Development of equipment to characterize soil attributes in different agricultural settings
Bibliographic record
Abstract
Agriculture intensification of crop production through heavy and uniform fertilization has harmful consequences. Precision agriculture (PA) offers a methodology to mitigate the negative effects of heavy fertilization by monitoring available soil nutrients to prevent excess fertilizing. Variable rate fertilizer application (VRA) provides information about the existing soil nutrients and their location in the field, thus, allowing farmers to apply fertilizer effectively. Georeferenced soil chemical information, such as soil pH and soluble nitrate, is important for successful VRA recommendation maps. This information can be measured using ion-selective electrode (ISE) sensors directly in-situ. Spot measurement using ISE can be done by making firm contact with the soil surface, known as direct soil measurement (DSM). Reliable automatic and manual soil sensing platforms are needed to perform DSM using ISEs. In this research, an automatic vehicle mounted on-the-spot soil analyzer (OSA) was developed. The OSA works by digging the soil to the depth defined by the operator, conduct DSM using multiple ISEs and cover the sampling hole and wash the ISEs when finished measuring. The OSA digging controller was able to manage the digging load accordingly under various field conditions. The OSA operation took 60 s per location to complete. Both the pH and nitrate ISEs were able to satisfactorly predict soil pH and nitrate with R2 of 0.59 and 0.72, respectively with RMSE of 0.43 pH and 0.16 pNO3. As an OSA alternative, a portable manual soil sensing platform was developed to accommodate the needs of small scale farmers and to provide a general overview of specific soil properties in the field. The manual soil sensing platform developed in this research offers a simple operation. Similar to standard soil core sampling, the operator pushes the platform into the soil. Then, conducting DSM by bringing the ISEs into contact with the sampled soil. The operator can monitor the ISE’s reading from the Arduino based DAQ connected to a generic Android Bluetooth terminal application. Manual pumping action will spray water to clean the ISEs after the measuring is finished. The developed manual soil analyzer provides georeferenced ISEs data and it is able to satisfactorily predict soil pH and nitrate with R2 between 0.53 to 0.88 with RMSE ranging from 0.17 to 0.36 pH for soil pH and R2 between 0.83 and 0.84 with RMSE of 0.21 to 0.29 pNO3 for soil nitrate prediction. Soil apparent electrical conductivity (ECa) maps provide valuable information for determining suitable sampling locations for an effective soil sensing platform operation. The interpretation of the ECa measurements is often site-specific, thus, needing an inversion to properly delineate the ECa data from the electromagnetic induction (EMI) sensor. In general, there are two approaches to invert the EMI measurements into depth specific soil ECa information: the finite element and fixed slice cumulative depth response approach. In this research, Brute-Force cumulative depth response inversion was developed due to its simplicity and straight forward calculation utilizing the low induction number (LIN) assumption of the EMI sensor. From filtered DUALEM-21S (Dualem, Inc. Milton, Ontario, Canada) ECa data, the Brute-Force ECa inversion produced two layer soil maps with their corresponding ECa and depth information. The software was tested successfully for characterizing the depth of the muck soil layer on typical horticulture farming operation in Quebec, Canada
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
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 source (direct Gemma or distilled Codex), 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".