Abstract. Different wavelength light sources are used in photodynamic therapy (PDT) of the skin to treat different conditions. Clinical studies show inconsistent results for the effectiveness of aminolevulinic acid (ALA)-PDT performed at different wavelengths. In order to understand the effect of treatment wavelength, a theoretical study was performed to calculate time-resolved depth-dependent distributions of PDT components including ground-state oxygen, sensitizer, and reacted singlet oxygen for different treatment wavelengths (405, 523, and 633 nm) using a numerical model of ALA-PDT of human skin. This model incorporates clinically relevant features of the PDT process including light attenuation, photobleaching, oxygen consumption, and diffusion, as well as tissue perfusion. The calculations show that the distributions of these quantities are almost independent of the treatment wavelength to a depth of about 1 mm. In this surface region, PDT-induced hypoxia is the dominant process. At greater depths, the production of singlet -oxygen is governed by the penetration of the treatment light. Two noninvasive PDT dosimetry approaches: the cumulative singlet oxygen luminescence (CSOL) and the fractional fluorescence bleaching metric, were investigated and compared for all three wavelengths. Although CSOL was more robust, both metrics provided correlations with the singlet oxygen dose in the upper dermis that were almost independent of treatment wavelength. This relationship breaks down at greater depths because light penetration depends on wavelength.