The WAVeTrends dataset is a 0.05 degree (5.55 km) vegetation change product, spanning the West African Sudano-Sahel region. It provides pixel-wise information on concurrent woody and herbaceous vegetation trends over a 32-year period (1982-2013). Change in woody vegetation was derived using long-term rain use efficiency (RUE) sensitivity, i.e., the per-pixel comparison of the difference of mean RUE between the first and last decades of the 32-year time series. Herbaceous vegetation change was defined by short-term RUE sensitivity, i.e., comparing the slope of the RUE relationship (productivity vs. precipitation) between both decades using per-pixel Analysis of Covariance (ANCOVA). Categorical vegetation change was then determined for each pixel using the direction of the change and a significance level of p<0.05. The use of RUE (the amount of biomass produced per unit of precipitation) for vegetation trend analysis in savanna regions relies on the assumption that rainfall is a significant positive driver of net production in drylands. Testing of this long-term productivity-rainfall relationship revealed that the assumption was not always met, therefore, validity flags are included for each pixel location.