文档介绍：
利用 Terra 和 Aqua 双星 MODIS 数据协同
反演日气温#
石晶晶,黄敬峰**
5
10
15
20
25
30
(浙江大学农业遥感与信息技术应用研究所)
摘要:本文分别采用分段三次 Hermite 插值法(PCHI)和小波分析方法(WT)两种方法对时间序
列 EVI 进行重构,基于 2000-2010 年 Terra 和 Aqua 的 MODIS LST 数据,采用多元逐步回归
法构建了长三角地区日平均气温的估算模型并进行验证。结果表明:WT 法可以有效去除因
气象条件、太阳观测角等因素带来的干扰,可用于植被物候监测,而 PCHI 法对 EVI 重构后
适合定量化研究。基于 Terra 和 Aqua 双星 MODIS LST 数据的日平均气温估算模型精度最高,
其 RMSE 为 ℃,基于单一时相 LST 建立的模型 RMSE 最高为 ℃。本文提出的利
用 Terra 和 Aqua 双星 MODIS 数据协同反演气温的算法充分体现了 MODIS 的高时间分辨率
的优势,可以获取长三角地区准确、详细的日平均气温空间分布。
关键词:MODIS;日气温;EVI 重构;长三角
中图分类号:TP79
Retrieval of daily air temperature by integrating MODIS
onboard Terra and Aqua
SHI Jingjing, HUANG Jingfeng
(Institute of Agricultural Remote Sensing and Information Application, Zhejiang University,
Hangzhou 310058)
Abstract: In this paper piecewise cubic hermite interpolation (PCHI) and wavelet transform (WT)
were used for time series EVI reconstruction. The estimation models for mean daily air
temperature in Yangtze River delta region were developed using multiple regression method based
on Terra and Aqua MODIS LST data. The results indicated that WT can effectively remove the
influence of weather condition, sun angle and other factors, and more suitable for vegetation
phenology detection, while PCHI was suitable for quantitative research. The bining
Terra and Aqua MODIS LST data has the highest accuracy with RMSE of ℃. The m