Basic Radiometry and SNR Equations for CCD, ICCD and EMCCD Imagers

Basic radiometry and SNR equations for CCD, ICCD and EMCCD imagers

1 Urban Brandstr¨ om,¨

1 Swedish Institute of Space Physics, Kiruna, Sweden Presentation at:

http://alis.irf.se/˜urban/AGF351/Braendstroem-UNIS.pdf

UNIS 2011-11-15 – p. 1 In memoriam

Professor Ingrid Sandahl (1949-2011) UNIS 2011-11-15 – p. 2 This is about taking pictures of darkness, or. . .

UNIS 2011-11-15 – p. 3 “Hunting photons with a spoon”

UNIS 2011-11-15 – p. 4 Radiometry

UNIS 2011-11-15 – p. 5 Radiometry vs. photometry

Holst [1998] deﬁnes the term radiometry, as the “energy or power transfer from a source to a detector”

UNIS 2011-11-15 – p. 6 Radiometry vs. photometry

Holst [1998] deﬁnes the term radiometry, as the “energy or power transfer from a source to a detector” while photometry is deﬁned as “the transfer from a source to a detector where the units of radiation have been normalised to the spectral sensitivity of the eye.”

UNIS 2011-11-15 – p. 6 Radiometry vs. photometry

UNIS 2011-11-15 – p. 6 Radiometry “Mathematics is often called the queen of the sciences. Radiometry should then be called the waiting maid or servant. It is not especially elegant; it is not very popular, has not been trendy; but it is essential in almost every part of optical engineering.” Wolfe [1998]

UNIS 2011-11-15 – p. 7 Solid angle The solid angle Ω sweeps out the area A on the unit sphere (4π) A Ω= [sr] r2

Think of it as a 3D generalisation of the radian (arc length on the unit circle)

UNIS 2011-11-15 – p. 8 Flux Photon ﬂux: ∂N photons Φ = γ ∂t s in energy units: hc ∂N Φ = [W] E λ ∂t

UNIS 2011-11-15 – p. 9 Radiance Also known as radiant sterance In energy units: ∂2Φ(λ) W LE = 2 ∂As∂Ω m sr In quantum units: λ photons L = L γ hc E sm2 sr

UNIS 2011-11-15 – p. 10 Spectral radiance Also known as spectral radiant sterance In energy units: ∂L W L = λE ∂λ m2 µm sr

In quantum units: λ photons L = L λγ hc λE sm2 µm sr

UNIS 2011-11-15 – p. 11 Spectral radiant emittance Also known as spectral radiant exitance ∂Φ photons Mλγ = = 2 ∂As s m

Flux per source area. What you get from a calibration source. In energy units: hc W M = M λE λ λγ m2 µm sr

UNIS 2011-11-15 – p. 12 Spectral irradiance Also known as spectral radiant incidance ∂Φ photons E = = λe ∂A s m2

Flux per detector area. What you get on a detector (or whatever) In energy units: hc W E = E λE λ λγ m2 µm sr

UNIS 2011-11-15 – p. 13 Transmittance

T = TX(λ)= TaToTf . . . YX ∀

UNIS 2011-11-15 – p. 14 Irradiance At apperture:

Φγapp LγAsTaΩds photons Eγapp = = = 2 Aapp Aapp s m

At image plane (assuming circular apperture):

2 Φγapp As πdapp photons (1) Eγi = = Lγ T 2 = 2 Ai Ai 4rs s m

UNIS 2011-11-15 – p. 15 Photometric units

750nm Φv = KM V (λ)Mp(λ)dλ [lm] Z380nm

scoptic—rods photoptic—cones After Holst [1998]

UNIS 2011-11-15 – p. 16 Photometric units

scoptic—rods (KM = 1746 lm/W) photoptic—cones UNIS 2011-11-15 – p. 17 (K = 683 lm/W) After Holst [1998] Photometric units

Φv lm luminous ﬂux 2 Lv cd/m or nits luminance 2 Mv lux or lm/m luminous emmitance 2 Ev lux or lm/m illumniance

UNIS 2011-11-15 – p. 18 The foot-lambert A foot-lambert or footlambert (fL, sometimes ﬂ or ft-L) is a unit of luminance in U.S. customary units and some other unit systems. A foot-lambert equals 1/π candela per square foot, or 3.426 candela per square meter (the corresponding SI unit). 1 cd cd 1 [ftL] = 3.426 π ft2 ≈ m2

UNIS 2011-11-15 – p. 19 The Rayleigh

UNIS 2011-11-15 – p. 20 The Rayleigh (1) Consider a cylindrical column of cross-sectional area 1 m2 extending away from the detector into the source. The volume emission rate from a volume element of length dl at distance l is 3 1 ǫ(l,t,λ) photons m− s− . The contribution to Lγ is given by: ǫ(l,t,λ) photons (2) dL = dl γ 4π sm2 sr

UNIS 2011-11-15 – p. 21 The Rayleigh (2) Integrating along the line of sight l [m]:

∞ (3) 4πLγ = ǫ(l,t,λ)dl Z0 This quantity is the column emission rate, which Hunten et al. [1956] proposed as a radiometric unit for the aurora and airglow.

UNIS 2011-11-15 – p. 22 The Rayleigh (3) In SI-units the Rayleigh becomes [Baker and Romick, 1976]: photons (4) 1 [Rayleigh] 1 [R] , 1010 ≡ sm2 (column) The word column denotes the concept of an emission-rate from a column of unspeciﬁed length, as discussed above. It should be noted that the Rayleigh is an apparent emission rate, not taking absorption or scattering into account.

UNIS 2011-11-15 – p. 23 The Rayleigh (4) However (unfortunately. . . )

“the Rayleigh can be used as deﬁned without any commitment as to its physical interpretation, even though it has been chosen to make interpretation convenient.” Hunten et al. [1956]

And then there is the clariﬁcations by: Baker [1974]; Baker and Romick [1976]; Chamberlain [1995]

UNIS 2011-11-15 – p. 24 By now you should realized that...

UNIS 2011-11-15 – p. 25 . . . God said: Go to, let us go down, and there confound their language, that they may not understand one another’s speech. [Bible Gen11:7]

UNIS 2011-11-15 – p. 26 . . . God said: Go to, let us go down, and there confound their language, that they may not understand one another’s speech. [Bible Gen11:7] And there was: stilb, Rayleighs, footlamberts, Irradiance, spectral-radiant sterance, lumens, lux, candela, radiometry, nit, luminance, illuminance, emittance, apostilb, phot, skot, lambert, foot-candle, photometry, DIN, ASA, ISO...

—Help! We are sinking!

UNIS 2011-11-15 – p. 26 and now. . .

UNIS 2011-11-15 – p. 27 The 4π confusion

UNIS 2011-11-15 – p. 28 The 4π confusion Therefore, we propose that photometric measurements of the airglow and aurora be reported in terms of 4πB rather than the surface brightness B itself. Further, we suggest that 4πB be given the unit “rayleigh” (symbol R), where B is in units of 106 quanta cm−2 s−1 sr−1. Hence −1 1R=106 quanta cm−2 (column) s−1. Hunten et al. [1956]

UNIS 2011-11-15 – p. 29 The 4π confusion Therefore, we propose that photometric measurements of the airglow and aurora be reported in terms of 4πB rather than the surface brightness B itself. Further, we suggest that 4πB be given the unit “rayleigh” (symbol R), where B is in units of 106 quanta cm−2 s−1 sr−1. Hence −1 1R=106 quanta cm−2 (column) s−1. Hunten et al. [1956]

So does both Hunten et al. [1956] and Chamberlain [1995] claim that 4π 106 = 106 ??? ×

UNIS 2011-11-15 – p. 29 Can we agree on this?

The apparent radiance (Lγ) can be obtained from the column emission rate I (in Rayleighs) according to Baker and Romick [1976]: 1010I photons (5) L = γ 4π sm2 sr

UNIS 2011-11-15 – p. 30 Can we agree on this?

The apparent radiance (Lγ) can be obtained from the column emission rate I (in Rayleighs) according to Baker and Romick [1976]: 1010I photons (7) L = γ 4π sm2 sr

Or is it:

photons (8) L = 1010I γ sm2 sr

UNIS 2011-11-15 – p. 30 Still confused. . .

. . . but at a different level. UNIS 2011-11-15 – p. 31 Signal

UNIS 2011-11-15 – p. 32 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

UNIS 2011-11-15 – p. 33 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

UNIS 2011-11-15 – p. 33 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

• Area of detector (pixel-area for imagers)

UNIS 2011-11-15 – p. 33 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

• Area of detector (pixel-area for imagers)

• Number of photoelectrons collected in a pixel

(15) − QE int neγ = (λ)Eγi t Apix e−

UNIS 2011-11-15 – p. 33 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

• Area of detector (pixel-area for imagers)

• Number of photoelectrons collected in a pixel

(17) − QE int neγ = (λ)Eγi t Apix e−

1010I (18) n − QE(λ)Ttint Apix e− eγ ≈ 16f 2 #

UNIS 2011-11-15 – p. 33 Noise

UNIS 2011-11-15 – p. 34 What is noise?

UNIS 2011-11-15 – p. 35 Some peoples noise are other peoples signal

UNIS 2011-11-15 – p. 36 Notation X 2 variance of X h i X standard deviation of X h i X mean value of X Photon arrival is Poisson distributed It can be shown that for a Poisson distributed signal variance is equal to the mean

UNIS 2011-11-15 – p. 37 CCD principle of operation

After Janesick et al. [1987]

UNIS 2011-11-15 – p. 38 E2V TECH CCD201

UNIS 2011-11-15 – p. 39 What is the SNR of an ideal photon detector?

UNIS 2011-11-15 – p. 40 CCD-noise sources

UNIS 2011-11-15 – p. 41 CCD noise

2 2 2 (19) ne− = ne− + ne− + ne− e− h CCD i h s i h r i h p i RMS q

UNIS 2011-11-15 – p. 42 Shot Noise

N (20) − CTE − 2 − 2 nes = neγ + ne = h i r h i h d i

N (21) CTE − − = neγ + ne r d ≈

(22) − − neγ + ne ≈ q d

UNIS 2011-11-15 – p. 43 CCD Noise sources

After Holst [1998] UNIS 2011-11-15 – p. 44 Pattern Noise

(23) − − 2 − 2 − nep = ne + ne ne h i qh F P N i h PRNU i ≈h PRNU i ≈

− neγ (24) − Uneγ eRMS− ≈ ≈ √ne− max

UNIS 2011-11-15 – p. 45 CCD Noise

2 (25) ne− ne− + ne− + ne− e− h CCD i ≈ γ d h r i RMS q

UNIS 2011-11-15 – p. 46 Signal-to-noise ratio for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ (26) SNRCCD = DN − noise ≈ ne h CCD i

UNIS 2011-11-15 – p. 47 Signal-to-noise ratio for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ (29) SNRCCD = DN − noise ≈ ne h CCD i

• thus for a CCD:

− neγ (30) SNRCCD ≈ − − − 2 neγ + ne + ner q d h i

UNIS 2011-11-15 – p. 47 Signal-to-noise ratio for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ (32) SNRCCD = DN − noise ≈ ne h CCD i

• thus for a CCD:

− neγ (33) SNRCCD ≈ − − − 2 neγ + ne + ner q d h i

• and for an ideal photon detector:

(34) SNR − γideal = neγ UNIS 2011-11-15 – p. 47 q Threshold of detection The threshold of detection is usually deﬁned as SNR =2 while the Noise Equivalent Exposure NEE, is obtained when SNR =1. For a CCD the maximum signal, or Saturation Equivalent Exposure, SEE is obtained when the charge well − capacity nemax [e−], is reached. This occurs when:

(35) ne− ne− ne− γ ≥ max − d In most cases the maximum charge-well capacity DN SEE [counts], is matched to the maximum ADC output DN max.

UNIS 2011-11-15 – p. 48 Dynamic range (1) The Dynamic Range is deﬁned as the peak signal divided by the RMS noise and the DC-bias-level, (if any). The minimum ADC output, is subtracted in the case of a signed integer output. DR is usually expressed in decibels.

DN SEE DN min DR = 20log10 − [dB] DN DC + DN NEE DN min (36) −

UNIS 2011-11-15 – p. 49 Dynamic range (2) An approximate theoretical value for DR is obtained by dividing the maximum signal by the total noise n − n − emax ed (37) DR 20log10 − [dB] ≈ n − h etot i

UNIS 2011-11-15 – p. 50 ICCD

UNIS 2011-11-15 – p. 51 ICCD

After Holst [1998] UNIS 2011-11-15 – p. 52 SNR for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ SNRCCD = DN − noise ≈ ne h CCD i

UNIS 2011-11-15 – p. 53 SNR for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ SNRCCD = DN − noise ≈ ne h CCD i

• For an ideal photon detector:

SNR − γideal = neγ q

UNIS 2011-11-15 – p. 53 SNR for a CCD

• Measured signal-to-noise ratio:

DN − signal neγ SNRCCD = DN − noise ≈ ne h CCD i

• For an ideal photon detector:

SNR − γideal = neγ q

• and for a CCD:

− neγ SNRCCD

≈ − − − 2 UNIS 2011-11-15 – p. 53 neγ + ne + ner q d h i SNR for an ICCD Noting that for an ICCD:

10 2 10 I n − QE (λ)Ttint M Apix e− eγ,pc ≈ pc F O 16f 2 RMS #

UNIS 2011-11-15 – p. 54 SNR for an ICCD

• The signal-to-noise ratio for an ICCD can be estimated as:

n − SNR eγ,pc ICCD 2 − + − ≈ ne ne 2 d h r i k (ne− + ne− )+ 2 r MCP γ,pc d,pc g

UNIS 2011-11-15 – p. 55 SNR for an ICCD

• The signal-to-noise ratio for an ICCD can be estimated as:

n − SNR eγ,pc ICCD 2 − + − ≈ ne ne 2 d h r i k (ne− + ne− )+ 2 r MCP γ,pc d,pc g

• As seen, increasing the gain of the image intensiﬁer makes the CCD noise-sources negligible, but does not increase the SNR beyond that. For very high gain, we see that:

− neγ,pc SNRICCD ≈ 2(n − + n − ) UNIS 2011-11-15 – p. 55 eγ,pc ed,pc q SNR for an emCCD The signal-to-noise ratio for an electron-multiplication CCD can be estimated as:

− neγ SNRemCCD 2 ≈ n − 2 2 h er i kem (ne− + ne− + ne− )+ 2 r γ d h cic i g

Please note: For an EMCCD kem √2 while kMCP (ICCD) is ≈ taken as √2 here, which is somewhat too good to be true. In real cases kMCP 1.6 ≥

UNIS 2011-11-15 – p. 56 SNR example: CCD vs. ICCD

t=16.7 ms, T=0.5, f/3.9, ALIS CCDCAM5, PAI ICCD 1000 Ideal CCD (a) ALIS CCD (b) PAI ideal CCD (c) PAI ICCD (d) PAI CCD (e)

100

10 SNR

0.1 10000 100000 1e+06 1e+07 1e+08 1e+09 1e+10 Column emission rate Rayleighs 557.7 nm

UNIS 2011-11-15 – p. 57 SNR vs. of integration time

T=0.5, f/3.5, ALIS CCDCAM5 557.7 nm 10000 1 MR (a) 100 kR (b) 10 kR (c) 1 kR (d) 100 R (e) 1000

100 SNR

0.1 0.001 0.01 0.1 1 10 100 1000 10000 Integration time [s]

UNIS 2011-11-15 – p. 58 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

UNIS 2011-11-15 – p. 59 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

UNIS 2011-11-15 – p. 59 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

• Area of detector (pixel-area for imagers)

UNIS 2011-11-15 – p. 59 Where are my photons?

• Transmittance (atmosphere, optics, ﬁlters. . . )

• Apperture of the optics

• Area of detector (pixel-area for imagers)

• Number of photoelectrons collected in a pixel

(44) − QE int neγ = (λ)Eγi t Apix e− 1010I (45) n − QE(λ)Ttint Apix e− eγ ≈ 16f 2 #

UNIS 2011-11-15 – p. 59 SNR and on-chip binning

t=1 s, T=0.5, f/3.5, ALIS CCDCAM5, PAI ICCD 1000 bin 1,1 (a) bin 2,2 (b) bin 4,4 (c) bin 8,8 (d) bin 16,16 (e)

100

10 SNR

0.1 1 100 10000 1e+06 1e+08 1e+10 Column emission rate Rayleighs 557.7 nm

UNIS 2011-11-15 – p. 60 SNR: CCD, ICCD and emCCD

t=1/25 s, T=0.5, f/1.6, bin=1x1 1000 Ideal SI003AB (a) SI003AB (b) Ideal PAI CCD (c) PAI ICCD (d) Ideal CCD201-20 (g) CCD201-20 (h) bin 2,2 CCD201-20 (h) 100

10 SNR

0.1 1000 10000 100000 1e+06 1e+07 1e+08 1e+09 1e+10 Column emission rate Rayleighs 557.7 nm

UNIS 2011-11-15 – p. 61 When do we need EM-gain?

UNIS 2011-11-15 – p. 62 Fast: 2900 photons/pixel ≈ EM ON vs. EM OFF at 10 MHz 100 Ideal CCD201-20 EM OFF 10 MHz CCD201-20 EM ON 10 MHz CCD201-20

1 SNR

0.1

0.01 1 10 100 1000 10000 photons/pixel (assuming 90% QE) UNIS 2011-11-15 – p. 63 Always for high temporal resolution

UNIS 2011-11-15 – p. 64 Slow: 42 photons/pixel ≈ Slow readout EM ON vs. Conventional Ampl. 100 Ideal CCD201-20 1 MHz Con. CCD201-20 EM ON 1 MHz CCD201-20

10 SNR

0.1 1 10 100 1000 photons/pixel (assuming 90% QE) UNIS 2011-11-15 – p. 65 Not always for low temporal resolution

UNIS 2011-11-15 – p. 66 Intercalibration

UNIS 2011-11-15 – p. 67 Intercalibration This is the process of intercalibrating calibration sources and to transfer absolute calibration information between different instruments and research groups.

Hans Lauches intercalibration photometer (responsible person: 1981–1999 Lauche, 1999-2011 Widell, SSC, 2011– Brändström, IRF) UNIS 2011-11-15 – p. 68 The European Rayleigh

UNIS 2011-11-15 – p. 69 Intercal. procedure

• Calibrators are compared at calibration workshops using a calibration-photometer with 7 ﬁlters and a reference source.

UNIS 2011-11-15 – p. 70 Intercal. procedure

• Calibrators are compared at calibration workshops using a calibration-photometer with 7 ﬁlters and a reference source.

• Last known absolute callibration of the calibration equipment against a national standard (NBS) was done by [Torr and Espy, 1981].

UNIS 2011-11-15 – p. 70 Intercal. procedure

• Calibrators are compared at calibration workshops using a calibration-photometer with 7 ﬁlters and a reference source.

• Last known absolute callibration of the calibration equipment against a national standard (NBS) was done by [Torr and Espy, 1981].

• Known calibration workshops at the optical meetings were: Aberdeen 1981, Lindau 1983, Lysebu 1985, Saskatoon 1987, Lindau 1989, Wien 1991, Lindau 1999, Stockholm 2000, Oulu 2001, Kiruna 2006, Andøya 2007 and Sodankylä 2011. UNIS 2011-11-15 – p. 70 Intercal. workshops

Number of participating calibrationsources in intercalibration workshops 1981-2011

0 1980 1985 1990 1995 2000 2005 2010

UNIS 2011-11-15 – p. 71 Sodankylä 2011

UNIS 2011-11-15 – p. 72 and the FMI-sphere

UNIS 2011-11-15 – p. 73 To be done here

After Sigernes et al. [2008]

UNIS 2011-11-15 – p. 74 Intercal. results

1000 Y275 L1614 920B

100

1 [Rayleighs/Angstrom] 0.1

0.01

0.001 3500 4000 4500 5000 5500 6000 6500 7000 7500 Wavelength [Angstrom]

UNIS 2011-11-15 – p. 75 Intercal. errors

Confusogram of calibration ratios [1985,1999,2000,2001 to 2006] 30 y275 1985 y275 1999 y275 2000 y275 2001 20 l1614 1985 l1614 1999 l1614 2000 l1614 2001 920b 1985 10 920b 1999 920b 2000 920b 2001

0 ratio [%]

-10

-20

-30 4000 4500 5000 5500 6000 6500 7000 Wavelength [Angstrom]

UNIS 2011-11-15 – p. 76 Calibration issues

UNIS 2011-11-15 – p. 77 Calibration Calibration is the process of answering the following two basic questions: 1. What physical value does the pixel represent? (absolute calibration) 2. How is each pixel mapped to the observed object? (geometrical calibration)

UNIS 2011-11-15 – p. 78 Abs. calibration (ALIS)

UNIS 2011-11-15 – p. 79 Challenge Read the “28. Appendix” and compare to Sigernes et al. [2008]. You might also want to compare to Torr and Espy [1981] Calculate R/Å for the IRF-UJO-Y275 radioactive source around 5577 Å using the result in “28. Appendix” and compare to latest intercalibration result from Sodankylä. (That source is marked 15 µlm

UNIS 2011-11-15 – p. 80 ALIS

UNIS 2011-11-15 – p. 81 ALIS 2009–2012

Norway

R A Finland

D S

N K E O Sweden M

I UNIS 2011-11-15 – p. 82 Spectroscopic imaging

4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 Wavelength [Å]

UNIS 2011-11-15 – p. 83 Selectable common volumes

Approximate field of view at 110 km [km]

250 50o 90o 100 50 Abisko 200 50 Abisko 0 Silkimuotka Silkimuotka NikaluoktaKiruna 0 NikaluoktaKiruna Merasjaervi Merasjaervi −50 150 −50 Tjautjas Tjautjas −100 −100

−150 −150 100 −200 −100 0 100 200 −100 0 100 200 surveilance mag_zen 50 50 Abisko Silkimuotka 100 0 NikaluoktaKiruna 0 −50 Merasjaervi 50 Abisko Tjautjas Silkimuotka −150 −100−500 100 200 [km] −100 0 NikaluoktaKiruna −150 −50 Merasjaervi −200 Tjautjas −100 −200 −100 0 100 200 −200 −100 0 100 south core S 400 300

300 200 200 100 100 β E Abisko β Abisko Silkimuotka W 0 Silkimuotka 0 NikaluoktaKiruna NikaluoktaKirunaMerasjaervi Merasjaervi z x Tjautjas Tjautjas −400 −200 0 200 400 −300 −200 −100 0 100 200 east−west north

400 400

300 300 α 200 200 aφ 100 100 Azimuth Abisko Abisko 0 NikaluoktaKirunaSilkimuotka 0 NikaluoktaKirunaSilkimuotka TjautjasMerasjaervi TjautjasMerasjaervi −200 0 200 −200 0 200 N eiscat heating y

UNIS 2011-11-15 – p. 84 Scientiﬁc results and capabilities

UNIS 2011-11-15 – p. 85 Auroral tomography

20:09:00 20:09:30 20:09:00 20:09:30

160 160 160 160

150 150 150 150

140 140 140 140

130 130 130 130

120 120 120 120

110 110 110 110 Altitude (km) Altitude (km) 100 100 100 100

90 90 90 90 −50 0 50 −50 0 50 −50 0 50 −50 0 50 20:10:00 20:10:30 20:10:00 20:10:30

160 160 160 160

150 150 150 150

140 140 140 140

130 130 130 130

120 120 120 120

110 110 110 110 Altitude (km) Altitude (km) 100 100 100 100

90 90 90 90 −50 0 50 −50 0 50 −50 0 50 −50 0 50 20:11:00 20:11:30 20:11:00 20:11:30

160 160 160 160

150 150 150 150

140 140 140 140

130 130 130 130

120 120 120 120

110 110 110 110 Altitude (km) Altitude (km) 100 100 100 100

90 90 90 90 −50 0 50 −50 0 50 −50 0 50 −50 0 50 North (km) North (km) North (km) North (km) 40 km west of Kiruna 70 km west of Kiruna 1997-02-16 ALIS/FAST/EISCAT

UNIS 2011-11-15 – p. 86 UNIS 2011-11-15 – p. 87 UNIS 2011-11-15 – p. 88 Auroral electron spectras,

from tomography, log electron energy flux 10

9 2 10 8.5

7.5

1 10 7

6.5 electron energy 6

5.5 0 10

4.5 200 400 600 800 1000 1200 time after 23:20:00 UT (s)

UNIS 2011-11-15 – p. 89 Auroral electron spectras,

Characteristic energy (keV) 4278 A

0.9 from tomography, 80 160 0.8 50 140 60 log electron energy flux 120 10 0.7 40 100 0 0.6 80 9 20 N−S distance 2 60 10 0.5 0 −50 8.5 40

N−S distance 0.4 −20 0 500 1000 8 0.3 time after 23:20:00 UT (s) −40 8446 A

7.5 0.2 −60 110 100 1 10 7 0 500 1000 50 90 time after 23:20:00 UT (s) 80 6.5 70 0 60 electron energy Oxygen scaling factor 6 N−S distance 50 0.35 −50 40 80 5.5 30 0 10 60 0 500 1000 0.3 5 time after 23:20:00 UT (s) 6300 A 40

4.5 0.25 250 200 400 600 800 1000 1200 20 time after 23:20:00 UT (s) 50 0 200 0.2 N−S distance −20 and from spectroscopic 0 150 −40 0.15 N−S distance 100

−60 −50 ratios (right panel). 50 0 500 1000 0 500 1000 time after 23:20:00 UT (s) time after 23:20:00 UT (s) After Gustavsson et al. [2001b], Phys. Chem. Earth 26.

UNIS 2011-11-15 – p. 89 UNIS 2011-11-15 – p. 90 ﬁlter/expose sequence

Filter/exposure sequence: sync−rapid−aeronomi

Bus

Kiruna

Optlab

Abisko

0 2 4 6 8 10 12 14 16 18 20 time (s)

UNIS 2011-11-15 – p. 91 UNIS 2011-11-15 – p. 92 UNIS 2011-11-15 – p. 93 UNIS 2011-11-15 – p. 94 Daylight aurora

After Rees et al. [2000], GRL, 27.

UNIS 2011-11-15 – p. 95 Radio-induced optical emissions

UNIS 2011-11-15 – p. 96 RIOE ALIS made the ﬁrst unambigous observation of high-latitude RIOE 1999-02-16 17:40:15 17:40:35 17:40:55 17:41:15 17:41:35 17:41:55 4000 3000 2000 1000 0

17:43:55 17:44:05 17:44:15 17:44:25 17:44:35 17:44:45 4000 3000 2000 1000 0

After [Brandstr¨ om¨ et al., 1999], GRL, 26.

UNIS 2011-11-15 – p. 97 Tomography of RIOE

ALIS made the ﬁrst tomographic estimate of volume distribution of RIOE.

After Gustavsson et al. [2001a], JGR

106, 29 UNIS 2011-11-15 – p. 98 UNIS 2011-11-15 – p. 99 UNIS 2011-11-15 – p. 100 Meteor research

UNIS 2011-11-15 – p. 101 A strange meteor trail

4 x 10 5000 10

9 130 4500

120 4000 7

6 110 3500 5 110

4 3000

100 3 100

95 2500 95 2

1 2000 4227 Å (left) 5893 Å (right)

After Pellinen-Wannberg et al. [2004, GRL 31], GRL 31.

UNIS 2011-11-15 – p. 102 Polar-Stratospheric clouds Triangulation

70 24 21 23 22 60 26

25 50

−20 −10 0

After Enell [2002], IRF Sci. Rep. 278

UNIS 2011-11-15 – p. 103 Future plans and challenges

UNIS 2011-11-15 – p. 104 Small structure The aurora is extremly rich in small structure

“With respect to understanding the dynamic coupling between the magnetosphere and the auroral ionosphere the observational bias toward bright aurora is physically unjustiﬁed” [Semeter 2001]

UNIS 2011-11-15 – p. 105 We do not understand:

• Creation of narrow arcs

UNIS 2011-11-15 – p. 106 We do not understand:

• Creation of narrow arcs

• Diffuse aurora

UNIS 2011-11-15 – p. 106 We do not understand:

• Creation of narrow arcs

• Diffuse aurora

• Pulsating aurora

UNIS 2011-11-15 – p. 106 We do not understand:

• Creation of narrow arcs

• Diffuse aurora

• Pulsating aurora

• The role of the ionosphere in the magnetosphere-ionosphere coupling

UNIS 2011-11-15 – p. 106 We do not understand:

• Creation of narrow arcs

• Diffuse aurora

• Pulsating aurora

• The role of the ionosphere in the magnetosphere-ionosphere coupling

• How are different scales related to each other?

UNIS 2011-11-15 – p. 106 We do not understand:

• Creation of narrow arcs

• Diffuse aurora

• Pulsating aurora

• The role of the ionosphere in the magnetosphere-ionosphere coupling

• How are different scales related to each other?

Thus we need instruments measuring different scales with high temporal and spatial resolution, e.g. Polar/VIS, ASC, ALIS, ASK

UNIS 2011-11-15 – p. 106 ALIS 2010–2014

• Electrodynamics of auroral structures: get most out of EISCAT-UHF

• ALIS/EISCAT/REIMEI

• Improve temporal resolution: EMCCD

• Review which sites to use

• Ionospheric sounding rockets?

• Collaboration for development of methods and models

• Calibration!!!

• Improve access to data

UNIS 2011-11-15 – p. 107 In particular we will work to answer the following speciﬁc questions: 1. What is the temporal and spatial scale distribution of small (less than a few km) auroral structures?. 2. What are the temporal and spatial variations of the primary particle distributions causing small auroral structures? 3. What is the detailed 3D electrodynamics of small auroral structures? 4. How does ionospheric feedback inﬂuence auroral structure?

UNIS 2011-11-15 – p. 108 and now. . .

UNIS 2011-11-15 – p. 109 My brain hurts!

Mr. T. F. Gumby:—Doctor! Doctor! DOCTOR! DOCTOR! Doctor! — Are you the brain specialist? — My brain hurts! http://www.mwscomp.com/mpfc/gumbrain.html

UNIS 2011-11-15 – p. 110 It’s

The end!

UNIS 2011-11-15 – p. 111 THE END!

UNIS 2011-11-15 – p. 112 The end?

Kiruna ASC 2007-02-05 17.39.00 UTC 10s exp. UNIS 2011-11-15 – p. 113 References

References

Baker, D. J., Rayleigh, the unit for light radiance, Applied Op- tics, 13, 2160–2163, 1974.

Baker, D. J., and G. J. Romick, The Rayleigh interpretation of the unit in terms of column emission rate or apparent radiance expressed in SI units, Applied Optics, 15, 1966–1968, 1976.

Brandstr¨ om,¨ B. U. E., T. B. Leyser, A.˚ Steen, M. T. Rietveld, B. Gustavsson, T. Aso, and M. Ejiri, Unambigous evidence of HF pump-enhanced airglow, Geophys. Res. Lett., 26, 3561– 3564, 1999.

Brandstr¨ om,¨ U., The Auroral Large Imaging System — Design, operation and scientiﬁc results, Ph.D. thesis, Swedish Insti- tute of Space Physics, Kiruna, Sweden, 2003, (IRF Scientiﬁc Report 279), ISBN: 91-7305-405-4.

Chamberlain, J. W., Physics of the aurora and airglow, Classics in geophysics, AGU (American Geophysical Union), 1995, (A reprint of the original work from 1961).

113-1 Enell, C.-F., Optical studies of polar stratospheric clouds and related phenomena, Ph.D. thesis, Swedish Institute of Space Physics, Kiruna, Sweden, 2002, (IRF Scientiﬁc Report 278), ISBN: 91-7305-307-4.

Gustavsson, B., T. Sergienko, M. T. Rietveld, F. Honary, A.˚ Steen, B. U. E. Brandstr¨ om,¨ T. B. Leyser, A. L. Aruliah, T. Aso, and M. Ejiri, First tomographic estimate of volume distribution of enhanced airglow emission caused by HF pump- ing, J. Geophys. Res., 106, 29,105–29,123, 2001a.

Gustavsson, B., A.˚ Steen, T. Sergienko, and B. U. E. Brand-¨ strom,¨ Estimate of auroral electron spectra, the power of ground-based multi-station optical measurements, Phys. Chem. Earth, 26, 189–194, 2001b.

Holst, G. C., CCD arrays, cameras and displays, second ed., The International Society for Optical Engineering, 1998, ISBN: 0-8194-2853-1.

Hunten, D. M., F. E. Roach, and J. W. Chamberlain, A photometric unit for the aurora and airglow, J. Atmos. Terr. Phys., 8, 345–346, 1956.

Janesick, J. R., T. Elliott, S. Collins, M. M. Blouke, and J. Free- man, Scientiﬁc charge-coupled devices, Optical Engineering, 26, 692–714, 1987.

113-2 Pellinen-Wannberg, A., E. Murad, B. Gustavsson, U. Brand-¨ strom,¨ C.-F. Enell, C. Roth, I. P. Williams, and A.˚ Steen, Op- tical observations of water in Leonid meteor trails, Geophys. Res. Lett., 31, 2004.

Rees, D., M. Conde, A.˚ Steen, and U. Brandstr¨ om,¨ The ﬁrst daytime ground-based optical image of the aurora, Geophys. Res. Lett., 27, 313–316, 2000.

Sigernes, F., J. M. Holmes, M. Dyrland, D. A. Lorentzen, T. Svenøe, K. Heia, T. Aso, S. Chernouss, and C. S. Deehr, Sensitivity calibration of digital colour cameras for auroral imaging, Optics Express, 16, 15,623–15,632, 2008.

Torr, M. R., and P. Espy, Intercalibration of instrumentation used in the observation of atmospheric emissions: Second progress report, Tech. Rep. 101, Utah State University, Cen- ter for atmospheric and space sciences, Logan Utah, 1981.

Wolfe, W. L., Introduction to Radiometry, vol. TT29, The Inter- national Society for Optical Engineering, 1998.

113-3