I have a confession to make… 

… Most of the clouds in my paintings weren't actually there.

Well, that's not entirely true. They were there, they did exist, just probably not above that particular hill, on that particular day, at that particular moment.

I have a sketchbook filled with drawings of clouds. I photograph them too, I've built up a whole library of them over the years. I have a photo folder full of clouds which I store in the cloud. Hundreds of images featuring great looming structures, little broken patches, stretched out strands that pull your eye across the sky. Then, when I'm in the final stages of marking a landscape up, I'll often rummage through my cloud folder or my cloud sketch book looking for the right one.

Many times the sky I actually walked beneath doesn't work with the painting. Perhaps it would pulls your eye in the wrong direction, makes the composition feel too heavy, doesn't carry the mood I remember from being there. So, I change it.

And now I’ve admitted to that, I suppose there are a few other things I should own up to…

… I’ve removed trees,

I've added boats,

I've made trains arrive at times when no train actually passed,

I've moved waterlilies along a river,

I've tidied up the mud,

I've even completely transformed a tree because the real one, however faithfully I tried to paint it, somehow made the whole painting feel wrong.

All this might all sound rather strange for a project built around walking to real places, sketching them and then making artwork from the experience. But I've never really thought of the paintings as records. They're memories. And memory, as I've explored in previous blogs, isn't particularly interested in accuracy.

When I painted Chee Dale, for example, I added waterlilies to a section of the river where they weren't actually growing. They were abundant elsewhere along the walk and had made such an impression on me that leaving them out felt stranger than putting them in. On the original pencil sketch you can just about see the hesitation, the experiment of placing those shapes on the page. 

At Lud's Church, things became a little cleaner and drier than they usually are in reality. Anyone who’s walked through there knows mud is hardly an unusual feature. But when I think of Lud's Church, mud isn't what comes back to me. I remember those enormous cavernous walls, the foliage and greenery, and that peculiar feeling of being enclosed by something ancient. I hope that's what grew in the painting.

Perhaps I'm idealising these places? Actually, I'm fairly certain I am. Some of the artists who have influenced me most have done something similar in their own ways. I've always been attracted to landscapes in which nature has been interpreted rather than simply recorded. The landscape might remain recognisable, but something of the artist has entered it too. I think that's inevitably happened to mine.

But that doesn't mean I remove everything imperfect.

On one walk through Lyme Park, I came across a fallen tree that had crashed through a dry stone wall. When I eventually made the artwork I took the tree out but, the damage to the wall remained. The great hole where the wall had collapsed now hints that something happened there, although the viewer doesn't know what. More importantly, it did something wonderful to the composition. You can almost step through the gap, like a doorway from the foreground into the landscape beyond.

I've had similar battles with fences which can be awkward things to paint. Include every single post and the artwork quite quickly becomes chaotic, so decisions have to be made. In my Shutlingsloe painting, it was the crookedness of those manmade structures that eventually became useful. I didn't need to describe every post, just choose the ones that would allow the viewer to understand the rest.

After all, who's counting the fence posts?

Pen y Gogarth gave me an even bigger problem in the town of Llandudno.

I'd never really painted a town into one of these landscapes before but looking down from the headland there was absolutely no pretending it wasn't there. For a while I genuinely didn't know what to do with it. Eventually I stopped trying to paint every building and instead picked out particular roofs and windows, finding the larger shapes and using blocks or overlays of colour to suggest everything in between. Just enough of the town remained to suggest the whole.

Recently, though, I've started wondering about the things I instinctively leave out. On Kinder Scout, vital restoration work has placed dams into eroded gullies to slow the water and help the damaged peatland recover. Some are conspicuously manmade things, like corrugated metal sheets, the sort of ‘ugly’ detail I might automatically remove from a painting of somewhere wild.

But should I? Those unsightly interruptions are part of helping the landscape heal. Take them away and perhaps I'm not just making the painting prettier, I'm removing part of what's actually happening there.

Even the great flagstone paths crossing parts of the moor present the same contradiction: something unmistakably manmade that divides opinion; some people despise them and some people see them as part of Kinder Scout's identity.

Maybe I've been asking the wrong question. I don't only paint the beautiful bits. I paint the bits that matter.

Sometimes that means adding waterlilies that were further down the river, or finding the right clouds from another day. Sometimes it means removing a fallen tree but keeping the broken wall it left behind. The painting isn't asking what deserves to stay because it's beautiful, but what deserves to stay because it says something about the place, the walk or my memory of being there.

Knowing what those strange dams on Kinder are doing changes how I see them. Would I paint them? I still don't know. But I think I'd find it much harder to simply paint them away.

I'm not trying to record every detail of a landscape, and I'm certainly not going to stop rearranging the clouds.

But maybe the beautiful bits aren't the only ones worth keeping.