Getting and Comparing KSSL Data
D.E. Beaudette
2023-09-26
1 Introduction
This document demonstrates how to use the soilDB package to download KSSL data from SoilWeb. These data are from the September 2018 snapshot, and will be updated as future snapshots are released. Comparisons are made via graphical summaries of key soil properties with depth, using data structures and functions from the aqp package.
1.1 Installation
With a recent version of R (>= 3.5), it is possible to get all of the packages that this tutorial depends on via:
# stable packages from CRAN
install.packages('aqp', dep = TRUE)
install.packages('soilDB', dep = TRUE)
install.packages('Gmedian', dep = TRUE)
install.packages('tactile', dep = TRUE)
install.packages('maps', dep = TRUE)2 Quick Example
KSSL and NASIS data are from the September 2018 snapshot. Details pending.
library(aqp)
library(soilDB)
# get lab and morphologic data
# simplify colors to 1 color (moist and dry) per horizon via mixing
# see ?mix_and_clean_colors for details.
s <- fetchKSSL(series = 'auburn', returnMorphologicData = TRUE, simplifyColors = TRUE)
# extract pedons into SoilProfileCollection
pedons <- s$SPCThe result is a list, check it out.
str(s, 2)## List of 2
## $ SPC :Formal class 'SoilProfileCollection' [package "aqp"] with 9 slots
## $ morph:List of 6
## ..$ phcolor :'data.frame': 105 obs. of 7 variables:
## ..$ phfrags :'data.frame': 27 obs. of 10 variables:
## ..$ phpores :'data.frame': 71 obs. of 5 variables:
## ..$ phstructure :'data.frame': 40 obs. of 7 variables:
## ..$ pediagfeatures:'data.frame': 24 obs. of 5 variables:
## ..$ rmf : logi [1, 1] FALSE# check out the "raw" morphologic data:
lapply(s$morph, head)## $phcolor
## phiid labsampnum colorpct colorhue colorvalue colorchroma colormoistst
## 1 820051 05N02182 NA 10YR 4 4 Dry
## 2 820051 05N02182 NA 10YR 3 3 Moist
## 3 820052 05N02183 NA 7.5YR 3 3 Moist
## 4 820052 05N02183 NA 7.5YR 4 6 Dry
## 5 820053 05N02184 NA 5YR 3 3 Moist
## 6 820053 05N02184 NA 5YR 4 4 Dry
##
## $phfrags
## phiid labsampnum fragvol fragkind fragsize_l fragsize_r fragsize_h fragshp
## 1 820051 05N02182 14 Metamorphic rock fragments 2 5 76 NA
## 2 820052 05N02183 10 Metamorphic rock fragments 2 20 76 NA
## 3 820053 05N02184 18 Metamorphic rock fragments 2 5 76 NA
## 4 570101 83P04043 5 <NA> 2 NA 75 NA
## 5 570102 83P04044 5 <NA> 75 NA 250 NA
## 6 570102 83P04044 10 <NA> 2 NA 75 NA
## fraground fraghard
## 1 Subrounded Indurated
## 2 Subrounded Indurated
## 3 Subangular Indurated
## 4 <NA> <NA>
## 5 <NA> <NA>
## 6 <NA> <NA>
##
## $phpores
## phiid labsampnum poreqty poresize poreshp
## 1 820051 05N02182 5 Fine Interstitial
## 2 820051 05N02182 7 Medium Tubular
## 3 820052 05N02183 5 Very fine Dendritic tubular
## 4 820052 05N02183 6 Medium Tubular
## 5 820053 05N02184 5 Very fine Dendritic tubular
## 6 519185 40A23590 21 Very fine and fine <NA>
##
## $phstructure
## phiid labsampnum structgrade structsize structtype structid structpartsto
## 1 820051 05N02182 Moderate Fine Granular NA NA
## 2 820052 05N02183 Moderate Medium Subangular blocky NA NA
## 3 820053 05N02184 Moderate Medium Subangular blocky NA NA
## 4 519185 40A23590 <NA> <NA> Massive NA NA
## 5 518423 40A23592 <NA> <NA> Massive NA NA
## 6 518425 40A23594 <NA> <NA> Massive NA NA
##
## $pediagfeatures
## pedon_key peiid featdept featdepb featkind
## 1 11634 122235 0 5 Ochric epipedon
## 2 11634 122235 5 43 Cambic horizon
## 3 11634 122235 43 0 Lithic contact
## 4 32586 158467 0 18 Ochric epipedon
## 5 32586 158467 10 33 Cambic horizon
## 6 32586 158467 33 41 Lithic contact
##
## $rmf
## [,1]
## [1,] FALSEGraphically check the results with soil profile sketches. The Auburn soil is typically shallow to bedrock, limit sketches to 60cm.
par(mar = c(1,0,0,2))
plotSPC(pedons, color = 'moist_soil_color', print.id = FALSE, name = 'hzn_desgn', width = 0.3, name.style = 'center-center', max.depth = 60)
Simplification of fragment data is still a manual process.
# simplify fragment data
s.frags <- simplifyFragmentData(s$morph$phfrags, id.var = 'labsampnum')
# merge fragment data into SPC via "left" join
horizons(pedons) <- s.frags
# check
par(mar = c(0, 0, 3, 2))
plotSPC(pedons, color = 'total_frags_pct', print.id = FALSE, name = 'hzn_desgn', width = 0.3, max.depth = 60)
addVolumeFraction(pedons, 'total_frags_pct', pch = 1, cex.min = 0.25, cex.max = 0.5)
2.1 Fragment Volume >= 100%
There are cases where errors in the source data (NASIS
Pedon Horizon Fragments table) suggest coarse fragment
volume > 100%. As of soilDB 2.0.2 and aqp
1.15.4, the simplifyFragmentData() function will report
such errors and the addVolumeFraction() function will fail
gracefully by truncating fragment volume at 100%.
Give it a try with an example (thanks to Brian Gardner for finding it).
# get lab and morphologic data
s <- fetchKSSL(series = 'kettenbach', returnMorphologicData = TRUE)
# extract pedons into SoilProfileCollection
pedons <- s$SPC
# simplify fragment data
s.frags <- simplifyFragmentData(s$morph$phfrags, id.var = 'labsampnum')
# merge fragment data into SPC via left join
horizons(pedons) <- s.frags
# flag horizons with >= 100% total fragment volume
pedons$fragment_errors <- factor(pedons$total_frags_pct >= 100)
# label pedons with user pedon ID
# label horizons with total fragment %
par(mar = c(0, 0, 3, 5))
plotSPC(pedons, color = 'fragment_errors', name = 'total_frags_pct', label = 'pedon_id', cex.name = 0.85)
addVolumeFraction(pedons, colname = 'total_frags_pct', pch = 1, cex.min = 0.25, cex.max = 0.5)## Warning: 160 is >= 100, likely a data error## Warning: 180 is >= 100, likely a data error