Update:
2021-06-10 Created by Tawan, T.
2021-06-28 Modify to handle incomplete runs
Why:
- Goal at the creation of this script is the create refined grids finer than the sub district area to present the population density obtained form facebook which comes in points of lat and lng.
What:
- Use Uber - h3 and Geopandas to extract boundary of each province from QGIS's shapefile of Thailand provinces; then, pass them on to Uber's H3 for griding
Avail Dataset:
1.General population
2.Youth population (15-24)
3.Elder population (>60)
4.Under five population
5.Men population
6.Women population
How:
1.Select h3_level to generate h3 grids
h3_level=9
Select if using the specific provinces
# if_all_provinces=1 => Use all provinces in boundary_data
# if_all_provinces=2 => Incase, previous run not complete, Continue running from what being left off from previous run.
if_all_provinces=1
2.Run Inegreated_Shapefile_Generation.py
Note: Code will read the boundary data file in folder boundary and generate shapefiles ( in csv ) to use with Kepler.gl and QGIS
in table : [TSR_ADHOC].[dbo].[H3_Kepler_Grid_Lv8_Province_2] and [TSR_ADHOC].[dbo].[H3_Grid_Lv8_Province_PAT] respectively.
=> Code uses h8-level grid which covers around 0.7 km2 on the map
tutorial on H3:
1.Study ipython files in ipython_notebook folder (to run use Folium environment and install kernel in jupyter notebook)
Note:
1.Facebook population density: Estimation of numbe of people in each 30 metres grid around the world
2.Data could be obtained from data.humdata.org
ref:
- https://dataforgood.fb.com/docs/high-resolution-population-density-maps-demographic-estimates-documentation/ (general information of fb population density dataset)
- https://data.humdata.org/organization/facebook (download data)
- https://h3geo.org/docs/api/traversal#hexring (H3 API documentation - H3 grid resolution level)
- https://kepler.gl/demo (Keler.gl)
DEMO
H3's grid Level 8 of BKK displayed on Kepler.gl
H3's grid Level 9 of BKK displayed on Kepler.gl
