Skip to content

Kruger National Park

Scientific Services

Remote Sensing Applications in the Kruger National Park

Fire mapping and monitoring

The KNP fire objectives are twofold:

  1. To understand the role of fire (and other important interacting co-drivers) in the KNP and its ecosystems.
  2. To realistically evaluate fire threats particularly to lives, human infrastructure and wildlife, generate contextually responsible attitudes, and respond appropriately at the individual, organizational, and regional levels.

To achieve these objectives, RS and GIS play a prominent role in providing the required spatial information. This include:

  • Burnt scar maps
  • Fuel load maps
  • Fire intensity modeling
  • Burnt area totals

MODIS imagery is used for detecting and mapping fires in the KNP. This came as part of The MODIS Land Rapid Response system developed by NASA to provide rapid access to MODIS data globally, with initial emphasis on 250 m colour composite imagery and active fire data. NASA Goddard Space Flight Centre (GSFC), in collaboration with the University of Maryland provide real time image subsets from aqua and terra satellites.

Using the MODIS imagery, a composite of MODIS Bands 7 (with a bandwidth of 2105-2155 nm), 2 (with a bandwidth of 841- 876 nm) and 1 (with a bandwidth of 620-670 nm), are assigned to the red, green, and blue portions of the digital image.

Using coordinates of ignition points as well as identified fire scars from field rangers, the fire scars are mapped.

Vegetation quality and quantity assessment

Research on vegetation quantity and quality has been done through research projects as well as by the KNP scientific staff. Information sources for this range from aerial photographs to hyperspectral imagery. Hyperspectral remote sensing has been used to map grass quality in the northern plains of the Kruger National Park. Ongoing research is extending this to cover the whole of KNP as well as testing the utility of relatively cheap satellite images such as ASTER.

Satellite images such as NOAA AVHRR and MODIS are being used to extrapolate herbaceous biomass - an important variable for explaining wildlife movement and distribution.

Available data sets and software

The GIS Lab has two copies of ERDAS Imagine (image processing software) as well as ILWIS (Integrated Land and Water Information Systems). There is also a substantial image database including aerial photography and satellite imagery from a wide variety of platforms and sensors. Most of the satellite imagery is from the Landsat series, ranging from MSS scenes in the 1970’s to current ETM+ scenes. Processed MODIS imagery (Bands 7.2.1) is supplied in JPEG and TIFF format everyday and is used for fire mapping and monitoring.

The aerial photographs vary in scale from 1:10,000 to 1: 60,000 and are almost entirely panchromatic (black and white), and mostly in analogue form. The Stevenson-Hamilton Knowledge Centre in Skukuza has archives of Photogrammetric Engineering and Remote Sensing since 1984 (1984-1993 irregular).

Research projects

Apart from in-house research, the GIS/RS section is also involved with the management of scientific projects from outside researchers (Universities, GIS/RS private organisations etc.). The projects underway include the following:

Principle investigator Project
Verbesselt, J Fire risk assessment by fuel status assessment using low resolution earth observation
Smit, IPJ Using Geographical information systems and remote sensing to assess the trampling and grazing impact of herbivores on vegetation in the vicinity of artificial water sources in the Northern Kruger National Park, South Africa
Rommens, R Development of a remote sensing based index for monitoring vegetation structure parameters
Wessels K Relationship between herbaceous biomass and 1km AVHRR NDVI in Kruger National Park, South Africa
Deman, K Development of a methodology to monitor fire fuel status and identity optimal fuel models using high resolution multi-spectral satellite imagery
Aplin, P Vegetation monitoring in the Kruger National Park using multiscale remote sensing analysis
Nackaerts, K Predicting and monitoring of ecosystem disturbance by fire
Lhermitte, S Development of a regreening index for monitoring vegetation regrowth after fires
Ferwerda, J Using hyperspectral remote sensing of plant chemicals to explain the interaction between secondary plant compounds and herbivores
Peel, M Landcover classifications for Private Nature Reserves
Mutanga, O Integrating remote sensing and spatial statistics to model biomass distribution in a tropical savanna