R: The Cook Agronomy Farm data set

cookfarm {GSIF}

R Documentation

The Cook Agronomy Farm data set

Description

The R.J. Cook Agronomy Farm (cookfarm) is a Long-Term Agroecosystem Research Site operated by Washington State University, located near Pullman, Washington, USA. Contains spatio-temporal (3D+T) measurements of three soil properties and a number of spatial and temporal regression covariates.

Usage

data(cookfarm)

Format

The cookfarm data set contains four data frames. The readings data frame contains measurements of volumetric water content (cubic-m/cubic-m), temperature (degree C) and bulk electrical conductivity (dS/m), measured at 42 locations using 5TE sensors at five standard depths (0.3, 0.6, 0.9, 1.2, 1.5 m) for the period "2011-01-01" to "2012-12-31":

SOURCEID: factor; unique station ID
Date: date; observation day
Port*VW: numeric; volumetric water content measurements at five depths
Port*C: numeric; soil temperature measurements at five depths
Port*EC: numeric; bulk electrical conductivity measurements at five depths

The profiles data frame contains soil profile descriptions from 142 sites:

SOURCEID: factor; unique station ID
Easting: numeric; x coordinate in the local projection system
Northing: numeric; y coordinate in the local projection system
TAXNUSDA: factor; Keys to Soil Taxonomy taxon name e.g. "Caldwell"
HZDUSD: factor; horizon designation
UHDICM: numeric; upper horizon depth from the surface in cm
LHDICM: numeric; lower horizon depth from the surface in cm
BLD: bulk density in tonnes per cubic-meter
PHIHOX: numeric; pH index measured in water solution

The grids data frame contains values of regression covariates at 10 m resolution:

DEM: numeric; Digital Elevation Model
TWI: numeric; SAGA GIS Topographic Wetness Index
MUSYM: factor; soil mapping units e.g. "Thatuna silt loam"
NDRE.M: numeric; mean value of the Normalized Difference Red Edge Index (time series of 11 RapidEye images)
NDRE.sd: numeric; standard deviation of the Normalized Difference Red Edge Index (time series of 11 RapidEye images)
Cook_fall_ECa: numeric; apparent electrical conductivity image from fall
Cook_spr_ECa: numeric; apparent electrical conductivity image from spring
X2011: factor; cropping system in 2011
X2012: factor; cropping system in 2012

The weather data frame contains daily temperatures and rainfall from the nearest meteorological station:

Date: date; observation day
Precip_wrcc: numeric; observed precipitation in mm
MaxT_wrcc: numeric; observed maximum daily temperature in degree C
MinT_wrccc: numeric; observed minimum daily temperature in degree C

Note

The farm is 37 ha, stationed in the hilly Palouse region, which receives an annual average of 550 mm of precipitation, primarily as rain and snow in November through May. Soils are deep silt loams formed on loess hills; clay silt loam horizons commonly occur at variable depths. Farming practices at Cook Farm are representative of regional dryland annual cropping systems (direct-seeded cereal grains and legume crops).

Author(s)

Caley Gasch, Tomislav Hengl and David J. Brown

References

Gasch, C.K., Hengl, T., Gräler, B., Meyer, H., Magney, T., Brown, D.J., 2015. Spatio-temporal interpolation of soil water, temperature, and electrical conductivity in 3D+T: the Cook Agronomy Farm data set. Spatial Statistics, 14, pp.70–90.
Gasch, C.K., D.J. Brown, E.S. Brooks, M. Yourek, M. Poggio, D.R. Cobos, C.S. Campbell, 2016? Retroactive calibration of soil moisture sensors using a two-step, soil-specific correction. Submitted to Vadose Zone Journal.
Gasch, C.K., D.J. Brown, C.S. Campbell, D.R. Cobos, E.S. Brooks, M. Chahal, M. Poggio, 2016? A field-scale sensor network data set for monitoring and modeling the spatial and temporal variation of soil moisture in a dryland agricultural field. Submitted to Water Resources Research.

Examples

## An example for 3D+T modelling applied to the cookfarm data set can be assesed via 
## demo(cookfarm_3DT_kriging)
## demo(cookfarm_3DT_RF)
## Please note that the demo's might take 10-15 minutes to complete.
library(rgdal)
library(sp)
library(spacetime)
library(aqp)
library(splines)
library(randomForest)
library(plyr)
library(plotKML)
data(cookfarm)

## gridded data:
grid10m <- cookfarm$grids
gridded(grid10m) <- ~x+y
proj4string(grid10m) <- CRS(cookfarm$proj4string)
spplot(grid10m["DEM"], col.regions=SAGA_pal[[1]])

## soil profiles:
profs <- cookfarm$profiles
levels(cookfarm$profiles$HZDUSD)
## Bt horizon:
sel.Bt <- grep("Bt", profs$HZDUSD, ignore.case=FALSE, fixed=FALSE)
profs$Bt <- 0
profs$Bt[sel.Bt] <- 1
depths(profs) <- SOURCEID ~ UHDICM + LHDICM
site(profs) <- ~ TAXSUSDA + Easting + Northing
coordinates(profs) <- ~Easting + Northing
proj4string(profs) <- CRS(cookfarm$proj4string)
profs.geo <- as.geosamples(profs)

## fit model for Bt horizon:
m.Bt <- GSIF::fit.gstatModel(profs.geo, Bt~DEM+TWI+MUSYM+Cook_fall_ECa
   +Cook_spr_ECa+ns(altitude, df = 4), grid10m, fit.family = binomial(logit))
plot(m.Bt)

## fit model for soil pH:
m.PHI <- fit.gstatModel(profs.geo, PHIHOX~DEM+TWI+MUSYM+Cook_fall_ECa
    +Cook_spr_ECa+ns(altitude, df = 4), grid10m)
plot(m.PHI)

[Package GSIF version 0.5-5 Index]