afspec: Sample soil spectroscopy data from Africa Sentinel Sites
In soil.spec: Soil Spectroscopy Tools and Reference Models

Description Usage Format Author(s) References Examples

Sample data (113 samples) containing absorbances in the Mid infrared region (MIR) collected at 57 sites (Africa Sentinel Sites).

1	data(afspec)

The afspec data set contains two data frames — "samples" and "ab". "samples" table contains the following columns:

SAMPLEID: factor; unique soil sample ID (duplicate IDs are possible as same soil sample can be measured twice)
Cluster: integer; Africa Sentinel Sites cluster number
Plot: integer; Africa Sentinel Sites plot number
Depth: factor; lower depth of the sampled horizon
Site: factor; site name
Country: factor; country name
Region: factor; region name
Longitude: numeric; longitude in decimal degrees on the WGS84 datum (Lat in the LDSF Field Database)
Latitude: numeric; latitude in decimal degrees on the WGS84 datum (Lat in the LDSF Field Database)

"ab" table contains the following columns:

SAMPLEID: factor; ...
m4003.5..m599.8: numeric; mid-infrared raw absorbance for denoted wavenumbers (Bruker-LTS instrument)
...: other raw absorbance values

Africa Soil Information Service (http://africasoils.net)

Vagen, T., Shepherd, K. D., Walsh, M. G., Winowiecki, L., Desta, L. T., & Tondoh, J. E. (2010) AfSIS technical specifications: soil health surveillance. World Agroforestry Centre, Nairobi, Kenya.
World Agroforestry Centre (http://worldagroforestry.org)

library(sp)
library(rgdal)
data(afspec)
sp <- afspec$samples[,c("Longitude","Latitude")]
coordinates(sp) <- ~Longitude+Latitude
proj4string(sp) <- CRS("+proj=longlat +datum=WGS84")
## prepare 'samples' table
samples <- cbind(afspec$samples["SAMPLEID"], 
    MID="AfSIS-MIR", 
    DateTime=Sys.time())
## convert to "SpectraPoints"
afspec.sp <- SpectraPoints(Spectra=Spectra(samples, afspec$ab), sp=sp)
summary(afspec.sp)
get("attributes", spec.opts)
## predict soil pH:
data(m.PHIHOX)
PHIHOX <- predict(afspec.sp, model=m.PHIHOX)
data(m.ORCCNS)
ORCCNS <- predict(afspec.sp, model=m.ORCCNS)
## Not run: 
library(plotKML)
afspec.pnts <- SpatialPointsDataFrame(sp, cbind(PHIHOX, ORCCNS))
data(SAGA_pal)
plotKML(afspec.pnts["PHIHOX"], colour_scale=rev(SAGA_pal[[1]]))
plotKML(afspec.pnts["ORCCNS"], colour_scale=SAGA_pal[[1]])
## plot with global soil legends:
library(GSIF)
data(soil.legends)
brks = c(soil.legends[["PHIHOX"]]$MIN[1]/10, soil.legends[["PHIHOX"]]$MAX/10)
pH.range <- cbind(soil.legends[["PHIHOX"]]$MAX, soil.legends[["PHIHOX"]]$MIN)
levs = paste(signif(rowMeans(pH.range)/10, 2))
afspec.pnts$Col <- cut(afspec.pnts$PHIHOX, breaks=brks, labels=levs)
plotKML(afspec.pnts["Col"], colour_scale=soil.legends[["PHIHOX"]]$COLOR)

## End(Not run)