ASTR 340 Lecture 07 Notes

Context and Why This Matters

Lectures 04-06 modeled an idealized detector; every real detector also carries a fixed electronic offset, a thermal signal, and a non-uniform pixel-to-pixel response that must be measured and removed before any of those idealized equations can be applied to real data. This lecture supplies the calibration procedure every subsequent lab and problem set in this course assumes has already been performed.

Learning Goals

Describe the physical origin and purpose of bias, dark, and flat-field calibration frames.
Derive the standard CCD calibration equation converting a raw science frame into a calibrated, instrument-independent image.
Explain the role of photometric standard stars in converting calibrated counts into physical flux or magnitude units.

Opening Phenomenon

A raw, uncalibrated CCD frame of an astronomical target contains not just the astronomical signal but also a fixed electronic bias level present even in a zero-length exposure, an accumulated dark-current signal proportional to exposure time, and a pixel-to-pixel sensitivity pattern (vignetting, dust shadows, gain variations) that can differ by tens of percent across the field -- all three must be measured separately and removed before a raw frame becomes a scientific measurement.

Terms Developed in Context

These terms form the vocabulary of the lecture's reasoning arc; each is tied to the model, the working equation, and the evidence used to test the instrument.

bias frame

Definition and use: Bias frame denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: The calibration equation removes the bias and dark signal by subtraction and the flat-field pattern by division: I_cal = (I_raw - B - D)/F, where B is a master bias frame, D is a matched-exposure master dark frame (with its own bias already subtracted), and F is a normalized master...

Development in this lecture: Representative calibration-frame pixel levels used in this course’s lab (illustrative, not from a specific real exposure): bias ≈ 400 ADU, dark (300 s, cooled CCD) ≈ 480 ADU, flat ≈ 30,000 ADU (normalized before division), raw light frame ≈ 18,500 ADU -- exactly the four levels this lecture’s figure compares.

Connections: It is used with dark frame to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

Boundary or common confusion: Subtracting a bias frame and a dark frame separately when the dark frame already includes the bias pedestal (the standard master-dark workflow for a matched-exposure calibration, as corrected in this curriculum’s ASTR 210 course for a real camera). Double-subtracting the bias level produces a systematically negative offset in the calibrated frame; the...

dark frame

Definition and use: Dark frame denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: The calibration equation removes the bias and dark signal by subtraction and the flat-field pattern by division: I_cal = (I_raw - B - D)/F, where B is a master bias frame, D is a matched-exposure master dark frame (with its own bias already subtracted), and F is a normalized master...

Development in this lecture: Dark frames taken at matched exposure time to a science frame show a signal that scales linearly with exposure time and depends strongly on detector temperature, consistent with the thermal-generation model of dark current from Lecture 05.

Connections: It is used with bias frame, flat-field frame to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

flat-field frame

Definition and use: Flat-field frame denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: A master flat-field frame is built from exposures of a uniformly illuminated source (twilight sky or an illuminated dome screen), normalized to a mean of 1.0, so dividing by it removes multiplicative pixel-to-pixel sensitivity variations without altering the frame’s overall signal level.

Development in this lecture: The calibration equation removes the bias and dark signal by subtraction and the flat-field pattern by division: I_cal = (I_raw - B - D)/F, where B is a master bias frame, D is a matched-exposure master dark frame (with its own bias already subtracted), and F is a normalized master flat-field frame.

Connections: It is used with dark frame, master calibration frame to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

master calibration frame

Definition and use: Master calibration frame denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: The calibration equation removes the bias and dark signal by subtraction and the flat-field pattern by division: I_cal = (I_raw - B - D)/F, where B is a master bias frame, D is a matched-exposure master dark frame (with its own bias already subtracted), and F is a normalized master...

Development in this lecture: A master flat-field frame is built from exposures of a uniformly illuminated source (twilight sky or an illuminated dome screen), normalized to a mean of 1.0, so dividing by it removes multiplicative pixel-to-pixel sensitivity variations without altering the frame’s overall signal level.

Connections: It is used with flat-field frame, photometric standard star to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

photometric standard star

Definition and use: Photometric standard star denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: A photometric standard star of known magnitude, observed on the same night under the same conditions as the science target, lets an observer solve for the instrumental zero point (the magnitude a source of 1 count/second would have), converting the science target’s calibrated counts into...

Development in this lecture: Photometric standard-star networks (e.g., Landolt 1992 UBVRI standards) have been observed repeatedly, at many observatories over decades, and are found to have stable, internally consistent magnitudes to a few millimagnitudes, which is exactly the property that makes them usable as an absolute photometric reference for any instrument, anywhere.

Connections: It is used with master calibration frame, zero-point magnitude to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

zero-point magnitude

Definition and use: Zero-point magnitude denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: A photometric standard star of known magnitude, observed on the same night under the same conditions as the science target, lets an observer solve for the instrumental zero point (the magnitude a source of 1 count/second would have), converting the science target’s calibrated counts into...

Development in this lecture: Landolt standard star SA 98-978 (Landolt 1992, standard published UBVRI photometric standard): V = 9.061, B-V = 0.593; observing this star and measuring its instrumental count rate on a given night lets an observer solve Z_pt = V + 2.5log₁₀(counts/s) for that night’s zero point.

Connections: It is used with photometric standard star, flat-field residual to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

flat-field residual

Definition and use: Flat-field residual denotes the physical quantity, process, or instrument component captured by this lecture-specific statement: A master flat-field frame is built from exposures of a uniformly illuminated source (twilight sky or an illuminated dome screen), normalized to a mean of 1.0, so dividing by it removes multiplicative pixel-to-pixel sensitivity variations without altering the frame’s overall signal level.

Connections: It is used with zero-point magnitude to interpret the lecture model. The quantitative connection is the working relation \(I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt}\).

Evidence

A zero-length "bias" exposure (shutter closed, minimum possible integration time) reliably shows a nonzero, spatially structured pixel-value pattern that is stable and repeatable across many such exposures at a given observatory, confirming it reflects a real, characterizable electronic offset rather than random noise.
Dark frames taken at matched exposure time to a science frame show a signal that scales linearly with exposure time and depends strongly on detector temperature, consistent with the thermal-generation model of dark current from Lecture 05.
Photometric standard-star networks (e.g., Landolt 1992 UBVRI standards) have been observed repeatedly, at many observatories over decades, and are found to have stable, internally consistent magnitudes to a few millimagnitudes, which is exactly the property that makes them usable as an absolute photometric reference for any instrument, anywhere.

Derivation and Model

The calibration equation removes the bias and dark signal by subtraction and the flat-field pattern by division: I_cal = (I_raw - B - D)/F, where B is a master bias frame, D is a matched-exposure master dark frame (with its own bias already subtracted), and F is a normalized master flat-field frame.
A master flat-field frame is built from exposures of a uniformly illuminated source (twilight sky or an illuminated dome screen), normalized to a mean of 1.0, so dividing by it removes multiplicative pixel-to-pixel sensitivity variations without altering the frame’s overall signal level.
A photometric standard star of known magnitude, observed on the same night under the same conditions as the science target, lets an observer solve for the instrumental zero point (the magnitude a source of 1 count/second would have), converting the science target’s calibrated counts into a physically meaningful, catalog-comparable magnitude.

Working Equation

\[ I_{\rm cal} = \dfrac{I_{\rm raw} - B - D}{F}, \qquad m = -2.5\log_{10}(\text{counts/s}) + Z_{\rm pt} \]

Worked Example

Representative calibration-frame pixel levels used in this course’s lab (illustrative, not from a specific real exposure): bias ≈ 400 ADU, dark (300 s, cooled CCD) ≈ 480 ADU, flat ≈ 30,000 ADU (normalized before division), raw light frame ≈ 18,500 ADU -- exactly the four levels this lecture’s figure compares.
Landolt standard star SA 98-978 (Landolt 1992, standard published UBVRI photometric standard): V = 9.061, B-V = 0.593; observing this star and measuring its instrumental count rate on a given night lets an observer solve Z_pt = V + 2.5log₁₀(counts/s) for that night’s zero point.
If SA 98-978 is measured at 4868164.0 e⁻/s (illustrative count rate, scaled from this course’s Lecture 06 scenario for a brighter standard star), the zero point is Z_pt = 9.061 + 2.5×log₁₀(4868164.0) = 25.779 mag, the calibration constant then applied to every other target observed that same night.

Common Misconception

Subtracting a bias frame and a dark frame separately when the dark frame already includes the bias pedestal (the standard master-dark workflow for a matched-exposure calibration, as corrected in this curriculum’s ASTR 210 course for a real camera). Double-subtracting the bias level produces a systematically negative offset in the calibrated frame; the correct model is I_cal=(raw-D)/F when D is a matched-exposure master dark that already includes the bias pedestal, or I_cal=(raw-B-D)/F only when D is a separately bias-subtracted dark-current-only frame -- the two workflows must not be mixed.

Active-Learning Guidance

Explain why a flat-field frame must be normalized to a mean of 1.0 before being used as a divisor, and what would happen to the calibrated image’s overall brightness scale if an un-normalized flat field (mean far from 1.0) were used instead.

Lab Connection

Lab 04 performs a full bias/dark/flat calibration sequence on a set of frames and derives a photometric zero point from a standard star, directly applying every equation in this lecture.

Synthesis Questions

What was measured directly, or taken from a published instrument specification, in this lecture’s worked example, and what was derived from a physical law?
Which assumption in the derivation would most change the interpretation if it were wrong?
How does this lecture’s technique connect to the lab and problem set that follow it?

References

OpenStax Astronomy 2e (local extracted reference copy: references/openstax-astronomy-2e-extracted.txt) covers observational instrumentation only descriptively (Chapter 6, astronomical instruments); this course derives the optics, detector, and error-budget physics explicitly, well beyond that introductory treatment.
Every numeric result above is computed programmatically in materials/ASTR340/src/generate_astr340_content.py, not hand-typed.