Authors: Leah A. Wasser, Kyla Dahlin
Reviewers: Edmund Hart, Megan A. Jones
Reviewers: Edmund Hart, Megan A. Jones
First, let’s load the required libraries.
# load libraries
library(raster)
library(rhdf5)
library(rgdal)
Load Functions
Once we have done the work to build our functions, we can perform routine tasks over and over using those functions.
Note: if you have not yet loaded the neonAOP package, please see directions
for how to do this at the top of the
Plot a Spectral Signature from Hyperspectral Remote Sensing data in R - HDF5 tutorial.
## call function library
library(neonAOP)
Calculate NDVI
Next, we can use the create_stack function to create a raster stack of the
red and near-infrared bands that we need to calculate NDVI.
# set working directory
# setwd("~/Documents/data/NEONDI-2016")
# setwd("~/data/NEONDI-2016") # Windows
# Define the file name to be opened
f <- "NEONdata/D17-California/TEAK/2013/spectrometer/reflectance/Subset3NIS1_20130614_100459_atmcor.h5"
# define CRS
epsg=32611
# Calculate NDVI
# select bands to use in calculation (red, NIR)
ndvi_bands <- c(58, 90)
#create raster list and then a stack using those two bands
ndvi_stack <- create_stack(ndvi_bands,
file = f,
epsg = epsg)
# calculate NDVI
NDVI <- function(x) {
(x[,2]-x[,1])/(x[,2]+x[,1])
}
ndvi_rast <- calc(ndvi_stack, NDVI)
# clear out plots
# dev.off(dev.list()["RStudioGD"])
plot(ndvi_rast,
main="NDVI for the NEON TEAK Field Site")
Export to GeoTIFF
# export as a GeoTIFF
writeRaster(ndvi_rast,
file="outputs/TEAK/ndvi_2013.tif",
format="GTiff",
overwrite=TRUE)
Plot NDVI
DSM <- raster("NEONdata/D17-California/TEAK/2013/lidar/TEAK_lidarDSM.tif")
slope <- terrain(DSM, opt='slope')
aspect <- terrain(DSM, opt='aspect')
# create hillshade
hill <- hillShade(slope, aspect, 40, 270)
plot(hill,
col=grey(1:100/100),
main="NDVI for the Lower Teakettle Field site",
legend=FALSE)
plot(ndvi_rast,
add=TRUE,
alpha=.3
)
