Canadian Forest Service Publications

Detecting change-point, trend, and seasonality in satellite time series data to track abrupt changes and nonlinear dynamics: A Bayesian ensemble algorithm. 2019. Zhao, K., Wulder, M.A., Hu, T., Bright, R., Wu, Q., Qin, H., Li, Y., Toman, E., Mallick, B., Zhang, Z., Brown, M. Remote Sensing of Environment, 232, 111181.

Year: 2019

Issued by: Pacific Forestry Centre

Catalog ID: 40019

Language: English

Availability: PDF (request by e-mail)

Available from the Journal's Web site.
DOI: 10.1016/j.rse.2019.04.034

† This site may require a fee

Mark record

Abstract

Satellite time-series data are bolstering global change research, but their use to elucidate land changes andvegetation dynamics is sensitive to algorithmic choices. Different algorithms often give inconsistent or some-times conflicting interpretations of the same data. This lack of consensus has adverse implications and can bemitigated via ensemble modeling, an algorithmic paradigm that combines many competing models rather thanchoosing only a single“best”model. Here we report one such time-series decomposition algorithm for derivingnonlinear ecosystem dynamics across multiple timescales—A Bayesian Estimator of Abrupt change, Seasonalchange, and Trend (BEAST). As an ensemble algorithm, BEAST quantifies the relative usefulness of individualdecomposition models, leveraging all the models via Bayesian model averaging. We tested it upon simulated,Landsat, and MODIS data. BEAST detected changepoints, seasonality, and trends in the data reliably; it derivedrealistic nonlinear trends and credible uncertainty measures (e.g., occurrence probability of changepoints overtime)—some information difficult to derive by conventional single-best-model algorithms but critical for in-terpretation of ecosystem dynamics and detection of low-magnitude disturbances. The combination of manymodels enabled BEAST to alleviate model misspecification, address algorithmic uncertainty, and reduce over-fitting. BEAST is generically applicable to time-series data of all kinds. It offers a new analytical option for robustchangepoint detection and nonlinear trend analysis and will help exploit environmental time-series data forprobing patterns and drivers of ecosystem dynamics.

Plain Language Summary

Satellite time-series data are bolstering global change research, but their use to elucidate land surface or vegetation dynamics is sensitive to algorithmic choices. Different algorithms often give inconsistent or sometimes conflicting interpretations of the same data. This lack of consensus has adverse implications and can be mitigated via ensemble modeling, an algorithmic paradigm that combines many competing models rather than choosing only a single “best” model. Here we report one such time-series decomposition algorithm for deriving nonlinear ecosystem dynamics across multiple timescales—A Bayesian Estimator of Abrupt change, Seasonal change, and Trend (BEAST). As an ensemble algorithm, BEAST quantifies the relative usefulness of individual decomposition models, leveraging all the models via Bayesian model averaging. Resultant highlights include:

  1. Deriving land dynamics from satellite time series is sensitive to algorithmic choice
  2. Satellite-derived ecosystem dynamics improved by embracing algorithmic uncertainty
  3. A Bayesian model averaging time-series decomposition algorithm (BEAST) developed
  4. BEAST is a generic tool to detect change point, trend, and seasonality in time series