Abstract
Ecological processes, like the rise and fall of populations, leave an imprint of their dynamics as a pattern in space. Mining this spatial record for insight into temporal change underlies many applications, including using spatial snapshots to infer trends in communities, rates of species spread across boundaries, likelihood of chaotic dynamics, and proximity to regime shifts. However, these approaches rely on an inherent but undefined link between spatial and temporal variation. We present a quantitative link between a variable’s spatial and temporal variation based on established variance-partitioning techniques, and test it for predictive and diagnostic applications. A strong link existed between spatial and regional temporal variation (estimated as Coefficients of Variation or CV’s) in 136 variables from three aquatic ecosystems. This association suggests a basis for substituting one for the other, either quantitatively or qualitatively, when long time series are lacking. We further show that weak substitution of temporal for spatial CV results from distortion by specific spatiotemporal patterns (e.g., inter-patch synchrony). Where spatial and temporal CV’s do not match, we pinpoint the spatiotemporal causes of deviation in the dynamics of variables and suggest ways that may control for them. In turn, we demonstrate the use of this framework for describing spatiotemporal patterns in multiple ecosystem variables and attributing them to types of mechanisms. Linking spatial and temporal variability makes quantitative the hitherto inexact practice of space-for-time substitution and may thus point to new opportunities for navigating the complex variation of ecosystems.
Highlights
The spatial texture of a landscape is a fundamental reflection of the ecological processes underpinning it
Because obtaining long time series is difficult, inferring temporal patterns from spatial data is used in such varied contexts as: (i) chronosequences, where gradients of different-aged sites are used to track how a process changes from one state to another over time [8,9,10,11], (ii) boundary dynamics, where spatial snapshots can reveal the rate of species spread [12], (iii) complex dynamics, where spatial data helps detect chaos [13], and (iv) regime or phase shifts, where changes in spatial variation can expose the incipient reorganization of an ecosystem [14,15,16]
Because these terms are linked to temporal variability, they may provide a new view of dynamics and their consequences for stability. Because they are commonly used in ecology, we extended our analytical framework to include common indices (Fig. 1B) like the Coefficient of Variation (CV), and indices of synchrony ( T) and persistence ( S)
Summary
The spatial texture of a landscape is a fundamental reflection of the ecological processes underpinning it. Spatial patterns are diagnostic when they are used to uncover hidden mechanisms in the landscape, and predictive when they indicate the likely future behavior of a process. Ecology is full of examples of the former, diagnostic approach where spatial patterns are mined for evidence of mechanisms like dispersal, competition or environmental structuring [4,5,6,7]. Because obtaining long time series is difficult, inferring temporal patterns from spatial data is used in such varied contexts as: (i) chronosequences, where gradients of different-aged sites are used to track how a process (e.g., succession) changes from one state to another over time [8,9,10,11], (ii) boundary dynamics, where spatial snapshots can reveal the rate of species spread [12], (iii) complex dynamics, where spatial data helps detect chaos [13], and (iv) regime or phase shifts, where changes in spatial variation can expose the incipient reorganization of an ecosystem [14,15,16]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call