Dunwoody, Ernest and Apan, Armando 
ORCID: https://orcid.org/0000-0002-5412-8881 and Liu, Xiaoye
(2013)
Effects of spatial resolution on measurement of landscape function using the landscape leakiness calculator.
In: Surveying and Spatial Sciences Conference (SSSC 2013): Collect, Connect, Capitalise, 17-19 Apr 2013, Canberra, Australia.
![[img]](http://eprints.usq.edu.au/style/images/fileicons/text.png)
|
Text (Published Version)
Dunwoody_Apan_Liu_SSSC2013_PV.pdf Download (456kB) |
|
|
Text (Documentation)
SSSC2013.pdf Download (562kB) | Preview |
Abstract
This study investigated the effect of changes in the scale of imagery on landscape elements and in turn on calculating the loss of resources in a rangeland catchment. The CSIRO Leakiness Calculator was used as the analytic tool to measure the leakiness of the experimental catchment under a variety of scenarios.
The approach was to use concurrent images of the same catchment area captured at different native resolutions and to upscale the high resolution image to match the lower resolution images to see the effect on calculated catchment leakiness. Two indices, the Stress-related Vegetation Index (STVI) and the Redness Index (RI) were used to measure land cover. The experimental catchment covered about 6,000 ha south-west of Charters Towers in North Queensland.
The two cover indices produced markedly different cover and leakiness results and these results varied with native image scale. The upscaled images also yielded different cover and leakiness results which did not coincide with comparable native scale image analyses.
The causes of the lack of agreement between the results were investigated using semivariance analysis techniques. The calculated leakiness of the native scale images showed no clearly defined relationship with the resolution-dependent amount of cover or the sill semivariance but it was strongly correlated with the negative power of the image resolution. The calculated leakiness of the upscaled image had a strong negative power correlation with the resolution-dependent amount of cover, image resolution and sill semivariance.
The findings highlight the importance of carefully considering all input variables including image preprocessing, type of cover index and resolution of the image when comparing the leakiness of different catchments using the Leakiness Calculator. These results indicate that it may be not possible to make meaningful comparisons of landscape leakiness of the same area between images of different scales or from different sensors.
|
Statistics for this ePrint Item |
Actions (login required)
 |
Archive Repository Staff Only |