NITESat: A High Resolution, Full-Co= lor, Light Pollution Imaging Satellite Mission
Ken Walczak1<=
/span>,*,
[1] 1Adler Planetarium, Chicago, Illinois 60605, USA
2City Colleges of Chicago, Chicago, IL USA
Received
16 December 2016, Accepted 08 May 2017
Abstract
The NITESat (Night Imaging and Tracking
Experiment Satellite) mission is a 2U CubeSat satellite designed for nightt=
ime
Earth imaging to quantify and characterize light pollution across the
Midwestern United States. It is accompanied and supported by an array of
ground-based light pollution observing stations called GONet (Ground Observ=
ing
Network). NITESat is a pilot mission testing the potential for a simple and
inexpensive (<$500,000) satellite to deliver high-resolution, three-color
regional data of artificial light at night. In addition, GONet will form the
core of an educational outreach program by establishing an array of all-sky=
monitors
covering the imaging region o=
f the
satellite with 20+ full sky light pollution citizen-operated stations. This
will provide synchronized data coinciding with the NITESat overpasses as we=
ll
as providing near continuous night sky quality monitoring. If the initial
mission is a success, the potential exists to expand the program into a low
cost constellation of satellites capable of delivering global coverage. NITESat is being designed, built a=
nd
will be operated by the Far Horizons program at the Adler Planetarium in
Chicago, Illinois. Far Horizons is a student and volunteer centered program
offering hands-on engineering and scientific research opportunities for
education.
Keywords: NITESat, GONet, CubeSat, Remote sensin=
g,
Satellite, Light pollution
1. Background
Light Pollution (LP) by its simplest definition =
is
the contamination of the natural nighttime environment by light of
anthropomorphic origin. Many studies over the past decades have shined a li=
ght
on the impacts and costs of light pollution. Much of the valuable data for
light pollution research beyond local measurements has come from orbital
satellite missions.
Studies have shown that excessive LP can adverse=
ly
impact human health [1], animal behavior and even entire ecological systems=
[2]
among other negative impacts. It is estimated 60% of the residents of Europe
and 80% of North American cannot see the Milky Way from their homes due to
light pollution [3]. This not only reduces the innate understanding and
connection we have to the Universe but it robs us of context of our place
within it.
Light pollution can also have economic implicati=
ons:
artificial light reaching space is wasted energy. The financial cost of this
waste can be quantified by careful observations from space. As an example, =
an
analysis of 2008 data shows that due to poor lighting practices and antiqua=
ted
fixtures Chicago, a city of 2.7 million, spends an estimated 30% ($5,400,000
USD) of its annual $18M USD public lighting budget solely in light
ineffectively sent up into the atmosphere [4]. This estimate is based on an
inventory of Chicago lighting fixtures and assumed efficiencies of those
fixtures. Space-based observations could help give a more precise figure of
those assumptions by measuring actual flux. Beyond the obvious financial bu=
rden
created by this inefficiency, the energy required to generate this light - =
in
most cases by the burning of fossil fuels - has direct and avoidable
environmental repercussions.
As awareness grows, preconceptions can begin to
fade. For example, much of the public believes the quantity of light
illuminating their nighttime environment correlates directly with safety. At
least one study has cast doubt on this assumption [5]. As researchers uncov=
er
relationships between LP and our environment the public is, in turn, gainin=
g an
awareness of its impact on their lives. Much of the growth in knowledge of =
the
effects of light pollution on biology, health, economy and ecology can be
attributed to the availability of large data sets quantifying LP from orbit=
al
satellite missions. The information provided by these large government-fund=
ed
missions has given researchers an opportunity to understand LP on a global =
and
regional level by quantifying LP emission.
Since 1973 the Optical Linescan System (OLS) ins=
trument
has been flown on board a continuous series of U.S. Air Force Defense
Meteorological Satellite Program (DMSP) missions. Once the data products fr=
om
the OLS were made publicly available in 1992, those interested in LP resear=
ch
for the first time benefited from full global maps of LP. The OLS data help=
ed
expand the field of LP research [6]. Although the OLS data provided much
valuable information, there are significant limitations for researchers. Am=
ong
the limitations from the OLS instrument are; a limited 5km/px resolution, a
less than favorable 19:30 local time overpass time, oversaturation in urban
areas, single panchromatic bandpass and no calibration [7].
Beginning at the end of 2011, the Suomi National
Polar-orbiting Partnership (Suomi-NPP) satellite was launched. It carries t=
he
Visible Infrared Imaging Radiometer Suite (VIIRS) instrument. The VIIRS Day
Night Band (DNB) data has addressed some of the shortcomings of the OLS dat=
a.
Although the data provided by the Suomi-NPP satellite has increased
researchers¡¯ ability to quantify LP, limitations still remain. Some of the=
se
limitations include: 742m/px resolution, a late 1:30AM local time overpass
time, a single panchromatic bandpass imagery [8] and a ¡°blue blindness¡± a=
bove
500nm that fails to detect significant emission from many styles of outdoor=
LED
lighting [3].
2. Mission Overview and Goals=
The
Night Imaging and Tracking Experiment Satellite (NITESat) mission aims to
improve the quality of LP data available from orbit. The mission consists of
two primary components: NITESat and GONet. NITEsat is an orbital nanosatell=
ite
with a sensitive, high resolution, RGB imaging system as a primary payload.=
The
Ground Observing Network (GONet) is an array of ground-based, all-sky imagi=
ng
stations. Both systems work in tandem to validate and support the data prod=
ucts
produced by the mission. NITESat and GONet have been specifically designed =
to
address many of the shortcoming of presently available on-orbit data as wel=
l as
exploring the future of LP observations with inexpensive nanosatellites.
To
achieve this objective, the design of the NITESat mission is driven by the
following goals:
2.1. Produc=
e a
complete regional map of LP
The
NITESat mission will produce a complete map of LP across a significantly la=
rge
region centered on the city of Chicago. The initial mission goal is to imag=
e a
1000x1000km area centered on Chicago (see Figure 1) multiple times througho=
ut
the mission lifetime of one to two years. Centering on Chicago (where the
mission is being designed and operated) will allow regional coverage of a
diverse range of LP affected environments from dense urban centers through =
to
relatively unaffected unpopulated areas. An additional advantage of imaging
this area is that it contains a large expanse of relatively pristine dark
locations over the two Great Lakes (Michigan and Huron). This allows the
opportunity to use it as a low-level calibration source. The size and locat=
ion
of this area also allows realistic and effective distribution and maintenan=
ce of
the GONet system for on-orbit to ground correlation.
The
primary need to limit the imaging footprint is due to the low downlink capa=
city
of the single Ground Station based at the Adler Planetarium. Mission
performance studies suggest that the addition of other Ground Stations run =
by
partnering institutions across the world, will allow expansion of the imagi=
ng
opportunities in nearly direct proportion to the available downlink capacity
(up to a maximum of 10-20 fold increase).
Fig. 1. Light pollution
intensity map for the 1000x1000km NITESat imaging area centered on Chicago.
This region covers 4 orders of magnitude in population density from rural to
one of the world¡¯s most light-polluted cities. The Great Lakes provide are=
as of
contrasting zero population. Image credit: F. Falchi, et al 2016 ADDIN CSL_CITAT=
ION {
"citationItems" : [ { "id" : "ITEM-1",
"itemData" : { "DOI" : "10.1126/sciadv.1600377&quo=
t;,
"ISSN" : "2375-2548", "author" : [ {
"dropping-particle" : "", "family" :
"Falchi", "given" : "Fabio",
"non-dropping-particle" : "", "parse-names" :
false, "suffix" : "" }, { "dropping-particle"=
:
"", "family" : "Cinzano", "given" :
"Pierantonio", "non-dropping-particle" : "",
"parse-names" : false, "suffix" : "" }, {
"dropping-particle" : "", "family" :
"Duriscoe", "given" : "Dan",
"non-dropping-particle" : "", "parse-names" :
false, "suffix" : "" }, { "dropping-particle"=
:
"", "family" : "Kyba", "given" :
"Christopher C M", "non-dropping-particle" : "&quo=
t;,
"parse-names" : false, "suffix" : "" }, {
"dropping-particle" : "", "family" :
"Elvidge", "given" : "Christopher D",
"non-dropping-particle" : "", "parse-names" :
false, "suffix" : "" }, { "dropping-particle"=
:
"", "family" : "Baugh", "given" :
"Kimberly", "non-dropping-particle" : "", &qu=
ot;parse-names"
: false, "suffix" : "" }, { "dropping-particle&quo=
t; :
"", "family" : "Portnov", "given" :
"Boris A", "non-dropping-particle" : "",
"parse-names" : false, "suffix" : "" }, {
"dropping-particle" : "", "family" :
"Rybnikova", "given" : "Nataliya A", "no=
n-dropping-particle"
: "", "parse-names" : false, "suffix" :
"" }, { "dropping-particle" : "",
"family" : "Furgoni", "given" :
"Riccardo", "non-dropping-particle" : "",
"parse-names" : false, "suffix" : "" } ],
"container-title" : "Science Advances", "id" :
"ITEM-1", "issue" : "6", "issued" :=
{
"date-parts" : [ [ "2016", "6", "10"=
; ]
] }, "page" : "e1600377-e1600377", "title" :
"The new world atlas of artificial night sky brightness",
"type" : "article-journal", "volume" :
"2" }, "uris" : [ "http://www.mendeley.com/documen=
ts/?uuid=3D16e4a7aa-8919-46ae-bd2e-1fb8f3436958"
] } ], "mendeley" : { "formattedCitation" :
"[3]", "plainTextFormattedCitation" : "[3]" },
"properties" : { "noteIndex" : 0 }, "schema" :
"https://github.com/citation-style-language/schema/raw/master/csl-cita=
tion.json"
}[3]
2.2. Collect
color information to distinguish LP sources
NITESat
imagery will provide color information sufficient enough to distinguish bet=
ween
major types of lighting used in municipal and commercial nighttime settings.
NITESat will be equipped with an RGB imaging system. With this data, major
types of light sources (ex., HPS, MV and most LED) will be distinguishable =
from
each other [9].
Surveys of specific artificial lighting environments across the imaging
footprint will be achievable. Direct analysis of comparable communities
utilizing different types of artificial light can also be made.
2.3. Improv=
e on
the resolution from VIIRS data
Improved
resolution compared to VIIRS data will add to our understanding of LP sourc=
es
and effects on a smaller, more defined scale. The present design of the mis=
sion
will achieve a GSD resolution of 100m/px, a more than 7-fold increase over =
the
VIIRS data. At this resolution, small-scale variations of artificial lighti=
ng
can be distinguished. A simulation of the expected results combining resolu=
tion
and RGB information of NITESat versus VIIRS data can be seen in Fig. 2
2.4. Provide
synchronized ground-to-orbit data
The
NITESat mission will be flown in conjunction with an array of ground-based =
observing
stations. These GONet stations will provide ground-truthing for the on-orbit
NITESat data. By synchronizing the GONet images precisely with the timing of
NITESat overpasses, ground-to-orbit correlations can be created. This data =
can
be valuable in modeling atmospheric scatter, aerosol distribution and other
parameters important for an understanding of LP.. Also, select GONet nodes will=
be
equipped with a high-intensity, calibrated light source to be triggered on
select NITESat overpasses. These sources will be designed to perform absolu=
te
calibration for the NITESat imaging system.
2.5. Capture
time-relevant data
The
mission¡¯s orbit will provide varied overpass times across the imaging
footprint. This will allow the collection of night luminance data across va=
ried
times of the night. As night progresses many towns and businesses begin tur=
ning
off or reducing the amount of lights they use. To capture an accurate
understanding of the impact of LP, the optimal time of data collection is
before 10pm local time when peak lighting can be observed [8].
Due to its varied overpass times, NITESat will be able to image areas during
the height of nighttime illumination as well as being able to distinguish
patterns in lighting change across night time hours. Additionally, NITESat =
is
expected to be operational for more than a year. This will allow the analys=
is
of seasonal changes or perhaps even monitoring the transitional changes in
communities or businesses that change their lighting fixtures or lighting u=
se
for better or worse.
2.6.
Educational outreach and engagement
A primary goal of the NITESat and GONet mission beyond research is educationa= l. The Far Horizons program at the Adler Planetarium has engaged students and volunteers for over a decade with hands-on scientific and engineering proje= cts. NITESat and GONet will leverage that experience in educational programming = and training to facilitate the design, testing and operations of the mission. <= o:p>
Additionally,
to implement the GONet array a program of LP education and awareness will be
used as the basis of training for schools, clubs and communities who host t=
he
GONet stations. This offers the opportunity to not only have the public
intimately involved in the science of the mission, but these same participa=
nts
can help raise the profile of LP awareness in their communities. This public
investment of effort can potentially turn into public advocacy for smart
lighting practices. One could imagine a scenario such as students at a rural
school, who host a GONet station, coordinating a town-wide ¡°Lights Out Nig=
ht¡±
to compare NITESat and GONet data and demonstrate the effects of different
lighting practices.
3. NITESat Design
NITESat
will be a 2U (10x10x20cm) CubeSat based on a commercially available (GOMSpa=
ce)
bus. It will include Commercial Off The Shelf (COTS) power, communications,
attitude control and flight computer modules. The main payload will be a
low-light capable imager based on the three color (RGB) PCO Edge 3.1 sCMOS
camera and a fast (FL=3D1.4) 25mm Lensation lens.
During
each favorable night-time overpass, multiple rapid (20ms exposure at a 0.1s
cadence) images will be taken and stacked by an on-board processor. Stacked
images will be stored and downlinked in losslessly compressed form. The tar=
get
performance is given in Table 1.
Table 1
|
Resolution |
Bands |
Sensitivity |
FOV |
Dynamic Range |
|
|
|
|
|
|
NITESat
will orbit at an altitude of approximately 500km with an inclination of 50¡=
Æ.
Even accounting for cloud cover averages, our target region should be fully
imaged multiple times in a year. In our reference orbit, imaging overpasses=
in
the optimal twilight to midnight time frame occur in clusters lasting
approximately 1 month separated by a 1 month period without imaging
opportunities. This cadence allows the downlink of stored data during lulls=
in
the imaging portion of the mission. The expected orbital lifetime of 1-2 ye=
ars
will allow the tracing of seasonal variations of the emission of artificial
light at night.
4. GONet Design
The
Ground Observing Network (GONet) is a complement to the satellite mission as
well as an independent means of valuable data collection about localized li=
ght
pollution. A primary component of the network is the GONet imager, which
consists of a 5.0 Megapixel Point Grey Blackfly GigE Aptina MT9P006 camera
equipped with a fisheye lens capable of capturing full 2𝜋
sky images. The GONet imager is controlled using a Raspberry Pi 3 with
images taken at regular intervals and processed for subsequent analysis.
The
goal is to design a system requiring as little maintenance and the greatest
ease of use as possible while still providing significant, useful data. The
optimal system will be as autonomous as possible, allowing installation in
relatively remote locations where power and data connectivity will not be
provided. With these design goals some of the baseline features planned for=
the
final instrument include:
¡¤
GPS for timing and location
data
¡¤
Magnetometer for absolute
orientation knowledge
¡¤
Weatherproof enclosure with
automated articulated cover to reduce cleaning of the imaging aperture
¡¤
Solar charging power supply
for long-term function with minimal maintenance
¡¤
Bluetooth connectivity to
allow the easy upload of imaging schedules and download of stored data
Calibration
is still a concern to be addressed. We are presently exploring a variety of
approaches to calibrate the system with hardware and or software solutions.
There is substantial research to inform potential solutions specifically for
all sky monitoring systems using fisheye lenses [10]–[12].
In the current design iteration, the imaging routine includes the automatic
capturing of calibration data such as dark frames and identification of sta=
rs
with known brightness and sky positions. Figure 3
shows the current optical and electronic system. Development of proper
packaging and field hardening is underway.
Fig. 3. Second GONet
prototype used during September 10, 2016 test observations
5. Preliminary Testing and
Results
An
extensive qualifying program is currently underway using stratospheric high
altitude balloon (HAB) flights and laboratory tests to ensure the main payl=
oad
meets required specifications, is space-ready and both the hardware and
software are fully integrated with the spacecraft bus. These tests used a
representative imaging system, a Point Grey Grasshopper 2 with a 25mm f/1.4
lens, to confirm resolution and sensitivity assumptions. The flights also s=
erve
as a preliminary test of the mission operations.
Using
HAB flights, we were able to conduct two validation tests with representati=
ve
model hardware in 2016. A prototype GONet system was set up in a remote and
relatively dark location along the predicted path of the HAB flight. For the
first test flight, the GONet station used a Raspberry Pi 2 and was equipped
with a Point Grey Firefly MV 13S2C-CS, a fisheye lens, GPS, and a calibrated
light source to help confirm the sensitivity and resolution of the in-flight
imaging system (see Fig. 4).
The light source was periodically covered when GONet took images by using a
stepper motor to place a 3d printed cap over it, and a luminosity sensor in=
the
cover was used to measure the intensity of the light source.
On
the second test flight, the GONet Firefly camera was replaced by the Blackf=
ly
camera because of better S/N with long exposures needed under dark conditio=
ns.
Both hardware and software improvements were made to increase the accuracy =
of
data collection and system efficiency. A magnetometer was added to de=
tect
the orientation of the images, and a python script was created to automate
magnetometer calibration which can be quickly performed at the observing si=
te.
Initial test results have indicated the system can collect data able to meet
the mission requirements. (see Fig. 5)
Fig. 4. Results of first =
HAB
NITELITE/GONet test flight showing flight path and select the images from
flight. Inset: Identification of the GONet calibration light in an image fr=
om
an altitude of 28 kilometers.
Fig.
5
The GONet test results have
helped guide the design of the next version of the system including
optimization of the exposure range, addition of in-image metadata and softw=
are
improvements for simplification. Information from the test flights
resulted in a third iteration of the software package. The latest
software update makes the GONet a fully autonomous system capable of
significantly reducing noise, determining accurate brightness measurements =
of
light sources, and producing both quantitative and qualitative visualizatio=
ns
of statistical data.
6. Conclusion
Small,
single-purposed nanosatellites (CubeSats) have the potential to provide
valuable data for light pollution research. The NITESat mission has been de=
signed
to take advantage of the relative low cost and simplicity of the CubeSat fo=
rm
factor while carefully maintaining the quality of the data collected. The
mission requirements of NITESat and GONet have been specifically generated =
to
demonstrate the feasibility of missions of this scale for research.
The
Far Horizon team intends to make its data publically available in reduced f=
orm.
We are also interested in collaboration with researchers in the light pollu=
tion
community, education outreach professionals and engineers to help optimize =
our
mission design and provide the most valuable data to best serve the communi=
ty¡¯s
goals. Partnerships with groups interested in providing additional downlink
services, i.e., communication Ground Stations, may allow additional imaging
opportunities beyond the mission target region.
References
[1]
Blask, D. E., Brainard, G. C., Dauchy, R. T., Hanifin, J. P., Davidson, L. =
K.,
Krause, J. A., Sauer, L. A., Rivera-Bermudez, M. A., Dubocovich,M. L., Jass=
er,
S. A., Lynch, D. T., Rollag, M. D., & Zalatan=
F.
(2005). Melatonin-depleted blood from premenopausal women exposed to light =
at
night stimulates growth of human breast cancer xenografts in nude rats. Can=
cer
research, 65(23), 11174-11184.
[2] Longcore, T., & Rich, C. (2004). Ecol=
ogical
light pollution. Frontiers in Ecology and the Environment, 2(4), 191-198.
[3] Falchi, F., Cinzano, P., Duriscoe, D., Ky=
ba, C.
C. M., Elvidge, C. D., Baugh,K., Portnov,B. A., Rybnikova, N. A., &
Furgoni, R. (2016). The new world atlas of artificial night sky brightness.
Science advances, 2(6), e1600377.
[4] Carhart, D. (2016). Chicago: A Study in O=
utdoor
Lighting Practices and Costs. illinoislighting.org, Online available: http:=
//www.illinoislighting.org/chicago.html.
(Accessed: 23-Nov-2016).
[5] Steinbach, R., Perkins, C., Tompson,L.,
Johnson, S., Armstrong, B., Green, J., Grundy, C., Wilkinson, P., &
Edwards, P. (2015). The effect of reduced street lighting on road casualties
and crime in England and Wales: controlled interrupted time series analysis.
Journal of epidemiology and community health, jech-2015.
[6] Elvidge, C. D., Erwin, E. H., Baugh, K. E=
.,
Ziskin, D., Tuttle, B. T., Ghosh, T., & Sutton, P. C. (2009, May). Over=
view
of DMSP nightime lights and future possibilities. In Urban remote sensing
event, 2009 Joint (pp. 1-5). IEEE.
[7] Huang, Q., Yang, X., Gao, B., Yang, Y., &=
amp;
Zhao, Y. (2014). Application of DMSP/OLS nighttime light images: A
meta-analysis and a systematic literature review. Remote Sensing, 6(8),
6844-6866.
[8] Elvidge, C. D., Baugh, K. E., Zhizhin, M.,
& Hsu, F. C. (2013). Why VIIRS data are superior to DMSP for mapping
nighttime lights. Proceedings of the Asia-Pacific Advanced Network, 35, 62-=
69.
[9] Elvidge, C. D., Keith, D. M., Tuttle, B. =
T.,
& Baugh, K. E. (2010). Spectral identification of lighting type and
character. Sensors, 10(4), 3961-3988.
[10]Rabaza,
O., Aznar-Dols, F., Mercado-Vargas, M. J., & Espín-Estrella, A. =
(2014). A
new method of measuring and monitoring light pollution in the night sky.
Lighting Research & Technology, 46(1), 5-19.
[11] Nievas Rosillo, M. (2013). Absolute photome=
try
and Night Sky Brightness with all-sky cameras. Máster Interuniversit=
ario en
Astrofísica UCM-UAM.
[12] Román, R., Antón, M., Cazorla=
, A., De Miguel,
A., Olmo, F. J., Bilbao, J., & Alados-Arboledas, L. (2012). Calibration=
of
an all-sky camera for obtaining sky radiance at three wavelengths. Atmosphe=
ric
Measurement Techniques, 5(8), 2013. –2024.
K. Walczak et al. /International Journal of Sustainab=
le
Lighting 20 (2017) 48-55