mikegreenimages

Musings on: using technology to pre-visualise images

By Mike Green on 27 May, 2011

I am now, or so I’m told by certain non-photographer friends, not only ‘fiddling with images’ in post-processing, but also ‘cheating’ by pre-visualising them with the aid of technology. Shocking! I refuted, or made a serious attempt to refute, the former charge in a previous post on the ethics of digital manipulation. I shall now refute the latter.

The charge goes something like the following:

“Using tools such as Google Earth and The Photographer’s Ephemeris (TPE) is cheating; you should just find places to photograph by wandering around.”

That is, of course, a paraphrase of what a non-photographer would say, but it’s what I’ve done for a couple of images, making it a reasonable one. In fact, I’ve been using TPE for well over a year now, it’s invaluable in working out where the Sun and Moon will be at a given point in time, at a specific location, and in determining both whether they will be visible and whether they’ll be lighting the landscape in the way that I would like them to. TPE, however, presupposes that I know where I’m going to be, whereas Google Earth, and in particular the relatively new ‘ground level view’ introduced in Google Earth 6, enables me to get a remarkably good idea of what things will look like when I’m standing at my chosen location. By combining the two it’s possible to go a long way to composing an image without even leaving the house. I don’t think that’s ‘cheating’, as such; I’d characterise it as taking advantage of technology to get a good result when it’s not practical to scout a location on foot and in a variety of lighting conditions

A real-world example using Google Earth ‘ground level view’

By way of example, I used both tools in creating the image below, which is of a valley in the centre of the Howgill Fells, in Cumbria, England, an area I’ve described in a post on its photographic potential. I’d noticed the interesting, interlocking spurs of hills on a previous walk in the area, but didn’t have a camera with me, and in any case the Sun was high in a clear, blue sky – less than ideal (and that’s why I didn’t have a camera…..). At the time, I just noted the location and then later, at home, investigated what could be done with it. It’s a non-trivial drive to the Howgills, followed by a walk-in of a few kilometres horizontally and about half a kilometre upwards, so I wasn’t keen on going up there at random and hoping it was worthwhile. To be honest, I probably would have done, but I felt much more confident that I was not about to waste my time, having planned it in some detail in advance.

The first thing I did was find the area in Google Earth, then I zoomed down to ground level, at which point the view shifts to the cunningly-named ‘ground level view’. This isn’t a tutorial on how to use the software, so suffice it to say that you can move around as if you’re walking and that the view you’d see is represented topographically. From my two ‘serious’ uses of this excellent facility, I can say that it’s sufficiently accurate to plan from – at least for the areas I’ve looked at. The following images are, on the left, a screen shot from the precise location I eventually stood to take the photo and, on the right, the image itself. The photo is zoomed a fair bit, so it’s of the top half of the Google Earth representation. No, they’re not identical, but they’re remarkably close if you look at the degree of overlap of the spurs and the shape of the river in the valley. Certainly, they’re close enough that relying on the software to aid visualisation saved me a good deal of hunting around on awkward terrain for a point to set up the tripod.

To show how close the representation is, I’ve deliberately done this backwards for the sake of this article, using the location data from the photograph to return to Google Earth and make the above screen shot. The on-line investigation I’d done in advance enabled me to mark a point which looked promising, and then drive / walk to it and be within fifty metres of where I ended up. More importantly, it let me play with compositions in advance. I was standing on what might be considered a steep slope which I had, of necessity, approached from above, and in inclement weather. Since I’d already determined, by ‘walking’ down the slope in Google Earth, the lowest level which improved the shot, I simply descended to that contour and then traversed the hillside until I found the composition I’d visualised on-screen as well as, more conventionally, in my head. Running up and down the hillside to see whether the composition would work better from lower down was something I was thoroughly happy to forego!

TPE

And what of TPE? Well, what I really wanted was sunlight on the right hand slopes, which means ‘sometime in the morning’ (the valley runs roughly north-south). More precisely, I wanted illumination, but not direct sunshine, so I’d used TPE to determine when the Sun would be low enough to not create harsh shadows anywhere – TPE showed me that this meant that I needed to be there within half an hour of sunrise. Unfortunately, the only place to park – the only sensible place to park anyway – is by an isolated house. Doing so at least an hour before dawn at any time of year might be considered anti-social; in late May, it really wasn’t an option. I settled for ‘any day with grey, high cloud’ instead. As it turned out the cloud was somewhat thicker than I wanted and it was decidedly dark in the valley, so the only thing which was still lacking precision was, as usual, the weather forecast! Oh, and it was raining and very windy too – a couple more things the weather forecast had assured me wouldn’t be the case…

TPE helped me work out what was best, I merely wasn’t able to follow its guidance on this occasion, though having now seen the way the various spurs in that valley lie, I’m sure the early option would be the best in terms of the end result – maybe I’ll return in winter.

That, then, is the combination of techniques, but I’ve not yet actually refuted the argument that using them is in some way ‘wrong’; all I’ve done is say that this sort of planning is very practical and effective, at least for ‘big vista’ type compositions where the overall shape of the land is important – Google has not as yet recorded sufficient detail to enable anyone to decide in advance which trees to include in an image – give it time though.

Conclusion

Would I feel in some way better about the image if I’d not worked out roughly where to stand before I got there? I think not. I might have lost some time in searching around, up and down the slope which, as I’ve mentioned, is verging on being very steep. I might even not have found the spot, or left insufficient time to find it and found the valley even darker than it was when I arrived. If either of those things were true, then I’d possibly have felt some aching of the legs afterwards too. More probably, not being confident of how good the composition would be in advance, I’d not even have left home at six in the evening, aiming for a vaguely-defined point two hours travel away, so I’d not actually have made the capture at all.

Yes, perhaps the process is marginally less Romantic than wandering the fells hopefully in search of surprise compositions, but I’d emphasise the ‘marginal’ aspect quite strongly. It was raining, cold, and windy up there, and getting dark; had I not been near-certain of a good composition, I suspect that I would have turned back. This refutation is perhaps a largely pragmatic one, but I think it’s also convincing. I’m certainly not advocating doing this for every image; that simply wouldn’t be possible. Using the technique for certain types of composition, however, at least as a means of getting an idea of whether it might ‘work’ and where to start, seems to me to be a very helpful addition to the set of methods for pre-visualising compositions.

I’d be interested in hearing your views, especially if you disapprove of using such technological techniques for pre-composition, or pre-visualisation.

Addendum

Since I first published this article, Stephen Trainor, who wrote TPE, and Bruce Percy have jointly published an e-book in which Stephen describes all the facets of TPE and Bruce relates this to how he uses it in his work, in combination with Google Earth. i.e. exactly the same topic as this article, but with rather more ‘how to’ detail! I recommend it if you’re interested in taking this approach to planning images. ‘Understanding Light with The Photographer’s Ephemeris’