Oikos 00: 1–14, 2009
doi: 10.1111/j.1600-0706.2009.18284.x
© 2009 he Authors. Journal compilation © 2009 Oikos
Subject Editor: José Alexandre Felizola Diniz-Filho. Accepted 16 November 2009
1
The virtual ecologist approach: simulating data and observers
Damaris Zurell, Uta Berger, Juliano S. Cabral, Florian Jeltsch, Christine N. Meynard,
Tamara Münkemüller, Nana Nehrbass, Jörn Pagel, Björn Reineking, Boris Schröder and
Volker Grimm
D. Zurell (damaris.zurell@uni-potsdam.de) and B. Schröder, Inst. of Geoecology, Univ. of Potsdam, Karl-Liebknecht-Str. 24/25, DE–14476
Potsdam, Germany. BS also at: ZALF e.V., Leibniz-Centre for Agricultural Landscape Research, Soil Landscape Modelling, Eberswalder Straße
84, DE–15374 Müncheberg, Germany. – U. Berger, Inst. of Forest Growth and Computer Sciences, Dresden Univ. of Technology, Pienner
Straße 8, DE–01737 harandt, Germany. – J. S. Cabral, F. Jeltsch and J. Pagel, Inst. for Biochemistry and Biology, Univ. of Potsdam,
Maulbeerallee 2, DE–14469 Potsdam, Germany. – C. N. Meynard, Inst. des Sciences de l’Evolution, Univ. de Montpellier II, UMR CNRS
5554, Place Eugène Bataillon, CC 065, FR–34095 Montpellier Cedex 5, France. – T. Münkemüller, Laboratoire d’Ecologie Alpine, Univ. J.
Fourier, UMR CNRS 5553, BP 53, FR–38041 Grenoble Cedex 9, France. – N. Nehrbass and V. Grimm, UFZ, Helmholtz Centre of
Environmental Research – UFZ, Dept of Ecological Modelling, Permoserstr. 15, DE–04318 Leipzig, Germany. Present address for NN:
Stünz-Mölkauer Weg 18, DE–04318 Leipzig, Germany. – B. Reineking, Biogeographical Modelling, BayCEER, Univ. of Bayreuth,
Universitätsstraße 30, DE–95440 Bayreuth, Germany.
Ecologists carry a well-stocked toolbox with a great variety of sampling methods, statistical analyses and modelling tools,
and new methods are constantly appearing. Evaluation and optimisation of these methods is crucial to guide method-
ological choices. Simulating error-free data or taking high-quality data to qualify methods is common practice. Here, we
emphasise the methodology of the ‘virtual ecologist’ (VE) approach where simulated data and observer models are used
to mimic real species and how they are ‘virtually’ observed. his virtual data is then subjected to statistical analyses and
modelling, and the results are evaluated against the ‘true’ simulated data. he VE approach is an intuitive and powerful
evaluation framework that allows a quality assessment of sampling protocols, analyses and modelling tools. It works under
controlled conditions as well as under consideration of confounding factors such as animal movement and biased observer
behaviour. In this review, we promote the approach as a rigorous research tool, and demonstrate its capabilities and practi-
cal relevance. We explore past uses of VE in diferent ecological research ields, where it mainly has been used to test and
improve sampling regimes as well as for testing and comparing models, for example species distribution models. We discuss
its beneits as well as potential limitations, and provide some practical considerations for designing VE studies. Finally,
research ields are identiied for which the approach could be useful in the future. We conclude that VE could foster the
integration of theoretical and empirical work and stimulate work that goes far beyond sampling methods, leading to new
questions, theories, and better mechanistic understanding of ecological systems.
Models permeate every ield in ecology. hey have become
an indispensable tool for a wide range of tasks, including
the understanding of mechanisms, capturing the processes
behind the emergence of ecological phenomena, quantify-
ing relationships between species presence or abundance
and environmental conditions, and forecasting efects of
changing environments on broad spatial and temporal scales
(DeAngelis and Mooij 2005, Araújo and Rahbek 2006,
huiller et al. 2008).
here is, however, a further important ield of applica-
tion of ecological models that so far has not been thoroughly
acknowledged in ecological research: evaluating methods for
data sampling, analysis and modelling methods by means
of virtual data. Here, the idea is to generate virtual data by
simulating not only ecological processes, but also the sam-
pling processes that are used to collect these data in real-
ity and the methodological tools used to analyse them. We
propose to call this the ‘virtual ecologist’ (VE) approach (see
Glossary). he virtue of this approach is its ability to rigor- he virtue of this approach is its ability to rigor-
ously test method performance against a known truth. he
VE approach is concerned with practical questions regard-
ing ecological methods: Is a method able to identify patterns
that we know exist (Grimm et al. 1999)? Can we infer the
mechanisms underlying these patterns given a certain set of
data (Tyre et al. 2001)? Can we correctly and reliably predict
future events (Zurell et al. 2009)?
To evaluate methods of data collection, statistical analy-
sis, and modelling we would ideally compare their outcome
to reality. his would allow us to assess whether existing
patterns were detected correctly, whether correct estimates
of process rates were obtained, or whether the distribu-
tion of a species was predicted correctly. However, we have
no privileged access to reality independent of and beyond ield
observations and analytical methods. he ability of ield data
to represent reality depends not only on the time interval and
the spatial extent of observation but also on the disturbances