18 February: fyi -- formal grammar and gardens, France &

Index of February 2010 | Index of year: 2010 | Full index


here is a postdoctoral fellow offer in computer science and
mathematical ecology (landscape modelling by the use of formal
grammars). This offer takes place into a French research project which
associates researchers from the diverse concerned specialities.

We are particularly interested in candidates with mathematical (formal
grammars), computational (java development) and ecological
(landscapes) skills, but we will examine every motivated proposition.

With all the best regards,

--
C?dric Gaucherel, INRA-AMAP (cedric.gaucherel@cirad.fr)
Marie-Laure Gu?not, LALIC / Universit? Paris-Sorbonne (mlguenot@gmail.com)

Postdoctoral fellow in mathematical ecology
The landscape language: a formal grammar developed and
implemented for landscape modelling.

Join an exciting team on a unique new project investigating landscape
dynamics and ecological modelling, based in Montpellier, South of
France. We are seeking enthusiastic candidates for a one-year
postdoctoral position studying landscape modelling using formal
language. Supervisor: Cédric GAUCHEREL (INRA - EFPA, UMR AMAP,
Montpellier) Context Patchy landscapes driven by human decisions
and/or natural forces are still a challenge to be understood and to be
modelled. Attempts to define a coherent framework such as a
mathematical description to derive landscape evolution rules are still
very scarce. This is the first objective of this postdoctoral
position. We propose to develop a new formalism of landscape modelling
based on rewriting rules belonging to the widest field of formal
grammars (Chomsky 1956). Patchy landscape units (patches) indeed may
be handled by a set of driving forces that are gradually changing
their states according to human decisions (e.g. agricultural of
forestry politics) or natural forcing (e.g. climate). We wish to adapt
formal grammar approaches to such specific two-dimensional mosaics. Do
have landscapes a common language? Which syntax would be the most
appropriate for rural landscapes? This work is based on the hypothesis
that landscapes have common features and dynamics that an adapted
language may capture. Such a language and stochastic rules should then
be used for landscape modelling and would greatly improve our
understanding of landscape dynamics. The second objective of this
work is to implement such grammar (reusing for instance existing
Relational Growth Grammar package such as GroImp (Kniemeyer et
al. 2008)), into an existing landscape modelling platform prototype
written in Java® and called DYPAL (Dynamic PAtchy Landscape). First,
this task supposes to develop a computing bridge between the grammar
editorial software and the existing prototype. This will require to
modify the existing platform kernel (based on landscape objects and
spatio-temporal operations (Gaucherel et al. 2006)) to make it
compatible with grammar rules. Finally, some well-chosen case-studies
will be implemented to illustrate new features of the system based on
various evolution scenarios with different impacts on landscape
patterns. The development of the platform will be performed in narrow
collaborations with agriculture and forestry specialists, providing
site data as well as detailed expertise about their studied
landscapes. Powerful landscape formalism as well as a user-friendly
modelling platform would emphasize the interest of formal grammar
approaches to model patchy landscapes, and hopefully other landscape
structures.



Materials

BiodivAgriM project data at landscape scales (this ~1 M project of the
ANR ­ Biodiversity call started in 2007 and is exploring landscape
modelling in order to preserve biodiversity in rural landscapes).
Searched skills

Qualifications:

A PhD doctorate in either applied mathematics and computing with some
biology/ecology skills or in biology/ecology with strong mathematical
skills would be appreciated. The candidate should demonstrate interest
in the broader questions in ecology and complexity theory studies and
a strong motivation. Research experience in landscape ecology, in
theoretical ecology, in stochastic processes and analytical/numerical
modelling are desirable. Employment Conditions: Initial appointment
would be for a period of one year, according to the INRA postdoctorat
salary (~ 25 k brut / yr). The starting date is negotiable between
January and March 1st, 2010. The postdoctorat may be transformed into
a more classical contract if necessary.

To Apply: Please send a single PDF file containing letter of
application with statement of interest, CV and two letters of
reference to Cédric Gaucherel (gaucherel at cirad.fr). For full
consideration, apply by March 1st, 2010.

References

ACL1. Chomsky, N. 1956. Three Models for the Description of
Language. IRE Transactions on Information Theory 2:113­124.

ACL2. Gaucherel, C., N. Giboire, V. Viaud, T. Houet, J. Baudry, and
F. Burel. 2006. A domain specific language for patchy
landscape modelling: the brittany agricultural mosaic as a
case study. Ecological Modelling 194:233-243.

ACL3. Kniemeyer, O., G. Barczik, R. Hemmerling, and
W. Kurth. 2008. Relational Growth Grammars - a parallel graph
transformation approach with applications in biology and
architecture. Pages 152-167 in A. Schürr, M. Nagl, and
A. Zündorf, editors. Applications of Graph Transformations
with Industrial Relevance AGTIVE'07, International Workshop
Kassel, Lecture Notes in Computer Science 5088. Springer,
Berlin

Index of February 2010 | Index of year: 2010 | Full index