[Artist's Concept of AXAF]

CXO Project Science:
Chandra Radiation Environment


This page addresses the radiation environment for the Chandra X-ray Observatory. It provides estimates of the external fluence of nonpenetrating radiation, which can enter the telescope's aperture and reach the focal plane. Such radiation has damaged the front-illuminated CCDs of the AXAF CCD Imaging Spectrometer (ACIS), substantially increasing their charge-transfer inefficiency (CTI). Current operating procedures, especially hiding the ACIS from viewing the telescope's aperture during radiation-belt passages, has dramatically reduced the rate of CTI increase.

Contributors to these estimates include members of the following organizations:

* Air Force Research Laboratory (AFRL) Space Hazards Branch (Sue Gussenhoven)
* Marshall Space Flight Center (MSFC) Chandra Project Science (Steve O'Dell & Doug Swartz)
* Chandra X-ray Observatory Center (Yousaf Butt, Rob Cameron, & Shanil Virani)


The prime suspect is moderate-energy (0.1-0.3 MeV) protons. Analysis of the ion transmission shows that such protons have just enough energy to penetrate the ACIS optical blocking filter and gate structure of the ACIS front-illuminated (FI) CCDs. Thus they produce most of their displacement damage in the FI-CCD's buried channel, effecting traps which increase the CTI. As a measure of the relevant proton fluence, we use the spectral intensity at 0.14 MeV.

Radiation belt

By far the largest contribution to the proton fluence in Chandra's orbit comes from the Earth's radiation belts.

i_rb = 3.7*10^11 p/(cm^2 sr MeV orbit) @ E = 0.14 MeV

We computed this value for Chandra's orbit in the AP8MAX environment, using ESA's SPENVIS (SPace ENVironment Information System) tool. As a cross-check, we also computed the fluence using AFRL's CRRESPRO tool, with CRRES-EPAS (Combined Radiation and Release Exffects Satellite - Electron Proton Angle Spectrometer) data sets. The result using the EPAS Quiet model agrees very well with the AP8MAX value; that using the Active model is nearly five (5) times higher.

ACIS went unprotected through eight (8) radiation-belt passes, during relatively quiet periods. Consequently, the concensus estimate for the external proton fluence which caused the observed CTI degradation, uses the AP8MAX (or nearly equal CRRES-EPAS Quiet) value.

I_o = 3.0*10^12 p/(cm^2 sr MeV) @ E = 0.14 MeV

This fluence constitutes the reference, with respect to which we normalize other contributions to the total proton fluence. Thus, we define a fractional degradation rate for each radiation environment.

FDR_env = ( i_env / I_o )

As an example, we compute the fractional degradation rate FDR_rb due to an unprotected radiation-belt passage.

FDR_rb = ( i_rb / I_o ) = 12.5%/orbit

Quasi-trapped region

The region of the magnetosphere immediately outside the classical radiation belts can quasi-trap particles released during geomagnetic storms. Here, we consider the proton environments at 7.5-9 R_E and at 9-11 R_E (geocentric), for the CRRES Quiet and Active models, extrapolated using data from NASA's POLAR mission. We treat the remaining major component of the magnetosphere --- namely, the geomagnetic tail --- separately (below).

For the inner (7.5-9 R_E) quasi-trapped region, the Quiet model gives the spectral intensity i_iq and fractional degradation rate FDR_iq; the Active model, i_ia and FDR_ia, respectively.

i_iq = 5.1*10^9 p/(cm^2 sr MeV orbit) = 7.0*10^11 p/(cm^2 sr MeV year) @ E = 0.14 MeV

FDR_iq = 0.17%/orbit = 23%/year

i_ia = 1.0*10^10 p/(cm^2 sr MeV orbit) = 1.4*10^12 p/(cm^2 sr MeV year) @ E = 0.14 MeV

FDR_ia = 0.35%/orbit = 48%/year

For the outer (9-11 R_E) quasi-trapped region, the Quiet model gives the spectral intensity i_oq and fractional degradation rate FDR_oq; the Active model, i_oa and FDR_oa.

i_oq = 5.4*10^8 p/(cm^2 sr MeV orbit) = 7.5*10^10 p/(cm^2 sr MeV year) @ E = 0.14 MeV

FDR_oq = 0.018%/orbit = 2.5%/year

i_oa = 4.3*10^9 p/(cm^2 sr MeV orbit) = 6.0*10^11 p/(cm^2 sr MeV year) @ E = 0.14 MeV

FDR_oa = 0.14%/orbit = 20%/year

Current operating procedures hide the ACIS from the entrance aperture when Chandra is in this quasi-trapped region. These estimates demonstrate that this operating procedure should continue, in order to reduce the risk of further significant degradation of the ACIS FI CCDs.

Solar wind

The NOAA Space Environment Center provides near-real-time solar-wind monitoring, using data from NASA's Advanced Composition Explorer (ACE), in L1 orbit. Of relevance here, the ACE Electron, Proton, and Alpha Monitor (EPAM) measures the intensity of moderate-energy (0.05-2 MeV) protons In particular, the EPAM P3 channel, centered on about 0.14 MeV, measures the spectral intensity of protons believed to be most responsible for damaging the ACIS FI CCDs.

Plot shows the daily averaged intensity of the ACE Electron, Proton, and Alpha Monitor (EPAM) P3 channel, centered on about 0.14 MeV. Online archived data are from the ACE Science Center.

Averaging over all the ACE EPAM-P3 data, we calculate the average solar-wind proton intensity i_sw.

i_sw = 3.1*10^8 p/(cm^2 sr MeV orbit) = 4.3*10^10 p/(cm^2 sr MeV year) @ E = 0.14 MeV

Thus, the time-averaged solar-wind contribution to the fractional degradation rate FDR_sw is as follows:

FDR_sw = 0.010%/orbit = 1.4%/year

To estimate the expected damage due to solar-wind protons, we must multiply this number by a duty cycle, the fraction of the time that ACIS is unprotected (i.e., in the focal plane) and Chandra is in the solar-wind environment. Further, the gratings, when inserted, provide some protection by stopping and diluting the proton intensity. Consequently, to the accuracy of our understanding of the environment and the sustained damage, solar-wind protons will effect measureable, but not significant, additional degradation of the FI CCDs, over the life of the mission. By hiding the ACIS during extreme solar events, with fluence of 10^10 p / (cm^2 sr MeV), we can further reduce this potential for damage. Hence, the Chandra X-ray Observatory Center has initiated a near-real-time solar-proton alert system, using ACE-EPAM data from the NOAA Space Environment Center.

Geomagnetic tail

Determination of the contribution of the geomagnetic tail to the external proton fluence experienced by Chandra is not yet complete. Because the geomagnetic tail swings around the Earth once a year, as the Earth revolves around the Sun, there is a seasonal dependence of the exposure of Chandra to this environment. Geomagnetic storms, initiated by variations in the solar wind, result in dramatic temporal variations in the proton intensity in the tail and back to the dusk sector of the magnetosphere.

Preliminary calculations indicate that the contribution of the geomagnetic tail to the time-averaged proton fluence exceeds that of solar wind, especially when Chandra's apogee is in the direction of the tail or into the dusk sector. Furthermore, the episodic nature of geotail activity suggests that an alert system, analogous to that used to monitor solar-wind protons, will be of value. Such an alert system will use near-real-time data from the ACE MAGnetometer instrument (MAG) and ACE Solar-Wind Electron, Proton, and Alpha Monitor (SWEPAM) to drive a Kp- or Dst-index predictor, which in turn will estimate the proton fluence accumulated in Chandra's orbit.


Return to our ACIS page, go to the Chandra Project Science Page or to X-ray Astronomy at MSFC

Contact information