The Evolution of Everything (EvE) and Genetic Programming 139
2.1 Example Uses of Fog Lifter
Some examples of the sort of use cases where Fog Lifter would be advantageous
are shown in Figs.1 and2.InFig.1, data from agricultural fields are collected from
sensors placed in or near fields and gathered into a model of each field which, when
collected together, produces a model for the entire farm. The data from these models
may be shared with other local farmers, who may be part of the same Farm Coop,
to give a local view of the crops being grown and the state of the soil and water use
in the area. These can then be aggregated regionally to inform local markets, crop
insurance agents and farm machine usage planning. Note that often this is not raw
data but processed results from the raw data collected from the fields.
Figure2 shows how this approach can be extended to a global scale if the data is
made available to other interested parties, up to, and including global entities. Here
local or regional data is not only used on larger geographical areas, but selected
portions of the analysis done at a local area used for different purposes.
For example, governmental water planners may not be interested in regional
crop production, but are very interested in agricultural water usage as part of
larger forecast of water needs that include industrial and civilian water use. This
information may also be integrated with weather forecasts that influence potential
agricultural water usage as well as the availability and state of water sources such
as aquifers, reservoirs and snow packs.
However, multinational NGOs may be interested not only in water usage and
long range weather models, but agricultural yields and the spread of pests in order
to predict where there are likely to be crop shortages and danger of starvation.
While this kind of planning already exists, by combining data from thousand
or millions of sensor sources related to farming, weather, insect control, and water
source and usage, the quality and accuracy of data and predictions could be greatly
improved.
Fig. 1 Aggregation of local analytics