Across scale temperature modelling using a simple approach for the characterization of high mountain ecosystem complexity

Abstract

A modelling approach was practised to characterize the complexity of high mountain ecosystems using a new simple model to simulate near-surface temperature variations. The heterogeneity of the investigated landscape led to an across scale procedure that combined vertical interactions at single locations (nano-scale), micro-spatial differentiations within small catchments (micro-scale), and altitudinal changes of an entire mountain system (meso-scale). The model was calibrated on detailed measurements for different sites. Simulation runs were satisfactory according to model validation based on data with high temporal resolution. In all cases, more than 80% of the variance of observed temperatures was explained by the model. We analysed correlations between general meteorological trends and local climatic differentiations. The hypothesis was that different overlying meteorological phenomena (i.e. different weather situations) would find their expression in micro-climatic conditions (especially in the variation of near-surface temperatures) superiorly determining ecosystem functioning. It turned out that the micro-spatial conditions resulted in complex principles of thermal changes along altitudinal broad-scale gradients. So, the adiabatic lapse rate, commonly used to describe the altitudinal zonation, did not explain the different mosaics of ecosystems. We illustrated the biases between measurements and common assumptions by means of correlation matrices. To conclude, the complexity of the ecosystem determining energy balance was described by complex differences of temperature gradients that were achieved from across scale multi-regression analysis. It was shown to which degree local site conditions corresponded with meteorological assumptions under different spatio-temporal conditions.