Friday, 13 April 2012

Climate Change? Shhh, Don't Tell Anyone. You'll Regret It

"Michael E. Mann is a member of the Pennsylvania State University faculty, holding joint positions in the Departments of Meteorology and Geosciences and the Earth and Environmental Systems Institute (EESI). He shared the Nobel Peace Prize in 2007 with other scientists who participated in the Intergovernmental Panel on Climate Change."

That's the editors note for an opinion feature by CNN on events that unfolded in relation to Mr Mann and about which I am going to write. The feature, dated 28th March can be found here. In fact, we can go further in looking at Mr Mann's credentials - he received undergraduate degrees in physics and applied mathematics from the University of California at Berkeley, a masters degree in physics from Yale University and a Ph.D. in geology and geophysics from Yale. He is the author of more than 140 peer-reviewed and edited publications.

His CV sounds impressive. He presents, by virtue of it, as someone who should be regarded as an expert in his field. Apparently, however, some disagree.

Mr Mann puts the scenario best: "Imagine you are sitting in your office simply doing your job and a nasty e-mail pops into your inbox accusing you of being a fraud. You go online and find that some bloggers have written virulent posts about you. That night, you're at home with your family watching the news and a talking head is lambasting you by name. Later, a powerful politician demands all your e-mails from your former employer."

This happened to him because he worked on Climate Change research in the late 90's and, when it became public, he was vilified, like many other science professionals, by what he describes as a "well-funded anti-science campaign". That resonates with me because this is not 1210 and we are not anticipating the passing of the Condemnations in Paris nor is it the Renaissance where we might expect to see Copernicus and, a century later, Galileo cowering under the might of the Catholic Church. In the 21st Century, science is practised by professionals interested in evidence. It is up to us how we interpret that evidence and how we do so is the most compelling demonstration of our objectivity there can be.

Mann calls it the "scientization of politics" [sic] - he's wrong about that, it's worse. It's the politicisation of science. Science is all about evidence - politics is all about the obfuscation of evidence. Politicians use evidence, or more correctly misuse, to bolster and underpin their (short-term) objectives. As Mann observes, commercial entities have a history of attacking science when it impacts their profits and, in a world when commercial entities have more wealth and power than some nation states, that should worry us greatly. In February 2012, the Guardian reported on how the Heartland Institute, a think tank retained in the past by Microsoft, GM and Exxon, had documents (it later said) stolen or forged - although its explanation actually revolved around employee mistake rather than actual theft but theft is so much more inflammatory and suggestive of illegal and dishonest conduct by those Heartland would appear to be seeking to discredit. Those documents allegedly demonstrated a clear policy to "discredit climate change". The scheme included, said the Guardian in a later article, spending $100,000 for spreading the message in selected schools that "the topic of climate change is controversial and uncertain - two key points that are effective at dissuading teachers from teaching science". Heartland issued a statement saying the documents were fake. We should be concerned to find the truth in this disparity.

The Guardian goes on to say "the Heartland Institute, founded in 1984, has built a reputation over the years for providing a forum for climate change sceptics. But it is especially known for hosting a series of lavish conferences of climate science doubters at expensive hotels in New York's Times Square as well as in Washington DC. Some might call that an incentive which might impact on credibility.

The law requires a high standard of proof before a case can be decided and it might be said that no one could sensibly suggest human beings adopt that cumbersome standard in their daily decision-making processes. But I have heard, over the years, many defence advocates urge juries to adopt the same thought processes when considering their verdict in a case as they would in the most important decisions in their day to day affairs. We all have important decisions to make and we all consider the options before making them. Evidence is a daily part of our lives and we judge evidence based on its credibility and weight - we just don't realise we're doing it. In fact, this is what any Judge will tell a jury to do when coming to a verdict in a criminal case.

The internet has been a game changer in every part of modern, civilised human society. Yet, when it comes to the internet, we know little about the credibility of the entity (because it is not always a single person, often it's an organisation) making the statement we are considering, nor do we know what weight to give that statement because we don't know where it really comes from. That's the danger of the internet - too many people accept evidence, the provenance of which they know nothing about. Who is delivering what you are reading? How are they doing it?  To what end? What basis in fact or experience do they have to attempt to influence you? And then it becomes viral - a wonderful new word which describes succinctly and eloquently the process by which (dis)information spreads like a disease.

Mann goes on:

"In the most infamous episode, somebody stole thousands of e-mails and documents from leading climate researchers, including me. They cherry picked key phrases from the e-mails and published them out of context, like a black-and-white political attack ad with ominous music. Fossil fuel industry-funded groups gleefully spread the e-mails online and badgered the mainstream media into covering the "controversy" they had manufactured. It was no accident that this happened on the eve of a major international climate change meeting. Saudi Arabia, the world's largest exporter of oil, was the first to call for an investigation.The dozen independent investigations that did follow -- all of which exonerated the scientists -- got much less media coverage than the original nonscandal. Last year, the inspector general of the National Science Foundation found the charges against me were all baseless and reaffirmed mainstream climate science."

Who do you believe? What evidence is the most compelling to you? Did anyone investigate Heartland in the same way or were they protected from investigation?

And as Mann rightly says, this is a dangerous way to have a climate change debate. His new book depressingly aptly entitled "The Hockey Stick and the Climate Wars: Dispatches From the Front Lines" is on precisely this point. We are at war with short-termists - for example, those who devastate rain forests for commercial gain knowing that trees produce the oxygen we, and every other living thing on this planet on which we depend, require to exist; those who produce cigarettes knowing they are addictive and cause lung cancer. I name only two but of course there are many more examples. 

Commercial entities are run by human beings, and human beings are motivated by self-interest. We are failing to deal with the issue because it is in the interests of so many to muddy the waters. We live in a world dominated by short-termism. Politics runs in cycles of four years - the question on the lips of those wielding power is 'will we get re-elected' not 'what can we do to change things'? In the UK, the trustworthiness of those we have elected was so starkly called into question in the last few years following the expenses scandal. Then, ambiguous rules were blamed - none of them realised that they should not have needed to have been told that what they were doing was fundamentally dishonest. How then, can you expect to find persuasive the words of individuals who have been so utterly dishonest in the past? Commercial entities have too much power for their undemocratic, unelected status and necessarily skewed priorities. Their impartiality is non-existent by definition.

What evidence you accept in your search for the truth demonstrates in the most clear terms your objectivity. A favourite question of a lawyer from a layperson is 'how can you defend someone you know is guilty?' We've all had it. The answer is, I suspect, the generally same for all honest legal professionals - who am I to determine guilt or innocence? That's not my role - it's yours. My role is to make sure the system works and the evidence is tested. The prosecution bring the case, the defence try to knock it over. When that process is finished, the jury have seen the evidence placed under the most exacting microscope and all its flaws and blemishes exposed. Every nuance and inference is considered, stretched and abused and only then are they able to determine whether there is enough evidence, and it is compelling enough, for them to be satisfied so that they are sure of a person's guilt. It's not about truth, it's about whether there is enough evidence, which is credible and persuasive, to prove the allegation. Or to quote Aaron Sorkin's character Daniel Kaffey in 'A Few Good Men': "It doesn't matter what I believe, it only matters what I can prove." Over the years, I've found no better précis of my role.

Unless we are able to evaluate the evidence before us by judging those giving it to us and assessing their creditability, we cannot attach to it the weight it deserves. We cannot deal with the real issues, there can be no real debate. We cannot answer the fundamental question - who is telling the truth? We are the ones needing to drive the debate and we need to make a statement about who we believe. Juries are asked to do exactly that each and every day up and down the country when they say "guilty" or "not guilty" but the evidence they are exposed to is regulated by rules that have largely been in place for hundreds of years, adapting as society changes, and analysed, after properly delivered forensic legal argument, by the greatest minds in legal history. What juries are exposed to is the 'best evidence' - the fairest evidence. 

The internet has no such sanctions, no such protections. Be careful who you listen to and be careful what you accept. These issues need to be decided and we need to decide them.

Saturday, 7 April 2012

On Exposure Part II: Aperture

While ISO certainly has an impact on a perfect exposure, for what Bryan Peterson called, is his book Understanding Exposure, a ‘creatively correct exposure’ aperture and shutter speed are the more important corners of the Exposure Triangle. A 'creatively correct exposure' is, in short, a 'perfect' exposure which follows a particular creative avenue chosen by you to inspire, in your viewer, a desired emotional reaction. You are attempting to recreate the drama which inspired you to make the exposure in the first place.

You will recall from my previous post - On Exposure Part I - that a 'perfect' exposure is one where the shutter speed and aperture, along with ISO, combine to make an exposure where the light meter sits on '0' or is 1/3 to 2/3 of a stop under (-) or over (+) exposed. An exposure made as a consequence of one 'perfect' aperture/shutter speed combination will be very different from another, also 'perfect', aperture/shutter speed combination. Which combination you select will depend on what feeling you are trying to inspire in the viewer and what type of image you're trying to create. In order to understand how to select an aperture setting, you need to understand what aperture is and what effects it creates.

What is Aperture?
Aperture values, expressed as ‘f-stops’, relate to the diameter of the opening through which light hits the camera’s sensor. The lower the numerical value, the wider the opening and vise versa. And, as we saw in Part I, the wider the opening, the more light can fall on the sensor.

Traditionally, aperture values or f-stops, are as follows, which reflect the diameter of the hole halving each time:

f/1.0, f/1.4, f/2.0, f/2.8, f/4, f/5.6, f/8, f/11, f/16, f/22 and f/32.

That said, most DSLR cameras have aperture settings in between these - my Canon 60D for example has among others F/9, f/10 and f/13, all of which can be quite useful. The lower and higher ends of the aperture scale are a function of the lens you have selected subject to your DSLR itself. High quality ‘fast’ lenses (those with f/2.8 or lower) are usually very expensive.

Depth of Field
The principal creative reason for changing aperture is to influence depth of field. A lower f-stop value (a wider opening) generates a shallow depth of field. The subject in focus will be sharp, but the background will be blurred. In fact, and more specifically, on my Canon 24-105 f/4 L-series lens, f/4 produces a very shallow depth of field indeed. In an image of my partner holding my son, at 100mm focal length, my son is in perfect focus but even my partner, inches away from him, is slightly blurred. 

Which leads me onto the other point about utilising a low f-stop in order to create what Peterson terms so vividly an ‘isolation’ shot - the longer the focal length selected (the higher the level of zoom used), the more blurred the background will be. Use a 300mm lens and zoom in to the fullest extent and the background beyond your subject may in fact be so indistinct as to be almost a single texture with just some limited separation in colour. 

For a variety of different 'textures', take a look at this exposure I recently took for a magazine feature on wild camping (f/5.6, 65mm) - the subject was the stove, but I wanted the far background very blurred but the near foreground and background also blurred but not to the same extent as I wanted to show the presence of the grass but not focus on it:

A higher f-stop value (also known as ‘stopping down’) produces significant depth of field. Not only will the subject be in focus but the foreground and background as well. This sort of aperture setting is commonly used in landscape shots. There are issues about retaining perfect sharpness throughout an exposure which revolve around calculating hyperfocal distance and where in the scene you should focus but I’ll deal with those in the next post on 
Depth of Field and Hyperfocal Distance Focusing. For the exposure below I used f/22, 17mm (and a 3 stop ND Grad filter - but more on that in another post) and focused on the second row of trees to the left of the exposure.

As we know, a high f-stop value, or narrow opening, requires a lower shutter speed to achieve a perfect exposure. Given many landscape shots are best taken, for the ‘creatively correct exposure’, in low-light conditions such as dawn or dusk, almost all landscape photographers need to use a tripod and remote shutter release.

A note of caution on high f-stop values - whilst f/16 and f/22 may produce prodigious depth of field, the exposures they generate also suffer, to varying degrees, from diffraction. This effect, a consequence of light being forced through a narrow opening, means the sharpness of the image begins to become compromised. The optimum sharpness on most lenses is two or three stops down (i.e. higher in numerical value) from the lowest aperture value on the lens. Usually, this is f/8. Because of this, many landscape photographers will select f/8 or f/11, or possibly f/13, for their landscape exposures. Whether diffraction is an issue for you will depend on how sharp you want your exposures to be. However, as long as you don't want subject in the immediate foreground to be sharply in focus, say within 3-4m of you, then f/8 or f/9 will have sufficient depth of field and produce sharp exposures. I'll tell you why this is in the next post.

For example for this exposure at RHS Wisley I wanted everything in focus but the foreground interest was 6m away. f/8 was fine at 29mm.

And for example, for this exposure, I wanted the tree branches to be sharp as well as the sky. The branches were around 8m away from me, so f/9 was perfect.

So, making exposures of the same scene, with different aperture settings, produces wildly different images. 

Next: Depth of Field and Hyperfocal Distance Focusing

The two books I recommend to learn more about Exposure are listed below. These are Amazon Associate affiliate links which means if you buy them here, I get some income from that purchase. I own both these books and endorse them completely. Peterson's book is a legendary starting point for beginner DSLR users. Freeman's book is hugely technical but tremendously detailed.

Tuesday, 3 April 2012

On Exposure Part I: The Photographic (or Exposure) Triangle

This first part of a four part series dealing with the fundamental principles of understanding exposure is both an introduction to the concept of exposure and to the foundation for understanding it - the photographic triangle. It also deals with the first element of that triangle - ISO.

The series is intended to set out, for beginners and in an understandable and accessible form, the essence of what I’ve learned as I have turned from the smaller cameras I normally take into the mountains to a digital SLR camera system. Of course, the principles are applicable to more than just the modern DSLR but that is what I am currently using so that will be my frame of reference. Note: I always use the Manual (M) setting on my DLSR which means each element of the Photographic (or Exposure) Triangle is set by me (rather than by the camera) for each and every exposure.

Exposure is the fulcrum of photography. As a term, it describes a number of linked concepts. At its most basic, it is used interchangeably with ‘image’ or ‘photograph’. Photographers refer to ‘making an exposure’. In reality, however, the most important use of the term exposure is to define the amount of light allowed to fall on film or a digital sensor.

Three main elements combine when determining the exposure of a particular photograph - ISO, Shutter Speed and Aperture. Taken together, they are elements of the Photographic (or Exposure) Triangle.

In crude summary, each setting controls exposure differently: Aperture controls the amount of light permitted to fall on the sensor; Shutter speed controls the amount of time the light is permitted to fall on the sensor and ISO is a measure of the sensitivity of your camera's sensor to a given amount of light.

Each of these three elements is affected by the choices made by the other two. In low-light conditions, for example, allowing as much as light as possible to fall on the sensor (aperture), for as long as possible (shutter speed), and making the sensor as sensitive to that light as possible (ISO), seems logical. Yet there are tradeoffs and it is not as simple as it might first seem from that statement.

Higher ISO settings can lead to grainy images - once used creatively in film photography, it is not quite so effective when it comes to digital sensors. A slower shutter speed, that is to say, having the shutter open for longer, could lead to blurring of the whole image if the camera moves at all during that time. Selecting a fully open aperture, to allow as much light in as possible, drastically reduces depth of field (the area in the image which is in focus and, hopefully, sharp) which means not everything in your exposure will be sharply focused. Any or all of these effects may be desired by you for creative reasons. Some, may not be.

By way of example (and don't worry if this is a bit technical right now, it will become clearer later and in subsequent posts) the image below, handheld and indoors, was at ISO 500, aperture f/4, shutter speed 1/15, focal length 50mm. It shows how aperture can isolate your subject (by blurring foreground and background) and how, even at ISO 500, images can look good. As it was handheld, and shutter speed was 1/15, it is not pin sharp but pretty close.

Understanding the Photographic Triangle means understanding why it is also called the ‘Exposure’ Triangle. Each element has a crucial role in determining whether you have captured a perfect exposure once the shutter is released. It is not all about light or dark either. A DLSR has a light meter which can be seen through the viewfinder. On the manual setting, that light meter will indicate whether the exposure is ‘under exposed’ (-) or ‘over exposed’ (+). It does this 
by virtue of a tracking spot which moves along the line towards the + and - sides as you alter whatever the camera is pointed at, and/or when you change aperture, ISO and shutter speed. Your aim, at the moment, is to ensure that your selection of the settings for the elements from the Exposure Triangle means that the tracking spot is bang on the middle (or 'perfect'), or within 1/3 or even 2/3 of a stop, of ‘perfect’. You might think that the image above is 1/3 underexposed. I probably agree with you.

All three elements, when used in concert, will create for you exposures which capture the emotion of the scene which moved you pull your camera out in the first place. However, in order to capture that scene, you need to understand what made you want to capture it in the first place and how each setting impacts on the final exposure. Changing one setting will require changes in at least one of the other settings to achieve a perfect exposure but each setting, on its own, has a creative influence.

For instance, a narrow aperture opening will mean much greater depth of field. A wider aperture opening will mean shallow depth of field. If you are intending on highlighting a particular subject by blurring the background beyond it, then an aperture setting of f/4 or lower would normally be required. This could potentially mean a high shutter speed. If you want everything to be sharp in a landscape image, then f/11 or higher might be required. This will necessitate a lower shutter speed which may preclude handholding. If you want the wispy, ghostly effect created by silky smooth waterfall images, a very low shutter speed is required. This will mean a consequential impact on aperture and depth of field. 

For the image below, again, I wanted to blur out the background and only have my son sharply in focus - so low aperture. I needed a high shutter speed to freeze the action as he was on a swing. I kept ISO at 500 even though there was plenty of light to get that high shutter speed. Again, this exposure was handheld. ISO 500, aperture f/4, focal length 96mm, shutter speed was 1/2500.

Aperture and Shutter Speed are discussed in more detail later, but it may be that you need to go outside the triangle to get the effect you want by deploying filters, for example.

ISO was first used in film photography and is an indication of how sensitive the particular film concerned was to light. It is measured through a numerical value and the lower the value the less sensitive the film and, importantly, the finer the grain in the exposures.

In digital photography, ISO reflects the sensitivity of the camera’s sensor and, again, the lower the value the less sensitive the sensor is (at that setting) and, importantly, the finer the grain in the exposures.

The baseline for exposures with very little grain, or ‘noise’, is 100 ISO. My Canon 60D does not have a setting below 100 ISO, but some high-end DSLR cameras do. A higher ISO, which makes the sensor more sensitive to light, will result in dramatically more grain or noise. As a result, you will usually only increase ISO from 100 if your desired aperture and shutter speed aren't otherwise obtainable. That said, a DLSR sensor, in contrast to that of a compact camera, will respond far more positively to that problem and even at ISO 500 or more, there will be little in the way of perceivable noise.

As I have said, your choice of ISO will impact on the aperture and shutter speed required for a perfect exposure. Increasing your ISO from 100 to 400 or 500 would mean a perfect exposure would require a faster shutter speed and narrower aperture. This might make handholding possible in low-light, for example. So, how do you choose ISO?

You must, ultimately, assess the light - do you need a high shutter speed for your exposure? If so, set the ISO as high as you are able so long as there is not too much noise. Indoor exposures will often require a higher ISO. Are you using a tripod? Shutter speed then becomes less of an issue so a lower ISO is possible. Is your subject stationary or moving? Are you attempting to achieve a particular effect as a consequence of high or low shutter speeds such as panning (where the subject is moving, but is kept sharp by moving the camera with it - this has the effect of blurring the background into movement lines and implying motion by the subject).

Next: Aperture.

The two books I recommend to learn more about Exposure are listed below. These are Amazon Associate affiliate links which means if you buy them here, I get some income from that purchase. I own both these books and endorse them completely. Peterson's book is a legendary starting point for beginner DSLR users. Freeman's book is hugely technical but tremendously detailed.