Interpolation of the Mean Anomalies for Cloud Filling in Land Surface Temperature and Normalized Difference Vegetation Index

Abstract

When monitoring time series of remote sensing data, it is advisable to fill gaps, i.e., missing or distorted data, caused by atmospheric effects or technical failures. In this work, a new method for filling these gaps called interpolation of the mean anomalies (IMA) is proposed and compared with some competitors. The method consists of: 1) defining a neighbourhood for the target image from previous and subsequent images across previous and subsequent years; 2) computing the mean target image of the neighbourhood; 3) estimating the anomalies in the target image by subtracting the mean image from the target image; 4) filtering the anomalies; 5) averaging the anomalies over a predefined window; 6) interpolating the averaged anomalies; and 7) adding the interpolated anomalies to the mean image. To assess the performance of the IMA method, both a real example, and a simulation study are conducted with a time series of MODIS TERRA and MODIS AQUA images captured over the region of Navarre (Spain) from 2011 to 2013. We analyze the land surface temperature (LST) day and night, and the normalized difference vegetation index (NDVI). In the simulation study seven sizes of artificial clouds are randomly introduced to each image in the studied time series. The square root of the mean squared prediction error (RMSE) between the original and the filled data is chosen as an indicator of the goodness of fit. The results show that the IMA method outperforms Timesat, Hants, and Gapfill in filling small, moderate, and big cloud gaps in both the day and night LST and NDVI data, reaching RMSE reductions of up to 23%.

A Comparison of SSEBop-Model-Based Evapotranspiration with Eight Evapotranspiration Products in the Yellow River Basin, China

“…The filling method for the LST missing value causes different errors [46]. This paper adopted the adaptive window method, which considers the relationship between high-quality pixel LST values and influencing factors (e.g., elevation, NDVI, and air temperature) to fill the missing values [77].…”

Downscaling ESA CCI Soil Moisture Based on Soil and Vegetation Component Temperatures Derived From MODIS Data

“…The coarse-resolution pixels that covered by high-resolution pixels with percentage higher than 70% were considered to be full covered by high-resolution pixels to reduce the influence of cloud, it might not be the truth when the surface coverage is complex and heterogeneous. Research on NDVI and LST reconstruction can fill the data gaps caused by cloud cover [42,46,47], which would improve the application of the proposed method. When removing the bias of the downscaled soil moisture, the Kriging method was applied in this study.…”

Spatial–Spectral Radial Basis Function-Based Interpolation for Landsat ETM+ SLC-Off Image Gap Filling

“…Essentially, the principle of gap filling for SLC-off images is similar to other data reconstruction issues, such as cloud removal and infilling [11], [42], [43]; all of them are performed by borrowing information from the spatially and temporally neighboring data of the gap image. Moreover, besides optical images, there are also missing data in a variety of quantitative products, such as global soil moisture [44], [45], land surface temperature (LST) [46], [47], and the normalized difference vegetation index (NDVI) [48], [49]. Given the competitive performance of SSRBF in this article, it has great potential for other data reconstruction problems.…”