Soil pH limits of forest vascular plants determine range size and threat level

• Published in: Journal of vegetation science 2016-05, Vol.27 (3), p.535-544
• Main Authors: Michaelis, Jana, Pannek, Angela, Diekmann, Martin
• Format: Article
• Language: English
• Subjects: Edaphic niche; Forest specialist; Germany; Huisman-Olff-Fresco (HOF) models; pH limit; Physiological threshold; Rare plant species; Soil pH; Species response curve
• ISSN: 1100-9233; 1654-1103
• DOI: 10.1111/jvs.12380

QUESTIONS: Knowledge about the response curves of species along soil variable gradients is scarce. Because environmental change may force species to live closer to their ecological limits than optima, we examined and compared the importance of response limits, response optima and Ellenberg indicator values for soil pH to determine the regional range sizes and threat levels of species. LOCATION: Deciduous forests in the lowlands and uplands of northern Germany. METHODS: We used Huisman–Olff–Fresco models to examine the species responses of forest vascular plants along a soil pH gradient with a particular focus on rare species. Separately for the two regions, the optima and limits across all species were related to range size, change in regional range size over the past decades and threat level. RESULTS: Lower pH limits showed an aggregation around pH 4, and were consistent across regions, whereas upper pH limits were not clumped and not significantly correlated between regions. In the lowlands, species with relatively large lower pH limits were less widespread, had decreased more over time and were more threatened than species able to grow on very acid soils. Lower limits were more closely related to regional range size and rarity than optima and Ellenberg R values. These patterns were not found for upland species, probably due to a more complex interaction between pH and other soil factors. CONCLUSION: The results reinforce the importance of soil variables and their interactions for the occurrence of plant species, and suggest that limits may be better predictors of current and future species distributions than optima and Ellenberg indicator values. Still, crucial information is missing even for common species. Therefore, there is an urgent need for more and better species occurrence data relative to edaphic factors in order to identify the species' needs and their potential responses to environmental change.