NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Data > TWikiUsers > RonMaddalena > CalibrationMeetings > CalMinutes22Feb06

Changes | Index | Contents | Search | Statistics | Jump to Topic

Calibration Meeting for February 22, 2006

Proposed agenda:

• Exp(tau*AirMass) atmospheric correction and associated uncertainties
• Tippping Primer
  • various models and approximations
  • limitations of the models
  • limitations of the measurements
• Air Mass estimations
• Tatm estimations used by Tippings

Minutes:

(1) Overall Summary:

We spent most of the time discussing various tippings models and their limitations and never got to the last two agenda items. These will be on the agenda for the next meeting.

We also brought up various related topics (linearities, calibration using planets or planetary nebula, forecasted opacities vs that derived from tippings, ...). I'm planning to devote future meetings to these topics. For now, they are noted as being hot topics that the group wants to discuss. Since we'll be returning to these other topics, I think it is best that I do not summarize here our not-yet-completed discussions

(2) Exp(tau*AirMass) atmospheric correction and associated uncertainties:

To give guidelines to how accurately one needs to estimate the atmospheric corrections, the percentage errors in Tr* (or S) is related to errors in the opacity and air mass estimates:

  deltaTr*/Tr* = tau*delta_airMass
  deltaTr*/Tr* = airMass * delta_Tau

Using the assumptions that observers will seldom observe when tau*airMass >~ 1, and that one wants to achieve a calibration accuracy of n percent (I'll assume 1%), we can derive the accuracy we need for estimates of airMass and Tau.

  delta_AirMass < 10*n ~ 0.1
  delta_Tau < n ~ 0.01

(3) Tippings

• Presented with three straw-man tipping models, each is making some broad assumptions but each is useful for many observing conditions.:
  • Tsys = Trcvr + Tatm*Tau*AirMass
  • Tsys = Trcvr + Tatm*(1 - exp(-Tau*AirMass))
  • The Kutner/Ulich model summarized at: http://www.gb.nrao.edu/~rmaddale/GBT/TipperAips++.pdf
• These basic models have room for improvement. For example, add Tspill, forward efficiencies, and Tbg to the second model.
• Our lack of a model for Tspill limits most tipping models. For example, Tspill is not centered at the same elevation at which the telescope points and, therefore, must be modelled differently than models in the literature for symmetric antenna.
• Suggest tippings should only be done over the elevation range of the expected observations.
• Suggest tippings should avoid elevation ranges where Tspill is problematic.
• Multi-layered tipping models in the literature don't extend down to cm-wavelengths
• Tippings might only be needed to determine Trcvr and then use the Tsys for each scan, and the tippings Trcvr to determine exp(tau*A).
• Suggest we need to make a list of typical observing 'cases', suggest data taking techniques and tippings models for each case, and provide estimates for the accuracy of the results.

Lots of ideas that we've not yet fully fleshed out. We'll return to Tipping models at a later meeting.

-- RonMaddalena - 14 Mar 2006