The remainder of this section will adapt the approach previously
described to a practical application which utilizes multiple
sources of data such as Landsat, meteorological, and agricultural.
The sample application described is a simple early warning pro-
cedure which utilizes the APU, a computer system, and a USDA
commodity analyst.
C. Application of APU and Gridded Data Base (Example)
The APU's and gridded data base will support an early warning system
as one of many applications. The purpose of this application is
to detect early warning of changes affecting production and quality
of commodities. It will signal the occurrence of an unusual event,
such as floods, drought, and winterkill, that can affect crop condi-
tion and rely on APU data for areal and impact boundaries.
USDA personnel are currently implementing an alarm application that
will alert the remote sensing analyst of an unusual event. This sys-
tem will give early warning alarms for the following:
Preseason soil moisture
Drought
Late/early freezes
Maximum air temperature
Lodging/harvest problems
Minimum air temperature
Planting soil temperature
Excessive precipitation
Insects/disease/erosion
OOOOOOOOO
Critical or threshold values will be established for each of the
above items. They will be specific for each grid cell, crop type,
and crop biological growth stage.
The logic of an alarm application is shown in Figure 6. In this
example, the system has alarmed for a potential winterkill problem.
The analyst verifies, using information within the master data set,
that winterkill has occurred. The secondary data set is then
activated and an analysis utilizing available spectral and agri-
cultural data is done to determine the impact of the adverse
condition. Landsat data are then used to track a situation until
a final production impact can be made.
The reader can visualize any number of potential applications
whether simple or complex using the defined approach. However, a
note of caution is warranted. We have discussed the utilization of
multispectral and meteorological data within a common data framework.
We have built no hypothesis that suggests these data and their
associated technologies are in any sense at an operational point.
Research and development will continue into the foreseeable future