Leveraging geospatial data to improve soil characterization for precision agriculture
Notice bibliographique
Résumé
Understanding field heterogeneity allows resources (e.g., fertilizers, irrigation water, etc.) to be accurately controlled so that wastage is minimized and profit is maximized. The overall goal of this dissertation was to evaluate current soil mapping methods using grid-based composite sampling and proximal soil sensing (PSS) and to develop optimization and economic strategies to support precision farming applications. The first objective was to evaluate the optimization of conventional soil mapping through interpolating grid-based composite sampling data. Three types of sampling schemes, i.e., center-point-grid, diagonal-grid-cell and z-pattern-grid-cell, and three types of spatial interpolation methods, i.e., ordinary kriging (OK), inverse distance weighting (IDW) and Tile were evaluated. To explore the potential spatial heterogeneity on mapping quality, six sets of dense proximal soil sensing based apparent soil electrical conductivity (ECa) data layers were used to simulate ECa mapping and one set of high-density soil sampling data collected from Nebraska, USA, were used to simulate the mapping quality of phosphorus and potassium mapping. Two agricultural fields in north-central Ukraine were used to implement the investigated sampling strategies for phosphorus and potassium mapping. Under the second objective, the optimization of single sensor-based mapping was evaluated by investigating strategies for qualitative calibration sampling through an example of optic-sensor based soil organic matter (SOM) estimation. Soil reflectance measurements produced using a commercial two-channel [i.e., visible red (RED) and near-infrared (NIR)] optical reflectance sensor were regressed against laboratory analyzed SOM of calibration samples in 15 agricultural fields across the USA. Simple linear regression and D-optimality analyses were used to outline requirements for sensor implementation to declare successful calibration models. The third objective was the optimization of multi-sensors based mapping though evaluating the newly developed Neighborhood Search Algorithm for the multi-sensor calibration sampling design. Multi-layers sensing data was collected using a commercial Mobile Sensor Platform (MSP) housing ECa, optic and pH sensors as well as a real-time kinematic (RTK) level global navigation satellite system (GNSS) receiver. Soil samples obtained from ten designated calibration locations were used to correlate sensor measures (i.e., ECa, topography, RED and NIR reflectance) with soil properties of interest [i.e., pH, buffer pH, cation exchange capacity (CEC), particle size distribution (percent clay and sand) and SOM] for two fields in eastern Ontario, Canada (NX, 40 ha; ST, 45 ha). Best subset multiple linear regressions were used to establish calibration models for sensor-based mapping. 1-ha grid-based maps derived from OK based interpolation of 35 (NX) and 45 (ST) samples was compared to sensor-based maps. An additional ten random soil samples were used at each site to qualify the overall accuracy of each map. Under the final, fourth objective, the optimization of agricultural resources and profitability using geospatial data was approached by developing an economic assessment tool for variable rate applications (VRA). Benefits of optimizing irrigation water using variable rate irrigation (VRI) technology were accessed by incorporating spatially dynamic yield response and decline functions. To demonstrate tool performance, the ECa map was used as a proxy of soil water storage potential (WSP) and the topography map was employed as a proxy of landscape induced water-logging eï¬ects for a 20-ha ï¬eld in Southern Alberta. A total of 62 irrigation management scenarios, including: (1) no irrigation (NI), (2) uniform management (UM), (3) VRI Speed Control (SC) and (4) VRI Zone Control (ZC), were compared in terms of anticipated profitability.
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 | 0,001 | 0,002 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,000 |
| Méta-épidémiologie (sens large) | 0,001 | 0,000 |
| Bibliométrie | 0,000 | 0,000 |
| Études des sciences et des technologies | 0,002 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,001 | 0,000 |
| Intégrité de la recherche | 0,001 | 0,001 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,000 | 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 ».