Variable renewable energy sources display different space-time variability driving the availability of energy generated from these sources. Complementarity among variable renewable energies in time and space allows reducing the variability of power supply and helps matching the electricity demand curve. This work investigates the temporal structure of complementarity along an alpine transect in North-East Italy, considering a 100% renewable energy mix scenario composed by photovoltaic and run-of-the-river energy. We analyze the dominant scales of variability of variable renewable energy sources and electricity demand. In addition, we introduce a new metric, the wavelet-based complementarity index, to quantify the potential complementarity between two different energy sources. We show that this index varies at different temporal scales and it helps explaining the discrepancy between demand and supply in the study area. Continuous and discrete wavelet analyses are applied to assess the energy balance variability at multiple temporal scales and to identify the optimal mix of renewable energies, respectively. This work describes therefore an effective approach to investigate the temporal-scale dependency of the variance in the energy balance and can be further extended to different and more complex situations.