INRA

Lancer l'impression

Fermer cette fenêtre

Pascal MONESTIEZ


Directeur de Recherche

 

  • ADRESS:

Email : Pascal.Monestiez@avignon.inra.fr
Tél: +33 (0)6.85.92.66.67


Research interests / Recent publications / PhD students / Teaching / Others

  • RESEARCH INTERESTS:

Methodological developments in Geostatistics :
    • Non-Gaussian variables (count data, binomial data, risk) and spatial hierarchical models
    • Modelling of non-stationarity (in covariance)
    • Spatial domains which differ from the plane : trees structures, river networks
    • Multivariate spatial modelling Multi-scale modelling in ecology

ANR projects :
    • IPSOS SEAL  (ANR VMC) 2008 -2011 : Investigation of the vulnerability of the biological Productivity of the Southern Ocean Subsystems to climate change: the Southern Elephant seal Assessment from mid to high Latitudes. Christophe Guinet, CEBC, CNRS
    • DIGISOL HYMED  Projet ANR Blanche  2009 - 2011: Cartographie Numérique des Sols par imagerie hyperspectrale pour la modélisation environnementale en région méditerranéenne ( http://www.umr-lisah.fr/digisolhymed/ ) Philippe Lagacherie, LISAH, INRA

Geostatistics for count data and spatial distributions of abundance :

Spatial hierarchical Models Count data especially when they contain a large proportion of zeros, are always a methodological problem for geostatistics as well as for any other spatial data modelling.
We propose a geostatistical approach that models the Poisson distributional nature of the counting process and significantly improves experimental variogram estimation, local mean prediction and error of prediction compared to classical ordinary kriging. This “Poisson kriging” is applied to fin whale count data recorded for 1993 to 2001 in the Northwestern Mediterranean Sea. Our approach is compared to Bayesian approaches introduced by Diggle et al. and developed by O. Christiensen, C Wilkle and others.

Bellier, E., Monestiez, P. Certain, G. Chadoeuf, J. and V. Bretagnolle 2009. Modeling species distribution with count data: nested variography accounts for the hierarchical structure of animal distributions (submitted)

E. Bellier and P. Monestiez, 2008. Model based block kriging to estimate spatial abundance of wildlife populations from count data. In Proceedings of the Eighth International Geostatistics Congress. Julian Ortiz & Xavier Emery (Eds.). GECAMIN. pp 119-128. ISBN:978-956-8504-18-2.

Monestiez, P., Dubroca, L., Bonnin, E., Durbec, J-P., Guinet, C., 2006. Geostatistical modelling of spatial distribution of Balaenoptera physalus in the Northwestern Mediterranean Sea from sparse count data and heterogeneous observation efforts. Ecological Modelling 193:615-628.

Monestiez, P., Dubroca, L., Bonnin, E., Durbec, J-P., Guinet, C., 2005. Comparison of model based geostatistical methods in ecology : application to fin whale spatial distribution in northwestern Mediterranean Sea. In Geostatistics Banff 2004, Leuangthong, O. and Deutsch, C. V. Eds, Springer, 777-786.

Geostatistics on tree structures, river networks

Classical geostatistical modelling can be extended to spatial supports as directed trees and applied to variables spatially structured along a river network. The topological properties of the tree structure imply specific model assumptions, proper variogram estimates and ad-hoc drift models. Once the covariance structure specified on the tree, kriging can be generalized for prediction. We also proposed a model-based simulation procedure to generate random functions on directed trees. Two case studies, fluvisol delineation for the Hérault river (South of France) and network properties of rural catchments, were explored. Clear spatial structures were exhibited and modelled using variogram based on upstream-downstream distance for both hydrographic networks. A drift was also modelled in both cases. This approach is presently extended to risk modelling of a disease whose pathogen is spread by water flow along a river network.

Vincent Garreta, Pascal Monestiez and Jay M. Ver Hoef, 2009. Spatial modelling and prediction on river networks: up model, down model or hybrid ? Environmetrics (on line)

Monestiez, P., Bailly J-S., Lagacherie P., Voltz M., 2005. Geostatistical modelling of spatial processes on directed trees : Application to fluvisol extent. Geoderma 128:179-191.

Bailly, J-S., Monestiez, P., Lagacherie, P., 2006. Modelling spatial processes on directed trees with geostatistics : Application to artificial network properties on rural catchment areas. Mathematical Geology 38: 515-539..

.

  • RECENT PUBLICATIONS (2005-2009):
Nerini, D., Monestiez, P. and Manté, C., 2009. Cokriging for spatial functional data. Journal of Multivariate Analysis (on line) *

Garreta, V., Monestiez P. and Jay M. Ver Hoef, 2009. Spatial modelling and prediction on river networks: up model, down model or hybrid ? Environmetrics (on line) *

Bellier, E., Monestiez, P. Durbec, J-P. and Candau, J-N., 2007. Identifying spatial relationships at multiple scales : Principal coordinates of neighbour matrices (PCNM) and geostatistical approaches. Ecography Ecography 30, 385–399. *

Bailleul, F., Charrassin, J-B, Monestiez, P., Roquet, F., Biuw, M.. and Guinet, Ch., 2007 - Successful foraging zones of southern elephant seals from Kerguelen Islands in relation to oceangraphic conditions.  Phil. Trans. R. Soc. B 362, 2169-2181. *

Bailly, J-S., Monestiez, P. and Lagacherie, P., 2006 - Modelling spatial processes on directed trees with geostatistics : Application to artificial network properties on rural catchment areas. Mathematical Geology 38: 515-539. *

Gaba S., Chadoeuf J., Monestiez P., Sauve C., Cortet J. and Cabaret J., 2006. Estimation of abomasum strongyle nematode infections in sheep at necropsy: Tentative proposals for a simplified technique. Veterinary Parasitology 140: 105-113.

Monestiez, P., Dubroca, L., Bonnin, E., Durbec, J-P., Guinet, C., 2006. Geostatistical modelling of spatial distribution of Balaenoptera physalus in the Northwestern Mediterranean Sea from sparse count data and heterogeneous observation efforts. Ecological Modelling 193:615-628. *

Chadoeuf, J., Goulard M., Lefèvre F., Monestiez, P., Pichot C., 2006. Poisson non-stationnaire et échantillonnage semi-raréfié; comment les cèdres du Luberon envahissent-ils l'espace ? In Analyse statistiques des données spatiales. Saporta G., J.-J. and Lejeune M. Eds., Editions Technip, Paris.

Monestiez, P., Dubroca, L., Bonnin, E., Durbec, J-P., Guinet, C., 2005. Comparison of model based geostatistical methods in ecology : application to fin whale spatial distribution in northwestern Mediterranean Sea. In Geostatistics Banff 2004, Leuangthong, O. and Deutsch, C. V. Eds, Springer, 777-786. *

Monestiez, P., Bailly J-S., Lagacherie P., Voltz M., 2005. Geostatistical modelling of spatial processes on directed trees : Application to fluvisol extent. Geoderma 128:179-191. *

Desassis, N., Monestiez, P., Bacro, J-N., Lagacherie, P., Robbez-Masson, J-M., 2005. Mapping unobserved factors on vine plant mortality. In Geostatistics for environmental applications, Renard Ph. et al. Eds, Springer, 125-136.

Harmelin-Vivien M.L., Bitar G. , Harmelin J-G., Monestiez P., 2005. Importance of Lessepsian Red Sea immigrant species in the littoral fish community of the Lebanese rocky coast (eastern Mediterranean Sea). Biological invasions 7:625-637. *

(*) For some references above, the pdf files cannot be available on Web and downloaded due to copyright policy, but they can be sent directly by the author. Ask for them at pascal.monestiez@avignon.inra.fr


  • SUPERVISION OF PhD STUDENTS:
    • Nicolas DESASSIS, PhD student until Oct. 2007

Thesis title : Modélisation et inférence du lien entre deux variables à partir d'observations géoréférencées et hétérotopes.

Thesis defended on October 23rd, 2007 at Montpellier, Statistics, Applied Mathematics, University of Montpellier II
Jury : Gilles Ducharme, Olivier Perrin (rapporteur), Carlo Gaetan (rapporteur), Liliane Bel, Michel Goulard, Jean-Noël Bacro, Pascal Monestiez

Presently : Research Scientist at the Center of Geostatistics, Ecole des Mines de Paris - Paris Tech
Adress: Centre de Géosciences, 35, rue Saint-Honoré,  77305 Fontainebleau cedex, France
 

Thesis title : Spatial distribution within hierarchical patch dynamics systems: identifying species-environment spatial relationships at multiscales.

French title: Distributions et structures spatiales au sein de systèmes en patchs hiérarchiques dynamiques: mise en évidence des relations spatiales espèces-environnement à échelles multiples.

Thesis defended on November 20th, 2007 at Avignon, Ecology, Environmental Sciences, Université de la Méditerranée, Marseille
Jury: Jean-Pierre Durbec, Marie-Josée Fortin (rapporteur), Christophe Guinet, Pascal Monestiez, Pierre Petitgas, Yvan Lagadeuc, Hans Wakernagel (rapporteur)


Presently :  Research Scientist at the Norwegian Institute for Nature Research
NINA Postal address: NO-7485 Trondheim, NORWAY Delivery/Visiting address: Tungasletta 2, NO-7047 Trondheim, Norway
Email: edwige.bellier@nina.no • Phone: +47 73 80 14 00  •  Fax: +47 73 80 14 01 • www.nina.no


SUPERVISION OF MASTER STUDENTS (2002-2009):


2009 : Benoît PAKLEPA : Monitoring à long terme de la distribution spatiale du rorqual commun en Méditerranée nord-occidentale : données multi-sources et aspects méthodologiques. Univ. de la Méditerranée.


2007 : Philipp NEUBAUER : Quantification par un modèle bayésien spatial de l’impact d’une réserve marine sur les populations et sur les captures par pêche en fonction de l’éloignement de la zone protégée. Univ. de la Méditerranée (PhD student at the Victoria University of Wellington, NZ)

 

2006 : Vincent GARRETA : Modélisation statistique et distribution spatiale de variables environnementales le long d'un système hydrographique. Univ de Montpellier II. PhD student at the CEREGE, Aix-Marseille ( www.cerege.fr )

 

2005 : Stéphanie PEDRON : Distribution spatiale des éléphants de mer (Minounga leonida) en relation avec les conditions océanographiques de l'océan austral. Univ. de la Méditerranée.

 

2003 : Nicolas DESASSIS : Modélisation et analyse statistique des structures spatiales des phénomènes de dégradation des sols et du dépérissement de la vigne à différentes échelles. Univ de Montpellier II. (Research scientist et the ENSMP, Paris tech).

 

2003 : Emilie BONNIN : Modélisation géostatistique spatio-temporelle de la répartition du rorqual commun (Balaenoptera physalus) de Méditerranée Nord-Occidentale, Univ. de la Méditerranée. (cf. Monestiez et al., 2006, Ecological Modelling)
 

  • TEACHING:
    • Master 2 de Biostatistique de Montpellier 2 (with Denis Allard) (2000 – 2004) (21h)
    • Master 2 d’Océanographie , OSU, Université de la Méditerranée, Aix-Marseille II, Spécialité : Biologie et écologie marine. UE 29 statistiques spatiales (2004 - 2009)  (30h - 6 credits)


  • OTHER ACTIVITIES:
    • Reflexives ® : core team member.
The Réflexives ® are an innovative methodological approach to develop cooperation for building research questions in real work situationsThe Réflexives ® seminars are spaces of dialogue, of reflective practice and openness which were created for and by INRA researchers; they have been organized for more than ten years within research departments and between departments to lay the foundations of a scientific and human cooperation among researchers, and between supervisors and PhD students. They are key events in the process when a PhD student must take ownership of its research subject and turn it into a research project, and develop into a researcher “by profession” ; they are also key in the training of supervisors who most of the time learn their trade from their peers. >> more on the web site http://www.reflexives-lpr.org/
    • Scuba Diving Instructor

Instructeur National #117 (Course Director) from the FFESSM (Fédération Française d'Etudes et de Sports Sous-Marins - French Underwater Federation).

 

Rédacteur : Biometrie
Date de création : 10/11/2006
Date de dernière mise à jour : 05/07/2009

INRA PACA. Domaine Saint Paul
Site Agroparc 84914 Avignon cedex 9
tél : (33) 04 32 72 20 00 - fax : (33) 04 32 72 20 42

copyright © 2008 | Crédits | Mentions légales