On Zimbabwe Idai recovery site, we should load population demographics

[akahure]

We should load the High Resolution Population Density Maps + Demographic Estimates from https://data.humdata.org/dataset/highresolutionpopulationdensitymaps-zwe to zimbabwe idai site here http://idairecovery-zimbabwe.onalabs.org/

[cKellyDesign]

From a Mapbox perspective this data source is a Raster layer type because the data is a high-resolution geotiff. If we can convert that to a tileset then we just need to add a new Layer Type (raster) to Gisida.

[cKellyDesign]

@akahure An update on the FB Population data raster layer:

TLDR

We're making concurrent progress on the Mapbox tileset and the gisida code changes to support raster layers.

What we're trying to do

We are still trying to get the FB Population geotiff to upload to Mapbox correctly so we can render it on the map. We have to do some conversions to get the file to be compatible with Mapbox and it's proving difficult to achieve. Slowing down the process is the fact that these are very large files and we are only getting errors after we have uploaded new tilesets.

To format the geotiff we are applying color conversions and mercator projection translations, and seem to have only done so in a way that is compliant with Mapbox uploads with a clipped version of the original data.

@KipSigei has added some core code to support raster tilesets, we know this code is working because it was able to render this clipped version of the geotiff successfully.

What we need

Aside from a geotiff that works, we also need to know what color to use for the layer. I suggest green so that additional blue (flooded damage) layers and red (collapsed damage) layers can blend where they overlap (purple = flooded + collapsed; yellow = dense population + collapsed; teal = dense population + flooded; blackish brown = dense population + collapsed + flooded;)

[mberg]

I've already created a vector for this. It will be easy to stylize.

Here you go

https://studio.mapbox.com/tilesets/ona.97ca3h1w/

On Tue, Feb 4, 2020 at 4:01 PM Conor Kelly notifications@github.com wrote:

@akahure https://github.com/akahure An update on the FB Population data raster layer: TLDR

We're making concurrent progress on the Mapbox tileset and the gisida code changes to support raster layers. What we're trying to do

We are still trying to get the FB Population geotiff to upload to Mapbox correctly so we can render it on the map. We have to do some conversions to get the file to be compatible with Mapbox and it's proving difficult to achieve. Slowing down the process is the fact that these are very large files and we are only getting errors after we have uploaded new tilesets.

To format the geotiff we are applying color conversions and mercator projection translations, and seem to have only done so in a way that is compliant with Mapbox uploads with a clipped version of the original data.

@KipSigei https://github.com/KipSigei has added some core code to support raster tilesets, we know this code is working because it was able to render this clipped version of the geotiff successfully. What we need

Aside from a geotiff that works, we also need to know what color to use for the layer. I suggest green so that additional blue (flooded damage) layers and red (collapsed damage) layers can blend where they overlap ( purple = flooded + collapsed; yellow = dense population + collapsed; teal = dense population + flooded; blackish brown = dense population + collapsed + flooded;)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/onaio/gisida/issues/411?email_source=notifications&email_token=AAAOEQKTX24YBJD4KO275ITRBHJTNA5CNFSM4KPEGXBKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKZFOXI#issuecomment-582113117, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAOEQMZSPVUXKODA5IKOHLRBHJTNANCNFSM4KPEGXBA .

[mberg]

note it's only visizble from zoom level 9 and down. if you want you can use tippecanoe to change the shapefile to make it a higher zoom. i'll upload the shape to gisida-views

On Tue, Feb 4, 2020 at 4:19 PM Matt Berg mberg@ona.io wrote:

I've already created a vector for this. It will be easy to stylize.

Here you go

https://studio.mapbox.com/tilesets/ona.97ca3h1w/

On Tue, Feb 4, 2020 at 4:01 PM Conor Kelly notifications@github.com wrote:

@akahure https://github.com/akahure An update on the FB Population data raster layer: TLDR

We're making concurrent progress on the Mapbox tileset and the gisida code changes to support raster layers. What we're trying to do

We are still trying to get the FB Population geotiff to upload to Mapbox correctly so we can render it on the map. We have to do some conversions to get the file to be compatible with Mapbox and it's proving difficult to achieve. Slowing down the process is the fact that these are very large files and we are only getting errors after we have uploaded new tilesets.

To format the geotiff we are applying color conversions and mercator projection translations, and seem to have only done so in a way that is compliant with Mapbox uploads with a clipped version of the original data.

@KipSigei https://github.com/KipSigei has added some core code to support raster tilesets, we know this code is working because it was able to render this clipped version of the geotiff successfully. What we need

Aside from a geotiff that works, we also need to know what color to use for the layer. I suggest green so that additional blue (flooded damage) layers and red (collapsed damage) layers can blend where they overlap ( purple = flooded + collapsed; yellow = dense population + collapsed; teal = dense population + flooded; blackish brown = dense population + collapsed + flooded;)

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/onaio/gisida/issues/411?email_source=notifications&email_token=AAAOEQKTX24YBJD4KO275ITRBHJTNA5CNFSM4KPEGXBKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEKZFOXI#issuecomment-582113117, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAAOEQMZSPVUXKODA5IKOHLRBHJTNANCNFSM4KPEGXBA .

[cKellyDesign]

@akahure is this still a requirement? If so, what's the priority?