This paper describes the application of a relatively new remote sensing tool, airborne video imagery, for distinguishing weed and brush species on rangelands. Plant species studied were false broomweed, spiny aster, and Chinese tamarisk. A multispectral video system that acquired color-infrared (CIR) composite imagery and its simultaneously synchronized three-band [near-infrared (NIR), red, and yellow-green] narrowband images was used for the false broomweed and spiny aster experiments. A conventional color camcorder video system was used to study Chinese tamarisk. False broomweed and spiny aster could be detected on CIR composite and NIR narrowband imagery, while Chinese tamarisk could be distinguished on conventional color imagery. Quantitative data obtained from digitized video images of the three species showed that their digital values were statistically different (P = 0.05) from those of associated vegetation and soil. Computer analyses of video images showed that populations of the three species could be quantified from associated vegetation. This technique permits area estimates of false broomweed, spiny aster, and Chinese tamarisk populations on rangeland and wildland areas.

Field spectroradiometric plant canopy measurements showed that broom snakeweed [Gutierrezia sarothrae (Pursh.) Britt. and Rusby # GUESA] and spiny aster (Aster spinosus Benth. # ASTSN) had lower near-infrared (0.85-μm) reflectance than did other associated rangeland shrubs and herbaceous vegetation. The low near-infrared reflectances of both species were attributed to their erectophile (erect leaf/stem) canopy structures. These low near-infrared reflectance values caused broom snakeweed to have a dark-brown to black image on color-infrared aerial photos (0.50- to 0.90-μm), whereas spiny aster had a dark reddish-brown to black image. Other rangeland plant species had light-brown, red, or magenta images. Computer-based image analyses of color-infrared film positive transparencies showed that broom snakeweed and spiny aster infestations could be quantitatively differentiated from associated rangeland species. Computer analyses can permit “percent land area infested” estimates of broom snakeweed and spiny aster infestations on rangelands.