This work investigates the channel changes of Alpine rivers from the end of the Little Ice Age (1850s) to the 1950s, with the aim to determine the possible role of climatic variations occurred in this period before the onset of anthropic pressures (i.e., dams, check-dams, bank protections, and gravel mining). The research was conducted on 17 river catchments of South Tyrol (northern Italy), glaciated and unglaciated. A multitemporal GIS analysis approach was adopted to assess the morphological changes (in terms of channel width and pattern) from three different sources: (i) Austrian cadastral map (1858), (ii) maps from the Italian Institute of Military Geography (1917–1925), and (iii) two aerial photo sets taken in 1945 and 1954. The analysed river network (a total of 480 km) was subdivided into 162 morphologically homogeneous reaches (76 confined, 81 partly confined, and 5 unconfined), with lengths ranging from 630 to 5500 m, slope from 0.3 to 24%, and drained area from 20 to ~ 4000 km2. The statistical relationships among morphological changes and reach- and basin-scale factors were analysed by univariate and multivariate methods, and the relationships between width changes and 36 controlling factors were explored using Principal Component Analysis. The variability in width and morphological pattern changes were very pronounced between and within single rivers, highlighting the value of such a large data set. Overall, the analysed rivers varied their morphological pattern, mostly exhibiting a shift from multithread/transitional to single-thread patterns, but unchanged planform types were also common. Variations in channel width varied substantially among the analysed rivers, which featured narrowing (slightly prevailing) and widening (the least common) as well as many cases of very limited changes. Channel width variations appear statistically, although weakly, related to some morphometric variables; and significant differences emerge comparing glaciated vs. unglaciated basins. Climate-related variations (glacier dynamics and channel disturbance frequency) are argued to be the dominant factors that affected channel variations.