Regional crop maturing stage prediction method based on accumulated temperature, radiation and soil water content

Abstract

The invention belongs to the field of agricultural remote sensing, and relates to a regional crop maturing stage prediction method based on accumulated temperature, radiation and soil water content. The specific steps of the method include: S1, synthesizing MODIS NDVI data of a to-be-predicted crop at a whole growing stage according to a time sequence, and extracting a planting region by employinga random forest algorithm; S2, constructing an accumulated temperature and radiation model from a heading stage to a maturing stage in a full irrigation condition; S3, constructing a Thiessen polygonby regarding agricultural weather stations as nodes; S4, calibrating a SWAP model of the to-be-predicted crop of a research region; S5, establishing a maturing stage correction factor lookup table ofthe soil water content at the whole growing stage and a maturing date of the to-be-predicted crop in a water stress condition; and S6, dynamically adjusting maturing stage model results of accumulated temperature and radiation grid unit by grid unit, and obtaining an optimal maturing stage prediction result of the to-be-predicted crop. According to the method, advantages of a statistical model and a mechanism model are combined, the maturing stage of the crop can be predicted in a large-region scale, and the prediction precision of the maturing stage of the crop is improved.

Description

Technical field [0001] The invention belongs to the field of agricultural remote sensing, and specifically relates to a method for predicting the mature period of regional crops based on accumulated temperature, radiation and soil water content. Background technique [0002] The maturity period of crops is an important agronomic parameter. Due to conditions such as light, temperature and water, the maturity period has strong spatial variability. In the prior art, statistical models are used for prediction, but the disadvantage is the lack of the comprehensive influence of light, temperature and water interaction on the maturity period. In the prior art, the mechanism model is used for prediction, and the disadvantages are that there are many input parameters and the amount of calculation is large, which is difficult to apply to the prediction of the mature period on a large regional scale. At present, there is a lack of a mature period prediction method that combines statistical...

AI Technical Summary

Problems solved by technology

Statistical models are used for prediction in the prior art, but the disadvantage is the lack of comprehensive influence of light, temperature and water interaction on the maturity period

In the existing technology, the mechanism model is used for prediction, and the disadvantages are that there are many input parameters and a large amount of calculation, so it is difficult to apply to the prediction of maturity period on a large regional scale

There is currently a lack of maturity prediction methods that combine statistical and mechanistic models

Images