Package index • soilKey

Core data structures

The unified profile carrier. Every classification, every provenance tag, every key-trace lives on one of these objects.

PedonRecord: PedonRecord: structured representation of a single pedon
DiagnosticResult: DiagnosticResult: structured outcome of a diagnostic test
ClassificationResult: ClassificationResult: structured outcome of running a key
FamilyAttribute: Classe S4-like para atributos de Familia (5o nivel SiBCS)
horizon_column_spec(): Canonical horizon column specification
make_empty_horizons(): Build an empty horizons data.table with the canonical schema
ensure_horizon_schema(): Coerce a horizons-like data.frame to the canonical schema
validate_horizon_geometry(): Validate horizon depth geometry

Classification entry points

One function per system. Each returns a ClassificationResult with the assigned name, the full key trace, the per-attribute provenance log, and the evidence grade.

classify_wrb2022(): Classify a pedon under WRB 2022
classify_sibcs(): Classifica um pedon segundo o SiBCS 5a edicao (1o + 2o + 3o + 4o niveis)
classify_sibcs_familia(): Classifica um perfil no 5o nivel categorico do SiBCS (Familia)
classify_usda(): Classify a pedon under USDA Soil Taxonomy (13th edition)
classify_all(): Classify a pedon across all three taxonomic systems
classify_from_documents(): Build a fully-classified `PedonRecord` from documents in one call
classify_from_photos(): Classify a soil profile from field photographs alone
classify_by_spectral_neighbours(): Classify a soil by spectral similarity to OSSL reference profiles
soil_classes_at_location(): Likely soil classes at a geographic location (spatial classification aid)
apply_soilgrids_depth_prior(): Fill missing horizon attributes from a SoilGrids depth prior
gapfill_within_pedon(): Fill interior missing horizon attributes by within-pedon depth interpolation
gapfill_derive_horizon(): Fill horizon attributes derivable BY DEFINITION from the same horizon
build_taxon_profiles(): Build per-taxon mean depth profiles for predicted-taxon gap-fill
gapfill_by_predicted_taxon(): Fill missing horizon attributes from the predicted taxon's mean profile
compute_per_attribute_evidence_grade(): Per-attribute provenance-aware evidence grade

Interoperability

Conversion to / from the canonical R soil-data ecosystem: aqp::SoilProfileCollection. Round-trip preserving.

as_aqp(): Convert one or more PedonRecord objects to an aqp SoilProfileCollection
from_aqp(): Convert an aqp SoilProfileCollection back to a list of PedonRecord

Interactive Shiny app

Drag-and-drop CSV web interface; renders all three classifications side-by-side and exports a self-contained HTML report.

run_classify_app(): Launch the soilKey interactive classification Shiny app

USDA Soil Taxonomy 13ed – diagnostic helpers

Soil Taxonomy diagnostic helpers used by the v0.9.27+ argillic / clay-films logic.

argillic_clay_films_test(): Test for clay-illuviation evidence (KST 13ed Ch 3 p 4)

Benchmark utilities

Benchmark drivers and KSSL/NASIS label normalisers.

canonicalise_kst13ed_gg(): Canonicalise a USDA Great Group label to a KST 13ed-compatible key
normalise_kssl_subgroup(): Normalise KSSL USDA subgroup labels for benchmark comparison

WRB 2022 – diagnostic horizons (Ch 3.1)

32/32 diagnostic horizons per IUSS Working Group WRB (2022) Ch 3.1.

argic(): Argic horizon (WRB 2022)
albic(): Albic horizon (WRB 2022)
cambic(): Cambic horizon (WRB 2022)
calcic(): Calcic horizon (WRB 2022)
duric_horizon(): Duric horizon (WRB 2022)
ferralic(): Ferralic horizon (WRB 2022)
ferric(): Ferric horizon (WRB 2022)
fragic(): Fragic horizon (WRB 2022): a high-bulk-density horizon with restricted rooting. v0.3.3: detects via bulk_density_g_cm3 >= 1.65 AND structure grade massive/very firm OR designation pattern x/Bx.
gypsic(): Gypsic horizon (WRB 2022)
histic_horizon(): Histic horizon (WRB 2022)
mollic(): Mollic horizon (WRB 2022)
natric_horizon(): Natric horizon (WRB 2022)
nitic_horizon(): Nitic horizon (WRB 2022)
petrocalcic(): Petrocalcic horizon (WRB 2022)
petroduric(): Petroduric horizon (WRB 2022): cemented duric.
petrogypsic(): Petrogypsic horizon (WRB 2022): cemented gypsic.
petroplinthic(): Petroplinthic horizon (WRB 2022): cemented plinthic.
pisoplinthic(): Pisoplinthic horizon (WRB 2022): pisolitic plinthic. v0.3.3 detects via designation pattern Bspl / Bvpi or via plinthite \>= 15% AND structure_type containing 'pisol'.
plinthic(): Plinthic horizon (WRB 2022)
salic(): Salic horizon (WRB 2022)
sombric(): Sombric horizon (WRB 2022): subsurface accumulation of humus that qualified neither as spodic nor as a true mollic-like horizon (low-base-saturation cool tropical highlands). v0.3.3 detects via designation pattern + OC criteria (BS < 50, OC > 0.6, depth > 25 cm).
spodic(): Spodic horizon (WRB 2022)
thionic(): Thionic horizon (WRB 2022): post-oxidation acid sulfate horizon. Requires sulfidic_s_pct >= 0.01 AND pH(H2O) <= 4.
umbric_horizon(): Umbric horizon (WRB 2022)
vertic_horizon(): Vertic horizon (WRB 2022 Ch 3.1)
chernic(): Chernic horizon (WRB 2022): the cherozemic-style mollic with very high biological activity (worm holes, casts, coprolites). v0.3.3: delegates to mollic + worm_holes_pct >= 50 (proxy for "biological homogenization").
tsitelic(): Tsitelic horizon (WRB 2022 Ch 3.1)
panpaic(): Panpaic horizon (WRB 2022 Ch 3.1)
limonic(): Limonic horizon (WRB 2022 Ch 3.1)
protovertic(): Protovertic horizon (WRB 2022 Ch 3.1)
anthraquic(): Anthraquic horizon (WRB 2022): puddled-rice / paddy plough layer. v0.3.3 detects via designation pattern Apl|Ap|Hh.
hydragric(): Hydragric horizon (WRB 2022): subsoil hydric horizon under anthraquic. v0.3.3 detects via designation pattern Bg|Brg immediately below an anthraquic-like topsoil.
hortic(): Hortic horizon (WRB 2022): garden / kitchen-midden topsoil. Diagnostic criteria: thickness \>= 20 cm, dark colour (mollic-like), high P (Mehlich-3 P >= 100 mg/kg or P2O5_1pct_citric >= 175 mg/kg), high SOC.
irragric(): Irragric horizon (WRB 2022): topsoil thickened by irrigation deposits. v0.3.3: thickness >= 20 cm + sediment-derived structure proxied via designation Apk|Apg|Au.
plaggic(): Plaggic horizon (WRB 2022): sod-derived topsoil >= 20 cm with low BD AND independent evidence of human input.
pretic(): Pretic horizon (WRB 2022): "Amazonian Dark Earth" (terra preta de indio) horizon – thick anthropogenic surface with high P, SOC, and incorporated charcoal / pottery.
terric(): Terric horizon (WRB 2022): topsoil thickened by long-term application of mineral material (sediment / sand additions). v0.3.3: thickness >= 20 cm + designation Au / Apc.

WRB 2022 – diagnostic properties (Ch 3.2)

abrupt_textural_difference(): Abrupt textural difference (WRB 2022 Ch 3.2.1)
albeluvic_glossae(): Albeluvic glossae (WRB 2022 Ch 3.2.2)
andic_properties(): Andic properties (WRB 2022)
anthric_horizons(): Anthric horizons (WRB 2022)
continuous_rock(): Continuous rock (WRB 2022 Ch 3.2.5)
gleyic_properties(): Gleyic properties (WRB 2022)
lithic_discontinuity(): Lithic discontinuity (WRB 2022 Ch 3.2.7)
protocalcic_properties(): Protocalcic properties (WRB 2022 Ch 3.2.8)
protogypsic_properties(): Protogypsic properties (WRB 2022 Ch 3.2.9): visible secondary gypsum \>= 1% but below the gypsic gate.
reducing_conditions(): Reducing conditions (WRB 2022 Ch 3.2.10) – per-pedon test wrapping test_reducing_conditions.
retic_properties(): Retic properties (WRB 2022)
shrink_swell_cracks(): Shrink-swell cracks (WRB 2022 Ch 3.2.12) – per-pedon test wrapping test_shrink_swell_cracks.
sideralic_properties(): Sideralic properties (WRB 2022 Ch 3.2.13)
stagnic_properties(): Stagnic properties (WRB 2022)
takyric_properties(): Takyric properties (WRB 2022 Ch 3.2.15) – per-pedon test wrapping test_takyric_surface.
vertic_properties(): Vertic properties (WRB 2022)
vitric_properties(): Vitric properties (WRB 2022 Ch 3.2.16)
yermic_properties(): Yermic properties (WRB 2022 Ch 3.2.17) – per-pedon test wrapping test_yermic_surface.

WRB 2022 – diagnostic materials (Ch 3.3)

aeolic_material(): Aeolic material (WRB 2022 Ch 3.3.1)
artefacts(): Artefacts (WRB 2022 Ch 3.3.2)
calcaric_material(): Calcaric material (WRB 2022 Ch 3.3.3): \>= 2% CaCO3 throughout the fine earth, primary carbonates from the parent material.
claric_material(): Claric material (WRB 2022 Ch 3.3.4): light-coloured fine earth with Munsell criteria.
dolomitic_material(): Dolomitic material (WRB 2022 Ch 3.3.5): \>= 2% Mg-rich carbonate, CaCO3/MgCO3 < 1.5. v0.3.3: detects via designation pattern kdo|do|magn as proxy when ratio data missing.
fluvic_material(): Fluvic material (WRB 2022)
gypsiric_material(): Gypsiric material (WRB 2022 Ch 3.3.7): \>= 5% gypsum that is primary (not secondary). Without a "secondary fraction" schema column, v0.3.3 treats any layer with caso4_pct >= 5 as gypsiric unless it explicitly carries gypsic-horizon designation.
hypersulfidic_material(): Hypersulfidic material (WRB 2022 Ch 3.3.8): \>= 0.01% inorganic sulfidic S, pH \>= 4, capable of severe acidification on aerobic incubation.
hyposulfidic_material(): Hyposulfidic material (WRB 2022 Ch 3.3.9): same inorganic sulfidic S and field pH as hypersulfidic but does NOT consist of hypersulfidic (criterion 3 – does not acidify to pH < 4 on aerobic incubation, usually self-neutralised by carbonate). Reachable from v0.9.128: when incubation_ph is measured, a sulfidic + pH>=4 layer that stays >= 4 on incubation is the set-complement of hypersulfidic_material and is reported here. Without an incubation pH the two cannot be told apart, so this returns empty (the layer is reported as potential hypersulfidic instead).
limnic_material(): Limnic material (WRB 2022 Ch 3.3.10): subaquatic deposits (coprogenous earth, diatomaceous earth, marl, gyttja). v0.3.3: detects via rock_origin %in% c("lacustrine", "marine") or designation pattern.
mineral_material(): Mineral material (WRB 2022 Ch 3.3.11): < 20% SOC AND < 35% volume artefacts containing >= 20% organic carbon. The complement of organic_material / organotechnic_material.
mulmic_material(): Mulmic material (WRB 2022 Ch 3.3.12): mineral material developed from organic material; \>= 8% SOC, with low BD, structural / chroma criteria.
organic_material(): Organic material (WRB 2022 Ch 3.3.13): \>= 20% SOC + recognisability criteria. v0.3.3: SOC threshold only.
organotechnic_material(): Organotechnic material (WRB 2022 Ch 3.3.14): \>= 35% volume of artefacts that themselves contain \>= 20% organic C. Soil itself has < 20% SOC.
ornithogenic_material(): Ornithogenic material (WRB 2022 Ch 3.3.15): bird-influenced topsoil. Mehlich-3 P >= 750 mg/kg + designation pattern Aornit|Bornit.
soil_organic_carbon(): Soil organic carbon (WRB 2022 Ch 3.3.16): organic C that does NOT belong to artefacts. v0.3.3: any layer with oc_pct >= 0.1 and artefacts_industrial_pct < 35.
solimovic_material(): Solimovic material (WRB 2022 Ch 3.3.17): hetero genous mass-movement material on slopes / footslopes (formerly "colluvic"). v0.3.3: detects via rock_origin == "colluvial" OR layer_origin == "solimovic".
technic_features(): Technic features (WRB 2022)
tephric_material(): Tephric material (WRB 2022 Ch 3.3.19): \>= 30% volcanic glass in 0.02-2 mm fraction AND no andic / vitric properties.

WRB 2022 – RSG-level gates

acrisol(): Acrisol RSG diagnostic (WRB 2022)
alisol(): Alisol RSG diagnostic (WRB 2022)
lixisol(): Lixisol RSG diagnostic (WRB 2022)
luvisol(): Luvisol RSG diagnostic (WRB 2022)
chernozem(): Chernozem RSG diagnostic (WRB 2022)
kastanozem(): Kastanozem RSG diagnostic (WRB 2022)
phaeozem(): Phaeozem RSG diagnostic (WRB 2022)
andosol(): Andosol RSG gate (WRB 2022 Ch 4, p 104)
chernozem_strict(): Chernozem RSG gate (strengthened, WRB 2022 Ch 4, p 111)
kastanozem_strict(): Kastanozem RSG gate (strengthened, WRB 2022 Ch 4, p 112)

SiBCS 5ª ed. – horizontes B diagnósticos (Cap 2)

Os gates de horizonte B públicos. As demais gates da chave SiBCS (carater_, horizonte_, atributos, qualifiers WRB, e os dispatchers por Ordem) são o motor determinístico interno, despachado por classify_sibcs() / classify_wrb2022(); veja a seção “Internal – motor taxonômico” no fim da referência.

B_textural(): Horizonte B textural (SiBCS Cap 2, p 54-57; v0.7 strict)
B_latossolico(): Horizonte B latossolico (SiBCS Cap 2, p 57-59; v0.7 strict)
B_nitico(): Horizonte B nitico (SiBCS Cap 2, p 61-62; v0.7)
B_espodico(): Horizonte B espodico (SiBCS Cap 2, p 62-65; v0.7)
B_incipiente(): Horizonte B incipiente (SiBCS Cap 2, p 59-61; v0.7)
B_planico(): Horizonte B planico (SiBCS Cap 2, p 65-66; v0.7)

SiBCS 5ª ed. – Família (5º nível, Cap 18)

O 5º nível categórico (família), via classify_sibcs(include_familia = TRUE).

classify_sibcs_familia(): Classifica um perfil no 5o nivel categorico do SiBCS (Familia)

USDA Soil Taxonomy 13ed – Família (5º nível)

Os modificadores de família (particle-size, mineralogy, CEC-activity, reaction, temperature, depth) via classify_usda(pedon, include_family = TRUE). Os gates de Ordem -> Subgrupo são internos (veja “Internal”).

classify_usda_family(): Classify the USDA family (5th level) of a pedon

Module 2 – VLM extraction

Schema-validated multimodal extraction via ellmer. The VLM never classifies; every extracted value carries provenance = extracted_vlm.

vlm_provider(): Construct a VLM provider chat object
MockVLMProvider: Mock VLM provider for testing
extract_horizons_from_pdf(): Extract horizons from a soil description PDF
extract_munsell_from_photo(): Extract Munsell color from a profile photo
extract_site_from_fieldsheet(): Extract site metadata from a field-sheet image

Module 3 – Spatial prior

SoilGrids + Embrapa raster. Used as a sanity-check; never overrides the deterministic key.

spatial_prior(): Spatial prior over RSGs (or Orders) at a pedon's location
spatial_prior_soilgrids(): SoilGrids spatial prior
spatial_prior_embrapa(): Embrapa national soil-class spatial prior (Brazil only)
prior_consistency_check(): Check consistency between a deterministic RSG assignment and a spatial prior

Module 4 – OSSL spectroscopy

Vis-NIR / MIR gap-fill. Predicted attributes carry provenance = predicted_spectra, which downgrades the evidence grade from A to B.

preprocess_spectra(): Pre-process Vis-NIR or MIR spectra
predict_ossl_mbl(): Memory-based learning prediction against the OSSL library
predict_ossl_plsr_local(): Local PLSR prediction against the OSSL library
predict_ossl_pretrained(): Pre-trained OSSL prediction
fill_from_spectra(): Fill missing soil attributes from spectra via OSSL
read_spectral_library(): Read a Vis-NIR / MIR reflectance + lab table into an OSSL-shaped library
pedons_from_spectral_table(): Build PedonRecords with attached Vis-NIR/MIR spectra from a table
benchmark_spectral_fill(): Benchmark the accuracy lift of spectral gap-fill (ON vs OFF), k-fold
pi_to_confidence(): Map a 95% prediction interval to a [0, 1] confidence score

Reporting

Render a complete pedologist-facing report from one or more ClassificationResult objects.

report(): Render a soilKey classification report
report_html(): Render a soilKey classification report as self-contained HTML
report_pdf(): Render a soilKey classification report as PDF
report_to_qgis(): Export a classification result + pedon to a QGIS GeoPackage

Layperson on-ramp

Zero-code Shiny GUI plus auto-detect helpers that remove the friction non-coders hit on a fresh install.

run_demo(): Launch the soilKey Shiny demo (one-screen GUI)
auto_set_proj_env(): Auto-detect PROJ_LIB and GDAL_DATA directories
vlm_pick_provider(): Pick the best available VLM provider
ollama_is_running(): Is the local Ollama HTTP API reachable?

Real-data benchmarks (v0.9.15+)

Loaders for the three external validation sets soilKey is benchmarked against in the v1.0 methods paper, plus FEBR / BDsolos label-normalisation helpers (v0.9.16).

load_kssl_pedons(): Load NCSS / KSSL pedons with reference USDA Soil Taxonomy classification
load_kssl_pedons_gpkg(): Load KSSL / NCSS pedons from the ncss_labdata GeoPackage
load_kssl_pedons_with_nasis(): Load KSSL pedons enriched with NASIS morphology
load_lucas_pedons(): Load EU-LUCAS / ESDB pedons with reference WRB classification
load_embrapa_pedons(): Load Embrapa dadosolos pedons with reference SiBCS classification
load_febr_pedons(): Load the Embrapa FEBR superconjunto into a list of PedonRecords
normalise_febr_sibcs(): Canonicalise FEBR SiBCS names to match soilKey rule outputs.
normalise_febr_wrb(): Normalise FEBR WRB taxon strings to RSG-only
normalise_febr_usda(): Normalise FEBR USDA taxon strings to USDA Soil Taxonomy Order
benchmark_run_classification(): Run a benchmark across one of the loaded pedon lists

Canonical fixtures

One-call constructors for the 31 canonical fixtures (one per WRB RSG, plus auxiliaries) used in the test suite.

make_acrisol_canonical(): Build the canonical Acrisol fixture
make_alisol_canonical(): Build the canonical Alisol fixture
make_andosol_canonical(): Build the canonical Andosol fixture
make_anthrosol_canonical(): Build the canonical Anthrosol fixture
make_arenosol_canonical(): Build the canonical Arenosol fixture
make_argissolo_canonical(): Perfil canonico de Argissolo (SiBCS 5a ed., Cap 5)
make_calcisol_canonical(): Build the canonical Calcisol fixture
make_cambisol_canonical(): Build the canonical Cambisol fixture
make_cambissolo_canonical(): Perfil canonico de Cambissolo (SiBCS 5a ed., Cap 6)
make_chernossolo_canonical(): Perfil canonico de Chernossolo (SiBCS 5a ed., Cap 7)
make_chernozem_canonical(): Build the canonical Chernozem fixture
make_cryosol_canonical(): Build the canonical Cryosol fixture
make_durisol_canonical(): Build the canonical Durisol fixture
make_espodossolo_canonical(): Perfil canonico de Espodossolo (SiBCS 5a ed., Cap 8)
make_ferralsol_canonical(): Build the canonical Ferralsol fixture
make_fluvisol_canonical(): Build the canonical Fluvisol fixture
make_gleissolo_canonical(): Perfil canonico de Gleissolo (SiBCS 5a ed., Cap 9)
make_gleysol_canonical(): Build the canonical Gleysol fixture
make_gypsisol_canonical(): Build the canonical Gypsisol fixture
make_histosol_canonical(): Build the canonical Histosol fixture
make_kastanozem_canonical(): Build the canonical Kastanozem fixture
make_latossolo_canonical(): Perfil canonico de Latossolo (SiBCS 5a ed., Cap 10)
make_leptosol_canonical(): Build the canonical Leptosol fixture
make_lixisol_canonical(): Build the canonical Lixisol fixture
make_luvisol_canonical(): Build the canonical Luvisol fixture
make_luvissolo_canonical(): Perfil canonico de Luvissolo (SiBCS 5a ed., Cap 11)
make_neossolo_canonical(): Perfil canonico de Neossolo Litolico (SiBCS 5a ed., Cap 12)
make_nitisol_canonical(): Build the canonical Nitisol fixture
make_nitossolo_canonical(): Perfil canonico de Nitossolo Vermelho (SiBCS 5a ed., Cap 13)
make_organossolo_canonical(): Perfil canonico de Organossolo (SiBCS 5a ed., Cap 14)
make_phaeozem_canonical(): Build the canonical Phaeozem fixture
make_planosol_canonical(): Build the canonical Planosol fixture
make_planossolo_canonical(): Perfil canonico de Planossolo (SiBCS 5a ed., Cap 15)
make_plinthosol_canonical(): Build the canonical Plinthosol fixture
make_plintossolo_canonical(): Perfil canonico de Plintossolo (SiBCS 5a ed., Cap 16)
make_podzol_canonical(): Build the canonical Podzol fixture
make_retisol_canonical(): Build the canonical Retisol fixture
make_solonchak_canonical(): Build the canonical Solonchak fixture
make_solonetz_canonical(): Build the canonical Solonetz fixture
make_stagnosol_canonical(): Build the canonical Stagnosol fixture
make_technosol_canonical(): Build the canonical Technosol fixture
make_umbrisol_canonical(): Build the canonical Umbrisol fixture
make_vertisol_canonical(): Build the canonical Vertisol fixture
make_vertissolo_canonical(): Perfil canonico de Vertissolo (SiBCS 5a ed., Cap 17)

Helpers and miscellaneous

run_all_benchmarks(): Run the full soilKey benchmark suite and (optionally) write a report
run_classify_app(): Launch the soilKey interactive classification Shiny app
run_demo(): Launch the soilKey Shiny demo (one-screen GUI)
run_sibcs_grande_grupo(): Resolve o grande grupo (3o nivel) de um pedon classificado em uma subordem SiBCS
run_sibcs_key(): Roda a chave SiBCS 5a edicao sobre um pedon
run_sibcs_subgrupo(): Resolve o subgrupo (4o nivel) de um pedon classificado em um Grande Grupo SiBCS
run_sibcs_subordem(): Resolve a subordem de um pedon ja classificado em uma ordem SiBCS
run_taxa_list(): Iterate a flat taxa list and evaluate tests in canonical order
run_taxonomic_key(): Run a taxonomic key (system-agnostic engine)
run_usda_great_group(): Run the USDA Great Group key for a given Suborder
run_usda_key(): Run the USDA Soil Taxonomy Order key over a pedon
run_usda_subgroup(): Run the USDA Subgroup key for a given Great Group
run_usda_suborder(): Run the USDA Suborder key for a given Order
run_wrb_key(): Run the WRB 2022 key over a pedon
compute_ki(): Ki (silica:alumina molar) – SiBCS Cap 1, p 32
compute_kr(): Kr (silica:sesquioxidos molar) – SiBCS Cap 1, p 32
compute_per_attribute_evidence_grade(): Per-attribute provenance-aware evidence grade
load_afsp_pedons(): Load Africa Soil Profiles (AfSP) v1.2 as PedonRecord objects
load_afsp_sample(): Load the bundled AfSP stratified sample (v0.9.77)
load_bdsolos_csv(): Load a BDsolos CSV export as a list of PedonRecord objects
load_embrapa_pedons(): Load Embrapa dadosolos pedons with reference SiBCS classification
load_febr_pedons(): Load the Embrapa FEBR superconjunto into a list of PedonRecords
load_kssl_nasis_sample(): Load the bundled KSSL + NASIS morphological-enriched sample (v0.9.75)
load_kssl_pedons(): Load NCSS / KSSL pedons with reference USDA Soil Taxonomy classification
load_kssl_pedons_gpkg(): Load KSSL / NCSS pedons from the ncss_labdata GeoPackage
load_kssl_pedons_with_nasis(): Load KSSL pedons enriched with NASIS morphology
load_kssl_sample(): Load the bundled KSSL/NCSS lab-data sample (v0.9.74)
load_lucas_pedons(): Load EU-LUCAS / ESDB pedons with reference WRB classification
load_lucas_soil_2018(): Load the LUCAS Soil 2018 Topsoil release as a list of PedonRecord objects
load_redape_pedons(): Load curated soil profiles from the Embrapa Redape GeoTab dataset
load_rules(): Load a soilKey rule set (YAML)
load_wosis_sample(): Load the bundled WoSIS South-America sample
load_wosis_stratified_sample(): Load the bundled WoSIS stratified RSG-balanced sample (v0.9.73)
save_ossl_models() load_ossl_models(): Save / load trained OSSL-backed PLSR models
apply_soilgrids_depth_prior(): Fill missing horizon attributes from a SoilGrids depth prior
format_wrb_name(): Format a WRB 2022 soil name with qualifiers
validate_horizon_geometry(): Validate horizon depth geometry
validate_pedon_json(): Validate a PedonRecord against the JSON schema
leptic_features(): Leptic features (WRB 2022)
planic_features(): Planic features (WRB 2022)
technic_features(): Technic features (WRB 2022)
andic_properties(): Andic properties (WRB 2022)
gleyic_properties(): Gleyic properties (WRB 2022)
protocalcic_properties(): Protocalcic properties (WRB 2022 Ch 3.2.8)
protogypsic_properties(): Protogypsic properties (WRB 2022 Ch 3.2.9): visible secondary gypsum \>= 1% but below the gypsic gate.
retic_properties(): Retic properties (WRB 2022)
sideralic_properties(): Sideralic properties (WRB 2022 Ch 3.2.13)
stagnic_properties(): Stagnic properties (WRB 2022)
takyric_properties(): Takyric properties (WRB 2022 Ch 3.2.15) – per-pedon test wrapping test_takyric_surface.
vertic_properties(): Vertic properties (WRB 2022)
vitric_properties(): Vitric properties (WRB 2022 Ch 3.2.16)
yermic_properties(): Yermic properties (WRB 2022 Ch 3.2.17) – per-pedon test wrapping test_yermic_surface.
`%||%`: Default-value-for-NULL operator
arenic_texture(): Arenic texture (WRB 2022)
cerosidade(): Cerosidade quantitativa (SiBCS Cap 13, p 207; Cap 1)
combine_priors(): Combine multiple spatial priors via weighted geometric mean
cryic_conditions(): Cryic conditions (WRB 2022)
distrofico(): Solo distrofico (SiBCS Cap 1, p 30)
duripa(): Duripa (SiBCS Cap 2, p 74; v0.7)
eutrofico(): Solo eutrofico (SiBCS Cap 1, p 30)
evaluate_rsg_tests(): Evaluate the test block of a single RSG
fibrico(): Material organico fibrico (SiBCS Cap 14)
fragipa(): Fragipa (SiBCS Cap 2, p 73-74; v0.7)
hemico(): Material organico hemico (SiBCS Cap 14)
oxic_usda(): Oxic horizon (USDA Soil Taxonomy)
vertic_horizon(): Vertic horizon (WRB 2022 Ch 3.1)
protovertic(): Protovertic horizon (WRB 2022 Ch 3.1)
panpaic(): Panpaic horizon (WRB 2022 Ch 3.1)
tsitelic(): Tsitelic horizon (WRB 2022 Ch 3.1)
limonic(): Limonic horizon (WRB 2022 Ch 3.1)
make_synthetic_pedon_with_spectra(): Build a synthetic PedonRecord with attached spectra (testing aid)
mudanca_textural_abrupta(): Mudanca textural abrupta (SiBCS Cap 1, p 30-31)
ossl_library_template(): Canonical schema for an `ossl_library` object
clear_ossl_cache(): Clear the soilKey OSSL cache
download_ossl_subset(): Download an OSSL subset and return an `ossl_library` artefact
download_ossl_subset_with_labels(): Download an OSSL subset and attach WRB / SiBCS / USDA labels
familia_mineralogia_argila_geral(): Familia: mineralogia da fracao argila (geral, nao-Latossolos)
ossl_demo_sa: Synthetic OSSL South America demo subset
planosol(): Planosol RSG gate (WRB 2022 Ch 4, p 107)
posterior_classify(): Bayesian posterior classifier (optional)
resolve_wrb_qualifiers(): Resolve WRB 2022 qualifiers for a Reference Soil Group
saprico(): Material organico saprico (SiBCS Cap 14)
soilgrids_usda_lut(): SoilGrids -> USDA Soil Order lookup table (placeholder)
soilgrids_wrb_lut(): SoilGrids -> WRB code lookup table
subgrupo_planossolo_espessos(): Subgrupo "espessos" de Planossolos (B planico profundo, > 100 cm)
subgrupo_planossolo_mesicos(): Subgrupo "mesicos" de Planossolos (B planico topo em [50, 100] cm)
subgrupo_plintossolo_endico_concrecionario(): Subgrupo "endico" de Plintossolos Concrecionarios (topo de horizonte concrecionario >= 40 cm)
subgrupo_plintossolo_endico_litoplintico(): Subgrupo "endico" de Plintossolos Litoplinticos (topo de horizonte litoplintico >= 40 cm)
subgrupo_plintossolo_espessos(): Subgrupo "espessos" de Plintossolos (horizonte plintico topo > 100 cm)

RSG-level gates and other materials

ferralsol(): Ferralsol RSG gate (WRB 2022 Ch 4, p 110)
gleysol(): Gleysol RSG gate (WRB 2022 Ch 4, p 103)
vertisol(): Vertisol RSG gate (WRB 2022 Ch 4, p 101)
acrisol(): Acrisol RSG diagnostic (WRB 2022)
alisol(): Alisol RSG diagnostic (WRB 2022)
lixisol(): Lixisol RSG diagnostic (WRB 2022)
luvisol(): Luvisol RSG diagnostic (WRB 2022)
chernozem(): Chernozem RSG diagnostic (WRB 2022)
chernozem_strict(): Chernozem RSG gate (strengthened, WRB 2022 Ch 4, p 111)
kastanozem(): Kastanozem RSG diagnostic (WRB 2022)
kastanozem_strict(): Kastanozem RSG gate (strengthened, WRB 2022 Ch 4, p 112)
phaeozem(): Phaeozem RSG diagnostic (WRB 2022)
andosol(): Andosol RSG gate (WRB 2022 Ch 4, p 104)
calcaric_material(): Calcaric material (WRB 2022 Ch 3.3.3): \>= 2% CaCO3 throughout the fine earth, primary carbonates from the parent material.
dolomitic_material(): Dolomitic material (WRB 2022 Ch 3.3.5): \>= 2% Mg-rich carbonate, CaCO3/MgCO3 < 1.5. v0.3.3: detects via designation pattern kdo|do|magn as proxy when ratio data missing.
hypersulfidic_material(): Hypersulfidic material (WRB 2022 Ch 3.3.8): \>= 0.01% inorganic sulfidic S, pH \>= 4, capable of severe acidification on aerobic incubation.
hyposulfidic_material(): Hyposulfidic material (WRB 2022 Ch 3.3.9): same inorganic sulfidic S and field pH as hypersulfidic but does NOT consist of hypersulfidic (criterion 3 – does not acidify to pH < 4 on aerobic incubation, usually self-neutralised by carbonate). Reachable from v0.9.128: when incubation_ph is measured, a sulfidic + pH>=4 layer that stays >= 4 on incubation is the set-complement of hypersulfidic_material and is reported here. Without an incubation pH the two cannot be told apart, so this returns empty (the layer is reported as potential hypersulfidic instead).
solimovic_material(): Solimovic material (WRB 2022 Ch 3.3.17): hetero genous mass-movement material on slopes / footslopes (formerly "colluvic"). v0.3.3: detects via rock_origin == "colluvial" OR layer_origin == "solimovic".
technic_hard_material(): Technic hard material (WRB 2022 Ch 3.3.18): consolidated human-made material (asphalt, concrete, worked stones).
tephric_material(): Tephric material (WRB 2022 Ch 3.3.19): \>= 30% volcanic glass in 0.02-2 mm fraction AND no andic / vitric properties.

Engine selection and dispatch (v0.9.65)

Per-pedon engine-selection heuristic, engine-aware dispatch (soilKey vs aqp), and side-by-side comparators.

pick_engine(): Choose the best diagnostic engine for a single pedon
pick_engine_batch(): Per-pedon batch engine recommendation
classify_with_engine_heuristic(): Classify a pedon with the engine chosen by `pick_engine()`
compare_engines(): Side-by-side comparison of soilKey vs aqp diagnostic engines
argic_aqp(): Argic / argillic horizon via aqp::getArgillicBounds()
cambic_aqp(): Cambic horizon via aqp::getCambicBounds()

Canonical references (v0.9.62 – v0.9.65)

Vendored WRB 2022 / KST 13 / Soil Taxonomy criteria and shared lookup helpers for pkg vs vendored data sources.

canonical_reference(): Load a canonical reference dataset from soilKey or SoilTaxonomy
wrb2022_canonical(): WRB 2022 canonical reference (parsed IUSS Working Group WRB 2022)
kst13_canonical(): Keys to Soil Taxonomy 13th edition canonical reference
kst13_codes(): Load the canonical KST 13ed code -> taxon-name lookup table
kst13_criteria(): Load the canonical KST 13ed criteria for a single taxon code
st_features_canonical(): USDA Soil Taxonomy diagnostic features canonical table
clear_kst13_cache(): Clear the in-memory KST13 cache
coverage_report(): Honest taxonomic-completeness report

SmartSolos / Embrapa AgroAPI integration (v0.9.54)

Cross-validation against Embrapa’s PROLOG SiBCS classifier.

classify_via_smartsolos_api(): Classify a PedonRecord via Embrapa's SmartSolosExpert REST API
compare_smartsolos(): Cross-validate the local SiBCS classifier against the SmartSolosExpert API

BDsolos loader and benchmarks (v0.9.55 – v0.9.60)

benchmark_bdsolos(): Benchmark soilKey classifiers against BDsolos national reference labels
download_bdsolos(): Download the BDsolos consulta-publica CSV (experimental, requires chromote)
inspect_bdsolos_csv(): Diagnostic inspection of a BDsolos CSV before loading

Argic strong-films audit (v0.9.83)

Empirical audit of the SiBCS Cap 18 latossolic-vs-argic precedence rule. Extracts the strong-clay-films decision into a reusable helper so the rule can be validated on any benchmark dataset (BDsolos RJ: 0.9% Latossolo false-positive exclusion rate, 37.6% Argissolo correct retention rate).

argic_with_strong_clay_films(): Test whether a pedon's argic horizon has strong clay films
audit_argic_strong_films(): Audit the strong-clay-films exclusion across a list of pedons

Lazy-fetch benchmark caches (v0.9.94)

The four large benchmark caches (AfSP, KSSL, KSSL+NASIS, WoSIS stratified, ~1 MB each) are no longer bundled in the CRAN source tarball. They are downloaded on demand from a versioned GitHub Release into the user cache (~/Library/Application Support/…/soilKey/data on macOS) on first call.

download_extdata_cache(): Download one or more soilKey lazy-fetch caches from GitHub Release

FEBR loader (v0.9.57)

UFSM Free Brazilian Repository readers (~249 datasets, ~36k horizons).

read_febr_pedons(): Load FEBR datasets as a list of PedonRecord objects
febr_index_munsell(): Curated index of FEBR datasets that carry Munsell colors

Redape curated GeoTab dataset (v0.9.71)

Embrapa Redape repository (DOI 10.48432/PYKKA7) – 96 hand- reviewed Brazilian soil profiles, suitable as a gold-standard benchmark for SiBCS classification.

download_redape_dataset(): Download the curated Redape GeoTab dataset (Vaz et al 2023)
load_redape_pedons(): Load curated soil profiles from the Embrapa Redape GeoTab dataset
benchmark_redape(): Benchmark soilKey SiBCS predictions against the Redape gold standard

WoSIS stratified WRB benchmark (v0.9.73)

ISRIC WoSIS bundled cache, 130 profiles balanced across 26 WRB Reference Soil Groups (5 per RSG). The first WRB benchmark with profile depth (analog of Redape but for WRB), pulled via RSG-filtered GraphQL queries.

load_wosis_stratified_sample(): Load the bundled WoSIS stratified RSG-balanced sample (v0.9.73)

USDA Soil Taxonomy <-> WRB cross-walk (v0.9.74)

KSSL/NCSS Lab Data Mart benchmark via IUSS WRB 2022 Annex 6 USDA Order/Suborder -> WRB RSG correlation. Provides the richest WRB benchmark to date (CEC 65%, Ca/Mg/K/Na 40-56% vs WoSIS-stratified ~30%).

usda_to_wrb_rsg(): USDA Soil Taxonomy <-> WRB Reference Soil Group correlation table
annotate_wrb_from_usda(): Annotate KSSL/NASIS pedons with a derived WRB Reference Soil Group
benchmark_wrb_vs_usda(): Benchmark soilKey WRB predictions against a USDA-derived ground truth
load_kssl_sample(): Load the bundled KSSL/NCSS lab-data sample (v0.9.74)

KSSL + NASIS morphological-enriched sample (v0.9.75)

Joins KSSL lab gpkg with NASIS Morphological sqlite – adds Munsell colours, structure, clay films, slickensides (Munsell 0% -> 89.6% vs lab-only).

load_kssl_nasis_sample(): Load the bundled KSSL + NASIS morphological-enriched sample (v0.9.75)

Africa Soil Profiles (AfSP) WRB benchmark (v0.9.77)

ISRIC AfSP v1.2 (Leenaars et al. 2014) – 18,533 georeferenced African profiles, ~7000 with WRB 2006 RSG. Loader + bundled 120-pedon stratified sample (5 per RSG x 24 RSGs). Achieves 28% Order accuracy with strong Cambisol/Histosol (100%) and Ferralsol/Solonetz (80%) recall.

load_afsp_pedons(): Load Africa Soil Profiles (AfSP) v1.2 as PedonRecord objects
load_afsp_sample(): Load the bundled AfSP stratified sample (v0.9.77)
benchmark_afsp(): Benchmark soilKey WRB predictions against AfSP ground truth
wrb06_code_to_rsg(): WRB 2006 RSG code -> 2022 RSG name

LUCAS Soil 2018 (v0.9.49 – v0.9.50)

benchmark_lucas_2018(): Run the LUCAS Soil 2018 / ESDB WRB benchmark
attach_lucas_spectra(): Attach LUCAS 2018 Vis-NIR spectra to a list of PedonRecord objects

Unified benchmark and robustness (v0.9.61)

benchmark_unified(): Unified cross-dataset benchmark across SiBCS / WRB / USDA
run_all_benchmarks(): Run the full soilKey benchmark suite and (optionally) write a report
benchmark_performance(): Run the soilKey performance benchmark
batch_robustness(): Batch robustness across many pedons
classification_robustness(): Robustness of classification under input perturbation
classify_with_uncertainty(): Posterior distribution over classification outcomes
get_perturbation_scale(): Monte-Carlo perturbation scale for an evidence grade

OSSL spectra: PLS training and prediction

Build PLS regressors from OSSL Vis-NIR/SWIR spectra and predict lab properties / Munsell / XYZ / generic targets.

train_pls_from_ossl(): Train pre-trained PLSR models from an OSSL library
predict_from_spectra(): Predict soil properties from spectra
predict_lab_from_spectra(): Predict CIE Lab from Vis-NIR reflectance spectra
predict_munsell_from_spectra(): Predict Munsell hue / value / chroma from Vis-NIR reflectance spectra
predict_xyz_from_spectra(): Predict CIE XYZ tristimulus values from Vis-NIR reflectance spectra
predict(<soilKey_pls_model>): Predict from a soilKey_pls_model
print(<soilKey_pls_model>): Print method for soilKey_pls_model
fill_munsell_from_spectra(): Fill missing Munsell colors on a PedonRecord from Vis-NIR spectra

Spatial / database lookups

ESDB attributes, SoilGrids and MapBiomas-Solos pulls.

available_esdb_attributes(): List ESDB Raster Library attributes available at a given root
lookup_esdb(): Look up an ESDB raster value at WGS84 coordinates
lookup_soilgrids(): Look up a SoilGrids 250m soil property at WGS84 coordinates
lookup_mapbiomas_solos(): Look up a MapBiomas Solos raster value at WGS84 coordinates

GlobalSoilMap depth harmonisation and aqp helpers

GSM_DEPTHS: Default GlobalSoilMap depth intervals (cm)
harmonize_to_gsm(): Harmonise pedons to GlobalSoilMap depth intervals
pedon_to_spc(): Convert a soilKey PedonRecord to an aqp SoilProfileCollection
pedon_json_schema(): JSON Schema for a soilKey PedonRecord
texture_class_from_pct(): NRCS texture-class shorthand from clay / silt / sand percent

Internal – motor taxonômico

Os preditores atômicos do motor determinístico – qualifiers WRB (qual_*), gates de Subgrupo/Grande-Grupo USDA (*_usda), atributos e horizontes diagnósticos SiBCS (carater_*, horizonte_*), e os dispatchers por Ordem. Continuam exportados e chamáveis, mas são despachados por nome pela chave (classify_*); não fazem parte da API pública curada.

MockVLMProvider: Mock VLM provider for testing
.BDSOLOS_COLUMN_PATTERNS: Canonical mapping from BDsolos column-name variants to soilKey schema
.BDSOLOS_SITE_PATTERNS: Site-level columns (BDsolos full export). Mapped at the site, not horizon, level.
.FEBR_TO_HORIZON_MAP: Map FEBR layer-table columns to soilKey horizon attributes
.GLEYIC_HUE_REGEX: Gleyic Munsell hue patterns (WRB 2022, Ch 3.1.13 redoximorphic features)
.KST13_CACHE: Package-level cache for the parsed KST 13ed JSON files
.REDAPE_API_BASE: Embrapa Redape Dataverse API endpoint
.REDAPE_GEOTAB_DOI: Default DOI for the Vaz et al. 2023 curated GeoTab dataset
.SIBCS_LEGACY_ORDER_MAP: Pre-2018 SiBCS Order names -> SiBCS 5a edicao plural Title Case map
.SMARTSOLOS_DRAINAGE_SCALE: SmartSolos drainage class scale (DRENAGEM, 1-8)
.SOILGRIDS_TO_HORIZON_MAP: Mapping of SoilGrids 250m property names to soilKey horizon columns
.SOILKEY_LAZY_FETCH_CACHES: Caches managed by the v0.9.94 lazy-fetch system
.SOILKEY_LAZY_FETCH_RELEASE: Versioned GitHub Release tag where the lazy-fetch caches are pinned
.WRB_LV1_NAME_BY_CODE: WRB Reference Soil Group code-to-name table
soilKey soilKey-package: soilKey: Automated Soil Profile Classification per WRB 2022 and SiBCS