Learning Goals

  • Understand the difference between land cover and land use
  • Basics principles in visual image interpretation of land cover and land use
  • Knowledge of the Land Use and Land Cover Survey (LUCAS)

Background

Mapping the Earth’s surface

The Earth’s surface is a mosaic of various land cover and land use types. Remote sensing based satellite imagery provide synoptic views in space and time and are an ideal source for mapping this diversity. Mapping with remote sensing requires a basic understanding of the difference between land cover and land use as well as the definition of an appropriate classification scheme that is ideally standardized.

Greater Berlin area as seen in January and July 2019 from the Sentinel-2 platform

Land cover and land use

The terms land cover and land use are often used interchangeably, however, they are distinct from each other. Land cover describes the biophysical composition of the Earth’s surface. Land use describes the anthropogenic use of the Earth’s surface. The differentiation of both terms is crucial in remote sensing as satellites provide imagery of land cover, whereas information on land use is mostly based on additional human interpretation.

Land cover: crops, land use: agriculture (left); land cover: needleleaf trees, land use: forest management (middle); land cover: building materials, land use: residential area (right)

Visual image interpretation

Visual image interpretation implies the human’s ability to analyze the content of images, e.g., land cover and land use from remote sensing imagery. Visual image interpretation encompasses two steps, first the perception of objects according to their external attributes and, second, the actual interpretation of their meaning.

Attribute Description (example) Interpretation (example)
Contrast, color, brightness Transition from light to dark blue Variations in water depth
Geometry (shape, size) Sinuous ribbon-like object River
Texture (structure of a surface) Rough surface with vertical line patterns Maize cultivation
Spatial context (functional interrelationship) Rail tracks that intersect a building Railway station
3D structure Visible facade and long shadow Tower

Example subsets used for interpretation in table above (Source: images from Google EarthTM)

LUCAS survey

The Land Use/Cover Area frame Survey (LUCAS) provides harmonized and comparable information on land use and land cover across the EU. The collection of LUCA data is based on a point survey and follows the standardized LUCAS classification scheme to ensure comparability between EU member states. The information provided by the LUCAS survey is used in various contexts, e.g., Common Agriculture Policy, EU-Biodiversity Strategy, Copernicus Earth Observation Program.

The LUCAS survey encompasses around 1,1 mio points with 2 km spacing covering the whole territory of the EU. In a fist phase, pre-defined land cover classes are assigned to each point based on photo-interpretation. In a second phase, a stratified subset of points is extracted for the field survey where samples are classified according to the full land use and land cover class definitions of LUCAS (e.g. in 2015: 273500 points, 750 field surveyors). The LUCAS survey takes place every three years (last 2018).

LUCAS survey points 2015 in Berlin/Brandenburg (Source: eurostat)

LUCAS and remote sensing

The LUCAS survey points provide a unique reference source for remote sensing. For example, Pflugmacher et. al (2018) used LUCAS as a training and validation source for wall-to-wall mapping of pan-European land cover based on Landsat satellite imagery.

Pan-European land cover map by Pflugmacher et al. (2018) based on Landsat imagery and LUCAS data

Session materials

Download the session materials from our shared repository.

Exercise

Land cover and land use

  • Provide one example for a land cover and a land use type for each image chip below.

LUCAS survey

  • Make yourself familiar with the LUCAS classification scheme below (a high resolution version is provided as PDF-document in the course materials).

Assignment

The objective of this assignment is to map land cover and land use based on visual interpretation of a very high resolution image covering a part of the Wuhlheide Park, Berlin.

Develop a classification scheme

  • Open the following data sets in QGIS:
    • Extent of the area on the Wulheide to be mapped.
    • Very high resolution satellite image (e.g., Google Satellite imagery using the QuickMapServices Plugin).
    • 10 m Sentinel-2 image (see materials session 2).
  • Display the data sets appropriately (e.g., Sentinel-2 as false-color RGB-composite).
  • You may use the MapSwipe Tool plugin for swiping layers.

False-color Sentinel-2 (R= band8/nIR, G=band3/green,B=band2/blue) and Very high resolution true-color satellite images of the Wuhlheide Park (Source: Copernicus Open Access Hub and Google.cn Satellite

  • Please use the following layout suggestion below to create a classification scheme for the following eight surface types:
Class Surface type
1 Building
2 Sealed non built-up
3 Rail track
4 Coniferous forest
5 Deciduous forest
6 Grass
7 Bare land (soil/sand)
8 Water
  • For each class, provide an image sample, attribute/class description as well as the LUCAS land cover and use nomenclature in the most detailed level.

Template to create the classification scheme

Digitize example areas

  • Digitize example areas for the eight classes according to your classification scheme. Please consider the following:
    • Two example areas per class.
    • Minimum mapping unit equals 0.05 ha (500 m2), objects smaller than that are included in the surrounding class.
  • Create a map of your results (including a legend).
  • Provide a bullet list of encountered problems and inaccuracies.

Submission

  • Upload your classification scheme, map and bullet list as a PDF file to moodle.

  • General submission notes: Submission deadline for the weekly assignment is always the following Monday at 10am. Please use the naming convention indicating session number and family name of all students in the respective team, e.g. ‘s01_surname1_surname2_surname3_surname4.pdf’. Each team member has to upload the assignment individually. Provide single file submissions, in case you have to submit multiple files, create a *.zip archive.


Copyright © 2021 Humboldt-Universität zu Berlin. Department of Geography.