Root system estimation based on satellite remote sensing: An applied study in Eastern Uganda

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceeding


The density of roots is an important factor influencing the rate and magnitude of landslides. Due to the increased variability in climate, mainly rainfall, Eastern Uganda is severely struck by an increasing number of these mass movements, often with human casualties as one of the negative impacts. The aim of this study is to explore the possibility to estimate the depth and density of the root system influencing the resistance to landslides, from satellite remote sensing data. 104 samples were collected in field, where the root system was classified into 5 different classes, from non-existing to dense and deep (forest). The study was carried out in the Mount Elgon area located at the Ugandan-Kenyan border. The field data were then compared with 30 m Landsat TM data, in order to investigate possible links between reflectance (single bands as well as indices) and ground truth data. The results indicate that, following this methodology, it is not possible to estimate the root system density based on the remotely sensed data, since the maximum Cohen’s kappa value of 0.081 is judged deficient.


  • Petter Pilesjö
  • Malin Ahlbäck
  • Maya Ahlgren
  • Hanna Ekström
  • Malin Hansson
  • Iris Mužić
  • Nike Rosenström
  • Franziska Weichert
External organisations
  • Lund University
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Remote Sensing


  • Landslides, Remote sensing, Uganda, Root system density, NDVI, EVI
Original languageEnglish
Title of host publicationGeospatial Technologies for All
Subtitle of host publicationshort papers, posters and poster abstracts of the 21th AGILE Conference on Geographic Information Science. Lund University 12-15 June 2018, Lund, Sweden
EditorsAli Mansourian, Petter Pilesjö, Lars Harrie, Ron van Lammeren
PublisherAGILE, Association of Geographic Information Laboratories for Europe
Number of pages5
Publication statusPublished - 2018 Jun 13
Publication categoryResearch
Event21st AGILE Conference on Geographic Information Science, 2018 - Lund, Sweden
Duration: 2018 Jun 122018 Jun 15


Conference21st AGILE Conference on Geographic Information Science, 2018