HOME | Research | Media | Careers | Contacts | Products | Search | Publications | Site Map
CSIRO Mathematics, Informatics and Statistics

 

Terrestrial Mapping and Monitoring

About us

Research Achievements

Applications Projects

Case Studies

Collaborators & Clients

Publications

Staff

Links

Mathematics for Mapping and Monitoring

Assessing forest condition

Dieback-affected jarrah forest

Long-term
sequences of aerial
photographs can be
combined with
elevation data to
identify vegetation
boundaries
and changes in vegetation
density over time

Dieback disease caused by phytphthora species is a major threat to the ecology and conservation of susceptible plant communities of southwestern Australia. During the relatively short time since its introduction, infection by phytphthora species has irreversibly destroyed the structure and diversity of many plant communities. Because no cure has been found for the disease, large area of trees, shrubs and smaller plants are laid to waste.

A reliable monitoring mechanism which can detect and quantify changes in vegetation is important for understanding and managing the dieback problem. The existing archive of aerial photographs is a resource which can be used for this purpose. With careful processing, sequences of aerial photographs can be analysed to give a representation of environmental changes that have occurred through a significant time frame. The resulting information can be used to update digital data bases on the spread of dieback.

In a study of an area of jarrah forest, a combination of historical aerial photography and elevation data was used to examine changes in the forest over the period. The changes may result from various causes. This study looked at the impacts of vegetation change attributed to dieback disease. The results show that the combination of historical aerial photography and elevation data may be used to identify vegetation boundary change due to dieback impacts.

Example study : Changes in jarrah forest 1972-1993

A sequence of historical aerial photographs was carefully processed to produce a multi-date image. Selected aerial photographs were digitally scanned. Abutting photos were then joined to form a mosaic for each date.

Differential rectification was used for each of the mosaics to rectify to a common base. Illumination correction and image calibration were performed, so that changes in different areas and times can be compared quantitatively.

Multi-date image (1972:blue, 1985:green, 1993:red) combined with elevation shows changes in forest condition

The colours in the above map indicate the location, timing and severity of changes over a twenty-year period. Blue shows areas where vegetation cover increased early in the period, such as pine plantations planted in the 1970's. Red areas indicate loss of vegetation since1985, either from clearing or gradual change such as forest disease movement. Yellow areas have shown gradual decreases in vegetation density, primarily indicating selective logging or forest disease.

The digital elevation model is incorporated to display the higher regions within the landscape in lighter colours. Contour lines are overlaid to aid the visual interpretation of the likely causes of change. The colour variations can be measured to determine rates of disease spread.

This work was produced by Graeme Behn , WA Conservation and Land Management.

To top

last updated August 05, 2005 12:52 PM

 

© Copyright 2010, CSIRO Australia
Use of this web site and information available from
it is subject to our
Legal Notice and Disclaimer and Privacy Statement