TIMED SEE Version 11 Data Product Release Notes ----------------------------------------------- Date: 1 October 2012 Notes by: D.L. Woodraska and T.N. Woods (PI), University of Colorado, Laboratory for Atmospheric and Space Physics Contact: don.woodraska@lasp.colorado.edu, 303-735-5617, or tom.woods@lasp.colorado.edu, 303-492-4224, LASP/CU, 1234 Innovation Dr., Boulder, CO 80303 See the README file for each SEE Version 11 data product for a more detailed description of the data product and information on how to read and plot the SEE data. See the "CHANGES SINCE VERSION 10" section below for information about changes to the SEE data processing algorithms. See the "KNOWN ISSUES/PROBLEMS" section below for caveats related to each data product. SEE OVERVIEW ------------ The Solar EUV Experiment (SEE) measures the solar ultraviolet irradiance for the NASA TIMED mission. The SEE instruments include the EUV Grating Spectrograph (EGS) for measuring the solar irradiance from 27 nm to 190 nm with 0.4 nm spectral resolution and the XUV Photometer System (XPS) for measuring the irradiance from 0.1 nm to 27 nm in 7-10 nm broad bands. TIMED was launched in December 2001, and SEE has made daily observations since January 22, 2002, except for a few dates listed below. The SEE Level 3 data product is considered the primary product from SEE as it combines the EGS and XPS results into a single product. SEE VERSION 11 DATA PRODUCTS: ---------------------------- * SEE LEVEL 3 DATA PRODUCT = daily-averaged solar spectral irradiance with flares removed in three forms: (1) spectra from 0-194 nm in 1-nm intervals, (2) 38 emission lines (with background removed) extracted from EGS Level 2 spectra, and (3) XPS photometer Level 2 results. The SEE Level 3 product is the most commonly used SEE product. * SEE LEVEL 3A DATA PRODUCT = ~3 minute observation-averaged solar spectral irradiance in three forms: (1) spectra from 0-194 nm in 1-nm intervals, (2) 38 emission lines (with background removed) extracted from EGS Level 2A spectra, and (3) XPS photometer Level 2A results. No flares are removed from Level 3A; however, only a few flares are observed by SEE because of its 3% duty cycle for solar observations. * SEE Space Weather (SpWx) DATA PRODUCT = 8 solar indices useful for space weather research and operations. These are provided as average over the SEE 3-minutes observations and are organized into a file per year. * EGS LEVEL 2 DATA PRODUCT = daily-averaged solar spectral irradiance with flares removed from 26-194 nm at 0.4 nm spectral resolution (in 0.1 nm intervals) * EGS LEVEL 2A DATA PRODUCT = ~3 minute observation-averaged solar spectral irradiance from 26-194 nm at 0.4 nm spectral resolution (in 0.1 nm intervals) No flares are removed from Level 2A; however, only a few flares are observed by SEE because of its 3% duty cycle for solar observations. * EGS Level 2B DATA PRODUCT = solar occultation data that are useful for deriving Earth's atmospheric density using the EGS data in the 27-190 nm range. These data provide the amount that the solar radiation is transmitted at a specific tangent height. This product is organized into a single file for the mission. * XPS LEVEL 2 DATA PRODUCT = daily-averaged solar spectral irradiance with flares removed from 0.1-34 nm and at 121.5 nm (in the bandpasses of the 9 photometers) * XPS LEVEL 2A DATA PRODUCT = ~3 minute observation-averaged solar spectral irradiance from 0.1-34 nm and at 121.5 nm (in the bandpasses of the 9 photometers) No flares are removed from Level 2A; however, only a few flares are observed by SEE because of its 3% duty cycle for solar observations. * XPS LEVEL 4 MODEL PRODUCT = 10-second model spectra from 0.05 nm to 39.95 nm in 0.1 nm intervals. The spectral model is scaled to match the signal (current) of the TIMED-SEE XPS photometer #1 or #2 (both Ti coated diodes for the 0.1-7 nm bandpass). The GOES-XRS X-ray measurements are also used to determine the temperature of the flare for use in the CHIANTI flare spectral model. KEY REFERENCES: -------------- * TIMED SEE Overview / Algorithms / Validation Paper: Woods, T. N., F. G. Eparvier, S. M. Bailey, P. C. Chamberlin, J. Lean, G. J. Rottman, S. C. Solomon, W. K. Tobiska, and D. L. Woodraska, The Solar EUV Experiment (SEE): Mission overview and first results, J. Geophys. Res., 110, A01312, doi: 10.1029/2004JA010765, 2005. * XPS Level 4 Algorithm / Validation Paper: Woods, T. N., P. C. Chamberlin, W. K. Peterson, R. R. Meier, P. G. Richards, D. J. Strickland, G. Lu, L. Qian, S. C. Solomon, B. A. Iijima, A. J. Mannucci, and B. T. Tsurutani, XUV Photometer System (XPS): Improved irradiance algorithm using CHIANTI spectral models, Solar Physics, 249, doi 10.1007/s11207-008-9196-6, 2008. * WHI Solar Irradiance Reference Spectra (SIRS) of March-April 2008: Woods, T. N., P. C. Chamberlin, J. W. Harder, R. A. Hock, M. Snow, F. G. Eparvier, J. Fontenla, W. E. McClintock, and E. C. Richard, Solar Irradiance Reference Spectra (SIRS) for the 2008 Whole Heliosphere Interval (WHI), Geophys. Res. Lett., 36, L01101, doi:101029/2008GL036373, 2009. Chamberlin, P. C., T. N. Woods, D. A. Crotser, F. G. Eparvier, R. A. Hock, and D. L. Woodraska, New, Higher Resolution Solar Extreme Ultraviolet (EUV) Irradiance Results for Solar Cycle Minimum Conditions on April 14, 2008, Geophys. Res. Lett., 36, L05102, doi:10.1029/2008GL037145, 2009. USING UNCERTAINTIES: -------------------- * Level 3 and 3A products have two propagated relative uncertainties (fractional): ERR_TOT - total accuracy, useful for comparing against other measurements and models ERR_MEAS - measurement precision, uncertainty based on count rates * Level 2 and 2A EGS and XPS products have three relative uncertainties (fractional): ERR_TOT - total accuracy ERR_MEAS - measurement precision STDEV_MEAS - standard deviation of the data averaged (over 24 hours for Level 2 and over 3 minutes for Level 2A) DATA COMPLETENESS: ------------------ * The TIMED SEE mission log is at http://lasp.colorado.edu/see/TIMED_SEE_mission_log.html * XPS Level 2 and 2A data are available starting 2002/022 (start of normal ops). * SEE Level 3, 3A, EGS Level 2, and 2A data are available starting 2002/039 (first calibration rocket flight) * Version 11 Release ends on June 30, 2012 (last calibration rocket on 23-June-2012) * Certain other data are not available as described below: * GAP-1: For day 2002/061, the TIMED spacecraft was in safe mode the entire day. SEE was not able to observe the sun, so none of the SEE data products can ever be generated for day 2002/061. The TIMED S/C was also in safe mode for a portion of the following days which have reduced data available: 2002/060 (6 orbits), 2002/078 (3 orbits), 2002/088 (12 orbits) * GAP-2: Due to an anomaly on day 2002/063 that only affected the EGS, no L2_EGS data product is available. It may be possible for us to recover some of the data from day 2002/063, so future versions may or may not have this day included. * GAP-3: New SEE flight software was loaded which reduced the data for the following two days: 2002/114 (7 orbits), 2002/115 (4 orbits) * GAP-4: Due to the XPS filter wheel anomaly, the XPS was powered off from day 2002/206-2002/210 while we studied the anomaly. To ensure the continued health and safety of the XPS component, normal XPS operations were suspended while the EGS continued collecting data. No L2_XPS data products can ever be generated from those days. The anomaly occured during day 205, so there is limited data available on that day. Operations for XPS resumed on day 2002/211, but at fixed filter wheel position 6. * GAP-5: During the 2002 Leonid meteor storm, the TIMED S/C was commanded into safe mode. The following days were affected: 2002/322 (5 orbits), 2002323 (10 orbits). * GAP-6: SEE observed a total or partial solar eclipse on one orbit in each of the following days: 2002/161, 2002/338, 2004/110. * GAP-7: TIMED S/C flight software load and testing during 2004/260-2004/265 and 2004/273-2004/275 * GAP-8: SEE operator error unintentionally left the HK data rate at 5 sec which caused the packet limit to be reached, losing 6 orbits on 2005/124. * GAP-9: 2006/020-2006/021 EGS anomaly resulting in no science being transferred. 7 orbits missed on 020, 11 missed on 021. Since there were 5 internal errors, this is believed to be caused by a particle event. Power cycling EGS returned normal operational status. * GAP-10: 2006/228-2006/230 Spacecraft demoted to safe-mode due to an anomaly with the TIMED reaction wheel #1. 8 orbits missed on 228, 15 (all) missed on 229, 8 missed on 230. * GAP-11: 2008/009-2008/015 Spacecraft demoted to safe-mode caused by an AIU zone keepout violation. Spacecraft was held in safe and new spacecraft software was loaded and TIMED now runs only in FC Only mode with 3 wheel control. * Gap-12: 2008/144-2008/145 Spacecraft demoted to safe-mode caused by an AIU zone keepout violation. * Gap-13: 2008189 SEE planning anomaly. Commands were sent to the MOC but were not uploaded due to a long holiday weekend. * Gap-14: 2009/165-2009/173 TIMED Safe mode. GNS software anomaly. Recovery triggered the default ODC on days 171-173 producing lower-quality observations. * Gap-15: 2009/336-2009/337 TIMED Safe mode. Some solar observations on DOY 337 (Dec 3). * Gap-16: 2011/331-2009/332 TIMED Safe mode. Some solar observations on each day. * WARNING-1: Mission Data Center contains noisy data: 2006/338-2006/339 * WARNING-2: Mission Data Center contains anomalous data: 2007/005-2007/008 CHANGES SINCE VERSION 10: ------------------------- * Changes affecting SEE Level 3: - Included new EGS degradation trend from 2004 to 2012 based on calibration rocket flights in 2008, 2010, 2011, and 2012 (SDO EVE calibration rocket experiments) and also verified by checking for solar cycle variability consistency across solar cycle minimum (expected solar activity level for 2004/232 and 2012/040). - Updated scattered light correction for EGS in the 27-40 nm range and also adjusted the EGS responsivity values near 30 nm so SEE He II 30.4 nm emission and SDO EVE rocket observations are in better agreement. - Included new EGS correction for the time period of 2008/252 to 2009/110 to account for additional degradation of the EGS Calibration channel due to observing the Sun twice per day with the Calibration channel instead of the normal twice per week cadence of the Calibration channel before and after this period. - Updated and extended EGS flatfield degradation, EGS gain, and EGS field of view (FOV) trends fits through 2011. - Updated the Quiet Sun (QS) and Active Region (AR) reference spectra in the XPS Level 4 model to use spectra in the 7-40 nm range derived from SDO EVE observations at 0.1-nm resolution. CHIANTI model spectra are still used for the flare spectra and also the QS and AR spectra in the 0-7 nm range. The XPS Level 4 results are also included in the SEE Level 3 and 3A spectra. - Replaced the Woods-Rottman VUV2002 model for the 1 nm bins around Lyman-alpha. The new linear model spans from 115-129 nm (excluding Lyman-alpha) and is based on 9 years of SEE EUV measurement correlations with SORCE SOLSTICE version 11 measurements. ----Relevant version 9 changes that apply to version 10---- - Updated EGS FOV correction to perform monthly averaging to remove day-to-day jumps in irradiance. This removes the need for the beta angle correction that was added into version 9. - Updated degradation algorithm to keep the normal channel in better agreement with the calibration channel. This improves the spectral shape in heavily degraded regions. - Updated EGS gain to be a function of wavelength and temperature. The present assumption is that enough degradation has occurred to affect the electrical resistance of the microchannel plates such that low-exposure regions on the detector are increasing with time. - Added a step correction for EGS data beyond 2004/263. Near this time period the spacecraft entered a prolonged safe mode demotion. An operational decision was made to point SEE down toward the spacecraft deck. We now believe this allowed a small amount of contamination to enter the aperture which decreased the sensitivity by approximately 8% in the EUV. - The responsivity of the XPS photometers was updated. - The second order subtraction of 30.4 has been adjusted to avoid zero flux. This is based on a curve fit to two gaussians, which has some problems, but is an improvement over previous versions. - Wavelengths longer than 190 nm are now zeroed out. Users interested in this range are encouraged to use the higher-precision SORCE-SOLSTICE data. ----Relevant version 8 changes that apply to version 9---- - An updated EGS flatfield degradation algorithm has been applied on a per-pixel basis that reduces noise. - A Level 0b algorithm was changed to perform better filtering of MDC problem data. - An updated long-term EGS responsivity change has been applied from 121 nm and longer using SORCE SOLSTICE, UARS SOLSTICE, and UARS SUSIM relative changes. - XPS data is now compared to GOES XRS measurements to ignore periods when particles are believed to be causing spurious high readings. This occurs at only certain orbits during certain times of the year. The upper range has been increased to accomodate large (>X10) flares. - SORCE XPS measurements are no longer used. Instead the CHIANTI spectral model in the XPS Level 4 data product is used for the Level 3 spectrum below 27 nm. ----Relevant version 7 changes that apply to version 8---- - An algorithm has been developed for the solar XUV irradiances between 0 and 27 nm and includes the use of the SORCE XPS data when available. This algorithm scales a solar model (Woods and Rottman, 2002) for the given date to match the broadband XPS measurements. This scaling factor is determined in the bands of 0-4 nm, 4-17 nm, and 17-27 nm. This new algorithm is applied for both TIMED SEE XPS and SORCE XPS data. The SORCE XPS data are used, if available, after day 2003/070. The TIMED SEE XPS data are used elsewhere. After the XPS filter wheel anomaly on day 2002/205, but before SORCE XPS data are available, the solar XUV irradiance between 4-27 nm is filled with a model of the solar variability based on the TIMED SEE XPS diode number 1 (XP#1). This solar XUV irradiance model calculation is a power-law empirically derived using the TIMED SEE XPS data before day 2002/205. One should use with caution the solar XUV irradiances in the 4-27 nm range between day 2002/205 and day 2003/070 as these "filled" data are from a model based on the TIMED SEE XP#1 diode measurement. - The XPS photometer data (Level 2) included in the Level 3 and Level 3A data products are from the TIMED SEE XPS only. Only XP#1, 5, and 10 have any valid data after day 2002/205, and the corrections for degradation for XP#5 and #10 are still problematic after 2003 in the Version 7 data products. The better XUV irradiance data to use are the irradiance values in the 1-nm spectrum, which are based on SORCE XPS measurements after day 2003/070 and TIMED SEE XPS measurements at all other times. - The XPS diode irradiances are forced to be non-negative, so the irradiances and currents may not match the Level 2 values. - The 1-nm spectrum around Lyman-alpha is filled with the Woods-Rottman VUV 2002 model scaled to match at the EGS Lyman-alpha measurement. This affects 114-129 nm. (Refer to KNOWN ISSUES/PROBLEMS below for the latest details.) * Changes affecting SEE Level 3A: - All changes to Level 3 also apply to Level 3A. * Changes affecting EGS Level 2: - All EGS changes listed in Level 3 also apply to Level 2. ----Relevant version 9 changes that apply to version 10---- - An improved mapping of the calibration channel spectrum to the normal channel channel spectrum has greatly improved the spectral shape in version 10. - An improved wavelength scale adjustment has been applied that affects integrations further from the center of the field of view. ----Relevant version 8 changes that apply to version 9---- - The long term sensitivity change corrections have been updated (see KNOWN ISSUES/PROBLEMS below). Long-term changes in the FUV sensitivity are removed based on relative comparisons with UARS SUSIM, UARS SOLSTICE, and SORCE SOLSTICE time series. Long-term changes in the EUV are removed based on a linear interpolation between the calibration sounding rocket flights for SEE (NASA 32.192 in Feb. 2002, NASA 36.205 in Aug. 2003, and NASA 36.217 in Oct. 2004). - An updated flatfield correction algorithm has been implemented (see KNOWN ISSUES/PROBLEMS below). - Artifacts not in the solar spectrum but appearing in the data near 134.5 nm and 135.5 nm have been removed from the EGS irradiances (see KNOWN ISSUES/PROBLEMS below). - The field-of-view (FOV) sensitivity correction for the EGS has been improved through a new algorithm based on data collected over a single day. The on-orbit FOV map data over the full mission have been included for the calibration channel FOV correction. * Changes affecting EGS Level 2A: - All EGS changes listed in Level 2 also apply for Level 2A. ----Relevant version 8 changes that apply to version 9---- - All changes to EGS Level 2 also affect EGS Level 2A. - The data structure has an EGS flare index tag. The index is the mean of several 0.1 nm bin irradiance ratios to the daily median values. Bins around the following lines are used for the calculation of the index: 49.9, 52.1, 58.4, 103.8, 108.5, 140.3, and 154.9 nm. This flare index can be less than 1 and sometimes exceeds 2 for the largest flares. The chosen lines sample a wide range of solar atmospheric temperatures and have shown large increases during flares. Note that while no flares are removed from Level 2A, only a few flares are observed by SEE because of its 3% duty cycle for solar observations. * Changes affecting XPS Level 2: - All XPS changes listed in Level 3 also apply to Level 2. ----Relevant version 8 changes that apply to version 9---- - The visible light correction algorithm has been improved. - The dark correction algorithm has been improved. - The degradation correction for a few of the XPS diodes has been improved (see KNOWN ISSUES/PROBLEMS below). A new second degradation function for the current (gain) conversion has been added. - The field of view maps have been improved. - Improved the data selection algorithm for calculating the averages. - An bug has been fixed relating to filter transmission history files. * Changes affecting XPS Level 2A: - All changes to XPS Level 2 also affect XPS Level 2A. ----Relevant version 8 changes that apply to version 9---- - The data structure has an XPS flare index tag. The index is the irradiance ratio to the daily median irradiance for XP#1, so values less than 1.0 are possible. For large flares, the XPS flare index sometimes exceeds 50. Note that while no flares are removed from Level 2A, only a few flares are observed by SEE because of its 3% duty cycle for solar observations. * Changes affecting XPS Level 4: - The XPS Level 4 spectra model was updated. CHIANTI spectral models were used in Version 10. Version 11 now uses Quiet Sun (QS) and Active Region (AR) reference spectra in the 7-40 nm range derived from SDO EVE observations at 0.1-nm resolution. CHIANTI model spectra are still used for the flare spectra and also the QS and AR spectra in the 0-7 nm range. The XPS Level 4 results are also included in the SEE Level 3 and 3A spectra. KNOWN ISSUES/PROBLEMS IN VERSION 11: ----------------------------------- * SEE Level 3 Issues: - All issues for EGS Level 2 and XPS Level 2 data products (listed above) apply to the SEE Level 3 data products. - The extraction of the EGS emission lines with background (continuum) subtracted is not well-tuned for all of the 38 reported lines. The 1-nm spectrum irradiance is more accurate than the emission line irradiance. - The XP#7 solar irradiances are filled after day 2002/205 using a linear relationship to XP#1 as determined using the XPS data prior to the XPS filter wheel anomaly on that date. - There is a spacecraft-to-Sun angle effect that manifests itself at the 1-3% level (most noticeable in the FUV) with a period of about 70 days. - Data corresponding to wavelengths longer than 190 nm should not be used. * SEE Level 3A Issues: - All above issues for SEE Level 3 apply for Level 3A. - Flares are not removed, so an average of Level 3A data will not necessarily agree with the daily average (Level 3). Because of SEE's low duty cycle of 3% for solar observations, the SEE data set only includes a few of the flare events. The NOAA GOES X-Ray Sensor (XRS) provides continuous sampling (100% duty cycle) if you need higher time cadence information about the solar flares. * EGS Level 2 Issues: - EGS data shortward of 27 nm and longward of 190 nm should NOT be used due to poor quality data (low signals) at those wavelengths. - EGS data between 114 nm and 129 nm, except for the 120-123 nm region, is set to zero (0.0) because of a neutral density filter in that region that permits the EGS to make a measurement of the bright HI Lyman-alpha (121.6 nm) emission, but masks neighboring dimmer emissions. - While EGS started normal operations on day 2002/022, only data starting from day 2002/039 (the date of the first SEE rocket calibration underflight) has been released with Version 7 and later because the current model of EGS degradation is not valid prior to that date. - The EGS data at 63-70 nm, at 109-114 nm, near 60.8 nm, and near 91.2 nm should be used with caution due to issues with the current algorithm for removing higher orders from the EGS data. In Version 7 and later, only the brighter emission lines seen in second and third grating orders are removed from the EGS spectra; therefore, the regions in the EGS spectra where the signal level is low, such as at 63-70 nm and 109-114 nm, might have irradiance values that are too high due to uncorrected higher order contributions to the signal. The correction of the bright HeII 30.4 nm emission at the second and third order (60.8 and 91.2 nm) is difficult because of blending with first order emissions at these wavelengths, thus the irradiances near 60.8 nm and 91.2 nm have much larger uncertainties than other nearby emissions. For these lines, the spectral shape has been scaled to compensate for the higher order line irradiance. - The EGS data have spectral artifacts near 134.5 nm and 135.5 nm which have been removed from the Level 2 product. These artifacts appear as line-like features and are probably not in the solar spectrum, but of instrumental origin. In Version 7 and later the artifacts are simply removed and replaced with constant irradiance values matching the average of the wavelength-neighboring irradiances. The source of the artifacts may be related to scatter off the EGS Lyman-alpha filter holder, but they appear to decrease with time. - The long-term FUV decreases observed in version 8 and previous data versions have been largely removed in version 9. The long term trend has been set to match the SORCE SOLSTICE measurements in 1 nm bins. - In 2004 several channels have started to show degradation which appears as quantized drops in irradiance near 60, 63, 89, 90, & 97 nm. These channels continue to make good solar observations, however the channels have increased noise. - There is a spacecraft-to-sun angle effect that manifests itself at the 2-5% level. This much less than the reported uncertainties, but a correction has been implemented in version 9 that may work better or worse in certain wavelengths at certain times. - There is a gain problem that remains uncorrected. Although a new time-dependent, improved gain correction has been implemented in version 9, this is based on the center of the FUV and applied to all wavelengths. Certain wavelengths are now observed to have a positive correlation and others a negative correlation with temperature which indicates that this needs further analysis. This could also be a manifestation of a wavelength shift at certain times. * EGS Level 2A Issues: - All above EGS Level 2 issues apply to Level 2A. - Flares are not removed, so an average of Level 2A data will not necessarily agree with the daily average (Level 2). Because of SEE's low duty cycle of 3% for solar observations, the SEE data set only includes a few of the flare events. * XPS Level 2 Issues: - Diode 1 is sensitive to particle noise that causes spurious signal increases. The algorithm for filtering out these particles uses the GOES XRS long channel and works well when there are valid signals for the GOES XRS. Unfortunately, near very quiet solar conditions the GOES XRS long channel does not respond, so particles are not filtered properly for the lowest solar conditions. The particle noise affects the diodes for short periods near the same time each year. - There are no solar XUV irradiance results from XUV Photometers (XP) numbers 4, 8, and 12 because these are visible-light (bare) photodiodes used to calibrate the window transmissions. - Results from XP#3 (with a bandpass of 17-26 nm) should NOT be used because of a problem with the field-of-view correction used in Version 6 and later processing. - Due to the XPS filter wheel anomaly on day 2002205, only the XP#1, #5, and #10 measurements continue to the present. All other XPS channels have NOT made solar irradiance measurements since 2002205. (More complete spectral coverage for the solar XUV irradiances is provided by the SORCE satellite which has an XPS on board. The SORCE XPS instrument began daily measurements in March 2003.) - Due to increased exposure on XP#5 and #10 caused by the filter wheel anomaly there has been a noticeably increased degradation in the field-of-view sensitivity of these two diodes. This time-dependent change has been corrected for in Version 7 and later, but negative irradiance values for these diodes sometime exist when solar activity is low. * XPS Level 2A Issues: - All above XPS Level 2 issues apply to Level 2A. - Flares are not removed, so an average of Level 2A data will not necessarily agree with the daily average (Level 2). Because of SEE's low duty cycle of 3% for solar observations, the SEE data set only includes a few of the flare events. * XPS Level 4 Issues: - All above XPS Level 2 and Level 2A issues apply to XPS Level 4. - The high resolution spectra in Level 4 is a MODEL not a direct measurement. The CHIANTI spectral model is scaled to match the signal (current) of XPS photometer #1 or #2. So the Level 4 irradiances over broad bands (7-10 nm) are considered accurate, but the irradiance in each 0.1 nm interval is not validated. See the XPS algorithm paper listed above (Woods et al., 2008). - The daily (non-flare) irradiance spectra in the XPS Level 4 have been greatly improved in Version 11 by switching to reference Quiet Sun and Active Region spectra in the 7-40 nm range derived from SDO EVE observations at 0.1-nm resolution. A calibration rocket flight on 23-June-2012 also provided a new X-ray spectral irradiance for the 0.1-7 nm range, and this preliminary result indicates that the XPS Level 4 in the 0-3 nm range has a spectral shape that is ten (10) times too high as compared to the rocket result. With the integrated 0-7 nm irradiance from XPS considered accurate, we expect that the 3-7 nm range of the XPS Level 4 model is too low. **** END OF FILE ****