【摘要】 针对当前利用 MODIS 数据反演地表温度过于复杂和缺乏简便易行的实地验证方法的不足,尝试了改进反演方法的研究,提出了反演结果的地面准同步测量验证方法和在沙漠化地区分类计算地表比辐射率的方法,简化了原劈窗算法所需参数的计算过程。利用改进后的方法对新疆沙漠化地区的地表温度进行了反演,并同步进行了该区域在各种天气条件和地表条件下的地表温度反演结果的验证。经分析表明,在天空晴朗无云,无风,植被稀疏,地表类型均一,面积较大,沙质土壤且观测时刻与卫星过境时刻接近的条件下,该方法精度较高。研究表明改进的利用 MODIS 影像数据反演沙漠化地区地表温度方法和地面验证方法具有一定的可行性,适合于对沙化地区地表温度大范围且快速的遥感监测。更多还原

【Abstract】 Considering that the present land surface temperature(LST)retrievals from the earth observing satellite(EOS)- moderate resolution imaging speetroradiometer(MODIS)data have the drawbacks of high computing complexity and lack- ing of real-time validation of retrieval products,the authors tried the study of improvement on LST retrieval,and proposed the validation method by synchronous large-scale ground measurement,and presented the classifiable computation of land surface emissivity for simplifying the split window algorithm’ s computation process.The improved method was applied to the LST retrieval from the MODIS data in the Xinjiang desertifieation region,and all the retrieval products obtained in various weather and land surface conditions were validated synchronously.The comparison analysis showed that the accu- racy of the improved method was pretty high in the conditions of sunny day,vegetation sparseness,uniformity in land sur- face type,bigger area,sandy soil,and simultaneousness in observation and the satellite transit.The research results showed that the improved method for retrieving I_ST in Aeolian desertification areas from MODIS data and the synchronous large-scale ground validation were feasible to a certain degree,which were suitable for quick LST monitoring at large- scale by remote sensing.更多还原