Detecting change-point, trend, and seasonality in satellite time series data to track abrupt changes and nonlinear dynamics: A Bayesian ensemble algorithm

REMOTE SENSING OF ENVIRONMENT

卷:232

Article

Zhaoa, Kaiguang^1,2  Wulder, Michael A.³  Hu, Tongxi²  Bright, Ryan⁴  Wu, Qiusheng⁵  Qin, Haiming²  Li, Yang²  Toman, Elizabeth²  Mallick, Bani⁶  Zhang, Xuesong⁷  Brown, Molly⁸ 

[1] Ohio State Univ, Ohio Agr Res & Dev Ctr, Sch Environm & Nat Resources, Wooster, OH 44691 USA

[2] Ohio State Univ, Sch Environm & Nat Resources, Environm Sci Grad Program, Columbus, OH 43210 USA

[3] Nat Resources Canada, Canadian Forest Serv, Victoria, BC V8Z 1M5, Canada

[4] Norwegian Inst Bioecon Res NIBIO, N-1431 As, Norway

[5] Univ Tennessee, Dept Geog, Knoxville, TN 37996 USA

[6] Texas A&M Univ, Dept Stat, College Stn, TX 77843 USA

[7] Univ Maryland, Joint Global Change Res Inst, Pacific Northwest Natl Lab, College Pk, MD 20740 USA

[8] Univ Maryland, Dept Geog Sci, College Pk, MD 20771 USA

关键词: Changepoint;    Bayesian changepoint detection;    Disturbance ecology;    Breakpoint;    Trend analysis;    Time series decomposition;    Bayesian model averaging;    Disturbances;    Nonlinear dynamics;    Regime shift;    Ensemble modeling;    Time series segmentation;    Phenology;   

DOI  :  10.1016/j.rse.2019.04.034

来源: Elsevier

【 摘 要 】

Satellite time-series data are bolstering global change research, but their use to elucidate land changes and 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. We tested it upon simulated, Landsat, and MODIS data. BEAST detected changepoints, seasonality, and trends in the data reliably; it derived realistic nonlinear trends and credible uncertainty measures (e.g., occurrence probability of changepoints over time) some information difficult to derive by conventional single-best-model algorithms but critical for interpretation of ecosystem dynamics and detection of low-magnitude disturbances. The combination of many models 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 robust changepoint detection and nonlinear trend analysis and will help exploit environmental time-series data for probing patterns and drivers of ecosystem dynamics.

【 授权许可】

Free