[OSGeo-Jobs] Fwd: Space-Time Modelling of Crop Rotation in Luxembourg (Luxembourg)

[Markus Neteler]

FYI


---------- Forwarded message ----------
From:  <jobs at tudor.lu>
Date: Mon, Jan 16, 2012 at 3:41 PM
Subject: Space-Time Modelling of Crop Rotation in Luxembourg (Luxembourg)
To: SJD-GEO at listserv.uni-heidelberg.de


SCIENCE-JOBS-DE Stellenausschreibung (vacancies)
----------------------------------------------------------------------
Wenn sie auf ein Job-Angebot antworten, achten sie bitte darauf, dass
die Antwort nicht an die Liste gerichtet ist.
If you reply to a job offering, please take care, that your reply is
not adressed to the list.
----------------------------------------------------------------------

Background and objectives

Within the project LUCAS (Indirect land use change effects in
consequential Life Cycle Assessment of bioenergy) there is the need to
determine the changes in the spatial patterns of land use induced by
an additional production of maize for biogas production.
The Luxembourgish agricultural and farming model being set up by CRTE
and the external contractor ORDECSYS will determine the new production
patterns coming from an economic equilibrium (opportunity cost
determination, revenue maximisation) at the farm level.
The goal is to add a spatial component to the optimization model
developed in LUCAS, to determine impacts of the induced land use
changes and to show the importance of accounting for spatial patterns
of land use change in crop production optimisation, i.e. where these
changes in land use patterns occur.
The objective of the internship is to analyse and model spatial
patterns of crop rotation for Luxembourg. This includes the production
of a consistent space-time data set on crop rotation of the
Grand-Duchy of Luxembourg using the agriculture reference parcel data
set of Luxembourg (FLIK) from the Agricultural Service of Luxembourg,
ASTA as well as the development of a simple spatial model to simulate
crop rotation accounting for crop specific environmental parameters.

Description of the internship

The trainee will follow a stepwise approach based on the following tasks:
1.      Develop an automated algorithm to clean geometries and
duplicates of existing agricultural plots and harmonise land register
and FLIK numbers in space and time.
2.      Develop an algorithm to estimate missing crops accounting for
neighbouring parcels. Generate typical crop rotations for Luxembourg.
Generate mixed crop plantation distributed on the same plot for the
same year.
3.      Build a spatial allocation model accounting for crop specific
information, such as soil properties, nutrients, fertiliser usage,
meteorological information and typical crop rotations, etc.
Depending on the skills of the trainee steps 1 and 2 should be
accomplished within the trainingship project. Step 3 can be added if
the trainee shows sufficient skills.
Various spatial methods will need to be considered to achieve the
outcomes of the project:
- Cleaning of geometries, removing of duplicates will be achieved by
standard GIS operations available in R, GRASS GIS and/or PostGIS;
- For harmonisation of land register and FLIK numbers matching
algorithms will need to be developed in R. The cleaned and harmonised
data set will be stored in a PostGIS data base;
- To estimate missing crop data on a per FLIK number basis, nearest
neighbour based algorithms could be used and/or developed, such as
simple neighbour statistics or more advanced methods such as Moran's
methods with the help of the spdep package;
- For the space-time analysis of typical crop rotation data and most
occurrences of specific crops in space and their evolution in time
standard statistical procedures such as cross-tables and principle
component analysis can be used to develop transition matrices for crop
rotation using R;
- To optimally allocate crops in space and time a spatial-temporal
algorithm will be built by using simplified probabilistic approaches
based on Markov chains relying on transition probabilities or more
advanced geostatistical methods.
The candidate is expected to start as soon as possible. The interview
process will continue until a suitable candidate will be found.

Profile and expected skills

The trainee will have a background in Geography, Geoinformatics,
Geomathematics, GIS or Spatial Statistics. He/she will work in an
interdisciplinary and multi-lingual/-cultural work environment and
therefore team spirit is expected.
Fluent knowledge of English is required. The successful candidate will
have some experience in:
- Programming in R;
- Spatial statistics and GIS;
- Knowledge on agricultural engineering and SQL are an asset.
The candidate is also expected to learn by doing.

Information

For further information about this internship, please contact:
- Ulrich Leopold, Phone: +352-425991-4618; Email: Ulrich.Leopold at tudor.lu
- Antonino Marvuglia, Phone: +352-425991-4652; Email:
Antonino.Marvuglia at tudor.lu
To apply for the internship position, please send your motivation
letter and CV to Mrs. Paula Couto at jobs at tudor.lu citing the title of
the advert.

----------------------------------------------------------------

Von:

Monika Grünwald
jobs at tudor.lu
CRP Henri Tudor
Luxembourg

Ansprechpartner: Paula Couto - jobs at tudor.lu

----------------------------------------------------------------

* science-jobs-de sublist: SJD-GEO