How spatial links form new business entities

In a previous post I showed how links can be created based on the spatial relationship between two business entities. In most cases, a spatial link is just a link as any other, defining a relationship between two geographical objects, answering question like how "far are these two objects apart" or "how many objects are in the vicinity of this location".

In some cases however, a spatial link might result in the creation of new business information.
Let's consider the following example. From time to time, heavy rainfall causes parcels near watercourses to flood. These floods cause damage to the crops that farmers grow on these parcels. In order to compensate the farmers, we determine which areas are flooded, how big they are, and what crops were grown. In business terms, we have two entities of data:

1. A collection of parcels, including information about ownership and crop growth.
2. A flooded area that is either digitized or modeled, containing information about the date, time and nature of the flood.

On the left, the parcels. On the right, the flooded area.

Both entities have a spatial attribute with its location, which in both cases are polygons. In a map, they look like this.

 

Data model for floods

The source data model is shown above, and is actually quite simple. We just have two source tables, each with some attributes and both with a geometry column. They are spatially connected based on the geometries being at the same spot on the earth's surface.

Data vault model for floods

Modeling this source data with the data vault approach gives us the following data vault model. Again, it's a very simple model. The two source tables result in two hubs that are spatially linked (see the previous post about spatial links here).

But in this particular case, the parcels and flood are not just on the same location, but they can interact with eachother. When we intersect the flood contour with the parcels, we get the areas of each parcel that is actually flooded. 

Flooded areas (in red)

In doing this, we're actually creating new business entities which possess their own unique spatial location and combination of attributes. As you can see above, the 3 red polygons form new business entities called 'flooded areas'. So, when overlaying two hub entities we're essentially creating new business information by intersecting their spatial attributes. Remember the 3 types of spatial references? The creation of new spatial business entities through spatial linking is mostly the result of overlap-type spatial reference.

These new business entities can be modeled as a link with satellite or as a new hub with a satellite attached to the exisiting link. There's currently some debate about the validity of the link satellite that centers around this question. I'm not going to weigh in on that debate here though. Either way, the resulting geometry of the intersection (the red polygons) are stored as new geometries in the satellite of either the link or the newly created hub. The resulting new geometry is called INTERSECTION_GEOMETRY. Below I've modeled both scenarios to give you an idea of how this looks, so you can take your pick.

New business entity modeled as link satellite

New business entity modeled as a hub

Comments or questions? Please let me know.