Skip to main navigation Skip to search Skip to main content

UAV-mounted multispectral vegetation surveys from 2019-2022 tropical savanna field campaigns

  • Tom Eames (Creator)
  • Adrian Kaluka (Creator)
  • Roland Vernooij (Creator)
  • Cameron Yates (Creator)
  • Jeremy Russell-Smith (Creator)
  • Guido van der Werf (Creator)

Dataset

Description

This repository contains drone data from vegetation surveys before and after fire events conducted from fieldwork in 2019, 2021 and 2022 in southern African and northern Australian savannas. These are split by instrument used to capture the images: a MicaSense RedEdge camera mounted on a DJI Matrice 100 (M100, 2019) and a DJI Phantom P4 Multispectral (P4, 2021 & 2022).

There are two files for each survey, both in GeoTIFF format. The bands file contains a stitched orthophoto, processed from multiple individual images captured from approximately 50m altitude with 80% overlap at each site. The classes file shows the same image classified into 5 separate vegetation categories. Information on these classes, as well as a description of the individual band contents is contained in the README file.

Also included is a csv file with ground measurements for above-ground biomass, in g/m². These are labelled with the survey ID code to which they correspond.

The two .qml files are suggested formatting for the classes GeoTIFF files, for use with QGIS software.

Please see the README file for further details on the data contents. For any queries on this dataset, please contact Tom Eames ([email protected]).
Date made available2 Jun 2025
PublisherVrije Universiteit Amsterdam

Cite this