Satellite Imagery vs. Aircraft Photography

For large volume data collection, which includes measuring many points, mapping, contouring, etc., one should consider photogrammetry as an efficient data collection tool (Lange 1996). Aerial photographs can be made with virtually any type of camera. Many successful applications have employed aerial photographs made from light aircraft with hand-held 35-mm cameras. The simplicity operation and low cost of purchase make the 35-mm camera ideal sensors for small area analysis. For large areas, over 100 different models of precision built aerial cameras are currently in use. They are specifically designed to expose a large number photographs in rapid succession with the ultimate in geometric fidelity (Lillesand and Kieffer 1994).

Clouds often block satellite images. The satellite images have coarser spatial resolution (20 to 1100-m compared to as little as 1-m resolution for aerial photography) that make within-field analysis difficult, if not impossible. Temporal resolution ranges from very good (morning and afternoon AVHRR) to only fair (every 16 days TM). So decisions can be made in a timely manner. But the images are usually expensive and take considerable time between image acquisition and delivery to the user. Satellite sensors have the advantages of good geometric and radiometric integrity. Accurate meteorological parameters such as cumulative growing degree-day maps, evapotranspiration maps, snow accumulation, rainfall and others are good satellite products. (Jensen 1996, Lillesand and Kiefer 1994)

Sensors aboard aircraft will be able to meet the requirements for fine spatial resolution (1-m), flexible and narrow spectral bands, inexpensive ($35/sq. mi.), frequent repeat coverage and quick turn-around times (Denison 1996). Image registration difficulties are a limiting factor. Quick assessment for crop damage control can be done with raw images. Comparisons over time and space require additional enhancement of the images. Calibrated data with values of reflectance, temperature and atmospheric corrections, are needed for most monitoring of seasonably variable soil and crop characteristics (Moran 1996).

The latest in technological tools could be a break through for both farming as well as soil and water quality. It’s called high resolution, infrared imagery.

Subtle differences in the field are detected with an aircraft mounted infrared sensor or hand-held 35-mm camera loaded with IR film (Lillesand and Kiefer 1994). Photo by Kodak

Key Benefits

• Farm and field specific imagery
• Great temporal resolution
• Great spatial resolution
• Vegetation stress early identification
• Time change analysis

Capabilities

These images are an electronic (digital) image. The image helps the grower identify and manage problems fast and efficiently to avoid serious losses (CTIC 1997).  The IR film processing used to take several weeks, so turn around time was not very good for detecting crop problems before they became severe. Today the IR film can be processed with the same system as regular slide film. It can be done at the local one-hour developer as soon as the plane lands, and be ready for scanning into a digital format a short time later. A Charged Coupled Display (CCD), otherwise known as a scanner, is used to convert the photograph into a digital image so the computer can work with the data (Jensen 1996). Each picture element (a pixel) is assigned a digital number depending on its brightness value. The values range from 0 to 255 in an 8-bit computer system (Lillesand and Kiefer 1994).

	Areas with drooping or curling leaves with reduced biomass may be suffering soil type related water stress, compaction of the earth around the roots, low levels of critical nutrients, pests or disease, or many other problems.
	Time change analysis using two images helps to identify areas of crop stress before visual signs appear. Now detective work needs to be done in the field to determine the cause(s).

The relative health of a crop is directly indicated by leaf structure. Leaves that are active in the photosynthetic process are the most reflective of near-infrared (NIR) wavelengths of the electromagnetic spectrum. The reflectance of leaves has been related to yield (Thomas et al. 1967).  Field investigations must be done to correctly identify the problems causing the vegetation stress. "Best management practices" (BMP) should be chosen that address the cause(s).

Leaf anatomy has also been related to photosynthetic activity (Hasketh and Baker 1969) and optical constants (Gausman and Allen 1973).

Normalized difference vegetation indices (NDVI) have been created since 1973 (Jensen 1996). NDVI is the difference between the near-infrared (NIR) and red (R) values for each pixel, divided by their sum.

NDVI = (NIR – R) / (NIR + R)

This formula is based upon the fact that healthy leaves reflect near-infrared light while absorbing red light. Photosynthesis and transpiration are two basic processes of vegetative canopies that are strongly dependent upon the amount of green leaf area. The NDVI is strongly correlated to the fraction of photosynthetically active radiation intercepted by the canopy. A near-linear relationship exists, and canopy geometry does not cause large variations in the relationship (Kanemasu et al. 1990).

"NDVI is widely used in satellite imaging." (Steven 1990) Regional vegetation NDVI temporal studies of crop health and biomass are regular features in the Journal of Advances in Space Research (Gupta et al. 1993) and International Journal of Remote Sensing (Derrien et al. 1992, Quarmby et al. 1993

University of California at Davis, CA agronomist Ford Denison has used computer image processing to derive site-specific information from aerial photographs of UC Davis experimental fields. "Aerial images can be processed to provide detailed information on spatial differences in soil properties and crop performance," said Denison, director of the Long Term Research on Agricultural Systems (LTRAS) project at UC Davis. "NDVI is widely used in satellite imaging, but we are not aware of any published accounts of its use with aerial photos." (Denison et al. 1996)

Remote Sensing Flowchart