Book Review: Landscape Modelling

A little earlier in the week the copy of Landscape Modelling by Barry Norman I ordered arrived in the mail. Like a lot of books I've bought recently it has been on my list for a long time, and finally one came up for sale online at a decent price.

It's a clean copy, and there isn't any indication inside of who the previous owner was or where this book may have resided before coming to Canada. All I can say is it was published by Wild Swan Publications of England in 1986. If you're doing any comparative reading you might want to dig up your copies of John Olson's 1983 book, Building an HO Model Railroad with Personality, and Malcom Furlow's 1984 book. HO Narrow Gauge Railroad You Can Build. Both were published by Kalmbach in the US. Even though Olson's and Furlow's book cover all aspects of model railway construction, it's interesting to compare their scenery modelling techniques to Norman's given that all three books were all published in the same era. 

The entire book is an enchanting mixture of prototype photos, model photos,
and expertly drawn line illustrations.

I liked this book right from the get-go. Bob Barlow, the editor of Model Railway Journal, has this to say in the Foreword which sets the tone for the pages to come:

"However, there's far more to this slim volume [JDL: it's only 66 pages] than useful modeller's hints and wrinkles; there is a deeper dimension which speaks of success out of failure, of observation blending with a determination to get things right, and, above all, of love for the subject - in this case, our own British landscape. Looking around and loving what you see is a taller order than taking it for granted as most of us do, especially when those surroundings are conventionally 'ugly' like so many urban landscapes today. But somewhere, even in the grimmest of city wastes, beauty exists. And if you're out there studying the landscape closely enough - in order to build an accurate model, for example - you will assuredly find it. Not just once, but dozens of times: images that will linger and give inner pleasure for many, many years. I find it comforting that this hobby of ours - which most will concede is a fairly insignificant part of whatever great scheme exists - can lead us to great insights into our relationship with our own small planet."

Inside is a thoughtful account of how to go about creating a landscape for a model railway. It discusses both design and implementation aspects, and the author's layout, 'Petherick', provides examples. It's well written, clear, and the modelling techniques are effective and easy to understand. All the photos are black-and-white, but that hardly seems to matter given the excellence of the text.

This book is also great for just dipping into and browsing. Well, all Wild Swan publications are, but this one especially so for me because I'm a sucker for well done line drawings, and there are a lot in this book. The layout of the book is excellent too. Inspired page layout is a rarity these days, so it's even more impressive to me when I just stumble across it.

Out on the back porch with my 'tree box' getting a start on the community garden trees,
and looking through my just arrived copy of Landscape Modelling.

The book showed up at an interesting time. I had started work on making trees for the community garden layout, and I had just heard of David Hockney's passing, so I was a bit preoccupied with thinking about landscapes.

Hockney was a favourite artist. I never pay much attention to the passing of media figures, but this one hit home. As silly as it sounds, he seemed like an indomitable presence who would live forever. His work was varied, but I especially appreciated his thoughts that direct observation and wrestling with the problems of depiction were key things to deal with in art, as it should be for railway modellers. Here is Sam Hamper's tribute to Hockney where he goes into that and many other things about his life:

At around the 7:23 mark Hockney makes an interesting observation about trees in a discussion about perspective, well, maybe it's an obvious observation, but I thought it was good he said it out loud:

"Any cameraman will tell you it's not so easy to film the tallness of trees for instance because you have to look up to see the tallness. It's the tallness that would give you the majesty of the tree isn't it? The majestic nature."

Which is why I think it's important to model trees to their correct scale height and volume, otherwise it's conventional toy train lollipop trees all the way down that are completely without majesty. 

There are a lot of Hockney videos at YouTube, but I'll simply point out this short one of Hockney painting a tree dominated scene in Yorkshire:

The ways of the YouTube recommendation algorithm are a mystery to me. As I was looking for Hockney videos it recommended some from a channel called Crime Pays But Botany Doesn't ! The videos at that channel are serious, hilarious, and scathing, but in line with Barlow's comment that "... somewhere, even in the grimmest of city wastes, beauty exists" have a look at this poignant one of a tour of an abandoned Southern Pacific station in West Oakland, California (Warning: This is serious stuff):

 

A year in the life of a community garden tree

26 May 2025

15 July 2025

17 August 2025

18 August 2025

21 September 2025

14 December 2025

14 February 2026

16 February 2026

4 April 2026

23 April 2026

15 May 2026

18 May 2026

The community garden has a few trees scattered here and there. One is going to be represented on the layout; likely this one. It'll be planted by the re-charging spur for the electric loco, providing shade for the loco while it recharges.

AI and I have aesthetic differences

Picture of the modification to the Loonar Module micro-layout created by Gemini

Recently Vince and I had a discussion about an interesting question he posed to Gemini: can you critique the layout in this photo (or some similar phrasing)? The layout and photo being referred to is this one, the Loonar Module:

The real Loonar Module awaiting dusting.

I'm told Gemini felt there were too many trees, and when asked to create a picture of a revised layout produced the one shown in the lead photo to this post. Its suggestion stripped away the core design I had in mind for this layout and reduced it to something that reminded me of what our environment looked like here after the Victoria Day 2022 derecho swept across Ontario.

I didn't act on Gemini's suggestion and have no intention to. 
The Loonar Module awaits good weather so I can assess its dustiness outside in sunlight.

I mention this AI encounter because I think it highlights something important with the LLM AI variants that are freely available to consumers like me: they're only as good as their training, and they can't get out into the world to see and reflect on things as we can. I know these are obvious points that no doubt many others have noted, but for obvious reasons this example cuts close to my heart.

The 'prototype' of the Loonar Module landforms and scenery.

The Loonar Module had two main aims: to allow me to build a diorama of an area of the province I love, and allow me space to practice making trees based on species found here. 

Secondarily, the layout allowed me to explore some ideas in a somewhat humorous manner I'd been kicking around in my head about highly fictional Ontario narrow gauge railways. I tried to capture some early thoughts in a couple of episodes of the Streetcar Noir novella series back in 2015: Gone to the country and If a phone rings in the forest.

Same 'prototype', different view.

The micro-layout aspect came later. I wasn't tuned into micro-layouts when I started work on the project, and only realized when it was done that it classified as one because it was built on a stack of pre-cut foam slabs I purchased that happened to measure 2' x 2'. It was a completely accidental entry into the field.

'Prototype' of the little offshore island Gemini stripped the trees from.

Longtime readers may remember The Dictionary of Non-Existent Model Railroad Terms series I ran here from 2019 to 2024. Although the entries were presented in a somewhat tongue-in-cheek manner, the principles and terms presented were dead serious. 

The dictionary has a few entries dealing with how to capture the spirit and atmosphere of a location as part of a model railway's design work. The three I use often, and used on the Loonar Module, are: Full Observation, Three Ws, and Start with Geography. Those entries encapsulate my views on getting out and observing things, something AIs can't do.

'Prototype' loons: mom, dad, and junior.

AIs don't approach design in the way those dictionary entries recommend. It seems they look for patterns in example model railways they're trained on. Trees are sparse on many layouts, maybe the majority, so I'd guess and say if sparse is the dominant training pattern, then sparse is what gets recommended. 

In my opinion tree density is already far sparser than the 'prototype', even with 23 trees.

After the layout was built I discovered that by insisting on trees that were scaled to match their real life counterparts height-wise, they act as effective view blockers so that the entire layout can't be observed in a single glance. Basically the tall trees allowed the creation of four distinct scenes. This lesson regarding the use of height as a way to create scenes is one I've continued to use, and it seems an important way to make micros seem bigger.

A corner of the 'prototype' community garden.

Trees also figured prominently on the Centennial Experimental Farm layout, and will on the Community Garden micro. So, with the Loonar Module they form something of a trilogy, or maybe, strictly speaking, a tree-logy :-) I hope my AI friends are taking note.

Dusty Trees

Over the weekend a friend got in touch and asked if I was going to the Ontario Narrow Gauge Show in May, and if so, would I consider bringing the Loonar Module? It's around a 500 km and 5 hour or so drive from here, so it's a long trip, but that aside, the Loonar Module is looking pretty dusty these days, especially the central tall pines and lake. I decided to take it down to the workshop to try and clean it up and revive it over the next few days to see if it can be made presentable. As well, I'm not sure how it will stand up to a long trip. 

Ok, first things first, dust removal and 'sprucing' up, logistics later. Well, while I'm at work, I'll leave you with the Dusty you're likely thinking about, Dusty Springfield.

27 Trees

Layout cranked up to nearly the ceiling.

When I work on my little layouts I do so at a standing desk that I've put on wheels, equipped with a 2' x 4' hardboard top, and has a crank so I can move that surface up and down. It comes in quite handy when it's time to add scenery as I can crank the top surface up and down as well as turn the whole thing to various angles since it's on wheels. This allows me to get a variety of views while I'm adding things.

Over the last few weeks I've been adding trees to fill out the scenes. There're 27 in total. Funny thing is back in May I thought I had made too many trees! Turns out I was right and wrong. I had a few too many tall ones and not enough small ones, so I had to make some little trees to balance out the scenes.

The other side.

These days I'm adding rock debris and shrubbery. At one time I thought I would put in the effort to accurately replicate chaparral found in the region, but I've given up on that and decided to keep with the more simplified and somewhat abstract look I've started with the trees. I'm focusing on size, placement, and colour instead of fidelity to 'prototype'. 

Adding shrubs and other foliage is taking a surprisingly long time. Ok, I'm the problem because as usual my method is to add a bit, have a look, add a little more, take a break, look again, and so on - my standard operating procedure on this layout. So, you can see why it's a long process :-)

First tree

The first tree off the assembly line for the Mt. Lowe layout. Needs a bit of fluffing and trimming here and there, but it's usable. I think I made more armatures than I'll need on this layout, but no doubt I'll have a use for them in the future.

Work stoppage at the tree factory

Back in March I made a start on twisting up trunk armatures for the trees that will appear on the Mt. Lowe layout. Since then I've been following the procedure for finishing the trunks that I used on the Loonar Module.

Once I had coated the trunks with 2 layers of sand to give them texture I started in on painting and trimming branches.

As I was working I started to test out tree placement on the layout to see how things might look. I quickly realized that just about every tree is going to need some custom trimming to make it look like it actually grew up in the area where I planted it. 

That being the case I figured I should stop working on trees until I was ready to install them as it didn't make sense to build up generic trees and then wind up trimming off a lot of work to get each one to fit. 

So, I've stopped work on trees and started making uprights for the circular bridge. Once the bridge is done, the layout's basic scenic components will all be in place and I'll be ready for adding details such as trolley wire support poles, pterodactyl petroglyphs, and trees.

Tree factory

I need a lot of trees. Maybe 50. Lots for the Mt. Lowe layout. Some for that lumber shed diorama. A few to have around just in case. Right now I'm twisting up armatures for conifers. It'll take awhile to make 'em, but time's not an issue.

Replanting a mighty tree

Vicki's tree after planting in January

Back in January I finished and planted the large tree for the front of the Centennial Experimental Farm layout.

The 'prototype' tree outside Vicki's 

Paul noted the tree seemed too large and the bifurcation was too high off the ground. At the time I was obsessed with how big the tree was, so I didn't do anything. But, the comment kept rattling around in the back of my mind. 

As I was working on the Mt. Lowe layout I moved the Centennial Farm layout to a corner of the workshop to give me some extra elbow room. The farm was never out of sight, and I could see that tree in all its extra tall glory every time I worked on the Mt. Lowe track.

Drilling a deeper and wider hole for the tree

That constant visual reminder - nagging? :-) - convinced me that once I got the Mt. Lowe track functioning I'd reposition the tree.

And so I did. 

This change seems to have made a big difference in my opinion of the layout. It looks more balanced. Yeah, I still need to work on my technique for shaping and colouring bark; however, the overall scene is better.

Watch out for that tree!

I finished the big tree and planted it outside Vickis Veggies, near the street. I took a celebratory loop around the farm to honour the occasion. If there isn't already a sub-genre of model railroad videos that centre on layout trees, well, here's one :-)

To be honest I was having a lot of trouble with this tree, and at one point decided to get it finished and move on instead of continuing with the constant fiddling I was doing. I learned a lot of lessons, but there are still a few I'm working on: getting the trunk colour and texture right is a big one that still needs improvement.

I've got a few more big trees I'd like to model, but I'm going to leave those projects until the spring when I've got enough distance from this one.

Well, even though I have my complaints with this tree, I rather like how it appears on the layout. It has the commanding presence I was looking for and just seems to pull the layout together. With this and Insectary 75D finished, the major elements on the layout are done. I have a lot of detail items in mind, so they'll be next up. I think they'll take me to the spring to complete, so once the good weather has returned I'll be ready to take the layout outside for photographs.

I'm quite convinced now that if a little layout has tall elements it needs an equally thick base to create some overall visual balance. To my eye the layout would look a little odd to have tall trees, or tall buildings, and a thin base to support it all - it just wouldn't look right.

While I was working on the tree, the Deoralow was staring at me from a shelf where it's collecting dust awaiting a new location. I was struck by how small its trees are. They're store-bought ones from many years ago. They're pipsqueaks compared to Vickis, which is to scale. Well, for newly planted trees on a suburban lot, they're fine, but to populate the wild portions of a layout, they're way too small. Although my tree modelling still needs lots of work, I at least think I've come a long way in my appreciation of what layout trees need to be.

Well, speaking of appreciation, and since it's Saturday morning, here's the opening to a Saturday morning cartoon that, strangely enough, both my father and I appreciated.

Bark & Root

After a lot of frustrating trial and error I think I finally got some bark on the tree to be planted outside of Vickis Veggies that looks more-or-less like bark.

The trick seemed to be to coat the armature with generous dollops of Golden Molding Paste and use a fairly stiff, 1/2" wide paint brush to shape the glop into a bark-like mass. In earlier attempts I was far too stingy with the paste, and didn't have enough on the wire to shape or properly cover up the under structure.

I also lengthened the support wire that will root the tree in the layout. The short stubby thing wasn't going to properly support the massive tree above. I took 3 lengths of 18 gauge wire and then wire wrapped them to the original root, followed by a generous application of white glue to help hold the whole thing together.

So, now it's onto painting and foliage. The end is in sight!

Armature for Vickis tree

I worked up the branch structure on the tree to be planted between Vickis Veggies and the street. Now it's on to barking :-)

Trees at the Centennial Experimental Farm

Fits nicely on the living room coffee table. I see a trend :-)

The story so far

Back in the summer I started work on some tree armatures for the Centennial Experimental Farm layout. Throughout the summer I continued twisting up tree armatures and refining them while vegging on the back porch. Once I had 7, 3 large and 4 small, I decided to move on to applying bark.

That's when the trouble started.

Apprehensions

I tried some methods presented over at Boomer Dioramas and in Gordon Gravett's Modelling Trees for creating bark on the armatures. I was having trouble getting the hang of things. The trunks weren't looking that good. I realized I was in over my head and stopped. Stopped for a good long time. Months. 

I finally decided that even though I was having trouble learning how to coat the armature and model realistic looking bark, I should proceed. At the very least I needed the practice, and I wanted to continue on building the layout in accordance with the vision I had in mind. 

More importantly, from a learning standpoint, I wouldn't properly understand what to improve if I didn't complete the trees I currently was working on to the best of my current ability.

This is all to say that what follows is more a documentation of a learning experience than it is an expert tutorial on how to make trees. 

One tree

Across from Vickis Veggies is a model of a tree that's a few blocks down the street from my house. I've admired it for years, and wanted to have a go of at least making a simple representation.

Here it is sometime in the summer.

I should note that every tree I model I base on some real tree I've seen and photographed. As well as the tree on the left, I've used the two black walnut trees in my backyard as prototypes and a number of smaller trees that line the boulevards in my neighbourhood. I think it's important to use real trees as starting points instead of dreaming something up because the mind can be full of stereotypes.

This winter photograph gives a very clear idea of the trunk's structure.

You can see it is quite complex. Modelling it exactly as seen in HO scale would be a lengthy and daunting task, so simplification was in order.

The first thing I do is try and verbalize what the features are I like the most to help start the simplification process. In this case I particularly like the way the branch on the left arches over the road, how the trunk bifurcates near the ground, and the broad arc of the top of the leaf canopy. I figured if I could capture those aspects, and approximate the rest, I'd be happy.

There's the armature ready for the application of the leaf canopy. Let me point out a few things.

(Argh, I should have flipped the armature around before photographing so it was easier to compare to the real thing - sorry about that.)

The 'bark' consists of 4 coatings of Golden Coarse Modelling Paste. All I've managed to do fairly well is to get the trunk diameters about right. The texture and evenness of application need a lot of work.

The upper branch wires have been left extra long at this stage so I had enough material for securing the foliage. They were trimmed as I proceeded to shape the leaf canopy.

The bark is painted a mixture of Tamiya's XF-53 Neutral Grey and XF-49 Khaki. This mix was one suggested in Modelling Trees and I found it a good one for many of trees located here.

This is the tree partway through the process of adding volume to the canopy.

The fluffy, spidery structure is teased out Woodland Scenics Poly Fiber. Frankly, I haven't teased it out enough. I'd say it needs to be far less dense than I've applied. I had this problem with pine trees, and have still not overcome my reticence to have a far thinner, lacier canopy. It's some sort of mental block.

No glue is applied at this stage to hold the fibre in place. The wire branches hold it quite well by themselves. I trim the wires as I build out the canopy.

All done and ready for leaves.

I use both SuperLeaf and Noch's leaf products for texturing the canopy.

In this case I mixed some medium green leaves from SuperLeaf and Noch 50:50 to give a little subtle colour variation. It's indeed subtle, both products are very close in colour. 

I also sprinkled a little Noch dark green leaf material in areas inside the canopy to simulate shadow, and a little light green on the canopy's top to simulate dappled sunlight.

The first step is to spray on Mod Podge matte medium. It goes on white, but dries clear.

First, sprinkle a few dark leaves inside the canopy. You may need to turn the tree upside down to do this.

Then sprinkle on the medium green leaf mix, making sure no clumps fall on the canopy. 

Let dry.

A second application maybe required to even out the leaf cover.

The last application is a sprinkle of light green from relatively high above the tree.

When dry I usually need to do a final pruning to get rid of loose threads and clumps.

As you can see in the above photo, and in the others of the finished tree previously shown, the arched part is going to need to be reshaped a bit. I thought it was ok, but closer examination through taking photos highlights that work needs to be done to restore the arc I like.

Although my trees need work regarding bark and canopy density I am happy with the height, shape, basic branching structure, canopy volume, sky holes, and trunk diameter. I believe these are crucial to obtaining realistic trees, especially height as it more-or-less determines everything else. Scale-wise, real trees are much bigger than the typical items for sale at hobby shops.

Composition with trees

Before I started making tree armatures for this layout I had a plan in mind for types, sizes, number, and where they should be planted. I had to modify that when I finally had finished trees because their canopies took up more volume than I had anticipated. I modified my original arrangement to get the pleasing views I was after and still be able to gain access to the interior of the layout. 

It turned out I only planted 2 of the 4 small trees I made otherwise the layout would have been overcrowded with small trees. The small ones are planted in the picnic area between the two curved sections of track (I should note there was already a maple planted near them.)

1 and 2 mark where two additional tall trees are needed

However, after planting and looking at the layout and thinking for awhile I realized I need two more trees to balance out the composition, at least to my eye.

Outside the actual Vickis Veggies in Prince Edward County is a massive old tree that I'd like very much to try to model, and have it over hang the street outside the farm. It'll be planted at spot #1.

Spot #2? Well, I'm not sure yet what's going to go there other than it'll be tall and it's canopy needs to be above the streetcar wires.

I've started the armature for Vickis tree. It's a massive thing by itself, but not out-of-scale with respect to the other trees on the layout. I have some ideas for how to proceed with bark on this one, so I'll see how that goes.

Ok, one last thing. You might be wondering why a farm has so many trees, and big ones too. It turns out Ottawa's Central Experimental Farm, which is the inspiration for the layout, has an associated arboretum, The Dominion Arboretum. Frankly, I like streetcars, the farm, the arboretum, and Kato's Unitram track, so I squeezed them into one layout as I don't have the space, time, or money to build something more true to prototype. One thing I do hope that is true is the good feeling I get from visiting these places.