detect_large_greenness_change_periods: implememnts an approximation of the methods of Zhang et al...

In Timothy-W-Hilton/VPRMLandSfcModel: R implementation of VPRM with parameter estimation

implememnts an approximation of the methods of Zhang et al (2003, 2006).

Description

locate periods of sustained greeness increase or decrease within an EVI time series

Usage

detect_large_greenness_change_periods(data, verbose = 0)

Arguments

data	data frame with (at least) columns "sitecode" (unique sitecode), "t" (chron object timestamps) and "EVI" (EVI observations). Data should contain EVI observations from one and only one site.
verbose	integer; causes silent execution if 0 (default); larger values cause increasingly verbose printouts. Currently has two options: silent if 0; prints site being processed if >0; this may change in the future to provide more granularity.

Value

data frame with columns "idx" (row indices of data where slope of EVI change changes sign), "date" (chron object containing the timestamps corresponding to idx), and "phen" (factor encoding the type of EVI change, with levels ginc, gdec, gmin, gmax.

Author(s)

Timothy W. Hilton

References

Xiaoyang Zhang, Mark A. Friedl, Crystal B. Schaaf, Alan H. Strahler, John C.F. Hodges, Feng Gao, Bradley C. Reed, Alfredo Huete, Monitoring vegetation phenology using MODIS, Remote Sensing of Environment, Volume 84, Issue 3, March 2003, Pages 471-475, ISSN 0034-4257, http://dx.doi.org/10.1016/S0034-4257(02)00135-9.

Zhang, X., M. A. Friedl, and C. B. Schaaf (2006), Global vegetation phenology from Moderate Resolution Imaging Spectroradiometer (MODIS): Evaluation of global patterns and comparison with in situ measurements, J. Geophys. Res., 111, G04017, doi:10.1029/2006JG000217.

Examples

data(Park_Falls)
names(PFa_evi) <- c('t', 'EVI', 'sitecode')
PFa_evi[['sitecode']] <- "US-PFa"
print(head(PFa_evi))
phen_transition_dates <- detect_large_greenness_change_periods(PFa_evi)
print(head(phen_transition_dates))

Timothy-W-Hilton/VPRMLandSfcModel documentation built on July 29, 2023, 8:43 p.m.