2. Updates and Ongoing Issues

2.1 Introduction

This chapter contains information on current caveats and things to watch out for when working with XRISM data. Please also check the XRISM web page and especially Things To Watch Out For (TTWOF) page, which are updated more frequently than this guide, for the latest on newly identified issues or solutions to previously identified issues.

2.2 History of the Software & Calibration Updates

The following list summarizes updates of the XRISM software and calibration database as of Jun 12, 2026. Please refer to the XRISM software and calibration pages for the latest information.

• 2026 March: XRISM CALDB, version 260315 (CalDB 13), is released for Resolve and Xtend. See the details in the Resolve and Xtend release notes.

• 2025 September: XRISM software is updated with HEASoft version 6.36. This software includes the new tool, rslmkrsp, which accounts for pixel-to-pixel spatial variation in branching ratios for bright sources. The software also includes major updates on xmatraceback and xtdpixclip, as well as various minor updates. See the details in the release note.

• 2025 September: XRISM CALDB, version 250915 (CalDB 12), is released for Resolve and Xtend. See the details in the Resolve and Xtend release notes.

• 2025 March: XRISM software is updated with HEASoft (version 6.35). This software includes new tools as well as updates to existing tools. See the details at the release note.

• 2025 March: XRISM CALDB, version 250315 (CalDB 11), is released for Resolve.
• 2024 November: XRISM CALDB, version 241115 (CalDB 10), is released for Xtend.
• 2024 August: XRISM software was included in HEASoft (version 6.34). First XRISM CALDB, version 240815 (CalDB 9), was released.

2.3 Anomalies during Observations

XRISM observations may experience anomalies in some detectors, and the affected data may need to be treated differently or discarded. The operations team may inform the PIs of highly impacted events, but we strongly recommend that users check the operation log before data analysis to identify potential anomalies during observations. The following list shows some (but not all) of these events.

1.

Resolve Pixel 7 experienced an increase in broadband noise over a fifty-day interval estimated to extend from 2025-5-29 10:00 UTC (XRISM MET: 202176000) to 2025-7-23 14:00 UTC (XRISM MET: 206978000). This resulted in a degradation of the spectral resolution from $\sim$4.5 eV to $>$7 eV, sometimes as high as 9 eV FWHM.

2.4 Current Data Reduction Issues

Resolve

• Pixel 27 shows irregular gain variations. We recommend not using the data for scientific analyses. We advise the user to check the Resolve energy scale quality reports of their observations for possible gain anomalies affecting other pixels.

• The onboard calibration found more Ls events than expected from the ground study. The so-called anomalous Ls events do not appear to originate from direct X-ray signals from celestial objects, but background particle events or secondary X-ray events. This problem causes an uncertainty in the telescope's effective area and, therefore, source flux measurements. This problem can be easily evaded for relatively weak sources, as these events almost exclusively originate from particle events, but not for bright sources. Section 6.4 “Removing Ls Events from Cleaned Event Files" describes the problem in detail.

• There are significant differences in the spectra extracted from different Resolve pixels. The magnitude and energy dependence of the differences vary from source to source and from pixel to pixel, and are not dependent on count rate. The origin of these spectral differences is not yet understood and is under investigation. Until the origin of the spectral differences is understood, they must be treated as systematic errors. See RSL-4 of TTWOF for details.

Xtend

• Cosmic ray echo events produce false events at affected pixels at every exposure frame until the CCD's dark level is initialized daily. These events show multiple high-count dots in Xtend images. Future software is expected to effectively remove these events. For now, we suggest that users avoid analyzing data below $\sim$0.6 keV or running the HEASoft tool xtdpixclip to remove these events.