A low-light radiative transfer model for satellite observations of moonlight and earth surface light at night

https://doi.org/10.1016/j.jqsrt.2020.106954Get rights and content

Highlights

  • New Low-Light Radiative Transfer Model with active surface light sources is developed.

  • The simulation of the new model has a minimal bias compared with the DISORT model.

  • The new model shows potential to simulate space-based low-light imager observations.

Abstract

Lunar sun-reflected light can be effectively measured through a low-light band or a day/night band (DNB) implemented on space-based optical sensors. Based on moonlight, nocturnal observations for artificial light sources at night can be achieved. However, to date, an open-sourced and mature Low-Light Radiative Transfer Model (LLRTM) for the further understanding of the radiative transfer problem at night is still unavailable. Therefore, this study develops a new LLRTM at night with the correction of the lunar and active surface light sources. First, the radiative transfer equations with an active surface light source are derived for the calculation based on the lunar spectral irradiance (LSI) model. The simulation from this new LLRTM shows a minimal bias when compared with the discrete ordinates radiative transfer (DISORT) model. The simulated results of radiance and reflectance at the top of the atmosphere (TOA) also show that the surface light source has a remarkable impact on the radiative transfer process. In contrast, the change in the lunar phase angle has minimal influence. Also, comparing with space-based DNB radiance observations, LLRTM shows the potential to simulate space-based low-light imager observations under an effective surface light source condition during the night.

Introduction

In a clear night sky, natural light remains strong because of a unique light source—the moon, which can reflect solar light to the Earth's surface directly. The moon's sun-reflected light can be effectively measured by using a low-light band or day/night band (DNB), which is implemented on some custom-designed and advanced satellite optical sensors, such as the Operational Linescan System (OLS) [1], [2], [3]. Compared with relatively stable solar spectral irradiances, the extremely low magnitudes of periodical variation of lunar spectral irradiance (from 10−9 to 10−7 W·m−2·sr−1) is correlated with lunar phase angle changes, Sun-Earth-Moon geometries and, in particular, the lunar phase-dependent moon surface materials (or albedo) [4,5]. Nevertheless, the nocturnal observation for artificial light sources at night using data collected by the U.S. Air Force Defense Meteorological Satellite Program (DMSP) Operational Linescan System has been investigated since 1976 [1,6,7]. For the low-light band, a Photo Multiplier Tube (PMT) was used to record the signal with a broad spectral response (440–940 nm). The visible band signal (500–650 nm) illuminated from the reflection of moonlight by clouds has the highest PMT sensitivity. In addition, this low-light band also covers the range for primary emissions from the most widely used surface lamps: mercury vapor (545 nm and 575 nm), high-pressure sodium vapor (from 540 nm to 630 nm), low-pressure sodium vapor (589 nm), etc. [2,6,8].

The successor of the DMSP/OLS, an advanced Day/Night Band (DNB) on the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the NOAA (National Oceanic and Atmospheric Administration) Suomi National Polar-orbiting Partnership (SNPP) satellite was successfully launched on 28 October 2011 [9,10]. The first dedicated nighttime light remote sensing satellite in the world, the LuoJia1-01 satellite, led by Wuhan University, was successfully launched on 2 June 2018 [11]. In addition, a new DNB will be onboard FengYun-3 (FY-3) Early-Morning-Orbit satellite's Medium Resolution Spectral Imager (MERSI), which is scheduled to launch in 2020 [12], [13], [14], [15]. The space-based DNB observations not only allow users to examine social and economic activities at urban scales or monitor hurricanes [2,[16], [17], [18], [19], but can also provide high-quality sub-pixel cloud test information for passive microwave and infrared hyperspectral sensors, which act as a critical data source for data assimilation modules in numerical weather prediction modeling [9,20,21].

According to previous research, compared with infrared band observations, the DNB observations can capture more details of the Earth's surface through relatively weak lunar reflected light at night [22]. It attempts to retrieve nighttime optical depth and microphysical features of aerosols and clouds by using the VIIRS/DNB measurements. This study shows the dependency of the retrieval on different regions, for example, those with and without artificial surface lights [23,24]. However, the increasing number of artificial lights at the surface, primarily in urban areas resulting in increasing radiance, has a notable impact on cloud and aerosol retrievals from space-based DNB observations [16]. Nevertheless, the basic radiative transfer model at night does not perform well in processing data containing surface light sources in the visible band, such as the retrieval of cloud nighttime optical depth [25,26]. The Low-Light Radiative Transfer Model (LLRTM) at night is one of the essential prerequisites and tools for promoting practical quantitative applications of space-based DNB data. However, the current version of LLRTM does not take into account the surface anthropological light source, and it is still a test version. [25].

Therefore, the primary goal of this investigation is to develop a new LLRTM with corrections for lunar and active surface light sources. The next sections will introduce the derivation of the low-light radiative transfer equations with lunar and active surface light sources. The validation results for the new LLRTM are shown in the Appendix section. The following sections will demonstrate the comparison of simulated radiance and reflectance at the TOA between LLRTM with and without effective surface light sources. The contrast between the developed LLRTM simulations and VIIRS/DNB observations will also be discussed. Finally, significant findings will be summarized.

Section snippets

Fundamentals of lunar radiative transfer

First, we reviewed the fundamentals of traditional atmospheric radiative transfer under a solar incident condition [27]. In essence, the physical processes of radiative transfer between a conventional solar reflective RTM and a new RTM with a lunar irradiance source at night are nearly the same, except for solar and lunar sources, which have different intensities of irradiance and distances with Earth [25]. Besides, a plane-parallel solar irradiance at the top of the atmosphere is assumed by

Radiances and reflectance at the TOA

Like previous study [25], we simulate the spectral radiance and reflectance (see Eq. (29)) of clear sky at the TOA from 0.4 to 2.0 µm with the lunar irradiance source using LLRTM under four lunar phase angles of θ = 0°, 45°, 90°, and 135° in Fig. 4. Lunar zenith angle, view zenith angle, and relative azimuth angle respectively equal to 20°, 10°, and 0° This model considers seven major atmospheric absorptive gases (H2O, CO2, O3, O2, CH4, CO, and N2O) based on a rigorous line-by-line (LBL)

Conclusion

This study attempts to develop a new low-light radiative transfer model with an active surface light source and a lunar irradiance source for space-based low-light band (or DNB) measurements at night. The primary objective of this model is to promote practical quantitative applications of space-based DNB data in the future. Here, we elucidate the lunar irradiance model, the active surface light irradiance model, and how to build this low-light radiative transfer model from basic theory to code.

CRediT authorship contribution statement

Min Min: Conceptualization, Data curation, Methodology, Investigation, Software, Funding acquisition, Resources, Writing - review & editing. Zheng Jianyu: Conceptualization, Methodology, Data curation, Investigation, Software, Writing - review & editing. Zhang Peng: Conceptualization, Methodology, Project administration, Supervision. Hu Xiuqing: Formal analysis, Resources, Supervision. Chen Lin: Formal analysis, Validation. Li Xi: Formal analysis, Resources. Huang Yu: Writing - review &

Declaration of Competing Interest

The authors have no affiliation with any organization with a direct or indirect financial interest in the subject matter discussed in the manuscript.

Acknowledgments

The authors would like to acknowledge Dr. Steven Miller and Dr. Christopher D. Elvidge for making his lunar irradiance model and light intensity measurements publicly available, respectively. We thank the STEVENS Institute of Technology Laboratory (http://lllab.phy.stevens.edu/disort) and Atmospheric and Environmental Research Company (http://rtweb.aer.com/lblrtm_frame.html) for freely providing the DISORT-V2.0 and LBLRTM models. Suomi NPP VIIRS/DNB and MODIS data are freely downloaded from

References (42)

  • X. Li et al.

    Potential of NPP-VIIRS nighttime light imagery for modeling the regional economy of China

    Remote Sens (Basel)

    (2013)
  • X. Li et al.

    Remote sensing of night-time light

    Int J Remote Sens

    (2017)
  • H. Kieffer et al.

    The spectral irradiance of the moon

    Astron J

    (2005)
  • S.D. Miller et al.

    A dynamic lunar spectral irradiance data set for NPOESS/VIIRS day/night band nighttime environmental applications

    IEEE Trans Geosci Remote Sens

    (2009)
  • S.D. Miller et al.

    Suomi satellite brings to light a unique frontier of nighttime environmental sensing capabilities

    Proc Natl Acad Sci

    (2012)
  • C.D. Elvidge et al.

    The NightSat mission concept

    Int J Remote Sens

    (2007)
  • C. Cao et al.

    Early on-orbit performance of the visible infrared imaging radiometer suite onboard the Suomi National Polar-orbiting Partnership (S-NPP) satellite

    IEEE Trans Geosci Remote Sens

    (2014)
  • S. Lee et al.

    The S-NPP VIIRS day-night band on-orbit calibration/characterization and current state of SDR products

    Remote Sens

    (2014)
  • Zhang G., Guo X., Li D., Jiang B. Evaluating the potential of LJ1-01 nighttime light data for modeling socio-economic...
  • X. Hu et al.

    Calibration for the solar reflective bands of medium resolution spectral imager onboard FY-3A

    IEEE Trans Geosci Remote Sens

    (2012)
  • Min M., Cao G., Xu N., Bai Y., Jiang S., Hu X., et al. On-orbit spatial quality evaluation and image restoration of...
  • Cited by (10)

    View all citing articles on Scopus
    View full text