Monitoring the phase space of ecosystems: Concept and examples from the Quillow catchment, Uckermark

Abstract Ecosystem research benefits enormously from the fact that comprehensive data sets of high quality, and covering long time periods are now increasingly more available. However, facing apparently complex interdependencies between numerous ecosystem components, there is urgent need rethinking our approaches in ecosystem research and applying new tools of data analysis. The concept presented in this paper is based on two pillars. Firstly, it postulates that ecosystems are multiple feedback systems and thus are highly constrained. Consequently, the effective dimensionality of multivariate ecosystem data sets is expected to be rather low compared to the number of observables. Secondly, it assumes that ecosystems are characterized by continuity in time and space as well as between entities which are often treated as distinct units. Implementing this concept in ecosystem research requires new tools for analysing large multivariate data sets. This study presents some of them, which were applied to a comprehensive water quality data set from a long-term monitoring program in Northeast Germany in the Uckermark region, one of the LTER-D (Long Term Ecological Research network, Germany) sites. The effective dimensionality was assessed by the Correlation Dimension approach as well as by a Principal Component Analysis and was in fact substantially lower than the number of observables. Continuity in time, space and between different types of water bodies was studied by combining Self-Organizing Maps with Sammon's Mapping. Groundwater, kettle hole and stream water samples exhibited some overlap, confirming continuity between different types of water bodies. Clear long-term shifts were found at the stream sampling sites. There was strong evidence that the intensity of single processes had changed at these sites rather than that new processes developed. Thus the more recent data did not occupy new subregions of the phase space of observations. Short-term variability of the kettle hole water samples differed substantially from that of the stream water samples, suggesting different processes generating the dynamics in these two types of water bodies. However, again, this seemed to be due to differing intensities of single processes rather than to completely different processes. We feel that research aiming at elucidating apparently complex interactions in ecosystems could make much more efficient use from now available large monitoring data sets by implementing the suggested concept and using corresponding innovative tools of system analysis.

[1]  P. Grassberger,et al.  Measuring the Strangeness of Strange Attractors , 1983 .

[2]  Juha Vesanto,et al.  SOM-based data visualization methods , 1999, Intell. Data Anal..

[3]  Michael Glemnitz,et al.  What governs macrophyte species richness in kettle hole types? A case study from Northeast Germany , 2012 .

[4]  Felix Müller,et al.  Long-Term Ecosystem Research Between Theory and Application – An Introduction , 2010 .

[5]  Christoph Merz,et al.  Data on geochemical and hydraulic properties of a characteristic confined/unconfined aquifer system of the younger Pleistocene in northeast Germany , 2015 .

[6]  Quanfa Zhang,et al.  Advancing Analysis of Spatio-Temporal Variations of Soil Nutrients in the Water Level Fluctuation Zone of China’s Three Gorges Reservoir Using Self-Organizing Map , 2015, PloS one.

[7]  Nitin Muttil,et al.  Machine-learning paradigms for selecting ecologically significant input variables , 2007, Eng. Appl. Artif. Intell..

[8]  Ralf Dannowski,et al.  Parameterization and regionalization of redox based denitrification for GIS-embedded nitrate transport modeling in Pleistocene aquifer systems , 2009 .

[9]  Romeo Gilbuena,et al.  Clustering spatio-seasonal hydrogeochemical data using self-organizing maps for groundwater quality assessment in the Red River Delta, Vietnam , 2015 .

[10]  G. Lischeid,et al.  Grasping the heterogeneity of kettle hole water quality in Northeast Germany , 2011, Hydrobiologia.

[11]  Ralf Wieland,et al.  Spatial Analysis and Modeling Tool (SAMT): 2. Applications , 2006, Ecol. Informatics.

[12]  Gunnar Lischeid,et al.  Non-linear visualization and analysis of large water quality data sets: a model-free basis for efficient monitoring and risk assessment , 2009 .

[13]  Pavel Praks,et al.  Information retrieval in hydrochemical data using the latent semantic indexing approach , 2007 .

[14]  T. Kohonen.  Self-organized formation of topographically correct feature maps , 1982 .

[15]  Gunnar Lischeid,et al.  Tracing biogeochemical processes in stream water and groundwater using non-linear statistics , 2008 .

[16]  Sang-Wook Yeh,et al.  Recent progress on two types of El Niño: Observations, dynamics, and future changes , 2014, Asia-Pacific Journal of Atmospheric Sciences.

[17]  Tom Addiscott.  Emergence or self-organization? Look to the soil population , 2011 .

[18]  Mikko Kolehmainen,et al.  Evaluating online data of water quality changes in a pilot drinking water distribution system with multivariate data exploration methods. , 2008, Water research.

[19]  Laura Toran,et al.  What's new with CUASHI? , 2004, Ground water.

[20]  Gunnar Lischeid,et al.  Tracing Biogeochemical Processes in Small Catchments Using Non-linear Methods , 2010 .

[21]  Thomas Kalettka,et al.  Hydrogeomorphic types of glacially created kettle holes in North-East Germany , 2006 .

[22]  Monika Wulf,et al.  Plant species richness of very small forests related to patch configuration, quality, heterogeneity and history , 2014 .

[23]  H. Vereecken,et al.  A terrestrial observatory approach to the integrated investigation of the effects of deforestation on water, energy, and matter fluxes , 2014, Science China Earth Sciences.

[24]  Niels Wessel,et al.  Correlation dimension analysis of heart rate variability in patients with dilated cardiomyopathy , 2005, Comput. Methods Programs Biomed..

[25]  Sebastian Schmidtlein,et al.  Revealing biogeographical patterns by nonlinear ordinations and derived anisotropic spatial filters , 2008 .

[26]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[27]  Corinne Le Quéré,et al.  Persistent growth of CO2 emissions and implications for reaching climate targets , 2014 .

[28]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[29]  Theiler,et al.  Efficient algorithm for estimating the correlation dimension from a set of discrete points. , 1987, Physical review. A, General physics.

[30]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[31]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[32]  T. Kalettka,et al.  “Potholes” in Northeast German Agro-landscapes: Functions, Land Use Impacts, and Protection Strategies , 2001 .

[33]  William J. Walsh,et al.  Marine Protected Area Networks: Assessing Whether the Whole Is Greater than the Sum of Its Parts , 2014, PloS one.

[34]  William N. Venables,et al.  Modern Applied Statistics with S , 2010 .

[35]  Stephen J. Fraser,et al.  On the scope and management of pesticide pollution of Swedish groundwater resources: The Scanian example , 2015, AMBIO.

[36]  G. Heckel,et al.  Breeding site fidelity and winter admixture in a long-distance migrant, the tufted duck (Aythya fuligula) , 2012, Heredity.

[37]  Boris Schröder,et al.  Predicting spatial and temporal habitat use of rodents in a highly intensive agricultural area , 2014 .