Landscape-scale factors driving plant species composition in mountain streamside and spring riparian forests

Abstract

Effective sustainable management of riparian forests requires an understanding of both the local physical setting and larger system at the landscape scale. Although site-specific influences on riparian forests are well studied, landscape-scale factors driving species composition are insufficiently recognized, especially regarding forests in spring-fed areas. In this study we ask three questions: (1) What are the main topographic factors driving the species composition in mountain riparian forests at the landscape scale? (2) What are the differences in environmental controls between riparian forests growing along river banks and in spring-fed areas? (3) Are species responses to environmental gradients in streamside and spring-fed environments consistent with plant traits or life history strategies? To address these questions, we used a dataset of vegetation plots sampled along 100 rivers in the Sudetes (southwestern Poland, Central Europe) and GIS-derived topographic variables, including the elevation, slope, topographic wetness index (TWI), stream power index (SPI), potential solar radiation and distance from river source. The significance of the variables was tested by the ordination technique (CCA). In addition, species responses (Huisman–Olff–Fresco models) were analysed for the species that fitted the ordination models best. The results showed that floristic variation in streamside forests is related to distance from the river source, altitude and SPI; while, floristic variation in spring forests is linked to altitude and solar radiation. Moreover, the frequency of particular response curve types varies between streamside and spring forests: the common presence of unimodal responses indicates that streamside forests are much more diverse in terms of community turnover than spring forests. The study highlights different landscape-scale environmental parameters that control the variation in species composition in streamside and spring forests. Streamside forests are the products of their catchments and considerably depend on the conditions that exist within the contributing area, while forests growing in spring-fed areas are mainly controlled by local conditions. We conclude, that effective management and conservation, especially approaches based on emulation of natural disturbances, should take these differences into consideration by developing site-specific management guidelines for both types of riparian forests.