Skip to content
/ xylookup Public

Webservice for looking up environmental and societal spatial information based on coordinates and optionally time and depth.

License

MIT license
0 stars 1 fork Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

iobis/xylookup

BranchesTags

Repository files navigation

xylookup

Webservice for looking up environmental and societal spatial information based on coordinates.

API

Documentation

https://iobis.github.io/xylookup/

Get all data for a point as json

https://api.obis.org/xylookup?x=2.90&y=51.2

Get all data for multiple points as json

https://api.obis.org/xylookup?x=2.90,2.1&y=51.2,51.1

Filtering the results can be done by excluding the different categories(areas, shoredistance, grids) e.g. to return results without the distance to the shoreline:

https://api.obis.org/xylookup?x=2.90&y=51.2&shoredistance=0

Note that msgpack data can also be used for sending/receiving data.

In R: obistools

devtools::install_github('iobis/obistools')

library(obistools)

?lookup_xy


xydata <- lookup_xy(abra, shoredistance = TRUE, grids = TRUE, areas = TRUE)
head(xydata)
shoredistance sstemperature sssalinity bathymetry                                                                                                   final_grid5
1            30      10.28631   34.76271       -4.0 United Kingdom, United Kingdom, F, T, eez, eez, United Kingdom: all, United Kingdom, United Kingdom: all, 221
2          1080      10.33242   34.90622       61.4 United Kingdom, United Kingdom, T, F, eez, eez, United Kingdom, United Kingdom: all, 221, United Kingdom: all
3          1184      10.72199   34.88896      122.2 United Kingdom, United Kingdom, T, F, eez, eez, United Kingdom, United Kingdom: all, 221, United Kingdom: all
4           290      10.79197   34.29342       20.6 United Kingdom, United Kingdom, F, T, eez, eez, United Kingdom: all, United Kingdom, United Kingdom: all, 221
5           259      10.72199   34.88896       51.0 United Kingdom, United Kingdom, F, T, eez, eez, United Kingdom: all, United Kingdom, United Kingdom: all, 221
6           506      10.77101   34.30700       32.4 United Kingdom, United Kingdom, F, T, eez, eez, United Kingdom: all, United Kingdom, United Kingdom: all, 221

In python: pyxylookup

Documentation

http://pyxylookup.readthedocs.io/en/latest/

Installation

pip install git+https://github.com/iobis/pyxylookup.git#egg=pyxylookup
import pyxylookup as xy

xy.lookup([[120,0], [-170,1]])

Dependencies

  • PostgreSQL and PostGIS (currently tested on 9.6 and 9.4)
  • Python 2.7 or 3.6 with
    • falcon
    • numpy
    • msgpack
    • psycopg2
    • gunicorn (webserver)
    • gdal (for the datapreparation scripts, not for the service itself)
    • test dependencies:
      • pytest
      • pytest-profiling
  • nginx

Deployment

  • Install postgresql
    • Load the needed areas tables into PostgreSQL
      • Currently only final_grid5 (see below)
  • Install nginx
  • Install python
  • Copy the source code from the service directory
  • Copy the datadir
  • Start gunicorn
    • gunicorn --reload app:api
    • gunicorn --pythonpath /usr/local/bin/python3 --timeout 120 --reload service.app:api
    • Preferably make sure that gunicorn restarts when the service restarts
  • Configure nginx so that it proxies all calls to gunicorn
  • Update service/config.py, set the datadir, PostgreSQL connection string and if needed the available areas.

Important commands

Manually running tests

python -m pytest

Starting the gunicorn service on the server

sudo systemctl start xylookup_service

Data pre-processing

Areas: final_grid5

Starting from the final.shp shapefile, perform the following steps:

  • Create grid layer in QGIS
QGIS => Vector => research tools => Vector grid
xmin -180.000001 ymin -90.000001
xmax 180.000001 ymax 90.000001
X: 5,000000030
Output grid as polygons
gridpoly_5.shp
  • Fix Antarctica
  • Intersect layer final and grid
QGIS => Vector => Geoprocessing => Intersect
final_fixed
gridpoly_5
final_fixed_grid5.shp
  • Upload gridded layer to PostgreSQL and apply fixes
shp2pgsql -I -s 4326 final_fixed_grid5.shp final_fixed_grid5 postgres > final_fixed_grid5.sql
psql -d xylookup -U postgres -p 5433 -f final_fixed_grid5.sql

ALTER TABLE final_fixed_grid5 DROP COLUMN sp_id;
ALTER TABLE final_fixed_grid5 DROP COLUMN gid;
ALTER TABLE final_fixed_grid5 DROP COLUMN marregion;
ALTER TABLE final_fixed_grid5 DROP COLUMN objectid_1;
ALTER TABLE final_fixed_grid5 DROP COLUMN mrgid;
ALTER TABLE final_fixed_grid5 DROP COLUMN iho_sea;
ALTER TABLE final_fixed_grid5 DROP COLUMN iho_id;
ALTER TABLE final_fixed_grid5 DROP COLUMN base;
ALTER TABLE final_fixed_grid5 DROP COLUMN iho_mrgid;
ALTER TABLE final_fixed_grid5 DROP COLUMN eez;
ALTER TABLE final_fixed_grid5 DROP COLUMN eez_id;
ALTER TABLE final_fixed_grid5 DROP COLUMN sovereign;
ALTER TABLE final_fixed_grid5 DROP COLUMN sov_id;
ALTER TABLE final_fixed_grid5 DROP COLUMN eez_mrgid;
ALTER TABLE final_fixed_grid5 DROP COLUMN longitude;
ALTER TABLE final_fixed_grid5 DROP COLUMN latitude;
ALTER TABLE final_fixed_grid5 DROP COLUMN area_m2;
ALTER TABLE final_fixed_grid5 DROP COLUMN layer;
ALTER TABLE final_fixed_grid5 DROP COLUMN path;
ALTER TABLE final_fixed_grid5 DROP COLUMN __xmin;
ALTER TABLE final_fixed_grid5 DROP COLUMN __xmax;
ALTER TABLE final_fixed_grid5 DROP COLUMN ymin;
ALTER TABLE final_fixed_grid5 DROP COLUMN ymax;

ALTER TABLE final_fixed_grid5 ADD COLUMN geog geography;
UPDATE final_fixed_grid5 SET geog = geom::geography;
CREATE INDEX final_fixed_grid5_geomix ON public.final_fixed_grid5 USING GIST (geom);
CREATE INDEX final_fixed_grid5_geogix ON public.final_fixed_grid5 USING GIST (geog);

update final_fixed_grid5 as f set id = t.id from (select row_number() over (order by name, country, type) as id, name, country, type 
from final_fixed_grid5 group by name, country, type order by name, country, type) t where f.name = t.name;
ALTER TABLE final_fixed_grid5 ALTER COLUMN id TYPE integer;

select * from final_fixed_grid5

create table abnj_grid5 as (select * from final_fixed_grid5 where type = 'abnj');
delete from final_fixed_grid5 where type = 'abnj';
CREATE INDEX abnj_grid5_geomix ON public.abnj_grid5 USING GIST (geom);
CREATE INDEX abnj_grid5_geogix ON public.abnj_grid5 USING GIST (geog);

Areas: EBSA

  • Fix EBSA layer by applying zero distance buffer
  • Intersect EBSA layer and grid
QGIS => Vector => Geoprocessing => Intersect
ebsa
gridpoly_5
ebsa_grid5.shp
  • Upload gridded layer to PostgreSQL and apply fixes
shp2pgsql -I -s 4326 ebsa_grid5.shp ebsa_grid5 postgres > ebsa_grid5.sql
psql -d xylookup -U postgres -p 5433 -f ebsa_grid5.sql

ALTER TABLE ebsa_grid5 DROP COLUMN ebsa_id;
ALTER TABLE ebsa_grid5 DROP COLUMN global_id;
ALTER TABLE ebsa_grid5 DROP COLUMN area_mw_km;
ALTER TABLE ebsa_grid5 DROP COLUMN gid;
ALTER TABLE ebsa_grid5 ADD COLUMN geog geography;
UPDATE ebsa_grid5 SET geog = geom::geography;
CREATE INDEX ebsa_grid5_geomix ON public.ebsa_grid5 USING GIST (geom);
CREATE INDEX ebsa_grid5_geogix ON public.ebsa_grid5 USING GIST (geog);

update ebsa_grid5
set id = 10000 + t.row_number
from (
	select name, row_number() over (
		order by name
	) from ebsa_grid5
	group by name
	order by name
) as t where t.name = ebsa_grid5.name

Areas: MWHS

  • Fix MWHS layer by applying zero distance buffer
  • Intersect MWHS layer and grid
  • Upload gridded layer to PostgreSQL and apply fixes
shp2pgsql -I -s 4326 mwhs_grid5.shp mwhs_grid5 postgres > mwhs_grid5.sql
psql -d xylookup -U postgres -p 5433 -f mwhs_grid5.sql

ALTER TABLE mwhs_grid5 DROP COLUMN longitude;
ALTER TABLE mwhs_grid5 DROP COLUMN latitude;
ALTER TABLE mwhs_grid5 DROP COLUMN sph_area;
ALTER TABLE mwhs_grid5 DROP COLUMN source;
ALTER TABLE mwhs_grid5 DROP COLUMN id;
ALTER TABLE mwhs_grid5 DROP COLUMN gid;
ALTER TABLE mwhs_grid5 ADD COLUMN geog geography;
UPDATE mwhs_grid5 SET geog = geom::geography;
CREATE INDEX mwhs_grid5_geomix ON public.mwhs_grid5 USING GIST (geom);
CREATE INDEX mwhs_grid5_geogix ON public.mwhs_grid5 USING GIST (geog);

ALTER TABLE mwhs_grid5 add COLUMN id integer;
update mwhs_grid5 set id = 20000 + refid;

ALTER TABLE public.mwhs_grid5 RENAME COLUMN full_name TO "name";

Areas: IHO

  • Fix IHO layer by applying zero distance buffer
  • Simplify
  • Intersect IHO layer and grid
  • Upload gridded layer to PostgreSQL and apply fixes
shp2pgsql -I -s 4326 iho_grid5.shp iho_grid5 postgres > iho_grid5.sql
psql -d xylookup -U postgres -p 5433 -f iho_grid5.sql

ALTER TABLE iho_grid5 DROP COLUMN id;
ALTER TABLE iho_grid5 DROP COLUMN gid;
ALTER TABLE iho_grid5 DROP COLUMN longitude;
ALTER TABLE iho_grid5 DROP COLUMN latitude;
ALTER TABLE iho_grid5 DROP COLUMN area;
ALTER TABLE iho_grid5 DROP COLUMN id_2;
ALTER TABLE iho_grid5 DROP COLUMN min_x;
ALTER TABLE iho_grid5 DROP COLUMN min_y;
ALTER TABLE iho_grid5 DROP COLUMN max_x;
ALTER TABLE iho_grid5 DROP COLUMN max_y;
ALTER TABLE iho_grid5 DROP COLUMN ymin;
ALTER TABLE iho_grid5 DROP COLUMN ymax;
ALTER TABLE iho_grid5 DROP COLUMN __xmin;
ALTER TABLE iho_grid5 DROP COLUMN __xmax;

ALTER TABLE iho_grid5 ADD COLUMN geog geography;
UPDATE iho_grid5 SET geog = geom::geography;
CREATE INDEX iho_grid5_geomix ON public.iho_grid5 USING GIST (geom);
CREATE INDEX iho_grid5_geogix ON public.iho_grid5 USING GIST (geog);

ALTER TABLE iho_grid5 add COLUMN id integer;
update iho_grid5 set id = 30000 + mrgid;

ALTER TABLE iho_grid5 DROP COLUMN mrgid;

Areas: LME

  • Fix LME layer by applying zero distance buffer
  • Intersect LME layer and grid
  • Upload gridded layer to PostgreSQL and apply fixes
shp2pgsql -I -s 4326 lme_grid5.shp lme_grid5 postgres > lme_grid5.sql
psql -d xylookup -U postgres -p 5433 -f lme_grid5.sql

ALTER TABLE lme_grid5 DROP COLUMN gid;
ALTER TABLE lme_grid5 DROP COLUMN objectid;
ALTER TABLE lme_grid5 DROP COLUMN shape_leng;
ALTER TABLE lme_grid5 DROP COLUMN shape_area;
ALTER TABLE lme_grid5 DROP COLUMN lat;
ALTER TABLE lme_grid5 DROP COLUMN lon;
ALTER TABLE lme_grid5 DROP COLUMN id;
ALTER TABLE lme_grid5 DROP COLUMN __xmin;
ALTER TABLE lme_grid5 DROP COLUMN __xmax;
ALTER TABLE lme_grid5 DROP COLUMN ymin;
ALTER TABLE lme_grid5 DROP COLUMN ymax;

ALTER TABLE lme_grid5 ADD COLUMN geog geography;
UPDATE lme_grid5 SET geog = geom::geography;
CREATE INDEX lme_grid5_geomix ON public.lme_grid5 USING GIST (geom);
CREATE INDEX lme_grid5_geogix ON public.lme_grid5 USING GIST (geog);

ALTER TABLE lme_grid5 add COLUMN id integer;
update lme_grid5 set id = 40000 + lme_number;

ALTER TABLE public.lme_grid5 RENAME COLUMN lme_name TO "name";

Shore distance

All data used was downloaded from Open Street Map.

In order to calculate the shoredistance we need two parts:

  1. water polygons in the database which are used to test whether points are on land
    • download water polygons
    • simplify
      • ogr2ogr water_polygons0_00005.shp water_polygons.shp -simplify 0.00005
    • upload to PostgreSQL
      • shp2pgsql -s 4326 water_polygons0_00005 public.water_polygons0_00005 > water.sql
      • psql -d xylookup -U postgres -p 5433 -f water.sql
  2. coastlines as geojson that are loaded in the python service for calculating the shoredistance
    • download coastlines
    • import coastline-split-4326 as table coastlines in database xylookup using the QGIS DBManager because the dbf file gives errors with shp2pgsql
shp2pgsql -I -s 4326 lines.shp coastlines postgres > coastlines.sql
psql -d xylookup -U postgres -p 5433 -f coastlines.sql

Then execute the following sql code in PostgreSQL:

CREATE OR REPLACE FUNCTION simplifyutm(geom geometry, tolerance float, max_segment_length float)
   RETURNS geometry AS
$BODY$
DECLARE
     bestUTM integer;
     geomCenter geometry;
     geomUTM geometry;
     geomBack geometry;
BEGIN
     geomCenter:= ST_centroid(geom);
     bestUTM:= floor((ST_X(geomCenter)+180)/6)+1;
     IF (ST_Y(geomCenter))>0 THEN
        bestUTM:= bestUTM+32600;
     ELSE
        bestUTM:= bestUTM+32700;
     END IF;
     geomUTM:= ST_Transform(geom,bestUTM);
     geomUTM:= ST_SimplifyPreserveTopology(geomUTM, tolerance);
     geomUTM:= ST_Segmentize(geomUTM, max_segment_length);
     geomBack:= ST_Transform(ST_setsrid(geomUTM, bestUTM),4326); -- set_srid as it is sometimes lost for some features ...
     RETURN geomBack;
END
$BODY$
   LANGUAGE plpgsql STABLE STRICT;

SELECT AddGeometryColumn('coastlines', 'geom_simple50', 4326, 'LINESTRING', 2);
UPDATE coastlines SET geom_simple50 = simplifyutm(geom, 50, 500);
COPY (SELECT st_asgeojson(geom_simple50) FROM coastlines) TO '<datadir>/coastlines50.jsonlines';

Rasters

Run dataprep/rasters.py, it prepares both the data and metadata needed. Data is prepared by storing them as uncompressed binary numpy array files which are later on read by using memorymapped files.

Some input source data will have to be downloaded manually such as the EMODnet and GEBCO bathymetry.

xylookup OpenAPI Specification

Build Status

Links

Warning: All above links are updated only after Travis CI finishes deployment

Working on specification

Installation

  1. Install Node JS
  2. Clone repo and cd
    • Run npm install

Usage

  1. Run npm start
  2. Checkout console output to see where local server is started. You can use all links (except preview) by replacing https://iobis.github.io/xylookup/ with url from the message: Server started <url>
  3. Make changes using your favorite editor or swagger-editor (look for URL in console output)
  4. All changes are immediately propagated to your local server, moreover all documentation pages will be automagically refreshed in a browser after each change TIP: you can open swagger-editor, documentation and swagger-ui in parallel
  5. Once you finish with the changes you can run tests using: npm test
  6. Share you changes with the rest of the world by pushing to GitHub

Data references and acknowledgements

Claus S., N. De Hauwere, B. Vanhoorne, F. Souza Dias, P. Oset García, F. Hernandez, and J. Mees (Flanders Marine Institute) (2017). MarineRegions.org.

The GEBCO_2014 Grid, version 20150318, http://www.gebco.net.

EMODnet Bathymetry Consortium (2016): EMODnet Digital Bathymetry (DTM). http://doi.org/10.12770/c7b53704-999d-4721-b1a3-04ec60c87238.

Tyberghein L, Verbruggen H, Pauly K, Troupin C, Mineur F, De Clerck O (2012) Bio-ORACLE: A global environmental dataset for marine species distribution modelling. Global Ecology and Biogeography, 21, 272–281.

Map data copyrighted OpenStreetMap contributors and available from https://www.openstreetmap.org.

About

Webservice for looking up environmental and societal spatial information based on coordinates and optionally time and depth.

Resources

Readme

License

MIT license
Activity
Custom properties

Stars

0 stars

Watchers

3 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages