Resolution and extent

Let’s add some more technical words to our growing vocabulary

Resolution

The level of detail in the dataset, indicating the size of the smallest object you can detect

Imagine you have a landscape or piece of art made up of tiles. You could make much more detailed pictures from tiny mosaic tiles (1cm across) than large flagstones. Resolution in raster data is determined by the size of grid cells. Resolution in vector data is determined by the precision of x,y coordinates i.e. how many decimal places the coordinate has, and therefore the minimum spacing at which you can place adjacent points or vertices

High resolution datasets contain a lot of detail within a given area, for example interview data from individual households, or aerial or drone photography on which you might be able to see individual zebra or animal burrows. Low resolution datasets have lower detail, for example national-level population counts, or imagery from early satellites such as Landsat (30m grid cells).

Also known as grain, or frequency. I prefer the terms ‘fine grain’ (think silt or sand) to indicate high resolution, and ‘coarse grain’ (think boulders or pebbles) for low resolution, as these phrases are more intuitive and easy to remember

Extent

The area covered by the dataset, also known as coverage

Contrast a global dataset (large extent) with data from a single survey location (small extent)

Contrasting spatial resolutions

Move the slider across the images below to see how the resolution affects the complexity and precision of spatial data

Spatial and temporal

The terms resolution and extent can be applied to the temporal dimension (time) as well as the spatial. In other words, you can describe a dataset as having high temporal resolution (e.g. observations repeated every minute), or low temporal resolution (e.g. observations spaced many years apart)

Dimension	Property	Example of a low value	Example of a high value
Spatial	Resolution	One value per country	Submetre grid cells
Spatial	Extent	Single survey location	Entire continent
Temporal	Resolution	Every decade	Every minute
Temporal	Extent	Single point in time	Spanning multiple decades

See it for yourself

To help you understand these concepts, load the following new datasets to your QGIS project:

1. OpenStreetMap fine grain vector data: add using Add Vector Layer...
2. Coarse-grained data from the Digital Chart of the World - select ‘Vietnam’ and the following two datasets from the DivaGIS data download page:
   1. Roads (shapefile)
   2. Land cover (virtual raster): add using Add Raster Layer... or drag & drop the .vrt file into your QGIS map view
3. VNM_cov.qml QGIS style file for the DCW landcover layer you just downloaded from DivaGIS - save it in the same folder as the VNM_cov.vrt and VNM_cov.grd files
4. Chế Tạo Nature Reserve boundary from Protected Planet:
   1. Click the green Download button
   2. Select SHP, then Non-Commercial use
   3. Unzip the downloaded file WDPA_WDOECM_Jun2021_Public_555594126_shp.zip
   4. Alongside many other folders/files, you’ll now see WDPA_WDOECM_Jun2021_Public_555594126_shp_0.zip. Unzip this second .zip file to extract the contents
   5. You can now add the .shp to your QGIS project

1. Compare the visual detail of the roads from OSM and DCW
2. Compare the fine-grain Che Tao Nature Reserve boundaries from Protected Planet with the corresponding polygon in the IUCN Redlist’s NomascusConcolor_Distribution layer
3. Compare the two landcover raster layers from Copernicus and DCW - consider both their spatial resolution, and how well their landcover classes (thematic values) overlap with each other

---