New features to the night sky radiance model illumina: Hyperspectral support, improved obstacles and cloud reflection

doi:10.1016/j.jqsrt.2018.02.033

Journal of Quantitative Spectroscopy and Radiative Transfer

Volume 211, May 2018, Pages 25-34

https://doi.org/10.1016/j.jqsrt.2018.02.033 Get rights and content

Highlights

•
The blue component of the Artifical Sky Brightness (ASB) is lower for distant observers.
•
Ground reflectance and atmospheric extinction attenuate the blue content of the ASB.
•
Full cutoff lamps reduce the ASB but decreases the ASB contrast from horizon to zenith.
•
Buildings and trees can reduce the ASB by a factor of 5 while clouds increase it by a factor of up to 24.
•
Cloudy sky ASB show clumpy structures highly linked to nearby light sources.

Abstract

Illumina is one of the most physically detailed artificial night sky brightness model to date. It has been in continuous development since 2005 [1]. In 2016–17, many improvements were made to the Illumina code including an overhead cloud scheme, an improved blocking scheme for subgrid obstacles (trees and buildings), and most importantly, a full hyperspectral modeling approach. Code optimization resulted in significant reduction in execution time enabling users to run the model on standard personal computers for some applications.

After describing the new schemes introduced in the model, we give some examples of applications for a peri-urban and a rural site both located inside the International Dark Sky reserve of Mont-Mégantic (QC, Canada).

Introduction

The main goal of this paper is to provide a description of new features incorporated in the sky brightness model named Illumina [1], [2], [3]. This radiative transfer model aim to simulate the measurements from a virtual spectrometer located anywhere in the world and looking toward any viewing angle. Illumina uses some ray tracing and statistical optimization techniques to reduce the computation time and a voxel-based simulation domain characterized by: 1) A radiant flux map of the light sources in the territory; 2) the angular photometry and spectral power distribution (SPD) of these light sources; and 3) a description of the spectral and geometrical properties of the environment (e.g. the spectral reflectance of the ground, the size and density of the obstacles, the orography). The model then calculates the path of a statistically selected set of photons exiting the light sources of the domain to every point in the line of sight. The first and second order of scattering is considered along with possible reflections on the ground surface. Illumina can be freely downloaded [4] and used by anyone since the code is released under GNU Public Licence.

The new features that will be presented here are: 1) The implementation of hyperspectral capabilities; 2) the improvement of the subgrid obstacle blocking scheme; and 3) the addition of a diffuse reflexion cloud scheme.

In this paper, case studies, applied to the Mont-Mégantic Dark Sky Reserve (QC, Canada), will be presented to illustrate the new possibilities provided by the novelties of the model.

Section snippets

Implementation of the hyperspectral capabilities

By doing hyperspectral computations for a number of users-defined spectral bands, it is possible to generate the sky brightness maps and spectra for any combination of lamp technologies. For each pixel, one can define the net spectrum and the net angular photometry describing the lighting infrastructure. This makes it easy to quickly see what would happen to the night sky spectra or to any of its integrations (e.g. scotopic or photopic integration) if the lighting technology is changed over a

Correction for subgrid obstacles

The subgrid obstacles are defined by three numbers for each pixel of the simulation domain, as seen in Fig. 1. These include both the averaged distance between the light source and the obstacles, and the averaged obstacle height. When combined with the averaged height of the light sources, also defined independently for each pixel of the domain, it allows for the determination of which light rays leaving the source, scattered by the atmosphere, or reflected by the ground will be blocked by

Scattering by overhead clouds

To incorporate clouds in the already existing model, we need to determine the clouds base height and assume that no light can penetrate through, making it the effective maximum height of the atmosphere. We also need to define an hemispherical reflectance function for the cloud base surface which will depend on the cloud type. The hemispherical reflectance of different types of clouds has already been studied long ago by Shapiro [7].

The parametrization proposed by Shapiro [7] is quite simple as

Applications for peri-urban and rural sites

In this section we will show some effects of the newly added features to the model for two different geographical cases, both located in the province of Québec in Canada. The first one, Sanctuaire de Beauvoir, is a site located at about 3 km from the edge of the city of Sherbrooke, Québec, Canada. The second one, Mont-Mégantic Observatory (MMO), is located on top of Mount Mégantic about 60 km away from Sherbrooke. Both sites lie inside the first international dark sky reserve of Mont-Mégantic.

Conclusion

This paper aimed to introduce to the scientific community three important new features of the sky brightness model Illumina. These features are opening new ranges of applications. We presented the results of a modeling experiment involving the variation of the related parameters for two very different contexts (rural and peri-urban) to show the novel possibilities associated with the new features added to Illumina.

Among the most important results we found were that the effects of Rayleigh

Acknowledgments

We applied the sequence-determines-credit approach for the sequence of authors. This work was supported by the Fond Québécois pour la Recherche sur la Nature et les Technologie (FQRNT). Computation time on Mammouth serial II was provided by Compute Canada and Calcul Québec.

References (14)

B. Holben et al.
An emerging ground-based aerosol climatology: aerosol optical depth from aeronet
J Geophys Res
(2001)
M. Aubé et al.
Light pollution modelling and detection in a heterogeneous environment: toward a night-time aerosol optical depth retreival method
Optics & Photonics 2005
(2005)
M. Aubé
Light pollution modeling and detection in a heterogeneous environment
Proceedings of Starlight 2007 conference, La Palma, Spain
(2007)
M. Aubé
Physical behaviour of anthropogenic light propagation into the nocturnal environment
Philos Trans R Soc Lond B
(2015)
Aubé M., Simoneau A.. Illumina users guide. 2017....
C.D. Elvidge et al.
Viirs night-time lights
Int J Remote Sens
(2017)
Y.J. Kaufman et al.
Algorithm for remote sensing of tropospheric aerosol from modis
NASA MODIS Algorithm Theoretical Basis Document
(1998)

There are more references available in the full text version of this article.

Cited by (38)

Assessing light pollution in vast areas: Zenith sky brightness maps of Catalonia
2023, Journal of Quantitative Spectroscopy and Radiative Transfer
Zenith sky brightness maps in the V and B bands of the region of Catalonia are presented in this paper. For creating them we have used the light pollution numerical model Illumina v2. The maps have a sampling of $5 \times 5$ km for the whole region with an improved resolution of $1 \times 1$ km for one of the provinces within Catalonia, Tarragona. Before creating the final maps, the methodology was tested successfully by comparing the computed values to measurements in nineteen different locations spread out throughout the territory. The resulting maps have been compared to the zenith sky brightness world atlas and also to Sky Quality Meter (SQM) dynamic measurements. When comparing to measurements we found small differences mainly due to mismatching in the location of the points studied, and also due to differences in the natural sky brightness and atmospheric content. In the comparison to the world atlas some differences were expected as we are taking into account the blocking effect of topography and obstacles, and also due to a more precise light sources characterization. The results of this work confirm the conclusion found in other studies that the minimum sampling for studying sky brightness fine details is of $1 \times 1$ km. However, a sampling of $5 \times 5$ km is interesting when studying general trends, mainly for vast areas, due to the reduction of the time required to create the maps.
HABLAN: Multispectral and multiangular remote sensing of artificial light at night from high altitude balloons
2023, Journal of Quantitative Spectroscopy and Radiative Transfer
Some studies have established statistical links between Artificial Light At Night (ALAN) and the risk associated with diseases such as hormone-dependent Cancers (HDC). These links have been identified in part under a yellowish lighting regime and with space-borne images that only detect light travelling upward and that are often blind to the blue or the shorter wavelengths of the visible spectrum. The use of colour images taken by astronauts at the International Space Station (ISS) can detect the blue component of ALAN but it is still restricted to the upward light emissions. It may be expected that the light travelling near horizontal is even more correlated to HDC. Therefore, it is essential to develop multispectral and multiangular remote sensing techniques of ALAN. To achieve this, we suggested a new experiment to observe ALAN from the stratosphere during High Altitude Balloon (HAB) flights. The system use multiple colour cameras to picture a city at different viewing angles and in the three RGB bands (Red, Green and Blue bands). The ultimate goal of the project is to provide better remote sensing data as inputs to numerical radiative transfer models. The model may then be used to infer health risks associated with ALAN. Stratospheric remote sensing makes it possible to measure large areas in a short period of time at low cost compared to space-borne methods. The system High Altitude Balloon Light At Night (HABLAN) was designed to achieve that goal. The first HABLAN experiments were carried out aboard balloon flights of the Canadian Space Agency (CSA) as part of their STRATOS program. In this paper, we present the experimental concept of the HABLAN system and we show some preliminary results.
Controlling the artificial radiance of the night sky: The Añora urban laboratory
2023, Journal of Quantitative Spectroscopy and Radiative Transfer
Citation Excerpt :
Modeling the artificial night sky brightness at medium to long distances from the sources is nowadays accomplished by using different point-spread functions (PSF) that describe the artificial brightness dependence on the source and detector properties, the state of the atmosphere, and the distance to the observer. The interested reader may consult among others the works of Aubé [2], Aubé and Simoneau [3], Bará et al. [4,5,9], Cinzano et al. [16], Cinzano and Falchi [17,18], Duriscoe [20], Falchi et al. [24], Falchi and Bará [26], Garstang [28–30], Kocifaj [35–37], and Simoneau et al. [54] for further details. The basic information about the spatial distribution of the sources and some of their emission properties is available in repositories of nighttime satellite imagery [38–41] as e.g. those of the VIIRS-DNB radiometers onboard the Suomi-NPP and NOAA-20 satellites [21–23, 51,58], the Luojia-01 [34], and the Crew Earth Observation program of the International Space Station [52,53,55].
We provide radiometric evidence of the relevance of nearby light sources on the artificial brightness of the night sky. To obtain the required data we developed a method based on the use of power-regulated urban lighting systems, which also provides relevant information on the propagation of light pollution at short distances from the sources. A controlled experiment was carried out in the Andalusian municipality of Añora (1526 inhab.) in which the radiant power of its regulated public outdoor lighting system was modified in a predefined way during the central part of the night while continuously recording the zenith sky brightness in the TESS-W photometric band from two separate locations inside the town. We determined that the power-regulated streetlight sources contribute a 60–62% to the total zenith sky brightness at these observing locations in clear and moonless nights when operated at their maximum power rating, being the remaining sky radiance due to constant local sources and sources from neighboring towns. These results, in combination with georeferenced information on the sources' location and properties, impose some constraints on the functional form of the effective point-spread functions (PSF) of the zenith sky brightness at short distances from the sources. For the conditions of our experiment we have found that the expected exponents of power-law PSFs are close to -1.
Keeping light pollution at bay: A red-lines, target values, top-down approach
2021, Environmental Challenges
Citation Excerpt :
This means that every pixel of the territory contributes to the artificial sky brightness in proportion to its absolute radiance emissions, being the constant of proportionality dependent on the angular emission pattern of the light sources, their spectral radiant density, the spectral reflectance of the pavements, the presence of obstacles blocking the propagation of light along some set of rays, the composition and concentration profiles of the molecular and aerosol atmosphere, and, of course, on the distance from the emitting sources to the observation point. A wide set of models for characterizing this propagation are available in the literature (Aubé, 2015; Aubé and Simoneau, 2018 Aubé et al., 2020; Cinzano and Falchi, 2012; Garstang, 1986, 1989, 1991; Kocifaj, 2007, M. 2016, M. 2018; Kocifaj and Bará, 2019). They basically provide the light pollution point spread function (PSF), that is the contribution of a unit radiance source to the sky radiance as a function of the distance and wavelength (with the remaining variables mentioned above acting as parameters of the model).
The prevailing regulatory framework for light pollution control is based on establishing conditions on individual light sources or single installations (regarding features like ULOR, spectrum, illuminance levels, glare, …), in the hope that an ensemble of individually correct lighting installations will be effective to somehow solve this problem. This "local sources" approach is indeed necessary, and shall no doubt be enforced; however, it seems to be clearly insufficient for curbing the actual process of degradation of the night, and for effectively attaining the necessary remediation goals. In this paper we describe a complementary (not substitutive) 'red-lines' strategy that should in our opinion be adopted as early as possible in the policies for light pollution control. It is based on setting maximum values for absolute light pollution indicators and using linear models relating the indicators to the source emissions in order to establish the maximum light emissions compatible with these red-lines. This top-down approach seeks to set definite limits on the allowable degradation of the night, providing the methodological tools required for making science-informed public policy decisions and for managing the transition processes. Light pollution abatement should routinely be included as an integral part of any territorial management plan. A practical application case-study based on the night sky brightness at zenith is described to illustrate these concepts.
Estimating the loss and fragmentation of dark environments in mammal ranges from light pollution
2021, Biological Conservation
A hallmark of the Anthropocene is the global expansion of pollution stemming from electric lighting. This evolutionarily novel phenomenon has left few spaces on Earth where natural light cycles remain unaltered. Assessing the exposure of species to light pollution is critical for developing conservation plans that address this expanding sensory pollutant. Here we used data from NASA's Visible Infrared Imaging Radiometer Suite to determine the contribution of full moonlight radiance to establish an ecologically-informed threshold of natural radiance. Using the geographic ranges of 351 mammals across the contiguous United States, we estimated the percentage of each species' range in which light pollution exceeded thresholds associated with direct emissions and skyglow, where it did not (“dark environment”), and the associated fragmentation of dark environments. Average mammal range contained 2.6% (95% CI: 2.3–3.2%) of area that consistently exceeded full moon radiance at point-source emissions, but with a large range among species (0–47.4%). Skyglow affected far greater percentages of ranges ( $\bar{X} =$ 24.3%; 95% CI: 22.1–26.8%). Nocturnal species had slightly greater exposure than diurnal species. Several families with the most impacted ranges included species of conservation concern (e.g., Molossidae; free-tailed bats). When assessing connectivity of dark environments, we found light pollution fragmented most mammal ranges and resulted in isolated dark refugia (e.g., 154% average increase in patches of dark environments). Identifying species with the greatest exposure to and dark environment fragmentation from light pollution is an important step for targeted conservation efforts of remaining dark refugia for light-sensitive species.
The impact of atmospheric aerosol particles on the brightness of the night sky
2020, Journal of Quantitative Spectroscopy and Radiative Transfer
Citation Excerpt :
Its features represent major improvements to previous light pollution models. ILLUMINA was used in a number of later publications analysing light scattering in the atmosphere at specific measurement points [22-25]. In 2016, Kocifaj and Komar presented a model of light scattering on atmospheric aerosols that allows for determination of their role in creating the sky glow, as well as its impact on suburban areas [26].
This paper analyses the impact of various types of aerosols, both of natural and anthropogenic origin, on the brightness of the night sky glow in southern Poland. The particles of particulate matter, related to the combustion of solid fuels in the winter, the volcanic or desert dust, as well as mists and haze, are considered as the artificial light scattering sources. Measurements of the brightness of the cloudless and moonless night sky were done in 2009–2016, both within the city of Krakow and in suburban areas, as well as in mountainous ones. The strong linear correlation between the brightness of such sky and the concentration of particulate matter is shown. The acoustic sounding of the atmosphere (SODAR) has indicated the possibility of a relationship between the brightness of the night sky and the amount of such particulates, which accumulate in atmospheric boundary layers. The usefulness of the theoretical model of horizontal transport of dust in the atmosphere (FAPPS) for forecasting the brightness of the night sky glow is also pointed out. A clear effect of the Saharan origin dust clouds on the brightness of the night sky glow is shown. This brightness, in the conditions of a low level of light pollution, is associated with the forecasted optical density of such clouds. It is also demonstrated, that with the thickening of mist, the impact of distant light sources on the brightness of the night sky decreases, but the one of a nearby sources becomes more significant. The conclusion states that anthropogenic particulate matter has the greatest impact on the brightness of the cloudless night sky glow in winter. In areas heavily polluted with light, fogs and mist are particularly important. In areas with low levels of light pollution, the clear impact of desert dust is visible.

View all citing articles on Scopus

View full text

New features to the night sky radiance model illumina: Hyperspectral support, improved obstacles and cloud reflection

Highlights

Abstract

Introduction

Section snippets

Implementation of the hyperspectral capabilities

Correction for subgrid obstacles

Scattering by overhead clouds

Applications for peri-urban and rural sites

Conclusion

Acknowledgments

J Geophys Res

Light pollution modelling and detection in a heterogeneous environment: toward a night-time aerosol optical depth retreival method

Optics & Photonics 2005

Light pollution modeling and detection in a heterogeneous environment

Proceedings of Starlight 2007 conference, La Palma, Spain

Physical behaviour of anthropogenic light propagation into the nocturnal environment

Philos Trans R Soc Lond B

Viirs night-time lights

Int J Remote Sens

Algorithm for remote sensing of tropospheric aerosol from modis

NASA MODIS Algorithm Theoretical Basis Document