Terra LUT Update History for MOD_PR02 The AlgorithmPackageVersion is used to describe the version of the LUTs. The first 3 digits of the AlgorithmPackageVersion are the same as the PGEVersion and the 4th digit is the LUT update number relative to that version. v2.5.5.1_Terra 02/12/2001 ============================================================================ QA LUT CHANGES: New serial number: Q006 2001:02:12:15:00 "Detector Quality Flag Values" bit 1, dead detector flag, was set to 1 for band 30, detectors 5 and 8 (product order). This means that EV pixels will not be calibrated for those detectors. Rather, the scaled integer will be interpolated from the nearest, live detectors. Previously, detector 5 of band 30 was flagged as having an out-of-family gain. v2.5.5.0_Terra 01/12/2001 ============================================================================ There was no significant change to the science content of the LUTs. However, there were some changes to the structure and number of LUTs. REFLECTIVE LUTS ADDED: "E_sun_over_pi" (RSR-weighted solar irradiance/pi for RSB detectors) EMISSIVE LUT DELETED: "Number of overlap scans for temperatures" EMISSIVE LUT NAME CHANGE: "Number of overlap scans for b1 and T BB" changed to "num_overlap_scans_b1" v2.5.4.0_Terra 12/01/2000 ============================================================================ Changes were made to the structure of the Reflective LUTs file as part of the code change implementing the new SWIR out-of-band (OOB) correction algorithm. The new algorithm derives a correction to dn based on a quadratic equation in terms of dn_28, the dn determined for band 28. The new SWIR LUTs are the coefficients of the quadratic relation. REFLECTIVE LUTS REMOVED: "Z_b28_to_5micron" "X_OOB" "RVS_5micron_EV" "RVS_5micron_SV" "RVS_5micron_BB" "RSR_5micron_OOB" "WL_RSR_5micron_OOB" "NWL_RSR_5micron_OOB" REFLECTIVE LUTS ADDED: "X_OOB_0" (constant, offset, term of correction) "X_OOB_1" (linear term of correction) "X_OOB_2" (quadratic term of correction) SCIENCE CHANGES: "Serial Number of Reflective LUT": R008 2000:12:01:18:00 "X_OOB_0" -- currently set to zero "X_OOB_1" -- consistent with the m1 and R* LUTs delivered as part of the last LUT update (v2.4.4.2). "X_OOB_2" -- currently set to zero Science values for the new LUTs are implemented only for the B-side. The correction switch is similar to v2.4.4.2, where the switch is ON for the B-side but OFF for the A-side. Thus, there will be no effective change in the A-side calibration. v2.4.4.2 (Terra) 11/30/2000 ============================================================================ Changes were made to the B-side SWIR out-of-band (OOB) correction LUTs to improve the calibration of the first subsample (in bands 5, 6, 7) and for band 26. The calibration of the second subsample of bands 5, 6, 7 remains suspect. The A-side LUTs remain unchanged. REFLECTIVE LUTS CHANGED: "Serial Number of Reflective LUT": R007 2000:11:29:18:00 Science changes were made in the reflective LUTs. "m1" (Reflectance calibration linear term) "R_star" (Radiance gain factor) "X_OOB" (OOB response reciprocal gain factor) "SWIR_OOB_correction_switch" (switch ON for B side, OFF for A-side) To implement the switch ON for B-side LUTs and OFF for A-side luts, this switch has a time dependence similar to those described under the v2.4.4.1 LUTs. v2.5.3.0_Terra (NOT USED IN OPERATIONS) 11/22/2000 ============================================================================ New code version. The LUT science content is the same as v2.4.4.1. (See the MOD_PR02 HISTORY file for a description of the code changes). v2.4.4.1 (Terra) 11/16/2000 ============================================================================ TIME DEPENDENCE: Due to the switch from side "A" to side "B", this LUT version incorporates two sets of values for some LUTs: 1) A-side LUTs for instrument configuration Itwk/Vdet = 110/226 2) B-side LUTs for instrument configuration Itwk/Vdet = 79/110 The appropriate set is selected automatically by the L1B code based on data collection time. The switch to B-side occurred on day 2000/304 at 14:00, which is used as the time stamp for the B-side LUTs in this version. However, after switching to B-side, there was a 2-hour warm-up period. All Level 1B products during this 2-hour warm-up period should be considered unreliable. Also, the instrument was initially configured at Itwk/Vdet = 79/190. The instrument was not configured for the current state of Itwk/Vdet = 79/110 until day 2000/305 at 19:12:36. The change in Itwk/Vdet is expected to only affect the PV bands (5,6,7,20-30). Thus, the Level 1B products for those bands between day 304, 16:00 and day 305, 19:12:36 should also be considered unreliable. The specific LUTs which have the above time dependence are: Emissive: A0, A2, Band_21_b1, Ucoeff_Calibr_resid Reflective: DN_sat, K_inst, m1, R_star, RSB_NEdL, Sigma_m1, Sigma_K_inst, Sigma_R_Star_Lin_Resid_Ucoeff, T_inst_ref QA: Detector Quality Flag Values All other LUTs in this version are constants (independent of data collection time or electronics side). EMISSIVE LUTS CHANGED: "Serial Number of Emissive LUT": E005 2000:11:16:09:00 Science changes were made in the emissive LUTs. "L_Max" (Max value of L for scaling to scaled integer) Values for some bands were raised to prevent large numbers of instances where L > L_Max. Changes include: band old value new value 24 0.374 1.0 25 0.968 1.76 28 4.906 5.86 30 6.974 12.68 33 7.216 7.9376 34 5.522 6.0742 35 4.862 5.3482 36 3.256 3.5816 The effect on the calculated radiance (after rescaling the product scaled integer) is typically less than 0.05%. These value have no time dependence (i.e., they apply to any data collection time). "A0", "A2" (emissive calibration quadratic coefficients) "Band_21_b1" (b1 value for all band 21 detectors) "Ucoeff_Calibr_resid" (residual uncertainty to calibration fit) B-side LUTs were added as described in the "TIME DEPENDENCE" section. The B-side LUTs were derived from on-board black-body data collected on days 305/306 with Itwk/Vdet = 79/110. REFLECTIVE LUTS CHANGED: "Serial Number of Reflective LUT": R006 2000:11:16:09:00 Science changes were made in the reflective LUTs. "DN_sat" (DN at saturation or where saturation begins) "K_inst" (Instrument temperature correction factor) "m1" (Reflectance calibration linear term) "R_star" (Radiance gain factor) "RSB_NEdL" (RSB noise equivalent delta radiances) "Sigma_m1" (uncertainty in m1) "Sigma_K_inst" (uncertainty in K_inst) "Sigma_R_Star_Lin_Resid_Ucoeff" (uncertainty due to linear fits) "T_inst_ref" (Instrument tepmerature reference) B-side LUTs were added as described in the "TIME DEPENDENCE" section. Some of the above, such as DN_sat, K_inst and T_inst_ref were derived from pre-launch B-side data. Other, such as m1, R_star and Sigma_m1 were derived from Solar Diffuser calibration data collected on orbit on day 305 with Itwk/Vdet = 79/110. QA LUTS CHANGED: "QA serial number": Q005 2000:11:16:09:00 Science changes were made in the QA LUTs. "Detector Quality Flag Values" (noisy, dead, out-of-family, ...) B-side LUTs were added as described in the "TIME DEPENDENCE" section. In particular, for the B-side with Itwk/Vdet = 79/110, there are no detectors that are flagged as dead (non-functional). v2.5.2.0_Terra (NOT USED IN OPERATIONS) 10/16/2000 ============================================================================ No change. The PGE version changed due to re-baselining with the new MCFs v2.4.4.0 (Terra) 09/26/2000 ============================================================================ No change. The PGE version changed due to code bug fix. v2.5.1.0 (Terra) (NOT USED IN OPERATIONS) 08/24/2000 ============================================================================ No change. The PGE version changed due to code bug fix. v2.5.0.0 (Terra) (NOT USED IN OPERATIONS) 08/18/2000 ============================================================================ No science changes from version 2.4.3.1. v2.4.3.1 08/10/2000 ============================================================================ EMISSIVE LUTS CHANGED: "Serial Number of Emissive LUT": E004 2000:08:11:16:00 This was changed to reflect that science changes were made in the emissive LUTs. "PC_XT" (PC bands cross-talk correction) PC bands (B32-36) optical leak correction coefficients are based on day 084 Moon's observation. "A0", "A2" (emissive calibration) These values are based on the OBC BB cool-down calibration on day 102 (Itwk/Vdet=110/226) data set. The A0 coefficient contains three coefficients of A00, A01, and A02 to compensate the variation on instrument temperature T_inst. Because T_inst has been fairly stable (2 deg. variations) during the mission as compared to the T/V test at different plateaus, the A01 and A02 are set to ZERO for this delivery. This means A0 is not instrument temperature dependent. Similar comments hold for A2. "RVS_1km_Emiss_EV" (earth-view response vs. scan angle) This update was made using the C-NAD data on day 118. Changes are made for mirror side 2 only. They are detector dependent and normalized at OBC BB angle. Here are the rules: For dead detectors: Do not update the RVS. (see QA LUTS CHANGED section for list of dead detectors) For Band 21: Apply the same value (to all detectors) from the average of Band 22 (dead detector excluded). For Band 25 det 9&10 (product order): Use the average from the other 8 detectors. For Band 27 det 9 (product order): Use the average from the other 9 detectors. "RVS_1km_Emiss_SV" (space-view response vs. scan angle) (similar comments as above for "RVS_1km_Emiss_EV") "Band_21_b1" (values for b1 for each band 21 detector) The new values are based on recent b1 trending on band 21. The b1 for the dead detector (#9 product-order) is not updated for this delivery. "L_Max" (upper limit of radiance for scaled integer) The values for bands 31 and 32 were increased to 26.2 and 22.7, respectively. These values correspond to a saturation temperature of about 388K. REFLECTIVE LUTS CHANGED: "Serial Number of Reflective LUT": R005 2000:08:11:16:00 This was changed to reflect that science changes were made in the reflective LUTs. "m1", "R_star" (reflective calibration) New values are based on solar diffuser data collected on Day 171. "Sigma_m1" (uncertainty in m1) New values are based on solar diffuser data collected on Day 171. "SWIR_OOB_correction_switch" (5.3 micron leak correction switch) The value was changed to 0 (OFF). This turns off the 5.3 micron out-of-band (OOB) leak correction for SWIR bands (5,6,7 and 26). QA LUTS CHANGED: "QA serial number": Q004 2000:08:11:16:00 This was changed to reflect that science changes were made in the QA LUTs. "ALGORITHMPACKAGEMATURITYCODE" (maturity of the algorithm as a whole) The value was changed to "PREL" -- preliminary. EOS platform is flying development code at best; frequently changing, not stable. "Detector Quality Flag Values" (noisy, dead, out-of-family, ...) This was updated to reflect current assessment of detectors. Bit 1: Non-functional (dead) detectors band detector (product order) 6 3 6 7 7 14 21 9 22 4 24 9 By setting these values, the corresponding scan lines will have interpolated values for scaled integers, if possible to obtain. (See "OTHER COMMENTS" section regarding other bits). "Moon Offset Limits" (defines keep-out-box for moon in SV port) The preflight values of this LUT were found to be too large. The limits (units of 1km pixels) for all bands were changed to: track: -15 to 15 scan: -55 to 55 This will reduce the number of scans flagged as having the Moon in the SVP without degrading calibration. OTHER COMMENTS: 1. Detector quality flag values for thermal emissive bands (20-25, 26-36) Bit 0 Noisy Detector (NEdL@L_typ > 2 X mission specification) None. Note: There are some detectors (B22, 24, 25, 33, 34, 36) which are close to 1 X mission specification. Bit 2 Out-of-Family Gain (Gain deviates > 10% from median) Results based on the b1 on-orbit trending. Bit 3 Dynamic Range (Saturates before mission specification Lmax) Band 21 expected saturation before its Lmax (@500K) Bit 4 Detector DN saturates on blackbody when Tbb=290K (DN > 4095) None. Bit 5 High calibration-fit residuals (1% deviation from the fit) Based on the OBC BB cool-down calibration on day 102, which is the same data set used for A0 and A2 delivery. Bit 6 Electrical cross-talk. All detectors in thermal bands 20-25, 32-36 were flagged. 2. Detector quality flag values for reflective bands (1-19, 26) Bits 3, 4, and 5 are unchanged from previous LUT. They are described below for completeness. Bit 0 Noisy Detectors. Determined from the day 2000171 SD Calibration collect data. The analysis was performed using FFT as was done for the SNR analysis. Bit 2 Derived from R* determined from day 171 data. Bit 3 Dynamic Range Set the following to 1: Bands 2, 5, 6, 13hi, 14hi, and 17 - all detectors. Band 7, detectors 1,2,3,4,6,7,8,9,10,13,14,15,16,17,19,20 Bit 4 Detector DN saturates on calibration source Set bands 8 through 16, all detectors to 1. Bit 5 High calibration fit residuals Set all RSB bands, all detectors to 1. Bit 6 Electrical cross-talk. All detectors in bands 5,6,7 and 26 were flagged. v2.4.3.0 (NOT USED IN OPERATIONS) 07/25/2000 ============================================================================ Scientifically, these LUTs are the same as versions 2.3.2.3, 2.4.0.0, 2.4.1.0 and 2.4.2.0. This set was created as part of PGE version 2.4.3. v2.4.2.0 06/09/2000 ============================================================================ Scientifically, these LUTs are the same as versions 2.3.2.3, 2.4.0.0 and 2.4.1.0. This set was created as part of PGE version 2.4.2. v2.3.2.5 (NOT USED IN OPERATIONS) 05/22/2000 ============================================================================ Not in operations. For testing by the land team only. EMISSIVE LUTS (same as V2.3.2.2) REFLECTIVE LUTS (same as V2.3.2.2 except for the following) New Serial number: R004 2000:05:22:08:00 LUTs changed: m1, R_star, Sigma_m1, SWIR_OOB_correction_switch The new calibration LUTs apply to: Day 055, 21:19 to Day 057, 15:00 (excluding the period on Day 056, 22:00 to 22:10). The SWIR_OOB_correction_switch was set to zero. This turns off the out-of-band leak correction algorithm. QA LUTS (same as V2.3.2.2) v2.4.1.0 (NOT USED IN OPERATIONS) 05/05/2000 ============================================================================ Scientifically, these LUTs are the same as versions 2.3.2.3 and 2.4.0.0. This set was created as part of PGE version 2.4.1. v2.4.0.0 (NOT USED IN OPERATIONS) 04/24/2000 ============================================================================ Scientifically, these LUTs are the same as version 2.3.2.3. This set was created as part of PGE version 2.4.0. v2.3.2.4 (NOT USED IN OPERATIONS) 04/20/2000 ============================================================================ Not in operations. For testing by MCST and Miami, only. EMISSIVE LUTS New Serial number: E003 2000:04:20:15:00 LUTs changed: A0, A2 These LUTs were generated to help resolve discrepancies noted by Miami in their images. They are not planned to be placed in operations at this time. REFLECTIVE LUTS (no changes) QA LUTS (no changes) ============================================================================ v2.3.2.3 03/15/2000 ============================================================================ EMISSIVE LUTS (Emissive_Lookup_Tables_file) ------------------------------------------- (no changes) REFLECTIVE LUTS (Reflective_Lookup_Tables_file) ----------------------------------------------- New Serial number: R003 2000:03:15:12:00 LUTs changed: DN_sat K_inst m1 R_star Sigma_K_inst Sigma_m1 X_OOB (see "SWIR LUTs" comments below) RSR_5micron_OOB (see "SWIR LUTs" comments below) WL_RSR_5micron_OOB (see "SWIR LUTs" comments below) NWL_RSR_5micron_OOB (see "SWIR LUTs" comments below) QA LUTS (QA_Lookup_Tables_file) ------------------------------- New Serial number: Q003 2000:03:15:12:00 LUTs changed: Detector Quality Flag Values This version has been updated to reflect the gains calculated from solar diffuser response. In addition, some detectors previously flagged as exhibiting crosstalk and noise have been cleared, per Bruce Guenther's instructions. Changes in "SWIR LUTs": I. X_OOB: The values of the X coefficients were computed using post launch data from day 057 (Feb 26,2000), during which the Vdet and Itwk values were set to 110 and 226. BB sector background corrected count rates from six granules MOD02OBC.A2000057.1705.002.2000061034610.hdf MOD02OBC.A2000057.1735.002.2000061053446.hdf MOD02OBC.A2000057.1935.002.2000061054508.hdf MOD02OBC.A2000057.1940.002.2000061055139.hdf MOD02OBC.A2000057.2055.002.2000061055417.hdf MOD02OBC.A2000057.2145.002.2000061071658.hdf were fitted to a linear function of blackbody and scan mirror temperatures as discussed in the memo Call #: M0850 Title: Prototype On-Orbit Correction Algorithm for Out of Band (5.4) Spectral Response in the MODIS SWIR Bands Author(s): Reichert, G., Godden, G., Xiong, (X.) Jack, Berriman, B., Guenther, B. Date: 2/7/2000 (see memo equation 24). Data were taken from subframe 1 for bands 5 and 7, and from subframe 2 for band 6. Data from both mirror sides were fitted simultaneously. The fits did not yield reasonable values for four detectors (in product order, band 6 detectors 2, 6, band 7 detectors 11, 13). Values for these detectors were set to zero. II. RSR_5micron_OOB: The band 7 RSR now covers the entire range of measurement above 2.3 microns. The values were rebinned to 27 wavelength bins. III. WL_RSR_5micron_OOB: The wavelength range for band 7 now contains 27 values covering the entire range above 2.3 microns over which the RSR was determined. IV. NWL_RSR_5micron_OOB The numbers of wavelengths for band 7 were changed to 27. ============================================================================ v2.3.2.2 02/25/2000 ============================================================================ EMISSIVE LUTS (Emissive_Lookup_Tables_file) ------------------------------------------- (no changes) REFLECTIVE LUTS (Reflective_Lookup_Tables_file) ----------------------------------------------- (no changes) QA LUTS (QA_Lookup_Tables_file) ------------------------------- New Serial number: Q002 2000:02:24:13:00 LUTs changed: "Detector Quality Flag Values" Bit 0 (Noisy Detector) and Bit 6 (Electrical Crosstalk) set for every detector in bands 5, 6, 7, and 20 - 30 inclusive. Bit 2 has been set for RSB detectors where appropriate. ============================================================================ v2.3.2.1 02/16/2000 ============================================================================ EMISSIVE LUTS (Emissive_Lookup_Tables_file) ------------------------------------------- New Serial number: E002 2000:02:16:09:00 LUTs changed: Band_21_Uncert_Lsat A0 A2 Ucoeff Ucoeff_Calibr_resid Band_21_b1 TEB_specified_uncertainty TEB_UI_scaling_factor REFLECTIVE LUTS (Reflective_Lookup_Tables_file) ----------------------------------------------- New Serial number: R002 2000:02:16:09:00 LUTs changed: T_inst_ref R_star K_inst m1 RVS_250m RVS_500m RVS_1km_RefSB Sigma_RVS_RSB Sigma_K_inst Sigma_R_Star_Lin_Resid_Ucoeff DN_sat T_FPA_ref delta_DN_RSB RSB_specified_uncertainty RSB_UI_scaling_factor QA LUTS (QA_Lookup_Tables_file) ------------------------------- (no science changes) ============================================================================ v2.3.2.0 01/05/2000 ============================================================================ Baseline set of LUTs for MOD_PR02 code version V2.3.2.