| Version 36 (modified by , 11 years ago) ( diff ) |
|---|
Schema column changes
FrameMeta
- Change numPhotoRef to nPhotoRef to match new convention.
ImageMeta
- Change numAstroRef to nAstroRef to match new convention.
- Change numPhotoRef to nPhotoRef to match new convention.
StackMeta
- Change nPhoRef to nPhotoRef to match new convention.
DiffMeta
- Add tess_id, projectionID columns to match other metadata tables.
- Change analVer to analysisVer to match other metadata tables.
DiffDetection
- Reorder ippObjID to be before ippDetectID to match other tables.
ForcedMeanLensing
- Drop rpsfE1, rpsfE2, ipsfE1, ipsfE2, zpsfE1, zpsfE2, ypsfE1, ypsfE2. These do not have g filter counterparts, and do not appear to be populated.
ForcedWarpMeta
- Add tessID, projectionID columns to match other metadata tables.
- Change analVer to analysisVer to match other metadata tables.
- Change nPhoRef to nPhotoRef to match new convention.
ForcedWarpMeasurement
- Add tessID, projectionID columns to match other metadata tables.
ForcedWarpExtended
- Add tessID, projectionID columns to match other metadata tables.
Irregularities
dvoRegionID
It appears that the REAL/-999 version is more correct?
dvoRegionID Detection dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID DiffDetObject dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID DiffDetection dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID ForcedWarpExtended dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID ForcedWarpLensing dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID ForcedWarpMeasurement dimensionless REAL -999 "Internal DVO region identifier." dvoRegionID ObjectThin dimensionless INT -1 "Internal DVO region identifier." dvoRegionID StackObjectThin dimensionless REAL -999 "Internal DVO region identifier."
ForcedWarpExtended
Should these be transitioned to Janskys?
flxR5 ForcedWarpExtended adu/seconds REAL -999 "Flux from forced photometry measurement within an aperture of radius r = 3.00 arcsec." flxR5Err ForcedWarpExtended adu/seconds REAL -999 "Error in flux from forced photometry measurement within an aperture of radius r = 3.00 arcsec." flxR5Fill ForcedWarpExtended dimensionless REAL -999 "Aperture fill factor for forced photometry measurement within an aperture of radius r = 3.00 arcsec." flxR5Std ForcedWarpExtended adu/seconds REAL -999 "Standard deviation of flux from forced photometry measurement within an aperture of radius r = 3.00 arcsec." flxR6 ForcedWarpExtended adu/seconds REAL -999 "Flux from forced photometry measurement within an aperture of radius r = 4.63 arcsec." flxR6Err ForcedWarpExtended adu/seconds REAL -999 "Error in flux from forced photometry measurement within an aperture of radius r = 4.63 arcsec." flxR6Fill ForcedWarpExtended dimensionless REAL -999 "Aperture fill factor for forced photometry measurement within an aperture of radius r = 4.63 arcsec." flxR6Std ForcedWarpExtended adu/seconds REAL -999 "Standard deviation of flux from forced photometry measurement within an aperture of radius r = 4.63 arcsec." flxR7 ForcedWarpExtended adu/seconds REAL -999 "Flux from forced photometry measurement within an aperture of radius r = 7.43 arcsec." flxR7Err ForcedWarpExtended adu/seconds REAL -999 "Error in flux from forced photometry measurement within an aperture of radius r = 7.43 arcsec." flxR7Fill ForcedWarpExtended dimensionless REAL -999 "Aperture fill factor for forced photometry measurement within an aperture of radius r = 7.43 arcsec." flxR7Std ForcedWarpExtended adu/seconds REAL -999 "Standard deviation of flux from forced photometry measurement within an aperture of radius r = 7.43 arcsec."
ImageMeta.nx/ny
Proposal: Drop these columns, as they are currently incorrectly population, they will be identical for all rows, and the confuse the number of detection column names.
nx ImageMeta pixels SMALLINT -999 "Image dimension in x." ny DiffDetObject dimensionless SMALLINT -999 "Number of difference detections in y filter." ny ImageMeta pixels SMALLINT -999 "Image dimension in y." ny ObjectThin dimensionless SMALLINT -999 "Number of single epoch detections in y filter."
posImageID/negImageID
I cannot find which columns these link to or reference.
posImageID DiffMeta dimensionless BIGINT NA "positive image identifier" negImageID DiffMeta dimensionless BIGINT NA "negative image identifier"
psfModelID
Is the VARCHAR(100) version correct?
psfModelID ForcedWarpMeta dimensionless VARCHAR(100) "PSF model identifier." psfModelID ImageMeta dimensionless INT -999 "PSF model identifier." psfModelID StackMeta dimensionless VARCHAR(100) "PSF model identifier."
skyCellID
Which version/default is correct? Proposal: This column (skycellID), tessID, and projectionID should index a table like the GPC1.skycell table, which lists tessellation, skycell_id, ra, dec, l, b, and the width and height of the skycell on the sky. This information is currently unavailable in PSPS, and would make skycellID more useful.
skyCellID DiffDetection dimensionless REAL -999 "Skycell region identifier." skyCellID DiffMeta dimensionless REAL NA "Skycell region identifier." skyCellID ForcedWarpExtended dimensionless REAL -999 "Skycell region identifier." skyCellID ForcedWarpLensing dimensionless REAL -999 "Skycell region identifier." skyCellID ForcedWarpMeasurement dimensionless REAL -999 "Skycell region identifier." skyCellID ForcedWarpMeta dimensionless REAL NA "Skycell region identifier." skyCellID ObjectThin dimensionless TINYINT 255 "Skycell region identifier." skyCellID StackMeta dimensionless TINYINT 255 "Skycell region identifier." skyCellID StackObjectThin dimensionless TINYINT 255 "Skycell region identifier."
ForcedMeanObject/DiffDetections/ForcedWarpMeasurement.Xflags
Can these all be converted to INT/0 columns?
gFlags ForcedMeanObject dimensionless BIGINT NA "Information flag bitmask indicating details of the photometry from forced single epoch g filter detections. Values listed in ObjectInfoFlags." gFlags MeanObject dimensionless INT 0 "Information flag bitmask for mean object from g filter detections. Values listed in ObjectFilterFlags." iFlags ForcedMeanObject dimensionless BIGINT NA "Information flag bitmask indicating details of the photometry from forced single epoch i filter detections. Values listed in ObjectInfoFlags." iFlags MeanObject dimensionless INT 0 "Information flag bitmask for mean object from i filter detections. Values listed in ObjectFilterFlags." rFlags ForcedMeanObject dimensionless BIGINT NA "Information flag bitmask indicating details of the photometry from forced single epoch r filter detections. Values listed in ObjectInfoFlags." rFlags MeanObject dimensionless INT 0 "Information flag bitmask for mean object from r filter detections. Values listed in ObjectFilterFlags." yFlags ForcedMeanObject dimensionless BIGINT NA "Information flag bitmask indicating details of the photometry from forced single epoch y filter detections. Values listed in ObjectInfoFlags." yFlags MeanObject dimensionless INT 0 "Information flag bitmask for mean object from y filter detections. Values listed in ObjectFilterFlags." zFlags ForcedMeanObject dimensionless BIGINT NA "Information flag bitmask indicating details of the photometry from forced single epoch z filter detections. Values listed in ObjectInfoFlags." zFlags MeanObject dimensionless INT 0 "Information flag bitmask for mean object from z filter detections. Values listed in ObjectFilterFlags." DinfoFlag DiffDetection dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags." FinfoFlag ForcedWarpMeasurement dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags." ginfoFlag StackObjectThin dimensionless INT -999 "Information flag bitmask indicating details of the g filter stack photometry. Values listed in StackDetectionFlags." rinfoFlag StackObjectThin dimensionless INT -999 "Information flag bitmask indicating details of the r filter stack photometry. Values listed in StackDetectionFlags." DinfoFlag2 DiffDetection dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags2." FinfoFlag2 ForcedWarpMeasurement dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags2." ginfoFlag2 StackObjectThin dimensionless INT 0 "Information flag bitmask indicating details of the g filter stack photometry. Values listed in StackDetectionFlags2." rinfoFlag2 StackObjectThin dimensionless INT 0 "Information flag bitmask indicating details of the r filter stack photometry. Values listed in StackDetectionFlags2." DinfoFlag3 DiffDetection dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags3." FinfoFlag3 ForcedWarpMeasurement dimensionless BIGINT -999 "Information flag bitmask indicating details of the photometry. Values listed in DetectionFlags3." ginfoFlag3 StackObjectThin dimensionless INT 0 "Information flag bitmask indicating details of the g filter stack photometry. Values listed in StackDetectionFlags3." rinfoFlag3 StackObjectThin dimensionless INT 0 "Information flag bitmask indicating details of the r filter stack photometry. Values listed in StackDetectionFlags3."
Qf/Cf
These seem inconsistent. After asking Gene, he claims that no Cf parameters will be measured (and are scheduled to be dropped from the schema), so all coverage factors in PSPS will be Qf.
gpetCf StackPetrosian dimensionless REAL -999 "Petrosian (1976) fit coverage factor for g filter stack detection." ipetCf StackPetrosian dimensionless REAL -999 "Petrosian (1976) fit coverage factor for i filter stack detection." rpetCf StackPetrosian dimensionless REAL -999 "Petrosian (1976) fit coverage factor for r filter stack detection." ypetCf StackPetrosian dimensionless REAL -999 "Petrosian (1976) fit coverage factor for y filter stack detection." zpetCf StackPetrosian dimensionless REAL -999 "Petrosian (1976) fit coverage factor for z filter stack detection." DpsfQf DiffDetection dimensionless REAL -999 "PSF coverage factor." FpsfQf ForcedWarpMeasurement dimensionless REAL -999 "PSF coverage factor." gpsfQf StackObjectAttributes dimensionless REAL -999 "PSF coverage factor for g filter stack detection." ipsfQf StackObjectAttributes dimensionless REAL -999 "PSF coverage factor for i filter stack detection." rpsfQf StackObjectAttributes dimensionless REAL -999 "PSF coverage factor for r filter stack detection." ypsfQf StackObjectAttributes dimensionless REAL -999 "PSF coverage factor for y filter stack detection." zpsfQf StackObjectAttributes dimensionless REAL -999 "PSF coverage factor for z filter stack detection."
ra/dec
Values seem to be consistently FLOAT, errors are consistently REAL, but the defaults are not consistent. Are these correct as is, or should the defaults be different?
dec Detection degrees FLOAT NA "Declination." dec DiffDetObject degrees FLOAT NA "Declination mean." dec DiffDetection degrees FLOAT NA "Declination." decBore FrameMeta degrees FLOAT -999 "Declination of telescope boresight." decErr Detection arcsec REAL NA "Declination error." decErr DiffDetObject arcsec REAL 0.0 "Declination standard deviation." decErr DiffDetection arcsec REAL NA "Declination error." decMean ObjectThin degrees FLOAT -999 "Declination mean from single epoch detections." decMeanErr ObjectThin arcsec REAL -999 "Declination standard deviation from single epoch detections." decStack ObjectThin degrees FLOAT -999 "Declination mean from stack detections." decStackErr ObjectThin arcsec REAL -999 "Declination standard deviation from stack detections." gra StackObjectThin degrees FLOAT NA "Right ascension from g filter stack detection." graErr StackObjectThin arcsec REAL NA "Right ascension error from g filter stack detection." raBore FrameMeta degrees FLOAT -999 "Right ascension of telescope boresight." raMean ObjectThin degrees FLOAT -999 "Right ascension mean from single epoch detections." raMeanErr ObjectThin arcsec REAL -999 "Right ascension standard deviation from single epoch detections." raStack ObjectThin degrees FLOAT -999 "Right ascension mean from stack detections." raStackErr ObjectThin arcsec REAL -999 "Right ascension standard deviation from stack detections." ra Detection degrees FLOAT NA "Right ascension." ra DiffDetObject degrees FLOAT NA "Right ascension mean." ra DiffDetection degrees FLOAT NA "Right ascension." raErr Detection arcsec REAL NA "Right ascension error." raErr DiffDetObject arcsec REAL 0.0 "Right ascension standard deviation." raErr DiffDetection arcsec REAL NA "Right ascension error."
Initial pass.
This may duplicate entries on the other lists. I have not checked for duplicates yet (I will strike through the entry here when I do so).
Schema Browser
- Detection table references DetectionCalib and DetectionFull tables, which do not appear to exist.
- Typos in table description: "unless one wantw to do the our the astrometric".
- Are the FWHM values really in arcseconds?
- No description text for DiffDetObject table.
- No description text for DiffDetection table.
- DiffDetection uses "pixels" and "raw pixels". This seems like an error. "raw pixels" seems to be the standard on other tables.
- Should DiffMeta/DiffToImage be listed in "Fundamental IPP Data Products" or "Observational Metadata and Supporting System MetaData Tables"?
- ForcedMeanObject lists gFPSFFluxStd, gFKronFluxStd, gFApFluxStd, etc for other filters as dimensionless. These should be set to the correct flux unit to match the gFPSFFlux/gFPSFFluxErr values.
- gFPSFFlux/gFPSFFluxErr claim to be AB magnitudes. They are fluxes, and therefore should not be in magnitudes. A database query suggests they are in Janskies.
- gFPSFFlux and other filters has confusing Description text.
- No table descriptions for DiffMeta, DiffToImage, ForcedMeanLensing, ForcedMeanObject, ForcedWarpMeasurement.
- Are the FWHM values really in arcseconds?
- MeanObject table has no table information. Unclear table description. Quadruple-quote typos.
- Column orderings are different between g and other filters. gMeanPSFMagNpt comes after MagMin/MagMax, but r-y have this before.
- ObjectThin has confusing table description.
- I do not think the objName is being calculated correctly. For example objID = 101273355750459367, (raMean, decMean) are (335.57505452, -5.60085831). This should convert to sexagesimal as (22:22:18, -5:36:03), which does not match the given name of "PS1.2 J222218.0--62356.9". I'm not sure if this is generically true, or simply a mishandling of negative declination.
- Is this even the correct format? No. From Gene, these should be decimal.
- objName description in DiffDetObject should probably match the version found here.
- consistencyFlag does not appear populated.
- nStackObjectRows does not appear to be populated.
- nStackDetections does not appear to be correct. Example objID =101683338006825170.
- nDetections seems wrong as well. objID = 101683337908223873
Unclear cx,cy,cz,lambda,beta,l,b descriptions.ra/decErr occasionally -999, which doesn't not seem correct. Example objID = 101283355856926633.
- I do not think the objName is being calculated correctly. For example objID = 101273355750459367, (raMean, decMean) are (335.57505452, -5.60085831). This should convert to sexagesimal as (22:22:18, -5:36:03), which does not match the given name of "PS1.2 J222218.0--62356.9". I'm not sure if this is generically true, or simply a mishandling of negative declination.
- StackApFlx table description does not state the convolved FWHM values.
- primaryDetection description is unhelpful.
- gflxR5 and other radii and filters: can this list the radius in physical units as well?
- !StackApFlxExGalCon6 has no table description.
- stackDetectRowID has typo in description. Is this a bitmask? A query to the database indicates this value is unpopulated with values -999.
- primaryDetection description is unhelpful.
- gc6flxR3 and other radii/filters: what are the radius units?
- !StackApFlxExGalCon8 has the same issues as !StackApFlxExGalCon6.
- !StackModelFitDeV has no table description.
- stackDetectRowID complaint.
- primaryDetection description is unhelpful.
- DeVaucouleurs inconsistently spelled.
- StackModelFitExp has no table description.
- stackDetectRowID/primaryDetection issue.
- Exponential likely not needed capitalized in row descriptions.
- StackModelFitExtra has no table description.
- stackDetectRowID/primaryDetection issue.
- glogC/other filters description does not close parentheses.
- StackModelFitSer has no table description.
- stackDetectRowID/primaryDetection issue.
- StackModelFitPet gpetCf is in pixels, and is dimensionless in other tables (see serCf in StackModelFitSer)
- StackObjectAttributes has no table description.
- stackDetectRowID/primaryDetection issue.
- gxPos/gyPos/other filters claim to be in arcseconds, but a query suggests it is in pixels (or "stack pixels" to keep consistent with ForcedWarpMeasurement/Detection)
- gxPosErr/gyPosErr/other filters also claims to be in arcseconds. The query again suggests pixels.
- gpsfMajorFWHM/gpsfMinorFWHM claims to be in arcseconds as well. I suspect this to be in pixels as well. This point is more thorough than those for Detection and ForcedWarpMeasurement because I didn't notice any issues until going back to check from here.
- gmomentXX/XY/YY/R1/RH claim to be in units arcsec squared/arcsec squared/arcsec/sqrt arcsec. This is inconsistent with Detection and ForcedWarpMeasurement, which list all five in units of pixels.
- gapFillFac/gexpTime/filters is unclear.
- gsky/filters is unclear in units between description and units.
- Should there be an aperture radius?
- StackObjectThin has no table description.
- randomStackObjID references a random objid that does not appear in this table.
- stackDetectRowID/primaryDetection issue.
- ginfoFlag/ginfoFlag2/ginfoFlag3/filters descriptions are unhelpful.
- Why is this table StackObjectThin when it contains more information than the ObjectThin table? Is there a definition of a Thin table?
- StackPetrosian has no table description.
- stackDetectRowID/primaryDetection issue.
- gpetRadiusErr/filters largely all -999.
- when gpetMag/filters = -999, the gpetR50/gpetR90 values and errors appear to be nonsense.
- gpetCf/filters has values greater than 1. This is listed as units of pixels, but a I was expecting a coverage factor to be a 0-1 like Detection.psfQf.
- ImageFlags looks correct.
- CameraConfig has an unhelpful description.
- DetectionFlags suggests that it is only applicable to the Detection table.
- I am assuming that all *ObjectView tables pull their row description information from the tables that comprise the view, and will not re-examine those.
- Filter has an unhelpful description.
- ForcedWarpExtended has no table description.
- skyCellID has a typo in the description.
- flxR5/radii should have radius units in the description.
- obsTime description references "positive detection" which is all that exists for a forced warp measurement.
- ForcedWarpLensing has no table description.
- skyCellID/obsTime as before.
- ForcedWarpMeta has no table description.
- This table is also the first place I've seen an "alphanumeric" unit, which have been simply "dimensionless" in other tables when a similar field was a string (qv DetectionFlags).
- ForcedWarpToImage has no table description.
- FrameMeta has typos in the table description.
~p1Recip/p2Recip/p3Recip do not appear to be populated.~- expStart is in "days". Previous entries for similar time fields have used Mean Julian Days.
- ctype2 has no unit description, neither alphanumeric nor dimensionless.
- crpix1 has "pc001001" as the units, which is a header keyword. crpix2 has units "dimensionless" which doesn't seem right either although it does match other uses (DiffMeta).
- telescopeID does not reference a table.
- ImageMeta.
- imageID has undefined reference to ALEX.
- I am suspicious that the bias and biasScat values are identical for all OTAs in an exposure.
- magSat and completMag do not appear to be populated (values=0, not the default value of -999).
- nx/ny are incorrect, or at best incomplete. They list values = (590, 598) which is the size of a single cell, not the full OTA that this table is describing. Are these useful at all given that they will be identical for every exposure?
- A similar question about psfModelID (=-999).
- My continuing confusion on FWHM units. These again have values (~4) that make me suspect they are actually in pixels. Similarly for psfWidMajor, psfWidMinor.
- momentM2C/M2S/M3/M4 units. These are all pixels for the Detection table.
- detrend1-8. If these are pulling specific header values (which they appear should be), then the descriptions should be updated to reflect which detrend they are mapping. In addition,
~DETREND.NOISEMAP, DETREND.NONLIN, DETREND.VIDEODARK~, and if it's not already included DETREND.FRINGE should be included. Pulling PTRN_ROW and PTRN_CON (booleans) might also be useful. - crpix1/2 have units of pixels here, and so FrameMeta should be changed to match.
- psfModelID is a string in StackMeta/ForcedWarpMeta.
- ObjectFilterFlags has an unclear table description.
- ObjectInfoFlags has an unclear table description.
- ObjectQualityFlags has an unclear table description.
- PhotoCal references a DetectionCalib table that does not exist.
- StackDetectionFlags references a StackDetection table that does not exist.
- StackMeta
~magSat does not appear to be populated.~- expTime references the units in the description.
- completMag,astroScat,photScat,nAstroRef,
~nPhoRef~, photoColor do not appear to be populated. - crpix1/2 have units dimensionless.
- psfFwhm_mean/_max units.
- TessellationType does not have a table description.
- Should nAstroRef and nPhoRef be changed to match the usage in ImageMeta, ie numAstroRef/numPhotoRef?
General issues.
- All tables seem to have columns like "randomStackObjID" that contain a random value between zero and one. These are not ID fields in any sense.
- Primary/secondary/best detections are not clearly defined.
- All tables seem to have columns like "randomStackObjID" that contain a random value between zero and one. These are not ID fields in any sense.
- What are the stack convolution target sizes?
- All tables seem to have columns like "randomStackObjID" that contain a random value between zero and one. These are not ID fields in any sense.
Note:
See TracWiki
for help on using the wiki.
