The Resident Shared Risk Observing Program

The WIDAR correlator and the EVLA will provide a vastly more powerful instrument than the VLA. The RSRO program offers participants early access to the growing capabilities of the EVLA as it is being commissioned, in exchange for a period of residence in Socorro to assist with the commissioning. It is intended to accelerate the development of the EVLA's full scientific capabilities by gaining enhanced resources and expertise through community participation. It will at the same time help quickly optimize the scientific productivity of the EVLA. The RSRO program will run for approximately two years (from March 2010 through the end of 2011), with up to 25% of the EVLA time available for astronomical observations allocated to the RSRO program, depending on demand and quality of science proposed. At the end of this period all access to the EVLA will be through the OSRO program, until full operations begin in 2013. The expected EVLA and WIDAR capabilities available to the RSRO program as a function of time are described further below, along with details of the program.

Expected EVLA capabilities for RSRO, 2010-2011

Antennas

VLA antennas are currently being retrofitted to EVLA specifications at a rate of approximately one antenna every two months. At the end of 2008 there were 18 EVLA antennas regularly used for astronomical observing, and it is planned that by the end of 2009 this number will have increased to 24. Despite the large instantaneous bandwidths and enhanced tuning ranges of the new EVLA RF and IF system and receivers relative to the VLA, as long as the VLA correlator continues in use the maximum instantaneous bandwidth that can be correlated is 50 MHz per IF. For this and several other technical reasons the VLA correlator will be turned off in January 2010, and all remaining VLA antennas will be decommissioned until they are retrofitted to become EVLA antennas. VLA antennas will not be used with WIDAR.

Receivers

When EVLA antennas are retrofitted they are currently outfitted with "interim" L, EVLA or "interim" C, VLA X, EVLA K, and EVLA Q-band receivers. [Interim receivers are EVLA receivers with VLA orthomode transducers. All interim receivers will be converted to full EVLA capabilities by the end of the project. The polarization purity and sensitivity of the interim receivers typically is good only over the traditional VLA tuning range.] As of August 2009 thirteen of the EVLA antennas also have EVLA Ka-band receivers. Figure 1 shows the expected rate of antenna retrofits and installation of the final EVLA receiver systems throughout the EVLA construction project. Figure 1 does not show the whole picture, however. Those antennas that do not have final EVLA receiver systems typically do have VLA or interim receivers, and Figures 2 and 3 show the total numbers of receivers at each band as a function of time. The maximum bandwidth that can be brought back from the antennas to the correlator depends on the rate at which the antennas are outfitted with fast 3-bit samplers (the "4 GHz BW" or "8 GHz BW" lines in Figures 2 and 3).

Figure 1: Availability of EVLA antennas and final EVLA receiver systems as a function of time.

Figure 2: Low-frequency receiver availability. The red line shows the number of VLA receivers, the light green line shows the interim receivers, and the blue line shows the EVLA receivers.

Figure 3: High-frequency receiver availability. The red line shows the number of VLA receivers, typically with more limited tuning capability, and the blue line shows the final EVLA receivers.

WIDAR correlator

Boards for a 4-station test version of the WIDAR correlator were tested in Socorro prior to the WIDAR critical correlator production review at the beginning of December, 2008. Delivery of production WIDAR boards will continue through the end of 2009, and boards will be installed and commissioned into 2010. The operational capabilities of the EVLA correlator will depend on the order in which the station boards (approximately equivalent to the number of antennas that can be correlated) and the baseline boards (approximately equivalent to the total bandwidth that can be correlated) are installed and commissioned in the correlator, and (ii) the delivery of correlator control software (simple modes will be available first, more complex ones later).

The first step in commissioning WIDAR will be to implement the functionality to be offered for the OSRO program. As additional baseline boards are added, the total bandwidth correlated by WIDAR will continue to grow, and all boards are expected to be in place by early 2010. Therefore, close to the full bandwidth will be available to participants of the RSRO program, subject to the outfitting of antennas with 3-bit samplers (Figure 1).

The precise capabilities available to the RSRO program are less well-defined than those for OSRO, in part because observing and correlator modes will be commissioned by RSR observers according to the needs of the science proposed. Nevertheless, we outline a possible growth path for RSRO capabilities below, to inform potential RSRO science proposals (see Table 1 for an overview). The following timeline is purely notional, since it assumes the commissioning resources provided by the RSRO program. If these resources are unavailable the correlator commissioning will likely follow the path outlined, although at a slower rate. Any questions about correlator capabilities should be directed to Michael Rupen.

Table 1: Overview of correlator capabilities for RSRO; further details are described below

Trimester 1, 2010 (D-configuration):

The maximum bandwidth available will be 2 GHz, using 8-bit samplers, implemented as 2x1 GHz basebands that can be tuned independently within a given receiver band. The number of antennas available with this bandwidth depends on the implementation of the wideband digital IF system, and is shown as the line marked "2 GHz BW" in Figures 1-3. This will be available with the receiver combinations shown in Figures 2 and 3. The correlator will provide 16 independently-tunable sub-band pairs. For this initial mode, each sub-band pair must be identical in bandwidth, number of channels, and number of polarization products, and can be configured as shown in Table 2, with the additional flexibility that the number of polarization products can be traded for an increase in the number of channels (e.g., correlating just the parallel hands doubles the number of channels per polarization product, and halves the spectral resolution).

Table 2: Initial correlator capabilities per sub-band for RSRO; the number of polarization products can be traded for an increase in the number of channels

Trimester 2, 2010 (C-configuration):

The maximum bandwidth available will be 2 GHz, implemented as 2x1 GHz basebands that can be tuned independently within a given receiver band. The correlator will have 64 independently-tunable sub-band pairs, which may be set up as shown in Table 2, with identical bandwidths, number of channels, and number of polarization products, to cover the maximum available bandwidth. It will also be possible to trade sub-bands for channels, so that the correlator resources for one sub-band might be used to double the number of channels in another sub-band. For example, one could use 64 sub-bands with 64 channels each, full polarization, or 16 sub-bands with 256 channels, full polarization.

Trimester 3, 2010 (B-configuration):

The maximum bandwidth available will be 2 GHz, implemented as 2x1 GHz basebands that can be tuned independently within a given receiver band. Recirculation (trading sub-band bandwidth for an increase in the number of channels per sub-band) will be enabled in the correlator, to give the capabilities per sub-band pair shown in Table 3 below. There will be 64 independently-tunable sub-band pairs, which can have identical bandwidths, number of channels, and number of polarization products, although it will also be possible to trade sub-bands for channels. The number of polarization products can be traded for an increase in the number of channels, up to a maximum of 16,384 channels per polarization product per sub-band pair.

Table 3: Correlator capabilities per sub-band with recirculation; the number of polarization products may be traded for number of channels

Trimester 1, 2011 (A-configuration):

After implementing recirculation the next step will be to introduce more flexibility by allowing each of the independently-tunable 64 sub-band pairs to have different bandwidths and numbers of channels. The number of polarization products will be the same for all sub-bands. The available bandwidths and numbers of channels are as shown in Table 3.

Trimester 2, 2011 (D-configuration):

At this time the array will be outfitted with fast 3-bit samplers to provide 8 GHz instantaneous bandwidth for all receivers capable of delivering this bandwidth.

Trimester 3, 2011 and following:

Up until this point the WIDAR capabilities expected to be available will be for more or less "standard" types of observations. Special modes will then be implemented and tested. The order in which these special modes are developed will, to a large extent, be driven by the proposed RSRO science and the presence of an expert user to help with the commissioning, subject to practicalities of hardware and software development. Indeed, some of these may be commissioned sooner if an expert user is available to help through the RSRO program, and the necessary software development can be accelerated. The special modes are given below (no priority is intended by the order of this list).

• (Even) more flexibility in the allocation of correlator resources
• Multiple subarrays
• Phased array and single-dish VLBI
• Solar mode
• Planetary mode
• Burst mode
• Radar mode
• Pulsar binning
• Pulsar gating
• On-the-fly mosiacing
• 7-bit correlation
• Fast switching between correlator set-ups
• Very fast dumps
• Mixing of 3-bit and 8-bit samplers
• More processing by the correlator back-end
• Enable more channels for single-polarization observations
• Online RFI flagging
• Multiple phase/delay tracking centers
• Output to user instruments

Although some very straightforward datasets may be reduced in AIPS it is expected that in general RSRO data will be reduced using CASA, and that testing the post-processing software will be part of the RSRO commissioning effort.

Participation in the RSRO program

The primary requirement of the RSRO program is that there be at least one expert from each participating group in residence in Socorro. These experts must be able to contribute effectively to commissioning while incurring as little overhead from EVLA staff as possible. Some support for accommodation may be available, as described below.

There will be two possible routes to obtaining observing time through the RSRO program:

1. The successful submission and acceptance of a proposal received in response to a Call for Resident Shared Risk Observing Proposals for the October 1, 2009 deadline, and subsequent proposal deadlines through the end of the RSRO program in 2011. Such proposals will be allowed an extra two pages compared with OSRO proposals and will have three parts:

   1. A scientific justification, to be peer reviewed as part of NRAO's current time allocation process, submitted through the Proposal Submission Tool. The RSRO resources will not be part of the Proposal Submission Tool, however. Instead, a separate web-based form will allow proposers to specify their expected observing set-up, data rate, data volume, and post-processing requirements.
   2. A technical section naming the personnel who will be involved in the residency and describing how their expertise will be used to address the critical priorities of EVLA commissioning. The proposed dates of the residency must be included, so that the residency can be matched to EVLA commissioning planning. This section will be reviewed by NRAO staff.
   3. A budget specifying the level and nature of any support requested from NRAO. It is expected that NRAO will be able to provide accommodation in the NRAO Guest House, subject to availability. Other support may be available separately through the NRAO Visitor's Program. Proposals that do not require Observatory support will have a substantial advantage over those that request NRAO resources.

   The acceptance of a RSRO proposal will depend on the merits of all three sections of the proposal, and will be judged in terms of the benefits to the EVLA commissioning process by NRAO staff.

   Up to 25% of the hours available for astronomical observations will be reserved for this program; the actual fraction of time assigned will depend upon demand, the quality of the proposed science as judged via the peer-review process, and upon the quality of the technical support offered to the EVLA project, as determined by NRAO. Throughout the transition from the VLA to the EVLA we have maintained astronomical observing at a rate of approximately 60% of all possible available hours. Assuming this efficiency is continued during WIDAR commissioning, we expect approximately 1200 hours per year to be devoted to the RSRO program.

   One month of resident commissioning effort will be required for every 20 hours of EVLA time awarded to a RSRO project, with a minimum residency of 3 months; this is considered the minimum time useful for becoming familiar with the EVLA commissioning effort and contributing effectively, although in exceptional circumstances NRAO is prepared to consider proposals for dividing this into two or three shorter visits. The period(s) of residency may occur in advance of the observing time awarded in order to decouple essential scientific requirements (such as array configuration) from other factors which may affect when personnel are available (such as teaching schedules). However, observers should be present for one week prior to their observations in order to become familiar with the latest commissioning developments and to set up their observations. In the special case of Target of Opportunity proposals an EVLA staff collaborator will be required for setting up observations on short timescales.

   It should be noted that having a member of the NRAO-EVLA commissioning staff as a collaborator on a RSRO proposal will not satisfy the residency requirement. Furthermore, graduate students will in general not satisfy the residency requirement, although there may be exceptional cases. Graduate students will be allowed to accompany their advisor as long as the advisor takes primary responsibility for managing the student's efforts. Resident personnel will work under NRAO management in order to optimize the overall commissioning effort. A set of deliverables will be agreed upon in advance of the start of the residency.

   The types of proposals considered under the RSRO program may include both large (>200 hours) and small (~10-200 hours) projects. Qualified large projects proposed by consortia will be considered as long as the residency requirements are met. A single individual may satisfy the residency requirement for several small projects.

   Resident observers will also be permitted to take part in the "exploratory proposal" section of the NRAO EVLA Commissioning Staff Observing program, in addition to the time allocated by the process described above.

2. A participant may arrange to visit Socorro to contribute to EVLA commissioning by submitting a proposal directly to the Assistant Director for NM Operations (nraonmad@nrao.edu) containing items (b) and (c) of the RSRO Proposal requirements described above, and obtain observing time either by submitting part (a) at the October 1, 2009 (or subsequent) proposal deadline, or through the "exploratory proposal" section of the NRAO EVLA Commissioning Staff Observing program while in residence. Such visits should conform to the residency requirements above, namely, that they provide a minimum of 3 months resident commissioning effort total, but which can be divided into two or three shorter visits if agreed in advance. Proposals to visit Socorro under this program may be submitted at any time, and residence may begin as early as summer 2009.

We emphasize the "shared risk" nature of the RSRO program. Since observers will be attempting to use the full capabilities of a system under development and in the process of being commissioned, NRAO can make no guarantee of the success of any observations made under this program, and no additional commitment is made beyond granting the hours actually assigned by the peer review process.

Commissioning needs of the EVLA project

The rationale for the RSRO program is driven in part by the need for additional personnel to help accelerate the commissioning of the EVLA, WIDAR, and related software systems. Users who visit Socorro as part of the RSRO program will be required to work on aspects of EVLA commissioning that would most benefit from the increased manpower, given the expertise available in the community. A list of potential commissioning tasks that could be addressed through RSRO participation is given below. As emphasized earlier, the dates associated with these commissioning tasks assume the resources provided by the RSRO program. If the resources are not forthcoming, the commissioning will proceed at a slower pace than that outlined here.

Table 1: Overview of EVLA commissioning, 2010-2011

One of the goals of the RSRO program is also to develop EVLA expertise in the user community, so all residents will in general be astronomers with a personal stake in both their observations and the results of their commissioning efforts. Some exceptions may be considered at the discretion of the Assistant Director for New Mexico Operations.

Questions regarding the RSRO program should be directed to Claire Chandler (cchandle@nrao.edu).

Copyright © 2009 Associated Universities, Inc.
The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.