Decompose {greenbrown} | R Documentation |
This function decomposes time series in different components using a simple step-wise approach.
Decompose(Yt, breaks = 0, mosum.pval = 0.05)
Yt |
univariate time series of class |
breaks |
maximal number of breaks in the trend component to be calculated (integer number). |
mosum.pval |
Maximum p-value for the OLS-MOSUM test in order to search for breakpoints. If p = 0.05, breakpoints will be only searched in the time series trend component if the OLS-MOSUM test indicates a significant structural change in the time series. If p = 1 breakpoints will be always searched regardless if there is a significant structural change in the time series or not. |
The decomposition of the time series is based on a simple step-wise approach:
The mean of the NDVI time series is calculated.
In the second step, monthly values are aggregated per year by using the average value and the trend is calculated based on annual aggregated values using TrendAAT
.
The mean of the time series and the derived trend component from step (2) are subtracted from the annual values to derive the trend-removed and mean-centred annual values (annual anomalies). If the trend slope is not significant (p > 0.05), only the mean is subtracted.
In the next step, the mean, the trend component and the annual anomalies are subtracted from the original time series to calculate a detrended, mean-centered and for annual anomalies adjusted time series. From this time series the seasonal cycle is estimated as the mean seasonal cycle.
In the last step, the short-term anomalies are computed. For this, the mean, the trend component, the annual anomalies and the mean seasonal cycle are subtracted from the original time series.
The function returns a multi-variate object of class ts including the time series components.
Matthias Forkel <matthias.forkel@geo.tuwien.ac.at> [aut, cre]
Forkel, M., N. Carvalhais, J. Verbesselt, M. Mahecha, C. Neigh and M. Reichstein (2013): Trend Change Detection in NDVI Time Series: Effects of Inter-Annual Variability and Methodology. - Remote Sensing 5.
# load a time series of Normalized Difference Vegetation Index data(ndvi) plot(ndvi) # decompose this time series ndvi.dc <- Decompose(ndvi) plot(ndvi.dc) ndvi.dc2 <- Decompose(ndvi, breaks=2, mosum.pval=1) plot(ndvi.dc2)