SMAP DATA RELEASE DR2 Contact: smap@astro.caltech.edu A. Conley, G. Marsden, M. Zemcov 2013-09-30 ####################### This is the README file for the SPIRE map files associated with the second Hershel-SPIRE/SAG-1/HerMES public data release (DR2). The HerMES program is presented in [1]. HerMES maps are created by the SMAP pipeline. The SMAP map-maker, SHIM (SPIRE-HerMES Iterative Mapper), iteratively removes a low-order polynomial baseline from each scan. At each iteration i a polynomial is fit to the time-stream residual R_i = S - M_{i-1}, where S is the time stream and M_{i-1} is the predicted time stream given the map calculated on the previous iteration. Additionally, each scan is given a weight based on the inverse variance of the time-stream residual. The order or the polynomial baseline varies from 0 to 9, depending on the scan length, with longer scans requiring higher order polynomials. The polynomial order is chosen using an automated algorithm which is only a function of the scan length. These maps are made with 20 iterations, which appears to provide sufficient convergence. The mapping algorithm is described in [3] and [4]. If you have any questions, comments or concerns please contact the SMAP team at smap@astro.caltech.edu ####################### FILE NAMES The file names indicate the level, field, image type, band, and version number (here DR2). ___SMAP_.fits For instance, L5_Bootes_image_SMAP250_dr2.fits is the DR2 image map at 250 µm of the Boötes (level 5) field. ####################### FILE CONTENTS Each fits file in this data release contains 4 extensions: 1 - signal map [Jy / beam] 2 - error map, based on propagation of time-stream weights [Jy / beam] 3 - exposure map 4 - mask map with the following values: 0: no mask (ie good data) 1: no data 2: no cross-linking 4: large noise region (Filter maps only). 5 - filter (Filter maps only) 6 - effective point spread function (Filter maps only) ####################### Definition of the FIELDS ------------------------- The fields released here (named as in Oliver et al. 2012, Table 1) are: Abell1689 Abell2218 MS0451.6-0305 MS1054.4-0321 RXJ0152.7-1357 Abell1835 Abell2219 Abell2390 Abell370 Cl0024+16 MS1358+62 RXJ13475-1145 GOODS-N COSMOS (combined) ECDFS (nested) CDFS-SWIRE (combined) Bootes-HerMES (combined) Lockman-North, Lockman-East-ROSAT (nested) Lockman-SWIRE (combined) Groth-Strip (nested) EGS-HerMES UDS, VVDS (nested) XMM-LSS-SWIRE ADFS ELAIS-N1-SWIRE ELAIS-S1-SWIRE FLS Nested fields ECDFS, Lockman North and Lockman Est-ROSAT, Groth-Strip, UDS and VVDS: ------------------------------------------------------------------------------------ In these fields the shallow map(s) overlaps the deep one(s): nested maps combining shallow and deep data are created (see Oliver et al. 2012 for details). For each of these fields, a DS9 region file corresponding to the footprints of the deep map(s) is provided (with a plot of the footprints in the HerMES_DR2_nested_fields.pdf file). The corresponding catalogues are listed below. The AOR set numbers are defined in Oliver et al. (2012) and are listed on the front page of the DR2 data release. - ECDFS: The nested map NE-CDFS-SWIRE_nested-image_SMAPXXX_DR2 is the combination of GOODS-S (AOR 13), ECDFS (AOR 15) and CDFS-SWIRE (AOR 27 and 33). - Lockman-North, Lockman-East-ROSAT: The map NE-Lockman-SWIRE_nested-image_SMAPXXX_DR2 is the combination of Lockman-SWIRE (AOR 28B), Lockman-North (AOR 19) and Lockman-East-ROSAT (AOR 18 & 18B). The map NE-Lockman-SWIRE_nested-image-full_SMAPXXX_DR2 combines the same fields as NE-Lockman-SWIRE_nested-image_SMAPXXX_DR2 with the addition of the AOR 28. The source extraction is performed on NE-Lockman-SWIRE_nested-image_SMAPXXX_DR2 (AOR 28B, 18, 18B & 19) - Groth-Strip: NE-EGS-HerMES_nested-image_SMAPXXX_DR2 is the combination of EGS-HerMES (AOR 29) and Groth-Strip (AOR 17) - UDS, VVDS: NE-XMM-LSS-SWIRE_nested-image_SMAPXXX_DR2 is the combination of XMM-LSS-SWIRE (AOR 36), UDS (AOR 23) and VVDS (AOR 24). Combined fields COSMOS, Bootes-HerMES, Lockman-SWIRE and CDFS-SWIRE: -------------------------------------------------------------------- These fields were observed several times with a similar (but not strictly identical) coverage, combined maps are produced from these multiple observations, the corresponding AOR sets are listed below: -COSMOS: AOR sets 22 & 22B -Bootes-HerMES: AOR sets 30 & 37 -Lockman-SWIRE: AOR sets 28 & 28B -CDFS-SWIRE: AOR sets 27 & 33 ####################### COMPLEMENTARY MAPS *** (not available for nested and combined fields)***: ------------------------------------------------------------------------ ang1: sub-map made from all scans in one orientation on sky (eg horizontal) ang2: sub-map made from all scans in other orientation on sky (eg vertical) bolo1: sub-map half of detectors in focal plane bolo2: sub-map other half of detectors in focal plane half1: sub-map made from first half (in time) of data half2: sub-map made from second half (in time) of data filter: full-data set, point source filtered (Chapin et al. 2011) this type of map is provided only for the deep fields pixel size for these maps is half the standard size (3, 4.167, 6") - please read [2] for a complete description. In some fields, there is more than one set of rotation angles. Here, we provide, e.g., ang1a/ang2a and ang1b/ang2b jack-knife maps where possible. The jack-knife sub-maps are not iterated, but simply use the offset and weights solution calculated for the full "image" map, as well as the cosmic ray information. Note: for FLS (and other fields observed with only 1 repeat), the half1/2 maps are not included, since they would be identical to the ang1/2 maps. ####################### Additional Notes ----------------- Changes to the SMAP algorithm as described in [3] and [4] compared with the DR1 release: * The algorithm for the temperature-correlation removal has been modified slightly to use more robust fitting. * One must use a full implementation of the WCS specification to use the astrometry for these (and all previous) SMAP maps. In particular, for far northern or southern fields, LATPOLE and LONPOLE do not have their standard values to reduce field distortion. Currently, this only affects the GOODS-N and MS1358 fields. ####################### References [1] Oliver, S.~J., Bock, J., et al. 2012, MNRAS, 424, 1614 [2] Chapin, E.L., et al. 2011, MNRAS, 411, 505 [3] Levenson, L., Marsden, G., Zemcov, M., et al. 2010, MNRAS, 409, 83 [4] Viero, M., et al. 2013, ApJ, 772, 77