The Deep Space Climate Observatory (DSCOVR) was launched in February 2015 into a Lagrange 1 (L1) orbit, approximately 1.5 million kilometers from Earth in the direction of the Sun. In many regards, the DSCOVR is a unique mission: for the first time, well-calibrated multi-spectral Earth observations are being delivered from an L1 point. This unique location allows viewing the entire sunlit part of our planet multiple times a day with one to 2 hours cadence, significantly expanding the space/time scales of Earth monitoring. The moderately high observation rate from DSCOVR results in cloud-free views of nearly the entire Earth’s land surface and global ocean, with significantly higher frequency than that available from the operational polar orbiters and at a near-global scale inaccessible from geostationary sensors.

In addition to providing continuous solar wind measurements for accurate space weather forecasting, DSCOVR operates two Earth science instruments: the Earth Polychromatic Imaging Camera (EPIC) and the NIST Advanced Radiometer (NISTAR). NISTAR measures the absolute irradiance integrated over the entire sunlit face of the Earth in four broadband channels every minute covering visible and IR wavelengths. EPIC has a 2048 × 2048-pixel CCD with sensitivity to UV, visible, and near IR (NIR) wavelengths. The filter wheel contains 10 narrow-band filters covering the range from 317 nm to 780 nm. The spatial resolution is about 10 km at the nadir. The Earth-observing geometry of the EPIC instrument captures a backscattering view geometry with a small-range scattering angle varying between 168° and 178°.

Due to technical issues with DSCOVR spacecraft, there were two periods during the mission time with data gaps from EPIC and NISTAR: June 27, 2019 to March 2, 2020 and July 15 to October 9, 2025.

The unique near-backscatter view geometry of EPIC led to the creation of several new Earth science products. One of them is the diurnal course of the “sunlit leaf area index” (SLAI) (Yang et al., 2017). It characterizes the area of green leaves at a given time, intercepting direct sunlight, and depends on canopy structural organization. Another one involves the specular reflection from horizontally oriented ice crystals in clouds (Marshak et al., 2017). The first operational glint product was released in May 2021 (https://epic.gsfc.nasa.gov/science/products/glint). Collected EPIC data are processed on a semi-operational basis providing a suite of atmospheric, land, and ocean data products. Many of the processing algorithms have been described in the earlier Research Topic Frontiers in Remote Sensing published 5 years ago, “DSCOVR EPIC/NISTAR: 5 years of observing Earth from the first Lagrangian point,” which featured 23 papers providing a holistic view of the Earth science gained from DSCOVR.

The current Research Topic, titled “Earth observations from the Deep Space: 10 years of the DSCOVR mission”, is a continuation of the previous Research Topic. It features 20 papers focused on science data analysis and various applications and covers the progress achieved in sensor calibration and processing algorithms.

Below is a brief overview of the scope of the Research Topic. Cede et al. discusses the radiometric stability of EPIC and NISTAR. The study found no drift over the entire mission in the EPIC 551 nm channel and NISTAR photodiode channel (200–1,100 nm). Changes in EPIC calibration were very small, ranging from a maximum degradation of 3% over 10 years in the short UV channels to smaller to negligible changes for other channels. Such remarkable performance is largely attributed to the L1 orbit located beyond the Earth’s radiation belt and the thermal stability provided due to the Sun’s positioning on the same shielded side of the spacecraft. This unique perspective highlights the value of the L1 orbit as a deep-space outpost for continued long-term Earth monitoring. Sutton et al. discusses the EPIC data processing pipeline including the list of science products with links to the long-term archive. Lacis et al. uses NISTAR photodiode measurements to confirm EPIC-derived data on the diurnal variability of Earth’s reflected radiation.

The tropospheric and total column ozone records from EPIC are discussed by Ziemke et al. and Herman et al. These papers summarize 10 years of tropospheric ozone observations and validate EPIC total ozone using both ground-based measurements and a comparison with similar products from OMPS, OMI, and TEMPO.

The aerosol research from EPIC is represented by four studies. Torres et al. discusses the decadal aerosol record and several significant wildfire and dust storm events using the near-UV EPIC aerosol product. Lyapustin et al. describes the new version 3 MAIAC EPIC algorithm for joint retrieval of aerosol optical depth, spectral absorption, and layer height (ALH). An extensive full mission validation analysis shows a high accuracy for both MAIAC spectral single scattering albedo comparable to AERONET uncertainty and the retrieved ALH with global rmse∼1 km as validated by CALIOP CALIPSO data. Using v3 MAIAC aerosol products, Choi et al. discusses the climatology and variability of smoke aerosols over North America over the past decade while Go et al. provides an in-depth analysis of dust aerosols over the globe.

Five papers in this Research Topic are dedicated to the analysis of cloud properties: Yang et al. summarizes what we learned about clouds from 10 years of observations of global daytime measurements, while Delgado-Bonal et al. analyze the effect of spatial resolution on derived fraction, height, and optical thickness of clouds. Two more papers (Várnai et al. and Kostinski et al.) derive statistics and constrain ice crystals’ orientation, respectively, using glints in atmospheric ice clouds detected by EPIC. Russell et al. uses EPIC observations to critically examine the performance of climate models at daily scales, in contrast to traditional analysis based on monthly-mean values where the climate variability at shorter time scales is lost.

Pisek et al. uses DSCOVR EPIC data over Australian TERN vegetation sites to empirically confirm a theoretically predicted relationship between the Directional Area Scattering Factor (DASF) and the clumping foliage index.

Several papers describe the use of DSCOVR/EPIC observations for planetary studies. Herman and Blank estimate Jupiter albedo and limb darkening. Jian et al. reports progress in exoplanet research, treating aggregated images of Earth from EPIC as a distant spatially unresolved world with characteristic temporal and spectral features. Analysis of different temporal scales of the Earth’s cumulative reflectance and radiative flux from EPIC is given by Wen et al. Constructed high resolution multispectral maps of both sides of the Moon in 10 EPIC channels helped Gorkavyi et al. to reveal mineralogical differences between the two sides of the Moon. The paper shows that the Moon images obtained by EPIC have a significant scientific value. The paper by Blank et al. discusses a deep space mirage (Gaia’s crown).

Beyond the individual advances reported in this Research Topic, several cross-cutting themes emerge. The sustained radiometric stability of EPIC and NISTAR demonstrates that deep-space platforms can support long-duration climate-relevant records. The continuous full-disk perspective has elevated the study of diurnal changes in clouds, surface processes, and atmospheric composition to a truly global scale, enabling more rigorous evaluation of global climate and Earth system models and clearer dynamical interpretation. In addition, co-located measurements of aerosols, cloud fields, and Earth’s reflected energy from a single vantage point provide stronger constraints on their mutual interactions than multi-platform approaches. Together, these developments define L1 observations as a coherent and complementary component of the modern Earth observation system.

Taken together, the contributions to this Research Topic mark a milestone in the maturation of deep-space Earth observation. Over the past decade, DSCOVR has demonstrated that viewing Earth from L1 offers a systems-level perspective that integrates variability across processes and time scales within a single, globally consistent framework. Looking ahead, the experience gained from this mission suggests that sustained L1 observations could play an expanded role in future climate observing architectures not only as independent reference records but also as integrative platforms linking polar and geostationary measurements. As the global observation system evolves, deep-space perspectives may become increasingly central to how we monitor and understand a changing planet.