We present a fast and statistically principled approach for land cover change detection. The approach is illustrated with a geographic application that involves analyzing remotely sensed data to detect changes in the normalized difference vegetation index (NDVI) in near real time. We use the Wal-Mart land cover change data set as a nontraditional way to monitor and validate known cases of NDVI change. A reference distribution has been justified to fit the available data. An adaptive metric based on the exponentially weighted moving average (EWMA) of normal scores derived from p-values is tracked for new or streaming data, leading to alarms for large or sustained changes. A heuristic algorithm based on the property of the metric is proposed for change point detection. The proposed framework performed well on the validation dataset.