Scientists have made progress in predicting cotton production in Xinjiang

Recently, the hyperspectral team of the National Engineering Research Center for Remote Sensing Satellite Applications of the Aerospace Information Innovation Institute of the Chinese Academy of Sciences (CAA) and the College of Agriculture of Xinjiang Shihezi University have made progress in predicting cotton production in the middle of the growing season in the main cotton area of Xinjiang.

Cotton is widely grown in many developing countries, including China, and provides a large number of natural fibers and edible oils for countries around the world. The development of the cotton industry contributes to the achievement of the United Nations 2030 Sustainable Development Goals (SDGs), reducing extreme poverty (SDG1) and achieving Zero Hunger (SDG2). Timely and accurate cotton yield forecasting plays an important guiding role in the formulation of national agricultural policies, and is of great significance in international trade, humanitarian aid, food security, etc.

To this end, the research team explored the feasibility of the spatio-temporal prediction (STP) product of SIF in cotton yield prediction by using multi-source remote sensing data such as Sentinel-2 optical remote sensing images and multi-scale satellite sunlight-induced chlorophyll fluorescence (SIF) products on the basis of the “One Map of Cotton” work. The research team first proposed the STP-SIF problem (Fig. 1), analyzed its underlying driving mechanism, and designed a time-series data-driven deep learning method (Fig. 2), and secondly, produced the STP-SIF product and explored the accuracy of the product in predicting the yield of the cotton growing season in the middle of the growing season.

Fig.1 Schematic diagram of sif spatiotemporal prediction (STP-SIF).

Figure 2 Design of the potential network structure for STP-SIF problems

The results show that the proposed method can accurately predict the SIF from the middle to the end of the growing season, and the cotton yield of STP-SIF products in the season can be accurately predicted based on the known SIF and the normalized differential moisture index NDWI (Fig. 3). The best prediction accuracy R2 of cotton yield can reach 0.70 (1 month) and 0.66 (2 months) 1-2 months before harvest, respectively.

Fig.3 Mapping of cotton yield forecast 1-2 months before harvest

This study is the first exploration of the STP-SIF problem, which preliminarily verifies the weak spatial scale dependence of the problem, and is expected to provide a feasible framework for the prediction of SIF in the middle and late stages of the crop growing season.

The research results were published in the top remote sensing journal “Remote Sensing of” with the title of “Regional-scale cotton yield forecast via data-driven spatio-temporal prediction (STP) of solar-induced chlorophyll fluorescence (SIF)”. Environment)》。 The Aerospace Institute is the first completion unit. Dr. Kang Xiaoyan of the Academy of Aerospace Sciences is the first author, researcher Huang Changping is the corresponding author, and researcher Zhang Lifu of the Institute of Aerospace Sciences, associate professor Zhang Ze, professor Lv Xin, and doctoral student Wang Huihan of Shihezi University participated in the research.

His research work has been supported by the “From 0 to 1” original innovation project of the Basic Frontier Scientific Research Program of the Chinese Academy of Sciences, the National Natural Science Foundation of China, the Science and Technology Project of the Xinjiang Corps, and the outstanding member of the Youth Innovation Promotion Association of the Chinese Academy of Sciences. (Source: Aerospace Information Research Institute, Chinese Academy of Sciences)

Related Paper Information:

Special statement: This article is reproduced only for the purpose of disseminating information, and does not mean that it represents the views of this website or confirms the authenticity of its content; if other media, websites or individuals reprint from this website, they must retain the “source” indicated on this website, and bear their own legal responsibilities such as copyright; if the author does not want to be reprinted or contact the reprint fee and other matters, please contact us.

Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button