Notice bibliographique
Résumé
Meeting food demand for ever increasing global population can be attained through sustainable management of soil resources. This requires a thorough understanding of soil properties and processes and calls for methods to quantify and display spatial variability of soil. Three dimensional digital soil mapping (3D-DSM) with its ability to quantify both the horizontal and the vertical variability has become popular in recent days. The state-of-the-art data mining techniques including 3D regression kriging (RK) has been used to uncover complex soil-landscape relationships but not assessed at small scales. In addition, recent advances in proximal soil sensing allow measurement and prediction of various soil properties simultaneously and rapidly at multiple depths and provide required information for DSM. Furthermore, sampling design (SD) plays a vital role in providing a reliable input for DSM, whereas its effectiveness on 3D-DSM has not been tested. A total of 148 sample locations, identified by six SDs, including grid sampling (GS), grid random sampling (GRS), simple random sampling (SRS), stratified random sampling (StRS), transect sampling (TS), and conditioned Latin hypercube sampling (cLHS), were used to collect vis-NIR spectra data to about 1-m depth in-situ using a commercial soil profiler from a small agricultural farm in Macdonald campus, McGill University. A subset of 32 sample locations were identified to collect soil cores down to 1-m depth and sampled at 10-cm depth intervals. A total of 251 samples were analyzed in laboratory for a range of soil properties. Partial least square regression was used to develop soil-spectral relationship model. Predicted soil and uncertainty maps for soil properties were developed using 3D-DSM with RK from the calibration dataset (103 locations) and assessed using validation dataset (45 locations). Further three regression techniques, including generalized linear model (GLM), regression tree (RT), and random forest (RF) were tested and compared for accuracy and efficiency. Maps developed using sub samples (45 locations) identified by six SDs were further compared with the original map produced by the full dataset (148 locations) and individually validated by the rest 103 locations.The results showed that a good prediction was obtained for soil organic matter (SOM) and water-related soil properties from in-situ vis-NIR spectra, while a fair prediction was obtained for other properties. RF outperformed GLM and RT by quantifying the non-linear soil-landscape relationship, displaying weak spatial structure of regression residuals, and resulting in a more robust prediction model with high accuracy and low uncertainty. The predicted maps clearly presented the soil spatial variability, reflected the interactions among soil properties, and displayed the associated soil forming processes. Among the SDs, StRS with both good spatial and feature space coverage better represented the distribution of original maps and showed a small prediction uncertainty, while cLHS produced higher validation accuracy. SRS resulted in good validation results, while requires further exploration for its robustness. The main contribution of this thesis was to assess and optimize the methods and techniques for 3D-DSM and associated SDs and quantify both the horizontal and vertical variability of multiple soil properties.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction distillée sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Apprise à partir de 10 348 étiquettes directes de Codex et de 10 348 étiquettes directes de Gemma. Le mode candidate est l'union des têtes enseignantes seuillées; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont ni des étiquettes humaines ni des étiquettes directes de modèles de pointe.
Scores Codex et Gemma par catégorie
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,000 | 0,000 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,000 | 0,000 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,000 | 0,000 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,001 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule tête enseignante, pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».