So as promised, the SKA Members met today (in Schiphol airport, lucky them) to review the recommendation made by the evaluation committee. You can read the announcement here. The decision is ….. we are going to think about it some more !!!
Just remember, it will be very groovy when it all happens. Two things that knocked me out at NAM, SKA-wise. First, Michael Kraemer, discussing the power requirement for cooling the SKA megacomputers, said that perhaps it doesn’t need its own nuclear power plant after all. Seeing as both potential sites are sunny deserts, what we need is a vast solar power array… and this exactly what these remote areas need anyway, and they can sell power back. So there ya go. Not only did radio astronomy apparently invent wifi, but they are going to save the world too. Hows that for impact.
Knockout fact number two. I went to the session with results from LOFAR, the SKA pathfinder. As you may know, it can act as a cosmic ray telescope, detecting radio waves from the air shower made by a particle hitting the top of the atmosphere. The shower arrives at an angle, so a kind of front moves across the array. LOFAR has nano-second time resolution and each station is a hundred metres across or something like that; so they watch the front move across the array. Holy shit, they can WATCH LIGHT MOVING.
The e-Astronomer 
Today’s terse statement has already been read in a way that the dreaded option of splitting the SKA antennas between Africa and Australia is now firmly on the table; I had heard both rumors in this direction and strong denials regarding such an outcome in recent months. From a purely scientific perspective: How much (on the SKA’s initial research agenda) would be lost this way – and what may, perhaps, even be gained?
Why is it a dreaded option? It would provide some longer baselines and there is a long tradition of international collaboration in radio astronomy (undoubtedly helped by the need to cooperate in VLBI). What are the disadvantages?
The SKA is a really great proposal, but our lunch time discussion focused on how limited the phase-1 science case might be to some: EOR (why not), H I absorption (of-course), pulsars (doesn’t everyone love pulsars) and transients (those aliens might be terrorists! – http://tinyurl.com/7p3qf6w).
For many, *their* science won’t come until later stags, when we have higher frequencies, longer baselines etc. and I appreciate the real difficulty the SPO has in building phase-1 without precluding other science. Nobody dares think about what happens if the budget over-runs..
Anyway, a clear suggestion (I’m not spreading rumours, this just makes sense to me) is for AU to do the low-frequency dishes/beam former surveys and SA to concentrate on (higher frequency) single dishes; and lets face it, this is what ASKAP/MWA and MeerKAT are already doing without the SKA Organisation!
So what happens to phase-1 if we get a split-site decision? Can two sites continue with the same plan?? And how will VLBI fit into all of this???
In fact why don’t most observatories start looking at solar power? They are, almost by definition, in sunny, remote locations so using systems (like those near Seville) that use concentrated solar power (CSP) with thermal storage to provide the electrical power at night would seem to be a way to go. Would it be an issue that CSP has upfront capital costs compared to diesel or grid electricity where the costs come from the operational budget? Observatories as exemplars of good, science based, environmental decisions on energy use anyone?
Space observatories have been doing it for years.
storage? most observatories (other than radio) operate at night…
Albert suggests that pumped water storage is the most efficient but it depends on your source of energy. Using PV or wind then thats probably so – PV to mechanical is efficient and wind can be coupled to a pump directly. But you have to move a lot of water a long way – 1kWh = 1tonne lifted by 360m. The concentrated solar systems I’m interested in store energy in liquid salts heated to ~600C. 1m^3 can store about 35kWh. So, I’m guessing now, a 4m telescope with an average load of 10kW running overnight for 10 hours would need ~3m^3. ELTs and SKA on the otherhand are MWs so would need 100s of cubic metres of storage.
The heat storage far away from the telescope please! And down wind. Heated soil can do interesting things to atmospheric stability. Same for wind power. Are solar panels seeing-neutral?
I think bureaucracy folds into this too, especially with government funded (ie most) observatories. The funding line rules don’t necessarily allow them to spend money on constructing solar farms (“we’re paying you to do astronomy, not alternative energy”), even if it would represent a cost saving in the long term. You can’t for example take money allocated to build an instrument and build a solar farm with it, then build the instrument by not having as big power bills.. 🙂
I’m not speaking officially and don’t know the current exact details, but I believe the observatory I work for in Hawaii is trying to do this at our base facility in Hilo by having a 3rd party company invest in covering our roof with photovoltaic panels and establishing the grid-tie agreement with the power company, then we the observatory will simply buy (some of our) power off them. At the base facility, we have a large power draw during the day (air conditioning, etc), which makes the concept more palatable to the grid company for grid-tie. Though of course doing it this way, there’s another company taking a cut (albeit in return for up-fronting the cost of the panels etc).
Storage is totally a killer for photovoltaic power. Batteries are big, heavy, expensive, inefficient and give safety people headaches in large quantities. Grid-tie (where you feed excess power into the grid, and draw from the grid at night) is the only sane solution on this kind of scale, but then you’re at the mercy of the power company. Power companies like night-time consumers but don’t really like day-time providers because the power company has to have the generating capacity to supply peak *evening* load when all the solar is off-line, but those big generators are idle or running at much reduced capacity at night. If everyone has solar, they’re standing idle during the day too, and of course no power wants to have huge generators that they only use for 2 hours a day. Thus power companies severely limit the amount of solar grid-tie generation that they will allow to connect in to their grid…
The most efficient storage is using the solar (or wind) power to pump water up to a large water reservoir. The water reservoir later provides hydroelectric power. Of course, a desert may not be the best place for this reservoir.
What about geothermal – e.g. is MK too far away / high up from a hotspot to exploit efficiently?
Crumbs – the usual SKA paranoia posting has turned into an energy policy discussion. I think to exploit geothermal you need a rather deep hole don’t you ?
A kilometer should do it. The cubic kilometer array?
Power requirements haven’t been a major consideration for astronomy projects in the past. But as for the LHC, it will be for the SKA. It isn’t called ‘high power computing’ for nothing. Computers are hungry beasts.
“Power requirements haven’t been a major consideration for astronomy projects in the past.”
As long as you don’t count the power of the rocket used to launch a satellite.
Update : there are claims that the two-site model is under serious consideration – see recent Nature News article
What a great discuss about astro. 😉 I try to read the other and catch up what you talking about deeply.
Why how awfully nice of you to say so old chap.
My buddy advisable I’d personally potentially like this website. He or she was previously altogether suitable. This kind of create really made my time. You cannot think about substantial amount I had wasted in this details! Cheers!
I like this one. Almost poetry. Its staying in.
[…] instead of going for a single winner, rumours on the blogs suggest that the SKA Organisation may opt for splitting the SKA antennas between Africa and […]