Native vegetation condition assessments are mandated by several federal and state legislations. Condition assessments use a suite of field survey derived indicators to assess native vegetation extent and characteristics including woody canopy cover. Within highly modified landscapes, such as agricultural systems, literature has demonstrated a correlation between remnant patch size and indicators of native vegetation condition. Despite this general association between small patch size and poor condition; small remnant patches are regarded as keystone landscape structures essential to ecosystem services. Remote sensing technologies are commonly used to provide landscape scale assessments of woody vegetation extent. This paper demonstrates that the spatial and spectral characteristics of remotely sensed observations have a significant influence on the accuracy with which small patches of remnant woody vegetation can be identified and delineated.

An assessment of non-woody/woody extent layers (derived from medium resolution satellite remote sensing) has been made by comparison with a subset of high resolution data sources, including full waveform LiDAR and aerial photography. Preliminary results indicate that small woody remnant patches are under-represented by up to 40% in current methods of vegetation mapping. This under-representation is of relevance to many ecological and conservation applications as landscape pattern configuration metrics (e.g. connectivity) are particularly compromised. In conclusion, this study quantifies the differences in woody vegetation detection as a function of remote sensing data source. This enables the identification of the optimum platform for mapping woody vegetation at a given scale and “processing” cost (e.g. data purchase and interpretation complexity).