【摘要】 本文提出了一种对玉米拔节期土壤墒情进行反演的改进水云模型;即:在经典水云模型基础上,考虑作物茎秆的散射因素,加入植被高度（h）和单位体积内植被叶和茎的雷达后向散射截面(σ_pq)两个参数,同时为扩大模型的适用范围,引入植被覆盖度,结合雷达Sentinel-1A多极化数据,通过AIEM模型构建土壤含水量与土壤表面后向散射系数之间的联系,最终建立雷达后向散射系数与植被覆盖下的土壤墒情之间的直接关系。结果表明:改进的水云模型在反演拔节期间玉米的土壤墒情时,反演精度比经典的水云模型有较大程度的提高,其中,VV极化效果更好,R²由0.532提高到0.645,RMSE从0.052降到0.039。结果验证了采用改进的水云模型能够有效地反演高秆作物覆盖地表的土壤墒情。更多还原

【Abstract】 An improved water-cloud model for inverting the soil moisture of corn jointing stage is put forward herein; that is to say, based on the classical water-cloud model, the scattering factor of crop stalk is considered along with adding both the parameters of the vegetation height（h） and the radar backscatter cross sections of vegetation leaves and stems per unit volume(σ_pq), and simultaneously,the vegetation coverage is introduced and combined with the radar multipolar data（Sentinel-1 A） through establishing the connection between the soil water content and the backscatter coefficient of soil surface for enlarging the application scope of the model, so as to make the direct relationship between the radar backscatter coefficient and the soil moisture under the vegetation coverage at last. The result shows that when the soil moisture of corn jointing stage is inverted with the improved water-cloud model, the inverting accuracy is largely enhanced if compared with that from the classical water-cloud model; in which the VV polarization effect is much better, i.e. R² being increased from 0.532 to 0.645 and RMSE being decreased from 0.052 to 0.039. The result demonstrates that the soil moisture of the soil surface covered by high-stem crops can be effectively inverted by the improved water-cloud model.更多还原