Monitoring and Prediction of Wild Blueberry Phenology Using a Multispectral Sensor
(1) Background: Research and development in remote sensing have been used to determine and monitor crop phenology. This approach assesses the internal and external changes of the plant. Therefore, the objective of this study was to determine the potential of using a multispectral sensor to predict p...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/17/2/334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | (1) Background: Research and development in remote sensing have been used to determine and monitor crop phenology. This approach assesses the internal and external changes of the plant. Therefore, the objective of this study was to determine the potential of using a multispectral sensor to predict phenology in wild blueberry fields. (2) Method: A UAV equipped with a five-banded multispectral camera was used to collect aerial imagery. Sites consisted of two commercial fields, Lemmon Hill and Kemptown. An RCBD with six replications, four treatments, and a plot size of 6 × 8 m with a 2 m buffer between plots was used. Orthomosaic maps and vegetative indices were generated. (3) Results: There were significant correlations between VIs and growth parameters at different stages. The F4/F5 and F6/F7 stages showed significantly high correlation values among all growth stages. LAI, floral, and vegetative bud stages could be estimated at the tight cluster (F4/F5) and bloom (F6/F7) stages with R<sup>2</sup>/CCC = 0.90/0.84. Variable importance showed that NDVI, ENDVI, GLI, VARI, and GRVI contributed significantly to achieving these predicted values, with NDRE showing low effects. (4) Conclusion: This implies that the F4/F5 and F6/F7 stages are good stages for making phenological predictions and estimations about wild blueberry plants. |
|---|---|
| ISSN: | 2072-4292 |