Dr. Karline Soetaert

Head of Department: Ecosystem Studies

Master in Zoology (1983), PhD in zoology (1988),
Master in Computer science (1985),
Visiting professor at University of Ghent,
Belgium (2001-present).
T. +31 (0) 113 577 487
F. +31 (0) 113 573 616 0

Karline.Soetaert(at)nioz.nl

Visiting address:

Korringaweg 7

4401 NT Yerseke

The Netherlands

Postal address:

Postbus 140

4400 AC Yerseke

The Netherlands

• Food webs
• Biogeochemistry
• Sediment-water interactions

• Modeling and data assimilation
• Scientific computing using R

I use four tools for my hobbies: my cello, my mountainbike, my wildlife garden and my computer.

Puzzling deep-water food webs

We are now at the brink of exploiting the deep sea, yet know very little about this ecosystem. Since a couple of years we investigate food webs in deep-sea sediments. This is not simple as we cannot just go to these remote places and investigate them in a way that we would do in shallow water. Rather we have to work with whatever piece of information we obtain from this environment, and use our modeling expertise, imagination and good sense to derive as much as we can from the data. By now we have a quite extensive data base of deep-water food webs. This work is mainly done with Dick van Oevelen. 

The role of sediments in shallow-water systems

Sediments are not just the ‘bottom’ of the water column. They are very active environments with their own peculiar characteristics and they may have a profound impact on the water column functioning. In a couple of projects, we are investigating the nitrogen and phosphorus cycle in Wadden Sea sediments, and their implications on the water column. In another project, we look at the organic matter exchanges between water and sediment in the North Sea.

Oceanographic applications using R

Oceanography is a discipline that works with a large variety of data sets, and good software to visualise or analyse such data is often rather expensive or has a restricted scope. A valuable alternative to the existing oceanographic data analysis tools is the open-source software environment R. I am writing a set of packages to visualize complex oceanographic data or the output of 3-D hydrodynamical models.

Keeping an eye on our backyard

For a long time in the past the Westerschelde has been used as a conduit to transport very organically polluted water to the North-Sea. With increasing efforts to reduce this eutrophication the system has been altered drastically and it is still changing. Our institute monitors the system on a monthly basis. The changes in the functioning of the estuary and the freshwater system till the end of 2004 were described in previous papers. Today we are taking a fresh look at any changes that occurred since then.

Please find all my publications, software packages, including downloadable PDFs, at google scholar

Here are a few selected papers, listed according to research topic

Nematodes.

Soetaert, K. & M. Vincx, 1987.  Six new Richtersia species (Nematoda, Selachinematidae) from the Mediterranean.  Zool. Scr. 16 (2): 125-142.
Soetaert, K. & C. Heip, 1989.  The size structure of nematode assemblages along a Mediterranean deep-sea transect.  Deep-Sea Res 36 (1): 93-102.
Soetaert, K., Muthumbi, A., C. Heip, 2002. Size and shape of ocean margin nematodes: morphological diversity and depth-related patterns. Mar. Ecol. Progr. Ser. 242: 179-193.

Freeliving marine nematodes were my first study subject, and I have been faithfully writing papers on them till very recently, although at a decreasing frequency. My first ecological paper was published in 1989 and described how, along a Mediterranean Margin, these small sediment-inhabiting worms become smaller with increasing water depth. A more recent publication from 2002 suggests that these changes in nematode size is an adaptation to sediment oxygen content. Also in this paper, I argue why there is an advantage for nematodes to be either short and plump or long and slender. My very first scientific paper (1987) described 6 new species belonging to the plump type.

Nematodes - part 2.

Soetaert, K., et al., 1997.  Nematode distribution in ocean margin sediments of the Goban Spur (North-East Atlantic) in relation to sediment geochemistry.  Deep-Sea Res. 44(9-10): 1671-1683.
Braeckman, U., Vanaverbeke, J., Vincx, M., van Oevelen, D., Soetaert, K, 2013. Meiofauna Metabolism in Suboxic Sediments: Currently Overestimated. PloS one 8 (3), e59289

A recurring theme in my nematode work is that I like to assess their importance in the sediment carbon cycle. My first estimates, published in 1996, attributed more than 10% of sediment carbon turnover to nematodes, slightly declining with water depth. Much later, in a presentation at the 14th international meiofauna conference in Ghent (2010) I argued that the estimated contribution of nematodes to system metabolism might be strongly overestimated. This result was recently published in 2013.

Zooplankton.

Soetaert, K. & P. Van Rijswijk, 1993.  Spatial and temporal patterns of the zooplankton in the Westerschelde estuary.  Mar. Ecol. Progr. Ser. 97: 47-59.
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. pdf 

After my work on free living nematodes, I studied the ecology of the meso-zooplankton in the Schelde estuary. In contrast to nematodes, estuarine zooplankton organisms are highly influenced by water movements. I observed (not for the first time) that each year, the marine part of the Westerschelde estuary is populated with typical marine species (Soetaert and Van Rijswijk, 1993). Later I used my own data as input to my own mathematical model to infer the driving mechanism (Soetaert and Herman, 1994). Apparently the marine zooplankton from the Westerschelde estuary is mainly imported from the North sea. We estimated that, on a yearly basis, some 1500 tonnes of dry weight is transported from the North Sea to the estuary where it dies; this is equivalent to about 4000 Dutch cows !  

Models.

Soetaert, K., P.M.J. Herman & J. Kromkamp, 1994.  Living in the twilight: estimating phytoplankton growth in the Westerschelde estuary (the Netherlands) by means of the ecosystem model MOSES.  J. Plankt. Res. 16: 1277-1301.
Soetaert, K., P.M.J. Herman & J. Middelburg, 1996.  A model of early diagenetic processes from the shelf to abyssal depths.  Geochim. Cosmochim. Acta 60(6): 1019-1040.
Soetaert, K., J. Middelburg, P.M.J. Herman, K. Buis, 2000. On the coupling of benthic and pelagic biogeochemical models.  Earth-Science reviews, 51, 173-201.
Soetaert, K, et al., 2004. Modeling growth and carbon allocation in two reed beds (Phragmites australis) in the Scheldt estuary (Belgium, The Netherlands). Aquatic Botany 79: 211-234.
Soetaert, K., van Oevelen, D., 2009. Modeling food web interactions in benthic deep-sea ecosystems: a practical guide. Oceanography (22) 1: 130-145.
K. Soetaert  and  P.M.J. Herman, 2009.  A practical guide to ecological modelling – using R as a simulation platform. Springer, 372 pp. web: .

Around 1993 I made my first mathematical model. I found this type of work so fascinating and insightful that I have used models in most of my subsequent research. There is an endless number of topics that can be investigated by means of mathematical models. My first model was on the ecology and biogeochemistry of the Schelde estuary (e.g. Soetaert et al., 1994), then I did some sediment diagenesis modeling (e.g. Soetaert et al., 1996), I described how to couple sediment models to pelagic models (Soetaert et al., 2000), I made an excursion to plant modelling (2004), … and more recently was active in food web modeling (2009). Of course my “piece de resistance” (highlight) on modelling is the book I wrote with Peter Herman in 2009.

Models part 2.

Soetaert, K., Hofmann, A.F., Middelburg, J., Meysman, F., Greenwood, J, 2007. The effect of biogeochemical processes on pH. Mar. Chem. 105: 30-51.
Soetaert, K., Middelburg, J., 2009. Modeling eutrophication and oligotrophication of shallow-water marine systems: the importance of sediments and stratification. Hydrobiologia, 629, 239-254. DOI 10.1007/s10750-009-9777-x
Soetaert, K., Herman, P.M.J. Middelburg, J. J. Heip, C. 1998.  Assessing organic matter mineralization rate, degradability and mixing rate in an ocean margin sediment (North-East Atlantic) by diagenetic modelling. Journ. Mar. Res. 56: 519-534.
Van Oevelen, D., G. Duineveld, M. Lavaleye, F. Mienis, K. Soetaert, C.H.R. Heip, 2009. The cold-water coral community as a hotspot of carbon cycling on continental margins: a food web analysis from Rockall Bank (northeast Atlantic). Limnology and Oceangraphy 54:1829-1844.

Models can be used in a variety of ways. Sometimes I use models for better understanding the effect of certain processes (e.g. Soetaert et al., 2007), or to investigate, in general terms how the environment operates (e.g. Soetaert and Middelburg, 2009). In other cases, mathematical models are the only way to get some quantitative information out of the data (e.g. Soetaert et al., 1998). The food web papers, written with Dick van Oevelen are a nice example of how we use models together with expensive but scarce data to get a complete picture of food webs from remote places such as the deep-sea.

Mathematical techniques.

Soetaert, K. & C. Heip, 1990.  Sample-size dependence of diversity indices and the determination of sufficient sample size in a high-diversity deep-sea environment.  Mar. Ecol. Progr. Ser. 59: 305-307. 
Van den Meersche , K., Soetaert, K., Van Oevelen, D., 2009. xsample: an R function for sampling over- and underdetermined linear inverse problems. Journal of Statistical Software, Code Snippets 30(1), 1-15. url  
Soetaert, K., Gregoire, M.L., 2011. Data assimilation of the carbonate pH system – a kalman filter tested. Ecol. Modelling 222, 1929-1942.
Soetaert, K., Cash, J.R., Mazzia, F., 2012. Solving differential equations in R. UseR, Springer, 248 pp. web.

Ever since the very beginning of my scientific work, I have been intrigued by mathematical techniques. As a very young scientist (Soetaert and Heip, 1990) I used monte carlo simulation to show that diversity estimates should be used with caution due to their dependence on sample size. Much later I worked on mathematical methods to sample underdetermined linear systems (van den Meersche et al., 2009), or investigated the usability of data assimilation techniques in aquatic systems (Soetaert and Gregoire, 2011). The most extreme example of my work in the field of mathematics, however, is my last book (Soetaert et al., 2012). This deals with both the theory of numerically solving differential equations and their implementation in the software R. I wrote the R example chapters, but also two theoretical chapters - I leave it to the reader to guess which.

SoftwaRe.

Soetaert, K., Petzoldt, T. and R.W. Setzer, 2010. Solving Differential Equations in R: Package deSolve,  Journal of  Statistical Software 33(9), 1-25.
Soetaert, K. and T. Petzoldt, 2010. Inverse modelling, sensitivity and Monte Carlo analysis in R using package FME. Journal of Statistical Software 33 (3), 1-28.
Soetaert, K. and F. Meysman, 2012. Reactive transport in aquatic ecosystems: rapid model prototyping in the open source software R. Environmental modelling and software 32, 49-60.

In 2008 I decided to get rid of most of the software I used at the time, and to focus on only one software package. Fate, and a student, led me to switch to R. Although it was not yet well known at the time I started using this software, by now, R is immensely popular and widely used for statistical analysis and graphics. I especially like it because it is free and easily extendible by means of user-contributed packages.
With the switch to R, I also decided that, if something could not (yet) be done in R, I would implement it - of course not necessarily in isolation!
The packages I made with various colleagues fall in several categories: visualization (e.g. shape, diagram, plot3D, plot3D.rgl, OceanView), numerical methods (deSolve, bvpSolve, rootSolve, deTestSet, limSolve, ReacTran, LIM), statistics (FME, BCE) and aquatic chemistry (marelac, AquaEnv). Some of these packages were described in a paper, and of course in my two books (see above).