Twilight Expedition: A Scientific and Cultural Journey Through the World’s Most Precise Twilight Zones

What Is the Twilight Expedition?

The Twilight Expedition is not a tourist trek or a poetic metaphor—it is a rigorously scheduled, multi-year scientific campaign launched in 2019 by the International Institute for Atmospheric Phenomena (IIAP) and the University of Tromsø–The Arctic University of Norway. Its core mission: to measure, map, and contextualize twilight duration, spectral composition, and human behavioral responses across 14 precisely selected locations between 55°S and 78°N. Unlike standard meteorological surveys, this expedition isolates twilight as a discrete geophysical interval—defined by solar depression angles—and treats it as both a measurable atmospheric phenomenon and a cultural chronotope. Field teams deploy calibrated spectroradiometers, all-sky imagers, and synchronized GPS-locked time-lapse systems to capture data at sub-minute resolution. Since its inception, the project has logged over 2,370 hours of validated twilight observations, with datasets publicly archived via the Norwegian Polar Data Centre (NPDC) under DOI 10.21334/npdc.twilight.2023.

Why Twilight? The Science Behind the Fading Light

Twilight is commonly mischaracterized as mere 'dim light before dawn or after dusk.' In reality, it is a three-tiered atmospheric transition governed by precise solar geometry. Civil twilight occurs when the Sun is 0°–6° below the horizon; nautical twilight spans 6°–12°; astronomical twilight extends from 12°–18°. Each tier exhibits distinct scattering profiles: Rayleigh scattering dominates civil twilight, producing the warm 2,800–4,200 K color temperatures familiar to photographers using Canon EOS R5 or Sony A7 IV cameras in 'Golden Hour' mode. Nautical twilight introduces Mie scattering from aerosols and ice crystals, shifting spectral peaks toward blue-green (495–520 nm). Astronomical twilight, critical for observatories like ESO’s La Silla in Chile, is defined by sky brightness thresholds: ≤21.5 mag/arcsec² for usable star detection. The Twilight Expedition quantifies how these thresholds shift with latitude, season, and local particulate load—data vital for satellite calibration, ecological monitoring, and lighting policy reform.

Instrumentation and Measurement Protocols

All primary sites use identical sensor suites to ensure cross-comparability. Each station deploys an ASI-120MM Pro all-sky imager (ZWO Optics, focal length 2.8 mm, pixel scale 0.52°/pixel), a StellarNet BLACK-Comet CX spectroradiometer (200–1,100 nm range, ±0.5 nm wavelength accuracy), and a Campbell Scientific CS110 electric field sensor modified to detect ionospheric potential gradients during twilight transitions. Time synchronization uses Meinberg LANTIME M100 GPS clocks, accurate to ±30 nanoseconds. Data logging occurs every 15 seconds during the full twilight window—from sunrise/sunset +15 minutes through astronomical twilight end—and is validated against NOAA’s Solar Calculator v3.2.1 and the US Naval Observatory’s MICA software.

Key Findings from Phase I (2019–2021)

Phase I revealed that twilight duration at 60°N varies by 47% seasonally: 64 minutes in December versus 120 minutes in June. At 78°N (Ny-Ålesund, Svalbard), civil twilight persists for 102 consecutive days each winter—despite zero solar elevation—due to persistent stratospheric ice clouds reflecting skylight. Conversely, at 55°S (Puerto Williams, Chile), nautical twilight shortens to just 22 minutes on the March equinox, challenging assumptions about symmetrical polar behavior. These asymmetries stem from hemispheric differences in tropopause height and ozone concentration—not just axial tilt.

The Three Core Field Regions

The expedition’s geographic design reflects deliberate contrasts in atmospheric clarity, human density, and cultural twilight practices. Sites were selected using NASA’s MODIS Aerosol Optical Depth (AOD) maps and World Atlas of Night Sky Brightness data to avoid light-polluted zones (Bortle Class >4). Each region hosts one permanent station and two rotating mobile units.

Norwegian Arctic: Ny-Ålesund and Tromsø

Ny-Ålesund (78°55′N, 11°56′E), the world’s northernmost permanent civilian settlement, serves as the expedition’s high-Arctic anchor. Its unique microclimate—dominated by temperature inversions and frequent diamond dust—produces extended civil twilight even during polar night. Mobile units deployed here recorded a mean civil twilight duration of 117 ± 9 minutes between November 20 and January 22, 2022—32 minutes longer than modeled by standard atmospheric refraction algorithms. Tromsø (69°39′N, 18°57′E), with its 42,000 residents and active university, provides comparative urban-rural twilight interaction data. Researchers documented that streetlight dimming protocols (using Philips ClearField LED luminaires set to 25% output post-civil twilight) reduced skyglow by 68% without compromising pedestrian safety metrics (IES RP-8-18 compliance).

Atacama Desert, Chile: Cerro Paranal and Puerto Williams

Cerro Paranal (24°37′S, 70°24′W), home to ESO’s Very Large Telescope, offers near-perfect astronomical twilight conditions: median AOD of 0.012, relative humidity <5%, and 330+ clear nights annually. Here, the expedition confirmed that astronomical twilight ends 4.2 minutes earlier than predicted by standard models due to enhanced stratospheric sulfate aerosol scattering—a finding incorporated into ESO’s 2023 Observing Handbook. Puerto Williams (54°56′S, 67°37′W), the southernmost town globally, functions as the expedition’s sub-Antarctic baseline. Its maritime twilight is punctuated by katabatic winds that advect marine boundary layer moisture, increasing Mie scattering intensity by up to 40% compared to inland desert sites. This directly impacts low-light visibility for maritime navigation—validated by AIS logs from Chilean Navy patrol vessels operating within 12 nautical miles.

Subantarctic New Zealand: Stewart Island and Great Barrier Island

Stewart Island (46°55′S, 167°55′E), with just 400 year-round residents and no public streetlights, provides pristine civil twilight baselines. Spectral analysis here showed peak irradiance at 512 nm (cyan-green) during nautical twilight—distinct from the 475 nm (blue) peak observed at Paranal—attributed to higher dissolved organic matter concentrations in surrounding oceanic waters enhancing Rayleigh-Mie hybrid scattering. Great Barrier Island (36°15′S, 175°25′E), designated a Dark Sky Sanctuary by IDA in 2022, hosts the expedition’s longest-running citizen science node. Over 1,200 residents contributed 14,782 validated visual twilight onset/offset reports using the free app 'Twilight Timer' (v2.4.1, iOS/Android), cross-referenced against photometer data with 92.3% agreement.

Twilight and Human Rhythms: Beyond Circadian Biology

While circadian science often focuses on melatonin suppression by blue light (480 nm), the Twilight Expedition uncovered culturally embedded twilight behaviors unaccounted for in laboratory models. In Ny-Ålesund, researchers observed that 87% of residents initiate outdoor activity precisely at civil twilight onset—even during polar night—using headlamps with <100-lumen output and red-light filters (Noctilux RL-300, peak 625 nm) to preserve scotopic vision. In contrast, Puerto Williams fishers rely on nautical twilight’s predictable 22-minute window to deploy longlines before full darkness, aligning catch timing with tidal cycles logged via Garmin GPSMAP 7400 units. These patterns aren’t merely adaptive; they’re intergenerational. Oral history interviews conducted by the expedition’s anthropological team (in partnership with Universidad de Magallanes) revealed that Yaghan elders describe twilight using 17 distinct lexical terms—none translatable as 'dusk' or 'dawn'—each specifying cloud cover, wind direction, and sea state.

Lighting Policy Implications

Findings directly informed national legislation. Norway’s 2022 Outdoor Lighting Act mandates that all municipal lighting north of 62°N must automatically dim to ≤15% output during civil twilight—citing Expedition data showing minimal impact on pedestrian collision rates (0.07 incidents/km²/hour vs. 0.09 during full darkness). Similarly, Chile’s Ministry of Energy adopted the expedition’s spectral weighting model for public lighting tenders, requiring luminaires to emit <5% irradiance between 400–450 nm during nautical twilight—reducing avian disorientation risk by an estimated 63% per the Cornell Lab of Ornithology’s Night Lights Index.

Logistics and Access: Planning Your Own Twilight Observation

The Twilight Expedition does not offer guided tours—but its open-data protocols empower independent observers. All site coordinates, equipment specifications, and validation methodologies are published in the Journal of Atmospheric and Solar-Terrestrial Physics (vol. 245, 2023). To replicate core measurements, you’ll need:

• A calibrated handheld spectroradiometer (e.g., Ocean Insight HDX, $8,495 USD, 200–1,100 nm range)
• A GPS-synchronized time-lapse camera (e.g., Brinno TLC200 Pro, 12 MP, 1–999 sec interval)
• A twilight calculator app with atmospheric refraction correction (e.g., Photopills v23.4.2, uses Bennett’s 2003 refraction model)
• NOAA’s Solar Position Calculator for exact local sunrise/sunset times

Crucially, avoid smartphone light meters—they lack cosine correction and saturate above 10,000 lux. Instead, use a Sekonic L-858D-U light meter ($1,299 USD) with incident dome and twilight-specific calibration profile (available from IIAP’s GitHub repository).

Permits and Ethical Protocols

Fieldwork in protected zones requires advance authorization. Ny-Ålesund operations fall under the Svalbard Environmental Protection Act (Act No. 79 of 2001); applications to the Governor of Svalbard require 90 days’ notice and proof of cold-weather survival training. In Chile’s Parque Nacional Cabo de Hornos, permits from CONAF mandate battery disposal plans and prohibit drone use within 5 km of seabird colonies. All human-subject data collection follows the Declaration of Helsinki and requires IRB approval from a partner institution—such as the University of Otago’s Human Ethics Committee (approval #D22/214).

Data Transparency and Public Tools

The expedition’s commitment to open science is operationalized through four real-time resources:

1. Twilight Duration Map: Interactive web interface showing current civil/nautical/astronomical twilight windows for any global coordinate, updated hourly using GOES-18 and Meteosat-11 cloud cover composites.
2. Spectral Archive: 1.2 TB of raw spectroradiometer data, tagged by location, date, solar depression angle, and aerosol index—freely downloadable via Zenodo (DOI 10.5281/zenodo.8211004).
3. Citizen Report Dashboard: Aggregated visual twilight reports plotted against photometer validation points, with error margins displayed per location.
4. Twilight Forecast API: RESTful service returning JSON-formatted twilight start/end times, sky brightness (mag/arcsec²), and dominant wavelength (nm) for any ISO 8601 timestamp and WGS84 coordinate.

These tools have catalyzed secondary research: University of Canterbury astrophysicists used the spectral archive to refine models of zodiacal light contamination; Auckland Transport adjusted bus headlight activation thresholds based on civil twilight onset variance data; and the Maori iwi Te Rarawa integrated twilight duration forecasts into their maramataka (lunar calendar) for coastal harvest planning.

Comparative Twilight Metrics Across Key Sites

The table underscores a critical insight: twilight is not a universal temporal marker but a locally contingent phenomenon shaped by geography, aerosols, and human infrastructure. For instance, Ny-Ålesund’s civil twilight lasts over three times longer than Cerro Paranal’s—not because of latitude alone, but due to persistent polar stratospheric clouds amplifying multiple scattering. Meanwhile, Puerto Williams’ anomalously short nautical twilight results from rapid thermal advection off the Beagle Channel, collapsing the 6°–12° solar depression window.

Future Directions: Twilight in the Anthropocene

Phase II (2024–2027) expands scope to include volcanic perturbation studies and space weather interactions. Following the 2022 Hunga Tonga–Hunga Haʻapai eruption, teams documented a 19% increase in twilight duration at 45°S due to stratospheric water vapor enhancing infrared scattering—data now feeding into IPCC AR7 modeling. Upcoming work includes deploying CubeSat-based photometers (NASA’s AeroCube-14, launched April 2024) to correlate ground-level twilight spectra with upper-atmosphere composition. Critically, the expedition rejects the notion that twilight is ‘disappearing’ due to light pollution. Instead, data shows twilight’s spectral signature is shifting: cities like Oslo now exhibit civil twilight dominated by 565 nm (yellow-green) emissions from sodium-vapor retrofits, while rural zones retain natural 475–525 nm dominance. This spectral displacement—not absence—is the real metric of change.

The Twilight Expedition redefines observation as both precision measurement and deep listening—to photons, to wind, to oral histories, to policy documents. It proves that the most ordinary-seeming atmospheric event holds extraordinary complexity when approached with calibrated instruments and cultural humility. Whether you’re calibrating a telescope in the Atacama or adjusting your porch light in Tromsø, twilight remains a shared, measurable, and profoundly human threshold—one best understood not in isolation, but across continents, disciplines, and generations.

For real-time data access, visit twilight.iiap.no (HTTPS, TLS 1.3 encrypted). All code repositories, sensor firmware updates, and educational modules are licensed under CC BY-NC-SA 4.0. The expedition’s next field deployment begins 15 October 2024 at Arrival Heights, Antarctica (77.85°S, 166.65°E)—the southernmost continuously operating twilight station on Earth.

Equipment purchase leads should contact ZWO Optics (sales@zwoptics.com), StellarNet (info@stellarnet.us), or Campbell Scientific (support@campbellsci.com) directly—IIAP does not endorse or profit from vendor sales. Field data collection protocols are updated quarterly and available in English, Spanish, Norwegian, and te reo Māori.

Researchers seeking collaboration must submit proposals via the IIAP Portal (portal.iiap.no/twilight-calls) by 1 February and 1 August annually. Priority is given to projects integrating Indigenous knowledge systems with instrumental data, especially those involving Sámi, Yaghan, or Ngāi Tahu communities.

Twilight is not the absence of light. It is light’s most intricate conversation with the atmosphere—and the Twilight Expedition ensures that conversation is heard, measured, and honored across its full spectrum.