for Ground-Based Astroparticle Detectors MPI Workshop, January 31, 2017 Instituto de Astrofísica de Canarias (IAC)
Detection Principle p, γ Development of Air Shower depends on density profile ρair = m air V = p M air RT 2
Detection Principle p, γ Development of Air Shower depends on density profile ρair = m air V = p M air RT Interactions with Atmosphere cause emission of photons (Fluorescence and Cherenkov) 3
Detection Principle p, γ Development of Air Shower depends on density profile V = p M air 15 40 k m RT 5 15 km Interactions with Atmosphere cause emission of photons (Fluorescence and Cherenkov) Photons ρair = m air Attenuation (scattering) depends on aerosols and molecular atmosphere I det I emit T mol T aer emitted measured Height (km) 4
Fluorescence Light 337nm M. Ave et al., NIMA 597 (2008) 357nm Particles excite nitrogen vibrational and/or rotational states Radiative transitions to lower states Isotropic emission of UV fluorescence light Non-radiative transition through collisions with (water) molecules Vapor quenching 5
Fluorescence Yield tot de dep dn γ = dx dx FY (λ, p, T, e) τ atm (λ, p,t, e) ϵ FD (λ) dλ Reduction of emitted light due to humidity up to 7 km a.s.l. Reconstructed energy without consideration of humidity too low Small change in shower maximum (dependent on zenith angle) B. Keilhauer et al., NIMA 597 (2008) Simulation Electron 0.85 MeV Simulation Fe, 1019 ev, 0 6
Cherenkov Photons Number of Cherenkov photons per charged particle dn γ 2πα (h, E) = dx ρair (h) (1 n 2 β 2) λ 2 d λ Shower particles exceed speed of light in air Cherenkov threshold 2 mc E thr h = 2 n 1 Bonardi et al., Exper.Astron. 38 (2014) 7
Scattering τ / sin ψ T=e Rayleigh Scattering (Molecular Atmosphere) Mie Scattering (Aerosols) 8
Molecular Atmosphere Effect on produced light Shower development depends on density Number of Cherenkov photons, Cherenkov threshold Fluorescence Yield Transmission to detector I det I emit T mol T aer Reconstructed energy scales with optical transmission 9
Molecular Atmosphere 10
Weather Balloon 11
Weather Station Ground Measurements T (ºC) Temperature Pressure Relative humidity Wind speed and direction Mean Night Day 12
Global Data Assimilation System Global measurements and numerical weather prediction GDAS data available for whole earth 1 grid (180 x 360 ) every 3 hours Pierre Auger Observatory Comparison with balloon data validates GDAS for Auger site 13
GDAS Advantages Balloon Monthly Model Balloon GDAS Improved systematic uncertainty compared to other models Replacement for balloon launches Save money for equipment and personnel 14
GDAS in La Palma Pressure (mbar) 14 60 40 12 20 785 0-4 -3-2 -1 0 1 2 3 10 4 p WS -p GDAS (mbar) 8 780 Pressure Comparison Entries Mean x Mean y RMS x RMS y 775 2384 787.8 787.4 3.326 3.291 : 0.97,S: 103.3 775 780 785 790 795 MAGIC Weather station Pressure (mbar) WS 6 20 15 Entries Mean x Mean y RMS x RMS y 2384 9.763 11.95 4.915 4.553 10 8 : 0.95,S: 91.0 6 10 counts Temperature Comparison GDAS 790 Humidity (%) GDAS1 vs. MAGIC Weather station (night-time 2013/14) 100 GDAS1 Humidity 2188m (%) 80 25 90 Accuracy MAGIC weather station: 2% Humidity Residuals Entries Mean RMS 80 70 60 80 2384-3.791 14.18 RMS x RMS y 50 40 30 20 Humidity Comparison Entries Mean x Mean y 2384 32.77 26.08 29.39 18.86 18 16 14 : 0.74,S: 46.1 20 10 60 0-50 -40-30 -20-10 0 10 20 30 40 12 50 RHWS -RHGDAS (%) 10 Accuracy MAGIC weather station: 0.3 C Temperature Residuals counts 16 Entries 2384 Mean -0.4149 0.792 RMS 100 GDAS Pressure Residuals 120 GDAS1 Temperature 2188m ( C) 795 Temperature (ºC) GDAS1 vs. MAGIC Weather station (night-time 2013/14) 18 Accuracy MAGIC weather station: 0.8 mbar Resolution MAGIC weather station: <2 mbar counts GDAS1 Pressure 2188m ( C) GDAS GDAS1 vs. MAGIC Weather station (night-time 2013/14) 140 Entries Mean RMS 120 5 100 2384 2.185 1.493 4 40 8 6 80 4 2 60 0 20 4 2 20 0-10 0 2 40-5 -5 0-8 -6-4 -2 0 2 4 6 8 10 TWS -T GDAS ( C) 5 10 15 20 25 MAGIC Weather station Temperature ( C) 0 WS 0 0 20 40 60 80 100 MAGIC Weather station Humidity (%) WS Comparison with MAGIC weather station Agreement in pressure Sys. offset in temperature (~ 2ºC ground effects) Humidity very dependent on location Cheap and reliable data source for CTA 15 0
Aerosols Aerosol enhancements close to ground and clouds Highly variable in altitude and time, scale of hours Transmission to detector I det I emit T mol T aer Strong energy dependence on cloud height 16
Aerosol Transmission and Clouds I det I emit T mol T aer Measuring instrument: LIDAR Light Detection and Ranging ( Light Radar ) Different kind of LIDARs Wavelength of scattered light (scatter center, scattering process) Location of laser and detector (collocated or separated) Each with advantages and disadvantages 17
Theory: Elastic LIDAR Light reaching detector L = C G L0 T mol T aer [βmol (π) + βaer (π)] T mol T aer Aerosol λ0 λ0 Aerosol backscattering unknown Size and number of aerosols unknown Need assumptions or scanning 18
Theory: Raman LIDAR Light reaching detector L = C G L0 T mol T aer R β N (π) 2 T R mol R T aer N2 λ0 λr Nitrogen backscattering Number density known Low Raman cross section Large amount of light needed 19
Raman LIDAR R&D system commissioned in SE Colorado Nd:YAG 355 nm 5 10 mj Raman Telescope Raman Receiver Laser 20
Raman LIDAR Data 21
Theory: Bi-static LIDAR Upper edge of FoV ~ 10 15 km Light reaching detector L = C L0 T mol T aer [βmol ( π ψ) + βaer ( π ψ)] 1/cos (ψ) [T mol T aer ] ψ 30 40 km Nd:YAG 355 nm 5 10 mj 22
Bi-static LIDAR (Receiver) 4 vertical columns, 16 PMTs each 1 FoV per pixel (old HiRes-II camera) UV bandpass filter Slow Control, Calibration Power, Readout, Internet 23
Bi-static LIDAR Main assumptions τ aer ( h) = ln N ref (h) ln N obs ( h) Reference night clear of aerosols Scattering out of beam is dominated by Rayleigh scattering Atmosphere horizontally uniform between laser and FD 1 + csc ϕ 2 Average Night 24
Bi-static LIDAR Data Hazy Night Cloud Above Laser 25
Comparison Raman / Bi-static Wiencke, Rizi, Will, in preparation (2016) preliminary Systematic difference Type of aerosols Reference night not clean Under investigation 26
Raman / Bi-static LIDAR at Auger km 0 3 40km 27
CTA Raman LIDAR Upgrade of Colorado system Continue measurements Characterize site Integrate into CTA Move to La Palma 2017 Later to southern site Minimize impact on CTA measurements 28
FRAM (Fotometric Robotic telescope for ) Passive measurement Stellar photometry 15 15 FoV, several 100 stars Integral extinction 10 years experience from Auger FRAM for CTA Prototype deployment in La Palma Aerosol maps in fixed FoV Altitude scans for aerosol profiles 11 inch MPI telescope for MAGIC Similar characteristics, spectrograph Transmission from spectrum differences of stars Deployed in La Palma before FRAM 29
CTA AllSky Camera AllSky camera on MAGIC counting house roof Czech construction (also used at Auger) 3 different filters (plus no filter) 60 seconds exposure 30
Cloud Detection Star detection with image filter Comparison with catalog star magnitude Determine cloud cover 31
Cloud Detection Star detection with image filter Comparison with catalog star magnitude Determine cloud cover 32
Cloud Tracking Identify single clusters in cloud maps Compare position of clusters between images Difference is direction movement of clouds 33
Cloud Tracking Identify single clusters in cloud maps Compare position of clusters between images Difference is direction movement of clouds 34
Cloud Tracking Identify single clusters in cloud maps Compare position of clusters between images Difference is direction movement of clouds 35
Cloud Tracking Identify single clusters in cloud maps Compare position of clusters between images Difference is direction movement of clouds Potential problems Clouds can change shape Clouds can split or merge Clouds can disappear Further improvement Wind speed from GDAS Cloud properties from LIDAR 36
CTA Air Shower Simulations Systematic uncertainties in CTA must not exceed 10% MAGIC: 11% due to atmosphere Auger: 6 7% due to atmosphere (14% total) Simulated response functions Need input from atmospheric measurement instruments Fast for online analysis Reliable and precise for offline analysis Instrumentation at Northern CTA site Weather stations, wind sensors, rain sensors, Dust counters, electric field mill,... LIDARs, FRAM,... 37
Summary Atmospheric parameters influence air shower detection Density profile influences development Cherenkov and fluorescence light production Transmission depends on molecular and aerosol scattering Measurement of atmospheric parameters Balloons, weather stations, model data for profiles LIDARs and passive photometry for transmission Cloud detection using AllSky cameras Apply lessons learned for CTA Combination of instruments to keep sys. uncertainties at 10% New models and unexplored sources for atmospheric data 38
GDAS Data Comparison Balloon Monthly Models Balloon GDAS Monthly Models include balloon data until 2008 Abreu et al., Astropart. Phys. 35 (2012) 591-607 39
Other Data Sources Dust measurement at TNG Other ORM telescopes Weather station data Seeing Dust concentration Dust (µg/m3) Automatic particle counter Lasair II 310B Particle concentration from laser scattering Size sensitivity 0.3, 0.5, 1.0, 3.0, 5.0, 10.0 µm 2h cumulative density in µg/m3 Time 40
Satellite Data and Forecasts Data available for La Palma site SKIRON AOD Forecast Weather forecasts Cloud satellite images Aerosol optical depth forecast EUMETSAT Cloud Image Roque de los Muchachos Mountain Forecast 41
Star Detection Blob detection with image filter Applied to each pixel Summing up all neighbor pixels Filter mask as weight matrix Returned value is filter response Chosen filter: Laplacian of Gaussian (LoG) Reduce noise by smoothing with Gaussian Laplacian filter: adjustable blob size, rotation invariant, fast computing speed, insensitive to linear brightness gradients Apply kernel for each star in catalog Avoid hot pixels Chose magnitude limit Take into account exposure, atm. absorption, lens distortions 42
Energy Threshold Cirrus (10 km) Volcanic debris (14 km) 43
Raman LIDAR Hatch Raman Telescope Detector Raman Receiver 39 km Laser Raman DAQ GPS antenna Hatch 44
Bi-static LIDAR DAQ PC PC UPS Radio Sounding Electronics DAQ Electronics Microwave 45
Bi-static LIDAR View from inside the detector Laser Camera 46
Bi-static LIDAR Time bins [50 ns] 47