TITLE: GCN GRB OBSERVATION REPORT NUMBER: 1556 SUBJECT: XRF 020903: Archival Optical Images DATE: 02/09/29 19:46:04 GMT FROM: Edo Berger at Caltech A. Gal-Yam (Tel Aviv University) reports: "We have extracted from the USNO plate archive (http://ftp.nofs.navy.mil/data/fchpix/) two blue plates, covering the location of the variable object possibly connected with XRF 020903, reported by Soderberg et al (GCN 1554). Inspecting these plates we find the following: 1) A plate obtained on UT 1977.6 reveals an object, coincident with the variable source reported by Soderberg et al. The B-band magnitude of this object appears to be about one magnitude fainter than that of the nearby galaxy noted as "G2" by these authors. 2) Comparison of the 1977 plate with the (much poorer) POSSI blue plate obtained on 1954.6 may suggest a variation of the flux ratio between the source coincident with the Soderberg et al. variable, and the galaxy G2, with G2 being brighter on the 1977 plate, but possibly fainter than the source reported by Soderberg et al., in 1954. This may hint that this source was variable in the past. We stress, however, that the 1954 data are quite poor, so past variability can not be ascertained at this stage. GIF Cutouts from both plates, showing the area around the Soderberg et al. variable, can be obtained from ftp://wise3.tau.ac.il/pub/avishay/xrf. These can be compared with the Palomar data shown by Soderberg at http://www.astro.caltech.edu/~ams/XrF.html. The source detected on the 1977 plate may be hard to interpret as a possible "host" galaxy for the Soderberg et al. transient, even if we disregard its possible past variability. Soderberg et al. report that this source is not detected on POSSII red plates, and estimate a magnitude of R=22 from their OT+SN+host model. This fact can be reconciled with the 1977 B-band data, showing this object only about 1 magnitude fainter than G2 (which has R=17 mag), if this object has a very blue B-R color. However, this appears to be in contradiction with the Magellan spectrum presented by Soderberg et al., which is flat, or possibly even red, in B-R. An alternative explanation fitting both the archival analysis and the entire data set presented by Soderberg et al. and Berger et al (GCN 1555) is that this source is a variable radio-loud AGN. We have recently shown (Gal-Yam et al. 2002, PASP, 114, 587) that such objects can appear to be very similar to optical transients associated with GRBs. Further inspection of archival data, as well as further observations (B-band especially), can distinguish between an OT+SN model, and an AGN, according to their variability properties. If an AGN, the source discovered by Soderberg et al. may still be related to XRF 020903. However, in that case, a SN component no longer seems to be required. Until this issue is clarified, we urge some caution in interpreting the Soderberg et al. data set as evidence for a connection between XRF 020903 and a SN. This work has made use of the USNOFS Image and Catalogue Archive operated by the United States Naval Observatory, Flagstaff Station (http://www.nofs.navy.mil/data/fchpix/)."