Jingcheng Zhang1,2,3,4, Yanbo Huang5, Lin Yuan1,2,3,4, Guijun Yang1,2,3,4, Liping Chen1,2,3,4, Chunjiang Zhao1,2,3,4. 1. Beijing Research Centre for Information Technology in Agriculture, Beijing, China. 2. National Engineering Research Centre for Information Technology in Agriculture, Beijing, China. 3. Key Laboratory for Information Technologies in Agriculture, the Ministry of Agriculture, China. 4. Beijing Engineering Research Center of Agricultural Internet of Things, Beijing, China. 5. USDA-ARS, CPSRU, 141 Experiment Station Road, Stoneville, MS 38776, USA.
Abstract
BACKGROUND: Armyworm, a destructive insect for maize, has caused a wide range of damage in both China and the United States in recent years. To obtain the spatial distribution of the damage area, and to assess the damage severity, a fast and accurate loss assessment method is of great importance for effective administration. The objectives of this study were to determine suitable spectral features for armyworm detection and to develop a mapping method at a regional scale on the basis of satellite remote sensing image data. RESULTS: Armyworm infestation can cause a significant change in the plant's leaf area index, which serves as a basis for infestation monitoring. Among the number of vegetation indices that were examined for their sensitivity to insect damage, the modified soil-adjusted vegetation index was identified as the optimal vegetation index for detecting armyworm. A univariate model relying on two-date satellite images significantly outperformed a multivariate model, with the overall accuracy increased from 0.50 to 0.79. CONCLUSION: A mapping method for monitoring armyworm infestation at a regional scale has been developed, based on a univariate model and two-date multispectral satellite images. The successful application of this method in a typical armyworm outbreak event in Tangshan, Hebei Province, China, demonstrated the feasibility of the method and its promising potential for implementation in practice.
BACKGROUND: Armyworm, a destructive insect for maize, has caused a wide range of damage in both China and the United States in recent years. To obtain the spatial distribution of the damage area, and to assess the damage severity, a fast and accurate loss assessment method is of great importance for effective administration. The objectives of this study were to determine suitable spectral features for armyworm detection and to develop a mapping method at a regional scale on the basis of satellite remote sensing image data. RESULTS: Armyworm infestation can cause a significant change in the plant's leaf area index, which serves as a basis for infestation monitoring. Among the number of vegetation indices that were examined for their sensitivity to insect damage, the modified soil-adjusted vegetation index was identified as the optimal vegetation index for detecting armyworm. A univariate model relying on two-date satellite images significantly outperformed a multivariate model, with the overall accuracy increased from 0.50 to 0.79. CONCLUSION: A mapping method for monitoring armyworm infestation at a regional scale has been developed, based on a univariate model and two-date multispectral satellite images. The successful application of this method in a typical armyworm outbreak event in Tangshan, Hebei Province, China, demonstrated the feasibility of the method and its promising potential for implementation in practice.