The Final Flux Calibration
of the STIS Echelle Modes
Alessandra Aloisi
(ESA/STScI)
• Description of the issues involved
• Data analysis and results
• Pipeline tools implemented for corrections
TIPS Meeting – 19 October 2006
1
The Final Flux Calibration
of the STIS Echelle Modes
Alessandra Aloisi
(ESA/STScI)
In collaboration with R. Bohlin & J. Kim-Quijano
• Description of the issues involved
• Data analysis and results
• Pipeline tools implemented for corrections
TIPS Meeting – 19 October 2006
1
Outstanding Issues
• Time dependent sensitivity
• Blaze function characterization
 Shifts vs. location (x,y) & t
 Shape vs. location (x,y) & t
• Updated F calibration of prime modes
• On-orbit F calibration of secondary modes
TIPS Meeting – 19 October 2006
2
Time Dependent Sensitivity (TDS)
• Sensitivity
of MAMAs
decreases
with time
(contaminants)
BD+28D4211 — E140M (1425)
1997.7
2000.9
2004.0
F errors < 15%
TIPS Meeting – 19 October 2006
3
Time Dependent Sensitivity (TDS)
• Analysis of Spectra with same MO & different t
= BD+28D4211 MO = (0,0)
t = 1997.7–2004
Prime Modes E140H (1416) E140M (1425)
E230H (2263)
E230M (1978, 2707)
 Results
* Similar trend as first-order modes
TIPS Meeting – 19 October 2006
4
Time Dependent Sensitivity (TDS)
• Analysis of Spectra with same MO & different t
= BD+28D4211 MO = (0,0)
t = 1997.7–2004
Prime Modes E140H (1416) E140M (1425)
E230H (2263)
E230M (1978, 2707)
 Results
* Similar trend as first-order modes
G140L
E140M (1425)
TIPS Meeting – 19 October 2006
4
Time Dependent Sensitivity (TDS)
• Analysis of Spectra with same MO & different t
= BD+28D4211 MO = (0,0)
t = 1997.7–2004
Prime Modes E140H (1416) E140M (1425)
E230H (2263)
E230M (1978, 2707)
 Results
* Similar trend as first-order modes
* Some scatter due to:
temperature ?
mis-centering in small aperture ?
blaze shift correction ?
TIPS Meeting – 19 October 2006
4
Time Dependent Sensitivity (TDS)
• TDS correction
from first-order
modes
• New TDS files
delivered to
the pipeline
 Apr. 2005
 F uncertainty < few %
TIPS Meeting – 19 October 2006
BD+28D4211 — E140M (1425)
FOS
no TDS correction
TDS correction
2004.0
5
Blaze Function (BF)
BD+28D4211 — E140M (1425)
• BF leads to
ripples in
calibrated
spectra
1998
 saw-tooth F
calibration
errors < 20%
TIPS Meeting – 19 October 2006
6
Blaze Function (BF)
BD+28D4211 — E140M (1425)
• BF leads to
ripples in
calibrated
spectra
2000.4
 saw-tooth F
calibration
errors < 20%
TIPS Meeting – 19 October 2006
6
Blaze Function (BF)
BD+28D4211 — E140M (1425)
• BF leads to
ripples in
calibrated
spectra
2003.4
 saw-tooth F
calibration
errors < 20%
TIPS Meeting – 19 October 2006
6
BF Variations
• Changes with time (contaminants & optics)
& location on the detector (MOs)
 What type of changes ?
* BF Shift
* BF Shape
 What causes the changes ?
* Effects due to location (x,y)
* Effects due to time (t)
TIPS Meeting – 19 October 2006
7
BF Shift
• MO produces BF shift
• Algorithm already implemented into pipeline
Bowers & Lindler (2002)
TIPS Meeting – 19 October 2006
8
BF Shift
• MO produces BF shift
• Algorithm already implemented into pipeline
Bowers & Lindler (2002)
TIPS Meeting – 19 October 2006
8
BF Shift
Parameter Model
• MO produces BF2shift
BZS = A x + B y
• Algorithm already implemented into pipeline
Bowers & Lindler (2002)
TIPS Meeting – 19 October 2006
8
BF Shift
Parameter Model
• MO produces BF2shift
Residuals
BZS = Avs.
xt + B y
• Algorithm already implemented into pipeline
Bowers & Lindler (2002)
TIPS Meeting – 19 October 2006
8
BF Shift
Parameter Model
• MO produces BF2shift
Residuals
vs. t= A Model
BLZSHIFT
x + B y
3 Parameter
• Algorithm already BZS
implemented
into pipeline
= A x+B y+C
t
Bowers & Lindler (2002)
TIPS Meeting – 19 October 2006
8
BF Shift
• MO produces BF shift
• Algorithm already implemented into pipeline
Bowers & Lindler (2002)
 CALSTIS 2.13b (Sep. 2002)
 BZS = A x + B y + C t
x & y from wavecals
A, B, & C from fit of data (1997-2001)
• MO switched off in Aug 2002
TIPS Meeting – 19 October 2006
8
BF Shape with Location
• Analysis of Spectra with different MO & same t
= BD+28D4211
5 MOs
Prime Mode E230H (2513)
 Results
t = Jul 2002
* No change in BF shape with location
* Average BF shift similar to pipeline BZS
once MO = (0,0) taken as reference
correct spatial BZS component
* No strong dependence on the order
TIPS Meeting – 19 October 2006
9
BF Shape with Location
E230H (2513)
• Analysis of Spectra with different MO & same t
= BD+28D4211
5 MOs
Prime Mode E230H (2513)
 Results
t = Jul 2002
* No change in BF shape with location
* Average BF shift similar to pipeline BZS
once MO = (0,0) taken as reference
correct spatial BZS component
* No strong dependence on the order
TIPS Meeting – 19 October 2006
9
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
E140M (1425) – order 95
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
 Results
* No change in BF shape with time
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
E140M (1425)
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
* Clear break between Side 1 and Side 2 operations
linear extrapolation of pipeline wrong
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for
TDS (1425)
E140M
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
* Clear break between Side 1 and Side 2 operations
linear extrapolation of pipeline wrong
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
* Clear break between Side 1 and Side 2 operations
linear extrapolation of pipeline wrong
* Clear (linear) dependence of BF shift on order
average BF shift not correct
TIPS Meeting – 19 October 2006
10
BF with Time
• Analysis of Spectra with same MO & different t
Same datasets used for TDS
 Results
* No change in BF shape with time
* BF shift linear with time
(once spatial component subtracted)
* Clear break between Side 1 and Side 2 operations
linear extrapolation of pipeline wrong
* Clear (linear) dependence of BF shift on order
average BF shift not correct
* Similar behavior of E230M(1978) & E230M(2707)
extrapolation to other modes
TIPS Meeting – 19 October 2006
10
Implementation of New
Blaze Shift Correction
E140H (1416)
• Linear fit vs time
BZS =  t + 
for every order
• Fit with zero- or firstorder polynomial of
 and  coeff. vs order
Side 1 & Side 2
TIPS Meeting – 19 October 2006
11
Implementation of New
Blaze Shift Correction
E140H (1416)
Constant
Linear Coeff.
E140H (1416)
Order
TIPS Meeting – 19 October
2006
E140M (1425)
Linear Coeff.
Constant
E140M (1425)
Order
12
Implementation of New
Blaze Shift Correction
E230H (2263)
Constant
Linear Coeff.
E230H (2263)
Order
TIPS Meeting – 19 October
2006
E230M (1978)
Linear Coeff.
Constant
E230M (1978)
Order
12
Implementation of New
Blaze Shift Correction
E230M (2707)
Constant
Linear Coeff.
E230M (2707)
Order
TIPS Meeting – 19 October
2006
E230M (1978)
Linear Coeff.
Constant
E230M (1978)
Order
12
Implementation of New
Blaze Shift Correction
• Implementation
of new BZS vs
order and side
• New PHT files
will be delivered
to the pipeline
 delivery next week
BD+28D4211 – E230M (1978)
FOS
Pipeline BZS
New BZS
2004.4
 F uncertainty from
~10% to < 1-2 %
TIPS Meeting – 19 October 2006
13
Updated Flux Calibration
of Prime Modes
• 2D algorithm for Scattered Light implemented
into the pipeline
Lindler & Bowers (2002)
 CALSTIS 2.9
(Dec. 2000)
 F uncertainty < 2%
(absorption lines)
ISR STIS 02-01
TIPS Meeting – 19 October 2006
14
Updated Flux Calibration
of Prime Modes
• Re-determination of Echelle flux calibration
for self-consistency
 no changes due to SL
 2-7 % due to other
• New PHT files
delivered to
the pipeline
BD+28D4211 — E140M (1425)
FOS
old PHT file
new PHT file
1997.7
 Apr. 2005
 F uncertainty < few%
TIPS Meeting – 19 October 2006
15
Updated Flux Calibration
of Prime Modes
• Multiple observations of 3 different standard stars
BD+75D325, BD+28D4211 & G191B2B at MO = (0,0)
< few % uncertainties introduced by:
averaging over different observation epochs (≠TDS and B shift)
use of standard stars with a lot of absorption lines
BD+75D325 & BD+28D4211
• Switched to a calibration that only averages over
exposures of the most suitable standard star
G191B2B, taken at the same time with MO ≠ (0,0)
• Same PHT file with new BZS correction
TIPS Meeting – 19 October 2006
16
On-Orbit Flux Calibration
of Secondary Modes
• Secondary Modes almost used
as often as Prime Modes
• Absolute F calibration of Secondary
Modes still based on pre-launch data
• TDS applied, but not BZS correction
TIPS Meeting – 19 October 2006
17
On-Orbit Flux Calibration
of Secondary Modes
• On-board data collected for G191B2B with MO≠(0,0)
All Prime & Secondary
Modes (Sep. 2001)
• On-orbit sensitivities
and BZS correction
introduced for
secondary modes
with new PHT files
 relative F uncertainty
from ~ 30% to < 1-2 %
TIPS Meeting – 19 October 2006
old PHT file
new PHT file
2003.4
HD158643 (pre-MS ) – E140H (1380)
18
ISRs in Preparation
“Time-Dependent Sensitivity of STIS Echelle Medium
and High-Resolution Modes” A. Aloisi
“Blaze Shift Correction for STIS Echelle Prime and
Secondary Modes” A. Aloisi
“On-Board Sensitivity Calibration of STIS Echelle Secondary
Modes (and Revision of Prime Mode Sensitivities)”
A. Aloisi, R. Bohlin, & J. Kim Quijano
TIPS Meeting – 19 October 2006
19