KVN as a Pathfinder for the ngVLA Richard Dodson 1 Maria Rioja - - PowerPoint PPT Presentation

1- ICRAR/UWA 2- CASS/CSIRO 3- OAN/IGN

KVN as a Pathfinder for the ngVLA

Richard Dodson1 Maria Rioja1,2,3

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

2

next generation VLA is the refresh of Radio Astronomy’s top observatory

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

2

next generation VLA is the refresh of Radio Astronomy’s top observatory Will connect the mm and sub-mm

• bservations of

ALMA to the cm regime.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

2

next generation VLA is the refresh of Radio Astronomy’s top observatory Will connect the mm and sub-mm

• bservations of

ALMA to the cm regime. Will span 1-116GHz, therefore fill the role

• f SKA-High

and SKA-mid!

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

2

SKA-2

next generation VLA is the refresh of Radio Astronomy’s top observatory Will connect the mm and sub-mm

• bservations of

ALMA to the cm regime. Will span 1-116GHz, therefore fill the role

• f SKA-High

and SKA-mid!

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

2

SKA-2

next generation VLA is the refresh of Radio Astronomy’s top observatory Will connect the mm and sub-mm

• bservations of

ALMA to the cm regime. Will span 1-116GHz, therefore fill the role

• f SKA-High

and SKA-mid! The proposal to be submitted after the decadal review. But will not be under SKA-O

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth Astro-Chemistry: Focus on biogenic molecules, test chirality

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth Astro-Chemistry: Focus on biogenic molecules, test chirality Galaxy Assembly: Tracing gas content in CO, HI

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth Astro-Chemistry: Focus on biogenic molecules, test chirality Galaxy Assembly: Tracing gas content in CO, HI Pulsars: In strong gravity regime frequency range allows views deep into G.C.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth Astro-Chemistry: Focus on biogenic molecules, test chirality Galaxy Assembly: Tracing gas content in CO, HI Pulsars: In strong gravity regime frequency range allows views deep into G.C. Black Holes: Black Hole Hunter to detect the number of binary BHs. Compare with LIGO results

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

3

Key Science Goals of ngVLA Planetary Disks: Follow on from ALMA, higher resolution larger dust grains, lower frequencies & optical depth Astro-Chemistry: Focus on biogenic molecules, test chirality Galaxy Assembly: Tracing gas content in CO, HI Pulsars: In strong gravity regime frequency range allows views deep into G.C. Black Holes: Black Hole Hunter to detect the number of binary BHs. Compare with LIGO results

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

On longer baseline the atmospheres (>VLA site) will be decorrelated - ngVLA is a VLBI machine. Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

On longer baseline the atmospheres (>VLA site) will be decorrelated - ngVLA is a VLBI machine. KVN offers a good platform to investigate methods and technologies for the high frequencies

Proposed Five Band Feed

Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

ngVLA AKA SKA-High

4

On longer baseline the atmospheres (>VLA site) will be decorrelated - ngVLA is a VLBI machine. KVN offers a good platform to investigate methods and technologies for the high frequencies

Proposed Five Band Feed Plan on fast switching

Baseline ngVLA covers New Mexico (~500 to 1000km) with 214, 18m, offset-Gregorian antennas. A dense core will cover the VLA site. The `Long Baseline’ enhancement replaces the VLBA, providing continental baselines and sub-mas resolution.

East Asia VLBI Workshop\, 2018

Goal of the Comm. Study

High freq. long baseline interferometry is very interesting & very difficult Sensitivity limitations come from high SEFD and short coherence times.

• Std. phase referencing (for increased sensitivity)

impossible Few sources can be studied (using self-calibration) Astrometry is unachievable so relationship to other sources/freq., motion on the sky, etc. can not be derived We have been tackling these (and other questions) to develop innovative calibration methods (see extra slide) Apply these to the ngVLA model

5

East Asia VLBI Workshop\, 2018

Frequency Phase Transfer

Correct the difficult mm-frequencies. Using phase solutions from easy lower cm-frequencies. For non-dispersive (tropospheric) terms simply just scale. This skips a lot of details! Full solution is called Source/Frequency Phase Referencing (SFPR) Two possible approaches: Fast Freq. Switching or Simultaneous Multi-band Two Radio Interferometers: Very Long Baseline Array & Korean VLBI Network

6

East Asia VLBI Workshop\, 2018

Frequency Phase Transfer

Correct the difficult mm-frequencies. Using phase solutions from easy lower cm-frequencies. For non-dispersive (tropospheric) terms simply just scale. This skips a lot of details! Full solution is called Source/Frequency Phase Referencing (SFPR) Two possible approaches: Fast Freq. Switching or Simultaneous Multi-band Two Radio Interferometers: Very Long Baseline Array & Korean VLBI Network

6

East Asia VLBI Workshop\, 2018

KVN Frequency Setup

7

22GHz 43GHz 86GHz 129GHz LPF1 LPF2 LPF3 Ellipsoidal Mirrors 1 Ellipsoidal Mirror 2 Ellipsoidal Mirror 3 Beams from antenna 22,43,86,129GHz 22GHz 43,86,129GHz 86,129GHz 129GHz 43GHz 86GHz

Korean VLBI Network has an innovative optical system that allows simultaneous observations.

East Asia VLBI Workshop\, 2018

KVN Frequency Setup

7

22GHz 43GHz 86GHz 129GHz LPF1 LPF2 LPF3 Ellipsoidal Mirrors 1 Ellipsoidal Mirror 2 Ellipsoidal Mirror 3 Beams from antenna 22,43,86,129GHz 22GHz 43,86,129GHz 86,129GHz 129GHz 43GHz 86GHz

Korean VLBI Network has an innovative optical system that allows simultaneous observations.

East Asia VLBI Workshop\, 2018

VLBA Frequency Setup

8

In comparison: Very Long Baseline Array has rapid switching between all the receivers. Facilitates for lots of interesting science. But also allows us to compare switching and simultaneous observational strategies.

East Asia VLBI Workshop\, 2018

Increased coherence times

Detection SNR for 3C279 at 129GHz for single baseline on KVN.

• Freq. Phase Transfer

provides good coherence times > 20min Source/Freq. Phase Referencing, or FTP- Squared, does even better > 1 day

9

(From Jung 12, Rioja 15, Dodson 17,Zhao 18)

East Asia VLBI Workshop\, 2018

Increased coherence times

Detection SNR for 3C279 at 129GHz for single baseline on KVN.

• Freq. Phase Transfer

provides good coherence times > 20min Source/Freq. Phase Referencing, or FTP- Squared, does even better > 1 day

9

(From Jung 12, Rioja 15, Dodson 17,Zhao 18)

East Asia VLBI Workshop\, 2018

Increased coherence times

Detection SNR for 3C279 at 129GHz for single baseline on KVN.

• Freq. Phase Transfer

provides good coherence times > 20min Source/Freq. Phase Referencing, or FTP- Squared, does even better > 1 day

9

10 10

1

10

2

10

3

0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Solution Time Interval (minutes) Fractional Flux Recovery

(From Jung 12, Rioja 15, Dodson 17,Zhao 18)

SFPR (or FPT2) FPT

East Asia VLBI Workshop\, 2018

Masers in AGB Stars (KVN)

Direct astrometric registration between (non-integer freq. ratio) maser transitions for R- Leo Min 22 (H2O maser) to 42.8,43.1 (SiO masers) GHz around AGB star

10

(From Dodson 2017, Yoon ’18 (Nature!)

East Asia VLBI Workshop\, 2018

Masers in AGB Stars (KVN)

Direct astrometric registration between (non-integer freq. ratio) maser transitions for R- Leo Min 22 (H2O maser) to 42.8,43.1 (SiO masers) GHz around AGB star

10

(From Dodson 2017, Yoon ’18 (Nature!)

22 (H2O maser) to 42.8, 43.1, 86, 129 GHz (SiO maser) around AGB star VX Sgr

East Asia VLBI Workshop\, 2018

Masers in AGB Stars (KVN)

Direct astrometric registration between (non-integer freq. ratio) maser transitions for R- Leo Min 22 (H2O maser) to 42.8,43.1 (SiO masers) GHz around AGB star

10

(From Dodson 2017, Yoon ’18 (Nature!)

22 (H2O maser) to 42.8, 43.1, 86, 129 GHz (SiO maser) around AGB star VX Sgr

East Asia VLBI Workshop\, 2018

Masers in AGB Stars (KVN)

Direct astrometric registration between (non-integer freq. ratio) maser transitions for R- Leo Min 22 (H2O maser) to 42.8,43.1 (SiO masers) GHz around AGB star

10

(From Dodson 2017, Yoon ’18 (Nature!)

22 (H2O maser) to 42.8, 43.1, 86, 129 GHz (SiO maser) around AGB star VX Sgr

East Asia VLBI Workshop\, 2018

Masers in AGB Stars (KVN)

Direct astrometric registration between (non-integer freq. ratio) maser transitions for R- Leo Min 22 (H2O maser) to 42.8,43.1 (SiO masers) GHz around AGB star

10

(From Dodson 2017, Yoon ’18 (Nature!)

22 (H2O maser) to 42.8, 43.1, 86, 129 GHz (SiO maser) around AGB star VX Sgr

P r

• b

e A s t r

• c

h e m i s t r y a s I S M i s s e e d e d

East Asia VLBI Workshop\, 2018

Masers in Proto-Planetary Nebulae (KVN)

Phase Referencing between frequencies provides: Registration of SiO (surrounding AGB) to water masers gives 3D structure of the archetypical Proto-Planetary Nebula

11

(From Dodson 18)

HST image of Calabash Nebula

Water masers in lobes, SiO around QX Pup Water maser expansion

• ver 17 years of obs.

East Asia VLBI Workshop\, 2018

Masers in Proto-Planetary Nebulae (KVN)

Phase Referencing between frequencies provides: Registration of SiO (surrounding AGB) to water masers gives 3D structure of the archetypical Proto-Planetary Nebula

11

(From Dodson 18)

HST image of Calabash Nebula

Water masers in lobes, SiO around QX Pup Water maser expansion

• ver 17 years of obs.

East Asia VLBI Workshop\, 2018

Masers in Proto-Planetary Nebulae (KVN)

Phase Referencing between frequencies provides: Registration of SiO (surrounding AGB) to water masers gives 3D structure of the archetypical Proto-Planetary Nebula

11

(From Dodson 18)

O u t

• fl
• w

s a n d e v

• l

u t i

• n
• f

P P N s

HST image of Calabash Nebula

Water masers in lobes, SiO around QX Pup Water maser expansion

• ver 17 years of obs.

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

MHD Simulations

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

MHD Simulations Predict deviation for B&K optical depth core-shift model

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

MHD Simulations Predict deviation for B&K optical depth core-shift model Perfect match to

• bservations

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

MHD Simulations Predict deviation for B&K optical depth core-shift model Perfect match to

• bservations

Made possible by VLBA freq agility: Fast switching between 22/43/86GHz Slower switching between 22/6/1.4GHz

East Asia VLBI Workshop\, 2018

Standing Shocks in AGNs (VLBA)

12

(From Dodson 2017, Molina 17)

Phase Referencing purely between frequencies: To uncover the transition from B&K core-shift to unveil the standing shock, for BL-Lac

MHD Simulations Predict deviation for B&K optical depth core-shift model Perfect match to

• bservations

Made possible by VLBA freq agility: Fast switching between 22/43/86GHz Slower switching between 22/6/1.4GHz

J e t P h y s i c s

• f

A G N s

East Asia VLBI Workshop\, 2018

Jet Physics of X-ray binaries (VLA)

Some signals change very fast … Shown are VLA observations of V404 Cygni in out-burst. Sub-minute differences between the light-curves carries information on the jet-width as a function of distance down the jet. Only sim observations will be able to follow this at mm-frequencies

13

(From Tetarenko `17)

East Asia VLBI Workshop\, 2018

Jet Physics of X-ray binaries (VLA)

Some signals change very fast … Shown are VLA observations of V404 Cygni in out-burst. Sub-minute differences between the light-curves carries information on the jet-width as a function of distance down the jet. Only sim observations will be able to follow this at mm-frequencies

13

(From Tetarenko `17)

p r

• b

e

• f

j e t s p e e d , s t r u c t u r e , e n e r g e t i c s , a n d g e

• m

e t r y

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

Increasing errors with freq.

Increasing errors with switching time.

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy Over plotted are Flux Recovery

X X

22/130 30sec 22/90 GHz 30sec

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

Increasing errors with freq.

Increasing errors with switching time.

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy

43/86GHz VLBA Rioja `11 22/43 GHz VLBA Rioja `14

Over plotted are Flux Recovery

X X

22/130 30sec 22/90 GHz 30sec

+ real results

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

Increasing errors with freq.

Increasing errors with switching time.

East Asia VLBI Workshop\, 2018

ngVLA simulation results

14

What are the observational losses from fast freq. switching? Characterising the performance of switching Use delays and rates to predict the next solution Scale for freq. and find no. > than 1/2 cycle error

20s 30s 40s 60s

Thermal noise-free case, but corrected phases are noisy

43/86GHz VLBA Rioja `11 22/43 GHz VLBA Rioja `14 22/43 GHz KVN Rioja `14

Over plotted are Flux Recovery

X X

22/130 30sec 22/90 GHz 30sec

+ real results

Error between Predicted Phase and Actual Phase

Phase Error (deg)

Time (min)

Increasing errors with freq.

Increasing errors with switching time.

43/130 GHz KVN Rioja `15

SFPR limitation in VLBI

SFPR works best when the frequency ratio (of the spectral reference point) is an integer. So — to avoid problems best that R is INTEGER or N is ZERO We don’t want limited frequency coverage, so we should ensure we can track the fringe phase. That is phase rate < 3E-13 for 100GHz & 30sec cycle Typical AllanStdDev 1E-13 — will loose good fraction of data Best solution is cycle time is zero .. but that is not the ngVLA design

SFPR limitation in VLBI

SFPR works best when the frequency ratio (of the spectral reference point) is an integer. So — to avoid problems best that R is INTEGER or N is ZERO We don’t want limited frequency coverage, so we should ensure we can track the fringe phase. That is phase rate < 3E-13 for 100GHz & 30sec cycle Typical AllanStdDev 1E-13 — will loose good fraction of data Best solution is cycle time is zero .. but that is not the ngVLA design

SFPR limitation in VLBI

SFPR works best when the frequency ratio (of the spectral reference point) is an integer. So — to avoid problems best that R is INTEGER or N is ZERO We don’t want limited frequency coverage, so we should ensure we can track the fringe phase. That is phase rate < 3E-13 for 100GHz & 30sec cycle Typical AllanStdDev 1E-13 — will loose good fraction of data Best solution is cycle time is zero .. but that is not the ngVLA design

SFPR limitation in VLBI

SFPR works best when the frequency ratio (of the spectral reference point) is an integer. So — to avoid problems best that R is INTEGER or N is ZERO We don’t want limited frequency coverage, so we should ensure we can track the fringe phase. That is phase rate < 3E-13 for 100GHz & 30sec cycle Typical AllanStdDev 1E-13 — will loose good fraction of data Best solution is cycle time is zero .. but that is not the ngVLA design

10 second cycle

Possible StrawMan System (?)

16

• Freq. Selective

Grating @ 50GHz 30—50GHz feed Incoming cm & mm frequencies

Quasi-optical system from KVN allows separation of

• freq. bands and therefore simultaneous co-observation.

Both allow precise calibration and therefore the target science. Very fast switching (5sec) with precise timing may also work

70—116GHz feed Feeds for >1cm

Conclusions:

Simultaneous multifreq observations allows for more and better science ngVLA performance will benefit greatly from this configuration Loss of signal == Fraction of observing time Taking this in to account the costs for Multi- Freq receiver are minimal in comparison We continue to strongly recommend that this

• ption is fully explored; our experience favours

the KVN-style over the VLBA-style

18

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11

East Asia VLBI Workshop\, 2018

Functional analysis of residual contributions

Atmosphere has Troposphere at ~10km and Ionosphere at ~100km (<8GHz) Both dynamic (fast changing) and static (slow changing) terms Dependent on Angular Sep, Residual Zenith path, Residual TEC and Switching time

19

Resisdual phase errors (deg) Static Contributions Dynamic Contributions

Troposphere Ionosphere

Asaki ’07 + Rioja ‘11