High Resolution Imagery
for Resource Management: ADAR

Satellite-based imagery, such as Landsat TM, has many uses for resource management applications but higher spatial resolution is needed when analyzing important watershed features such as riparian buffers, field boundaries, crop rotations, impervious surfaces, or urban landuse.

In October of 1999, Positive Systems flew over the Washington D.C. area and acquired 4,691 frames of ADAR (Airborne Data Acquisition and Registration) imagery. The ADAR System 5500 SN8 is based on the Kodak DCS 420-IR digital camera. This camera setup acquires a picture in four regions of the electromagnetic spectrum. Three of these are in the visible range, and the fourth is in the near infrared range.

The image below shows the position of each frame, represented by a red dot. Washington D.C. is in the lower left-hand corner and a portion of the Chesapeake Bay is on the right hand edge.

Each frame of the ADAR imagery is a rectangle of approximately 900 meters by 1300 meters. Each pixel within the frame represents 0.64 square meters on the ground. This high-resolution data allows us to view ecological details previously difficult to obtain.

Below is a portion of western Washington D.C. from a August 1998 Landsat TM image. Each pixel represents an area of 900 square meters. The 3 black dots within the image represent 3 ADAR frame centers.

The image below is a mosaic of 3 ADAR frames of the same area shown above. The increase in resolution and discrimination of features is evident. This image gives a sense of the utility of very high resolution imagery compared to moderate resolution Landsat imagery. Clicking within the white box will bring up a detailed image.

These images represent two popular ways of viewing the ADAR data. The "natural color" image uses the red, green, and blue regions of the spectrum collected by the camera to replicate the way we see the landscape. The "color infrared" image shifts the spectrum collected to include a portion of the near-infrared. This combination helps in determining the health of plants and aids in discriminating water and urban features from plants.

A useful advantage of the ADAR imagery is that the spectral ranges collected are virtually identical to that of Landsat TM. Many algorithms have been developed for vegetation analysis using Landsat as the data source. Similar spectral response with the ADAR instrument will allow us to conduct research on the scalability of algorithms developed for a 30 meter resolution image to that of a 0.8 meter resolution sensor. One example of this is in the scaling of impervious surface calculations from TM scale to ADAR scale. Research into the identification of impervious surfaces using Landsat is being carried out at the watershed scale. Impervious surfaces are important in watershed management to gain insight into the amount of urban runoff entering the Chesapeake bay. Access to high resolution data such as ADAR will significantly improve our calculations for this parameter, and can augment current for field sampling work.