Category: Time series and analysis
Gap Filling
Gap filling estimates missing values in time series caused by clouds, shadows, or data dropouts.
Also known as: imputation, missing data filling
Expanded definition
Gap filling is any method that estimates values where observations are missing. In optical time series, gaps are often caused by cloud masking and cloud shadow.
Simple approaches include temporal interpolation or compositing, while more advanced methods use multi-sensor fusion or model-based estimation. The key point is that filled values are not direct observations, so they should be treated as estimates.
For monitoring and analytics, gap filling can improve continuity, but it can also hide uncertainty. Good products document whether a pixel is observed or filled and avoid using future observations to fill the past when the goal is real-time monitoring.
Related terms
Cloud Mask
A cloud mask labels pixels likely affected by clouds so they can be excluded or handled differently.
Data Fusion
Data fusion combines multiple data sources or sensors to create a more complete or consistent product.
Temporal Integrity
Temporal integrity means an output for a given date is built only from information available on or before that date.
Time Series
A time series is a sequence of observations over time for the same location, used for monitoring and change detection.