NDMI与TCW在西藏低植被区土壤湿度监测中的比较

Comparison of NDMI and TCW for Soil Moisture Monitoring in Tibet Low-vegetative Area

土壤湿度的遥感监测在农牧业生产中发挥着重要作用。基于Landsat TM影像,利用研究区不同深度的土壤湿度实测数据,对两种可以反映地表湿度的指标归一化差异湿度指数（Normalized Difference Moisture Index,NDMI）和缨帽变换的湿度分量（Tasseled Cap Wetness,TCW）在西藏低植被覆盖区土壤湿度的监测结果进行精度验证与对比分析。结果表明：在0~5 cm、10 cm和20 cm深度处,NDMI与土壤湿度实测值回归方程的回归系数分别为0.6181、0.6853、0.5764,TCW与土壤湿度实测值回归方程的回归系数分别为0.4586、0.5101、0.5628。此外,NDMI、TCW与实测值的变化趋势一致,但NDMI对土壤湿度变化更加敏感。通过对比NDMI与TCW反演结果的空间分布情况也表明,NDMI对土壤湿度的变化具有更好的区分能力,尤其是在高海拔的低植被覆盖区。该研究可为高原低植被覆盖区土壤湿度监测方法的选择与误差分析提供参考依据。

Soil moisture monitoring by re mote sensing plays a significant role in agricultural and stock raising production. Based on Landsat TM image, Normalized Difference Moisture Index（NDMI） and Tasseled Cap Wetness（TCW） were applied to obtain soil moisture in the Tibet low-vegetative area. Then, the observed soil moisture data in the study area were used to compare and validate NDMI and TCW. The results showed that the correlation coefficient between NDMI and observed soil moisture data was 0.6181, 0.6853 and 0.5764 at depths of 0- 5, 10 and 20 cm,respectively, and the correlation coefficient between TCW and observed data was 0.4586, 0.5101 and 0.5628,respectively. Moreover, the variation trends of NDMI and TCW were both similar to observed data, but NDMI was more sensitive to soil moisture changes. The comparison of spatial distribution of NDMI and TCW also indicated that NDMI was better in distinguishing the difference of soil moisture, especially in high altitude areas. This research may provide valuable information for method selection and error analysis in soil moisture monitoring of low-vegetative area by remote sensing.