Using remote sensing and spatial information technologies to detect and map two aquatic macrophytes.
This paper describes the light reflectance characteristics of water hyacinth (Eichhornia crassipes) and hydrilla (Hydrilla verticillata) and the application of airborne videography with global positioning system (GPS) and geographic information system (GIS) technologies for distinguishing and mapping the distribution of these two aquatic weeds in waterways of southern Texas, USA. Field reflectance measurements made at several locations showed that water hyacinth generally had higher near-infrared (NIR) reflectance than associated plant species and water. Hydrilla had lower NIR reflectance than associated plant species and higher NIR reflectance than water. Reflectance measurements made on hydrilla plants submerged below the water surface had similar spectral characteristics to water. Water hyacinth and hydrilla could be distinguished in colour-infrared (CIR) video imagery where they had bright orange-red and reddish-brown image responses, respectively. Computer analysis of the imagery showed that water hyacinth and hydrilla infestations could be quantified. An accuracy assessment performed on the classified image showed an overall accuracy of 87.7%. Integration of the GPS with the video imagery permitted latitude/longitude coordinates of water hyacinth and hydrilla infestations to be recorded on each image. A portion of the Rio Grande River in extreme southern Texas was flown with the video system to detect water hyacinth and hydrilla infestations. The GPS coordinates on the CIR video scenes depicting water hyacinth and hydrilla infestations were entered into a GIS to map the distribution of these two noxious weeds in the Rio Grande River.