Satellite Remote Sensing to Predict Potential Distribution of Dyers Woad (Isatis tinctoria)

Abstract

Landsat-5 Thematic Mapper spectral data and field survey information were combined to identify specific land cover types associated with 1741 dyers woad infestation points on the Cache National Forest in northern Utah. Ten of 60 possible cover types were strongly associated with current dyers woad infestations, and therefore considered highly suited to establishment and survival of the weed. Non-infested land of the same “highly suited” cover types comprised an additional 24 714 ha within the 149 021-ha study area, suggesting the potential for an eventual 124-fold increase in dyers woad incidence on the forest.

Discrimination of blackberry (Rubus fruticosus sp. agg.) using hyperspectral imagery in Kosciuszko National Park,NSW, Australia

“…The potential of remote sensing to identify and monitor weed infestations has been studied using aerial photography (Arnold et al, 1985;Everitt et al, 1996), multispectral imagery, such as Landsat Thematic Mapper (Dewey et al, 1991;Gardiner et al, 1998;Taylor, 1990) and multispectral digital video systems (Lamb et al, 1999;Lass and Callihan, 1997;Richardson et al, 1985) with mixed success (McGowen et al, 2001). In particular, Ullah et al (1989) and Frazier (1998) showed that high-spatial-resolution multispectral data could be utilized to produce accurate maps of blackberry distribution.…”

Mapping invasive plants using hyperspectral imagery and Breiman Cutler classifications (randomForest)

“…Both spatial and spectral resolutions impact the accuracy with which individual species are mapped. Moderate resolution satellite imagery is more suited to mapping at the community level because the spatial resolution is generally too coarse to distinguish individual species unless represented as a monoculture (Dewey et al, 1991;Sohn & McCoy, 1997).…”

Mapping Invasive Aquatic Vegetation in the Sacramento-San Joaquin Delta using Hyperspectral Imagery

“…A range of airborne and spaceborne imagery has been investigated for mapping terrestrial invasive species (Dewey et al, 1991;Everitt et al, 1995Everitt et al, , 1996Williams and Hunt, 2002;DiPietro et al, 2002;Underwood et al, 2003), however, in contrast to their terrestrial counterpart, the application of remote sensing techniques to invasive aquatic vegetation has been limited. This likely reflects the numerous additional challenges of detecting species in aquatic ecosystems, such as differentiating submerged species from water when present at low density or at high water turbidity, or detecting rooted submerged invasives when high tides cause the canopy to be substantially below the water surface.…”