We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Detecting ecological regime shifts from transect data.
- Authors
Ward, Delphi F. L.; Wotherspoon, Simon; Melbourne‐Thomas, Jessica; Haapkylä, Jessica; Johnson, Craig R.
- Abstract
Timely detection of ecological regime shifts is a key problem for ecosystem managers, because changed ecosystem dynamics and function will usually necessitate a change in management strategies. However, currently available methods for detecting regime shifts depend on having multiple long time series data from both before and after the regime shift. This data requirement is prohibitive for many ecosystems. Here, we present a new approach for detecting regime shifts from one‐dimensional spatial (transect) data from just a single time step either side of the transition. Characteristic length scale (CLS) estimation is a method of attractor reconstruction combined with nonlinear prediction that enables identification of the emergent scale at which deterministic behavior of the system is best observed. Importantly, previous studies show that a fundamental change in ecosystem dynamics, from one domain of attraction to another, is reflected in a change in the CLS, i.e., the approach enables distinguishing regime shifts from variability in dynamics around a single attractor. Until now the method required highly resolved two‐dimensional spatial data, but here we adapted the approach so that the CLS can be estimated from one‐dimensional transect data. We demonstrate its successful application to both model and real ecosystem data. In our model test cases, we detected change in the CLS in cases where the shape (topology) of the interaction network had changed, leading to a shift in community composition. In an examination of benthic transect data from four Indonesian coral reefs, changes in the CLS for two of the reefs indicate a regime shift. This new development in estimating CLSs makes it possible to detect regime shifts in systems where data are limited, removing ambiguity in the interpretation of community change.
- Subjects
ECOSYSTEMS; ECOLOGY; BIOMES; TIME series analysis; BIODIVERSITY; ECOSYSTEM dynamics
- Publication
Ecological Monographs, 2018, Vol 88, Issue 4, p694
- ISSN
0012-9615
- Publication type
Article
- DOI
10.1002/ecm.1312