SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2019 | Register Today

SPIE Defense + Commercial Sensing 2019 | Call for Papers

2019 SPIE Optics + Photonics | Call for Papers



Print PageEmail PageView PDF

Remote Sensing

Satellites capture first glimpse of ‘milky-sea’ glowing-water phenomenon

Rare glowing-water effects known as ‘milky seas’ are generally attributed to bacterial bioluminescence, but their causes are still a mystery. Low-light satellite measurements have now provided the first overview of a milky sea, observed off the Somali coast and spanning ~15,000km2. The findings offer new hope for learning about these extraordinary displays.
7 February 2006, SPIE Newsroom. DOI: 10.1117/2.1200601.0093

‘Milky seas’ are rare nocturnal events where the ocean gives off a radiant glow that is reminiscent of a snow field and bright enough to read by. Witnessed over the centuries by mariners crossing the Indian Ocean,1 milky seas have been associated more with legends of the sea than scientific knowledge.

More recently they have been attributed to bacterial bioluminescence2 but their cause, distribution, and role remain mysteries. In particular, the environmental circumstances supporting the seemingly unrealistic bacterial populations needed to produce milky seas have been a topic of debate in the microbial ecology community for decades. Some evidence points to bacterial colonization upon surface slicks of organic matter,3 but observations of milky seas under higher wind conditions (where slicks would break apart) challenge this hypothesis.

One point of consensus is the fundamental importance of additional in situ observations to help understand the cause of these elusive macro-scale processes. The problem has always been finding a practical way to locate them. Satellite sensors represent a previously unexplored tool for this study.4

Perhaps surprisingly, the night-time hours feature a rich diversity of visible light, both natural (moonlight reflection, lightning, fires, and aurora) and artificial (cities).5 The Defense Meteorological Satellite Program6 (DMSP) series of military weather satellites carries the operational linescan system (OLS). This instrument is capable of detecting very weak light sources (roughly one million times fainter than the solar disk). Despite this sensitivity, the OLS cannot detect the more common bioluminescence events associated with disturbed-water (breaking waves, ship wakes) due to their, typically small, extent. However, the steady, widespread glow of milky seas makes them a better candidate for OLS detection.

Correlating with ship reports

We searched a collection of ship reports for milky sea sightings and found an encounter by the British merchant vessel S. S. Lima in 1995 off the Somali coast that provided sufficient detail to pursue. An archive of OLS data collected since 1992 is available at the National Geophysical Data Center. We used this to match the ship observations to any discernable coherent feature in the OLS night-time imagery (after pre-processing to remove instrument noise). As the OLS signal-to-noise characteristics vary greatly across its swath, some luck was needed to capture the low-light feature in the right place and at the right time.

An OLS pass collected at around 30min into the S. S. Lima encounter revealed a faint anomaly in a patch of water east of Somalia. The data were digitally processed by subtracting the mean noise, applying a ∼10×10 km (3×3 pixel) coherency filter (requiring two-thirds of the data to exceed a noise-floor threshold), and running a ∼20 km (6 pixel) boxcar smoothing process. Ship positions for entry/exit of the glowing waters were overlaid, revealing a close match to the OLS observations (see Figure 1). The bright feature persisted over three nights, displaying a rotational evolution consistent with local sea-surface-current analyses.

Figure 1. (A) Raw digital data for 25 January 1995 at 1836 GMT, collected by the Defense Meterorological Satellite Program's operational linescan system. (B) Noise-filtered image with the reported entry (a) and exit (b) coordinates of the British merchant ship S. S. Lima overlaid. Adapted from Miller et al. 2005.
Calculating potential bacterial population

Under the assumption that bacteria were indeed responsible for milky seas, we estimated the minimum population that would be detectable by the OLS sensor. Lab cultures of bioluminescent bacteria with emission properties that are similar to commonly-found free-living species were grown. Their spectrum and per-cell photon emission were recorded. Given the OLS minimum detectable power per unit area (adjusted to account for atmospheric attenuation and partial overlap between the OLS response and the bacterial spectra), the total equivalent bacteria per unit area was computed. This allowed for extrapolation to the total population based on the glowing surface area that was observed by the satellite.

Our conservative estimate of 4×1022 cells participating in this event is roughly the same as the estimated total background free-living bacteria present in the upper 200m of all the oceans of the world.7

Satellite remote sensing has provided our first objective view of the scale, structure, and—in some respects—integrity of a maritime legend. Indeed, the space perspective has enabled an ‘infusoria’ population estimate dismissed as impossible by the knowledgeable Professor Aronnax, protagonist of the Jules Verne's classic novel 20,000 Leagues Under the Sea. 8 More importantly, this capability instills hope within a research community hindered by the elusive nature of milky seas.

Information available from the satellite observations is inherently limited so additional details on milky seas must await in situ observations by research vessels properly equipped for depth-resolved chemistry, organics, and radiometry. The next step in our research is to follow several new leads that have emerged since the announcement of our findings. Our focus will then shift towards the National Polar-orbiting Operational Environmental Satellite System (NPOESS), which is due to come into operation in around 2010. This system is scheduled to carry low-light sensors, which are currently being examined to determine their potential for bioluminescence detection. These could hold promise for future observations of milky seas and possibly the first satellite-directed ship excursions into the surreal.

Steven Miller 
Naval Research Laboratory, Monterey, CA
Dr. Steven Miller is a satellite applications meteorologist at the Naval Research Laboratory in Monterey. He received his M.S. (1997) and Ph.D. (2000) degrees in atmospheric science from Colorado State University. His current interests center around design, development, and implementation of value-added environmental characterization algorithms based on near real-time multispectral/multisensor satellite observations.
Steven Haddock
Monterey Bay Aquarium Research Institute, Moss Landing, CA
Dr. Steven Haddock is a biologist at the Monterey Bay\break Aquarium Research Institute in Moss Landing, CA. After undergraduate studies in engineering, he received his Ph.D. from the University of California, Santa Barbara. He studies bioluminescence, fluorescence and deep-sea gelatinous zooplankton observations.