So when five years ago the world learned that bees in America and Canada were dying in large numbers, and hives were becoming defunct, the agricultural community and the beekeepers and just plain people became alarmed. Hives were deserted, the bees gone, presumably dead, honey production stopped, and the bee industry crippled.

The problem was called Colony Collapse Disorder, and it threatened California’s very profitable almond industry, which is dependent on bees to pollinate the trees and keep the nuts growing. And not just almonds: 130 crops in California alone depend on honey bees. Beekeepers from around the nation load their hives on trucks and bring them to California and rent them out to growers. As the disease, or whatever it is, spread, the price of renting ever-more-scarce bees went up.

Honey bee hive at UC Davis

Once the news media started reporting heavily on the plight of the bees and the beekeepers, interest soared. Researchers at universities around the country started looking into the problem; money was donated to figure out what was killing the bees. Stories appeared frequently about the scientific efforts to figure out what was causing Colony Collapse Disorder (CCD) and what would cure it. With all that attention you’d think they would have solved the problem.

But what the scientists have discovered is that they really don’t know very much about bees. They don’t have a baseline of what goes on on the microscopic level in the hives. What viruses already exist in healthy colonies? You’ve got know that before you can start to understand if a virus is normal or abnormal and may be killing bees. Scientists like Joe DeRisi at the University of California San Francisco say they’ve made great strides, even though they haven’t found a culprit.

DeRisi: “I think there’s been tremendous progress. One of the frustrating things with CCD is it doesn’t look like there’s any one single agent or culprit that you can point the finger to that’s causing all of these problems. It looks to be a confluence of things that is several different pathogens or situations or environmental conditions that are coming together to cause losses that are more than would be expected. And that’s what’s frustrating people. What has occurred because of the interest in honeybees and because of the large losses caused by CCD is people like myself and other researchers around the country applying new techniques and tools to honeybees which they normally would not have done so, and so we’ve learned an incredible amount about the ecosystem in the bee and around the bee. And what we now know is that there’s a whole host of pathogens no one knew anything about and that certain combinations of these appear to be associated with higher losses than would otherwise be expected during the season. “

DeRisi’s lab discovered four new viruses that exist in healthy hives they never knew existed before. But that didn’t solve the problem at hand.

The disease remains a serious threat, with about a third of all bee colonies affected, and no cure in sight. But many among the other two-thirds of the beekeeping community think they have it under control, because their hives are doing well. They claim they take better care of their bees, feed them better, and use various medicines and techniques to keep the hives healthy.

One technique some beekeepers swear by is splitting the hives every year or even more frequently. That means taking half the bees out, getting a new queen (you can buy queens!), and making two hives out of one. Eric Mussen, a university extension bee specialist at the University of California at Davis, thinks splitting works – up to a point:

Mussen: “When you make these splits, you more or less take the pathogen load, all the problems, you kind of split it in half and then you’ve got these little colonies that have to build up really quickly and when that happens frequently they can outrun some of the parasites. They can outrun some of the disease problems for awhile, so those colonies get up and they make it and they’re, they’re good for a season. Okay, had you not split it, it seems like in many cases the microbes and the parasites become overwhelming and the colony dies, so my terminology is starting from packages, making splits, if you could keep your colonies forever young it looks like that’s a, a way that helps deal with the problem. Nothing’s perfect.

Q: Why hasn’t that completely eradicated this problem then? Why isn’t everybody splitting?

Mussen: Well, a number of people are splitting, either by default or some by design. They’ve, they’re now understanding what the problem is and, and how this helps. But the problem is that I think some of the equipment has or whatever the CCD problem is, is kind of innate in the equipment and so it really doesn’t matter what bees you put in and how you deal with them, it’s always right there, right on the edge ready to create a real problem. So you do the best that you can to try to just stay a little bit ahead of that.”

The research goes on – and so does pollination. The almond industry is surviving, and in fact, thriving. Last year was the largest crop ever. The crisis mentality seems to have passed, but the problem remains. While beekeepers are used to cycles where their bees die off, and then come back, Colony Collapse Disorder seems to be more persistent than previous die-offs, and shows little sign of abatement. While it hasn’t been decoded nor cured, it has focused attention on a unique part of agriculture that seems to need the attention. And that’s not honey-coating the progress that has been made.

Infrared image of a zebra from the London Zoo.
Credit: Steve Lowe

Why this particular mountain? There are essentially three reasons:

It's dark
It's clear
It doesn't make the stars twinkle

The first two reasons are so obvious that I am almost embarrassed. The last reason is not quite so intuitive. What makes a star twinkle and why do we care? This goes back to a post I made a few months ago.

The basic idea here is that the churning atmosphere blurs your astronomical image. Local geography and weather patterns can either mitigate or exaggerate this effect. It is difficult to predict and many measurements need to be done to determine what is actually happening. Cameras were placed all around southern Utah on various mountain tops to observe the North Star over the course of the year. The mountain top that produced the highest resolution image of the star won the competition. That was Frisco Peak.

The telescope that will be placed on Frisco Peak was built by a very specialized company. This is quite rare–more typical are either large custom-made telescopes or small amateur telescopes. This telescope falls in the middle. It is bought off the shelf but is far superior to the commercially made amateur telescopes.

We are now discussing plans for this telescope, like the type of cameras that should be used. There is a strong interest in building an infrared camera. This allows us to see through large clouds of dust and allows us to see very distant galaxies.

Like most people, I am much more experienced with cameras in the visible spectrum. I work on CCDs in Berkeley and have barely used anything in the infrared. CCDs are made of silicon which is sensitive to light that can be seen with the naked eye (plus a little more red than what can be seen).

However, there is a lot of information in the sky that is too red to be seen with the naked eye and too red to be detected with a silicon detector. New materials are required for detectors in this wavelength range. One of the major new materials for infrared detectors is a blend of mercury, cadmium and telluride, usually called Mer-Cad-Tell in the astro community. The wavelength range of the detector can be tuned by changing the amount of mercury in the blend.

Clearly, a lot of the legwork has been done for this new telescope. We have the funding, we have a vendor, and we have a location. Now all that's left is to prioritize our science goals and to figure out how to get our hands on some mer-cad-tell.

Kyle S. Dawson is engaged in post-doctorate studies of distant supernovae and development of a proposed space-based telescope at Lawrence Berkeley National Laboratory.

37.8768 -122.251

Last week I went to a talk given by the leader of the Supernova Factory collaboration at LBNL. What is SN factory? This is an ambitious project to study supernovae like never before. I mentioned this project briefly in a previous post , now that they are so close to releasing their results I want to discuss it a bit more.

The main idea of this project is to study several hundred nearby supernovae using an instrument known as the Supernova Integral Field Spectrograph, or SNIFS. This type of instrument is essentially a blend between a traditional imaging camera and a spectrograph.

The resolution in an integral field spectrograph is defined in spaxels instead of the pixels that have become all too familiar with the advent of digital cameras. A spaxel is quite similar to a pixel, there aren't nearly as many and each one carries at least a 1000 times as much information.

In your digital camera, the light passes through the lens and directly onto the CCD. Each pixel on the CCD counts the number of photons in the red, the blue, and the green. Typically, there are millions of pixels, each counting photons from a slightly different region of the subject of your photograph.

Now imagine that instead of just counting red, green, and blue, that each pixel counts the entire rainbow of light from your subject. Now you have a spaxel. In an intregral field unit, the light passes through an array of microlenses and prisms before landing on the detector. We would call each set of microlenses and prisms a spaxel. The resulting image carries information about every wavelength of light from every region of your target.

Spectrum of the first SN observed with SNIFSThe advantage to an integral field spectrograph like SNIFS is that you gain a lot more information than either an imager or spectrograph alone. With an integral field spectrograph you can basically identify and organize every photon that reaches the telescope.

Specifically designed to observe supernovae, SNIFS is being operated at the 88-inch telescope on Mauna Kea. Spaxels are quite expensive – this particular instrument has only 225. However, this is more than enough to observe the entirety of a galaxy, a supernova, and the background.

The members of the SN Factory have now observed over 100 SNe using this new camera. Last Thursday, I saw the data from the first 25 well-calibrated supernovae and was very impressed. The data showed the evolution of each supernova and the properties of the host galaxy in great detail. I'm sure the supernova community will be equally impressed when they first see these new results.

Kyle S. Dawson is engaged in post-doctorate studies of distant supernovae and development of a proposed space-based telescope at Lawrence Berkeley National Laboratory.

37.8768 -122.251

Star or Comet?

Remember the solar wind I wrote about a few weeks ago? This stream of protons does more than create comet tails and aurora, it also destroys all of those fancy electronics we work so hard to put into orbit.

The protons streaming from the sun carry a lot of energy, and they leave a lot of this energy behind as they pass through satellites and astronauts that don’t have the Earth’s atmosphere to protect them. The energy released wrecks havoc on the system, throwing electrons and atoms around like a game of ping-pong. This is one form of radiation damage.

Definitely a comet!
This radiation damage is harmless over short periods of time, much like an occasional X-ray at the dentist. However the solar wind becomes a problem for something like the Hubble Space Telescope or our proposed satellite SNAP which are exposed for many years.

To understand how a telescope degrades from exposure to radiation, let me give an extremely quick explanation of how we gather astronomical images. A telescope is very similar to a camera you buy in the store. The large mirror is equivalent to the lens on your camera. The part that suffers the most radiation damage is the Charge Coupled Device, also known as a CCD.

The CCD is essentially the same as the 8-megapixel chip in your digital camera. This serves as an electronic version of film, recording the image through the photoelectric effect rather than through a chemical reaction. If you can still remember how photography was in the days of film, I'm sure you can appreciate the relief of going digital. Astronomers realized this early on and were pioneers in the use of CCDs.

The photons from the subject of the photograph collide with electrons in the silicon of a CCD, knocking them free from their parent atom. The free electrons are then collected in a well near the site of the collision. Once the exposure is complete, charge is moved one well (or pixel) at a time toward a transistor which then reports the number of electrons found. This process is usually described through the analogy of a bucket brigade passing buckets of water from a reservoir to a fire.

When the CCD is brand new, the bucket brigade performs almost perfectly. If I want to observe a star, the image comes out crystal clear. However, after enough time in space and in the solar wind, the CCD begins to show its wear. The bucket brigade gets sloppy at work and has to contend with an increasingly difficult obstacle course, spilling a little bit of water (or electrons) during each transfer. That same star now leaves a trail of charge behind and begins to look more like a comet.

Now, if I am observing a star, I want my image to look like a star, not like a comet. Is that really too much to ask? Unfortunately, the CCD will inevitably deteriorate in space and astronomers have to find ways to predict and correct for this deterioration. This is what we spent yesterday discussing. We passed around some pretty good ideas but still have a bit of work to do before we can prove a new method for correcting the images. I just hope we it figured out before our satellite launches in 2015!

Kyle S. Dawson is engaged in post-doctorate studies of distant supernovae and development of a proposed space-based telescope at Lawrence Berkeley National Laboratory.

latitude: 37.8768, longitude: -122.251