A4059 was observed by Chandra for 24.6 ksec on September 24,
2000 and for 20.1 ksec on January 4, 2001, using the ACIS detector.
The central radio source, PKS 2354
35, was centered 1arcmin from
the aim point on the S3 back-illuminated ACIS chip (7) such that the
core of the cluster could be imaged entirely by the S3 chip. Here, we
analyze data from this chip only. The data were filtered so as to
include only events with ASCA grades 0,2,3,4, and 6. The gain for the
first observation was reprocessed using acis_process_events
with the latest version of the gain file appropriate for data taken
with a focal-plane temperature of 
(acisD2000-08-12gainN0003.fits). Some periods during the first
observation were affected by background flares. These were removed by
excluding all times for which the background rate exceeded the
quiescent rate by a factor of 1.2. This left a total exposure time of
10.1ksec and 17.7ksec for the two observations, respectively.
Using the blank sky background fields of the same part of the detector
we created a background image and spectrum for each observation. Each
background dataset was processed using the same gain file in the A4059
observation and the aspect solutions of each observation were applied
to the background dataset. All ACIS data reduction was performed
using the Chandra Interactive Analysis of Observations (CIAO)
version 2.2.1, and spectral analysis was performed using the XSPEC fitting package. CIAO 2.2.1 uses Calibration Database (CALDB)
version 2.12.
For both imaging and spectral analyzes, we did not use the
overall merged event list for the two observations since it does not
contain a valid aspect solution. This would lead to slight errors in
the corresponding exposure maps and insufficient information to build
correct Response Matrix Files (RMFs) and Ancillary Response Files
(ARFs). Therefore, for imaging analysis, we also generated separate
flux-calibrated images for each observation and then combined them to
produce a final image. To produce an image giving the integrated flux
over the full 
keV band, we first computed flux-corrected
images in several narrow energy bands and then summed those
flux-calibrated images. This is necessary since the effective area of
the telescope is both energy and position dependent. To do this, we
built several exposure maps which were weighted according to both
photoelectric absorption and hot diffuse gas emission models for the
incident source spectrum and therefore make it possible to compute the
correct surface brightness having less significant spectral variation
over the entire image. For spectral analysis, we generated separate
source and background spectra (together with the corresponding RMF and
ARF files) for each of the two observations and then analyzed them
simultaneously. These data reduction steps were facilitated using the
CIAO threads provided by the Chandra Science Center.
Astrometric information was obtained directly from the aspect solution provided with the Chandra data release. There are two sources in the Chandra data that can be clearly identified in other datasets that have accurate astrometry. Firstly, the brightest of the small-scale peaks in the centralmost regions of A4059 lie within 0.5arcsec of the radio core of PKS2354-35. Secondly, a bright X-ray source to the east of the cluster center lies within 1arcsec of the 2MASS source 2MASXJ23570418-3448121. Thus, we conclude that the absolute Chandra astrometry for this observation is correct to within 1arcsec.