资源环境科技发展态势分析平台

Due to their expected higher resilience following disturbances and adaptive potential to new climatic conditions, interest in uneven-aged mixed forests has increased in recent years. It is, however, unclear how to best quantify their site-specific growth potential, particularly at a time when there is a pressing need to consider the effects of a changing climate on tree and forest growth. Here, we address these topics using growth models for Norway spruce (Picea abies (L) Karst.) and Silver fir (Abies alba Mill.), based on long-term observations from uneven-aged mixed forests in southwestern Germany.

We used a linear mixed-effects framework for modeling basal area increment of individual trees. We accounted for site quality using a phytocentric index based on the past growth of dominant trees (growth index) and three classes of geocentric environmental descriptors: physiographic, edaphic and climatic (temperature means and precipitation sums aggregated over 5, 15 or 30 years). For a subset of the data where it was possible to determine site index, growth index proved to be better predictor of tree increment than site index. When considering the entire dataset, climate variables had the single largest positive impact on model fit, yet cross-validation results suggested that no improvement in predictive ability occurred unless physiographic variables were also added. Higher levels of spring and growing season precipitation stimulated growth for Norway spruce and Silver fir, respectively. Temperature-growth relationships were predominantly positive for Silver fir and negative for Norway spruce. Aggregating climate variables over progressively longer time spans clearly reduced model fit for Norway spruce, yet a similar pattern was not apparent for Silver fir.

Our results indicate that without rigorous testing, tried-and-trusted decision tools developed for even aged, single-species stands cannot be transferred to uneven-aged mixed forests. Precipitation- and temperature-based variables provide dynamic proxies which may allow us to better grasp the complexity of climate-growth relationships. This understanding is essential for reducing the uncertainty around predictions of climate change impacts on forest ecosystems. (C) 2017 Elsevier B.V. All rights reserved.