2020 / Day 03: polygons

Data

• Estonian Weather Service location-based forecast

Tools

Python, PostGIS, QGIS.

Summary

The dataset used for this is the same as for the previous lines-map only this time around looking at wind forecast. With wind we have to different parameters which should be shown at the same time: wind speed and wind direction. For this (think polygons!) I decided to draw ellipses around each hex bin centroid with their major-axis radii corresponding to wind speed + a little displacement from the centroid (trial-and-error with this) based on wind speed aswell and then - to show wind direction - rotate them in the direction the wind is blowing.

First a function in PostGIS to draw ellpises (off PostGISUsersWiki)

CREATE OR REPLACE FUNCTION ST_Ellipse(
  x double precision, y double precision,
  rx double precision, ry double precision,
  rotation double precision DEFAULT 0.0,
  quadSeg integer DEFAULT 8)
RETURNS geometry AS
$$
  SELECT ST_Translate( ST_Rotate( ST_Scale( ST_Buffer(ST_Point(0,0), 0.5, quadSeg), rx, ry), rotation), x, y)
$$
LANGUAGE 'sql';

and then

drop table if exists w.wind_bubbles;
create table w.wind_bubbles as
select
  oid, grid_oid,
  st_rotate(
    st_setsrid(
      st_ellipse(
        st_x(geom),
        st_y(geom),
        2500,
        2500 + (coalesce(wind_speed_mps,0)*1000),
        0,
        16
      ),
      3301
    ),
    -1*(radians(wind_direction_deg) + pi()),
    geom
  ) as geom,
  wind_speed_mps, wind_direction_deg, valid_from, valid_to
from (
  select
    oid, grid_oid,
    st_transform(
      st_project(
        grid::geography,
        coalesce(wind_speed_mps,0)*500,
        /* to get the angles correct. st_project goes anti-clockwise*/
        radians(coalesce(wind_direction_deg, 0))-pi()
      )::geometry(point, 4326),
      3301) as geom,
    wind_speed_mps, wind_direction_deg, valid_from, valid_to
  from
    w.phenomen
) f;

where w.phenomen.grid is the centroid of a hex bin cell and where w.phenomen.oid the identifier for this wind speed/direction forecast location instance for valid_from-valid_to interval.

And the result as seen with the help of QGIS TimeManager:

hi-res

One more giffing submission to #30DayMapChallenge for Day 3 call to polygons: the forecast wind speed and direction as (ellipsoidal) polygons (khm) stretched out in the wind direction. Data off Estonian Weather Service locality forecast (oct 29 - nov 01).

QGIS, PostGIS pic.twitter.com/dq4DJeHIFh

— Tõnis Kärdi (@tkardi) November 5, 2020