Page 110 - Digital Analysis of Remotely Sensed Imagery

Overview of Remotely Sensed Data 81

automatically classified. However, shadow also facilitates the
identification of geologic structures and appreciation of terrain
relief. The chance of having more cloud cover is higher during
winter. The user must be knowledgeable in the maximum amount
of cloud cover allowable in an image. This limit should be explicitly
spelled out in signing contracts to purchase programmed data.

2.8.3 Cost of Data
The cost of remote sensing data varies enormously from absolutely
free to prohibitively expensive for academics. Although the price
charged by a data supplier or its approved agent is not negotiable, it
still pays to consider a few factors before an order is placed. In general,
the older the data, the cheaper they become. Data suppliers usually
sell several-years-old data at a heavily discounted price. So it may be
worthwhile to consider buying archived data that were recorded
years back. It is also cheaper to buy multitemporal, multiple-sensor,
multiple-scene data from the same data supplier. If the required data
are not archived, then an order must be accompanied by a special
programming request. Usually, the supplier charges a programming
fee in addition to the regular data cost. In this case the user can specify
the minimum acceptable level of cloud cover.
When inspecting the quoted prices, the user needs to be aware if
they include all processing and handling charges. The cost for processed
data that tend to be more accurate is higher. It is certainly cheaper to
have the processing done in-house for those users who have access to
a competent image processing system (refer to Chap. 4).

2.8.4 Mode of Data Delivery
Once the needed data have been identified, the next step is to place
an order. Prior to this, the analyst needs to be aware of the time it
takes to process the order and how quickly the ordered data are
delivered. If the requested data are to be recorded via special
programming, it may take a while to obtain the suitable ones. By
comparison, archived data can be delivered rather quickly. The
urgency attached to an order depends upon the nature of the
application. A higher priority should be attached to data for disaster-
related applications (e.g., fire monitoring and emergency response),
as they must be delivered almost instantaneously, even though this
means a higher cost. For other nonemergency applications, a delay of
a few days probably will not make much difference.
In placing an order, the user may have to decide the appropriate
way of data delivery. Data delivered via courier takes longer than
direct downloading from the vendor’s Web site. Courier delivery
involves more cost associated with data storage media and handling.
If this method is chosen, the data should be stored in a media that can
be read into the user’s computer properly. If accessibility to broadband