Articles
Evaluating different multispectral vegetation indices for assessing the nitrogen status of white cabbage (Brassica oleracea var. capitata) under organic farming conditions
Article number
1360_12
Pages
91 – 98
Language
English
Abstract
The horticulture sector still widely lacks applicable remote sensing methods for assessing the nitrogen (N) status during crop growth.
The objective of this study was to evaluate different common vegetation indices (NDVI, GLI, NDYI, CCCI, NDRE) as indicator for the N status of organically managed white cabbage.
In two field experiments, a wide range of cabbage N supply levels (21 to 253 kg N ha‑1) derived from 12 preceding green manure crops differing in sward management and the proportion of legumes were examined.
During cabbage crop growth, plots were nadir-scanned four times with a 5-channel multispectral camera (MicaSense RedEdge MX) mounted on a quadcopter drone.
To obtain the reflectance behaviour of the plant cover only, picture areas with bare soil were excluded by image processing.
Initial and harvest N supply was quantified by soil and destructive plant sampling and validated by an associated soil incubation experiment.
For comparison of indices, index values were correlated with the respective incubation experimental data and total above-ground N at harvest.
The closest correlations were obtained in 2021 54 days after planting resulting in R2 values of 0.95 for CCCI, GLI and NDRE and of 0.92 for NDVI. In contrast, at this time no correlation was found for NDYI (R2<0.1). The results indicate that relative index suitability varies with cabbage growth stage and overall N supply level.
The objective of this study was to evaluate different common vegetation indices (NDVI, GLI, NDYI, CCCI, NDRE) as indicator for the N status of organically managed white cabbage.
In two field experiments, a wide range of cabbage N supply levels (21 to 253 kg N ha‑1) derived from 12 preceding green manure crops differing in sward management and the proportion of legumes were examined.
During cabbage crop growth, plots were nadir-scanned four times with a 5-channel multispectral camera (MicaSense RedEdge MX) mounted on a quadcopter drone.
To obtain the reflectance behaviour of the plant cover only, picture areas with bare soil were excluded by image processing.
Initial and harvest N supply was quantified by soil and destructive plant sampling and validated by an associated soil incubation experiment.
For comparison of indices, index values were correlated with the respective incubation experimental data and total above-ground N at harvest.
The closest correlations were obtained in 2021 54 days after planting resulting in R2 values of 0.95 for CCCI, GLI and NDRE and of 0.92 for NDVI. In contrast, at this time no correlation was found for NDYI (R2<0.1). The results indicate that relative index suitability varies with cabbage growth stage and overall N supply level.
Authors
F. Besand, K.-U. Katroschan
Keywords
vegetation index, unmanned aerial vehicles, drone imaging, remote sensing, vegetable production, N mineralization
Groups involved
- Division Precision Horticulture and Engineering
- Division Plant-Environment Interactions in Field Systems
- Division Vine and Berry Fruits
- Division Tropical and Subtropical Fruit and Nuts
- Division Plant Genetic Resources, Breeding and Biotechnology
- Division Temperate Tree Nuts
- Division Temperate Tree Fruits
- Division Vegetables, Roots and Tubers
- Division Sustaining Horticulture in a Changing World
- Working Group Mechanization, Digitization, Sensing and Robotics
Online Articles (45)