LandGIS Open Land data service - global stack of environmental layers

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

LandGIS Open Land data service - global stack of environmental layers

Tomislav Hengl-5

We have recently released a webmapping system to serve global consistent
environmental and Earth science layers at spatial resolutions from 10 km
to 250 m (hopefully also soon at 100 m). This is an Open Data system as
majority of layers are distributed under the Open Data Commons Open
Database License (ODbL) and/or Creative Commons Attribution-ShareAlike
4.0 International license (CC BY-SA). Read more about LandGIS in:
http://opengeohub.org/about-landgis.

You can access the web app at:

https://landgis.opengeohub.org

The system currently (Jan 2019) serves about 300+ layers from relief and
geology, to vegetation indices, climatic images, soil properties and
classes and potential and actual vegetation. Complete overview of
available layers is available at:

https://github.com/Envirometrix/LandGISmaps

In addition to the web-mapping app, data can be accessed using the:

- Geonode installation at https://maps.opengeohub.org,
- LandGIS REST API services at https://landgisapi.opengeohub.org,
- LandGIS WCS at https://geoserver.opengeohub.org/landgisgeoserver/web/,

A copy of all layers is also available via Zenodo.org i.e. via an unique
URL.

To access data at point locations best use the REST API. For example, to
access monthly precipitations at a location X, Y you can use:

https://landgisapi.opengeohub.org/query/point?lat=7.58033&lon=35.6561&coll=layers1km&regex=clm_precipitation_imerge.(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)_m_1km_s0..0cm_.*_v0.1.tif

which returns a GeoJSON (table) with precipitation values in mm.

To access values of LandGIS layers at multiple points you can use:

curl -X POST --form "points=@test_points.geojson" --form
"layer=pnv_fapar_proba.v.jul_d_1km_s0..0cm_2014..2017_v0.1.tif"
https://landgisapi.opengeohub.org/query/points -o results.json

where test_points.geojson is the input GeoJSON file containing
coordinates of points. The multi-point access is currently limited to
max 20 points, but we hope to increase this number gradually. More
examples of how to construct spatial queries are available at:
https://landgisapi.opengeohub.org

In addition to the REST access, you can also access the LandGIS data
using the Web Coverage Service (WCS) functionality of the Geoserver e.g.
to subset layers using a bounding box. For example, to download surface
temperature for July for an area of about 300 by 300 km you can use:

https://geoserver.opengeohub.org/landgisgeoserver/ows?service=WCS&version=2.0.1&
request=GetCoverage&
coverageId=layers1km:clm_lst_mod11a2.jul.day_m_1km_s0..0cm_2000..2017_v1.0&
subset=Lat(41,45)&subset=Long(32,35)

The read limit for WCS is 4GB and response size limit is 200MB. This
means that WCS might fail if you try to fetch too large bounding boxes.
If this happens we recommend instead downloading whole GeoTIFFs from Zenodo.

We are currently preparing R functionality to allow users fetching data
from LandGIS in a more systematic way (import, overlay, subset,
plot...). If you would like to contribute to this initiative, especially
to testing the R functions, please send me an email. Also we would
appreciate if you would report any bug or inconsistency you discover
via: https://github.com/Envirometrix/LandGISmaps/issues

If you are currently producing any similar types of data (e.g.
environmental layers at resolutions from 100 m to 1 km for a global land
mask) and if you would like to publish this data on LandGIS, please
forward a proposal for publishing your global layers:
https://opengeohub.org/submitting-global-layers-inclusion-landgis and we
will get on it asap.

thank you,

Tom Hengl
https://opengeohub.org/people/tom-hengl

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: LandGIS Open Land data service - global stack of environmental layers

Tomislav Hengl-5

Just added an example with multipoint query to the LandGISmaps rep:

https://github.com/Envirometrix/LandGISmaps/tree/master/tutorial

REST now works for up to 50 points. The impressive speed is mainly
thanks to using SSD and cloud GeoTIFF settings
(https://github.com/Envirometrix/LandGISmaps#cloud-optimized-geotiff).

library(rjson)
library(rgdal)
library(fossil)
library(plotKML)

## REST multipoint query ----
path <- geopath(lon1=4.9, lon2=4.9, lat1=52.3, lat2=35.1)
writeOGR(as(path, "SpatialPointsDataFrame"), "test_points.geojson",
layer="test_points", driver="GeoJSON")
## overlay points and grids:
browseURL('https://landgis.opengeohub.org/#/?base=Stamen%20(OpenStreetMap)&center=49.6466,9.1126&zoom=7&opacity=80&layer=veg_fapar_proba.v.*_d&time=July')
system('curl -X POST --form "points=@test_points.geojson" --form
"layer=pnv_fapar_proba.v.jul_d_1km_s0..0cm_2014..2017_v0.1.tif"
https://landgisapi.opengeohub.org/query/points -o results.json')
df <- data.frame(matrix(unlist(rjson::fromJSON(file="results.json")),
ncol = 3, byrow = TRUE))
str(df)
plot(df[,2], df[,3], type="l")
## 255 is the missing value

Let me know if you experience any problems.

Tom Hengl
https://opengeohub.org/people/tom-hengl

On 1/15/19 10:55 PM, Tomislav Hengl wrote:

>
> We have recently released a webmapping system to serve global consistent
> environmental and Earth science layers at spatial resolutions from 10 km
> to 250 m (hopefully also soon at 100 m). This is an Open Data system as
> majority of layers are distributed under the Open Data Commons Open
> Database License (ODbL) and/or Creative Commons Attribution-ShareAlike
> 4.0 International license (CC BY-SA). Read more about LandGIS in:
> http://opengeohub.org/about-landgis.
>
> You can access the web app at:
>
> https://landgis.opengeohub.org
>
> The system currently (Jan 2019) serves about 300+ layers from relief and
> geology, to vegetation indices, climatic images, soil properties and
> classes and potential and actual vegetation. Complete overview of
> available layers is available at:
>
> https://github.com/Envirometrix/LandGISmaps
>
> In addition to the web-mapping app, data can be accessed using the:
>
> - Geonode installation at https://maps.opengeohub.org,
> - LandGIS REST API services at https://landgisapi.opengeohub.org,
> - LandGIS WCS at https://geoserver.opengeohub.org/landgisgeoserver/web/,
>
> A copy of all layers is also available via Zenodo.org i.e. via an unique
> URL.
>
> To access data at point locations best use the REST API. For example, to
> access monthly precipitations at a location X, Y you can use:
>
> https://landgisapi.opengeohub.org/query/point?lat=7.58033&lon=35.6561&coll=layers1km&regex=clm_precipitation_imerge.(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)_m_1km_s0..0cm_.*_v0.1.tif
>
>
> which returns a GeoJSON (table) with precipitation values in mm.
>
> To access values of LandGIS layers at multiple points you can use:
>
> curl -X POST --form "points=@test_points.geojson" --form
> "layer=pnv_fapar_proba.v.jul_d_1km_s0..0cm_2014..2017_v0.1.tif"
> https://landgisapi.opengeohub.org/query/points -o results.json
>
> where test_points.geojson is the input GeoJSON file containing
> coordinates of points. The multi-point access is currently limited to
> max 20 points, but we hope to increase this number gradually. More
> examples of how to construct spatial queries are available at:
> https://landgisapi.opengeohub.org
>
> In addition to the REST access, you can also access the LandGIS data
> using the Web Coverage Service (WCS) functionality of the Geoserver e.g.
> to subset layers using a bounding box. For example, to download surface
> temperature for July for an area of about 300 by 300 km you can use:
>
> https://geoserver.opengeohub.org/landgisgeoserver/ows?service=WCS&version=2.0.1& 
>
> request=GetCoverage&
> coverageId=layers1km:clm_lst_mod11a2.jul.day_m_1km_s0..0cm_2000..2017_v1.0&
> subset=Lat(41,45)&subset=Long(32,35)
>
> The read limit for WCS is 4GB and response size limit is 200MB. This
> means that WCS might fail if you try to fetch too large bounding boxes.
> If this happens we recommend instead downloading whole GeoTIFFs from
> Zenodo.
>
> We are currently preparing R functionality to allow users fetching data
> from LandGIS in a more systematic way (import, overlay, subset,
> plot...). If you would like to contribute to this initiative, especially
> to testing the R functions, please send me an email. Also we would
> appreciate if you would report any bug or inconsistency you discover
> via: https://github.com/Envirometrix/LandGISmaps/issues
>
> If you are currently producing any similar types of data (e.g.
> environmental layers at resolutions from 100 m to 1 km for a global land
> mask) and if you would like to publish this data on LandGIS, please
> forward a proposal for publishing your global layers:
> https://opengeohub.org/submitting-global-layers-inclusion-landgis and we
> will get on it asap.
>
> thank you,
>
> Tom Hengl
> https://opengeohub.org/people/tom-hengl

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo