适应光照环境变化的多种农作物的GPP遥感估测模型研究

Crop gross primary production (GPP) is a result of the canopy interception of solar radiation by photosynthesis, which is important for evaluating the sink of atmospheric CO2. Remote sensing is a powerful and expedient tool for assessing crop GPP over large areas, providing valuable information for regional and global crop growing status. Differences in light climate and vegetation species are two factors resulting in uncertainties of GPP estimates by remotely sensed data. Croplands in China are usually planted with vegetation of different species, and under complicated light climates. Therefore, the study on remote estimation of crop GPP resistant to variations in species and light climate is essential for monitoring agriculture development in China. .In this study, several typical crop species in China will be selected. Based on lab experiments, model simulated data and in-situ measurements, we will study crop GPP variations due to changing light climate and species. Conditions of maximum light, optimal light and insufficient light will be simulated. Based on the observations, appropriate adjustment parameters will be included to improve the model performance. Using crop spectral reflectance in visible to near-infrared region, together with meteorology data, we will develop an optimized model capable of estimating crop GPP in condition of different species and light environment.

农作物总初级生产力（GPP）体现了植物的光合作用能力，是构成碳汇功能的重要过程。利用遥感技术实现大尺度的监测农田GPP的时空分布和变化，为区域性的农情分析提供了重要的基础数据。光照环境和植被类型的差异是导致农作物GPP遥感估测模型不确定性的两个重要因素。针对我国农田作物混种，光照环境多变的特点，建立一种能适应不同光照环境和作物类型的GPP遥感估测模型，将为合理的利用和管理农业环境提供更可靠的辅助信息。.本项目将选取我国几种典型的农作物，模拟光照不足，光照适宜和光照过强三种情况，基于实验室测量，模型模拟数据和野外试验田测量方法，从遥感光学机理上理解农作物GPP对植被类型和光照环境变化的响应规律。将分析所观测的规律加入合适的调节因子，利用植被在可见光到近红外光谱范围内的反射率数据，结合气象环境数据，建立一个适用不同光照条件，无需先验作物类型知识，能准确估算含有混合作物区域的GPP遥感模型。

农作物总初级生产力（GPP）体现了植物的光合作用能力，是构成碳汇功能的重要过程。利用遥感技术实现大尺度监测农田GPP的时空分布和变化，可为区域性农情分析提供重要的基础数据。光照环境和植被类型的差异是导致农作物GPP遥感估测模型不确定性的两个重要因素。针对我国农田作物混种，光照环境多变的特点，建立一种能适应不同光照环境和作物类型的GPP遥感估测模型，将为合理的利用和管理农业环境提供更可靠的辅助信息。本项目选取我国几种典型的农作物，基于实验室测量、模型模拟数据和野外试验田测量方法，从遥感光学机理上理解影响农作物GPP变化的环境因素和植被参量。将分析所观测的规律加入合适的调节因子，利用植被在可见光到近红外光谱范围内的反射率数据，通过植被指数和机器学习的方法，建立一个适用不同光照条件，无需先验作物类型知识，能准确估算含有混合作物区域的GPP遥感模型。