Lecture 02: Observing Planning and Error Budgets

ASTR 210 Observational Astronomy and Data Reduction

Context and Why This Matters

Telescope time is scarce and expensive, and a plan built without an error budget fails quietly: the data come back, but they cannot answer the question that was asked. Professional proposals are reviewed partly on whether the requested exposure times and target altitudes are realistic.

Learning Goals

  1. Compute airmass from altitude and estimate the resulting extinction penalty.
  2. Identify the planning decisions (target altitude, exposure time, calibration cadence) that dominate a night’s success.
  3. Distinguish a planning error from a reduction error in a failed observation.

Opening Phenomenon

A target visible at only 35° altitude looks like a fine plan on a star chart, but the light must pass through 1.74 times as much atmosphere as it would at the zenith, dimming the source by roughly 0.26 magnitudes before it even reaches the detector.

Terms Developed in Context

These terms form the measurement chain for Observing Planning and Error Budgets: each connects an observable to a model choice, and each has a boundary that must be respected when making a scientific claim.

airmass

The atmospheric path length toward a target relative to the path at the zenith. Airmass predicts how strongly extinction and atmospheric systematics affect the planned exposure.

In the plane-parallel approximation \(X\approx\sec z\), and extinction contributes \(\Delta m=kX\). The secant approximation becomes inaccurate near the horizon, where curvature and refraction matter.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

zenith angle

The angular separation \(z\) between the target and the point directly overhead; it equals 90° minus altitude. It is the geometric input used to estimate airmass and therefore the target’s atmospheric penalty.

As \(z\) grows, \(\cos z\) decreases and \(X\approx1/\cos z\) rises steeply. Zenith angle is not the same as altitude; confusing them reverses the observing geometry.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

extinction coefficient

The magnitude attenuation per unit airmass in a specified passband under stated atmospheric conditions. It turns target geometry into an expected loss of flux and helps decide whether an observing window is viable.

The first-order model is \(m_{obs}=m_0+kX\), with \(k\) inferred from standards observed over a range of airmass. The coefficient is not universal: wavelength, aerosols, water vapor, and time can change it.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

exposure time

The interval during which a detector integrates incident photons for one recorded frame. It sets accumulated signal, saturation risk, time resolution, and part of the noise budget.

Source and sky counts usually scale with time, while read noise is incurred once per exposure; this creates an optimization rather than a simple longer-is-better rule. Longer exposure does not remove systematic errors and can blur time variability or saturate bright sources.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

overhead

Observing time spent on readout, slewing, acquisition, focusing, calibration, or file handling rather than photon collection. Ignoring overhead makes a schedule impossible even when the summed exposure times fit the night.

Efficiency is \(t_{exp}/(t_{exp}+t_{overhead})\), so short exposures can lose a large fraction of the available time. Overhead is not wasted time when it is required for calibration or reliable target acquisition.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

calibration cadence

The planned frequency and timing with which standards or calibration frames are obtained. It determines whether changing atmosphere and instrument response can be tracked during the science sequence.

Calibration cadence should be shorter than the timescale on which zero point, wavelength solution, or detector state drifts. One calibration at the start of a night cannot diagnose changes that occur later.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

observing window

The interval when a target satisfies the project’s altitude, darkness, timing, and instrument constraints. It converts a target’s visibility into the subset of time when useful data can actually be acquired.

The window is set by hour angle and airmass together with twilight, Moon, cadence, and required calibration time. Being above the horizon is not equivalent to being scientifically observable.

Developed in this lecture: The term helps interpret this evidence: Standard-star measurements repeated across an night at different altitudes show systematically fainter magnitudes at low altitude even though the star has not changed. It also constrains the working model: Airmass X approximates sec(z), where z is the zenith angle; X = 1 at the zenith and grows quickly below about 30° altitude.

Evidence

Model

Working Equation

\[ X \approx \sec z = \dfrac{1}{\cos z}, \qquad \Delta m_{\rm ext} = k\,X \]

Worked Example

  1. A target is at altitude 35°, so the zenith angle is z = 90° − 35° = 55°.
  2. Airmass: X = 1/cos(55°) = 1.74.
  3. Using a typical V-band extinction coefficient k = 0.15 mag/airmass, the extra dimming is 0.15 × 1.74 = 0.26 mag relative to the zenith.
  4. A 0.26-magnitude penalty is a roughly 24% flux loss — large enough to change whether a faint target is detectable at all.

Common Misconception

Choosing an exposure time based only on the target’s catalog brightness, ignoring sky brightness, airmass, and detector read noise, all of which change the achievable signal-to-noise.

Active-Learning Guidance

Given a target rising through the night, plot airmass versus time and identify the observing window where airmass stays below 2.0.

Lab / Observing Connection

Lab 01 opens with a short planning checklist before any calibration arithmetic begins, so a bad plan is caught before data are "collected."

Synthesis Questions

References