Foss4g.be

From OpenStreetMap Wiki
Jump to: navigation, search

Foss4g.be is the yearly event from the OsGeo.be community. It is a day of presentations and workshops, focusing on open source geo applications. The match with OpenStreetMap is obvious, and there have been many OSM-inspired sessions in the past. In 2017, there will be a full OSM track for the first time.

You can use the space below to elaborate on your proposed OSM ideas, though you should also register through the website (before Sept 4th 2017).

OSM Landuse: Land-use/land-cover in Belgium

What do you mean by land-use / land-cover?

Land-use refers to the utilisation of the land by humans, land-cover refers to the actual occupation of the land. For instance, the land-use term "forest" denotes a land that is usually used to produce timber wood, but also used for recreational purposes, tourism, outdoor sports, and/or nature conservation, etc, while the land-cover term "woodland" denotes a land that is covered by trees. A `woodland` (landcover) can actually be a `park` (landuse). After a clear-cut, a `forest` (landuse) has no more trees and is a `shrub` (landcover). A `residential` area (landuse) can be a `woodland`.

How do we map land-use / land-cover features in OSM?

Long story. There are [landuse=\*](https://wiki.openstreetmap.org/wiki/Key:landuse) and [landcover=\*](https://wiki.openstreetmap.org/wiki/Key:landcover) keys in OSM but actually many other tags map land-use/land-cover features, as [natural=\*](https://wiki.openstreetmap.org/wiki/Key:natural), [leisure=\*](https://wiki.openstreetmap.org/wiki/Key:leisure), or even barrier=hedge, water=pond,... See the [osmlanduse.org](http://osmlanduse.org) for an implementation of a landuse/landcover map based on OSM.

Draft of the program

  • State of the land-use

We plan to present some statistics about land-use/land-cover mapping in OSM in Belgium. How far is our country from being complete? How does it compare against authoritative land-use dataset? How land-use/land-cover features are typically handled by Belgian mappers?

  • Round table

The goal of this round table is to gather everyone inside and outside the OSM community who is specifically interested in the land-use and land-cover mapping and data consuming. A perspective of this workshop is to get some kinds of Belgian convention to map and improve the mapping of land-use and land-cover features. We believe that global OSM conventions must sometimes be adapted/interpreted to local conditions. It is not only about how to tag but also how to map, e.g., how to draw continuous but parcelized land-use.

  • External contributions

We'd hear some people professionally involved in the production of land-use/land-cover maps or in the consumption of land-use data. What are their practices? What do they think about OSM data in this context?

1. Julien Radoux (UCLouvain), producer of a land-cover map (lifewatch.be) (TBC)

Abstract submission

Title: State of land-use/land-cover mapping in Belgium in OpenStreetMap

By allowing its contributors to define themselves the classes, OpenStreetMap (OSM) has a colorful but challenging data model. Being a generalist geographical database, OpenStreetMap does support the description of land-use and land-cover (LULC) features, but clear specifications of how to map LULC features were not designed at the early stage of the OSM history. In particular, while the landuse and landcover keys exist, there are a lot of other tags that are actually used for describing LULC features: natural=*, leisure=*, water=*, barrier=hedge, etc. Like for other kinds of geographic objects, there is a need of local conventions with respect to the local LULC characteristics for guiding the local OSM communities in their mapping.

The aim of this talk is to discuss LULC mapping in OSM in Belgium. We present an overview of how such features are mapped in Belgium, both in terms of geometry and classification (tagging): How far is our country from being complete? How does it compare against authoritative land-use dataset? How LULC features are typically handled by Belgian mappers? Then, we invite the OSM community to discuss LULC mapping in Belgium: What are the best practices? Should we import land-use open data into OSM in Belgium? How this data can be exploited by end-users? This process would ultimately ends up with the redaction of some mapping conventions at the Belgium OSM wiki.