Dissecting the space-time structure of tree-ring datasets using the partial triadic analysis

Community ecology in 3D: Tensor decomposition reveals spatio-temporal dynamics of large ecological communities

Understanding spatio-temporal dynamics of biotic communities containing large numbers of species is crucial to guide ecosystem management and conservation efforts. However, traditional approaches usually focus on studying community dynamics either in space or in time, often failing to fully account for interlinked spatio-temporal changes. In this study, we demonstrate and promote the use of tensor decomposition for disentangling spatio-temporal community dynamics in long-term monitoring data. Tensor decomposition builds on traditional multivariate statistics (e.g. Principal Component Analysis) but extends it to multiple dimensions. This extension allows for the synchronized study of multiple ecological variables measured repeatedly in time and space. We applied this comprehensive approach to explore the spatio-temporal dynamics of 65 demersal fish species in the North Sea, a marine ecosystem strongly altered by human activities and climate change. Our case study demonstrates how tensor decomposition can successfully (i) characterize the main spatio-temporal patterns and trends in species abundances, (ii) identify sub-communities of species that share similar spatial distribution and temporal dynamics, and (iii) reveal external drivers of change. Our results revealed a strong spatial structure in fish assemblages persistent over time and linked to differences in depth, primary production and seasonality. Furthermore, we simultaneously characterized important temporal distribution changes related to the low frequency temperature variability inherent in the Atlantic Multidecadal Oscillation. Finally, we identified six major sub-communities composed of species sharing similar spatial distribution patterns and temporal dynamics. Our case study demonstrates the application and benefits of using tensor decomposition for studying complex community data sets usually derived from large-scale monitoring programs.

The spatiotemporal pattern of earthworm community in the grass savannas of Lamto (Ivory Coast)

BACKGROUND

The impact of earthworms on both soil physical properties and soil organic matter dynamics has been well documented (Lavelle and Spain 2001). There is a wealth of literature dedicated to the biological mechanisms at work or to empirical approaches based on field data. Assessing the functional role of a species or community implies establishing both time and space scales at which it is effectively the primary determinant of the process(es) at hand. In that context, space-time data analyses are powerful tools to process community data collected on numerous occasions but are, however, not widely disseminated in the community of ecologists. Although computer resources are available, one difficulty is that ad hoc field data are not always easily available which hinders the percolation of the methods.

NEW INFORMATION

We provide the results of a 5 dates survey of earthworm community in a grass savanna of Lamto (Ivory Coast) conducted between 1995 and 1997. At each sampling date, earthworm community was assessed by hand-sorting a set of 100 soil monoliths distributed on a regular grid of 5 m mesh. These data were analyzed in Rossi (2003a) and are published here with the aim that they could be reanalyzed using new statistical tools (e.g. MEM analyses see Jiménez et al. 2014) or serve as example for researchers that train on space-time statistical methods.

A Methodological Approach for Spatiotemporally Analyzing Water-Polluting Effluents in Agricultural Landscapes Using Partial Triadic Analysis

Multivariate techniques for two-dimensional data matrices are normally used in water quality studies. However, if the temporal dimension is included in the analysis, other statistical techniques are recommended. In this study, partial triadic analysis was used to investigate the spatial and temporal variability in water quality variables sampled in a northeastern Spain river basin. The results highlight the spatiality of the physical and chemical properties of water at different sites along a river over 1 yr. Partial triadic analysis allowed us to clearly identify the presence of a stable spatial structure that was common to all sampling dates across the entire catchment. Variables such as electrical conductivity and Na and Cl ions were associated with agricultural sources, whereas total dissolved nitrogen, NH-N concentrations, and NO-N concentrations were linked to polluted urban sites; differences were observed between irrigated and nonirrigated periods. The concentration of NO-N was associated with both agricultural and urban land uses. Variables associated with urban and agricultural pollution sources were highly influenced by the seasonality of different activities conducted in the study area. In analyzing the impact of land use and fertilization management on water runoff and effluents, powerful statistical tools that can properly identify the causes of pollution in watersheds are important. Partial triadic analysis can efficiently summarize site-specific water chemistry patterns in an applied setting for land- and water-monitoring schemes at the landscape level. The method is recommended for land-use decision-making processes to reduce harmful environmental effects and promote sustainable watershed management.