Spectral estimation of the aboveground biomass of cotton under water-nitrogen coupling conditions.

Abstract

Aims Hyperspectral remote sensing technology can quickly obtain above-ground biomass (AGB) information of cotton, playing an important role in realizing accurate management for cotton cultivation. Methods Using Tahe-2 as the research object, nitrogen application rates and irrigation amounts were set to 0 (N 0 ), 100 (N 1 ), 150 (N 2 ), 200 (N 3 ), 250 (N 4 ) kg ha - 1 and 4500 (W 1 ), 6000 (W 2 ), 7500 (W 3 ) m³ ha - 1 under the coupled conditions of water and nitrogen. Through correlation analysis between cotton AGB and canopy spectral reflectance, the intersection of feature wavelengths screened by the successive projection algorithm (SPA) and highly significant wavelengths was used as the input vector for modeling. Support vector machine (SVM), regression tree (RT), and convolutional neural network (CNN) were employed to verify the accuracy. Results The results revealed the following: (1) The AGB of cotton at the bud stage was highest under the W 1 N 2 gradient. At the flowering stage, the highest AGB was observed under the W 3 N 2 gradient. At the boll stage, the highest AGB was under the W 3 N 0 gradient. (2) The optimal spectral model based on SVM for cotton AGB identification had higher R 2 values and lower RMSE values at the boll stage, with R 2 = 0.76, RMSE = 0.35 g and RPD = 17.59. The optimal spectral model based on RT had higher R 2 values and lower RMSE values at the bud stage, with R 2 = 0.79, RMSE = 0.24 g and RPD = 16.18. The optimal spectral model based on CNN also had higher R 2 values and lower RMSE values at the bud stage, with R 2 = 0.70, RMSE = 0.42 g and RPD = 4.50. These results indicated that the inversion effect at the bud stage was better than at other stages. Conclusions In terms of model testing, the RT model was found to be the most accurate for estimating cotton AGB, outperforming SVM and CNN.

Plant phenotyping relevance

綿花の地上部バイオマスをハイパースペクトル反射から推定するモデルを開発・比較し、精度検証しており、植物表現型取得手法が研究の中心である。

Code and data availability

The article describes cotton AGB spectral measurements and SVM/RT/CNN modeling in MATLAB R2022a, but contains no data availability statement, no public dataset deposit, no author code/scripts or trained model release, and no supplementary material with phenotyping data. Only the license and DOI URLs appear, neither of哪

No evidence-backed public reproduction asset is currently recorded.