Search Results for ""

To enable observer planning for RFI avoidance, here is an archive of the most recent RFI scans to give the observer an idea of spectral occupancy as seen by the GBT receivers. Four zoom levels are provided starting at 100 Jy of the average RFI seen during the scan taken by the GBT.

Prime Focus 1 (342 MHz) – 02/18/2020

Prime Focus 1 (800 MHz) – 05/27/2021

L-Band – 07/04/2021

S-Band – 01/24/2021

C-Band – 07/09/2021

X-Band – 01/17/2021

Ku-Band – 11/18/2020

K-Band Focal Plane Array – 01/19/2021

Ka-Band – 09/20/2020

Q-Band – 03/03/2020

RFI scans are performed routinely by the operators during gaps between astronomical observations. The aim of the technique is to do the best job of monitoring narrow-band RFI coming from the horizon (which comes at the cost of monitoring changes in the RFI from satellites, nearby planes, etc.). The GBT, which can’t point below an elevation of 5 deg (typically many beamwidths), has very little sensitivity to horizon-based RFI in its forward direction. The sidelobes in the forward direction are also not uniform. The telescope is much more sensitive to radiation that comes from the horizon and that enters the sidelobes of the feeds. To make the sensitivity of the feed patterns uniform around the horizon we position the elevation of the antenna so that the flange of the feed is parallel to the horizon. The feed sidelobes also have uniform sensitivity as they cover a very large solid angle. Gregorian receivers require a different elevation than PF receivers to put the feeds into this orientation. However, the single telescope feed arm will introduce azimuthal diffraction patterns on top of the feed sidelobe patterns. To smooth out this azimuthal dependence, the telescope moves at near its top speed from Az=0 to 180 (or Az=180 to 0 if that’s a more efficient route). With this tactic, one can’t expect the monitoring of RFI that comes from the forward direction (satellites, etc.) to be anything more than hit and miss.

The data reduction uses the average of the raw bandpass data across the full slew and the average of the two polarizations (if the receiver has dual polarization). The raw bandpasses are put through a high-pass filter (with an upper frequency of 0.1 channels-1), which removes the overall bandpass shape. The use of a high-pass filter does the best job of depicting narrow-band RFI (our primary aim), but which comes at the cost of compromising the detection of wide-band RFI if it is significantly wider than about 10 channels.

Since the noise diode flickers throughout the observing, the bandpass power is converted into units of antenna temperatures using the ratio of the detected total power to the change in power when the diode is on. Since observers want to know the level at which the RFI would contaminate their observations, the signal strength observers would see is mimicked by converting Ta to Jy using the antenna’s main-beam gain.

The most recent GBT (raw, unfiltered) RFI data is stored as .fits files at:/home/gbtdata/TRFI_MMDDYY_RN where MM is month, DD is the day, YY is the last two digits of the year, R is the receiver letter designation (see below) and N is the session number for that particular day (1,2, 3 etc.) GBTIDL may be used to examine these plots dynamically and in more detail.

Further plots and files can be found at https://science.nrao.edu/facilities/gbt/interference-protection/ipg/rfi-scans . Please direct any questions to bgregory@nrao.edu

Like any other ground-based millimeter continuum observations, our default data processing attempts to remove the Earth’s atmosphere with either a common-mode subtraction or something quite similar to it (e.g. PCA).

An early memo on filtering compared the results using two different (Lissajous daisy) scan sizes and two different sets of filtering parameters, with an executive summary here and a more in-depth tabulation here.

• 

• 

The above plots show transmission functions for a broad range of (Lissajous) scan sizes (from 2.5′ to 5.0′) with a fairly gentle filtering (3 components subtracted from the timestreams via PCA and a windowed filter, keeping frequencies between 0.06 Hz and 41.0 Hz). Below are the same scan sizes filtered with 5 components (via PCA) and a window between 0.08 Hz and 41.0 Hz.

• 

• 

Splitting the scan sizes among two plots below that show the differences between these two reductions:

• 

• 

A filtering with a highpass of 0.06 Hz is, unfortunately, a bit more gentler than we find is necessary. Rather, a highpass at 0.07 or 0.08 Hz often results in acceptable noise in our maps. There are still other datasets which require still more aggressive filtering, either a highpass at 0.09 Hz, or even 0.1 Hz.

A repository of transfer functions (ascii files) is available here.

The MUSTANG-2 team is able to produce the following data products for all MUSTANG-2 projects:

• A calibrated map either in Jy/beam or Kelvin (main beam).
  • Each scan is gridded individually. Maps are stacked via weighted averaging.
• An associated noise map
  • The default noise map, for a single scan, flips every other detector. The combined noise map is the stack of all individual scans.
• An associated SNR map
  • The above map and noise map are smoothed by some amount (generally 9″). The weight map is scaled according to the RMS in the noise map

Additionally, the following products will help in using the above data products:

• A transfer function
  • This accounts for the filtering we perform on the data. The above calibrated map does not preserve signals on all scales; a transfer function (along with convolution of the beam shape) must be taken into account when modelling the intrinsic astronomical signal.
  • More information on how this is calculated and can be used can be found here: https://safe.nrao.edu/wiki/bin/view/GB/Pennarray/MUSTANG_CLASH
• A stacked beam map
  • From the calibrators, observed every ~30 minutes, we can compute an average beam for a given science target. The stacked beam is normalized to have a peak of 1. Double Gaussian (azimuthally symmetric) fits are included in the header (among the last cards in the header). The user can also produce their own fits to the stacked beam, if they so choose.

THE SCIENCE CENTER EXHIBITS AND INDOOR ACTIVITIES REMAIN CLOSED AT THIS TIME.

Tickets are NOW available for NEW outdoor programs, reservations are required to keep group sizes safe and small. See dates, times, and tickets available for Guided Walking Tours, Seated Presentations, and FREE Docent Activities. Need help buying your ticket online? Watch a video here.

Take a self guided walking tour of the site. Download a map here.

Learn more about biking onsite to explore our instruments and trails here.

Read the full announcement

2021 Closed Dates

November 23 – November 26 for Thanksgiving Holiday
December 21 – December 25 for Christmas Holiday
December 28 – December 31 for New Year’s Holiday