College of Geological Engineering and Geomatics, Chang'an University, Xi'an, China.
School of Land Engineering, Chang'an University, Xi'an , 710064, China.
Environ Sci Pollut Res Int. 2022 Oct;29(47):70882-70898. doi: 10.1007/s11356-022-20771-4. Epub 2022 May 19.
Remote sensing dynamic monitoring methods often benefit from a dense time series of observations. To enhance these time series, it is sometimes necessary to integrate data from multiple satellite systems. In particular, the Landsat and Sentinel series provide a rich source of data for Earth observations. National Aeronautics and Space Administration (NASA) scientists proposed a method that creates global fixed per-band transformation coefficients to reduce the reflectance difference between Landsat-8 and Sentinel-2 for the harmonized Landsat and Sentinel-2 (HLS) surface reflectance product. However, the coefficient has yet to be further validated in the target study area and the coefficient can only be used for Landsat-8 and Sentinel-2, and is not useful for other sensors. The purpose of this study is to evaluate the potential of integrating surface reflectance data from Landsat-7, Landsat-8, and Sentinel-2. Some differences in the surface reflectance of the sensor pairs were identified, based upon which a cross-sensor conversion model was proposed, i.e., a suitable adjustment equation was fitted using an ordinary least squares (OLS) linear regression method to convert the Sentinel-2 reflectance values closer to the Landsat-7 or Landsat-8 values. The results show that the model adjusted the Sentinel-2 surface reflectance to match Landsat-7 or Landsat-8. The maximum MRE of the adjusted sensor for surface reflectance was reduced from 17.96 to 12.15%. Differences in reflectance produce corresponding differences in estimates of biophysical quantities, such as NDVI, with MRE as high as 18.33%. However, adjusting the Sentinel-2 sensor was able to reduce this part of the discrepancy to about 12.56%. The study believes that despite the differences in these datasets, it appears feasible to integrate these datasets by applying a linear regression correction between the bands.
遥感动态监测方法通常受益于密集的时间序列观测。为了增强这些时间序列，有时需要整合来自多个卫星系统的数据。特别是，Landsat 和 Sentinel 系列为地球观测提供了丰富的数据来源。美国国家航空航天局（NASA）的科学家提出了一种方法，该方法创建了全球固定的波段变换系数，以减少 Landsat-8 和 Sentinel-2 之间的反射率差异，从而生成协调的 Landsat 和 Sentinel-2（HLS）地表反射率产品。然而，该系数尚未在目标研究区域进一步验证，并且该系数仅可用于 Landsat-8 和 Sentinel-2，对其他传感器则没有用。本研究的目的是评估整合 Landsat-7、Landsat-8 和 Sentinel-2 地表反射率数据的潜力。根据传感器对之间地表反射率的差异，提出了一种交叉传感器转换模型，即使用普通最小二乘法（OLS）线性回归方法拟合合适的调整方程，以将 Sentinel-2 反射率值转换得更接近 Landsat-7 或 Landsat-8 值。结果表明，该模型调整了 Sentinel-2 地表反射率以匹配 Landsat-7 或 Landsat-8。调整后的传感器对地表反射率的最大 MRE 从 17.96%降低到 12.15%。反射率的差异会导致对生物物理量（如 NDVI）的估计产生相应的差异，其 MRE 高达 18.33%。然而，通过在波段之间应用线性回归校正，可以将 Sentinel-2 传感器的这部分差异降低到约 12.56%。该研究认为，尽管这些数据集存在差异，但通过在波段之间应用线性回归校正，似乎可以整合这些数据集。
