Monday, 28 September 2015

Resurrecting megafauna: the various problems of de-extinction

The record-breaking success of Jurassic World proves that if there's anything a lot of people want to see in the animal kingdom it is species that are both large and fierce. Unfortunately, in these post-glacial times that type of fauna has been much reduced and will no doubt wane even further - not that I particularly wish to encounter an apex predator at close quarters, you understand.

Hollywood, of course, has much to answer for. There was plenty of poor science in the original Jurassic Park movie - the use of gap-filling frog DNA being a far worse crime in my book than the over-sized velociraptors (think Achillobator and similar species) but the most recent film in the franchise has pointedly ignored the advances in dinosaur knowledge made in the intervening period. Perhaps a CGI test of a feathered T-Rex looked just to comical?

In contrast, the amount of publically-available material discussing de-extinction has increased exponentially in the two decades since Jurassic Park was released, with the line between fact and fiction well and truly blurred. That's not to say that an enormous amount hasn't been learned about the DNA of extinct species during this period. I recently watched a rather good documentary on the National Geographic channel (yes, it does occasionally happen) about the one-month old baby mammoth Lyuba, recovered in Siberia almost forty-two thousand years after she died. The amount of genetic information that has been recovered from mammoths is now extremely comprehensive, but then they were alive until almost yesterday at geological timescales. Needless to say the further back in time a creature existed, the more problematic it is to retrieve any genetic material.

A lot has been written about the methods that have been, or could in the near future, be used to resurrect ancient animals. Some procedures involve the use of contemporary species as surrogate parents, such as elephants standing in for mother mammoths. But it seems fair to say that all such projects are finding difficulties rather greater than originally planned. One common misconception is that any resurrected animal would be a pure example of its kind. Even the numerous frozen mammoth carcasses have failed to supply anywhere near a complete genome and of course it isn't just a case of filling in gaps as per a jigsaw puzzle: one primary issue is how to know where each fragment fits into the whole. Our knowledge of genetics may have advanced enormously since Watson and Crick's landmark 1953 paper, but genetic engineering is still incredibly difficult even with species that are alive today. After all, Dolly the sheep wasn't a pure clone, but had nuclear DNA from one donor and mitochondrial DNA from another.

Therefore instead of resurrecting extinct species we would be engineering hybrid genomes. Jurassic World took this process to the extreme with Indominus rex, a giant hybrid of many species including cuttlefish! Some research suggests that the most of the original genes of any species over a million years old – and therefore including all dinosaurs – might never be recovered. Something  terrible lizard-ish may be built one day, but it would be closer to say, a chicken, with added teeth, a long bony tail and a serious attitude problem. In fact, George Lucas has been a key funder of the chickenosaurus project with aims along these lines. Let's hope he doesn't start building an army of them, totally obedient clones, ready for world domination…oh no, that was fiction, wasn't it?

But if – or more likely, when – creating variants of extinct species becomes possible, should we even attempt it? Apart from the formidable technical challenges, a lot of the drive behind it seems to be for populating glorified wildlife parks, or even worse, game reserves. The mock TV documentary series Prehistoric Park for example only contained large animals from various periods, frequently fierce carnivores, with no attention given to less conspicuous creatures or indeed flora. This gee-whiz mentality seems to follow a lot of the material written about de-extinction, masking some very serious long-term issues in favour of something akin to old-style menageries. Jurassic Park, in fact.

A big question that would be near impossible to answer in advance is whether such a species would be able to thrive or even survive in a climate far removed from the original, unless there was major genetic engineering just for such adaptive purposes. Again, the further back the animal lived, the less likely it is that there is a contemporary habitat close to the original. It may be possible to recreate glacial steppes suitable for some mammoth species, but what about the Earth of ten million or one hundred million years ago? Prehistoric Park got around the issue for its Carboniferous megafauna by housing them in a high oxygen enclosure, which is certainly a solution, if something of a fire hazard!

Any newly-created animal will lack the symbiotic microbial fauna and flora of the original era, but I've not seen much that tackles this issue. I suppose there could be a multi-stage process, starting with deliberate injections of material in vitro (or via the host /mother). But once the animal is born it will have to exist with whatever the local environment/habitat has to offer. The chimerical nature of the organism may help provide a solution, but again this takes the creature even further from the original.

Then there is the rather important issue of food. To his credit, Michael Crichton suggested in Jurassic Park that herbivorous dinosaurs swallowing gizzard stones might accidentally eat berries that their metabolism couldn't handle. It would be extremely expensive to maintain compounds large enough for megafauna that are constantly kept free of wind-blown, bird-dropped and otherwise invasive material dangerous to the animals.

If the hybrids were allowed free reign, what if they escaped or were able to breed naturally? Given a breeding population (as opposed to say, sterilised clones) evolution via natural selection may lead them in a new direction. It would be wise to consider them as an integral part of the ecosystem into which they are placed, remembering Darwin's metaphor of ten thousand sharp wedges. Is there a possibility that they could out-compete modern species or in some other way exacerbate the contemporary high rate of extinction?

I've previously discussed the dangers of deliberate introduction of foreign species for biological control purposes: surely introducing engineered hybrids of extinct species is the ultimate example of this process? Or would there be a complete ban on natural reproduction for resurrected species, with each generation hand-reared from a bank of genetic material? At this point it should be clear that it isn't just the nomenclature that is confusing.

Some research has been undertaken to investigate the de-extinction of species whose demise during the past few centuries can clearly be blamed on humans, obvious examples being the Tasmanian tiger and the nine species of New Zealand moa. It could be claimed that this has more to do with alleviating guilt than serving a useful purpose (assuaging crimes against the ecosystem, as it were) but even in these cases the funds might be better turned towards more pressing issues. After all, two-thirds of amphibian species are currently endangered, largely due to direct human action. That's not to say that such money would then be available, since for example, a wealthy business tycoon who wants to sponsor mammoth resurrection - and they do exist - wouldn't necessarily transfer their funding to engineering hardier crops or revitalising declining pollinating insect species such as bees.

As it happens, even species that existed until a few hundred years ago have left little useable fragments of DNA, the dodo being a prime example. That's not to say that it won't one day be retrievable, as shown by the quagga, which was the first extinct species to have its DNA recovered, via a Nineteenth Century pelt.

As Jeff Goldman's chaos mathematician says in Jurassic Park, "scientists were so preoccupied with whether or not they could that they didn't stop to think if they should". Isn't that a useful consideration for any endeavour into the unknown? If there's one thing that biological control has shown, it is to expect the unexpected. The Romans may have enjoyed animal circuses, but we need to think carefully before we create a high-tech living spectacle without rather more consideration to the wider picture than appears to currently be the case.

Wednesday, 19 August 2015

Stars in the city: an introduction to urban astrophotography

As a twelve year old astronomy nut, I was lucky enough to receive a small refracting telescope. Almost immediately, I utilised scrap timber to build an observatory in my back garden, just about large enough for two children (plus star charts, a moon map and at least as important in my opinion, a flask of hot chocolate). I recall it even had a sliding roof, thanks to a pair of dismantled wardrobe doors.

Although the imaging wasn't too bad - I lived in a small town, so light pollution was relatively low - I soon discovered that good optics are only part of the story: without a proper mount, a telescope can be next to useless. In this particular case, I obviously hadn't read the brief introduction to mounts in my trusty The Observer's Book of Astronomy by Patrick Moore. At any rate, I clearly didn't understand the difference between proper equatorial or alt-azimuth mounts and the piece of junk that allowed my refractor to sit on a table top. Therefore, except for getting to know the lunar landscape, I saw little that I couldn't more easily view with my 20x50 binoculars.

Jump forward thirty or so years and courtesy of a large tax refund I found myself in possession of a small reflector, complete with equatorial mount and right ascension motor. After some months getting to know it I started buying accessories, aiming to learn the ins and outs of astrophotography. Thanks to numerous websites I picked up some useful techniques and excellent free software - and as importantly, how to use the assemblage - and now feel it's about time I offered a one-stop-shop guide to getting the best images on a low budget in your own backyard. Of course there are plenty of books available, but most are at least one to two hundred pages long and often specify expensive kit, so this post is an attempt to cover the gap for those wanting an astrophotography 101 with the absolute minimum of basic equipment. Of course, it's entirely my approach, so there are no doubt plenty of other tutorials out there. But at least mine's short!

1. Equipment

I have to admit that I order all my kit from overseas, since New Zealand has few astronomy retailers and those there are appear to have a fairly limited range, often at uncompetitive prices. However, it is possible to accumulate a decent beginner's assortment for around a NZ$1000 / £500. I would always recommend a reflector as a first telescope, being far cheaper than a refractor with similar capability. The Newtonian is the most common, least expensive and easiest to maintain type of reflector, mine being a Sky-Watcher 130. As per the name, the primary mirror is 130mm (about five and a half inches in old money), which is really the minimum useful size for a reflector.

The telescope came with a red dot finder scope, several okay-ish eyepieces, a right ascension motor drive, a poor 2x Barlows and a reasonably stable equatorial mount. Since then I've bought a planetary camera, a good quality 2.5x Barlows, a compact camera adaptor, an adjustable polarising filter and a collimating eyepiece*. I've also made my own Bahtinov mask, courtesy of a website that supplies patterns for various diameter/focal length combinations. Although 'go-to' mounts are available, I agree with the general consensus that the best way to learn the night sky is by manually pointing the telescope, not just programming a target and letting the telescope slew into position for you.

*For complete newbies, a Barlows is a cheap method for increasing magnification with only a limited number of eyepieces, fitting into the eyepiece holder below the eyepiece. A collimator is used to check and correct misalignment between the primary and secondary mirrors, whilst a Bahtinov mask is a simple focussing aid.

I'm lucky to live in the 'winterless north' of New Zealand, but for those in colder climates it's probably wise to make or purchase a dew cap, or rather one for the main tube and another for the finder scope. A rubber eyecup for the eyepiece might also be a good idea; there's not much point in trying to observe anything if water is condensing on the mirrors or lenses.

I would recommend a CCD or CMOS telescope camera or modified webcam, since they are a lot cheaper than a digital SLR and far lighter. The EQ2 mount supplied with the Sky-Watcher needs adjusting on both axis depending on the combination of items in the eyepiece holder, otherwise at high angles it has a tendency to droop. The EQ2 counterweight can just about handle the long tube: experiments with a compact digital camera in a purpose-built mount have confirmed that additional off-centre mass requires regular fine-tuning to retain balance. Incidentally, I use a colour planetary camera since I tend to have short sessions - around two hours - and so only want to film each pass once rather than repeating in triplicate for colour filters, even if mono cameras achieve better resolution.

2. Where to observe?

Of course this is the least flexible part of astrophotography, since you are restricted by the buildings and trees in your garden - or any other convenient location. Not only is your view of the night sky limited by physical obstructions but pollution can severely impact viewing. As I have discussed previously, light pollution is the most obvious form, with street lighting often worse than that of buildings. I've found that even as low as ten percent cloud cover can degrade astrophotography, due to the artificial light reflecting off the clouds.

Heat pollution may be less obvious but can also severely reduce image quality. Therefore, try to avoid pointing the telescope directly above nearby rooftops or you will be looking through a rising column of hot air, either the radiating heat from earlier that day or leaking from poorly-insulated buildings that are heated at night. Also, never set the telescope up indoors and point it through an open window: the thermal variations will generate shimmering galore. Wind above the lightest of breezes cannot be recommended either, not just for 'scope instability but also because dust and particulates can deteriorate the viewing. High water vapour content is bad for the same reason; here in humid Auckland I'm frustrated by the hours before and after rain, meaning the best seeing I've ever had has been in high summer after a rain-free week.

Before using a reflecting telescope, it needs to be set up outdoors well in advance of the viewing session in order to allow the mirror to cool down to the ambient temperature. The cooling time is directly proportional to the primary mirror diameter, which for my 130mm is usually about one hour.

3. What to photograph?

For urban astrophotography I've found the moon and planets to be by far the best targets. By planets I mean just Mars, Jupiter and Saturn. Venus may be both large and bright but due to its cloud cover will never present anything other than a featureless crescent or globe.

The moon is endlessly fascinating, best observed between new moon and first or last quarter (i.e. half full). During these periods, the low-angle sunlight generates shadows that model the features without being overly bright. When observing closer to full moon I always use a polarising filter to reduce the incredibly intense light, but since sunlight is then perpendicular there is little modelling to give relief to the geology.

Jupiter is by far the best planetary target for small telescopes; in addition to the cloud patterns you can see some or all of its four largest moons (Ganymede, Callisto, Europa and Io), their number and position changing on a nightly basis. Saturn is an excellent target too, the angle of the rings varying widely. I've also found Mars to be surprisingly worthwhile even when not at its closest to Earth, with the major features clearly visible in reasonable seeing conditions.

The problem with deep sky objects in urban astronomy is that they are both difficult to locate and their light is easily degraded by light pollution and particulates. I've attempted to get images of more familiar DSOs such as the Orion Nebula with several cameras, but the results are hopeless.

Once you have some experience under your belt, you may want to attempt photographing the International Space Station. Various websites list details for near-future visible passes over any location, when it is easy to spot due to being obviously brighter than any other man-made orbiting object. However, since the ISS will only be visible for around four minutes each pass you have to quickly manoeuver the telescope whilst keeping it in an area that is only about thirty arc seconds in diameter. If I manage to get any image at all, it is usually a few dozen frames resembling an out of focus capital 'H', so it's definitely a target for those with a lot of patience - and good hand-eye co-ordination.

4. Locating targets

Although I'm against beginners using go-to mounts, there are various planetarium programs and mobile apps that are extremely convenient for locating target objects. I use Stellarium, excellent freeware that can be set to any location on Earth and has a night time (i.e. red on black) mode to help keep your eyes sensitive to the dark.

Northern Hemisphere observers are at an advantage compared to their counterparts south of the equator due to the ease with which the North Celestial Pole can be found. Not only is Sigma Octantis slightly further from the SCP than Polaris is from the NCP, it is considerably dimmer. Therefore I've always had great difficulty in lining up the telescope to the South Celestial Pole for setting circles with the polar axis motor drive. There are telescope-camera combinations that allow use of auto guiding software but I prefer the manual approach to finding your way around the night sky. Besides which, spotting the closer planets is pretty easy, the most common potential mix-up being Mars with the red star Antares (whose name after all means 'equal to Mars')! All in all, manually slewing the telescope using a printed or online star chart as a guide is the best way to learn.

5. Harvesting ancient light

I tend to take 20-60 seconds of video or still sequences when imaging the moon and planets, depending on various factors such as target brightness and seeing conditions. Planetary cameras allow some manual adjustments such as exposure length and gain, with shorter exposure lengths usually better so as to minimise degradation within a single image. When the seeing is reasonable I stack the planetary camera on top of the 2.5x Barlows, which gives a decent angular size for the planets. I've also used a compact CCD camera with an eyepiece and Barlows combination, but the camera adaptor is fiddly to align on three axis with the eyepiece and the extra weight can mean regular adjustments to the mount, depending on telescope angle.

6. Image processing

Once you have the raw video or sequence of stills there is a lot that can be done to improve the image quality, initially by aligning and stacking the best individual frames and discarding the rest. Again, there is a lot of freeware available to help with this. I use RegiStax, often creating 3 or 4 permutations from each sequence and then loading the best one in Photoshop for final tweaks. (If you cannot afford the latter, then GIMP - GNU Image Manipulation Program - is a great freeware alternative.) It can take a while to understand how to use the likes of RegiStax, but there are YouTube tutorials covering various processes and I always consider a trial and error approach to be a good way to learn!

So what sort of results can you expect from all this effort? The biggest factor in quality is undoubtedly the seeing conditions, which are outside of your control. However, just occasionally you get a perfect night. I find that it can take a few sessions to generate a half-decent image, so it definitely takes perseverance.  Since a picture is worth a thousand words, you can judge the results for yourself here.