Case Study 11 – Fairway Bunker Design – AppsinCadd

This Case study was written by Dr John Strodachs.

Following on from the previous Case Studies on a local golf course, this is a small exercise but has some interesting challenges. It will finish with setting out and 3D visuals, including the placement of tree objects rather than the spikes seen here. Some of the data used in earlier case studies will be used here, including LiDAR and georeferenced images.

The 17^th fairway runs parallel to the practice ground, which was covered in a previous case study, as shown here with new driving and chipping areas.

Longcliffe Gold Club - Updating Practice Ground

A new bunker on the 17^th fairway has been proposed, being positioned 253yds from and visible from the yellow tee box, strategically to the left of the fairway with new heather growth planned to the right requiring good course management.

A design was carried out by a golf course architect using digital data created in ⁿ4ce and supplied in AutoCAD DWG file format. The resulting bunker design was supplied in PDF file format.

We already have 1m LiDAR grid models, one showing the heights of trees and vegetation and the other filtered. The height difference of these models can give an approx. tree height used to scale tree 3D objects shown below. The Unfiltered LiDAR file represents trees as columns.

The proposed bunker was positioned offset from a tree and level with the 14^th green, as shown below.

The Golf Course

The shape of the bunker was sketched out by the architect, using contour lines to provide elevation detail. We will need a true-to-scale detail before carrying on.

A paper copy of the sketch was printed, 2.5m grid lines were drawn based on the scale bar shown above, and scanned back to a PDF file before importing it into ⁿ4ce as17^th Fairway bunker – cropped.

The bottom left grid origin was identified with coordinates (0,0) and the top right (10,10). This provides control that can be used in a coordinate transformation.

Open a new Model folder, calling it Digitise, and then go into graphics. Use the Properties panel on the left and in Backcloths, select the JPG image called 17^th Fairway bunker – cropped from the left window moving it to the right Selected window.

After zooming to extents, this image now appears on the screen as a backcloth. Move along the top menu bar until you see the Tools menu, from which you can select the Transformation Options under Backcloth Images. Select Transform.

Two sample points from the image, whose coordinates are known, are now selected. These are used to scale and orientate the image, adopting the same coordinate system as the sampled points.

Zoom bottom left when prompted and select the (0,0) point, followed by the top right (10,10). Right-clicking brings up a dialog box requesting you to enter the known coordinates for the selected points, as shown below. Enter the two pairs of coordinates, followed by OK to complete the transformation.

If Tooltips is set in the Properties panel, moving your cursor around the transformed image, you will see the size and coordinates are now based on the sampled points. The image is true to scale (1:1).

The next step is to digitise the shape of the new bunker with coded points and heights. From these, a 3D surface (DTM) can be created, which will be used to generate its own contours, sections and volume quantities.

We can deduce that contours were sketched out at 0.25m intervals. Unfortunately, the golf course architect overlapped some of the contours, which is technically incorrect, so a little artistic license was adopted here when digitising these lines. Using the Points -> insert menu option and Sketch -> Constant, each contour was given a CT code and appropriate contour height. Simply follow each contour to form a feature string. The two high points were picked up with an SP code.

The digitised contours are correct to scale and have their own coordinate system. This is fine for creating a DTM using DTM -> Create -> Normal.

Use Groups -> Settings to define two new groups, with Sand being yellow and Banking green.

The Group -> Add options were used to isolate these different parts of the model, as shown below. This allows us to generate individual quantities.

From the DTM menu, select Volumes by Prisms and set the reference surface to zero Datum Height, as shown below.

We are using a zero-datum height here to show the amount of material above and below the ground. There is 23m³ forming the banking behind the sand trap, but since the Sand was set to zero, it does not appear in this table.

If we measure the surface area of the sand, using Query -> Line, this comes out to be 44m². If we take out 0.5m depth of material from the sand surface to provide for the foundation, drainage and sand base, then we see that the volume removed comes to 22m³, a perfect balancing out of cut and fill!

Setting Out and Visualisations

We now move on to the visualisation of the new bunker. The first port of call is the 3D viewer, as identified by the yellow cube in the Home menu bar. There are various controls available here found in the Properties panel. 3D movement of the model is via mouse buttons and roller ball.

This model is not part of the overall scene, having a different coordinate system to the golf course.

LiDAR grid models and georeferenced aerial photography are available for the golf course. We now need to add the designed bunker to the 17^th fairway at the right distance from the tee and with correct levels.

The bunker is to be placed approximately. 253 yards from the yellow tee. A line needs to be drawn from the yellow tee 253 yards along the fairway to give reference for its placing. We will use the Dedicated CAD Backcloth here. Click on the single Pencil icon to take you there.

In Backcloths from the Properties Panel, add the georeferenced image called 2019 Aerial Photography. Zoom to extents and centre the image.

Using the Lines menu, select Single Seg and draw a line from the tee box using the Tooltip to identify the new bunker position (converting 253yds to 230m). This identifies the new bunker position.

We now have the position for the new bunker, but how do we move the bunker model to this position with the same coordinate system as the golf course?

First, let’s think about setting out the bunker, once correctly located. We could export coordinates to a CSV file, import them into a GNNS receiver for setting out. But let’s look at a simpler method that the greens staff can use, using appropriate technology - a tape!

Using the Dedicated CAD Backcloth of the bunker model, sketch over the sand outline with a green CAD polyline and the banking outline with a blue polyline (or use the option called commit to CAD, found in the Lines -> Move menu). Turn off the model, just leaving these CAD lines.

Still in CAD, from the Lines menu add a baseline in red followed by perpendiculars at 2m spacing, as shown below, adding labels and dimensions for setting out. See the Text and Dimensions menus.

The bunker is 14m wide, as identified by the labels. Add 7m perpendicular offsets to the 230m sight line in the Dedicated CAD Backcloth of the previously sketched 230m sight line. Use the Dimensions menu to add coordinates to their ends, as shown below (use Locking Mode E). These will be used to transform the bunker model into the same coordinate system as the golf course survey.

Returning to the bunker model, with its local coordinate system, turn on the Dedicated CAD Backcloth, with its setting out. We need this to reference the transformation. Add SP coded model points to the ends of the 14m baseline, locking onto their ends using Lock Mode E.

Use these two points, with coordinates as identified above in a 2D coordinate transformation. This brings the bunker into the same coordinate system as the golf course survey.

We’re nearly there with positioning, just one more item to fix: the levels. We know two spot heights in the banking are 1.1m and 1.3m above the existing ground.

With the LiDAR model included in the backcloth, use the DTM Height Difference tool to query the two points, making note of their levels.

A report will appear giving details of the point you have queried, the triangles surrounding this point and the coordinates of the LiDAR ground points, including height differences.

The ground levels at the two queried points are 85.040m and 85.043m, so we need to move the bunker model 85.0415m (averaged) vertically to match the LiDAR model.

Using Points -> Move -> Height -> Raise with the Pick Mode set to Rectangle, raise the heights of the bunker by 85.0415m. Everything will move, including the DTM, generating new contours.

Should you wish to look at the coordinates for setting out using GNNS, then these can be highlighted using List and sent to a CSV file. Note the Trimble Link option here.

There may be a better way to set out the new bunker, especially when work is to be carried out by green staff, who may be more familiar with string lines and tape measurements.

A laser range finder was used to measure out 253yds from the yellow tee box, and a paint mark was made on the left-hand side of the fairway, as identified in the original sketches. 7m perpendicular offsets were then identified from this baseline.

A 14m string line with 2m marked intervals was pegged between these points and offsets marked out as identified earlier. This allowed the green staff to remove the turf and begin excavating the sand area and building the bank at the back. This setting-out baseline can be re-established to mark out the sand and high points in the banking.

Here we can see the head greenkeeper marking out the new bunker, showing the setting out baseline. The image below shows the completed bunker in its correct position, excluding sand.

With material excavated from the sand area, the mound came in around the 23m³mark, as identified in our calculations. It’s important that the bunker can be seen from the 17^th tee. Sections could be used here.

Visualising the New Bunker

Let’s now look at how the transformed model fits in with the overall view. This can be done by using the transformed bunker model and adding the georeferenced image and LiDAR data as backcloths.

Below are both a plan and 3D visual of how the new bunker fits in with the existing layout. You will see spikes where the LiDAR grid picked up the tops of trees.

The golf architect has provided an artist's impression of the new bunker and images based on an existing bunker, as shown below. Let’s see if we can improve on this.

We’ve already located the new bunker in relation to the 17th fairway and its distance from the yellow tee box. Trees are stretched vertically into columns in the 3D view. This is due to the 1m LiDAR grid and the draping of the aerial photogram. Ideally, we would like to replace these columns with a fully 3D object, as shown below, but there is an issue with their heights.

ⁿ4ce allows the addition of 3D objects into a scene. If we have tree objects, these could replace the extruded column trees. But a little more work is required here.

This is just one of the objects provided within a 3D library with ⁿ4ce and found in n4ce Support Files (see icon on your desktop). Other 3D objects, including trees, can be downloaded from the internet, but you may have issues with scales!

Trees have different heights, so how can we take this into account? In its very basic form, we could count contour rings in the 3D view, as shown below, from the DSM LiDAR file.

LiDAR files, as supplied by the Environment Agency (EA), come in two forms: DTM and DSM. The DTM file has vegetation and trees removed, whilst the DSM shows the tops of vegetation and trees. We can use these files to work out ground levels and the heights of trees!

Note: The EA no longer provide ASC LiDAR files but instead point clouds in the form of LAZ files with and without vegetation. These can be imported into ⁿ4ce as point clouds, and subsequent LiDAR grids can then be created. ⁿ4ce also has powerful tools that can remove vegetation in point clouds.

Next, we need to look at how we place the tree objects. We use the Code Table here defining a code for a tree and attaching an object. This is done under the Symbols tab. Here we are using the TE code for a 1-Pt Scaled symbol, where its spread is defined by the Dimension S=, in Fields and height by SH=, identified as a height scaling factor in the 3D viewer.

The Dimensions S= and SH= have Default values on 1.0m. Also note here that R= is used for the radius of the bole, using a shape. This has a default value of 0.25m

We have access to both DTM and DSM LiDAR grids for the site.
A useful tool in Design -> Dimension -> Delta Z calculates the height difference from the current model to a reference the adds the value into a nominated Dimension (SH=). This will be used to scale the Tree object in the 3D viewer.

First, we need to create a new model called 3D Trees. We will digitise points where we can see them in the aerial photo in a backcloth. The spread Dimension will be added, and the level taken from the DTM LiDAR file.

Once the trees have been digitised, the heights can be added using the Delta Z tool mentioned above to provide the height of the trees to the DSM LiDAR file and look at this model in 3D! The aerial photo is draped over the DTM LiDAR file. The grid is added to show relief – if necessary.