J ovian E xtinction E vents JEE2014 Call for Observations

1. Jovian Extinction Events JEE2014 Call for Observations Modeling the Jovian dust field, moon atmospheres, Europa water geysers, and Io’s Torus 2014 Annual IOTA Meeting, July 12-13, 2014 Baltimore, Maryland

3. (Filter), Color, Wavelength nm Infrared (not visible) (R) Red Orange Yellow (V, G) Green (B) Blue Violet (UV) Ultra Violet (not visible) >750 nm 620-750 nm 590-620 nm 570-590 nm 495-570 nm 450-495 nm 380-450 nm < 380 nm White light spectrum of colors

4. Galileo probe found 600 nm particles around Io. Our multicolor photometry of Io particles demonstrate the scattering effect. Blue photons were significantly scattered while red photons were barely scattered.

5. Rayleigh and Mie scattering of photons 620-750 nm Raleigh 590-620 nm 570-590 nm 495-570 nm Mie 450-495 nm 380-450 nm 600 nm PARTICLE

6. The Hubble Challenge Question: Why can’t I see the atmosphere surrounding Io when it transits Jupiter? Answer: Absolutely you can… you just have to do your homework!

7. The tiny but important details: • HST filter should sample 500 nm or shorter wavelength photons (the shorter the better). • The use of a narrow band filter will only sample a narrow band of the total represented scattered photons (meaning the derived magnitude loss will never equal the broadband magnitude loss). • Jupiter is an illuminated sphere with a constant intensity gradient (it makes a horrible backdrop!). It is essential to establish the trend of this background gradient trend to normalize Io’s atmosphere transit data. • Europa’s atmosphere covers 44% of Jupiter during transit making it impossible to derive a background trend. • Lots of pixel binning statistics increases S/N of JEE trends. • Io transits are ideal (if the appropriate wavelength light is sampled).

8. U2YHA305T_D0M_PC1.TIF Figure 2 Europa transit (Mallama, 2013) • Mallama (2013) Fig 2 of a Europa transit incorrectly identified the observed wavelength as being 410 nm when in fact it was 544 nm where little if any extinction would be detected (tiny detail #1 & #4). • No background trend was measured to normalize the derived “scanned luminosity” (tiny detail #3 & #5). • Mallama (2013) incorrectly states that the JEE Campaign claims a trend should be visible equivalent to the red line simulation when we have made no such claim for this Hubble image, nor would we based on the above facts.

9. Rayleigh and Mie scattering of photons 620-750 nm 590-620 nm U2YHA305T 544 nm 570-590 nm 495-570 nm 450-495 nm 380-450 nm 600 nm PARTICLE D0M_TIFF_DOCUMENT = "U2YHA305T_D0M_PC1.TIF" CENTER_FILTER_WAVELENGTH = 0.5439 <micron> BANDWIDTH = 0.1228 <micron> START_TIME = 1995-10-04 T20:19:16 STOP_TIME = 1995-10-04 T20:19:16 EXPOSURE_DURATION = 0.2 <SECOND>

10. U3AP0308T_D0M_PC1.TIF Figure 1 Io transit (Mallama, 2013) • Mallama (2013) Fig 1 of an Io transit incorrectly identified the observed wavelength as being 555 nm when in fact it was 409 nm making this Hubble image an ideal candidate to detect Io’s atmosphere. (tiny detail #1 & #6). • No background trend was measured to normalize the derived “scanned luminosity” (tiny detail #3 & #5). • Mallama (2013) incorrectly states that the JEE Campaign claims a 15% extinction trend should be visible equivalent to the red line simulation. HST used a very narrow bandwidth filter for this image, so the expected detected extinction would realistically be a fraction of the total broadband extinction (tiny detail #2).

11. Rayleigh and Mie scattering of photons 620-750 nm 590-620 nm 570-590 nm 495-570 nm 450-495 nm U3AP0308T 409 nm 380-450 nm 600 nm PARTICLE D0M_TIFF_DOCUMENT ="U3AP0308T_D0M_PC1.TIF" CENTER_FILTER_WAVELENGTH = 0.4088 <micron> BANDWIDTH = 0.0147 <micron> START_TIME = 1996-10-21T07:56:08 STOP_TIME = 1996-10-21T07:56:16 EXPOSURE_DURATION = 8. <SECOND>

12. background trend Use background to normalize Io atmosphere 16 Io radii width intensity profile 4 Io radii vertical binning HST Image U3AP0308T_DOM_PC1 http://pds-rings.seti.org/vol/HSTUx_xxxx/HSTU0_6452/DATA/VISIT_03/

13. The IMCCE Challenge Question: Why can’t I see JEE trends in PHEMU Campaign mutual events? Answer: Absolutely you can… you just have to do your homework!

14. Rayleigh and Mie scattering of photons PHEMU IMCCE Jovian mutual events 620-750 nm 590-620 nm 570-590 nm 495-570 nm 450-495 nm 380-450 nm 600 nm PARTICLE PHEMU Tech notes recommend using a V, R, or I filter to observe Jovian mutual events, with an emphasis on preferably using an R filter. This would exclude sampling of the wavelengths with dominant JEE scattering.

15. There are datasets in the IMCCE database that demonstrate JEE trends. Here are two random examples fitted with JPL Horizons ephemeris

16. Rayleigh and Mie scattering of photons PHEMU IMCCE Jovian mutual events X 620-750 nm 590-620 nm U2YHA305T 544 nm 570-590 nm X 495-570 nm 450-495 nm U3AP0308T 409 nm < 500 nm dominate JEE detectability 380-450 nm 600 nm PARTICLE SUMMARY

17. Typical wing data outside of the mutual event submitted to IMCCE is usually 6 to 10 minutes in length. Typical JEE measurements are observed 10s of minutes outside the mutual event. • Below highlights this omission of JEE data. The two lightcurves below are the exact same event. The one on the left is with 6 minutes of wing data while the one on the right is +/- 60 minutes of center of the mutual event.

18. Popular statement: “The 1971 occultation of Beta Scorpii C by Io showed no extinction trend. Therefore JEE can’t be real.”

19. 30.00 Look closely at the length of this lightcurve prior to ingress. It is 30 seconds of time. The extinction event would have begun 1260 seconds prior to ingress. 30 seconds is not enough data to resolve the miniscule magnitude change in only 30 seconds of time. In 30 seconds Beta Scorpii C only moved 0.2 Io radii. To detect a full extinction event you would have to go out 9 or more radii. If you can find data from May 14, 1971 that starts a minimum of 21 minutes prior (this is where Beta Scorpii C was at 9 Io radii), and preferably 30 minutes prior to get good wing data, then one can begin to look at this as a valid argument against JEE. This event has insufficient data to make an argument for or against JEE.

20. Uninformed statement #1 about data from video: “Video cannot provide accurate photometry.”

21. The Video Challenge I have issued a challenge to others to name a photometric target to pit video photometry against CCD photometry. To date no one has been willing to accept the challenge (so my scientific viewpoint is that you do not have a valid argument against video photometry if you aren’t willing to put it to the true test!). So I thought to myself, what is the hardest target I could think of to challenge myself…. How about an exoplanet transit?! Reply:

22. Video data (left) compared to CCD data (right) Exoplanet transit HD189733 b Ummm, I believe the answer for video is not only YES, the quality difference speaks for itself.

23. Video data compared to calibrated CCD FITS are identical in trend for the same type JEE.

24. Types of noise in video and how to reduce them: • Systematic noise (removed by subtracting calibration frame) • Hot pixels • Thermal gradient across CCD chip • Random noise (removed by binning multiple data points into one) • Electronic readout noise frame to frame • Random thermal photons • Electronic noise from internal circuit (“snow”) • NTSC video produce 29.97 frames per second, PAL @ 25 frames per second at 8 bit resolution. • Using carefully placed background and measurements apertures in video photometry reduction software such as LiMovie we obtain a background corrected photometry measurement for each individual frame. • To significantly reduce scintillation and other noise contributions we then bin 10 seconds of data into a single data point, i.e. 300 frames for NTSC or 250 frames for PAL into one point. • This yields 256 x 300 = 76,800 or > 16 bit statistical resolution, easily reaching 0.010 magnitude stnd dev. Video wins at photometry by the sheer volume of data greatly leveraging the statistics to increase the S/N!

25. Uninformed statement #2 of video data: “Video data cannot be calibrated.” With 5 simple lines of code in an AVISynth script I was able to subtract the calibration frame (left) from the raw video (top left) and create a new calibrated video (top right). (http://scottysmightymini.com/JEE/HowToCalibrateVideo.htm)

26. Sampling the intensity profile diagonally across the raw video shows the non-flat response (mostly due to thermal heat on the CCD array). • After subtracting the calibration image from the raw video a test of the same region of intensity shows the response has been flattened (and the previous slide shows the hot pixels and defects were also removed).

27. Uninformed statement #3 about JEE video: “The glare from Jupiter makes it impossible to get accurate photometry.” We routinely reduce lunar occultations near the bright limb of our moon by carefully configuring the background aperture to be tangent to the bright limb. The same applies with Jupiter. Note the yellow measurement aperture right up against Jupiter measures zero ADU.

28. Uninformed statement #4 about video data: “Gamma is the source of JEE trend in our lightcurves.” We have multiple lightcurves of simultaneous observations with some cameras with a gamma = 1 and another = 0.45, and both lightcurves demonstrate JEE dimming.

29. Uninformed statement #5 about video data: “Camera response from merging intensities cause our JEE trends.”

30. Some will declare allJEE invalid because a portion of a single video was saturated. It is easy to see where this lightcurve trend goes flat as the merging moons enter saturation.

31. We can toss out the saturated portion of this video and the JEE trend is still prominent.

32. Saturated data is rare in our data archives and is discarded when identified.

33. Saturation theory as source of extinction trend doesn’t apply here. This entire video is not in saturation. Note the peak from Europa shrinks in size relative to Io. This coincided with Europa passing behind Io line of sight.

35. There is no mechanism by which a camera can randomly pick the moon in back to diminish its intensity every time two moons merge intensities. All of our extinction trends involving two moons have been identified as directly linked to the moon behind another moon having a known atmosphere.

36. Combined photometry 80mm finderscope Separate photometry 14” Meade

37. I II III I+II • This event randomized every aspect of observation and reduction: • one large field of view to combine Io and Europa under one aperture and normalized to Ganymede • the other a small FOV to derive separate photometry of Io and Europa and then normalizing Europa by Io. • The geometry of the intensities on the CCDs are drastically different decoupling the JEE trend from Point Spread Function or any other type of detector response of merging intensities.

38. 20 Europa radii Europa • A conjunction with no merging moons (we have many of these). • 2 independent observers recorded the same JEE trend. • Inverting the lightcurve by superimposing the extinction data on JPL Horizon ephemeris shows the trend fits in our approximate 20 Europa radii atmosphere.

39. Europa 20 Europa radii Horizons ephemeris displaying path of probing objects behind Europa

40. Europa 20 Europa radii Extinction data tracing out approximately 20 radii atmosphere

41. Jovian Extinction Event data trends are not reporting anything new. • JEE data models match published data for expected detection. • JEE observing merely presents an alternative observing modality. • If the source of JEE data was just noise or other camera response we would not get the following consistent results from varied observers…

44. + • Schneider et al. “MUTUAL EVENT OBSERVATIONS OF IO'S SODIUM CORONA (figure 7)” Dividing the number of JEE scattered photons in the volume by the column depth we have derived a first order assumption (1 particle per 1 photon at about 3 radii): Io column density of approximately 1.11E+11 cm-2(+) (Our detected density matches published densities)

45. How to best observe a JEE • Download predictionsfrom JEE site: • This will give you the best way for planning when to start and stop an observing run. • Try to acquire up to 15 minutes (or longer) of data outside the anticipated time frame of JEE data. • If you use JME predictions from OccultWatcher just observe +/- 5 times the occultation duration of Io occulting a moon and +/- 10 times the occultation duration when Europa is occulting a moon. • Exposure: • Pixel intensity for the target moon and all reference moons should be between 40-80% of maximum intensity fill throughout the entire video. • If you camera does not have variable gain then you can use an aperture mask to dim the moons of interest. (an aperture mask is preferred over defocusing for JEE work). • Wavelength: • If you don’t have filter capability observe broadband unfiltered. • If you can only observe with one filter use B. • If you can do two or more alternate between R-B, V-B, or I-B. • Camera type: • Use a video camera with as high a frame rate as possible. • Use a CCD camera with as high an image cadence rate as your system can provide.