Crop yield forecasting method based on crop model, history and meteorological forecasting data

Abstract

The invention belongs to the field of agricultural remote sensing, and particularly relates to a crop yield forecasting method based on crop models, history and meteorological forecasting data, whichcomprises the following steps: calibrating sensitive parameters by taking county as a reference partition; Screening high-quality reflectivity data in the MODIS reflectivity product in the growth period; coupling PROSAIL and WOFOST models, and constructing a cost function to establish a data assimilation framework so as to realize regional parameter optimization; Taking 50 groups of meteorologicalforecast data provided in the ECMWF as future 15d meteorological input of the WOFOST model; calculating the year with the highest similarity in the meteorological data of the same period of historical years of the pixel by using a generalized vector included angle method by using a vector formed by the highest daily temperature, and taking the meteorological data of the year as meteorological input of the WOFOST model in the future period of time; eriving the coupling model by using 50 groups of different meteorological input data to obtain 50 different forecast yields to realize probabilityforecast; and operating grid by grid to obtain a regional crop yield probability forecasting statistical graph. The method provided by the invention can accurately forecast the probability of high yield or real estate in partitions.

Description

technical field [0001] The invention belongs to the field of agricultural remote sensing, in particular to a crop yield forecasting method based on crop models, history and weather forecast data. Background technique [0002] With the deepening of the application and research of crop growth models, crop models have expanded from the earliest single-point-scale simulation to regional-scale simulation, and are especially widely used in climate change scenario simulation and remote sensing data assimilation. In the field of agricultural meteorology, there has always been an application demand for crop yield estimation, especially for accurate forecasting of large-scale crop yields. Therefore, it is an urgent problem to realize yield forecasting based on crop model assimilation. [0003] In terms of crop model application, the crop model requires high time resolution of the input meteorological elements, and needs to input the complete meteorological elements of the whole growth...

AI Technical Summary

Problems solved by technology

In existing studies, some studies directly use the data generated by regional climate models as monthly or quarterly weather forecast data. Due to certain errors in the generated meteorological data, it is necessary to correct and correct the output of the climate model. Day-to-day weather data simulated by high-resolution climate models can also cause mismatches in temporal and spatial scales

Images