Glacier Terminus Estimation from Landsat Image Intensity Proﬁles Joseph Usset 1 , Arnab Maity 2 , Ana-Maria Staicu 3 and Armin Schwartzman 4 Abstract Mountain glacier retreat is an important problem related to temperature increase caused by global climate change. The retreat of mountain glaciers has been studied from the ground, but there exists a need for automated methods to catalog glacial change with a wider scope. A viable approach is to extract intensity proﬁles from Landsat images along the glacial ﬂowline and follow the terminus location over time. We propose a new robust and accurate statistical algorithm to estimate the movement of glacial termini over time from these extracted image intensity proﬁles. The method we propose ﬁrst uses regression splines to smooth the image intensity proﬁles. For each proﬁle the glacial terminus location is assumed to lie near a point of high negative change in the smoothed proﬁles. An approximate path of termini locations over time is obtained by an algorithm that seeks to minimize the cumulative ﬁrst derivative value across the proﬁles. Spline smoothing is applied to this pilot path for estimation of long-term terminus movement. The predictions from the method are evaluated on simu- lated data and compared to available ground measurements for the Nigardsbreen, Gorner, Rhone, and Franz Josef glaciers. Keywords: Change-Point Estimation; Cross-Validation; Non-parametric Regression; Satellite Imagery; Spline smoothing. 1 Department of Biostatistics, University of Kansas, Kansas city, USA (E-mail: jusset@kumc.edu). 2 Department of Statistics, North Carolina State University, Raleigh, USA (E-mail: amaity@ncsu.edu). 3 Department of Statistics, North Carolina State University, Raleigh, USA (E-mail: astaciu@ncsu.edu). 4 Department of Statistics, North Carolina State University, Raleigh, USA (E-mail: aschwar@ncsu.edu). 1