Table 1.

Summary of remote sensing applications for non-native plant studies.

Moderate spatial/ spectral	Sensor specifications and examples: Spatial: 10-100 m. Temporal: Long (16-26 days). Spectral: < 20 bands. ASTER, SPOT, TM/ETM+.
	Species traits and remote sensing strategies: Large stands. Different phenology to co-existing plants. Selection of images acquired in the right season.
	Limitations: Coarse spatial and spectral resolutions unable to extract non-native species from a mixture of different plants.
High spatial	Sensor specifications and examples: Spatial: < 10 m. Temporal: Short (1-4 days). Spectral: ∼5 bands. Aerial photographs, QuickBird, IKONOS.
	Species traits and remote sensing strategies: Unique spatial patterns. Pronounced flowering season. Selection of images acquired in the right season.
	Limitations: Inflexibility of airborne data collection. Pixel spacing still not fine enough to observe plants at the species level. Unable to detect plants with no distinct flowering pattern due to the coarse spectral resolution. Impractical for large scale monitoring due to the time intensive approach (e.g., visual inspection), and small spatial extents.
High temporal	Sensor specifications and examples: Spatial: ≥ 250 m. Temporal: Very short (1-2 days). Spectral: < 40 bands. AVHRR, MODIS.
High temporal	Species traits and remote sensing strategies: Unique phenological characteristics. Combination of models and time-series vegetation indices derived from the images.
	Limitations: Insufficient spectral bands to extract non-native species from large pixels covering other plants, surface soils and senescent vegetation. Time-series vegetation pattern obscured by cloud and snow requiring a statistically sounded smoothing algorithm for noise removal. Difficult to conduct field validation due to the large plot size. Overwriting non-native species signals by climatic variations such as precipitation.
Hyperspectral	Sensor specifications and examples: Spatial: Varied (0.5-30 m). Temporal: Varied. Spectral: > 100 bands. AVIRIS, Hyperion.
	Species traits and remote sensing strategies: Unique signatures in the hyperspectral space. Spectral mixture analysis. Biochemical analysis at the canopy level.
	Limitations: No direct link between invasion mechanism and sophisticate hyperspectral analyses. A small swath width of data collected from aircraft restricting the ability for large spatial scale monitoring. Inflexibility of airborne data collection. High similarity in the spectral space among species.
Active remote sensing	Sensor specifications and examples: Spatial: 0.5-100 m. Temporal: Varied. Spectral: 1 band. 3-D view. LiDAR, RADARSAT.
	Species traits and remote sensing strategies: Large and pure stand. Monitoring the progress of species invasion from data acquired at two time points.
	Limitations: No spectral information and data only useful with good field knowledge.
Image fusion	Sensor specifications and examples: Spatial: ∼0.5 m. Temporal: Varied. Spectral: 200+ bands. 3-D view. Pushing the limits of spatial, spectral and dimensional resolutions in modern remote sensing. CAO.
	Species traits and remote sensing strategies: Unique signatures in the hyperspectral space. Detection of non-native species of different height-levels (overstory and understory) at the very fine spatial scale.
	Limitations: High cost of data collection. Requirement of high performance computing power. Inflexibility of airborne data collection. Small spatial extents restricting very large spatial-scale monitoring.