IV. NEOWISE Data Processing

IV.2. Scan/Frame Pipeline Updates

IV.2.d. Photometric Calibration

The calibration of NEOWISE photometry is tied to the AllWISE Source Catalog rather than the sparse network of standard calibration stars used during the WISE primary mission. NEOWISE source fluxes are measured in instrumental units (DN) and converted to calibrated magnitudes by adding a band-dependent instrumental zero point (ZP) in the same way that was done during the primary mission (see IV.4.h.iii in the WISE All-Sky Release Explanatory Supplement). The NEOWISE photometric ZPs are computed such that the average magnitudes of bright point sources will agree with those in the AllWISE Catalog, as described below.

As with the previous WISE data releases, the magnitudes reported in the NEOWISE Source Database are in the Vega system, and represent total in-band brightness. Conversion of NEOWISE magnitudes to flux-densities, AB magnitudes, and color corrections are described in section IV.4.h.i of the WISE All-Sky Release Explanatory Supplement.

IV.2.d.i. Initial photometric zero point determination

The initial photometric ZPs for NEOWISE Reactivation were derived by first calibrating profile-fit photometry of high SNR sources from a selection of 130 test scans acquired during the first 15 days of the survey using an estimated, constant ZP. The average differences between NEOWISE and AllWISE Source Catalog magnitudes for all high SNR sources in each of the test scans were computed, and are plotted versus scan number in Figure 1. The rapid decline in the magnitude residuals correlates to the period of rapid cooling of the focal planes at the beginning of the NEOWISE survey, as demonstrated in Figure 2a. Toward the end of test scan period, the photometric residuals stabilize to the levels indicated by the horizontal magenta lines in Figure 1. The estimated ZPs were adjusted by these levels to define the initial survey photometric ZPs.

Figure 1 - W1 and W2 average [NEOWISE-AllWISE] photometry differences per scan plotted vs. scan number for ~130 test scans taken at the start of the NEOWISE Reactivation survey. The rapid decline in the average magnitudes correlates with the rapid cooling of the focal planes. The magenta horizontal line marks the adjustment to the photometric zero points once the temperature stabilized.

After setting the initial ZPs, small adjustments were made a number of times during NEOWISE operations by adding a correction, dZP, that was keyed to time (or scan number):

A summary of the W1 and W2 photometric ZPs used during the seven years of the NEOWISE survey is given in Table 1. The ZPs for the scans acquired during the initial rapid cooldown phase were adjusted frequently using the linear fits of the average residuals as a function of temperature shown in Figure 2a. The resulting average [NEOWISE-AllWISE] photometric residuals after the ZP adjustments are shown in Figure 2b. The later ZP adjustments listed in Table 1 were usually made in conjunction with updates in the image calibration products (IV.2.a.i.2).

Figure 2a - [NEOWISE-AllWISE] photometric residuals vs. BSA temperature for ~130 test scans. A linear fit algorithm designed to eliminate bias due to clustering was used to characterize the relationship in each band.	Figure 2b - W1 and W2 [NEOWISE-AllWISE] residuals vs. scan for ~130 test scans. Black points represent NEOWISE magnitudes without time-varying zero point adjustment. Red points represent NEOWISE magnitudes with corrected time-varying zero points.

IV.2.d.ii. Photometric zero point stability

The stability of the W1 and W2 photometric calibration throughout the NEOWISE survey is illustrated in Figure 3 in which is shown the trimmed average [NEOWISE-AllWISE] photometric residuals for all high SNR sources in each scan plotted as a function of time. The sense of this diagram is that larger residual values indicate lower sensitivity or system throughput. The vertical red lines in Figure 3 denote times at which adjustments were made to the photometric ZPs. The adjustment that was made on MJD=57000 (scan 55174b) is now known to have been slightly incorrect. As a result, the W2 residuals in the interval 57000<MJD<57071 (scans 55174b to 57333a) are offset from the regular seasonal variations.

Figure 3 - W1 and W2 average [NEOWISE-AllWISE] photometric residuals plotted as a function of time (MJD). The red vertical lines indicate times when adjustments were made to the photometric ZPs points.

The W1 calibration is very stable over the full survey, with an overall RMS residual of 0.0026 mags. There is a small (<<0.01 mag) systematic increase over the full span of the survey that is correlated with focal plane temperature, as can be seen in Figure 4. The periodic negative "spikes" that appear every six months in the W1 residual trend plot are related to a spatially fixed bias in AllWISE W1 photometry, and not the NEOWISE photometry.

The W2 residuals in Figure 3 exhibit both a systemic rise of ~0.01 mag over the duration of the survey, and a significant seasonal variation that rises to over 0.02 mag later in the survey. The seasonal variations track the behavior of the focal plane (BSA) temperatures that are shown in Figure 11 of I.2.c.iii. As the temperatures rise around the summer and to lesser extent winter solstices, the [NEOWISE-AllWISE] residuals increase, indicating that the W2 throughput decreases with increasing temperature. The overall RMS of the W2 residuals is 0.0061 mag, but the measured flux value for a given object can vary by up to 0.025 mags depending on precisely when it takes place.

The average [NEOWISE-AllWISE] photometric residuals in each scan are shown plotted as a function of focal plane temperature in Figure 4. With the exception of the lowest temperatures that occurred at the very start of the survey, the average residuals increase linearly with temperature. Hence, the NEOWISE system sensitivity decreases with increasing temperature. The slightly discrepant W2 points near 73.5K correspond to the period when the incorrect ZP adjustment was in place, 57000<MJD<57071. Over most of the survey, though, the sensitivity variations are small relative to the photometric uncertainties of the individual single-exposure measurements for the great majority of sources (c.f. II.1.c.iii).

Figure 4 - W1 and W2 average [NEOWISE-AllWISE] photometric residuals plotted as a function of focal plane (BSA) temperature.

Last update: 14 March 2023