Researchers and development practitioners in remote mountain areas rely on elevation data to study vegetation dynamics, natural hazards, land use, and other patterns. However, despite advances in technology, accurate digital elevation models (DEMs) with spatial resolution <30 m do not exist for most of the world's montane regions. We used a low-cost GPS-based protocol to construct a high-resolution (10 m) DEM for a rugged, remote mountain site in the northern Peruvian Andes. Elevation data were collected with handheld GPS units and combined with digitized and interpolated points within a Geographic Information System to generate a 10 m DEM. Additional DEMs were generated using 50%, 20%, and 15% of the surface points collected and from a 1∶100,000 topographic map and ASTER GDEMv2 data. Estimated absolute vertical accuracy of the GPS surface-point DEMs was significantly lower than that of the ASTER GDEMv2 and topographic map DEMs. Relative vertical accuracy, a better measure of DEM quality, was considerably lower for all 6 DEMs than absolute vertical accuracy. Depending on project budget, time, and labor availability, this method can be used to produce DEMs with high spatial resolution and substantially improved relief maps for research, visualization, and communication purposes. Implementation of this method is practical in locations without access to electricity or post-processing correction facilities, open-canopy land covers, and projects with small budgets that involve local participants.