Predicting In-Season Corn Grain Yield Using Optical Sensors
In-season sensing can account for field variability and improve nitrogen (N) management; however, opportunities exist for refinement. The purpose of this study was to compare different sensors and vegetation indices (VIs) (normalized difference vegetation index (NDVI); normalized difference red edge...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/12/10/2402 |
| _version_ | 1827652337609474048 |
|---|---|
| author | Camden Oglesby Amelia A. A. Fox Gurbir Singh Jagmandeep Dhillon |
| author_facet | Camden Oglesby Amelia A. A. Fox Gurbir Singh Jagmandeep Dhillon |
| author_sort | Camden Oglesby |
| collection | DOAJ |
| description | In-season sensing can account for field variability and improve nitrogen (N) management; however, opportunities exist for refinement. The purpose of this study was to compare different sensors and vegetation indices (VIs) (normalized difference vegetation index (NDVI); normalized difference red edge (NDRE); Simplified Canopy Chlorophyll Content Index (SCCCI)) at various corn stages to predict in-season yield potential. Additionally, different methods of yield prediction were evaluated where the final yield was regressed against raw or % reflectance VIs, relative VIs, and in-season yield estimates (INSEY, VI divided by growing degree days). Field experiments at eight-site years were established in Mississippi. Crop reflectance data were collected using an at-leaf SPAD sensor, two proximal sensors: GreenSeeker and Crop Circle, and a small unmanned aerial system (sUAS) equipped with a MicaSense sensor. Overall, relative VI measurements were superior for grain yield prediction. MicaSense best predicted yield at the VT-R1 stages (R<sup>2</sup> = 0.78–0.83), Crop Circle and SPAD at VT (R<sup>2</sup> = 0.57 and 0.49), and GreenSeeker at V10 (R<sup>2</sup> = 0.52). When VIs were compared, SCCCI (R<sup>2</sup> = 0.40–0.49) outperformed other VIs in terms of yield prediction. Overall, the best grain yield prediction was achieved using the MicaSense-derived SCCCI at the VT-R1 growth stages. |
| first_indexed | 2024-03-09T20:54:05Z |
| format | Article |
| id | doaj.art-f9af75b6d3974b27a33b9e227b9ad607 |
| institution | Directory Open Access Journal |
| issn | 2073-4395 |
| language | English |
| last_indexed | 2024-03-09T20:54:05Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj.art-f9af75b6d3974b27a33b9e227b9ad6072023-11-23T22:26:32ZengMDPI AGAgronomy2073-43952022-10-011210240210.3390/agronomy12102402Predicting In-Season Corn Grain Yield Using Optical SensorsCamden Oglesby0Amelia A. A. Fox1Gurbir Singh2Jagmandeep Dhillon3Plant and Soil Sciences, Mississippi State University, Starkville, MS 39762, USAPlant and Soil Sciences, Mississippi State University, Starkville, MS 39762, USAPlant Science & Technology, University of Missouri, Novelty, MO 64399, USAPlant and Soil Sciences, Mississippi State University, Starkville, MS 39762, USAIn-season sensing can account for field variability and improve nitrogen (N) management; however, opportunities exist for refinement. The purpose of this study was to compare different sensors and vegetation indices (VIs) (normalized difference vegetation index (NDVI); normalized difference red edge (NDRE); Simplified Canopy Chlorophyll Content Index (SCCCI)) at various corn stages to predict in-season yield potential. Additionally, different methods of yield prediction were evaluated where the final yield was regressed against raw or % reflectance VIs, relative VIs, and in-season yield estimates (INSEY, VI divided by growing degree days). Field experiments at eight-site years were established in Mississippi. Crop reflectance data were collected using an at-leaf SPAD sensor, two proximal sensors: GreenSeeker and Crop Circle, and a small unmanned aerial system (sUAS) equipped with a MicaSense sensor. Overall, relative VI measurements were superior for grain yield prediction. MicaSense best predicted yield at the VT-R1 stages (R<sup>2</sup> = 0.78–0.83), Crop Circle and SPAD at VT (R<sup>2</sup> = 0.57 and 0.49), and GreenSeeker at V10 (R<sup>2</sup> = 0.52). When VIs were compared, SCCCI (R<sup>2</sup> = 0.40–0.49) outperformed other VIs in terms of yield prediction. Overall, the best grain yield prediction was achieved using the MicaSense-derived SCCCI at the VT-R1 growth stages.https://www.mdpi.com/2073-4395/12/10/2402active sensorscornnitrogenremote sensing |
| spellingShingle | Camden Oglesby Amelia A. A. Fox Gurbir Singh Jagmandeep Dhillon Predicting In-Season Corn Grain Yield Using Optical Sensors Agronomy active sensors corn nitrogen remote sensing |
| title | Predicting In-Season Corn Grain Yield Using Optical Sensors |
| title_full | Predicting In-Season Corn Grain Yield Using Optical Sensors |
| title_fullStr | Predicting In-Season Corn Grain Yield Using Optical Sensors |
| title_full_unstemmed | Predicting In-Season Corn Grain Yield Using Optical Sensors |
| title_short | Predicting In-Season Corn Grain Yield Using Optical Sensors |
| title_sort | predicting in season corn grain yield using optical sensors |
| topic | active sensors corn nitrogen remote sensing |
| url | https://www.mdpi.com/2073-4395/12/10/2402 |
| work_keys_str_mv | AT camdenoglesby predictinginseasoncorngrainyieldusingopticalsensors AT ameliaaafox predictinginseasoncorngrainyieldusingopticalsensors AT gurbirsingh predictinginseasoncorngrainyieldusingopticalsensors AT jagmandeepdhillon predictinginseasoncorngrainyieldusingopticalsensors |