|
| 基于微波散射传输和盐渍化土壤介电混合模型的土壤水盐联合反演 |
| Joint retrieval of soil moisture and salinity based on microwave scattering transfer and a dielectric mixing model for saline soils |
| 投稿时间：2026-03-31 |
| DOI：10.13254/j.jare.2026.0366 |
| 中文关键词: 土壤水分，土壤盐分，土壤介电混合模型，物理散射模型，主动微波遥感 |
| 英文关键词: soil moisture, soil salinity, soil dielectric mixing model, physical scattering model, active microwave remote sensing |
| 基金项目: |
| 作者 | 单位 | E-mail | | 闵肖肖 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | | | 段四波 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | duansibo@caas.cn | | 鲁志威 | 航天东方红卫星有限公司, 北京 100094 | | | 韩文静 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | | | 申晔琳 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | | | 吴双飞 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | | | 陈奕羽 | 北方干旱半干旱耕地高效利用全国重点实验室, 中国农业科学院农业资源与农业区划研究所, 北京 100081 | | | 陈颂超 | 浙江大学环境与资源学院, 杭州 310058 | | | 史舟 | 浙江大学环境与资源学院, 杭州 310058 | |
|
| 摘要点击次数: 446 |
| 全文下载次数: 86 |
| 中文摘要: |
| 土壤水分与盐分协同监测对盐碱耕地高效利用至关重要，然而现有盐分遥感估算多依赖机器学习，主动微波遥感土壤水盐联合物理反演框架仍不完善，盐渍化土壤介电混合模型的可靠性有待验证。为此，本文以黄河三角洲典型滨海盐渍土区——山东省东营市垦利区为研究区，基于两期野外采样数据和哨兵1号雷达影像，利用原位介电观测对Dobson介电混合模型中盐分影响介电损耗的校正参数进行局地化优化标定，并将其与微波物理散射传输AIEM–Oh和WCM模型耦合，构建了土壤水盐联合微波物理反演框架。结果表明，校准后的介电模型显著提升了垦利区盐渍化土壤介电模拟精度，相比于原始Dobson模型，秋季采样期的复介电常数实部和虚部模拟R2分别提高0.640和4.212，RMSE分别降低5.00和33.31，明显改善了介电常数虚部高估问题。基于联合反演框架实现了研究区植被覆盖期土壤水盐同步反演，土壤水分反演验证精度R2达到0.621，RMSE为0.035 cm3·cm-3，土壤盐分反演验证精度R2达到0.750，RMSE为1.191 g·kg-1。盐分影响校准后的土壤介电混合模型能够支撑基于微波散射传输机理的土壤水盐联合反演，所构建的面向哨兵1号高分辨率雷达的联合反演框架可用于滨海盐渍土区土壤水盐协同监测。 |
| 英文摘要: |
| The synergistic monitoring of soil moisture and salinity is essential for the efficient utilization of salt-affected cropland. However, existing remote sensing approaches for soil salinity estimation largely rely on machine learning, while phy sics-based joint retrieval frameworks for soil moisture and salinity using active microwave observations remain underdeveloped, and the reliability of dielectric mixing models for saline soils requires further validation. To address these issues, this study selected Kenli District, Dongying City, Shandong Province, a typical coastal saline soil region in the Yellow River Delta, as the study area. Based on two field sampling campaigns and Sentinel-1 radar imagery, in situ dielectric observations were used to locally optimize the salinity-related correction parameters controlling dielectric loss in the Dobson dielectric mixing model. The calibrated dielectric model was then coupled with microwave phy sical scattering transfer models, including AIEM-Oh and WCM, to construct a phy sically based joint microwave inversion framework for soil moisture and salinity. The results showed that the calibrated dielectric model substantially improved the simulation accuracy of dielectric properties for saline soils in Kenli District. Compared with the original Dobson model, the R2 values for the real and imaginary parts of the complex dielectric constant increased by 0.640 and 4.212, respectively, while the corresponding RMSE values decreased by 5.00 and 33.31 in the autumn sampling period. A marked improvement in correcting the overestimation of the imaginary part of the dielectric constant was also observed. The proposed joint inversion framework enabled the simultaneous retrieval of soil moisture and salinity during the vegetation-covered period. The validation accuracy reached R2=0.621 and RMSE=0.035 cm3· cm-3 for soil moisture, and R2=0.750 and RMSE=1.191 g·kg-1 for soil salinity. These results demonstrate that the salinity-calibrated dielectric mixing model can effectively support phy sically based joint inversion of soil moisture and salinity through microwave scattering mechanisms. The salinity-calibrated soil dielectric mixing model can support the joint retrieval of soil moisture and salinity based on microwave scattering transfer mechanisms. The developed joint retrieval framework for Sentinel-1 high-resolution radar data can be applied to the synergistic monitoring of soil moisture and salinity in coastal salt-affected soils. |
| HTML
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
