Classifying land use using satellite temperature data

1. A computer implemented method for classifying land use, the method comprising:

receiving, by one or more computer processors, geographic data and land use data for a first geographic area;

receiving, by the one or more computer processors, surface temperature data for the first geographic area;

mapping, by the one or more computer processors, the geographic data and temperature data to a first set of map grid cells;

determining, by the one or more computer processors, temperature statistics for each map grid cell of the first set of map grid cells according to the surface temperature data;

training, by the one or more computer processors, a machine learning model according to the land use data and the temperature statistics, wherein the training comprises minimizing a loss function for neural network node weight values which accurately associate the land use data and the temperature statistics;

receiving, by one or more computer processors, geographic data for a second geographic area

receiving, by the one or more computer processors, surface temperature data for a second geographic area;

mapping, by the one or more computer processors, the geographic data for the second geographic area and the surface temperature data for the second geographic area to a second set of map grid cells;

determining, by the one or more computer processors, temperature statistics for each map grid cell of the second set of map grid cells according to the surface temperature data;

classifying, by the one or more computer processors, land use for the map grid cells of the second geographic area using the machine learning model and the temperature statistics for each of the map grid cells of the second geographic area, wherein the second geographic area lacks land use data; and

generating, by the one or more computer processors, a land use map for the second geographic area according to the map grid cells of the second geographic area and

providing the land use map of the second geographic area.

2. The computer implemented method according to claim 1 , further comprising:

training, by the one or more computer processors, the machine learning model according to population data associated with the first geographic area.

3. The computer implemented method according to claim 1 , further comprising:

training, by the one or more computer processors, the machine learning model according to physical data associated with the first geographic area.

4. The computer implemented method according to claim 1 , further comprising generating, by the one or more computer processors, a weather forecasting model according to the land use map of the second geographic area.

5. The computer implemented method according to claim 1 , further comprising:

ranking, by the one or more computer processors, the map grid cells of the second geographic area according to temperature statistics of the second geographic area; and

providing, by the one or more computer processors, an indication of the map grid cells of the second geographic area rankings as part of the land use map of the second geographic area.

6. The computer implemented method according to claim 1 , further comprising:

ranking, by the one or more computer processors, the map grid cells of the second geographic area according to temperature statistics of the map grid cells of the second geographic area;

generating, by the one or more computer processors, the land use map for the second geographic area according to the machine learning model; and

providing, by the one or more computer processors, an indication of the map grid cells of the second geographic area rankings as part of the land use map of the second geographic area.

7. A computer program product for classifying land use, the computer program product comprising one or more computer readable storage devices and collectively stored program instructions on the one or more computer readable storage devices, the stored program instructions comprising:

program instructions to receive geographic data and land use data for a first geographic area;

program instructions to receive surface temperature data for the first geographic area;

program instructions to map the geographic data and temperature data to a set of map grid cells;

program instructions to determine temperature statistics for each map grid cell according to the surface temperature data;

program instructions to train a machine learning model according to the land use data and the temperature statistics, wherein the training comprises minimizing a loss function for neural network node weight values which accurately associate the land use data and the temperature statistics;

program instructions to receive geographic data for a second geographic area

program instructions to receive surface temperature data for a second geographic area;

program instructions to map the geographic data for the second geographic area and the surface temperature data for the second geographic area to a second set of map grid cells;

program instructions to determine temperature statistics for each map grid cell of the second set of map grid cells according to the surface temperature data;

program instructions to classify land use for the map grid cells of the second geographic area, using the machine learning model and the temperature statistics for each of the map grid cells of the second geographic area, wherein the second geographic area lacks land use data;

program instructions to generate a land use map for the second geographic area according to the machine learning model; and

program instructions to provide the land use map of the second geographic area.

8. The computer program product according to claim 7 , the stored program instructions further comprising:

program instructions to train the machine learning model according to population data associated with the first geographic area.

9. The computer program product according to claim 7 , the stored program instructions further comprising:

program instructions to train the machine learning model according to physical data associated with the first geographic area.

10. The computer program product according to claim 7 , the stored program instructions further comprising program instructions to generate a weather forecasting model according to the land use map of the second geographic area.

11. The computer program product according to claim 7 , the stored program instructions further comprising:

program instructions to rank the map grid cells of the second geographic area according to temperature statistics for the map grid cells of the second geographic area; and

program instructions to provide an indication of the map grid cells of the of the second geographic area rankings as part of the land use map of the second geographic area.

12. The computer program product according to claim 7 , the stored program instructions further comprising:

program instructions to rank the map grid cells of the second geographic area according to temperature statistics for the map grid cells of the second geographic area;

program instructions to generate the land use map for the second geographic area according to the machine learning model; and

program instructions to provide an indication of the map grid cells of the second geographic area rankings as part of the land use map of the second geographic area.

13. A computer system for classifying land use, the computer system comprising:

one or more computer processors;

one or more computer readable storage devices; and

stored program instructions on the one or more computer readable storage devices for execution by the one or more computer processors, the stored program instructions comprising:

program instructions to receive geographic data and land use data for a first geographic area;

program instructions to receive surface temperature data for the first geographic area;

program instructions to map the geographic data and temperature data to a set of map grid cells;

program instructions to determine temperature statistics for each map grid cell according to the surface temperature data;

program instructions to train a machine learning model according to the land use data and the temperature statistics, wherein the training comprises minimizing a loss function for neural network node weight values which accurately associate the land use data and the temperature statistics;

program instructions to receive geographic data for a second geographic area

program instructions to receive surface temperature data for a second geographic area;

program instructions to map the geographic data for the second geographic area and the surface temperature data for the second geographic area to a second set of map grid cells;

program instructions to determine temperature statistics for each map grid cell of the second set of map grid cells according to the surface temperature data;

program instructions to classify land use for the map grid cells of the second geographic area, using the machine learning model and the temperature statistics for each of the map grid cells of the second geographic area, wherein the second geographic area lacks land use data;

program instructions to classify land use for map grid cells of a second geographic area according to the machine learning model;

program instructions to generate a land use map for the second geographic area using the machine learning model, wherein the second geographic area lacks land use data; and

program instructions to provide an output mapping of the land use classifications of the second geographic area.

14. The computer system according to claim 13 , the stored program instructions further comprising:

program instructions to train the machine learning model according to population data associated with the first geographic area.

15. The computer system according to claim 13 , the stored program instructions further comprising:

program instructions to train the machine learning model according to physical data associated with the first geographic area.

16. The computer system according to claim 13 , the stored program instructions further comprising:

program instructions to rank the map grid cells of the second geographic area according to the temperature statistics; and

program instructions to provide an indication of the map grid cell ranking as part of the land use map of the second geographic area.