Despite intensive research into the coupling between tectonics and surface processes, our ability to obtain quantitative information on the rates of tectonic processes from topography remains limited due primarily to a dearth of data with which to test and calibrate process rate laws. Here we develop a simple theory for the impact of spatially variable rock-uplift rate on the concavity of bedrock river profiles. Application of the analysis to the Siwalik Hills of central Nepal demonstrates that systematic differences in the concavity of channels in this region match the predictions of a stream power incision model and depend on the position and direction of the channel relative to gradients in the vertical component of deformation rate across an active fault-bend fold. Furthermore, calibration of model parameters from channel profiles argued to be in steady state with the current climatic and tectonic regime indicates that (1) the ratio of exponents on channel drainage area and slope (m/n) is ∼0.46, consistent with theoretical predictions; (2) the slope exponent is consistent with incision either linearly proportional to shear stress or unit stream power (n = 0.66 or n = 1, respectively); and (3) the coefficient of erosion is within the range of previously published estimates (mean K = 4.3 × 10−4 m0.2/yr). Application of these model parameters to other channels in the Siwalik Hills yields estimates of spatially variable erosion rates that mimic expected variations in rock-uplift rate across a fault-bend fold. Thus, the sensitivity of channel gradient to rock- uplift rate in this landscape allows us to derive quantitative estimates of spatial variations in erosion rate directly from topographic data.