Open Forest GIS Data: 7 Datasets in QGIS

Introduction

In October 2023, Japan's Forestry Agency released data on forests and terrain as “high-precision forest resource information.” The data can be downloaded from the G-Spatial Information Center, and anyone can use it freely. The release covers three prefectures: Tochigi, Kochi, and Hyogo.

This article introduces all seven types of forest and terrain data released by the Forestry Agency, visualizing each one in QGIS.

This article is a revised version of the “QGIS Blog” content that was published on the MIERUNE Inc. website.

Tree species polygons

Tree species polygons classify the distribution of cedar forests, cypress forests, broadleaf forests, and so on. They let you visualize the distribution of different tree species and the diversity of the forest on a map.

The tree species polygon data is distributed as vector tiles. Vector tiles are a modern format for handling map data efficiently. Map elements (points, lines, and polygons) are stored as vector data and divided into tiles, each covering a fixed area. In QGIS, vector tiles load quickly and can be displayed and manipulated dynamically on the map.

To load vector tiles into QGIS, right-click Vector Tiles in the Browser panel and click New Generic Connection.

The Vector Tiles Connection dialog opens. Enter a name of your choice for the connection and the URL of the vector tile source. The tree species polygon vector tiles released this time also come with a Style.json that defines colors by tree species. Enter the URL of this Style.json in the Style URL field to apply the style. You can leave this field empty, but adding it makes the vector tiles look better. After you fill in each field, click OK to finish registering the vector tiles.

After registration, the name you entered appears as a sub-item under Vector Tiles in the Browser panel. Double-click it to display the vector tiles (tree species polygons) on the QGIS map canvas. Once the vector tiles are registered in QGIS, you can browse them easily without downloading the polygon data each time.

Forest resource aggregation mesh

The forest resource aggregation mesh covers the forest with 20 m mesh polygons and summarizes the representative tree species, number of standing trees, average tree height, timber volume, and more for each mesh.

Because it holds a wide range of attribute information on forest resources, this is a highly versatile dataset for forest data analysis.

However, the way the mesh is divided seems to differ by prefecture. Of the three prefectures in this release, the Tochigi mesh starts from 20 m intervals and is divided further by tree species. When consistency and uniformity with the data of other prefectures are required, you need to take these different ways of dividing the mesh into account.

By the way, vector tiles are also distributed for this data, just like the tree species polygons. These two types of forest polygon data use the GeoPackage data format. GeoPackage is a modern format that can store multiple geographic data types, such as vector, raster, and tiles, in a single file. It is the default format in QGIS. Until now, Shapefile has been the common vector data format for forest information, but it consists of multiple related files, which often made it cumbersome to handle. GeoPackage, in contrast, combines all of this information into one file, so the data is much easier to handle, move, and share.

Digital elevation model (DEM)

A digital elevation model (DEM) is data showing the height of the terrain. However, if you load the DEM released this time into QGIS as is, the coloring looks odd.

This DEM is in the “Terrain-RGB” format, a special tile set that encodes terrain elevation data into the RGB (red, green, blue) color channels. Because of this distinctive representation, it can be distributed efficiently in common image formats such as JPEG, which enables fast downloading and rendering on the web. QGIS also loads this format efficiently and renders it quickly. However, if you load it in QGIS as is, it looks like the image above, so some setup is needed.

When setting up the XYZ connection, set Interpretation to MapTiler Terrain RGB.

The layer now appears completely black. This is because the band 1 values of the tiles vary widely, so the values need to be adjusted.

As a test, set Min and Max to a range of 0 to 3000, and the layer takes on the familiar black-and-white look.

Micro-topography map (CS relief map)

The micro-topography map (CS relief map) illustrates the relief and slope of the land, so you can grasp detailed terrain features. Depressions (valleys) are shown in blue and raised areas (ridges) in red. Gentle slopes are shown in light colors and steep slopes in dark colors.

Next, combine the DEM from earlier with this micro-topography map. First, in the Layer Styling panel, set the DEM to Hillshade.

Then set the Blending mode to Multiply and place the micro-topography map below the DEM layer. The micro-topography map now appears with a three-dimensional feel.

You can also try the 3D display feature of QGIS. In the menu bar at the top of the window, click View → 3D Map Views → New 3D Map View.

The 3D view panel opens, but nothing is displayed in 3D yet. Click the wrench button, then click Configure.

The 3D Configuration dialog opens. Select the Terrain item and check Terrain. Switch Type to DEM (Raster Layer) and select the DEM layer from earlier for Elevation. Finally, set Tile resolution to 256 px and click OK.

The micro-topography map now appears in 3D.

You can clearly see how the coloring differs with the shape of the terrain. Incidentally, “CS relief map” is the common name in the fields of location information and GIS, but in forestry it is often called a “micro-topography map.” The Forestry Agency's “Manual for Terrain Interpretation Using CS Relief Maps” describes in detail how to interpret terrain with this micro-topography map (CS relief map).

Slope classification map

The slope classification map shows land slope divided into 5-degree intervals. The closer to red, the steeper the slope, and the closer to blue, the flatter the area. At a resolution of 5.0 m, it is somewhat coarse and not suited to detailed analysis, but it is effective for getting an overview of general terrain features and land slope over a wide area.

If you switch to the 3D display with the same steps as before, you can grasp the slope more intuitively.

Laser forest type map

The laser forest type map uses color to show the characteristics of tree species and crown shapes, based on the reflection intensity of laser pulses from airborne laser surveying. Compared with the orthoimages (aerial photographs) generally used for forest type interpretation, it has no cloud shadows, so its advantage is that it is easier to distinguish the distribution of tree species and the shapes of tree crowns in the forest.

If you overlay the tree species polygons introduced earlier with boundaries and tree species labels, you can see that the coloring differs by tree species.

Digital canopy height model (DCHM)

The digital canopy height model (DCHM) is the digital surface model (DSM) minus the digital elevation model (DEM). It does not show the elevation of the terrain but the height of the crowns of standing trees (the parts where leaves and branches grow thickly), so it can visualize the height and shape of the forest canopy.

With more advanced analysis of this data, it is also possible to extract “treetop points,” which give the position of each individual tree.

As with the DEM earlier, convert the digital canopy height model to hillshade and multiply it with the laser forest type map, and the shapes of the standing trees become easier to see.

In the 3D display, elevation is not included, so tree heights are easy to compare. The standard 3D view of QGIS did not display this data well, so the 3D display here uses the “Qgis2threejs” plugin. (* The height scale is set to 4 times to make it easier to see.)

Trial release of a forest web GIS

MIERUNE assisted with processing the forest information data released this time. MIERUNE also developed a site called “Forest Information Web-GIS,” where you can easily view this data on a web map, and it is currently open to the public on a trial basis.

Even people who cannot use QGIS can check the forest information data in a browser.

Conclusion

Until now, the forest information data available to the public was limited to little more than polygon data of national forests. Data that covers private forests as well, with forest resource volumes released too, can be called a major step toward “open data for forest information.” It is also excellent progress that forest information data has been adopted in next-generation geospatial data formats such as GeoPackage, vector tiles, and Terrain-RGB. It would be desirable to use this as a chance to explore moving away from older location data formats such as the traditional Shapefile.

We look forward to more prefectures releasing forest information data in the future.