What SPHEREx found and how it works
SPHEREx, short for Spectro-Photometer for the History of the Universe, Epoch of Reionization, and Ices Explorer, launched in March 2025 on a two-year mission to survey the entire sky in near-infrared light. The instrument captures data across 102 spectral bands spanning roughly 0.75 to 5.0 micrometers. That wavelength window is critical because frozen molecules coating interstellar dust grains each absorb infrared light at distinct frequencies, effectively leaving chemical fingerprints that SPHEREx can read. A technical paper posted to arXiv in spring 2026 details the new spectral maps. The team identified absorption features at 3 micrometers for water ice, 4.27 micrometers for carbon dioxide ice, and 4.67 micrometers for carbon monoxide ice, alongside a 3.28-micrometer emission feature tied to polycyclic aromatic hydrocarbons (PAHs), complex carbon-rich molecules found throughout the galaxy. The sky areas covered include the Cygnus X star-forming complex and the North American Nebula, both dense regions where new stars are actively assembling from collapsing gas and dust. Previous infrared observatories like Spitzer and Japan’s Akari satellite detected ices in molecular clouds, but only in narrow, targeted pointings. SPHEREx’s all-sky design and broad spectral coverage let it trace continuous structures of frozen material threaded through entire clouds rather than sampling isolated patches. By measuring how strongly ice absorbs background starlight at specific wavelengths, the telescope estimates relative column densities of each frozen species along every line of sight, building a large-scale picture of where the coldest, densest pockets of material sit inside these clouds. An official NASA image product, published in the agency’s Science Photojournal, renders water ice as bright blue structures overlaying dark lanes of interstellar dust throughout Cygnus X. The visualization reveals filaments and knots of ice-rich gas that may mark the future birthplaces of clusters of sunlike stars.Why “interstellar glaciers” matter for planet formation
The ice detected in Cygnus X and the North American Nebula sits in the same type of dense molecular cloud environment where our own Sun likely formed roughly 4.6 billion years ago. Understanding how much frozen water exists in these clouds, and where it concentrates, bears directly on a question planetary scientists have debated for decades: how did Earth get its water? One leading hypothesis holds that icy grains in the Sun’s parent cloud survived the collapse into a protoplanetary disk and were later delivered to the inner solar system by comets and asteroids. If SPHEREx’s maps show that water ice is abundant and widespread in star-forming regions, that strengthens the case that frozen water is a standard ingredient in planet-building, not a lucky accident. The relationship between ice and the PAH emissions detected in the same fields adds another layer. PAHs typically trace regions exposed to ultraviolet radiation from young, massive stars, while thick mantles of ice prefer shielded, colder environments deeper inside molecular clouds. Whether PAH-rich zones inversely correlate with ice concentrations, as some theoretical models predict, has not yet been confirmed in the SPHEREx data. If ultraviolet exposure does strip ice from dust grains in certain environments, that would influence how much solid material survives in the inner regions of emerging planetary systems, right where rocky planets tend to form.What scientists still need to pin down
These maps represent early-release data from SPHEREx’s initial sky passes. The mission is designed to survey the full sky at least four times over its two-year primary mission, and each additional pass will improve signal-to-noise ratios and reduce calibration uncertainties. Early observations often rely on provisional calibrations; systematic effects such as scattered light, detector nonlinearity, or imperfect background subtraction are better characterized only after months of repeated measurements. SPHEREx’s processing pipeline converts raw observations into calibrated spectral images distributed through NASA’s Infrared Science Archive. A separate arXiv paper on the data system describes a “quick release” approach intended to put preliminary products in scientists’ hands relatively soon after acquisition. That means the current ice maps may be subject to reprocessing that subtly shifts measured absorption depths or the apparent sharpness of ice-rich structures as the pipeline matures. The spatial resolution of SPHEREx, while sufficient for mapping large-scale ice distribution, cannot match the fine detail that the James Webb Space Telescope achieves on individual star-forming cores. Cross-matching SPHEREx’s broad, 102-band spectra with JWST’s detailed views of selected filaments will let researchers test whether the large-scale ice patterns hold up under closer inspection. Ground-based submillimeter observatories like ALMA can add complementary data on the gas and dust that accompany the ice.A new atlas of the galaxy’s frozen chemistry
For all the caveats that come with early data, the scale of what SPHEREx has produced is genuinely new. No previous space telescope has mapped frozen water, carbon dioxide, and carbon monoxide simultaneously across hundreds of light-years of active star-forming territory. The result is less a snapshot and more the first draft of an atlas, one that connects large-scale galactic structure to the small-scale chemistry of the clouds where stars and planets are born. As SPHEREx continues scanning the sky through its planned mission lifetime into 2027, each successive pass will sharpen that atlas. The “interstellar glaciers” mapped so far in Cygnus X and the North American Nebula are almost certainly just the beginning. The same frozen chemistry should appear wherever dense molecular clouds block and absorb starlight across the Milky Way, and SPHEREx is built to find it all. More from Morning Overview*This article was researched with the help of AI, with human editors creating the final content.
