lsat_evaluate_phenological_max: Evaluate estimates of annual phenological maximum

In logan-berner/lsatTS: An R package to facilitate retrieval, cleaning, cross-calibration, and phenological modeling of Landsat time-series data

Evaluate estimates of annual phenological maximum

Description

Assess how the number of annual Landsat measurements impacts estimates of annual maximum vegetation greenness derived from raw measurements and phenological modeling. The algorithm computes annual maximum vegetation greenness for each site using years with at least a user-specific number of measurements and then compares these with estimates derived when using progressively smaller subsets of measurements. This lets the user determine the degree to which annual estimates of maximum vegetation greenness are impacted by the number of available measurements.

Usage

lsat_evaluate_phenological_max(
  dt,
  si,
  min.frac.of.max = 0.75,
  zscore.thresh = 3,
  min.obs = 6,
  reps = 10,
  outdir = NA
)

Arguments

dt	Data.table output from lsat_fit_phenological_curves().
si	Character string specifying the spectral index (SI) to evaluate (e.g., NDVI).
min.frac.of.max	Numeric threshold (0-1) that defines the "growing season" as the seasonal window when the phenological curves indicate the SI is within a specified fraction of the maximum SI. In other words, an observation is considered to be from the "growing season" when the SI is within a user-specified fraction of the curve-fit growing season maximum SI.
zscore.thresh	Numeric threshold specifying the Z-score value beyond which individual measurements are filtered before computing the maximum SI.
min.obs	Minimum number of site-level measurements needed each year to be included in the evaluation (Default = 10).
reps	Number of times to bootstrap the assessment (Default = 10).
outdir	If desired, specify the output directory where evaluation data and figure should be written. If left as NA, then output is only displayed in the console and not written to disk.

Value

A data.table and a figure summarizing how estimates of annual maximum SI vary with the number of Landsat measurements made during each growing season.

Examples

data(lsat.example.dt)
lsat.dt <- lsat_format_data(lsat.example.dt)
lsat.dt <- lsat_clean_data(lsat.dt)
lsat.dt <- lsat_calc_spectral_index(lsat.dt, 'ndvi')
# lsat.dt <- lsat_calibrate_rf(lsat.dt, band.or.si = 'ndvi', write.output = FALSE)
lsat.pheno.dt <- lsat_fit_phenological_curves(lsat.dt, si = 'ndvi') 
lsat_evaluate_phenological_max(lsat.pheno.dt, si = 'ndvi')

logan-berner/lsatTS documentation built on Oct. 21, 2024, 12:23 a.m.