Moon observations from the Lagrange point L1 by the EPIC/DSCOVR spectrometer

DSCOVR/EPIC, located at the Sun-Earth Lagrange point (L1) around 1.5 million kilometers away from Earth, can capture images of the near and far sides of the Moon in the multiple UV-VIS-NIR wavelengths. These observations were previously used only for calibration purposes. In this study, for the first time, images of the Moon taken by EPIC are treated as scientific data with a unique set of characteristics: 1. They were acquired under full-disk illumination of the Moon. 2. They were taken in 10 narrow wavelength bands—from the ultraviolet (317 nm) to the near-infrared (780 nm). 3. At each wavelength the entire lunar disk is imaged simultaneously. 4. The images can be oversampled to reduce noise levels and increase spatial resolution. These features of the lunar images allow the creation of high-quality maps of the far and near sides of the Moon in 10 quasi-monochromatic wavelength channels. These maps will serve as a reference for comparison with data from other satellites in lunar orbit. The study of multispectral images of the Moon presented in this paper reveals a significant mineralogical difference between the farside and the nearside of the Moon. We interpret the studied spectral features of the Moon as indicating an increased concentration of ilmenite (a titanium-iron oxide mineral, FeTiO₃) on the nearside of the Moon, particularly in the Sea of Tranquility.

1 Introduction

Multispectral images of the Moon contain valuable information about its mineralogical and chemical composition. Quasi-monochromatic images that include the ultraviolet (UV) and near infrared (NIR) wavelengths are particularly important, because observations of the Moon in these narrow spectral bands with ground-based telescopes are difficult. Obtaining such images of the near side of the Moon is challenging due to the Earth’s atmosphere, while the far side is not visible neither from the Earth surface nor from a near Earth orbit. The only sources of such information are lunar orbiting satellites, such as Clementine (Lemelin et al., 2013) and Lunar Reconnaissance Orbiter (LRO) (Keller et al., 2016), or space telescopes (Robinson et al., 2007). A significant challenge is that images taken from a near lunar orbit cover only a small portion of the Moon’s surface, requiring thousands of individual images to be stitched together to create a complete map of both lunar hemispheres. This process involves addressing issues of images alignments and corrections, as the images are captured under varying lighting conditions, as well as dealing with gaps caused by various factors.

The Earth Polychromatic Imaging Camera (EPIC) aboard the Deep Space Climate Observatory (DSCOVR) is the only exception. NASA-NOAA DSCOVR mission was launched in 2015 and is located near the Sun-Earth Lagrange-1 (L1) stable point (million miles away from Earth towards the Sun), 3–4 times farther than the Moon. EPIC captures 13–22 images of the sunlit side of Earth per day. EPIC operates in 10 narrow spectral bands from ultraviolet to near-infrared (317–780 nm) – see Table 1.

Table 1

Table 1. Specifications of the EPIC filters (Cede et al., 2021).

EPIC provides data on total ozone column (REF), volcanic sulfur dioxide (SO₂) and ash index (Carn et al., 2018), aerosols (Torres et al., 2020), clouds, surface UV irradiance (Herman et al., 2020), ocean color, and vegetation (Herman et al., 2018; Marshak et al., 2018). It uses images of the Moon’s near side for calibration (Geogdzhayev and Marshak, 2018; Doelling et al., 2019; Cede et al., 2021; Geogdzhaev et al., 2021; Haney et al., 2021). During solar eclipses on Earth, the far side of the Moon also falls within the instrument’s field of view (Figure 1). Such transits, where the Moon moves between DSCOVR and Earth, have occurred several times since observations began in 2015 (see https://epic.gsfc.nasa.gov/galleries). Lunar observations with EPIC hold significant interest for both science and the public, offering an unusual perspective on the Earth-Moon system.

Figure 1

Figure 1. Image of Earth and the Moon taken by EPIC on 5 July 2016. Left: Earth and the Moon image taken with the 551 nm spectral filter (The image was generated using the Panoply software). Right: Naturally-looking color image of Earth and the Moon passing between DSCOVR and the Earth. It was created combining the three EPIC spectral bands accounting for the human visual sensitivity.

2 Materials and methods

Thanks to EPIC’s unique vantage point, high stability (Valero et al., 2021), and multi-spectral capabilities, images of the Moon taken by EPIC can serve various scientific purposes: (a) calibration of lunar maps produced by other missions under different observing times and conditions; (b) generation of geological maps of the Moon based on spectral reflection characteristics of different minerals; and (c) creation of a new type of composite maps through fusion of data from other missions.

On 09/27/2015, 09/01/2016, and 04/23/2024, a total of 78 maps of the near side and 30 maps of the far side of the full illuminated Moon were obtained. The authors of this paper initiated a program of additional lunar observations, during which EPIC acquired 117 images of the fully illuminated far side of the Moon in three sessions in 2025 (April 28, May 26, and June 25). The resulting database of 225 multispectral maps of the Moon obtained by EPIC has significant scientific value and can be used to determine the mineralogical composition of the lunar surface and to identify regions with the highest concentrations of economically valuable minerals, such as ilmenite (FeTiO₃). This database is composed only of lunar images that do not overlap with images of the Earth, since such overlap increases stray light.

2.1 Observations of the far side of the Moon

The image of the far side of the Moon occupies up to 600 pixels on EPIC’s two-dimensional (2D) CCD detector, meaning that one pixel corresponds to approximately ∼6 km at the center of the near disk of the Moon (Figure 2, left).

Figure 2

Figure 2. An image of the far side of the Moon captured by EPIC on 16 July 2015, at a wavelength of 680 nm (left, with contours added for better details; the image was generated using the Panoply software; file epic_1a_20150716210604_03.h5). The axis values correspond to the pixels of the EPIC detector array. The image on the left corresponds to a 600 × 600 pixels square. The spatial resolution at the center of the lunar disk is 6 km per pixel. The high-resolution image on the right was taken by the Lunar Reconnaissance Orbiter (LRO) spacecraft (NASA/Goddard Space Flight Center/Arizona State University; https://science.nasa.gov/resource/lunar-far-side-2/). The resolution of the LRO WAC cameras in the visible range is ∼100 m. EPIC image is an instantaneous snapshot of the Moon; LRO image is composed of individual frames taken at different times. We do not change the orientation of the images obtained by EPIC, but we indicate the location of the Moon poles.

A quasi-monochromaric EPIC image of the far side of the Moon taken on 16 July 2015, centred at 680 nm is shown in Figure 2 (left). The right panel shows a higher-resolution image of the far side of the Moon assembled from data taken by NASA’s Lunar Reconnaissance Orbiter (LRO) (https://science.nasa.gov/resource/lunar-far-side-2/). The image on the right was oriented to compensate for the different observation geometry and libration effects (Gorkavyi et al., 2023). Unlike the near side of the Moon, characterized by dark maria (volcanic plains), the far side of our natural satellite is heavily cratered and lacks large maria.

Comparing lunar images captured by EPIC with those from the Lunar Reconnaissance Orbiter (LRO) or ground-based telescopes can refine the calibration of instruments operating in overlapping spectral ranges. This ensures the consistency of lunar reflectance models, which is critical for calibrating data from different Earth observing missions and nocturnal aerosol measurements from the Earth’s surface (Berkoff et al., 2011). Images of the Moon captured by EPIC are highly stable, showing virtually no changes over time (Geogdzhaev et al., 2021). Figure 3 shows two images of the far side of the Moon at 443 nm taken on 16 July 2015 (left) and 5 July 2016 (right) on different parts of the CCD detector.

Figure 3

Figure 3. The far side of the Moon as seen by EPIC at a blue wavelength of 443 nm for two dates: at 6:04UTC on 16 July 2015 (a) and 5:43UTC on 5 July 2016 (b) (epic_1a_20160705054354_03.h5). The images were generated using the Panoply software. The black square highlights a ∼100 × 100 pixel homogeneous region zoomed in Figure 4.

Figure 4 compares two zoom-in images of the same region on the far side of the Moon (shown with the black square in Figure 3b and also in Figure 8b), taken 30 min apart using red filter centred at 680 nm. The difference between the images shows that the registration noise is random, and noise level is low (note the smaller scale of the difference image). Figure 4 shows, in a different color palette, a portion of the South Pole–Aitken Basin with dark large craters and bright small craters that formed later. The linear structure in the lower right part of the image is likely ejecta from a large crater located outside the field of view.

Figure 4

Figure 4. Reflectance of the South Pole–Aitken Basin region on the far side of the Moon, as measured by EPIC on 1 September 2016, at a wavelength of 680 nm. (a) Image taken at 14:03 UTC (size 100 × 100 pixels or ∼700 × 700 km). (b) Image taken at 14:33UTC. (c) Difference between images a,b demonstrating the low noise level.

EPIC offers data on the Moon’s broad spectral reflectance. Figure 5 shows the significant dependence of the far side spectral reflectance by comparing blue (443 nm) and red (670 nm) band images with the same reflectance scale. A notable feature is the increase in reflectance with wavelength (Figure 5). The different positions of the Moon relative to the Earth’s disk in Figures 3 (right), 5 are due to the fact that the photos were taken at different times of the day.

Figure 5

Figure 5. Image of the far side of the Moon captured by EPIC against the Earth disc at two wavelengths on 5 July 2016 (epic_1a_20160705042828). Interestingly, while the reflectance of the Moon increases with wavelength, the reflectance of the Earth’s ocean is decreasing. The slight shift between the images, noticeable in some details, is due to the fact that the photos at different wavelengths were taken sequentially at different times (∼4 min between 443 nm and 680 nm). The images were generated using the Panoply software.

2.2 Observations of the near side of the Moon

The image of the near side of the Moon occupies up to 340 pixels on EPIC’s CCD detector, meaning that one pixel corresponds to approximately ∼10 km at the center of the disk of the Moon. This is because EPIC observes the near side of the Moon from a greater distance than the far side (the difference is equal to the diameter of the Moon’s orbit).

To evaluate the spectral properties of the near side of the lunar surface, we also consider images taken by EPIC on 23 April 2024 (Figure 6) when the Moon was captured behind Earth.

Figure 6

Figure 6. Left: EPIC image of the Moon’s near side obtained with UV filter (centered at 325 nm) on 23 April 2024. The image on the left corresponds to a 340 × 340 pixels square. The white square marks typical mountainous terrain, while the black one indicates a lunar sea area (maria). Right: Moon image taken from the NASA’s website for 22 June 2024 (https://moon.nasa.gov/moon-observation/daily-moon-guide/?intent=011#1719028800000::1::event; the panel is oriented to match the EPIC image as closely as possible). The image was generated using the Panoply software.

Images of the Moon during its transits captured by EPIC allow to study how the Moon’s reflectance changes depending on the phase angle, which varies across different points of the image from 0 to 90° (Figures 5, 6). This improves our understanding of the scattering properties of lunar regolith. Such data can also be used to enhance surface reflectance models for exoplanets, as the Moon serves as an analogy for rocky planetary bodies.

3 Results

The EPIC L1 vantage point provides an excellent opportunity to compare the reflectance spectra of the different surfaces on the near and far sides of the Moon such as lunar maria (e.g., black square in the Sea of Showers in Figure 6) and the highland regions (e.g., shown as white square near the Crater Tycho in Figure 6).

3.1 Spectra of the near and far sides of the Moon

Figure 7 compares average spectral reflectance from different regions on the Moon. Curves 3 and 4 show the whole Moon-average spectra of the far (red curve 3) and near side (blue curve 4) obtained from EPIC data. It follows from Figure 7 that the far side of the Moon is 15%–20% brighter than the near side. This is related to the fact that the near side of the Moon contains significantly darker lunar mare. The spectra of the lunar mare (violet curve 5) and highland (black curve 2) regions are also shown in Figure 7. These regions are shown in Figure 6 (left) with black (mare) and white (highland) squares, respectively. It is clearly seen that the brightness of the highland regions exceeds that of the dark mare plains at all EPIC wavelengths by a factor of two.

Figure 7

Figure 7. Spectra of the lunar highlands (black curve 2) and mare (5, violet) (see Figure 6A) for 10 EPIC wavelengths averaged by 2,500 pixels (50 × 50 pixels). The red curve 3 shows the average reflectance of the far side of the Moon, while the blue curve 4 shows the average reflectance of the near side of the Moon. The green (1) and black with open squares (6) curves correspond to the average reflectance of the highland anorthosite (calcium rich white rock) regions and mare basalt (a dark-coloured rocks formed by lava) regions, respectively (Zou et al., 2004).

Also shown in Figure 7 are spectral data from Zou et al. (2004), where the spectrum of the lunar mare (curve 6) and lunar mountains (curve 1) in the range of 300–1,000 nm were calculated based on: (a) Spectral measurements of lunar soil samples collected by the Apollo missions from the regions of the lunar mare and mountains; (b) Calibration of the obtained spectrum against the observed reflectance of the entire Moon at 300 nm. We can see that despite the qualitative agreement between the spectral shapes there are significant quantitative differences between the spectra. For example, according to the model of Zou et al. (2004), the brightness of the lunar highlands is 10%–20% higher than that measured by EPIC. In contrast, the reflectance of the lunar maria, according to EPIC data, approximately agrees with the Zou et al. (2004) model in the ultraviolet range but exceeds that model by about 20% in the visible range. Although the definitions of highlands and maria used in this paper do not exactly coincide with those in Zou et al. (2004), it is unlikely that this noticeable discrepancy in the results can be attributed to this difference alone. We note that for the spectral analysis of the lunar maria and highlands in the EPIC images, we selected two representative regions of mare and highland terrain, each 50 × 50 pixels in size (approximately 500 × 500 km), marked by black and white squares in Figure 6 (left). In the work by Zou et al. (2004), the entire set of lunar maria and highland regions was considered according to a lunar map in which about 25% of the surface corresponds to maria and 75% to highlands.

3.2 Multispectral images of the Moon and reflectance ratio 340 nm/551 nm

Ilmenite (FeTiO₃) is a key mineral in lunar mare regions, is scientifically important and critical forin-situ resource utilization because it contains oxygen, titanium, and iron, and can retain solar-wind-derived hydrogen and helium-3. Since the extraction these elements is essential for establishing a sustained lunar base, mapping the distribution of ilmenite across the Moon is highly valuable for the future lunar expeditions.

A distinctive feature of ilmenite is its unusual spectral behavior: unlike the average lunar surface, whose reflectance increases from the ultraviolet to visible wavelengths, ilmenite exhibits higher reflectance at ∼340 nm than at ∼550 nm (see Figure 8). This inverse UV-VIS spectral slope makes the 340/551 nm reflectance ratio a useful diagnostic for identifying ilmenite-rich materials. High values of this ratio therefore provide a first-order indication of regions enriched in ilmenite-bearing basalts. For reference, ilmenite has a characteristic ratio of ∼1.15, while the global lunar average is ∼0.59 (Figure 7).

Figure 8

Figure 8. (a) Spectral signature of plagioclase (Plg); orthopyroxene (Opx), clinopyroxene (Cpx), olivine (Ol), and ilmenite (Ilm) from the RELAB spectral database (Figure 1 from Lemelin et al., 2013). The hatched rectangle highlights the spectral region where EPIC performs its measurements. It can be seen that within this spectral region the reflectance of all typical lunar minerals (with the exception of ilmenite) increases with increasing wavelength. (b) The spectrum of ilmenite (black line) shows a decrease in the wavelength range of 300–500 nm, which allows it to be distinguished from basalts (blue and red lines) (https://www.lroc.asu.edu/about/objectives). The LRO wide area camera (WAC) bandpasses are marked with green lines, and the Clementine band centered at 415 nm is indicated by the black horizontal line. The vertical red lines indicate the eight EPIC wavelengths (two NIR channels centered at 763.7 nm and 779.2 nm are outside the bounds of this figure). For preliminary analysis we selected images of the Moon taken by EPIC at 340 nm and 551 nm wavelengths (indicated by red arrows).

To apply this diagnostic, we constructed a 340/551 nm reflectance ratio (R_340/551) map using EPIC observations (Figure 9) for both near and far sides of the Moon. Areas displaying high ratios in this map (e.g., Sea of Traquility, Figure 9c) are interpreted as potential ilmenite-rich terrains, offering a preliminary view of how ilmenite is distributed across the lunar surface.

Figure 9

Figure 9. Top: Reflectance of the Moon’s (a) near side obtained by EPIC at 340 nm on 23 April 2024, and (b) far side obtained at 340 nm on 1 September 2016. Bottom: (c,d) reflectance ratio: 340 nm/551 nm. The average ratio for both hemispheres is 0.59 according to EPIC data (see Figure 7). Red areas on the near and far sides of the Moon show regions with the enhanced reflectance ratio R_340/551 indicating titanium-rich (ilmenite) regions. The Sea of Tranquility (c) is likely to have the highest concentration of ilmenite.

3.3 The case of the Sea of Tranquility

The Sea of Tranquility, located near the equator on the near side of the Moon, has the highest concentration of ilmenite (Figure 9c). This is confirmed by comparison with the detailed map of this region created using composite of LROC data (Figure 10a). The bluish tint on the LROC map indicates titanium-rich areas, which closely correspond to the areas with high 340/551 nm ratios in the EPIC map (Figure 10b).

Figure 10

Figure 10. (a) LRO/LROC wide area camera (WAC) false-color synthetic image (i.e., mosaic using 320 nm, 415 nm, 689 nm) (https://www.lroc.asu.edu/images/223). The bluish color area of the Sea of Tranquility indicates a high titanium content in ilmenite (FeTiO₃). The image contrast has been enhanced. The inset at the top shows the area around Proclus crater (marked by a circle), which exhibits a bluish tint of the bright ejecta. The image is presented in a selenochromatic format (https://en.wikipedia.org/wiki/Proclus_(crater); methodology by Aldo Ferruggia and the Gruppo Astrofili William Herschel). (b) EPIC map of the reflectance ratio at 340 nm and 551 nm, R_340/551, with the same scale shown in Figure 8. The distribution of this ratio closely matches the titanium distribution on the LROC map. The Proclus crater with a diameter of 27 km, occupies 2-3 pixels on the EPIC map. The EPIC map is slightly rotated for comparison with the LROC map.

The LROC map has a better spatial resolution (∼100 m per pixel in the visible part of the spectrum and ∼400 m per pixel in the ultraviolet range) but reflects difficulties with stitching of individual “snapshots” taken under different illumination conditions. In contrast, the EPIC map in Figure 10b, generated using just two full Moon images at different wavelengths, has no issues with gaps or stitching, as it captures the entire disc of the Moon instantaneously for each wavelength. Accumulating more images of both the near and far sides of the Moon and merging them into high-quality multispectral maps will enhance spatial resolution (e.g., through oversampling techniques). This improvement will increase our understanding of the distribution of minerals on the lunar surface, critically important for lunar exploration and the Artemis program.

We note that in Figure 10a the bright ejecta from fresh craters (for example, from the crater Proclus near the right edge of the image, at northern latitudes 15°–16° and eastern longitude around 47°) do not exhibit a bluish tint. This suggests that the ejecta contains little ilmenite. However, in Figure 10b this ejecta is characterized by a large reflectance ratio R_340/551. We associate this difference with the fact that Figure 10b uses the reflectance at 551 nm, which is close to the minimum of ilmenite reflectance (see Figure 8b); therefore, this wavelength is highly sensitive to the presence of ilmenite. Crater ejecta is also rich in bright plagioclase with a reflectance of about 0.6 at 689 nm (see Figure 8a). The interesting case of the bright rays from Proclus crater requires a detailed analysis. We propose a preliminary hypothesis that bright plagioclase, with a reflectance peak in the red part of the spectrum, masks the ilmenite signature in the 340 and 551 nm bands. The relative abundance of ilmenite has certainly decreased (a shift of the parameter R_340/551 from the limiting value of 0.7 to 0.65 apparently corresponds to roughly a factor-of-two reduction in the ilmenite fraction in the reflectance), but it is still present in an amount significantly exceeding the average level, which corresponds to a value of R_340/551 about 0.59. However, due to the contribution of bright plagioclase, the bluish tint in the image constructed from the 320 nm, 415 nm, and 689 nm bands (where the plagioclase contribution at 689 nm is large) disappears. As can be seen from the inset in Figure 10a, a different method of producing color images reveals a bluish оттенок of the bright ejecta around Proclus crater. Therefore, the reflectance ratio R_340/551 provides a more accurate indicator of ilmenite.

4 Discussion

The images of the whole Moon captured simultaneously by EPIC at different wavelengths can be helpful for calibrating data from various satellite instruments. EPIC’s lunar data can be used to validate models and measurements from other missions. For instance, the spectral and photometric data of the Moon can complement observations from instruments on the LRO (Chin et al., 2007; Keller et al., 2016) or Chang’e (Qian et al., 2021) missions, helping studies of the lunar surface and reflectance.

During its first decade of observations from 2015 to 2025, EPIC captured 78 maps of the entire lunar near side and 147 maps of the far side (e.g., https://epic.gsfc.nasa.gov/galleries/2021/lunar_transit); we count only images of the Moon against the background of space; images taken against the Earth are excluded because of additional stray light). We suggest that summing a large number of lunar images into a single map could increase the resolution in the central regions of the disk to about 1 km per pixel. These data could provide valuable information about the Moon’s surface chemical composition.

In particular, global lunar maps in the ultraviolet spectral range were previously obtained by LROC WAC at two wavelengths—320 nm and 360 nm. EPIC data provide lunar maps at four different ultraviolet wavelengths: 317, 325, 340 and 388 nm. We demonstrate that the data obtained by EPIC are in good qualitative agreement with results reported by previous missions and with the LROC WAC maps. Therefore, our EPIC data complement other missions and confirm known spectral features. This is an essential step in the validation of this new data set. The complete scientific value of EPIC’s lunar maps will emerge through further detailed studies. Comparing EPIC data with LROC’s global multispectral data (300 nm–680 nm) to characterize lunar resources, especially ilmenite (Keller et al., 2016; Petro et al., 2017), will mutually enrich these datasets.

A future Earth looking observatory placed at the Lagrange point L1 and specifically adapted for high-resolution imaging of the Earth and Moon could provide a wealth of valuable information. For example, using a higher 4000 × 4000 CCD detector would allow imaging the entire illuminated hemisphere of the Moon with an inherent resolution of about 1 km at the center of the disk for a single image. Simultaneously, such a satellite, equipped with an appropriate radio or laser communication equipment, could perform a critical function as a relay for signals from robotic spacecraft and crewed modules on both near and far side of the Moon.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: https://epic.gsfc.nasa.gov/.

Author contributions

NG: Software, Visualization, Writing – original draft, Writing – review and editing. NK: Investigation, Supervision, Writing – review and editing. AM: Conceptualization, Investigation, Project administration, Writing – review and editing.

Funding

The author(s) declared that financial support was not received for this work and/or its publication.

Acknowledgements

The authors thank Harrison Schmitt, Oreste Reale, Karin Blank and reviewers for the helpful discussions and Marshall Sutton and the EPIC team for providing the optimized lunar data.

Conflict of interest

Authors NG are employed by Science Systems and Applications, Inc.

The remaining author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Berkoff, T. A., Sorokin, M., Stone, T., Eck, T. F., Hoff, R., Welton, E., et al. (2011). Nocturnal aerosol optical depth measurements with a small-aperture automated photometer using the moon as a light source. J. Atmos. Ocean. Technol. V. 28, 1297–1306. doi:10.1175/JTECH-D-10-05036.1

CrossRef Full Text | Google Scholar

Carn, S. A., Krotkov, N. A., Fisher, B. L., Li, C., and Prata, A. J. (2018). First observations of volcanic eruption clouds from the L1 earth-sun lagrange point by DSCOVR/EPIC. Geophys. Res. Lett. 45 (11), 456–464. doi:10.1029/2018GL079808

CrossRef Full Text | Google Scholar

Cede, A., Huang, L. K., McCauley, G., Herman, J., Blank, K., Kowalewski, M., et al. (2021). Raw EPIC data calibration. Front. Remote Sens. 2. doi:10.3389/frsen.2021.702275

CrossRef Full Text | Google Scholar

Chin, G., Brylow, S., Foote, M., Garvin, J., Kasper, J., Keller, J., et al. (2007). Lunar reconnaissance orbiter overview: the instrument suite and mission. Space Sci. Rev. 129, 391–419. doi:10.1007/s11214-007-9153-y

CrossRef Full Text | Google Scholar

Doelling, D., Haney, C., Bhatt, R., Scarino, B., and Gopalan, A. (2019). The inter-calibration of the DSCOVR EPIC imager with Aqua-MODIS and NPP-VIIRS. Remote Sens. 11, 1609. doi:10.3390/rs11131609

CrossRef Full Text | Google Scholar

Geogdzhaev, I. V., Marshak, A., and Alexandrov, M. (2021). Calibration of the DSCOVR EPIC visible and NIR channels using multiple LEO radiometers. Front. Remote Sens. 2. doi:10.3389/frsen.2021.671933

CrossRef Full Text | Google Scholar

Geogdzhayev, I. V., and Marshak, A. (2018). Calibration of the DSCOVR EPIC visible and NIR channels using MODIS and EPIC lunar observations. Atmos. Meas. Tech. doi:10.5194/amt-2017-222

CrossRef Full Text | Google Scholar

Gorkavyi, N., Krotkov, N., and Marshak, A. (2023). Earth observations from the Moon's surface: dependence on lunar libration. Atmos. Meas. Tech. 16, 1527–1537. doi:10.5194/amt-16-1527-2023

CrossRef Full Text | Google Scholar

Haney, C., Doelling, D., Su, W., Bhatt, R., Gopalan, A., and Scarino, B. (2021). Radiometric stability assessment of the DSCOVR EPIC visible bands using MODIS, VIIRS, and invariant targets as independent references. Front. Remote Sens. 2, 765913. doi:10.3389/frsen.2021.765913

CrossRef Full Text | Google Scholar

Herman, J. R., Huang, L., McPeters, R. D., Ziemke, J., Cede, A., and Blank, K. (2018). Synoptic ozone, cloud reflectivity, and erythemal irradiance from sunrise to sunset for the whole Earth as viewed by DSCOVR spacecraft from the earth-sun Lagrange-1. Atmos. Meas. Tech. 11, 177–194. doi:10.5194/amt-11-177-2018

CrossRef Full Text | Google Scholar

Herman, J., Cede, A., Huang, L., Ziemke, J., Torres, O., Krotkov, N., et al. (2020). Global distribution and 14-year changes in erythemal irradiance, UV atmospheric transmission, and total column ozone for2005–2018 estimated from OMI and EPIC observations. Atmos. Chem. Phys. 20, 8351–8380. doi:10.5194/acp-20-8351-2020

CrossRef Full Text | Google Scholar

Keller, J. W., Petro, N. E., and Vondrak, R. R., and the LRO team (2016). The lunar reconnaissance orbiter mission – six years of science and exploration at the moon. Icarus 273, 2–24. doi:10.1016/j.icarus.2015.11.024

CrossRef Full Text | Google Scholar

Lemelin, M., Morisset, C.-E., Germain, M., Hipkin, V., Goïta, K., and Lucey, P. G. (2013). Ilmenite mapping of the lunar regolith over mare australe and mare ingenii regions: an optimized multisource approach based on hapke radiative transfer theory. J. Geophys. Res. Planets 118, 2582–2593. doi:10.1002/2013JE004392

CrossRef Full Text | Google Scholar

Marshak, A., Herman, J., Szabo, A., Blank, K., Cede, A., Carn, S., et al. (2018). Earth observations from DSCOVR/EPIC instrument. Bulletin of the American Meteorological Society. doi:10.1175/BAMS-D-17-0223.1

CrossRef Full Text | Google Scholar

Petro, N. E., Keller, J. W., and Gaddis, L. R. (2017). Editorial introduction: lunar reconnaissance orbiter, part II. Icarus 283, 1. doi:10.1016/j.icarus.2016.11.018

CrossRef Full Text | Google Scholar

Qian, Y., Xiao, L., Wang, Q., Head, J. W., Yang, R., Kang, Y., et al. (2021). China's Chang'e-5 landing site: geology, stratigraphy, and provenance of materials. Earth Planet. Sci. Lett. 561, 116855. doi:10.1016/j.epsl.2021.116855

CrossRef Full Text | Google Scholar

Robinson, M. S., Hapke, B. W., Garvin, J. B., Skillman, D., Bell III, J. F., Ulmer, M. P., et al. (2007). High resolution mapping of TiO₂ abundances on the moon using the hubble space telescope. Geophys. Res. Lett. 34, L13203. doi:10.1029/2007GL029754

CrossRef Full Text | Google Scholar

Torres, O., Bhartia, P. K., Taha, G., Jethva, H., Das, S., Colarco, P., et al. (2020). Stratospheric injection of massive smoke plume from Canadian boreal fires in 2017 as seen by DSCOVR-EPIC, CALIOP and OMPS-LP observations. J. Geophys. Res. Atmos. 125, e2020JD032579. doi:10.1029/2020JD032579

CrossRef Full Text | Google Scholar

Valero, F. P. J., Marshak, A., and Minnis, P. (2021). Lagrange point missions: the key to next generation integrated Earth observations. DSCOVR innovation. Front. Remote Sens. 2, 745938. doi:10.3389/frsen.2021.745938

CrossRef Full Text | Google Scholar

Zou, Y. L., Liu, J. Z., Liu, J. J., and Xu, T. (2004). Reflectance spectral characteristics of lunar surface materials. Chin. J. Astron. Astrophys. 4, 97–104. doi:10.1088/1009-9271/4/1/97

CrossRef Full Text | Google Scholar