{"id":115,"date":"2020-03-19T22:00:30","date_gmt":"2020-03-19T22:00:30","guid":{"rendered":"http:\/\/blogs.qub.ac.uk\/gis\/?p=115"},"modified":"2020-03-19T21:56:51","modified_gmt":"2020-03-19T21:56:51","slug":"whats-under-the-trees-visualising-opendatani-lidar-in-a-geographical-information-system","status":"publish","type":"post","link":"https:\/\/blogs.qub.ac.uk\/gis\/2020\/03\/19\/whats-under-the-trees-visualising-opendatani-lidar-in-a-geographical-information-system\/","title":{"rendered":"What’s Under The Trees? – Visualising OpenDataNI LiDAR in a Geographical Information System."},"content":{"rendered":"

Our friends at\u00a0OpenDataNI<\/a>\u00a0\u00a0have been busy securing some fantastic spatial data sets for public use. in particular LiDAR<\/a>\u00a0surface elevation data collected by various NI Government Departments over the last 12 years. \u00a0I’ve covered LiDAR technology in a previous post<\/a>\u00a0showing how it is used in our research here at SNBE<\/a>, but I thought it might be useful to show how you to can, process, visualise and interrogate the data using free Geographical Information Systems (GIS) software applications. <\/em><\/p>\n

(This blog QGIS 3.10 – Long Term Release<\/a>)<\/em><\/p>\n

Spurred on by the release of Environment Agency England <\/a>LiDAR elevation data, OpenDataNI has collated and uploaded LiDAR data from the Department of Infrastructure\u00a0<\/a>(Rivers and Flooding\/TransportNI) and the Department of Communities<\/a>\u00a0(Historic Environment Division). At this stage it is important to note that the LiDAR survey data covered in this blog was originally collected independently and for different output roles; flood prediction and mapping, and identification, analysis and management of archaeological landscapes. As a result the output data sets do not cover \u00a0the whole of NI and are at different ground resolutions (Rivers c1m and Historic c0.2m). For example researchers are better able to identify small archaeological surface features at higher resolutions (<1m) but modelling a river basin and catchment for flood flow prediction does not necessarily require these very high resolutions (>1m).<\/p>\n

\"riverlidar\"<\/a><\/p>\n

River Basin Coverage \u00a02004 -2014 (OpenDataNI)<\/em><\/pre>\n
\"\"<\/a><\/p>\n
Historic Division Coverage 2008 - 2014 (OpenDataNI)<\/em><\/pre>\n
Ok so you have identified the liDAR coverage area you area interested in and have downloaded the data from OpenDataNI (be careful LiDAR data is large!). \u00a0What can you do with it? How can we visualise and analyse the surface elevation that has been surveyed? \u00a0I’m lucky here at QUB as a GI Scientist, we have lots of great ‘commercial’ GIS applications we can access to for teaching and Research (ESRI<\/a>). \u00a0But what if you want to have a go at finding a new geological or archaeological surface feature that has yet to be mapped! \u00a0Thankfully there are some great free and ‘Open Source’ GIS that you can install at home! \u00a0One of the most widely used free GIS packages is QGIS<\/a>. Download and install to your favourite OS flavour (Win\/OSX\/Linux). I’m using the 3.4 version for this blog post.<\/p>\n
\"\"

QGIS Web and Download Site - www.qgis.org<\/a><\/em><\/pre>\n
Once installed<\/a> prepare your LiDAR data (I’m using the Cave Hill data<\/a> in this example – 2014). \u00a0You will notice that the data has been downloaded as one large compressed ‘ZIP’ file, \u00a0you need to extract this file to access the LiDAR data. Once extracted, delete the downloaded ZIP file to save space. \u00a0Open the data folder, in this case we have 4 sub-folders, 2x 0.2m (20cm) resolution Digital Surface Model data sets and 2x 0.2m (20cm) Digital Terrain Model data sets. \u00a0[I’ve covered the difference between DSM and DTM \u00a0and how they are created in the previous blog<\/a>]. \u00a0We will work with the DTM<\/strong> Irish National Grid <\/a>(ING) coordinate system data. \u00a0This data has been\u00a0projected<\/a> to Irish National Grid coordinates and has been processed with vegetation and buildings removed (anthropocentric filtering), helping us to see the ground surface below. \u00a0The DSM includes all the LiDAR data and is useful for mapping tree and building heights! The\u00a0Irish Transverse Mercator<\/a> (ITM) projection has been applied to the other data copies and is\u00a0 common in the Republic of Ireland.<\/p>\n
\"f1\"<\/a>\u00a0\"f2\"<\/a><\/p>\n
LiDAR folders and data files (CG GIS QUB)<\/em><\/pre>\n
Coordinate Reference System<\/em><\/p>\n
Before we ‘build’ the LiDAR data surface we will set the project Map Projection<\/a> or Coordinate Reference System (CRS) to Irish Grid. Click on the CRS button on the bottom right of the QGIS window to activate the CRS popup. In the ‘Filter’ box, type ‘Irish<\/strong>‘ and select TM75 \/ Irish Grid –\u00a0<\/strong> Traverse Mercator 1975 is the current coordinate system for Northern Ireland and supported by the Ordnance Survey of Northern Ireland<\/a>.<\/p>\n
\"\"<\/a><\/p>\n
CRS Popup (CG GIS QUB)<\/pre>\n
Build the LiDAR VRT Surface\u00a0<\/em><\/p>\n
The LiDAR surface elevation data is stored as ASCII format files (.asc), a Raster<\/a>\u00a0grid format file. \u00a0Each tile covers a fixed area depending how the data has been segmented, with 1000’s of 20cm cells recording height. \u00a0The tiles are seamless with joining tiles (‘edge to edge’). \u00a0We could add the individual tiles to the GIS as required but we are going to use a more productive virtual ‘Mosaicing’ process to create a seamless, continuous elevation surface to analyse. Open QGIS desktop and from the menu select Raster > Miscellaneous > \u00a0Build Virtual Raster. \u00a0<\/strong>(VRT)<\/p>\n
[Make sure Menu > Plugins > Installed > Processing is checked]<\/p>\n
\"\"<\/a><\/p>\n
QGIS Build Virtual Raster\u00a0(CG GIS QUB)\u00a0<\/em><\/pre>\n
In the ‘Input Layers<\/strong>‘ box click on the open folder button (…<\/strong>) to open the ‘Multiple Selection<\/strong>‘ popup and click on ‘Add Files<\/strong>‘ and browse to the downloaded and extracted LiDAR folder that we are going to mosaic (eg. 20cm_DTM_ESRI_grids_ING).\u00a0 Order the files based on ‘type’ and using the shift key on your keyboard select all<\/strong><\/span> the .asc<\/strong> files together and click Open<\/strong>. In the Multiple Selection popup a list of the selected files will be detailed on the left.\u00a0 Click OK<\/strong> to add the selection to the Input Layers, you should have 126<\/strong> files (elements) added if using the Cave Hill raster tile layers. Finally uncheck<\/u><\/b>\u00a0the ‘Place each input into a separate band<\/strong>‘\u00a0 option and leave all options as the defaults.<\/p>\n
As an optional step in the ‘Advanced Parameters<\/strong>‘ and to insure the output CRS is Irish Grid you can select TM75 \/ Irish Grid\u00a0 from the ‘Override Projection..<\/strong>‘ drop down option.\u00a0 Finally click Run<\/strong>\u00a0to start the VRT process!<\/p>\n
\"\"<\/a><\/p>\n
QGIS Build Virtual Raster\u00a0(CG GIS QUB)<\/em><\/em><\/p>\n
Depending on the spec of your computer it may take a few minutes for the process to complete and display the new VRT mosaiced dataset.<\/p>\n
\"\"<\/a><\/p>\n
Completed Cave Hill LiDAR layer Mosaic in QGIS (CG GIS QUB)<\/em><\/em><\/pre>\n
QGIS will now display the LiDAR mosaic as a ‘Grey Scale’ colour ‘render’ by default, and list the new VRT layer in the ‘Layer Panel’ on the left. The grey scale Raster layer displays the elevation range \u00a0(76m – 352m) as low (black) to high (white). Elevation is metres above sea level, in this case Belfast Lough. \u00a0This colour render can be changed in the layer properties (Symbology) if you want to experiment!<\/p>\n
We now have a continuous elevation surface at 0.2m resolution. Unfortunately this is not a very informative visualisation at the moment, even if you zoom in to the data! \u00a0Thankfully the GIS has a few neat tools to help out! First we will create a Hillshade Model<\/a>, the ‘go to’ technique for a first look at a LiDAR elevation surface.\u00a0A hillshade is a grayscale representation of the surface, with the sun’s relative position taken into account to calculate and model shadows (shading) on the elevation surface. We can set the elevation of the sun above the horizon (Altitude) and the compass bearing (Azimuth). \u00a0This process creates ‘virtual shadows’ that can help outline surface topography including geological, geomorphological and archaeological features. Buy experimenting with different sun orientation settings we can help identify features that are very difficult to see in the DTM surface alone.<\/p>\n
\"q6\"<\/a><\/p>\n
Azimuth and Altitude (ESRI.com)<\/em><\/pre>\n
To generate a Hillshade surface In QGIS, from the Menu select\u00a0Raster >\u00a0 Analysis > Hillshade<\/strong>, with the Cave Hill VRT selected as the Input Layer<\/strong> (this will create a temporary Hillshade layer unless you specify a name and folder location to save to). The default settings in QGIS for the Hillshade Sun elevation (altitude) is 40 degrees and the Azimuth (compass)\u00a0 is 315 degrees. \u00a0These can be changed to model differing Sun angles.<\/p>\n
Click Run<\/strong> to start the analysis. After processing completes the new layer will be added to QGIS!<\/p>\n
\"\"<\/a><\/p>\n
Hillshade Layer in QGIS\u00a0(CG GIS QUB)<\/em><\/em><\/pre>\n
The Hillshade shows a shaded ‘greyscale ‘colour value range between 0 and 200 (black to white) and does not<\/strong><\/span> represent the surface elevation. \u00a0Because we have used a very high resolution DTM (20cm) we should be able to see features as small as 20cm on the ground! \u00a0 Use the Zoom and Pan tools to explore the new topography layer and examine all the ‘humps n’ bumps’ not normally visible! If you tick off [x<\/strong>] the LiDAR VRT surface in the layer below, the map will render faster.<\/p>\n
\"q8\"<\/a><\/p>\n
Hillshade Layer in QGIS. Zoomed in showing shading relief and surface features. 'Relic\u00a0farmland' \u00a0on the South\u00a0Slope. \u00a0(CG GIS QUB)<\/em><\/em><\/pre>\n
We can also add background Aerial Imagery (Orthophotos) and mapping to help with our geography and landscape analysis using the ‘XYZ<\/strong> Tiles’<\/strong> layer in QGIS. In the ‘Browser Pane’<\/strong> layer list double click on the XTZ Tiles<\/strong> layer to open the mapping options.\u00a0 In this case click Google Satellite<\/strong> or Bing VirtualEarth <\/strong>to add background aerial imagery. These new layers will be added to the Layer list and you can switch the layers on an off (x<\/strong>)\u00a0 to compare the Hillshade and ‘current’ aerial images of the location, or move layers in the list to rearrange their order. \u00a0You need a web connection to connect to background mapping.\u00a0 Here are some other XYZ Tile<\/a> options.<\/p>\n
\"snip_20161103112600\"<\/a><\/p>\n
Hillshade Layer and XYZ Tile Google 'Satellite' Maps background imagery. (CG GIS QUB)<\/em><\/em><\/pre>\n
It is also possible to set the ‘Transparency’ of layers to aid visualisation. Double click over the layer in the list to open the Properties Window.<\/strong>\u00a0Combining a new colour ramp for the elevation data and setting the transparency of the Hillshade is a great way to visualise both the LiDAR elevation data and Hillshade together and can help interpret surface features. \u00a0Open the LiDAR Data layer Properties<\/strong> window (double click on layer) and change the Render Type<\/strong>\u00a0to Singleband pesudocolour. <\/strong>In the Colour <\/strong>drop down\u00a0<\/strong>\u00a0below select a colour ramp, RdYlGn<\/strong>\u00a0(Red Yellow Green) works well and check the Invert<\/strong> box. \u00a0Click OK<\/strong>. \u00a0Double click on the Hillshade layer and set the layer transparency to 30% and click OK.<\/p>\n
Save your QGIS LiDAR project and start searching for surface features!\u00a0 \ud83d\ude42<\/p>\n
\"\"<\/a><\/p>\n
Hillshade Layer and LiDAR DTM overlay. Large landslide features, East face of Cave Hill. \u00a0(CG GIS QUB)<\/em><\/em><\/pre>\n
View the LiDAR surface in 3D!\u00a0<\/em><\/p>\n
Finally we can use the 3D viewer built into QGIS to visualise the LiDAR surface and your visualisation.\u00a0 3D map views are useful way to add a more realistic \u2018context\u2019 to you GIS work and can bring extra understanding to your data that may not be apparent in more ‘traditional’ 2D views.<\/p>\n
QGIS uses a dedicated 3D viewer with mouse and Tool navigation. Views are based on the active checked\/symbolised data layer in the Layer Pane and can be exported as images for lots of uses including sharing cool landscape features on social media platforms!<\/p>\n
In Menu > View> New 3D Map View<\/strong><\/p>\n