Curriculum Vitae
Education:
MSc. Zoology, University Ghent, 1983. MSc. Computer science, University Ghent, 1985. PhD in Zoology, University Ghent, Belgium, 1988.
Jobs:
Research fellow of Belgian national Fund for Scientific Research, University of Ghent, 1984 – 1988. (one year of maternal leave – enough to realise I wanted to stay in science 1989). Postdoctoral researcher Belgian national Fund for Scientific Research, Free University of Brussels, 1990– 1991. Several short-term grants (duration 3 to 9 months) at University Ghent, and NIOO-CEME,1991-1992. Longer-term EU-funded research 1993-1999. 1999-now: senior researcher at NIOO-CEME.
Guest professor at the university of Ghent and the Free university of Brussels - course in ecological modelling.
Expertise:
First research was on deep-sea nematodes, systematics to start with, then more ecologically oriented (1984-1988). The first postdoc brought an abrupt switch to estuarine zooplankton (1990-1991), then another abrupt switch to estuarine modelling (1991-1993) with a short intermezzo (9 months) on estuarine nematodes (1992).
The modelling microbe that hit me was fatal. From 1993 on most research included modelling: diagenetic modelling (1994-1998), water-column oriented modelling (1999-..) or food web box modelling (2002-..), with an occasional excursion to other types of models (e.g. plant modelling). Most fascinating I find how models can be used for the quantification of unmeasured things or to better understand the functioning of ecosystems. This means that although you will (almost) never find me in the lab/field, there exists a close coupling between my models and data. Nematodes and zooplankton remain fascinating and occasionally irresistible though.
Projects
R-projects
A couple of years ago we decided to use R for scientific computing, data analysis and graphics. We chose R because it is freely available (http://cran.r-project.org/), and rapidly expanding due to contributions of a large number of enthusiasts. After having worked with it for a couple of years now, I cannot imagine how I would cope without it. Admittedly, it takes a while -and a lot of initial frustration- to learn how to work with it, but it pays off ten times afterwards.
Freedom (to use and modify) was an important element to reach this decision. Although other programmes would have been equally -at the time perhaps more- suited, these were found to be too expensive. Many of my students at the university of Brussels come from third world countries, and it would not have been proper to teach them how to work with a software, that, when back into their country they cannot afford any more.
However, the consequence of choosing R was that, if there is some mathematical routine that we need and which is not part of R, I have to implement it. This however, is a small price to pay for free software.
By now, we have almost everything we need to solve our models. Below you will find more details on the packages that we created ...
HERMES; Hotspot Ecosystem Research on the Margins of European Seas
This is an integrated European research project designed to gain new insights into the biodiversity, structure, function and dynamics of ecosystems along Europe's deep-ocean margin. We are responsible for the modeling of the five main ecosystems studied: open slopes, canyons, coral reefs, cold seeps, and anoxic systems.
http://www.soc.soton.ac.uk/CHD/HERMES/
WESTBANKS: Trophic structure of the southern North Sea.
In collaboration with my former Belgian colleagues from the University of Ghent, we study the sediment communities in the Southern Bight of the North Sea. Our task is to ascertain how the sediments function both from a biogeochemical point of view and from the point of view of the animals living there. All the information gathered will be used as input to an inverse food web model for the quantification of food web flows. (OSTC-funded research, Belgium).
http://www.vliz.be/projects/westbanks
CoralFISH
This is a EU-funded project, dedicated to the role of deep-water corals in fisheries. It aims to support the implementation of an ecosystem-based management approach in the deep-sea by studying the interaction between cold-water coral habitat, fish and fisheries. Our task in this project is to develop coral ecosystem models that include fishes to better understand coral fish-carrying capacity.
Selected Publications
Rather than giving a list of all publications (can be found via the web of science), I prefer to select few of them...
Soetaert Karline and Peter M.J. Herman 2009
A Practical Guide to Ecological Modelling. Using R as a Simulation Platform
372 pp
Publisher: Springer
Full article: http://www.springer.com/life+sci/ecology/book/978-1-4020-8623-6
This book, which is based on lecture notes written in 2001, has been the major topic of my scientific work for at least a year.
Peter an me have written it because we felt that a comprehensive tutorial on how to make models, why they are useful and how to apply them did not yet exist. Modeling is not part of most biologists' basic curriculum, which is a shame. Because of that, we feel, they miss training in deductive science whilst they get an overdose of correlative and descriptive approaches. In the end then, our students may be experts in pattern recognition, but lack the basic skills of deducing the processes that lead to these patterns.
Modeling is a powerful way to clean up thinking. And it is fun to do! That is the basic message that we try to convey. We hope that this book will reduce the barrier for those that are willing to become a modeler but do not dear because they fear that it might be too complicated. If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K., van Oevelen, D., 2009.
Modeling food web interactions in benthic deep-sea ecosystems: a practical guide.
Oceanography (22) 1: 130-145.
Although we prefer to use dynamic mechanistic models, often this is not possible because of lack of data. Food webs for instance are often seriously under sampled, i.e. there are much too few data to fully quantify them. This paper deals with a technique, linear inverse methods, to overcome these limitations. They do not allow obtaining one food web, rather the food web flows can be quantified within ranges, or their probability density function assessed.
http://www.tos.org/oceanography/issues/issue_archive/issue_pdfs/22_1/22-...
Soetaert, K., Hofmann,A., Middelburg, JJ, Meysman, FJR, Greenwood, J. 2007
The effect of biogeochemical processes on pH
Marine Chemistry 105, 30-51
In this paper, the effect of various processes on pH is estimated. It is shown that some processes tend to always increase, while other tend to always decrease pH. Intriguingly, some processes converge to a specific pH, whilst others diverge from it.
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K., Middelburg, JJ, Heip, C, Meire, P., Van Damme, S., Maris, T. 2006
Long-term change in dissolved inorganic nutrients in the heterotrophic Scheldt estuary (Belgium, the Netherlands).
Limnol. Oceanogr. 51: 409-423
Full article: http://aslo.org/lo/toc/vol_51/issue_1_part_2/0409.pdf
Here we use > 40.000 datapoints from Dutch and Belgian sources to document long-term changes in the water quality of the Westerscheldt estuary. We show the improvement of the water quality, in terms of oxygen and nutrients, the changes in nutrient ratios with a doubling of the N:P ratio from 1980 to 2000, but also that the functioning the estuary as a filter has changed considerably during that period.
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K, Hoffmann, M., Meire, P., Starink, M., van Oevelen, D, Van Regenmortel, S., Cox, T. 2004.
Modeling growth and carbon allocation in two reed beds (Phragmites australis) in the Scheldt estuary (Belgium, The Netherlands)
Aquatic Botany 79: 211-234
A mathematical model is used to make a carbon budget for reed and to examine what determines the summer biomass reached by this plant. I have had reed in my garden for many years; it took the development of a model to finally understand why it grows so vigorously !
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K., Muthumbi, A., Heip, C. 2002
Size and shape of ocean margin nematodes: morphological diversity and depth-related patterns.
Mar. Ecol. Progr. Ser. 242: 179-193.
With this paper, my work on nematodes has come full circle. The first ecological paper I ever wrote (in 1989) dealt with nematode size and how these creatures become smaller with water depth. Almost 15 years later, new data from the NE Atlantic allowed us to finally understand why. In addition, we show that nematodes are either fat, short and ornamented or long, slender and smooth.
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K., P.M.J. Herman, J.J. Middelburg, C. Heip, C.L. Smith, P. Tett & K.Wild-Allen 2001
Numerical modelling the shelf break ecosystem: integrating pelagic and benthic field measurements.
Deep-Sea Res. II 48: 3141-3177
Here a 1-D pelagic model has been coupled to a 1-D diagenetic (sediment) model and the model fitted against a substantial pelagic and benthic data set from a shelf site (200m). Although benthic profiles, measured in spring and fall, indicated temporal constancy in the sediment, the model demonstrated substantial temporal variation. Apparently the best period for sampling oceanic sediments is winter!
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K. & P.M.J. Herman, 1995.
Nitrogen dynamics in the Westerschelde estuary (S.W. Netherlands) estimates by means of the ecosystem model MOSES.
Hydrobiologia 311(1-3): 225-246.
This work infected me with the modelling bug. One of the figures in this paper, which documents the sources, transformations and fate of nitrogen in the Westerschelde estuary (from Belgian toilets and Dutch cows to the Northsea and into the air) has found its way to text books of marine and estuarine ecology.
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K. & P.M.J. Herman 1994
One foot in the grave: zooplankton drift into the Westerschelde estuary (the Netherlands).
Mar. Ecol. Prog. Ser. 105: 19-29.
This is the only work where I used my own data to fit my own model. Amongst other things we calculated the fate of marine zooplankton in the Westerschelde estuary. Some 1500 tons dry weight of these animals enter the estuary each year where they die - this is the weight equivalent of 4000 Dutch cows !
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Soetaert, K. & M. Vincx 1987
Six new Richtersia species (Nematoda, Selachinematidae) from the Mediterranean.
Zoologica Scripta 16 (2): 125-142.
This was my first scientific paper! At that time the scientific tools that I used were the microscope and lots of Chinese ink (to draw the new species).
One of the new species in this paper was named after my husband. Consequently my children now have some tiny worm in a remote place that shares their name. By the way, they are NOT AT ALL impressed by that.
If available then a reprint or PDF can be requested at library@nioo.knaw.nl
Links
A Practical Guide to Ecological Modelling - Using R as as Simulation Platform
http://www.springer.com/life+sci/ecology/book/978-1-4020-8623-6
The official page to the book: Soetaert Karline and Peter M.J. Herman, 2009.
A Practical Guide to Ecological Modelling - Using R as a Simulation Platform.
Springer, ISBN 978-1-4020-8623-6
Using R for scientific computing - R lecture notes
http://cran.r-project.org/doc/contrib/Soetaert_Scientificcomputing.zip
This ZIP file contain lecture notes to introduce you in the wonderful world of R-computing. R is an open-source software, extremely useful for scientific computing.
Once a year, me and Filip Meysman (VUB) teach ecological modelling at the universities of Ghent and Brussels. We use R as a simulation environment to implement and solve these models. Before starting the ecological modelling lectures, we spend one day guiding our students through parts of these lecture notes. As such, the students get a 'flavor' of what R can do, and how to best work with the language.
The ZIP contains the notes "Using R for scientific computing", an R reference card and a data set.
R lecture notes - answers
http://cran.r-project.org/web/packages/marelac/vignettes/Answers.pdf
The answers to the questions in the "Using R for scientific computing" lecture notes.
It is probably tempting to download this file together with the R lecture notes, but it is much more efficient if you try the exercises first by yourself, without peeking at the answers!
R-package deSolve
http://cran.r-project.org/web/packages/deSolve/index.html
deSolve is an R-package, created with colleagues Woody Setzer from EPA (US) and Thomas Petzoldt from Univ. Dresden (Germany) that contains functions to solve differential equations in R.
Essentially, this is the R-solver for dynamic models. The package contains routines designed for solving ordinary differential equations, differential algebraic equations but also routines to solve uni-and multicomponent 1-D and 2-D reactive transport models.
R-package rootSolve
http://cran.r-project.org/web/packages/rootSolve/index.html
This R-package contains functions to find the roots of nonlinear equations, to perform equilibrium and steady-state analysis of ordinary differential equations.
It was developed to solve some of the problems that are discussed in the book "A Practical Guide to Ecological Modelling".
R-package limSolve
http://cran.r-project.org/web/packages/limSolve/index.html
Many models consist of simple linear equations. For instance, many of our food web models are linear. In addition, estimating the diet of a grazer based on the stable isotopic composition of the animal and its food, or the algal composition based on pigments also lead to simple linear models.
Package limSolve was created together with my colleagues Dick van Oevelen and Karel van den Meersche to solve such models. It may seem rather technical, but is includes some nice features, e.g. mimicking chemtax or isosource software.
R-package LIM
http://cran.r-project.org/web/packages/LIM/index.html
Wheresas package limSolve contains functions to solve linear models, very often the creation of the linear equation is problematic, especially for large food web models, or for large reaction networks.
R-package LIM contains software to create these equations, based on ascii files in which the problem is specified in a natural and comprehensible way.
R-package LIM contains many examples of food web models, reaction network models (e.g. the E.coli metabolic network), or other problems (e.g. blending problems), that can be solved with the functions in R-package limSolve. It was created together with my colleague Dick van Oevelen.
Of all the R-packages that I made, this was the one I liked least, as it was the most technical (reading text files, parsing,... all very tedious). Nevertheless it is a great help to setup linear models.
R-package BCE
http://cran.r-project.org/web/packages/BCE/index.html
This is the Bayesian composition estimator, which estimates sample (taxonomic) composition from biomarker data. It is the R-version of "Chemtax".
R-package NetIndices
http://cran.r-project.org/web/packages/NetIndices/index.html
This R-package estimates network indices, including trophic structure of foodwebs. It was made together with -at the time- a Kenyan student of mine, Julius Kones, from the University of Nairobi.
R-packages shape and diagram
http://cran.r-project.org/web/packages/diagram/These are two graphical packages
Package shape plots graphical shapes (spheres, cylinders, arrows, etc...)
Package diagram contains functions for visualising simple graphs (networks), or plotting flow diagrams.
It may seem weird that a scientist spends (her free) time in creating such packages. However, they allow to avoid using mouse-driven click and drag drawing programs (whose name I will not tell), which are particularly bad for my health. Since I use R and generate figures by programming them, my RSI problems have been (almost completely) annihilated.
R-package ecolMod
http://cran.r-project.org/web/packages/ecolMod/This is the R-package that contains all the examples and R-code to produce the figures of our book ("A practical guide to ecological modelling - using R as a simulation platform").
Figures from each chapter are included as a demo, named after the chapter (e.g. demo("chap1") will show the figures from our chapter 1).
RPC-The R package creator
ftp://ftp.nioo.knaw.nl/Pub/aknuijt/R Package Creator Setup.exe
This software was made by Adri Knuijt, the programmer from our department. It is meant to help in creating R-packages. We found it necessary to have one person that knows how to make packages, as it is quite a challenge to make a first R-package.
For windows users, the “R package creator” (RPC) is a great aid to do this. The program installs all software that is needed to make a package. (cygwin, Perl, minGW, Microsoft HTML compiler) provides a windows interface that guides you through package creation.
Once a package has been created, RPC provides tools that can be triggered by a right-click in the windows explorer. Just select the directory with the package that was altered, right-click and choose one of the options (to build a tar.gz, zip, or the package PDF file, to check the R-package or to simply install the package into R).
FEMME Flexible Environment for Mathematically Modelling the Environment
http://www.nioo.knaw.nl/en/projects/femme
Downloadable models, visualisation software etc..
Downloads
Figures from Soetaert and Herman (2009)
figuresbook.ppt (2.57 MB)
This powerpoint presentation file contains all figures from our book.
R examples from Soetaert and Herman (2009)
R-examplesBook.zip (125.29 kB)
This zip file contains all the R-examples from our book.
Ecological modelling - lecture notes
marelac lecture notes.pdf (3.9 MB)
The lecture notes for the course in ecological modelling, given at the University of Ghent and the Free University of Brussels. From the very basics of modelling to rather advanced applications.
Ecological modelling - exercises in EXCEL - the questions
Exercises Questions.pdf (696.0 KB)
Ecological modelling - exercises in EXCEL - the answers
Exercises Answers.pdf (748.9 KB)
Ecological modelling - exercises in EXCEL - the spreadsheets
Marelac exercises sheets.zip (566.7 KB)
Ecological modelling - exercises in EXCEL - presentations
marelac_ecol_presentation.ppt (3.6 MB)
example of exam
examen2003.pdf (28.4 KB)
The exam of year 2003
Mixing models
MIXING.zip (354.8 KB)
This ZIP includes all the files necessary for running the non-local exchange diagenetic models, as described in Soetaert et al., (1996) - Journ. Mar. Res. 54(6): 1207-1277.
Lecture notes on databases
Databases.zip (1.9 MB)
This zip file contains lecture notes and excercises (including answers) for a database workshop we once organised in Ghent.
Using R for scientific computing - R lecture notes
Scientificcomputing.zip (681.2 KB)
This ZIP file contain lecture notes to introduce you in the wonderful world of R-computing. R is an open-source software, extremely useful for scientific computing.
Once a year, me and Filip Meysman (VUB) teach ecological modelling at the universities of Ghent and Brussels. We use R as a simulation environment to implement and solve these models. Before starting the ecological modelling lectures, we spend one day guiding our students through parts of these lecture notes. As such, the students get a 'flavor' of what R can do, and how to best work with the language.
The ZIP contains the notes "Using R for scientific computing", an R reference card and a data set.
Using R for scientific computing - answers to questions
Rlecture answers.zip (5.9 KB)
The answers to the questions in the "Using R for scientific computing" lecture notes
It is probably tempting to download this file together with the R lecture notes, but it is much more efficient if you try the exercises first by yourself, without peeking at the answers!
Tinn-R version that works with R
Tinn-R_1.19.2.2_setup.exe (2.1 MB)
Tinn-R is the ideal editor to work with R.
Unfortunately, in the later versions, Tinn-R does not communicate well with R.
Here is a version of Tinn-R (1.19.2.2) that we work with.
A half-a-day introduction to ecological modelling in R
NIOOdagen.zip (159.7 KB)
A half-a-day tutorial on how to use R for ecological modelling.
It was taught (with Peter Herman) at the NIOO days in 2008 for my NIOO-colleagues.
| 
