Analysis of prediction maps and data separation methods for site-specific management of wild blueberry
Bibliographic record
Abstract
Wild blueberry (Vaccinium angustifolium Ait.) is a key crop in the Lac-Saint-Jean region of Quebec. The industry totals $45 million annually. Wild blueberry is a lowbush species which flourishes in heterogeneous agronomic conditions where conventional crops cannot. It grows in areas of varying topography on sandy, acidic soils where competition with other plants is limited. Rhizome establishment takes years to develop, thus, bare spots are a common feature of young or poorly managed fields. Given the variation of soil, topography, and crop density, wild blueberry production would benefit from site-specific management, where levels of nutrient input are tailored to local needs based on within-field variation. A classic approach to site-specific management is the delineation of management zones, sub- field areas of relatively homogenous agronomic properties with uniform management rates. A second SSM approach is regression based, where a prescription regression equation based on sampled variables and known crop response to treatment is used for more continuous targeted treatment within the field. This thesis articulates the thematic mapping of agronomic variables and the comparison of two site-specific management strategies for wild blueberry using conventional soil sampling, proximal soil sensors, and multispectral satellite imagery. Two experimental sites were selected, one of varying topography and the other relatively at. Soil samples were collected in a 33-m grid scheme and tested for chemical and granulometric attributes. Soil apparent electrical conductivity (ECa) was collected with the non-contact DUALEM-21S sensor (Dualem Inc., Milton, ON) and the contact Veris 3100 sensor (Veris Technologies, Salina, KS). Elevation was mapped with real-time-kinematic (RTK) level global navigation satellite system (GNSS) receiver. Multispectral imagery acquired from the SPOT6 archive was radiometrically and atmospherically corrected, and a number of vegetation indices were derived from the image to map bare spots and compare VIs prediction of vigor to the sampled yield. Thematic maps were predicted from the sampled data using the Ordinary Kriging (OK) method and cross-validated to determine the strength of various data layers in predicting spatial patterns within-field. Classical statistics and geostatistics were performed on all sampled data. A classic approach to site-specific management through unsupervised classification of management zones was compared with a new regression-based approach which targets four field condition scenarios. Means of all properties in each of four scenarios were tested with ANOVA and Tukey's post-hoc test. In both MZA and the regression-based method, field conditions were most contrasted between scenarios EClow & Elevhigh and EChigh & Elevlow. The regression-based method separated data similarly or better than the MZA approach, while providing more precise areas to develop a regression-based prescription map.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".