I periodically "process" the NCSS-KSSL characterization data snapshot (usually quarterly) into a consolidated chunk of data that are used within SoilWeb and by fetchKSSL(). This snapshot is typically delivered as an Access database and contains a mixture of: the latest "lab" data from LIMS, and the latest taxonomic and spatial data from NASIS. The resulting "processed" data include over 50 attributes, split into chunks that roughly approximate the "site/pedon" and "horizon" objects.
There is a new ESRI FileGeoDatabase KSSL snapshot available. There will soon be an SQLite version as well.
Snapshots:
- KSSL lab, taxonomic, and location data: 2018-09-25 QC results
- NASIS morphologic data: 2018-10-23
- 2018-12-21: new NASIS snapshot from October 2018
- 2018-09-26: new KSSL snapshot from September 2018 (64732 pedons, 416567 horizons)
- 2018-04-26: now more
wrd_l2values, thanks to Jacob Isleib, closes issue #4 - 2017-09-13: new KSSL / NASIS snapshot from September 2017 (64153 pedons, 413385 horizons)
- 2017-09-11: complete overhaul, now based on FGDB -> TXT -> SQLite workflow
- 2017-04-24: fixed 2 major bugs: 1) data were shifted to the right by one column after "frags", 2) smarter estimated OC calculation
- 2016-11-17: added VG parameters estimated by Rosetta
- 2016-09-06: new KSSL snapshot from 2016-09-06 (63571 pedons, 408290 horizons)
- 2016-04-22: added new code + web-service for basic NASIS morphologic data; data returned as JSON
- 2016-01-15: added a new column with fragments (percent by weight) > 2mm. use these values with caution
- 2015-12-18: started processing new snapshot "December 2015" (62922 pedons, 401427 horizons) RaCA sites no longer included
- 2015-12-11: loaded snapshot "August 2015" (64071 pedons, 402199 horizons)
As part of the "processing" of these data, a number of data cleaning operations are performed.
A new field caled taxonname is added. This field is set to the correlated_as value when not null and not "unnamed". Otherwise, the taxonname field is set to the value in sampled_as.
- Missing lower horizon depths are replaced with the corrosponding top depth + 1cm.
- "O" horizons using the old-style notation, with top depths > bottom depths, are removed. Sorry.
State and MLRA codes are added to the data using spatial overlay with the most recent US states and NRCS MLRA maps.
- K saturation is computed via
ex_k / base_sum - total C and N values of 0 are replaced by NA
- organic C estimated via
with(h, ifelse(is.na(oc), c_tot - (ifelse(is.na(caco3), 0, caco3) * 0.12), oc)) - organic matter estimated via
estimated_oc * 1.724 - C:N estimated via
h$estimated_oc / h$n_tot - pH (1:1 H2O) estimated when missing via saturated paste pH (pedotransfer function)
- base saturation (pH 8.2) calculated when missing:
h$bs82.computed <- with(h, (ex_ca + ex_mg + ex_na + ex_k) / (ex_ca + ex_mg + ex_na + ex_k + acid_tea)) * 100
Horizon Attributes:
-
pedon_key(internal ID) -
labsampnum(links KSSL and NASIS records) -
hzn_top(horizon top depth: cm) -
hzn_bot(horizon bottom depth: cm) -
hzn_desgn(horizon designation, updated) -
hzn_desgn_old(original horizon designation) -
lab_texture_class(lab determined texture class) -
sand(sand, wt. %) -
silt(silt, wt. %) -
clayClyT_p: Total clay is the soil separate with <0.002 mm particle diameter. Clay size carbonate is included. Total clay is reported as a weight percent of the <2 mm fraction. -
co3_cly(carbonate clay ??) -
silt_f_psa(fine silt, wt. %) -
silt_c_psa(coarse silt, wt. %) -
vfs(very fine sand, wt. %) -
fs(fine sand, wt. %) -
ms(medium sand, wt. %) -
cs(coarse sand, wt. %) -
vcs(very coarse sand, wt. %) acid_teabase_sum-
al_kcl: KCl extractable aluminum approximates the exchangeable Al, and is a measure of the active acidity present in soils with a 1:1 water pH less than 5.5. It relates to the immediate lime requirement and the CEC of the soil. cmol(+)/kg -
cec7(CEC measured at pH 7, ... details) -
cec82(CEC measured at pH 8.2, ... details) -
ecec: cmol(+)/kg -
al_sat: Aluminum saturation is calculated by (AL_KCL/(Sum of bases))*100. It provides some inference of potential Al toxicity problems, although many other factors influence Al toxicity, % -
bs82(base saturation, pH 8.2, ... details) -
bs7(base saturation, pH 7, ... details) ex_caex_mgex_naex_kph_h2oph_cacl2ph_kclph_sp-
ph_ox(pH, Oxidized) gypl20caco3ec_12presar-
ocOrganic carbon is a measure of all organic forms of carbon in the soil, including organic carbon within minerals. % wt c_totn_tot-
fe_dith(weight percent, <2mm fraction) -
fe_ox(weight percent, <2mm fraction) -
al_dith(weight percent, <2mm fraction) -
al_ox(weight percent, <2mm fraction) p_olsnp_nzdb_13bdb_odCOLEwswhcw3cldw15l2w15clycec7_cly-
frags(weight percentage > 2mm) -
wrd_l2(Water Retention Difference, 1/3 to 15 Bar, <2mm: usually calculated from other properties) -
estimated_oc(wt. %) see above for details -
estimated_om(wt. %) see above for details -
estimated_c_to_n(unitless) -
ex_k_saturation(%) estimated_ph_h2o-
theta_r$\theta_{r}$ residual water content, parameter to van Genuchten model, as estimated by Rosetta -
theta_s$\theta_{s}$ saturated water content, parameter to van Genuchten model, as estimated by Rosetta -
alpha$log_{10}(\alpha)$ inverse of air entry suction, parameter to van Genuchten model, as estimated by Rosetta -
npar$log_{10}(n)$ measure of pore-size distribution, parameter to van Genuchten model, as estimated by Rosetta -
Ks$K_{s}$ saturated hydraulic conductivity, as estimated by Rosetta KoLpar
Site/Pedon Attributes:
- pedon_key
- pedlabsampnum
- pedon_id
- x
- y
- sampled_as
- correlated_as
- correlated_taxon_kind
- pscs_top
- pscs_bottom
- pedon_completeness_index
- ssl_taxsubgroup
- taxonname
- state
- mlra
Morphologic (field-described) Attributes:
- horizon colors (phcolor table)
- rock fragments -- volume percent (phfrags table)
- pores (phpores table)
- structure (phstructure)
- NASIS pedon/site/siteobs distillation
- NASIS taxonomic history disillation