SUPPORTED Several camera/filter combinations of HST do not sample the point spread function with a sufficiently small sampling interval to retrieve the full spatial resolution possible at the wavelength of the filter. This is especially true for the NIC3 camera. Dithering observations by sub-pixels is the standard solution. However, the extra overhead time of dithering can be too long to make dithering feasible for many solar system targets where rotation or relative motion is noticeable during the overhead time. I suggest to test a new method where a single exposure yields 13-22 dithered images which provide sufficient subpixel information to recover the whole spatial capabilities of HST. I suggest to test this method with Jupiter imaged by six NIC3 filters. Still exposures in the same six filters of NIC1 or NIC2 will be added to provide the standard for comparison of spatial resolution. The method lets Jupiter smear across NIC3 by about 4-6 pixels during the whole exposure by c hanging HST's tracking rates. Each interval between readouts of NIC3 provides a dithered image. It is difficult to predict how well the reduced NIC3 images will compare with the still NIC1 and NIC2 images with respect to spatial resolution, but one orbit of HST can test the method. If this method works well, it could be applied to many other future observations.

I propose to image Jupiter during one orbit with NIC1, NIC2, and NIC3. NIC1 and NIC2 Imaging: During the last (or first) part of the orbit, one exposure just north of the equator of Jupiter will be taken with six filters each, with exposure times in parenthesis: NIC1: F108N (16 s), F113N (32 s). F166N (32 s), F190N (16 s), NIC2: F212N (32 s) and F215N (32 s). These exposure times are expected to give sufficient signal-to-noise ratios for a comparison with the NIC3 images. NIC3 Imaging: During the main part of the orbit, four or five exposures each will be taken in six NIC3 filters, the same filters as for NIC1/NIC2: F108N (2.8 s), F113N (38 s), F166N (28 s), F190N (3 s), F212N (104 s), and F215N (44 s). Each of 13-22 readout intervals is expected to give signal-to-noise ratios of about 50 per pixel. The total accumulation after the last readout will be typically 100, 000 to 200, 000 electrons, mostly below the saturation at 185, 000 electrons, although the brightest pa rt s of Jupiter may saturate during the last readouts. The readout modes will be SCAMRR, STEP2, and STEP8. The telescope will drift with respect to Jupiter so that at each main readout, the pointing has shifted by 0.2- 0.4 pixels. The smear angle will be 18 degrees from the detectors x-axis or y-axis which makes sure that each readout probes a different fractional pixel. Ideally, 10 positions shifted by 0.32 pixels each along the tracking axis generate a perpendicular, eqidistant grid with 10 times as many pixels, and thus the effective pixel size reduced by a factor of square-root 10, to 0.32 pixels. Each actual grid will have a somewhat different spacing, but with the additional readouts (13-22 instead of the minumum required of 10), the sampling will mostly be sufficient. Tracking of HST is done by defining a fake satellite near the center of Jupiter with an orbital radius of 1800 to 2400 km and an appropriate orbital period. Unless the satellite is caught near its sta ti onary points, the multiple readouts will provide sampling about every 0.2-0.4 pixels (0.04-0.08 arc-seconds) which records all available spatial information. Note that the main readouts of each MULTIACCUM exposure with STEP2 and STEP8 are considered those with equal temporal spacing. Extra readouts of these sequences are not used. Telescope Time: The total telescope time is 6 min for acquisition, 18 min exposing on Jupiter, 12 min for exposure overheads, 11 min for both changes between the three cameras, 6 min for pointing the telescope at moving targets, totaling 53 min, which is the maximum time available for one orbit on Jupiter. Time-Critical Observations: It is requested to schedule the observations when a Galilean satellite is in the field of view of NIC3 and a shadow transits over Jupiter's disk. The satellite and shadow will provide useful test objects. Preferentially, another satellite and/or shadow should be imaged simultaneously for an even better test. The best possible test is with three shadows and two satellites in transit simultaneously. Such an event occurs only about two times per century. One such event occurs during Cycle 12, on March 28, 2004. OFFSET JUPITER PLANET STD=JUPITER TYPE=POS_ANGLE,RAD=16.0,ANG=312.0,REF=NORTH OFFSET JUPITER PLANET Wobbles by 3.5 arc-seconds at a rotational period of 28 seconds for exposure times near 2.8 seconds. STD=JUPITER TYPE=SAT,A=5800.0,EPOCH=28-Mar-2004:08:30:00.0,N=1100000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 OFFSET JUPITER PLANET Wobbles by 3.5 arc-seconds at a rotational period of 30 seconds for exposure times near 3 seconds. STD=JUPITER TYPE=SAT,A=5800.0,EPOCH=28-Mar-2004:08:30:00.0,N=1020000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 OFFSET JUPITER PLANET Wobbles by 1.1 arc-seconds at a rotational period of 56 seconds for exposure times near 28 seconds. STD=JUPITER TYPE=SAT,A=1800.0,EPOCH=28-Mar-2004:08:30:00.0,N=550000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 OFFSET JUPITER PLANET Wobbles by 1.3 arc-seconds at a rotational period of 76 seconds for exposure times near 38 seconds. STD=JUPITER TYPE=SAT,A=2200.0,EPOCH=28-Mar-2004:08:30:00.0,N=410000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 OFFSET JUPITER PLANET Wobbles by 1.5 arc-seconds at a rotational period of 88 seconds for exposure times near 44 seconds. STD=JUPITER TYPE=SAT,A=2400.0,EPOCH=28-Mar-2004:08:30:00.0,N=353000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 OFFSET JUPITER PLANET Wobbles by 1.2 arc-seconds at a rotational period of 208 seconds for exposure times near 104 seconds. STD=JUPITER TYPE=SAT,A=2000.0,EPOCH=28-Mar-2004:08:30:00.0,N=150000.0,L=0.0,I=4.2,O=138.0,EQUINOX=J2000 Note that either Visit 1 or Visit 2 needs to get exectuted. Orient 140.7 to 141.7 would work almost as well. Other windows with less satellite shadows on Jupiter are: 05-Mar-2004:07:05 to 05-Mar-2004:07:25, 08-Mar-2004:20:05 to 08-Mar-2004:20:40, 12-Mar-2004:09:05 to 12-Mar-2004:09:55, 15-Mar-2004:22:00 to 15-Mar-2004:23:10, 19-Mar-2004:11:05 to 19-Mar-2004:12:30, 23-Mar-2004:00:05 to 23-Mar-2004:01:35, 26-Mar-2004:13:00 to 26-Mar-2004:14:30, 30-Mar-2004:02:10 to 30-Mar-2004:03:30. Orient 140.7 to 141.7 would work almost as well. PI submitted two visits. STScI is to execute the one that fits into the period we will execute the proposal in. Currently this looks to be visit one. This visit will be with drawn once visit one executes. 1/30/04 wj