geoTS-package: Methods for Handling and Analyzing Time Series of Satellite...

In geoTS: Methods for Handling and Analyzing Time Series of Satellite Images

Methods for Handling and Analyzing Time Series of Satellite Images

Description

We provide tools for handling time series of satellite images as well as some statistical methods for spatio-temporal analysis

Tools for handling time series of satellite images

transfer_bin_raster transfers data from images originally recorded in a binary format to images in any of the formats allowed by the raster-package. Similarly, transfer_raster_RData extracts the entries of images originally recorded in tiff format, virtually stores them in an array object and, finally, this array is saved in an RData file. split_replace allows us to split Raster* objects, which can be arguably large, into smaller chunks. These chunks can be saved in any of the formats allowed by writeRaster. Often, satellite images come with missing values (or fill values assigned by other computer programs), split_replace allows to replace these values by values of users' convenience; see also reclassify. raster_intersect_sp allows us to obtain data in the intersection of Raster* and SpatialPolygonsDataFrame objects.

Methods for analyzing time series of satellite images

haRmonics allows us to fit harmonic regression models to numeric vectors; the method hants is based on Roerink et al. (2000) whereas the method harmR is based on Jakubauskas et al. (2001). The wls_harmR is the weighted least squares method which requires pre-estimation of heteroscedastic variance; hetervar allows for heteroscedastic variance estimation for numeric vectors extracted from time series of satellite imagery.

geoTS include the following datasets:

• master: RasterLayer with a land mask of eastern Yucatan Peninsula, Mexico.

• MOD13Q1_NDVI_2000129_009: A spatial subset of NDVI measurements taken over the eastern Yucatan Peninsula, Mexico in 2000.

• MOD13Q1_NDVI_Mohinora: RasterStack containing 23 spatial subsets of 16-day NDVI images of Cerro Mohinora acquired in 2001.

• shp_mohinora: SpatialPolygonsDataFrame delimiting the smallest Protected Area of Flora and Fauna in Mexico (Cerro Mohinora).

Author(s)

Tecuapetla-Gomez, I. [e-mail]

References

Roerink, G.J., Menenti, M., Verhoef, W. (2000). Reconstructing clodfree NDVI composites using Fourier analysis of time series, Int. J. Remote Sensing, 21(9), 1911–1917.

Jakubauskas, M., Legates, D., Kastens, J. (2001). Harmonic analysis of time-series AVHRR NDVI data, Photogrammetric Engineering and Remote Sensing, 67(4), 461–470.

The Matlab implementation of HANTS can be found here.

geoTS documentation built on Nov. 18, 2022, 1:08 a.m.