Implement allocation of built-up area

[jipclaassens]

No description provided.

[jipclaassens]

From email of Lewis:

If we want to set local maxima, we could use the following thresholds:

• If a cell has the highest local share of built-up area in T1, it can only increase by 11 pp in T2 (or by 50% - local share if this is smaller than 11 pp)
• If a cell doesn’t have the highest local share of built-up area in T1 but has neighbouring cells with built-up, it can only increase in T2 to the local maximum in T1 + 6 pp (or by 50% - local maximum if this is smaller than 6 pp)
• If a cell is not built-up and doesn’t have any built-up in its neighbouring cells in T1, it can only increase by 4 pp in T2

We would just have to decide if we use an area of 3x3 or 5x5 (minus corners)

NB:

• use wgs84 ~1km grid
• 0.0% == non-builtup
• 0.01% == builtup

[jipclaassens]

• We need suitability for built-up and population
• Household size could change over time, and have effect on the number of people per cell. We assume these change to be homogeneously per country.
• Do we need to model non-residential use? Implement non-residential land use as static #18
• Per cell, residential share/non-residential share/non-builtup. More possibility if cell is relatively a lot unbuilt.

Q:

• Do we assume redevelopment? Of is the number of dwellings per built-up area initially fixed? Can we model a new Manhattan in developing countries?
• can we get projections about household size?