Extraction of Velocity Time Series With an Optimal Temporal Sampling From Displacement Observation Networks

Today, more and velocity observations are available online or on-demand. However, this amount of data is complex to analyze since span different temporal baselines. Velocities obtained from a small baseline close the derivative displacement but likely be contaminated by noise. long approximate mean between two dates can affected decorrelation. Having short baselines provides redundancy that needs properly considered. In article, we propose method aims extract short-term time series with regular sampling all observations. The proposed relies on inversion based an improved closure observation network. Two criteria determine optimal study variations. To take unequal uncertainty into account, done iterative reweighted least square using well-established weighting function, without preprocessing. results in sampling, coverage, reduced uncertainty, no redundancy. studied area Kyagar glacier, North Karakoram range characterized strong variations originated glacier surge additional seasonal variability.