near-El Niño years, and each reached the Hazardous band. The 2015 episode ran for weeks.
IV. How this connects to the frame and the outdoor cards
The chain runs one way: a warm eastern Pacific (El Niño) weakens the rains over the Maritime Continent; a positive Indian Ocean Dipole (IOD) adds to the drying; dry peat burns; the monsoon sets the general direction of the smoke, with day-to-day swings when winds strengthen. Fires sit downstream of the climate drivers, which is why the map shows ENSO and IOD first and the fire counts last. ENSO stands for El Niño–Southern Oscillation, the umbrella name for the whole Pacific cycle, of which El Niño and La Niña are the two extremes; the IOD is the Indian Ocean's equivalent, the temperature difference between its western and eastern halves. The fires do feed back at a global scale, through carbon released from peat, but for this project they are the last link before the air arrives at the Daedalus.
V. What we are watching for
A very strong El Niño peaking in December, as forecast for 2026, means the coming dry months carry above-normal fire risk. The map, the fire counts and the wind arrow exist so that, when the haze comes, the record shows what reached the frame.
PSI and AQI
Every air quality index on this page starts from the same measurements: Singapore's national monitoring network, published by the National Environment Agency (NEA). The indices differ in how they summarise those measurements over time and in where they draw their category lines.
I. The averaging window
The headline 24-hour PSI is a rolling average of the past 24 hours across six pollutants. It is designed to describe a day's exposure, which is the right basis for planning decisions made in advance. Its cost is that a change in the air takes many hours to show fully in the number.
The US EPA AQI shown by most phone apps (including the AQICN marker on this page) uses a method called NowCast, which weights the most recent one to three hours heavily when conditions are changing. It is designed to describe the air right now. Its cost is that it moves quickly, up and down, and can overstate a brief spike.
To give the immediate picture as well, NEA also publishes a 1-hour PM2.5 concentration in micrograms per cubic metre. This page carries both windows on their own cards: PM2.5 over 24 hours and PM2.5 over the last hour, each with its NEA band, so the slow and the fast concentration sit side by side with the slow index (PSI) and the fast index (AQI). The 24-hour values are logged every 30 minutes.
II. The category lines
Both scales run from 0 to 500. They divide the middle differently.
| Range | Singapore PSI | US EPA AQI |
|---|---|---|
| 0 to 50 | Good | Good |
| 51 to 100 | Moderate | Moderate |
| 101 to 150 | Unhealthy | Unhealthy for Sensitive Groups |
| 151 to 200 | Unhealthy | |
| 201 to 300 | Very Unhealthy | Very Unhealthy |
| Above 300 | Hazardous | Hazardous |
The scales also convert PM2.5 concentration to index points with different breakpoints, so the same air can map to a different number on each, not only a different label. Because most consumer apps convert Singapore's hourly PM2.5 into the US scale, a day reported as "Moderate" on the 24-hour PSI can appear as "Unhealthy for Sensitive Groups" on a phone. That is the definitions at work.
III. How Singapore uses each measure
Public guidance uses both, for different purposes. The 24-hour PSI forecast supports decisions made ahead of time, such as whether outdoor programmes go ahead the next day. The 1-hour PM2.5 reading supports adjustments on the day itself. NEA groups the 1-hour reading into bands: Band I (0 to 55) normal, Band II (56 to 150) elevated, Band III (151 to 250) high, Band IV (above 250) very high, each with its own activity guidance.
Under MOH health advisories, the "Unhealthy" PSI band (101 to 200) is where the general population is advised to reduce prolonged or strenuous outdoor exertion, and where schools scale back outdoor physical activity. Full school closure is reserved for the Hazardous band, and classrooms are equipped with air filtration for that reason.
IV. What this page does with them
The PSI chart shades the Singapore bands and plots the 24-hour PSI as a filled marker. The AQICN value is drawn as a hollow marker, on purpose, because it sits on a different scale than the bands behind it. The two PM2.5 cards show the underlying concentration in both windows, which is the number comparable across countries and against the WHO and US EPA guidelines. All of them are logged, so the research record keeps the fast signal, the daily signal and the raw measurement side by side.
V. Why?
This page exists because of an art print, sealed inside a microclimate frame to protect it as best we can from every environmental element. Air pollutants are one of those elements. The Control Postcard hangs unprotected and takes the full brunt of them.
Fine particulates are a serious health matter, and one we feel is poorly understood; this page, though, is about their effect on art. Smoke from peat fires carries sulphate, nitrate and organic aerosols that settle on paper and pigment; the acidic fraction is a known driver of yellowing and embrittlement in cellulose, and soot deposits are difficult to remove from a matte surface without damage. For the sealed frame, a haze episode is a test of the enclosure, and only time will tell whether the seal is doing its work when analyzed against the two print controls.
VI. Reading the frame's own sensor against these
The "PM2.5 at the frame" card is a concentration that measured every minute by an optical sensor near the Control Postcard. It is the same quantity as NEA's 1-hour figure and should be read against that card first: the gap between them is the sensor's humidity over-read, which we are calibrating over its first weeks in place and will correct on the page once the fit is known.
It is not the same quantity as PSI or AQI. Those are indices, computed by passing a 24-hour (PSI) or NowCast-weighted (AQI) concentration through a published lookup table. The card's note shows what the frame's reading would score on both tables if the air stayed at that level for the full averaging window, as a way of placing it on the same scale, labelled "if sustained" because a one-minute reading is not a daily average. A high frame reading with a lower PSI usually means the haze arrived recently; the reverse means it is clearing.
For the artwork, the concentration matters more than either index. Dose to the exposed print is micrograms per cubic metre multiplied by hours, and that is what the log keeps.
All of this partially explains the slight obsession. If you made it this far, you now know the differences between PSI, AQI and the local guidelines. Together with the fire detections and the wind on the map above, they let a future reader of this data say, for any change found in the prints, what the air was doing that day, where the smoke came from, and how long it stayed.