Marked Assignment: Project I - Comprehensive Image Collection

This worksheet introduces your first project which aims in the compilation of a comprehensive dataset for further analysis in this and the parallel Geo Information Systems course.

Things you need for this worksheet

  • R — the interpreter can be installed on any operation system.
  • RStudio — we recommend to use R Studio for (interactive) programming with R.
  • Git environment for your operating system. For Windows users with little experience on the command line we recommend GitHub Desktop.
  • Datasets provided by the course instructors.
  • Many other things depending on your specific workflow and analysis design.

The big picture

The overall goal of this and the parallel GIS course is to compile a remote sensing based dataset that provides information on

  1. the tree species on an individual tree basis and
  2. the forest structure and competition between trees in the respective forest areas.

The graded course certificate will be an article following the type “Practical Tools” (but up to 2,000 words) of Methods in Ecology and Evolution that describes the project from its motivation and intention over the methods and results to the discussion of the final information. There will only be one article for both courses (this one and the GIS course).

The datasets required for project will be compiled within this and the parallel GIS course:

  1. Comprehensive dataset for species prediction (this course, project 1),
  2. Tree crown segmentation (GIS course, project 1),
  3. Prediction of tree species (this course, project 2),
  4. Forest structure and tree competition (GIS course, project 2).

This cross-linking of the project works has an important implication on the planing of this project. One cannot specify the actual dataset requirements at this stage but one must iteratively adjust the requirements based on insights and knowledge gained in the parallel work. Hence, this dual project workflow acts quite similar to other knowledge gaps one might have at the beginning of a project, e.g related to computing images based on focal operations. Acting under this uncertainty is a key competence and nothing to panic about as long as one keeps it in mind.

More information on the tree crown assignment and the forest structure project will be supplied in the parallel GIS course. To get an idea on the remote sensing based species prediction or spatial prediction project in general, the publication of Ludwig et al. 2016 on detecting woody vegetation in South Africa can be one starting point (by the way, this publication is the result of the B.Sc. thesis of Annika).

Project for building a comprehensive remote sensing dataset

Please compile a comprehensive dataset based on spectral and structural information derived from both multi-channel aerial and LiDAR datasets which is suitable for the parallel and upcoming project work mentioned above. The dataset should provide information on

  • spectral properties,
  • structural properties,
  • spatial patterns of spectral and structural properties.

Please use the upcoming spotlight units as a sort of guideline and a source of inspiration for the compilation.

Please start working on your article in parallel, focusing on the introduction and the methods chapter. Both chapters must be presented in session 8.

Updated: