← Back to the catalog
Private

ossl lucas woodwell mir soil all y

ossl · NIR

ossl lucas woodwell mir soil all y. v2.0 standardized NIRS package: 1 spectral source(s), 53 declared target(s). Auto-generated from dataset_card.json (verify before publication).

nirv2ossl
🔒
Private dataset. Full metadata and metrics are shown, but the bytes are not redistributed here — exporting the data requires a Dataverse token. The identity card carries no spectra, only descriptive statistics.
589
samples
1,701
wavelengths
1
sources
53
targets
31
metadata
NIR
family

Dataset property explorer

Mean profile risk0.48
Highest axisArtefacts locaux · 1.00
Diagnostics8
Sources profiled1
ossl lucas woodwell mir soil all y property profile0.250.50.751integritynoiseartefactsbaselinePCA outliersreferencerepeatabilitystructureossl lucas woodwell mir soil all y profileintegrity: 0.00noise: 0.00artefacts: 1.00baseline: 0.86PCA outliers: 0.44reference: 0.98repeatability: 0.00structure: 0.58ossl lucas wood…0 center · 1 outer ring · outward = stronger anomaly / heterogeneity signal

Profile axes

Intégrité0.00
Artefacts locaux1.00
Bruit0.00
Outliers PCA0.44
Distance à la référence0.98
Répétabilité0.00
Baseline / forme0.86
Structure multi-régimes0.58
Diagnostic hypotheses00.250.50.751hypothesis scoreSplice / raccord détecteursSplice / raccord détecteurs: 0.820.82Erreur calibration / référenc…Erreur calibration / référence blanche: 0.670.67Erreur interpolation / réécha…Erreur interpolation / rééchantillonnage: 0.620.62Signature VERA25-likeSignature VERA25-like: 0.600.60Fond différentFond différent: 0.600.60Différence de sonde / géométr…Différence de sonde / géométrie: 0.540.54Dataset multi-régimesDataset multi-régimes: 0.450.45Spectre hors domaine valideSpectre hors domaine valide: 0.450.45
DiagnosticScoreForceSignauxInterprétation probable
Splice / raccord détecteursX0.82forteSpike rate 1.00, Jump rate 1.00, SNR non dégradé 1.00Rupture aux jonctions de détecteurs, calibration locale ou sonde différente.
Erreur calibration / référence blancheX0.67moyenneartefacts locaux 1.00, RMS/SAM référence 0.98, Baseline/mean/area 0.86Décalage systématique entre campagnes, instruments ou référence blanche.
Erreur interpolation / rééchantillonnageX0.62moyenneSpike rate 1.00, Jump rate 1.00, SNR normal/élevé 1.00Artefacts numériques ou traitement spectral incorrect.
Signature VERA25-likeX0.60moyenneSpike rate 1.00, Jump rate 1.00, RMS/SAM référence 0.98Combinaison possible changement de sonde + splice, amplifiée par géométrie, fond ou calibration.
Fond différentX0.60moyenneRMS/SAM référence 0.98, Baseline/mean/area 0.86, PCA Q 0.44Effet systématique du support, blanc/noir, transflectance ou environnement de mesure.
Différence de sonde / géométrieX0.54moyenneRMS/SAM référence 0.98, Baseline/mean/area 0.86, PCA Q 0.44Modification de l'illumination, collecte, angle ou distance sonde-échantillon.
Dataset multi-régimesX0.45moyenneRMS/SAM référence 0.98, Structure PCA 0.58, PCA Q 0.44Mélange de campagnes, opérateurs, lots, setups ou sous-populations spectrales.
Spectre hors domaine valideX0.45moyenneRMS/SAM référence 0.98, Structure PCA 0.58, Mahalanobis / T2 0.44Variété, espèce, lot ou condition différente mais physiquement plausible.

Spectral sources

recovered_spectra

X · NIR · unknown
recovered_spectra spectra1.01.52.02.53.001,0002,0003,0004,000q05-q95 envelopeq25-q75 envelopemedian spectrummedianq25–q75q05–q95wavelength / none600none — median 2.202 (q25–q75 2.055–2.341)624none — median 2.244 (q25–q75 2.116–2.381)648none — median 2.268 (q25–q75 2.125–2.395)674none — median 2.249 (q25–q75 2.107–2.376)698none — median 2.34 (q25–q75 2.241–2.458)722none — median 2.217 (q25–q75 2.075–2.323)746none — median 2.171 (q25–q75 2.058–2.287)772none — median 2.167 (q25–q75 2.047–2.281)796none — median 2.196 (q25–q75 2.065–2.301)820none — median 2.229 (q25–q75 2.142–2.312)844none — median 2.069 (q25–q75 1.973–2.17)870none — median 2.036 (q25–q75 1.926–2.139)894none — median 2.045 (q25–q75 1.918–2.141)918none — median 2.093 (q25–q75 1.965–2.198)942none — median 2.086 (q25–q75 1.964–2.188)966none — median 2.076 (q25–q75 1.942–2.187)992none — median 2.053 (q25–q75 1.918–2.198)1,016none — median 2.016 (q25–q75 1.875–2.179)1,040none — median 1.957 (q25–q75 1.809–2.128)1,064none — median 1.896 (q25–q75 1.702–2.07)1,090none — median 1.841 (q25–q75 1.636–2.054)1,114none — median 1.874 (q25–q75 1.661–2.078)1,138none — median 1.858 (q25–q75 1.657–2.061)1,162none — median 1.929 (q25–q75 1.733–2.114)1,188none — median 1.843 (q25–q75 1.645–2.043)1,212none — median 1.786 (q25–q75 1.609–1.99)1,236none — median 1.831 (q25–q75 1.671–2.028)1,260none — median 1.965 (q25–q75 1.804–2.14)1,284none — median 2.102 (q25–q75 1.96–2.223)1,310none — median 2.186 (q25–q75 2.069–2.285)1,334none — median 2.218 (q25–q75 2.063–2.332)1,358none — median 2.205 (q25–q75 2.056–2.344)1,382none — median 2.199 (q25–q75 2.064–2.335)1,408none — median 2.171 (q25–q75 2.072–2.3)1,432none — median 2.156 (q25–q75 2.057–2.273)1,456none — median 2.115 (q25–q75 2.013–2.23)1,480none — median 2.055 (q25–q75 1.956–2.183)1,506none — median 2.031 (q25–q75 1.931–2.145)1,530none — median 2.018 (q25–q75 1.922–2.144)1,554none — median 1.97 (q25–q75 1.865–2.112)1,578none — median 2.024 (q25–q75 1.923–2.17)1,602none — median 2.127 (q25–q75 2.027–2.262)1,628none — median 2.173 (q25–q75 2.057–2.305)1,652none — median 2.134 (q25–q75 2.024–2.286)1,676none — median 2.072 (q25–q75 1.968–2.206)1,700none — median 1.943 (q25–q75 1.843–2.071)1,726none — median 1.82 (q25–q75 1.726–1.932)1,750none — median 1.737 (q25–q75 1.654–1.838)1,774none — median 1.718 (q25–q75 1.655–1.825)1,798none — median 1.739 (q25–q75 1.658–1.83)1,824none — median 1.623 (q25–q75 1.562–1.7)1,848none — median 1.621 (q25–q75 1.555–1.692)1,872none — median 1.674 (q25–q75 1.591–1.774)1,896none — median 1.626 (q25–q75 1.549–1.717)1,920none — median 1.531 (q25–q75 1.474–1.606)1,946none — median 1.521 (q25–q75 1.46–1.593)1,970none — median 1.523 (q25–q75 1.458–1.602)1,994none — median 1.521 (q25–q75 1.451–1.601)2,018none — median 1.472 (q25–q75 1.409–1.545)2,044none — median 1.404 (q25–q75 1.336–1.486)2,068none — median 1.358 (q25–q75 1.288–1.456)2,092none — median 1.345 (q25–q75 1.274–1.444)2,116none — median 1.335 (q25–q75 1.264–1.435)2,142none — median 1.337 (q25–q75 1.267–1.433)2,166none — median 1.317 (q25–q75 1.246–1.42)2,190none — median 1.306 (q25–q75 1.236–1.412)2,214none — median 1.309 (q25–q75 1.24–1.411)2,238none — median 1.313 (q25–q75 1.247–1.411)2,264none — median 1.301 (q25–q75 1.231–1.402)2,288none — median 1.284 (q25–q75 1.216–1.395)2,312none — median 1.28 (q25–q75 1.208–1.392)2,336none — median 1.282 (q25–q75 1.211–1.392)2,362none — median 1.277 (q25–q75 1.205–1.39)2,386none — median 1.274 (q25–q75 1.202–1.385)2,410none — median 1.274 (q25–q75 1.2–1.384)2,434none — median 1.272 (q25–q75 1.2–1.382)2,458none — median 1.277 (q25–q75 1.203–1.388)2,484none — median 1.304 (q25–q75 1.227–1.408)2,508none — median 1.338 (q25–q75 1.245–1.456)2,532none — median 1.332 (q25–q75 1.242–1.446)2,556none — median 1.319 (q25–q75 1.237–1.428)2,582none — median 1.322 (q25–q75 1.242–1.43)2,606none — median 1.322 (q25–q75 1.247–1.425)2,630none — median 1.314 (q25–q75 1.238–1.427)2,654none — median 1.315 (q25–q75 1.241–1.431)2,680none — median 1.324 (q25–q75 1.247–1.444)2,704none — median 1.336 (q25–q75 1.255–1.454)2,728none — median 1.347 (q25–q75 1.267–1.465)2,752none — median 1.36 (q25–q75 1.279–1.479)2,776none — median 1.375 (q25–q75 1.295–1.497)2,802none — median 1.399 (q25–q75 1.314–1.518)2,826none — median 1.426 (q25–q75 1.34–1.543)2,850none — median 1.487 (q25–q75 1.398–1.602)2,874none — median 1.503 (q25–q75 1.416–1.62)2,900none — median 1.514 (q25–q75 1.428–1.639)2,924none — median 1.552 (q25–q75 1.466–1.687)2,948none — median 1.552 (q25–q75 1.465–1.685)2,972none — median 1.56 (q25–q75 1.467–1.686)2,998none — median 1.556 (q25–q75 1.461–1.684)3,022none — median 1.569 (q25–q75 1.471–1.7)3,046none — median 1.594 (q25–q75 1.489–1.72)3,070none — median 1.615 (q25–q75 1.51–1.748)3,094none — median 1.634 (q25–q75 1.533–1.774)3,120none — median 1.66 (q25–q75 1.551–1.804)3,144none — median 1.686 (q25–q75 1.573–1.83)3,168none — median 1.711 (q25–q75 1.597–1.862)3,192none — median 1.738 (q25–q75 1.615–1.883)3,218none — median 1.763 (q25–q75 1.64–1.91)3,242none — median 1.782 (q25–q75 1.653–1.931)3,266none — median 1.796 (q25–q75 1.665–1.946)3,290none — median 1.809 (q25–q75 1.674–1.957)3,316none — median 1.819 (q25–q75 1.686–1.971)3,340none — median 1.832 (q25–q75 1.697–1.986)3,364none — median 1.846 (q25–q75 1.704–1.996)3,388none — median 1.853 (q25–q75 1.71–2.004)3,412none — median 1.853 (q25–q75 1.708–2.003)3,438none — median 1.85 (q25–q75 1.705–2.003)3,462none — median 1.84 (q25–q75 1.696–1.995)3,486none — median 1.829 (q25–q75 1.686–1.983)3,510none — median 1.818 (q25–q75 1.677–1.972)3,536none — median 1.819 (q25–q75 1.681–1.972)3,560none — median 1.824 (q25–q75 1.687–1.979)3,584none — median 1.837 (q25–q75 1.692–1.989)3,608none — median 1.879 (q25–q75 1.714–2.023)3,634none — median 1.889 (q25–q75 1.708–2.041)3,658none — median 1.784 (q25–q75 1.598–1.946)3,682none — median 1.596 (q25–q75 1.446–1.767)3,706none — median 1.482 (q25–q75 1.339–1.682)3,730none — median 1.265 (q25–q75 1.181–1.351)3,756none — median 1.197 (q25–q75 1.133–1.277)3,780none — median 1.175 (q25–q75 1.115–1.245)3,804none — median 1.163 (q25–q75 1.105–1.24)3,828none — median 1.156 (q25–q75 1.098–1.234)3,854none — median 1.15 (q25–q75 1.093–1.23)3,878none — median 1.147 (q25–q75 1.094–1.225)3,902none — median 1.146 (q25–q75 1.093–1.223)3,926none — median 1.145 (q25–q75 1.093–1.221)3,952none — median 1.144 (q25–q75 1.091–1.219)3,976none — median 1.142 (q25–q75 1.089–1.219)4,000none — median 1.14 (q25–q75 1.086–1.218)

Sampling

Wavelengths1,701
Axis range600–4,000 none
Mean spacing2 none
Griduniform
Observations589

Signal & quality

Value range0.892 – 3.48
Mean range1.16 – 2.35
Mean level1.704
Area5792
PTP1.188
Noise RMS0.00021192
SNR8e+03
SNR dB8e+01 dB
Dynamic range1.19
Smoothness0.007521
Saturated0.0%
X-outliers282

Integrity & artefacts

NaN ratio0.00%
Inf count0
Zero ratio0.00%
Spike count54,080
Spike rate5.40%
Jump count28,163
Jump rate2.81%
Clip fraction0.00%

Shape & reference

Baseline slope-0.73146
Curvature RMS0.0061241
D1 RMS0.0089085
RMS to mean0.14622
RMS p950.33927
SAM to mean0.062539
SAM p950.11101
Affine offset p950.37946
Affine gain p95 Δ0.22554
Affine residual p950.20094
Xcorr lag p952

Outliers & repeatability

PCA Q p95/median3.5
Hotelling T2 p95/median3
Mahalanobis H p95/median1.7
Repeat groups0

Dimensionality (PCA)

Effective rank3.9
PCs → 95% var6
PCs → 99% var14
Top-10 cum. var98.3%
Computed metric scores 29worst 1.00
FamilleMétrique calculéeValeurScoreNiveauInterprétation datasetCauses typiquesCalcul / scoring
Intégrité des donnéesNaN ratiointegrity.nan_ratio0%0.00faibleSpectre completErreur acquisition/exportcount(isnan(X)) / X.sizealert = min(1, nan_ratio / 0.05)
Intégrité des donnéesInf countintegrity.inf_count00.00faibleNormalCalculs invalidescount(isinf(X))alert = min(1, inf_count / 1)
Intégrité des donnéesZero ratiointegrity.zero_ratio0%0.00faibleNormalExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceamplitude.mean_reflectance1.70360.86fortValeur atypique: Trop clair / fond visible ou Trop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveamplitude.area_under_curve5792.20.86fortValeur atypique: Différence d'éclairement ou NormalDistance sondetrapezoid(mean_spectrum, spectral_axis)alert reuses baseline/shape drift because area scale depends on axis and units
Amplitude globalePeak-to-peak (PTP)amplitude.peak_to_peak1.18790.00faibleVariabilité forteSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceamplitude.variance0.145120.00faibleNormal ou hétérogèneMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSnoise.noise_rms0.000211920.00faibleStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRnoise.snr8038.90.00faibleBon signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRnoise.bandwise_snr_min68.230.00faibleZone fiableDétecteurmin(abs(mean_spectrum) / local second-derivative noise)alert decreases with worst-band SNR dB; >=35 dB is treated as low alert
Artefacts locauxSpike countartefacts.spike_count54,0801.00fortArtefactsCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateartefacts.spike_rate5.4%1.00fortSpectre suspectInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countartefacts.jump_count28,1631.00fortRaccord détecteurSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateartefacts.jump_rate2.81%1.00fortProblème spectralCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionartefacts.clip_fraction0.0002%0.00faibleNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopeshape.baseline_slope-0.731460.86fortDériveÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSshape.curvature_rms0.00612410.36faibleLisseFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSshape.d1_rms0.00890850.10faiblePlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)outliers.pca_q_ratio3.49530.44moyenSpectre atypiqueArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²outliers.hotelling_t2_ratio3.04120.38faibleCentralVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis Houtliers.mahalanobis_h_ratio1.74390.44moyenOutlier globalDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumreference.rms_to_mean_spectrum_p950.339270.80fortSpectre différentDomain shiftp95 RMS distance to dataset mean spectrumalert = RMS_p95 / signal_scale, saturated at 25%
Comparaison à référenceSpectral Angle Mapper (SAM)reference.sam_to_mean_spectrum_p950.111010.32faibleSimilaireFond, géométriep95 spectral angle to dataset mean spectrumalert = min(1, SAM_p95 / 0.35 rad)
RépétabilitéRMS intra-IDrepeatability.rms_intra_id0.00faibleStablePositionnementmedian RMS distance to repeated-sample centroidalert = RMS_intra_ID / signal_scale, saturated at 10%
RépétabilitéSAM intra-IDrepeatability.sam_intra_id0.00faibleStableAcquisitionmedian SAM to repeated-sample centroidalert = min(1, SAM_intra_ID / 0.15 rad)
RépétabilitéCV intra-IDrepeatability.cv_intra_id0.00faibleStableOpérateurmedian within-ID band CValert = min(1, CV_intra_ID / 0.25)
Structure du datasetPCA score densitystructure.pca_score_density0.428590.58moyenSous-populationsLots différents1 / median kNN distance in PCA score spacealert follows density_cv/profile structure complexity, not raw density alone
Structure du datasetLocal Outlier Factor (LOF)structure.local_outlier_factor_p951.75150.38faiblePopulation normaleCas raresp95 approximate LOF from PCA-score kNN distancesalert = min(1, max(0, LOF_p95 - 1) / 2)
Structure du datasetIsolation Forest scorestructure.isolation_forest_score_p950.531860.58moyenSpectre atypiqueDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
X PCA score plot-20-1001020-1001020PC1 -1.834 · PC2 -1.763PC1 -2.621 · PC2 0.7936PC1 -1.074 · PC2 0.9302PC1 -3.498 · PC2 0.3643PC1 12.98 · PC2 1.378PC1 7.371 · PC2 0.09939PC1 8.846 · PC2 0.8271PC1 -0.1193 · PC2 -0.5367PC1 10.68 · PC2 1.424PC1 17.46 · PC2 3.46PC1 5.332 · PC2 -0.3475PC1 -6.005 · PC2 -2.714PC1 0.8062 · PC2 -1.276PC1 -2.232 · PC2 -0.5873PC1 0.2644 · PC2 3.529PC1 1.12 · PC2 11.48PC1 -2.223 · PC2 -2.866PC1 3.778 · PC2 4.422PC1 3.599 · PC2 9.524PC1 -9.012 · PC2 1.555PC1 -4.394 · PC2 0.7254PC1 3.652 · PC2 15.09PC1 3.902 · PC2 7.243PC1 -7.588 · PC2 -0.3741PC1 -4.821 · PC2 -0.4583PC1 3.812 · PC2 0.4344PC1 -5.17 · PC2 0.4012PC1 2.147 · PC2 0.8247PC1 1.69 · PC2 -0.8037PC1 -7.96 · PC2 1.416PC1 13.85 · PC2 2.195PC1 -7.557 · PC2 0.4994PC1 -6.15 · PC2 0.1007PC1 -9.967 · PC2 -3.792PC1 3.065 · PC2 -0.1734PC1 2.717 · PC2 -0.05024PC1 -5.32 · PC2 -2.035PC1 9.179 · PC2 -3.677PC1 -1.034 · PC2 1.361PC1 -4.165 · PC2 -1.923PC1 4.805 · PC2 13.72PC1 -1.902 · PC2 -1.036PC1 -1.478 · PC2 -1.402PC1 14.5 · PC2 0.7757PC1 3.387 · PC2 0.1187PC1 0.3478 · PC2 -1.137PC1 7.629 · PC2 2.513PC1 -9.015 · PC2 -1.836PC1 -5.725 · PC2 -2.103PC1 0.2261 · PC2 11.18PC1 1.721 · PC2 0.8848PC1 0.4613 · PC2 -0.05352PC1 -3.339 · PC2 1.641PC1 3.346 · PC2 4.499PC1 11.92 · PC2 -0.5725PC1 -4.602 · PC2 0.2788PC1 0.6589 · PC2 1.053PC1 5.193 · PC2 5.829PC1 -2.014 · PC2 0.5523PC1 -5.701 · PC2 1.919PC1 -3.521 · PC2 -0.9196PC1 -8.485 · PC2 -1.383PC1 5.282 · PC2 -1.818PC1 0.6625 · PC2 1.967PC1 4.113 · PC2 2.308PC1 3.756 · PC2 11.5PC1 3.028 · PC2 10.25PC1 -4.086 · PC2 -0.481PC1 0.5283 · PC2 3.752PC1 -3.683 · PC2 2.704PC1 -7.264 · PC2 -0.2305PC1 -3.157 · PC2 2.89PC1 -4.011 · PC2 -0.1119PC1 -2.595 · PC2 0.2217PC1 3.935 · PC2 7.457PC1 0.1092 · PC2 1.845PC1 -2.646 · PC2 -0.2335PC1 -6.531 · PC2 -2.075PC1 1.014 · PC2 -0.4078PC1 -13.96 · PC2 -1.415PC1 -6.487 · PC2 -2.479PC1 1.289 · PC2 2.787PC1 3.775 · PC2 2.314PC1 -1.238 · PC2 -1.818PC1 0.8892 · PC2 -1.874PC1 6.251 · PC2 -0.1756PC1 2.555 · PC2 0.8736PC1 2.781 · PC2 9.984PC1 -4.273 · PC2 2.832PC1 -4.876 · PC2 -0.6589PC1 1.512 · PC2 10.43PC1 -7.595 · PC2 -1.576PC1 9.384 · PC2 11.16PC1 6.016 · PC2 12.62PC1 -9.103 · PC2 3.996PC1 4.474 · PC2 8.745PC1 2.151 · PC2 -1.346PC1 4.509 · PC2 -0.989PC1 3.969 · PC2 2.469PC1 -7.606 · PC2 -0.6049PC1 1.56 · PC2 -1.463PC1 1.679 · PC2 -0.3821PC1 3.188 · PC2 5.539PC1 -8.963 · PC2 -3.084PC1 -2.758 · PC2 -0.9802PC1 4.907 · PC2 1.448PC1 -2.391 · PC2 -0.5905PC1 -0.4144 · PC2 -0.6072PC1 -3.853 · PC2 -0.6387PC1 -1.048 · PC2 -0.5794PC1 -4.281 · PC2 1.707PC1 -6.306 · PC2 1.522PC1 3.192 · PC2 0.5342PC1 -0.2199 · PC2 14.02PC1 2.492 · PC2 0.5192PC1 3.721 · PC2 -0.09397PC1 -3.349 · PC2 1.099PC1 -3.13 · PC2 5.416PC1 3.501 · PC2 4.918PC1 -4.205 · PC2 -3.832PC1 -3.579 · PC2 -3.178PC1 -4.938 · PC2 -2.662PC1 -3.946 · PC2 -5.641PC1 5.812 · PC2 -6.803PC1 -0.6156 · PC2 1.443PC1 3.145 · PC2 -4.553PC1 1.519 · PC2 2.896PC1 -1.062 · PC2 -1.547PC1 2.002 · PC2 1.04PC1 5.816 · PC2 3.011PC1 3.689 · PC2 -1.481PC1 -2.714 · PC2 -4.197PC1 -2.21 · PC2 -4.254PC1 1.714 · PC2 2.822PC1 2.377 · PC2 -2.225PC1 9.944 · PC2 -0.5035PC1 -2.469 · PC2 -2.837PC1 3.753 · PC2 0.2696PC1 4.837 · PC2 -0.1156PC1 -4.609 · PC2 3.026PC1 2.421 · PC2 -2.352PC1 2.676 · PC2 -3.334PC1 -0.8729 · PC2 -2.682PC1 -1.485 · PC2 -1.137PC1 7.49 · PC2 -0.2349PC1 4.788 · PC2 3.766PC1 -4.275 · PC2 3.298PC1 9.908 · PC2 -1.197PC1 0.05272 · PC2 -1.929PC1 3.799 · PC2 -4.999PC1 2.27 · PC2 0.2015PC1 -6.602 · PC2 5.427PC1 -4.522 · PC2 -0.959PC1 1.822 · PC2 -3.442PC1 8.479 · PC2 -2.984PC1 -2.932 · PC2 -2.364PC1 12.84 · PC2 -3.897PC1 3.762 · PC2 -2.777PC1 2.855 · PC2 -2.745PC1 -3.255 · PC2 0.6818PC1 12.72 · PC2 -1.028PC1 1.205 · PC2 -3.159PC1 -2.675 · PC2 0.2039PC1 0.09857 · PC2 -1.402PC1 -3.22 · PC2 -0.6734PC1 -9.034 · PC2 2.408PC1 -5.406 · PC2 -0.8908PC1 -2.643 · PC2 -0.08517PC1 -6.623 · PC2 3.673PC1 -11.37 · PC2 0.1961PC1 5.35 · PC2 1.931PC1 -1.225 · PC2 -2.819PC1 -2.86 · PC2 -1.641PC1 -1.876 · PC2 -2.561PC1 6.097 · PC2 -1.925PC1 1.642 · PC2 -2.595PC1 9.246 · PC2 1.229PC1 -0.4123 · PC2 1.462PC1 -3.089 · PC2 -4.699PC1 -6.589 · PC2 -0.4983PC1 -10.1 · PC2 -0.3545PC1 -2.022 · PC2 -3.477PC1 -10.16 · PC2 2.272PC1 8.459 · PC2 -5.618PC1 -3.038 · PC2 -2.1PC1 1.515 · PC2 -2.829PC1 -0.7658 · PC2 -3.314PC1 -7.465 · PC2 3.187PC1 2.74 · PC2 -4.609PC1 9.137 · PC2 -5.36PC1 10.69 · PC2 0.6917PC1 -5.469 · PC2 -3.231PC1 -4.343 · PC2 -3.115PC1 0.3248 · PC2 2.199PC1 -1.349 · PC2 -2.173PC1 1.03 · PC2 -0.1879PC1 -3.224 · PC2 -0.7817PC1 16.87 · PC2 2.487PC1 -2.577 · PC2 -1.444PC1 -4.232 · PC2 -2.798PC1 1.323 · PC2 3.909PC1 -4.723 · PC2 -2.406PC1 3.119 · PC2 -2.288PC1 3.527 · PC2 -2.77PC1 -2.513 · PC2 2.962PC1 4.387 · PC2 -0.6168PC1 0.3711 · PC2 2.651PC1 0.8073 · PC2 -2.894PC1 -1.476 · PC2 0.03862PC1 -7.136 · PC2 4.305PC1 -3.057 · PC2 4.29PC1 -1.328 · PC2 -1.141PC1 0.08337 · PC2 3.332PC1 -6.702 · PC2 3.963PC1 10.22 · PC2 1.988PC1 12.26 · PC2 0.5364PC1 14.21 · PC2 -0.6316PC1 2.301 · PC2 -1.291PC1 -4.733 · PC2 5.371PC1 0.4816 · PC2 9.879PC1 1.629 · PC2 -2.154PC1 14.05 · PC2 3.451PC1 -1.785 · PC2 -3.465PC1 -0.3247 · PC2 -3.074PC1 5.689 · PC2 -2.16PC1 -0.01867 · PC2 -2.945PC1 1.828 · PC2 -3.413PC1 5.327 · PC2 -1.101PC1 0.3649 · PC2 3.108PC1 -1.36 · PC2 -1.917PC1 -4.987 · PC2 0.1032PC1 7.956 · PC2 -2.297PC1 0.2207 · PC2 -5.664PC1 -6.946 · PC2 0.409PC1 -3.601 · PC2 -3.37PC1 3.327 · PC2 -0.2675PC1 -9.121 · PC2 1.06PC1 -3.036 · PC2 -1.179PC1 -2 · PC2 -4.195PC1 2.697 · PC2 0.008589PC1 0.5826 · PC2 -2.712PC1 -4.234 · PC2 -3.705PC1 -4.569 · PC2 -3.213PC1 -2.944 · PC2 -1.302PC1 -8.508 · PC2 0.4229PC1 15.21 · PC2 2.433PC1 1.521 · PC2 -1.276PC1 -5.289 · PC2 2.636PC1 -11.18 · PC2 -3.363PC1 -3.505 · PC2 1.21PC1 -6.379 · PC2 -2.52PC1 2.428 · PC2 -5.718PC1 -0.2157 · PC2 -0.9267PC1 -1.4 · PC2 0.3275PC1 3.245 · PC2 -8.021PC1 1.281 · PC2 -4.941PC1 2.998 · PC2 -2.188PC1 -3.866 · PC2 2.187PC1 -4.836 · PC2 -2.129PC1 -0.4157 · PC2 -1.891PC1 -3.985 · PC2 -0.7179PC1 -3.668 · PC2 -1.224PC1 -8.428 · PC2 4.033PC1 -6.742 · PC2 3.878PC1 3.598 · PC2 -2.107PC1 -0.3259 · PC2 -0.3888PC1 13.42 · PC2 -0.5685PC1 13.3 · PC2 -4.67PC1 -1.364 · PC2 -1.393PC1 -2.915 · PC2 -4.64PC1 -10.56 · PC2 -1.559PC1 -1.413 · PC2 -3.036PC1 -2.854 · PC2 -3.452PC1 -3.158 · PC2 -0.9426PC1 -1.292 · PC2 -3.922PC1 -0.441 · PC2 -0.5585PC1 9.314 · PC2 0.5003PC1 1.187 · PC2 4.793PC1 -1.968 · PC2 -3.085PC1 0.1798 · PC2 1.85PC1 -5.866 · PC2 2.268PC1 7.121 · PC2 -1.265PC1 0.4794 · PC2 -4.342PC1 2.303 · PC2 -0.935PC1 14.4 · PC2 0.6032PC1 -6.452 · PC2 2.235PC1 -5.643 · PC2 -2.29PC1 -5.407 · PC2 -0.3929PC1 -2.058 · PC2 -3.777PC1 3.463 · PC2 3.25PC1 -5.089 · PC2 4.282PC1 -0.8548 · PC2 -3.029PC1 1.772 · PC2 0.2561PC1 1.314 · PC2 -5.269PC1 -3.566 · PC2 -2.623PC1 0.8528 · PC2 -1.491PC1 -0.3839 · PC2 -1.293PC1 -3.305 · PC2 -3.064PC1 -5.448 · PC2 -1.214PC1 10.5 · PC2 2.01PC1 12.77 · PC2 1.655PC1 7.789 · PC2 -2.355PC1 -2.068 · PC2 -4.003PC1 2.369 · PC2 -1.938PC1 -6.894 · PC2 1.344PC1 -10.25 · PC2 0.874PC1 9.163 · PC2 2.815PC1 -3.818 · PC2 -1.483PC1 0.0072 · PC2 -4.351PC1 3.05 · PC2 -1.425PC1 3.78 · PC2 -1.655PC1 -0.344 · PC2 0.9112PC1 4.182 · PC2 -0.7187PC1 2.346 · PC2 2.161PC1 1.646 · PC2 -3.282PC1 -3.914 · PC2 -3.774PC1 -0.2523 · PC2 -5.613PC1 -3.649 · PC2 -2.957PC1 -6.55 · PC2 0.4219PC1 -3.488 · PC2 1.911PC1 0.3076 · PC2 -1.677PC1 4.144 · PC2 3.449PC1 -0.137 · PC2 -2.055PC1 -0.8861 · PC2 -1.298PC1 -2.371 · PC2 -0.8699PC1 13.55 · PC2 0.4799PC1 3.603 · PC2 0.319PC1 -8.903 · PC2 -1.637PC1 -5.742 · PC2 4.727PC1 -2.965 · PC2 -3.702PC1 19.52 · PC2 2.624PC1 -5.53 · PC2 1.307PC1 -6.319 · PC2 2.87PC1 -0.0963 · PC2 -0.4385PC1 -3.034 · PC2 3.201PC1 7.742 · PC2 -3.369PC1 12.88 · PC2 2.182PC1 7.807 · PC2 -2.002PC1 7.125 · PC2 -1.384PC1 2.146 · PC2 -2.936PC1 -3.158 · PC2 -3.308PC1 12.86 · PC2 0.5264PC1 -4.461 · PC2 0.415PC1 -3.871 · PC2 -6.79PC1 3.267 · PC2 2.459PC1 6.286 · PC2 -0.9265PC1 -1.347 · PC2 -3.748PC1 2.925 · PC2 -1.936PC1 -3.393 · PC2 -1.545PC1 -4.511 · PC2 -3.302PC1 -5.236 · PC2 -1.718PC1 9.669 · PC2 -1.544PC1 0.3279 · PC2 -2.61PC1 -9.349 · PC2 -2.613PC1 -5.048 · PC2 2.106PC1 -1.033 · PC2 2.344PC1 15.83 · PC2 1.827PC1 -5.317 · PC2 0.1521PC1 -7.088 · PC2 0.295PC1 -10.23 · PC2 1.531PC1 -0.2311 · PC2 -0.9512PC1 8.432 · PC2 1.029PC1 -12.67 · PC2 -1.126PC1 -3.196 · PC2 -0.3454PC1 -10.05 · PC2 2.953PC1 -3.078 · PC2 0.04476PC1 3.399 · PC2 5.092PC1 -10.35 · PC2 5.443PC1 -4.095 · PC2 3.243PC1 -3.601 · PC2 -2.262PC1 -2.79 · PC2 -1.618PC1 8.44 · PC2 -2.519PC1 5.978 · PC2 -2.011PC1 -3.208 · PC2 0.9006PC1 -10.18 · PC2 2.081PC1 -1.265 · PC2 -4.471PC1 10.85 · PC2 0.3987PC1 -1.726 · PC2 -1.655PC1 1.172 · PC2 4.399PC1 -0.7812 · PC2 0.4317PC1 8.233 · PC2 -1.351PC1 4.264 · PC2 0.5416PC1 -6.621 · PC2 -1.276PC1 0.4419 · PC2 -0.1043PC1 3.57 · PC2 -3.4PC1 -2.891 · PC2 -3.618PC1 0.3894 · PC2 -1.464PC1 -1.833 · PC2 4.007PC1 1.507 · PC2 -3.024PC1 -9.819 · PC2 -0.04612PC1 5.423 · PC2 -2.863PC1 2.427 · PC2 -2.195PC1 -4.752 · PC2 -2.889PC1 6.357 · PC2 -1.432PC1 2.015 · PC2 -1.462PC1 -2.531 · PC2 0.7502PC1 1.471 · PC2 -3.684PC1 4.845 · PC2 -0.2952PC1 6.209 · PC2 -1.11PC1 2.694 · PC2 -1.591PC1 12.27 · PC2 1.703PC1 -7.714 · PC2 4.069PC1 -5.29 · PC2 -0.9349PC1 -4.167 · PC2 2.213PC1 2.291 · PC2 -3.609PC1 -9.219 · PC2 1.272PC1 -1.888 · PC2 -4.227PC1 3.481 · PC2 -2.401PC1 -15.97 · PC2 1.435PC1 -7.639 · PC2 3.873PC1 -0.4199 · PC2 -2.748PC1 15.24 · PC2 3.372PC1 1.228 · PC2 -3.314PC1 -0.02531 · PC2 -0.3058PC1 1.457 · PC2 -3.4PC1 3.921 · PC2 -5.013PC1 -0.4434 · PC2 0.348PC1 -10.31 · PC2 5.678PC1 -4.692 · PC2 -0.7201PC1 -9.762 · PC2 -2.491PC1 -6.48 · PC2 5.35PC1 0.6909 · PC2 -0.5686PC1 -11.94 · PC2 5.023PC1 -5.643 · PC2 2.95PC1 -7.86 · PC2 2.683PC1 -5.426 · PC2 4.764PC1 6.814 · PC2 -0.9146PC1 1.867 · PC2 0.5393PC1 -5.669 · PC2 3.855PC1 -3.684 · PC2 -0.7241PC1 12.82 · PC2 1.865PC1 -7.918 · PC2 2.858PC1 13.37 · PC2 1.379PC1 6.769 · PC2 0.1948PC1 8.762 · PC2 1.41PC1 -4.249 · PC2 3.024PC1 -0.2674 · PC2 0.0756PC1 -1.106 · PC2 -1.957PC1 16.82 · PC2 2.377PC1 -10.3 · PC2 1.998PC1 -1.61 · PC2 -2.32PC1 -0.1727 · PC2 0.6016PC1 -1.444 · PC2 -0.6078PC1 11.18 · PC2 1.315PC1 -4.418 · PC2 0.1223PC1 -4.482 · PC2 2.675PC1 9.072 · PC2 3.029PC1 -3.806 · PC2 -0.9624PC1 -11.52 · PC2 -1.72PC1 0.7677 · PC2 1.823PC1 -0.1876 · PC2 -1.333PC1 -0.6205 · PC2 -4.506PC1 -8.107 · PC2 6.522PC1 -4.507 · PC2 -1.32PC1 10.24 · PC2 -2.205PC1 -5.553 · PC2 0.000219PC1 3.824 · PC2 -2.265PC1 10.55 · PC2 1.321PC1 4.289 · PC2 0.3631PC1 -0.8772 · PC2 0.1443PC1 8.345 · PC2 -5.889PC1 1.907 · PC2 0.6349PC1 -7.036 · PC2 2.51PC1 -2.894 · PC2 0.8411PC1 -2.582 · PC2 1.85PC1 -4.748 · PC2 6.143PC1 10.33 · PC2 4.021PC1 -3.5 · PC2 3.328PC1 -4.74 · PC2 2.218PC1 -6.799 · PC2 -0.1011PC1 -5.781 · PC2 -3.701PC1 -0.6033 · PC2 -1.242PC1 12.3 · PC2 1.37PC1 -15.07 · PC2 3.353PC1 2.68 · PC2 2.134PC1 6.294 · PC2 -0.2057PC1 5.228 · PC2 1.055PC1 -7.164 · PC2 4.381PC1 -12.37 · PC2 3.998PC1 9.909 · PC2 1.006PC1 -6.377 · PC2 5.315PC1 -4.857 · PC2 4.451PC1 3.722 · PC2 3.104PC1 -0.6024 · PC2 -0.8135PC1 16.07 · PC2 2.495PC1 -4.058 · PC2 3.758PC1 4.622 · PC2 -2.575PC1 -3.029 · PC2 -1.061PC1 -3.037 · PC2 0.5929PC1 -1.849 · PC2 2.144PC1 -1.961 · PC2 -0.4637PC1 9.662 · PC2 -2.848PC1 5.687 · PC2 -1.132PC1 0.4186 · PC2 -1.747PC1 0.5215 · PC2 -0.9308PC1 3.421 · PC2 -3.07PC1 4.33 · PC2 -1.745PC1 -3.228 · PC2 -1.28PC1 -4.395 · PC2 3.455PC1 11.92 · PC2 3.483PC1 1.305 · PC2 -0.9727PC1 -2.294 · PC2 1.483PC1 2.924 · PC2 -0.1197PC1 4.395 · PC2 -2.869PC1 3.351 · PC2 -2.893PC1 0.1376 · PC2 -2.3PC1 1.558 · PC2 2.315PC1 0.4439 · PC2 -3.657PC1 1.693 · PC2 -3.479PC1 -9.344 · PC2 0.7527PC1 1.673 · PC2 -0.6687PC1 4.039 · PC2 -1.397PC1 -2.787 · PC2 -2.691PC1 -7.868 · PC2 3.326PC1 -10.09 · PC2 4.651PC1 10.15 · PC2 3.436PC1 13.2 · PC2 1.593PC1 -11.23 · PC2 4.162PC1 7.133 · PC2 -0.2066PC1 -3.852 · PC2 -0.6065PC1 2.674 · PC2 -1.303PC1 -7.784 · PC2 0.2852PC1 1.225 · PC2 -2.147PC1 -7.599 · PC2 -2.147PC1 0.2398 · PC2 -2.686PC1 5.742 · PC2 -3.207PC1 -3.349 · PC2 -1.517PC1 -0.03299 · PC2 -5.105PC1 3.882 · PC2 4.276PC1 -1.516 · PC2 -3.013PC1 -1.387 · PC2 -1.842PC1 -2.727 · PC2 0.3542PC1 -0.059 · PC2 -2.415PC1 -4.741 · PC2 -2.532PC1 11.58 · PC2 -0.2531PC1 1.077 · PC2 -6.629PC1 8.091 · PC2 0.7901PC1 -2.616 · PC2 -0.7119PC1 -4.477 · PC2 -2.29PC1 -5.6 · PC2 -2.061PC1 -2.065 · PC2 -4.31PC1 -3.076 · PC2 -3.434PC1 -8.355 · PC2 -1.813PC1 -8.943 · PC2 3.553PC1 5.273 · PC2 -0.8938PC1 -8.783 · PC2 -3.554PC1 3.088 · PC2 -4.404PC1 -2.645 · PC2 1.602PC1 10.12 · PC2 0.7033PC1 4.865 · PC2 -3.149PC1 1.511 · PC2 0.6948PC1 -4.495 · PC2 -1.725PC1 -2.915 · PC2 0.2068PC1 -1.74 · PC2 3.3PC1 -4.941 · PC2 4.597PC1 -3.344 · PC2 -3.514PC1 -0.2739 · PC2 -2.516PC1 14.37 · PC2 1.346PC1 -6.249 · PC2 -0.4062PC1 -4.663 · PC2 -1.132PC1 -0.6208 · PC2 4.363PC1 9.486 · PC2 1.423PC1 13.7 · PC2 1.81PC1 -7.808 · PC2 4.965PC1 -0.2528 · PC2 0.8526PC1 -0.9984 · PC2 3.05PC1 6.86 · PC2 -1.715PC1 -0.7971 · PC2 -2.828PC1 1.75 · PC2 -3.011PC1 -8.395 · PC2 2.948PC1 -1.488 · PC2 1.368PC1 6.57 · PC2 -1.776PC1 2.445 · PC2 -1.853PC1 -0.7709 · PC2 -2.132PC1 -5.615 · PC2 4.359PC1 3.774 · PC2 -4.65PC1 -1.16 · PC2 -3.299PC1 -0.9198 · PC2 -0.2177PC1 9.547 · PC2 -1.203PC1 4.908 · PC2 -2.81PC1 -4.911 · PC2 0.579PC1 6.095 · PC2 -4.151PC1 -6.935 · PC2 3.533PC1 -6.705 · PC2 -0.3621PC1 -2.073 · PC2 1.597PC1 -13.68 · PC2 -3.099PC1 6.01 · PC2 -1.716PC1 1.109 · PC2 0.1394PC1 -6.726 · PC2 -0.08113PC1 (61.4%)PC2 (16.7%)589 scores
PCA explained variance0%25%50%75%100%PC1: 61.4% (cumulative 61.4%)1PC2: 16.7% (cumulative 78.2%)2PC3: 7.2% (cumulative 85.4%)3PC4: 5.1% (cumulative 90.5%)4PC5: 3.4% (cumulative 93.9%)5PC6: 2.2% (cumulative 96.1%)6PC7: 0.9% (cumulative 97.0%)7PC8: 0.5% (cumulative 97.5%)8PC9: 0.4% (cumulative 97.9%)9PC10: 0.4% (cumulative 98.3%)10cumulative explained variancePC variancecumulativeprincipal component · cumulative (dashed)
X-Y spectral correlation 12
X · caco3_usda_a54_w_pct spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,0004,000|r|signed raxis · Pearson correlation scale
X · cec_usda_a723_cmolc_kg spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,0004,000|r|signed raxis · Pearson correlation scale
X · cf_usda_c236_w_pct spectral correlation-1-0.500.51absolute correlation envelopesigned correlationabsolute correlation01,0002,0003,0004,000|r|signed raxis · Pearson correlation scale
Targetmax |r|axis @ maxmean |r||r| ≥ .5
caco3_usda_a54_w_pct0.7581,3100.1423.2%
cec_usda_a723_cmolc_kg0.5863,3480.37420.3%
cf_usda_c236_w_pct0.251,3920.1030.0%
clay_tot_usda_a334_w_pct0.5623,6520.1953.0%
k_ext_usda_a725_cmolc_kg0.3341,2100.1480.0%
n_tot_usda_a623_w_pct0.7562,9220.43145.0%
oc_usda_c729_w_pct0.7862,9200.42545.0%
p_ext_usda_a274_mg_kg0.1981,5560.0950.0%
ph_cacl2_usda_a481_index0.5991,3180.1732.5%
ph_h2o_usda_a268_index0.6161,3180.1822.7%
sand_tot_usda_c60_w_pct0.6571,1780.349.6%
silt_tot_usda_c62_w_pct0.5221,3840.232.6%

Metric interpretation reference

Metric catalog 29
FamilleMétriqueCe qu’elle détecteForte valeur =Faible valeur =Causes typiquesCalcul / score
Intégrité des donnéesNaN ratioDonnées manquantesSpectre corrompuSpectre completErreur acquisition/exportcount(isnan(X)) / X.sizealert = min(1, nan_ratio / 0.05)
Intégrité des donnéesInf countValeurs infiniesCorruptionNormalCalculs invalidescount(isinf(X))alert = min(1, inf_count / 1)
Intégrité des donnéesZero ratioColonnes ou cellules nullesSpectre tronquéNormalExport, saturationcount(X == 0) / count(finite X)alert = min(1, zero_ratio / 0.05)
Amplitude globaleMean reflectanceNiveau moyenTrop clair / fond visibleTrop sombreFond, géométriemean(X finite)alert reuses baseline/shape drift because absolute reflectance ranges are technology-dependent
Amplitude globaleArea under curveIntensité globaleDifférence d'éclairementNormalDistance sondetrapezoid(mean_spectrum, spectral_axis)alert reuses baseline/shape drift because area scale depends on axis and units
Amplitude globalePeak-to-peak (PTP)DynamiqueVariabilité forteSpectre platSaturationmax(mean_spectrum) - min(mean_spectrum)alert increases when dynamic range is abnormally flat
Amplitude globaleVarianceVariabilité spectraleNormal ou hétérogèneSpectre platMauvais contactvar(X finite)alert increases when variance/dynamic range is abnormally flat
BruitNoise RMSBruit haute fréquenceBruitéStableLampe, détecteurmedian MAD(second derivative) * 1.4826 / sqrt(6)alert = noise_rms / signal_scale, saturated at 5%
BruitSNRQualité signalBon signalMauvais signalAcquisitionmean(abs(X)) / noise_rmsalert decreases with SNR dB; >=40 dB is treated as low alert
BruitBandwise SNRBruit localiséZone fiableZone problématiqueDétecteurmin(abs(mean_spectrum) / local second-derivative noise)alert decreases with worst-band SNR dB; >=35 dB is treated as low alert
Artefacts locauxSpike countPics étroitsArtefactsSpectre propreCosmic rays, splicecount robust outliers in second derivativealert follows spike_rate, saturated at 1%
Artefacts locauxSpike rateDensité de picsSpectre suspectNormalInterpolationspike_count / (n_samples * (n_features - 2))alert = min(1, spike_rate / 0.01)
Artefacts locauxJump countDiscontinuitésRaccord détecteurContinuSplicecount robust outliers in first derivativealert follows jump_rate, saturated at 1%
Artefacts locauxJump rateFréquence de sautsProblème spectralNormalCalibrationjump_count / (n_samples * (n_features - 1))alert = min(1, jump_rate / 0.01)
Artefacts locauxClip fractionSaturationClippingNormalDétecteur saturéfraction of finite cells equal to repeated min/max extremaalert = min(1, clip_fraction / 0.01)
Forme spectraleBaseline slopePente globaleDériveStableÉclairementlinear slope of mean_spectrum over normalized axisalert = abs(slope / signal_scale), saturated at 0.5
Forme spectraleCurvature RMSCourbureForme inhabituelleLisseFond, splicemedian RMS(second derivative per spectrum)alert = curvature_rms / signal_scale, saturated at 1%
Forme spectraleD1 RMSVariabilité localeSpectre structuréPlatBiologie ou artefactmedian RMS(first derivative per spectrum)alert = d1_rms / signal_scale, saturated at 5%
Outliers multivariésPCA Q (SPE)Non expliqué par PCASpectre atypiqueConformeArtefact, mélangep95(Q/SPE residual) / median(Q/SPE residual)alert = min(1, pca_q_ratio / 8)
Outliers multivariésHotelling T²Extrême dans PCAExtrême mais cohérentCentralVariabilité naturellep95(Hotelling T2) / median(Hotelling T2)alert = min(1, hotelling_t2_ratio / 8)
Outliers multivariésMahalanobis HDistance au nuageOutlier globalPopulation normaleDomaine différentp95(sqrt(T2)) / median(sqrt(T2))alert = min(1, mahalanobis_h_ratio / 4)
Comparaison à référenceRMS to mean spectrumDistance moyenneSpectre différentTypiqueDomain shiftp95 RMS distance to dataset mean spectrumalert = RMS_p95 / signal_scale, saturated at 25%
Comparaison à référenceSpectral Angle Mapper (SAM)Différence de formeForme différenteSimilaireFond, géométriep95 spectral angle to dataset mean spectrumalert = min(1, SAM_p95 / 0.35 rad)
RépétabilitéRMS intra-IDReproductibilitéMauvaise répétabilitéStablePositionnementmedian RMS distance to repeated-sample centroidalert = RMS_intra_ID / signal_scale, saturated at 10%
RépétabilitéSAM intra-IDVariation de formeInstableStableAcquisitionmedian SAM to repeated-sample centroidalert = min(1, SAM_intra_ID / 0.15 rad)
RépétabilitéCV intra-IDVariabilité interneMauvais contrôleStableOpérateurmedian within-ID band CValert = min(1, CV_intra_ID / 0.25)
Structure du datasetPCA score densityClustersSous-populationsHomogèneLots différents1 / median kNN distance in PCA score spacealert follows density_cv/profile structure complexity, not raw density alone
Structure du datasetLocal Outlier Factor (LOF)Anomalie localeSpectre isoléPopulation normaleCas raresp95 approximate LOF from PCA-score kNN distancesalert = min(1, max(0, LOF_p95 - 1) / 2)
Structure du datasetIsolation Forest scoreAnomalie globaleSpectre atypiqueNormalDiverses causesp95 IsolationForest anomaly score on PCA scoresalert follows structure complexity; raw score is implementation-dependent
Technology-specific extensions
TechnologieAdaptations / métriquesAnomalies cibléesCommentaire pratique
UV-Vis 300-1000 nmBaseline, pente globale, dérive aux bords 300-350 et 900-1000; métriques par zonesLumière parasite, mauvais blanc, saturation, faible signal aux extrémitésLes bords sont souvent instables; calculer aussi des scores edge/middle.
UV-Vis 300-1000 nmSaturation / clipping proche absorbance max ou réflectance maxSignal écrêtéTrès important si absorption forte.
UV-Vis 300-1000 nmRed-edge, position de maximum, ratios de bandes si végétalDécalage biologique ou artefact optiqueAide à distinguer changement réel et problème d'acquisition.
UV-Vis 300-1000 nmSmoothness / roughness indexBruit haute fréquenceSouvent plus informatif que le SNR seul.
MIR / ATR-FTIRATR contact quality index: intensité globale, aire totale, profondeur des bandes clésMauvais contact cristal-échantillonCrucial: beaucoup d'anomalies viennent du contact ATR.
MIR / ATR-FTIRCO2 / H2O atmospheric bandsMauvaise correction atmosphériquePics parasites fréquents.
MIR / ATR-FTIRBaseline curvature / rubber-band residualDiffusion, contact, dérive baselineTrès utile avant PCA.
MIR / ATR-FTIRPeak position shiftMauvais alignement spectral / calibrationImportant en FTIR car de petits shifts comptent.
MIR / ATR-FTIRBand area ratios sur bandes connuesSpectre chimiquement incohérentÀ adapter par matrice: polysaccharides, protéines, lipides, etc.
HS-MSTotal Ion Current (TIC), Base Peak Intensity (BPI)Injection faible, ionisation instableÉquivalent MS du niveau global spectral.
HS-MSNombre de pics détectésSpectre pauvre ou trop bruitéTrop peu = mauvais signal; trop = bruit/contamination.
HS-MSMass accuracy / m/z driftProblème calibration masseFondamental en HRMS.
HS-MSRetention time drift si LC/GC-MSDérive chromatographiqueÀ suivre sur standards/QC pools.
HS-MSBlank contamination scoreContaminants / carry-overComparer échantillons vs blancs.
HS-MSInternal standard CVVariabilité instrumentaleTrès robuste si standards disponibles.
HS-MSMissingness par featureInstabilité de détectionCrucial pour filtrer les variables.
Avec répétitionsRMS intra-échantillonRépétabilité globaleApplicable à toutes les technologies.
Avec répétitionsSAM / corrélation intra-échantillonRépétabilité de formeTrès utile pour spectres.
Avec répétitionsCV intra-échantillon par bande / featureRépétabilité localeDétecte les zones instables.
Avec répétitionsICC ou variance componentsPart variance échantillon vs techniqueTrès utile si plusieurs répétitions par sample.
Avec répétitionsDistance au centroïde intra-IDRépétition aberrantePermet de flagger la mauvaise répétition plutôt que le sample entier.
Bug-hunting / supervised audits
Famille de bug potentielMéthodes à ajouterCe que ça détecteÉtat dans l’explorateur
Shift spectral globalCorrélation spectre moyen inter-dataset, DTW, cross-correlation, comparaison positions de picsDécalage en longueur d'onde, mauvais alignement, interpolation différentePartiellement calculé: cross-correlation lag et dispersion des positions de pics vs spectre moyen.
Baseline / offset / gainRégression chaque spectre vs spectre moyen: x = a + b ref + residual; suivi de a, b, RMS résiduelOffset additif, effet multiplicatif, dérive de baselineCalculé dans reference.affine_*.
Mélange de lignes / mauvais appariement X-M-YVérification index, hash des lignes, duplication ID, distance spectrale intra-ID, labels incohérentsLignes mélangées, metadata mal alignées, Y attribué au mauvais spectrePartiellement couvert par répétabilité intra-ID; checks index/hash à ajouter au pipeline canonical.
Fuite d'information / répétitions mal splitéesGroupKFold par sample_id vs StratifiedKFold random; audit des partitions par sample_idPerformance artificiellement bonne due aux répétitionsNécessite splits et benchmark modèle; non calculé par la carte descriptive.
Label bugsÉchantillons proches en X mais Y différents, confident learning, erreurs systématiques FP/FNY inversés, erreurs de saisie, classes ambiguësNécessite Y et/ou modèle; recommandé pour l'explorateur supervisé.
Sous-domaines cachésPCA/UMAP/t-SNE + clustering non supervisé + association avec dataset/Y/date/operatorLots, campagnes, sondes, backgrounds non renseignésPartiellement calculé par structure PCA/LOF; UMAP/t-SNE hors carte statique.
Artefacts localisés inconnusCarte wavelength x dataset: différence moyenne, différence variance, KS par longueur d'ondeRégions spectrales anormales non anticipéesÀ calculer au niveau banque quand plusieurs datasets partagent un axe spectral.
Ruptures instrumentalesDiscontinuités dans dérivées, changepoint detectionSplice, raccord détecteur, saut local non prévuCalculé par jump/spike rates; changepoint plus avancé à ajouter.
Mélange / contamination spectraleNMF / unmixing / reconstruction par convex hullComposante externe: fond, plastique, solNon calculé automatiquement; nécessite hypothèses de composants ou grande bibliothèque.
Features instables mais prédictivesImportance modèle vs instabilité QC par variableModèle qui apprend un artefact plutôt qu'un signal biologiqueNécessite modèle supervisé; recommandé pour rapports de benchmark.

Variables

Targets 12

caco3_usda_a54_w_pct

target · numeric
caco3_usda_a54_w_pct distribution02004006000 – 2.788: 4502.788 – 5.575: 165.575 – 8.363: 228.363 – 11.15: 1411.15 – 13.94: 1413.94 – 16.73: 916.73 – 19.51: 1019.51 – 22.3: 622.3 – 25.09: 925.09 – 27.88: 827.88 – 30.66: 830.66 – 33.45: 333.45 – 36.24: 336.24 – 39.02: 239.02 – 41.81: 441.81 – 44.6: 244.6 – 47.39: 147.39 – 50.18: 250.18 – 52.96: 252.96 – 55.75: 055.75 – 58.54: 058.54 – 61.33: 261.33 – 64.11: 064.11 – 66.9: 2020406080
n / missing589 / 0
Mean ± SD4.707 ± 10.6
Median0.1
Range0 – 66.9
CV2.24
Skew / kurtosis3 / 9.5
Normal?no

cec_usda_a723_cmolc_kg

target · numeric
cec_usda_a723_cmolc_kg distribution01002001 – 9.004: 1889.004 – 17.01: 14117.01 – 25.01: 10025.01 – 33.02: 7433.02 – 41.02: 4241.02 – 49.02: 1349.02 – 57.03: 357.03 – 65.03: 665.03 – 73.04: 673.04 – 81.04: 281.04 – 89.05: 189.05 – 97.05: 197.05 – 105.1: 2105.1 – 113.1: 1113.1 – 121.1: 2121.1 – 129.1: 1129.1 – 137.1: 0137.1 – 145.1: 1145.1 – 153.1: 1153.1 – 161.1: 2161.1 – 169.1: 0169.1 – 177.1: 1177.1 – 185.1: 0185.1 – 193.1: 1050100150200
n / missing589 / 0
Mean ± SD20.53 ± 22.2
Median14.6
Range1 – 193.1
CV1.08
Skew / kurtosis3.7 / 19
Normal?no

cf_usda_c236_w_pct

target · numeric
cf_usda_c236_w_pct distribution0501001500 – 2.708: 512.708 – 5.417: 1085.417 – 8.125: 788.125 – 10.83: 4510.83 – 13.54: 6113.54 – 16.25: 5016.25 – 18.96: 2618.96 – 21.67: 3021.67 – 24.38: 3424.38 – 27.08: 2427.08 – 29.79: 1629.79 – 32.5: 1732.5 – 35.21: 1235.21 – 37.92: 737.92 – 40.62: 840.62 – 43.33: 543.33 – 46.04: 746.04 – 48.75: 348.75 – 51.46: 151.46 – 54.17: 154.17 – 56.88: 056.88 – 59.58: 359.58 – 62.29: 062.29 – 65: 2020406080
n / missing589 / 0
Mean ± SD14.34 ± 11.7
Median12
Range0 – 65
CV0.819
Skew / kurtosis1.3 / 1.7
Normal?no

clay_tot_usda_a334_w_pct

target · numeric
clay_tot_usda_a334_w_pct distribution0501000 – 3.292: 963.292 – 6.583: 436.583 – 9.875: 559.875 – 13.17: 5113.17 – 16.46: 4116.46 – 19.75: 3919.75 – 23.04: 4523.04 – 26.33: 1826.33 – 29.62: 2129.62 – 32.92: 2732.92 – 36.21: 1936.21 – 39.5: 2039.5 – 42.79: 2642.79 – 46.08: 2046.08 – 49.38: 1549.38 – 52.67: 1052.67 – 55.96: 1155.96 – 59.25: 1459.25 – 62.54: 462.54 – 65.83: 765.83 – 69.12: 369.12 – 72.42: 372.42 – 75.71: 075.71 – 79: 1020406080
n / missing589 / 0
Mean ± SD21.54 ± 17.8
Median17
Range0 – 79
CV0.826
Skew / kurtosis0.78 / -0.29
Normal?no

k_ext_usda_a725_cmolc_kg

target · numeric
k_ext_usda_a725_cmolc_kg distribution02004006000 – 1.565: 4481.565 – 3.13: 1003.13 – 4.695: 244.695 – 6.259: 96.259 – 7.824: 37.824 – 9.389: 09.389 – 10.95: 110.95 – 12.52: 012.52 – 14.08: 014.08 – 15.65: 015.65 – 17.21: 117.21 – 18.78: 018.78 – 20.34: 020.34 – 21.91: 021.91 – 23.47: 023.47 – 25.04: 025.04 – 26.6: 026.6 – 28.17: 128.17 – 29.73: 029.73 – 31.3: 031.3 – 32.86: 032.86 – 34.43: 034.43 – 35.99: 135.99 – 37.56: 1010203040
n / missing589 / 0
Mean ± SD1.314 ± 2.67
Median0.7284
Range0 – 37.56
CV2.03
Skew / kurtosis9.8 / 1.2e+02
Normal?no

n_tot_usda_a623_w_pct

target · numeric
n_tot_usda_a623_w_pct distribution01002003000 – 0.1392: 2290.1392 – 0.2783: 2250.2783 – 0.4175: 570.4175 – 0.5567: 80.5567 – 0.6958: 90.6958 – 0.835: 80.835 – 0.9742: 70.9742 – 1.113: 91.113 – 1.252: 31.252 – 1.392: 51.392 – 1.531: 21.531 – 1.67: 11.67 – 1.809: 11.809 – 1.948: 31.948 – 2.087: 42.087 – 2.227: 22.227 – 2.366: 62.366 – 2.505: 32.505 – 2.644: 12.644 – 2.783: 12.783 – 2.922: 12.922 – 3.062: 23.062 – 3.201: 03.201 – 3.34: 201234
n / missing589 / 0
Mean ± SD0.3236 ± 0.508
Median0.16
Range0 – 3.34
CV1.57
Skew / kurtosis3.5 / 13
Normal?no

oc_usda_c729_w_pct

target · numeric
oc_usda_c729_w_pct distribution02004000.1 – 2.322: 3442.322 – 4.543: 1354.543 – 6.765: 306.765 – 8.987: 108.987 – 11.21: 911.21 – 13.43: 713.43 – 15.65: 515.65 – 17.87: 317.87 – 20.1: 320.1 – 22.32: 322.32 – 24.54: 224.54 – 26.76: 226.76 – 28.98: 128.98 – 31.2: 231.2 – 33.43: 333.43 – 35.65: 535.65 – 37.87: 337.87 – 40.09: 540.09 – 42.31: 342.31 – 44.53: 344.53 – 46.76: 346.76 – 48.98: 548.98 – 51.2: 151.2 – 53.42: 20204060
n / missing589 / 0
Mean ± SD5.231 ± 9.77
Median1.96
Range0.1 – 53.42
CV1.87
Skew / kurtosis3.2 / 9.7
Normal?no

p_ext_usda_a274_mg_kg

target · numeric
p_ext_usda_a274_mg_kg distribution01002003000 – 12: 23812 – 24: 14624 – 36: 6136 – 48: 5148 – 60: 3260 – 72: 1872 – 84: 1584 – 96: 1196 – 108: 4108 – 120: 5120 – 132: 2132 – 144: 1144 – 156: 2156 – 168: 0168 – 180: 1180 – 192: 0192 – 204: 0204 – 216: 0216 – 228: 0228 – 240: 1240 – 252: 0252 – 264: 0264 – 276: 0276 – 288: 10100200300
n / missing589 / 0
Mean ± SD23.8 ± 30.9
Median15.4
Range0 – 288
CV1.3
Skew / kurtosis2.8 / 14
Normal?no

ph_cacl2_usda_a481_index

target · numeric
ph_cacl2_usda_a481_index distribution0501002.85 – 3.058: 83.058 – 3.266: 103.266 – 3.474: 143.474 – 3.682: 213.682 – 3.89: 233.89 – 4.098: 184.098 – 4.305: 304.305 – 4.513: 244.513 – 4.721: 204.721 – 4.929: 294.929 – 5.137: 225.137 – 5.345: 295.345 – 5.553: 195.553 – 5.761: 245.761 – 5.969: 345.969 – 6.177: 166.177 – 6.385: 166.385 – 6.593: 226.593 – 6.8: 176.8 – 7.008: 247.008 – 7.216: 557.216 – 7.424: 837.424 – 7.632: 297.632 – 7.84: 212510
n / missing589 / 0
Mean ± SD5.694 ± 1.37
Median5.77
Range2.85 – 7.84
CV0.241
Skew / kurtosis-0.25 / -1.2
Normal?no

ph_h2o_usda_a268_index

target · numeric
ph_h2o_usda_a268_index distribution02040603.55 – 3.758: 33.758 – 3.967: 113.967 – 4.175: 204.175 – 4.383: 214.383 – 4.592: 234.592 – 4.8: 214.8 – 5.008: 205.008 – 5.217: 345.217 – 5.425: 205.425 – 5.633: 225.633 – 5.842: 335.842 – 6.05: 326.05 – 6.258: 296.258 – 6.467: 236.467 – 6.675: 236.675 – 6.883: 206.883 – 7.092: 237.092 – 7.3: 207.3 – 7.508: 407.508 – 7.717: 427.717 – 7.925: 477.925 – 8.133: 318.133 – 8.342: 258.342 – 8.55: 612510
n / missing589 / 0
Mean ± SD6.306 ± 1.32
Median6.32
Range3.55 – 8.55
CV0.209
Skew / kurtosis-0.19 / -1.2
Normal?no

sand_tot_usda_c60_w_pct

target · numeric
sand_tot_usda_c60_w_pct distribution02550750 – 4.083: 704.083 – 8.167: 358.167 – 12.25: 4512.25 – 16.33: 2716.33 – 20.42: 2920.42 – 24.5: 2624.5 – 28.58: 2128.58 – 32.67: 2432.67 – 36.75: 2636.75 – 40.83: 2740.83 – 44.92: 2044.92 – 49: 2249 – 53.08: 2253.08 – 57.17: 2357.17 – 61.25: 1961.25 – 65.33: 2965.33 – 69.42: 2269.42 – 73.5: 1773.5 – 77.58: 1577.58 – 81.67: 1681.67 – 85.75: 1185.75 – 89.83: 1589.83 – 93.92: 1193.92 – 98: 170255075100
n / missing589 / 0
Mean ± SD38.58 ± 28.5
Median35
Range0 – 98
CV0.738
Skew / kurtosis0.35 / -1.1
Normal?no

silt_tot_usda_c62_w_pct

target · numeric
silt_tot_usda_c62_w_pct distribution02040600 – 3.458: 543.458 – 6.917: 166.917 – 10.38: 2210.38 – 13.83: 1213.83 – 17.29: 3117.29 – 20.75: 2420.75 – 24.21: 2324.21 – 27.67: 4027.67 – 31.12: 5831.12 – 34.58: 4034.58 – 38.04: 4238.04 – 41.5: 3341.5 – 44.96: 3344.96 – 48.42: 3648.42 – 51.88: 3051.88 – 55.33: 2655.33 – 58.79: 1758.79 – 62.25: 2462.25 – 65.71: 1065.71 – 69.17: 869.17 – 72.62: 172.62 – 76.08: 376.08 – 79.54: 279.54 – 83: 40255075100
n / missing589 / 0
Mean ± SD32.58 ± 18.7
Median33
Range0 – 83
CV0.574
Skew / kurtosis0.02 / -0.57
Normal?no

Metadata 4

ID_sample

metadata · categorical
n / missing589 / 0
Classes589
Balance (entropy)1
Imbalance ratio1
Top class917158aeab35339a96567ca8882da8f2 (1)

scan_local_id

metadata · categorical
n / missing589 / 0
Classes589
Balance (entropy)1
Imbalance ratio1
Top class11070 (1)

raw_label

metadata · categorical
raw_label classes0.00.0: 1891890.10.1: 1171170.20.2: 54540.30.3: 18180.50.5: 990.40.4: 660.60.6: 660.90.9: 662.42.4: 550.70.7: 55+10 more+10 more: 3333
n / missing589 / 0
Classes138
Balance (entropy)0.62
Imbalance ratio189
Top class0.0 (189)

reference_value

metadata · numeric
reference_value distribution02004006000 – 2.788: 4502.788 – 5.575: 165.575 – 8.363: 228.363 – 11.15: 1411.15 – 13.94: 1413.94 – 16.73: 916.73 – 19.51: 1019.51 – 22.3: 622.3 – 25.09: 925.09 – 27.88: 827.88 – 30.66: 830.66 – 33.45: 333.45 – 36.24: 336.24 – 39.02: 239.02 – 41.81: 441.81 – 44.6: 244.6 – 47.39: 147.39 – 50.18: 250.18 – 52.96: 252.96 – 55.75: 055.75 – 58.54: 058.54 – 61.33: 261.33 – 64.11: 064.11 – 66.9: 2020406080
n / missing589 / 0
Mean ± SD4.707 ± 10.6
Median0.1
Range0 – 66.9
CV2.24
Skew / kurtosis3 / 9.5
Normal?no
Constant metadata 19
  • SpectralRep1
  • datasetOSSL snapshot v1.2
  • collection_nameossl_mir
  • dataset_codeLUCAS.WOODWELL.SSL
  • dataset_titleLUCAS 2009, 2015 topsoil data
  • dataset_ownerEuropean Soil Data Centre (ESDAC), European Commission, Joint Research Centre
  • dataset_slugLUCAS.WOODWELL.SSL
  • task_typeregression
  • trait_headercaco3_usda_a54_w_pct
  • trait_header_originalcaco3_usda.a54_w.pct
  • spectral_kindmir
  • scan_model_nameBruker Vertex 70 with with PikeAutoDiff accessory
  • scan_model_codeBruker_Vertex_70.PikeAutoDiff
  • scan_opticsKBr beamsplitter, Gold mirror, Mirror background
  • upper_depth_cm0
  • lower_depth_cm20
  • feature_count_per_dimension1,701
  • dimensions1D
  • wavelength_noteossl_mir_600_4000_cm_minus_1_step_2

49 variable(s) omitted (no recorded values).

Alignment

Alignment levelobservation
Sample id availableno
Samples589
Observations (total)589
Reps per samplemin 1 · mean 1 · max 1

Splits

originalall: 589 documented · not applied

Provenance & citation

Governance & integrity

Tierprivate
LicenseLicenseRef-not-cleared
Permitted useResearch and benchmarking; private use only.
Access policyManual download / private-use-only per source.
RedistributionRecovered from local initial-source exports; rights not cleared for redistribution.
Content version1.0.0
Schema / protocol2.0
Content hash4ba9f75fc3d7d51f…
Processing hash3e02a8bc27d359c3…
Metadata hash8951917932a694dd…

Load this dataset

# pip install nirs4all-datasets
from nirs4all_datasets import get

# private dataset — export requires a Dataverse token
ds = get("ossl_lucas_woodwell_mir_soil_all_y", token="…")
X, y = ds.x(), ds.y()
print(X.shape, y.shape)

Metadata downloads are available for public datasets only. The dataset bytes are never served here — fetch them from the origin / DOI above.