An Upgrade to TAWS Urban Illumination Model Richard Shirkey Army Research Laboratory Computational & Information Sciences Directorate Battlefield Environment Division WSMR, NM richard.shirkey@us.army.mil Ph: (575) 678-5470 FAX: (575) 678-4449 WIDA 2012 13-15 March Approved For Public Release; Distribution Unlimited OUTLINE • Purpose • Beginnings • Walker • Treanor • Objective • Garstang’s improvements • AIM v2 • Spectral Radiance • V&V 2 Approved For Public Release; Distribution Unlimited Purpose To provide urban illumination levels for use in/for • Simulations • IWARS • Target Acquisition • TAWS • Night Vision Devices Approved For Public Release; Distribution Unlimited Beginnings: Walker’s Model Population x 106 Walker took sky brightness measurements of various cities to determine the distance where the artificial sky brightness was 0.1 mag at the zenith. From this he constructed the first Population-Distance relation. 3.0 2.0 1.0 20 40 60 80 Distance (miles) Garstang reformulated Walker’s P-D relation as B = CPD-2.5. Known as Walker’s law, it is the simplest model for light pollution. C does not depend on P and D, but depends on factors such as the light emission per head of the population and the reflectivity of the ground 4 Approved For Public Release; Distribution Unlimited 100 Beginnings: Treanor’s Model Treanor proposed a simple model for scattering along the direct beam (TZ) by restricting scattering to a small angle m. This restricts scattering to the particles which are contained within the figure of revolution generated by rotating the arc TQZ about TZ. This scattered light then undergoes a final scattering at Z into the observer’s LOS. when 5 ' > m scattering is not considered Approved For Public Release; Distribution Unlimited Objective criterion • A quick running model that accurately portrays both broadband and spectral illumination received from urban sources by an arbitrarily placed observer • point source • observer 2.25 x city radius • extended source • observer 2.25 x city radius • internal to the city • Easily obtainable inputs • Atmospheric variables • aerosols • clouds : acceptable error D: distance to source R: radius of source E: irradiance L': radiance Eex = πL'R 2 / (R 2 + D 2) Ept = πL'R 2 / D 2 = (Ept – Eex)/Eex = R2/D2 D R -½ ( 20% D 2.25 R) Approved For Public Release; Distribution Unlimited Garstang’s Model Assumptions for Garstang’s model are: • Flat Earth • All downward-directed light from fixtures does not undergo scattering or absorption wrt the atmosphere © International Dark-Sky Association • No interaction with buildings or vegetation • Aerosol scattering is predominantly in the forward direction 7 Approved For Public Release; Distribution Unlimited Garstang’s Model Garstang further modified Treanor’s model by introducing an exponential atmosphere and directly relating the atmospheric molecular content to the aerosol content Na a = 11.11 Nm m K e-cH where N is the number density, is the particle cross section, H is the height of the city and K is Garstang’s “clarity factor”. 8 Approved For Public Release; Distribution Unlimited Garstang’s Model The city emission function is defined by I = LP/2 {2G (1-F) cos + 0.554 F 4} City Emission Function where I is the upward intensity, F is the fraction of light that “escapes” the luminaries shielding in an upward direction, G is the surface albedo and is the angle between the upward normal and an upward light ray. Cinzano has found that the shape of the emission function is not important 9 Approved For Public Release; Distribution Unlimited “Double Scattering” General Scenario Final Equations Brightness Small Angle Approximation 10 Approved For Public Release; Distribution Unlimited Garstang’s Model: Details Broadband Brightness in Garstang’s model is represented by b = Nm R e-cH (dx dy / R2) du x 0 Iup S-2 EFxq EFqo SAA { e-ch (1 + cos2[ + ]) 3/(16) + e-ah 11.11 K f( + )} 11 Approved For Public Release; Distribution Unlimited AIM v2 • Original model tenets maintained • brightness ~ population • SSA • City population data base included • Cloud effects added • 8 cloud types • reflection • Lighting data base added • Airglow included 12 Approved For Public Release; Distribution Unlimited Natural Background Radiation In the absence of moonlight and artificial light, the night sky still is not “dark”. S10 is a bizarre astronomical unit of surface brightness corresponding to one 10th-magnitude star per square degree 13 Approved For Public Release; Distribution Unlimited Natural Background Radiation In general: Bnsb = 27.78 – 2.5 x log10(Bairglow + Bzodiacal + Bstarlight), where Bairglow, Bzodiacal, and Bstarlight are functions of the solar sunspot cycle, the ecliptic latitude, and the galactic latitude respectively. It was deemed reasonable to set the value of Bnsb = 17.6 nL rather than carry out additional computations for only a small additional contribution to the total night sky brightness For Denver at a distance of ~70 km the model predicts that the brightness is ~18 nL. 14 Approved For Public Release; Distribution Unlimited Cloud Effects • Transmission through clouds determined using ILLUMA • 8 cloud types in 3 layers • High - clear, thin and thick Ci/Cs • Medium - clear, As/Ac • Low - clear, clear (f/k), Cu/Cb or St/Sc • allowance for partial cloudiness • Cloud reflection • cloud and angle dependent reflectivity taken from Shaprio • simple attenuation in upward and downward paths • from lowest layer only 15 Approved For Public Release; Distribution Unlimited AIM Input • city name, city radius, city altitude, lumens/head • distance from observer to source, observer altitude, zenith angle, azimuth angle • clarity factor, surface albedo, fraction of light in upward direction • month, day, year, UTC • light type percents • cloud type & cloud amount for layer i, i = 1,3 16 Approved For Public Release; Distribution Unlimited Validation Clear Skies Sky brightness due to Denver as a function of zenith angle at a distance of 40 km* Sky brightness due to Denver as a function of distance for a zenith angle of zero* * Garstang ‘86 17 Approved For Public Release; Distribution Unlimited Brightness Distribution Denver 8.6 km from city center Zenith = 0 X Observer 18 Approved For Public Release; Distribution Unlimited Brightness Distribution Boulder Zenith = 65 Azimuth = 180 Distance = 0 km Boulder Zenith = 65 Azimuth = 0 Distance = 0 km X X Boulder Zenith = 65 Azimuth = 45 Distance = 0 km Boulder Zenith = 65 Azimuth = 45 Distance = 2 km X X marks observer position 19 X Approved For Public Release; Distribution Unlimited Spectral Radiant Energy 20 Approved For Public Release; Distribution Unlimited Spectral Radiant Energy Broadband illumination is broken into spectral components dependant on the City’s light types and amounts Approach Determine 1) Light types 2) Percent contribution 3) Spectral radiance values for each light type 21 Approved For Public Release; Distribution Unlimited Spectral Radiant Energy Light Types and their default percentages* Light Type Clear Mercury Default Value (%) Pi 20.2 Low Pressure Sodium 0 High Pressure Sodium 77.9 Metal Halide 1.9 white LED 0 Standard Fluorescent 0 Incandescent 0 Liquid Kerosene 0 Pressurized Propane 0 * El Paso Electric Co. 22 Approved For Public Release; Distribution Unlimited Light Type Spectra (1 nm resolution) Mercury Vapor Metal Halide Incandescent 23 HPS LPS Fluorescent LED Kerosene Propane Approved For Public Release; Distribution Unlimited j 1 Li (bin) li Spectral Radiant Energy Technique j For Light Type i 1) Identify total contribution over wavelength band S1 to S2 S1 Ltotal S2 L S S2 wavelength band under consideration HPS 1 2) Bin contribution LJth bin J 1 L J 24 Approved For Public Release; Distribution Unlimited HPS j 1 Li (bin) li j Spectral Radiant Energy Technique 3) Fractional radiance Lbin Li Pi Ltotal 4) Final spectral radiance sum over light type N B B Li i 1 25 Approved For Public Release; Distribution Unlimited Spectral Radiance Spectrum of Los Angles Zenith view Courtesy of Dr. Martin Aubé http://lightpollution.no-ip.org/~lightpol/data_usage.html Approved For Public Release; Distribution Unlimited Aubé AIM v2 10% Clear Mercury 5% LPS 55% HPS 20% Metal Halide 10% Incandescent No Airglow 27 Aubé AIM v2 10% Clear Mercury 5% LPS 55% HPS 20% Metal Halide 10% Incandescent SAB + Airglow Note increase in background level 28 V&V Night Illumination Test Evaluation (NITE) A Joint Undertaking with the UK Met Office 29 Approved For Public Release; Distribution Unlimited NITE: Location x 30 Approved For Public Release; Distribution Unlimited Night Illumination Test Evaluation (NITE) 31 Approved For Public Release; Distribution Unlimited Night Illumination Test Evaluation (NITE) x 32 Approved For Public Release; Distribution Unlimited Night Illumination Test Evaluation (NITE) x 33 Approved For Public Release; Distribution Unlimited NITE: Instrumentation Microtops Sunphotometer 5 channel hand-held sunphotometer for measuring aerosol optical thickness easily, accurately and dependably. 34 Approved For Public Release; Distribution Unlimited NITE: Instrumentation LMT Hoffman Albedo stand All-Sky camera first day out (brrr) 35 Approved For Public Release; Distribution Unlimited NITE: Meteorology • Met data • WSMR04 • Holloman AFB • mesonetwest.utah.edu Approved For Public Release; Distribution Unlimited NITE: Satellite Pictures 37 Approved For Public Release; Distribution Unlimited What’s Next? • Variable turbid atmosphere • add assortment of aerosol types • Inclusion of horizontal single scattering effects • zenith angles > 80 produce dubious results • City reflection from partially cloudy skies • reflection only allowed from lowest layer • Terrain/Vegetation blocking • NITE data reduction for V&V 38 Approved For Public Release; Distribution Unlimited Acknowledgements Warren Lewis* (for suggesting NITE) Damian Wilson* (for tolerating numerous snafus) Jeff Passner§ (for satellite pics) Ed Creegan§ (for setup and data acquisition) Will Lee-Rodgers* (for just about everything else) * UKMO § ARL 39 Approved For Public Release; Distribution Unlimited ? 40 Approved For Public Release; Distribution Unlimited