A Proposal for Land-Based, High Resolution, Continuous Coverage Visible Light Imaging of Solar System Objects
Original proposal: Apr 13, 2009
Updated: June 29 2010
Proposer
This project is proposed by the author (Anthony Wesley). I am an Australian citizen with backgrounds in science, electronics, computing (hardware and software) and a keen amateur astronomer.
I am perhaps best known for the recent pair of discoveries on Jupiter of an impact by a small asteroid on July 19 2009 and first detection from Earth of a Jovian fireball on June 3 2010.
Both of these discoveries come as a direct result of the ongoing, high resolution monitoring campaign that I am operating on Jupiter and Saturn, as a private citizen and amateur astronomer.
This proposal predates both of these discoveries, however it has been updated with additional content to reflect recent events.
My contact details are as follows:
Anthony Wesley
82 Merryville Drive
Murrumbateman, NSW 2582
Australia
+61 2 62270891
+61 0458090971 (mobile)
Section 1 – Introduction and Overview
1.1 Overview
1.2 Project Goals
1.3 Background and Discussion
1.4 Project Timeline
Section 2 – Hardware
2.1 Optical Evaluation and Design
2.2 Cameras
2.3 Filters
2.4 Thermal Environment and Thermal Control
2.5 Adaptive optics
2.6 Weather Sensing
2.7 Other Infrastructure
2.8 Estimated Infrastructure Costs
Section 3 – Software
3.1 Software Assisted Collimation
3.2 Target Acquisition and Tracking
3.3 Real-Time Data Acquisition and Analysis
3.4 Post-processing
3.5 Advanced Image Stacking and Modeling
Appendix 1 – Sample images by the author
Appendix 2 – Information about the author
Section 1 – Introduction and Overview
1.1 Overview
1.2 Project Goals
1.3 Background and Discussion
1.4 Project Timeline
1.1 Overview
This proposal describes a project to design, construct and operate a network of low-cost, remote controlled (or autonomous) telescopes optimised for high-resolution imaging of bright solar system targets such as the giant planets Jupiter and Saturn in bands such as visible (RGB), methane absorption (889nm) , near-IR (750-900nm) and UV (350-400nm).
This network of instruments would also be well positioned to provide continuous coverage of other targets of opportunity such as earth-crossing or near-earth asteroids and other planets such as Mars, Uranus, Neptune, but with reduced functionality and resolution.
Each telescope in this network would be located at a site of known good seeing and would be accompanied by a set of advanced software systems for image processing to provide high-resolution visible-light data suitable for analysis by planetary and atmospheric scientists, and would assist in the study and modeling of atmospheric systems on planets such as Jupiter and Saturn.
I would like to be employed by your agency to undertake this project as one member of a small team who would oversee and manage the project through its lifetime. I am an Australian citizen and I would be willing to relocate if necessary to take part in or direct this project.
1.2 Project Goals
- Provide continuous coverage of dynamic solar system objects such as Jupiter, Saturn and other planets as well as the larger asteroids and other targets of opportunity.
- Provide sustained, reliable, high resolution imaging of these targets using a combination of advanced hardware and software techniques, achieving resolutions of < 0.5 arc seconds on a routine basis (eg by combining simultaneous data from several individual sites).
- Provide full 3 dimensional maps of the gas giants Jupiter and Saturn on a routine basis, and other targets of opportunity such as Mars when possible.
- Act as a prototype and development platform for a next generation of more powerful land-based planetary imaging systems.
1.3 Background and Discussion
The scientific community has grown accustomed to the stunning, high resolution planetary images that have been provided by the NASA/ESA Hubble Space Telescope of Mars, Jupiter and Saturn, however with the end of the Hubble telescope coming in approximately 5 or 6 years from now we will be left with no space-based visible light instruments. This work will then be left to the ground-based telescopes.
With many of the larger professional observatories fully booked in their normal lines of research it has fallen mostly to the amateur astronomy community over the last several years to provide regular, high resolution images of planets such as Jupiter, Mars and Saturn. Rapid advances in camera technology and personal computer speeds and software mean that extremely good high resolution images are now possible with equipment that is within the budget of a dedicated amateur planetary astronomer. (refer to my sample images in Appendix 1).
Indeed in the last few years the amateur imaging community has provided much of the visible light research data on targets such as Jupiter. This data has been used in papers published in peer-reviewed journals(1). It is only when the amateur results show some dynamic feature of interest that time is taken on larger instruments, including the HST, to capture higher quality images.
In this regard the amateur community has shown what is possible with modern, commercially available cameras and imaging techniques that are now routinely outperforming the large telescopes of the previous generation in this domain at a small fraction of the cost .
The advances in ccd imaging and compute speeds make software algorithms and techniques possible now in (or near to) real time to produce images such as those in Appendix 1 on a routine basis from locations of good seeing.
My proposal is to construct a network of some small number - perhaps 6 – telescopes that are optimised for high resolution planetary imaging. They would be operated remotely, with as much of the systems to be automated as possible. These instruments should be located around the world in regions of known good seeing to provide continuous overlapping coverage of planets such as Jupiter for at least 6 months of the year and possibly as much as 9 months, losing time on any particular target only when the target is close to and in solar conjunction.
These instruments would each be in the small- to medium- size category, perhaps with a mirror diameter in the 0.5m to 1m range (to be decided), making it possible for them to be co-located at existing observatories in a small self-contained dome. They would use, as far as possible, commercially available hardware to lower the cost of construction and ongoing maintenance.
They would each employ a number of advanced techniques to optimise their high resolution imaging capability, including a minimal central obstruction and associated small field of view, a highly thermally controlled environment to minimise internally generated thermal noise and wavefront distortion, leading to a high strehl and suitable long focal length (matched to the chosen camera).
The proposed network would also act as a testbed and development platform for the next generation of larger, more powerful land based high resolution imaging systems. Advances in adaptive optics and atmospheric modeling can be trialled on this network to provide feedback to the development teams in a fast, time efficient manner.
(1) Sánchez-Lavega, A. et al. Nature 451, 437–440 (2008).|Article|
1.4 Project Timeline
By preferring commercially available components where possible the overall construction and maintenance costs of this network can be reduced.
The equipment costs will depend on final decisions abut the size of these instruments and whether certain systems such as Adaptive Optics can be usefully employed. There are many variables to be considered in the costing, including decisions about the final desired resolution of the resulting images.
There are too many unknowns at this stage of the proposal to attempt an accurate costing, however a rough costing is given in 2.8, with the assumption that all parts are purchased commercially. In practice it is hoped that some cost savings can be made by using (or re-using) parts and expertise internally available.
The expected project timeline may go as follows:
Stage 1 – Identification of first site, design and construction of prototype instrument, development of software.
Time allowed – 1 year.
Stage 2 – Identification of at least 2 more sites, construction of instruments from the prototype.
Bring the operational network to 3 instruments total.
Time allowed – 1 year.
Stage 3 – Identification of the last 3 sites, construction and commissioning.
Bring the operational network to 6 instruments total, giving overlapping coverage.
Time allowed – 1 year.
At the end of the 3 year development cycle the project should be assessed, including options for upgrading specific components such as the high resolution cameras and communications systems to the current state of the art. It is hoped that ongoing operational funding can be obtained for whatever useful lifespan is envisaged for this network, including further equipment and software upgrades as required.
Stage 4 – Ongoing operation
Time – as long as ongoing funding is provided. Regular upgrades would be done to keep up with the current state of the art in commercially available components such as cameras and optics.
Section 2 – Hardware and Infrastructure
2.1 Optical Evaluation and Design
2.2 Cameras
2.3 Filters
2.4 Thermal Environment and Thermal Control
2.5 Adaptive optics
2.6 Weather sensing
2.7 Other Infrastructure
2.8 Estimated Infrastructure Costs
2.1 Optical Evaluation and Design
Much of this section is open to discussion and debate.
There are many possible optical designs to be considered, however the specific design criteria for high resolution (ie long focal length) planetary imaging imposes some natural limits on the choice of optics and mount.
One possible scenario would be the use of classic newtonian optics and design to give an instrument with a primary mirror diameter of 0.5m and native FL of around 2m, ie f/4, increasing this to the desired FL by use of high quality focal extenders (barlows). This arrangement minimises the central obstruction and provides considerable flexibility in adjusting focal length by changing the optics near the camera. I have experience with focal extenders ranging from 2x to 8x, and for this type of imaging – ie on-axis, small fov, they have proven to be very suitable.
This arrangement also minimises the overall cost as many standard fabrication techniques can be employed in addition to the purchase of commercial newtonian optical components.
It is recognised that other optical arrangements are possible, and are open to discussion – eg the cassegrain style that is more common in larger instruments, including systems such as the Dall-Kirkham. The final choice of optical system will be directed by the goals of achieving the desired resolution and functionality while minimising the overall cost and maintenance.
These issues are all known and solved in existing large observatory installations, however one of the the goals of this proposal is to lower the costs of the instrument so that a network of many instruments can be constructed and operated, with associated lowering of maintenance and repair costs. This may necessitate some compromise in the choice of optical and other systems.
The short exposure nature of this imaging, combined with the intended processing software (see 3.5 Advanced Stacking and Modeling) would allow the use of either equatorial or alt-az mounting systems since the individual frames of data would be rotated and merged onto an appropriate 3-dimensional surface when stacking.
2.2 Cameras
Each instrument will be fitted with at least two cameras – the High Resolution Imaging Camera and the Low Resolution Target Acquisition camera. These will be chosen from the current crop of commercially available high-sensitivity, fast frame rate machine vision cameras.
The exact specifications for the High Resolution camera are to be determined as part of the specifications for the project, however there are a number of machine vision cameras commercially available that would fit the requirements, including cameras from Canadian company Point Grey Research. I have used three of the high sensitivity, high framerate cameras from this company – the Dragonfly Express and Dragonfly2 cameras as well as their most recent model, the Flea3 that has enhanced red and near-IR sensitivity – with good results.
The Low Resolution Target Acquisition camera uses a low-cost, low-resolution and low data rate commercial camera as a widefield target acquisition device. It is assumed that the pointing accuracy of the instrument as a whole will be sufficient for the target to appear in the Low Resolution camera and then the instrument control systems can command the scope to bring the target to the centre of the field by monitoring the images acquired from the Low Resolution camera.
Once in the centre of the field the target will be visible in the High Resolution camera, which will take over the fine-grained pointing and control of the instrument to keep the target in the centre of the field.If the target is lost then the Low Resolution camera is again used to reacquire the target.
2.3 Filters
The primary goal of this system is to provide high resolution visible-light images and so the filter set will be optimised around high quality RGB filters, augmented with UV, IR, CH4 and other narrowband filters such as O3 or others.
Extreme narrowband filters have proven to be useful on targets such as the moon where there is plenty of illumination available and the extremely narrow bandpass provides an increase in effective resolution by reducing any remaining chroma or aberrations once the position of best focus is achieved.
It is expected that a 5 or 6 position filter wheel equipped with the best available commercial filters would be sufficient.
It is possible that the exact set of filters may vary between sites in the network in order to maximise the coverage of any particular target across more filters than can be selected at any individual site.
2.4 Thermal Environment and Thermal Control
I have considerable experience in attempting high resolution imaging in environments which are not thermally stable – ie at altitude where the air temperature drops severely after sunset and continues to drop steadily until well after midnight. This is the normal environment around my home observatory, which is not at all optimal for high resolution imaging.
As a result I have spent the last several years actively researching and exploring thermal control systems for my equipment in order to overcome this problem as far as possible.