Georg-August University of Göttingen

Chair for Computer Graphics and Ecological Informatics

Büsgenweg 4, 37077 Göttingen, Germany

http://www.uni-goettingen.de/en/67072.html

 

 

                

               Tutorial and Workshop

               "Modelling with GroIMP and XL"

 

        combined with the 5th GroIMP user and developer meeting

 

 

Date:              27 – 29 February 2012

 

XL and GroIMP tutorial for beginners:                  27 Feb., 9:15 h – 15:00 h

Tutorial for advanced XL and GroIMP users:        27 Feb., 16:00 h – 28 Feb., 12:00 h

GroIMP user and developer meeting:                  28 Feb., 13:00 h – 14:00 h

Workshop dinner:                                             28 Feb., 19:00 h

Workshop:                                                       28 Feb., 14:00 h – 29 Feb., 12:00 h

 

 

Location:        University of Göttingen, Department Ecoinformatics, Biometrics and

                        Forest Growth, Büsgenweg 4, 37077 Göttingen, Germany

                   Room CIP I

 

 

 

Contact:  Winfried Kurth, Chair for Computer Graphics and Ecological Informatics

Georg-August University of Göttingen, Büsgenweg 4, 37077 Göttingen, Germany

Tel. +49–551–399715,  Fax +49–551–393465, E-mail wk(at)informatik.uni-goettingen.de

 

 

Programme

 

Monday, 27 February 2012

 

9:00     Opening, Introduction

9:10     W. Kurth: Introduction to rule-based programming, L-systems and XL  [pdf]

10:15   Coffee break

10:30   W. Kurth: Basic examples in XL (part 1)  [pdf]

11:30   R. Hemmerling and K. Smoleňová: How to model a daisy in 1/2 hour  [textures, archived files]

12:00   Lunch

 

13:30   W. Kurth: Basic examples in XL (part 2)  [pdf]

14:30   Simple branched structures and rendered plants modelled with XL (team work)

15:00   Coffee break  and/or Small team sessions

 

15:45   W. Kurth: Interpretive rules and instantiation rules  [pdf]

16:15   M. Henke: A closer look at some examples from the grogra.de gallery

16:45   W. Kurth and G. Buck-Sorlin: Using the radiation model of GroIMP  [pdf]

17:30   Using the radiation model (team work)

18:00   Questions and answers and/or Small team sessions

- ca. 18:30

 

 

Tuesday, 28 February

 

9:00     R. Hemmerling: The ODE framework of GroIMP

10:00   G. Buck-Sorlin: GPUFlux – a new radiation model using the GPU  [pdf]

10:30   Coffee break

10:45   Y. Ong:  MTG format, import and visualization in GroIMP, and statistical analysis

11:30   GroIMP user and developer meeting

12:30   Lunch

 

Workshop "Modelling with GroIMP and XL – recent results"

 

13:30   J. Evers: Simulating light interception in Australian wide-row wheat systems

14:15   V. Sarlikioti and L. Pagès: Exploring root developmental plasticity to soil nutrients in three-dimensional architectural modelling of the root system

15:00   Coffee break

15:15   R. Hemmerling: Specification of chemical formulæ in XL with operator overloading

16:00   T. W. Chen: Modelling cucumber plants (project outline)

16:15   K. Smoleňová, M. Henke and C. Ding: GreenLab in XL – usage and more

16:45   Coffee break

17:00   M. Henke and K. Smoleňová: Component-based modelling within GroIMP

17:30   G. Buck-Sorlin, P. Guillermin, M. Delaire, Fatoumata Sané, Christian le-Morvan: Towards a multi-scaled functional-structural model of apple, linking ecophysiology at the fruit and branch scales

- ca. 18:15

 

19:00   Workshop dinner @ Kartoffelhaus, Goetheallee 8

 

Wednesday, 29 February

 

9:00     F. Schöler: Introducing basic boolean set operations for GroIMP

9:30     P. Surový, A. Yoshimoto, N. A. Ribeiro: Comparison of pruning regimes for Stone pine (Pinus pinea L.) using a functional-structural plant model written in XL  [pdf]  (remote video / audio life presentation from Japan)

10:15   Coffee break

10:45   S. Paulus: The benefit of plant models for laser generated data  [pdf]

11:15   All: Conclusions and future projects

 

End of tutorials/workshop: 29 Feb., 12:00 h

 

 

 

 

Abstracts of workshop contributions

 

Gerhard Buck-Sorlin, Pascale Guillermin, Mickaël Delaire, F. Sané, Ch. le-Morvan (Agro­campus Ouest, Angers):

 

Towards a multi-scaled functional-structural model of apple, linking ecophysiology at the fruit and branch scales

 

A lot of data on eco-physiological processes in apple (Malus x domestica) exists, concerning various aspects of fruit growth and development, fruit quality, leaf photosynthesis etc. How­ever, despite the wealth of data and studies many processes leading to (interannual and intra-arboreal) heterogeneity in quantity of fruit production as well as fruit quality are only poorly understood at the branch level. Current Functional-Structural Plant Models of apple have targeted canopy architecture, i.e. development of vegetative structures. Here we will present a concept to apply the FSPM paradigm to the simulation of assimilation (source), transport and consumption (sink) of carbon in the context of a static structure representing the fruit-bearing branch.

 

Jochem Evers (Wageningen UR):

 

Simulating light interception in Australian wide-row wheat systems

 

To optimize the use of water of a wheat crop in water-limited environments, wheat can be sown in wide rows of closely spaced wheat plants. This is done to delay water use and avoid terminal water stress during grain filling. However, light interception in such systems differs from regular crop row distances. Therefore, to be able to simulate such wide-row systems using crop models such as APSIM, its light extinction routines need to accommodate alter­native crop configurations, such as wide rows. Here we show on-going work on using a GroIMP-based architectural model of wheat development to analyse the relationship between crop configuration and light interception. The aim of this work is to improve cal­culation of light interception in APSIM such that wide-row wheat systems can be simulated more accurately.

 

 

Reinhard Hemmerling (University of Göttingen):

 

Specification of chemical formulæ in XL with operator overloading

 

Operator overloading is a feature of many programming languages and allows to define the meaning of operator symbols on user-define types. In this presentation we will show how certain operators can be defined in XL that allow to parse chemical reactions, and then solve the reactions numerically. As an interesting side-effect, specification of chemical reaction equations can be seen as part of the XL language, and syntax checking of the compiler is applied to these equations. Two examples will be shown to demonstrate its use and to verify the results.

 

 

Michael Henke and Katarína Smoleňová (University of Göttingen):

 

Component-based modelling within GroIMP

A framework for interactive, visual modelling – new tools to tackle problems in plant science


Modelling has become an essential part of all kinds of scientific activities. In fact, it could be stated that modelling is the common way to obtain a better understanding of a system and problems related to it. The interactive modelling platform GroIMP is a general 3D-modelling software, which has been tailor-made for use in functional-structural plant modelling. By using the integrated language XL (a Java-based extended L-system language), various different models have been developed over the last years, yielding a pool of more or less specialized, openly accessible models, from which the user community could actually benefit directly. However, when applied to complex tasks, models still tend to develop in an in­transparent manner, rendering an efficient re-use of existing components next to impossible. Methods and tools which are commonly used in software engineering (e.g., component-based programming, graphical software visualization) can help to overcome this bottleneck.

 

 

Stefan Paulus (University of Bonn):

 

The benefit of plant models for laser generated data

 

Laserscanning of plants provides highly accurate resolution together with non-invasive measure­ments. The resulting point cloud can be used to extract parameters describing plant status in respect to, e.g., drought stress. Problems result from plant motion, dimension of deformation caused by growth and outer influences as well as from insufficient 3D cor­respondence of scan parts among each other. This complicates fusion of different sensors and scan parts and also comparison of data from different dates. Plant models give solutions to these problems by providing plant-applied coordinate systems that enable spatio-temporal registration.

 

 

Vaia Sarlikioti and Loïc Pagès (INRA Avignon):

 

Exploring root developmental plasticity to soil nutrients in three-dimensional architectural modelling of the root system

 

A simple dynamic root model proposed by Pagès (2011) was translated to the modelling language XL, using the GroIMP platform and was used as a base. The objectives of this work were to introduce a new simplified root architectural model that includes the recognized plasticity mechanisms of 1) Chemotropism, 2) Inter-Branch distance and 3) Branching Hierarchy and to investigate the effect of such mechanisms on root architecture and nutrient uptake for four different nutrient distributions in the soil.

 

 

Florian Schöler (University of Bonn):

 

Introducing Basic Boolean Set Operations for GroIMP

 

This talk will be about a technical approach to introduce basic Boolean set operations for the modelling platform GroIMP. Boolean set operations are important in many fields of appli­cation. Their results can, for example, be used to compute the volume of generated structures. Based on the volume, other physical properties like the weight or inertia, and also semantic properties like construction costs (e.g., in CAD) can be derived. It is shown how one can unify or intersect two or more geometric primitives and how one can compute their difference. The operations may be inserted as nodes into the RGG-Graph that automatically gather all accessible primitives. The geometric primitives are amended by a description of a closed surface as a set of triangles. As an example application, it is shown how one can compute the volume of complex objects generated by the Boolean set operations.

 

 

Katarína Smoleňová, Michael Henke and Cong Ding (University of Göttingen):

 

GreenLab in XL – usage and more

 

GreenLab is a functional-structural plant model that has already been applied to a number of various plant species. Plant organogenesis is desribed by the dual-scale automaton for­malism and growth is driven by source-sink relations between plant organs. It will be shown how the formalism of GreenLab and the growth process can be translated into the language XL. Currently several features of the deterministic and stochastic version of the GreenLab model are included.

To provide a user-friendly way for parameter input, an additionally implemented graphical user interface was developed as part of the modelling platform GroIMP. Supported is also import and export of parameter files (*.sci) and AMAPSymbol object shape files (*.smb), both being in use in the open-source GreenLab software called GreenSciLab.

 

 

Peter Surový, A. Yoshimoto and N. A. Ribeiro:

 

Comparison of pruning regimes for Stone pine (Pinus pinea L.) using a functional-structural plant model

 

Functional-Structural Plant Models (FSPM) are becoming important tools for modelling the structure and growth of plants, including complex organisms like trees. These models combine the advantages of empirical, mechanistic, and structural models to simulate the growth of individual plant structures (branches, buds, leaves, etc.). This approach enables realistic evaluation of the plant’s response – including changes in structure and growth – to different environmental conditions. We demonstrate the potential use of these models to evaluate individual tree growth under different management regimes (pruning). The data used in this study were obtained from 3-D measurements taken with a FASTRAK Polhemus digitizer, with specific attention given to bud creation and branching. Each branch segment was analyzed to estimate its age, enabling us to document annual structural changes. We use the XL programming language and a GroIMP environment to simulate and compare different pruning scenarios.