NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Dynamic > DynamicSchedulingGlossary > WeatherQueue

Changes | Index | Search | Go

Weather Queue and Sensitivity Calculations for the GBT - Under Development

• Introduction
• Details - The Weather Queue Choices
  • Opacity and Airmass
  • Wind
  • Time of Day
• PSC Decisions and DSS "long-term" Schedules - How much time to each queue?
• Current Documentation

Introduction

Determining the suitability of a weather queue for a given project with the GBT is a difficult task, as there are many factors in play. Opacity, cloud cover, outside temperature, wind, and time of day all are factors in deciding the suitability of running a given project. Also, though, one has to consider the desired sceintific outcome of a project. That is, if the project is a spectral line detection experiment of a narrow line, the investigators will likely have very different weather criteria then someone wishing to obtain continuum data of a small (point-like) object.

Additionally, any weather queue that is built for the GBT must be flexible enough to allow for seamless changes in the GBT weather criteria as the PTCS project progresses. E.g. if the PTCS project improves the pointing accuracy in wind, or increases the viability of observing high frequencies during the daytime, that needs to be seamlessly added into the queue information, without the investigators needed to know the fine details of the GBT PTCS models.

For the first stage of dynamic scheduling, we propose to simply create weather queues which the investigator can choose for her project. Accompanying the weather queues choices needs to be full documentation explaining how the investigator can convert the choices handed her to astronically interesting quantities (e.g. wind speed -> efficiency degradation due to pointing errors, Tsys -> rms). We also need to include enough information so that the investigators and the PSC understand how much time is available on the GBT for the various weather queues. This information should be used both to allow observers to make sensible choices about their chosen queues (e.g. not ask for 500 hours in a queue that may only have 300 hours available), and also allow the PSC to make sensible choices in assigning ranking to the proposals.

Details - The Weather Queue Choices

Opacity and Airmass

Opacities and airmass combine to increase the attenuation a signal will experience at a given elevation. Here, the signal is attenuated by the factor ^{airmass*opacity}. Opacity and and airmass, along with spillover and cosmic microwave background contributions, (what we'll call here Tsys') can be determined from the weather information available through Ron Maddelena's weather statistics. These contributions, added to the receiver temperature, provide the Tsys for the observations. As it is a readily calcuable value, Tsys' is an appropriate choice to use for one of the weather queue choices.

In an effort to keep all choices for the observer as frequency independent as possible, we propose to make four queues here, based off the amount of time a given Tsys' is available. That is, the four choices in this category will be: 25%, 50%, 75%, and 'any' Tsys' quartile, as shown in the middle panel of the following chart:

Here we will need full documentation explaining how the Tsys' value can be translated to sensitivity for spectral line, pulsar, and continuum (bolometer) measurements. This explanation will also have to include information on the recevier temperatures, to allow investigators to fully determine Tsys.

Wind

Wind speeds directly affect pointing, which in turn affects the telescope efficiency. Depending on their project, an observer my find wind speeds to be extremely important (e.g. a high frequency spectral line observation of a point source) or only minimally important (e.g. an detection experiment of an extended source). As such, investigators will also need to choose their wind speed category.

To again keep this as frequency independent as possible, we will use the PTCS criteria for wind speed, giving the observers three categories to choose from: good, usuable, or poor. These will translate directly to the PTCS tables for wind speed, which are replicated below. Note that these tables need to be updated from the PTCS ones to include Ka-band and to change the frquency of interest for Q band. Also, the numbers for the "poor" conditions need to be added.: MJM: These tables imply that the defintion of "Good" wind for X-band is different than that of "Good" for Q-band" (naturally). So for example, if the PI has specified good conditions and X-band (forget about other criteria for simplicity), then the DSS must consider receiver (or frequency , if one prefers) during the session selection/priority phase.

Table 1: Requirements, limits and observing strategies for "good" performance - (5% rms flux errors).

Receiver	Frequency	Wavelength	beam FWHM	rms tracking error	axial focus error	wind limit (m/s)	wind limit (mph)
L-band	1.4GHz	21cm	530"	75"	53mm	Any	Any
S-Band	2.0GHz	15cm	370"	50"	38mm	Any	Any
C-Band	5.0GHz	6cm	150"	20"	15mm	11 m/s	25 mph
X-Band	10.0GHz	3cm	75"	10"	7mm	8.0 m/s	18 mph
Ku-Band	15.0GHz	2cm	50"	7"	5mm	6.3 m/s	14 mph
K-Band - lower	20.0GHz	1.5cm	37"	5"	3.5mm	5.1 m/s	11 mph
K-Band - higher	25.0GHz	1.2cm	30"	4"	3.0mm	4.2 m/s	9 mph
Q-Band	52.0GHz	5.8mm	14"	2.0"	1.5mm	Note 1	Note 1

Table 2: Requirements, limits and observing strategies for "usable" performance (10% rms flux errors).

Receiver	Frequency	Wavelength	beam FWHM	rms tracking error	axial focus error	wind limit (m/s)	wind limit (mph)
L-band	1.4GHz	21cm	530"	105"	105mm	Any	Any
S-Band	2.0GHz	15cm	370"	75"	75mm	Any	Any
C-Band	5.0GHz	6cm	150"	30"	30mm	14 m/s	32 mph
X-Band	10.0GHz	3cm	75"	15"	15mm	10 m/s	23 mph
Ku-Band	15.0GHz	2cm	50"	10"	10mm	8.0 m/s	18 mph
K-Band - lower	20.0GHz	1.5cm	37"	7"	7mm	6.3 m/s	14 mph
K-Band - higher	25.0GHz	1.2cm	30"	6"	6mm	5.8 m/s	13 mph
Q-Band	52.0GHz	5.8mm	14"	2.8"	3mm	3.0 m/s	7 mph

Table 3: Requirements, limits and observing strategies for "poor" performance (20% rms flux errors).

Receiver	Frequency	Wavelength	beam FWHM	rms tracking error	axial focus error	wind limit (m/s)	wind limit (mph)
L-band	1.4GHz	21cm	530"	105"	105mm	Any	Any
S-Band	2.0GHz	15cm	370"	75"	75mm	Any	Any
C-Band	5.0GHz	6cm	150"	30"	30mm	?	?
X-Band	10.0GHz	3cm	75"	15"	15mm	?	?
Ku-Band	15.0GHz	2cm	50"	10"	10mm	?	?
K-Band - lower	20.0GHz	1.5cm	37"	7"	7mm	?	?
K-Band - higher	25.0GHz	1.2cm	30"	6"	6mm	?	?
Q-Band	52.0GHz	6mm	14"	2.8"	3mm	?	?

Key to columns:

Notes

Time of Day

The last choice the investigators will have is the time of day for the observations, with the two choices being "night" and "day and night". Here, night is defined as sunset+3 hours to sunrise+2 hours. We will also have a definition called "work day" for maintenance like activities.

Differential heating and cooling of the telescope alters the surface of the telescope, resulting in degradation of telescope efficiencies, and 'bends' the telescope, resulting in pointing changes. At high frequencies, these affects are important. The current recommendations are that, for best work, observing above 40 GHz should only be done at night, from 3 hours after sunset to 2 hours after sunrise. Low frequency observers may want to consider night time observing for two reasons. RFI is usually lower at night; and, in some cases, the sun has a slight negative impact on baseline shapes.

PSC Decisions and DSS "long-term" Schedules - How much time to each queue?

The GBT Proposal Scheduling Comittee (PSC) needs to have a rough idea of how much time is available in each weather queue in order to accurately asses proposal ratings without leaving the GBT with far too high a time request for a given queue. Additionally, the DSS needs to make telescope schedules (for the 1 and 2 week "look aheads"). Rather than having the PSC understand all 18 queues, and force the DSS to make 9 schedules, we will break the queues down into something simpler, along the lines of "excellent weather", "good weather" and "poor weather", with the exact definitions to be decided. MJM: Per Karen: we are temporarily tabling how many queues the DSS requires for look ahead and observer notification purposes. For session selection purposes, the DSS must evaluate all weather criteria. As it is currently understood, the best wind conditions are the least likely to occur. Therefor when promoting sessions during the session selection process, wind is the first criteria for session promotion. Promotion only occurs when there is are no sessions requiring the current weather and wind conditions.

Current Documentation

The best documentaion we currently have on the choices faced by observers are available here:

• Ron's information on receiver performance (from Dec 2005) - http://www.gb.nrao.edu/~rmaddale/GBT/ReceiverPerformance/PlaningObservations.htm
• The PTCS astronomy pages - http://wiki.gb.nrao.edu/bin/view/PTCS/PointingFocus
• An explanation of the files used by CLEO's forecast utility, and the effort to load these into a database: NOAAWeatherForecastData
• Tsys' predictions: TsysPredictions

-- KarenONeil - 25 May 2006