Image

Documentation

Learn how to use our API and UI and more.

graphical divider

The Wealth Change Data Layer

Understand how Atlas AI tracks the changing prosperity of settlements and how you can make use of this insight.

Overview

This data layer illustrates how the wealth of an area has changed over a given period of time. This data is based on Atlas AI’s Wealth layer describedhere. and computes the gradient of change based on the Wealth datasets included in the time range.

This data layer is useful as it allows one to understand and quantify long-term trends in prosperity at a glance. Our Wealth Change layer has been used to study:

  • The growth of settlements and communities in response to mineral and natural resource discoveries.
  • The deletrious impact of civil war in West Africa.

Data availability

The spatial resolution of the Wealth Change layer is the same as the Wealth Layer it is based on: 2km per pixel. This can also be aggregated up to district, regional and country levels using a weighted average in the same way that the Wealth layer can be. Since the Wealth Change layer is designed to show a change between at least two different wealth sets the temporal resolution of the Wealth Change layer can be no less than eight years and can be defined for any 4-year multiple higher than that for which we have data (e.g 12 years or 16 years).

User Type Geographic scale Temporal scale Granularity
Demo access Kenya-subregion 2003-2018 16 years / 1912m
Registered free access Kenya 2003-2018 16 years / 1912m
Full paid access Africa 2003-2018 16 years / 1912m

Details

Wealth: Wealth: The underlying Wealth values we are using are the Wealth composite values computed for the Wealth layer: more details can be found in their documentation.

Measuring Wealth Change: The computation of the change in wealth is done at the pixel level. For each pixel we take all the values of the Wealth composite falling into the time range. For example, if the time period is 2003-2018 we would have 4 wealth values w1 ,w2 ,w3 and w4 corresponding to 2003-2006, 2007-2010, 2011-2013, 2014-2017 respectively. Converting the 4 time periods to year offsets from the start of the time range we get y1, y2, y3 and y4 equal to 0, 4, 8, 12 respectively and then do a regression analysis with the wealth values wi as the independent variable and the year offers yi as the dependent variable. The returned gradient of the regression gives us the annual change in Wealth composite per year.

API access

For fuller documentation on the Atlas AI API click here

Data layers in the Atlas AI API are identified by numeric ids, the id for the Wealth Change layer is 4. Using this id as a parameter allows you to find all available Wealth datasets:

curl --header "Content-Type: application/x-www-form-urlencoded" \
--header 'token: yourapitoken' \
-G \
--data-urlencode 'indicatorId=4' \
https://data.atlasai.co/api/v1/datasets
                    

This will return all the distinct datasets relevant to the Wealth Change data layer. For example, we have shown a (partial) response for one of the possible Wealth Change datasets here:

[
  {
    "id":170,
    "dataType":"raster",
    "aggregationLevelId":1,
    "indicator":"Wealth Growth Rate",
    "aggregationLevel":"Pixel",
    "Version":"v1.1.0",
    "timePeriod":"2003-2018",
    "Available":true,
    ...
  }
]                   

Once you have established the id of the Wealth dataset you are interested in downloading the data is the same as for any other data layer, as detailed in the API documentation.

Icon For Arrow-up