Minutes of 21 April 2003 Telecon of USSKA Consortium Written 22 April by JMC In attendance: T. Beasley, D. Bock, J. Cordes, G. Cortes, D. DeBoer, R. Fisher, A. Goodman, L. Greenhill, D. Jones, N. Kassim, K. Kellerman, P. Kronberg, J. Lazio, C. Lonsdale, B. Preston, Y. Terzian, D. Wilner, J. Welch Main topic: our ``response'' document to EMT & ISAC questions and assessments. Lonsdale: will flesh out answer to question 1 from the EMT about surface brightness sensitivity on small spatial scales. JMC: read Dickey's email comments re putting in one of Colin's figures (new version of Figure 3.4 of the 2002 whitepaper) showing controllability of large-scale sensitivity. Also, Dickey commented that we should emphasize the flexibility of the LNSD concept in allowing additional antennas (e.g. low-frequency dipoles) to use the same overall infrastructure. JMD's comments are: " ... but it might be worth adding a short paragraph of conclusion, either of this document or of section 3.3 alone, that makes the point that the LNSD array provides all the infrastructure, data transmission, analog and digital signal processing, correlator and signal path hardware needed for a super-LOFAR that would be capable of detecting and mapping the EoR signals. Since the antenna and amplifier costs are very low at these frequencies, it is this infrastructure that is critical for an SKA operating below 150 MHz. The LNSD concept is thus naturally suited to extension to low frequencies by adding a separate antenna system. My personal guess is that LOFAR may detect the EoR signal in some form, but that a larger and more powerful array will be needed to map it spatially and in frequency so as to resolve the blobs that correspond to proto-galaxies. This may or may not be kept as a level-0 science goal for the SKA, if it is it will be obvious that all design concepts will need add-ons of some kind to accomplish it." TRANSIENTS: Welch: raised the question about slow vs. fast transients, that slow transients can be handled through standard correlation-based imaging. JMC: commented that fast transients and pulsars can be dealt with through fast dumps of a correlator (e.g. 10^3 channels, 100 microsec dump rates) but most likely only for an inner core array, whose size is a function of how many digital operations can be done. To process an inner array of 1 km (diameter) requires 10^4 separate beams to be formed from the correlations. Lazio: brought up giant pulses from other galaxies as an example. JMC: need wide-field sampling in searches of nearby galaxies (e.g. M31 and M33) but alternatively could target individual supernova remnants in those galaxies. Overall assessment: should flesh out the transients discussion in the document re how to approach different kinds. Discussion about hybrids and how much to emphasize: Welch: antennas can work at 100 MHz, which would obviate the need for ancillary dipoles. Lonsdale: ability to hybridize the LNSD concept should be pitched as a positive aspect. Kellerman: we will be faulted if we ask for too much! (i.e. adding dipoles to the LNSD concept as a means to achieve the goals will only increase cost) JMC: at this point we should simply advertise any and all positive aspects of the LNSD concept. The ISAC currently weights evenly the 18 Level-1 science areas in Table 1 (and the Compliance Matrix). Only when these are prioritized can we really weigh the costs of different science areas. Lazio: colocation of dipoles and dishes may introduce cross talk that would be anethema to EoR studies. Overall sentiment: emphasize flexibility of LNSD concept but don't push the idea of hybrids too hard at this point. SHADOWING: Jones: brought up the shadowing issue and suggested that we might want a plot of degree of shadowing vs. azimuth as well as elevation. Lonsdale: said people were too worried about shadowing. It can be dealt with through the necessarily very capable calibration software. Kronberg: polarization an issue at short spacings. Beasley: should mention the Galactic Center and that the configuration can be stretched if the central site is in the North. FREQUENCY RANGE: Welch: what frequency range should be selected for the room-temperature receiver and its corresponding prime-focus feed? Current proposal: 0.15 to 1.5 GHz But spillover will cause 30 K extra Tsys with the PF feed. Given that the 21 cm line is important to z = 2, we need the best system possible down to about 500 MHz. That suggests that the prime focus feed/Rx should cover (e.g.) 100 to 500 MHz. Others: questions about how large a subreflector can be used; what the relative efficiency will be at 100 MHz; about degradation of high-frequency performance from the PF feed. Welch: the ATA has a 2.4m secondary that can go to 500 MHz and the same can be true for the 12m dishes of the LNSD concept. Estimate efficiency down by 20 to 30% at 100 MHz from what it is at higher frequencies. The PF feed hides behind the secondary when not in use so in principle should not degrade high-frequency performance. overall: we should emphasize that we can go down to 100 MHz without any new antennas (just appropriate feeds). However there are lingering doubts about how good the system will be and whether it is the right approach for investigating the EoR. STRUCTURING THE DOCUMENT: Preston: 1. need short introduction and conclusion sections 2. suggest adding a new column to Table 1 that gives our own score of the LNSD concept. Beasley: should extract from the text the important points (flexibility and overall capability) of the LNSD concept and consolidate into an intro section.