R Package for harmonic modelling of time-series data

R Package for harmonic modelling of time-series data

written by Marius Philipp

April 23, 2020

Sentinel-2 NDVI time-series over the Steigerwald. Left: Original satellite scenes after cloud, cloud shadow and snow maksing. Right: Interpolated time-series using a harmonic modelling.

In order to fully exploit the monitoring potential of the satellite systems, challenges such as noise and data gaps must be effectively addressed. These quality losses are mainly caused by sensor artifacts, clouds, cloud shadows and other weather conditions (Verbesselt et al., 2012). In the context of time series analysis, harmonic modeling is a powerful tool to fill these gaps and reduce noise by smoothing the original signal (de Jong et al., 2011).

For that purpose, I created an R-package rHarmonics which enables the user to perform a harmonic analysis on a given time-series data set.

To calculate the harmonic fitted curve of a periodic signal, ordinary least squares regressions are computed using coupled sine and cosine curves on time-series data. The underlying algorithm which is based on Shumway & Stoffer (2017) equations 4.1 – 4.2 can be seen below:

MODIS NDVI time-series data together with a harmonic fitted curve using 3 cylces per year.



Literature:

de Jong, R., de Bruin, S., de Wit, A., Schaepman, M. E., & Dent, D. L. (2011). Analysis of monotonic greening and browning trends from global ndvi time-series. Remote Sensing of Environment, 115 (2), 692–702.

Shumway, R. H., & Stoffer, D. S. (2017). Time series analysis and its applications: with r examples. Springer.

Verbesselt, J., Zeileis, A., & Herold, M. (2012). Near real-time disturbance detection using satellite image time series. Remote Sensing of Environment, 123 , 98–108.

you may also like:

new publication on biodiversity metrics to measure from Space

new publication on biodiversity metrics to measure from Space

Our new Nature Eco Evol article is finally available online. Lead by Andrew Skidmore with an impressive list of great co-authors we stress the importance of earth observation to monitor biodiversity (change) from space in addition to ground based methods. We mainly...

WASCAL MSc Program course on spatial data analysis

WASCAL MSc Program course on spatial data analysis

As part of the project WASCAL-DE-Coop, the Department of Remote Sensing at the University of Würzburg conducted the second course on Spatial Data Analysis, embedded in the WASCAL Master Research Program "Climate Change and Informatics" at the University Université...

most recent news:

new publication on biodiversity metrics to measure from Space

new publication on biodiversity metrics to measure from Space

Our new Nature Eco Evol article is finally available online. Lead by Andrew Skidmore with an impressive list of great co-authors we stress the importance of earth observation to monitor biodiversity (change) from space in addition to ground based methods. We mainly...

WASCAL MSc Program course on spatial data analysis

WASCAL MSc Program course on spatial data analysis

As part of the project WASCAL-DE-Coop, the Department of Remote Sensing at the University of Würzburg conducted the second course on Spatial Data Analysis, embedded in the WASCAL Master Research Program "Climate Change and Informatics" at the University Université...