Spatiotemporal dynamics of plant assemblages under changing climate and land-use regimes in central Nepal Himalaya

Abstract

The trajectories of vegetation and landscape dynamics have been redirected globally by climate change and land-use change. The drivers and mechanisms of the spatiotemporal changes are likely scale dependent and are most probably confounded. Himalayan landscapes are under-explored and it is particularly crucial to understand the mechanisms and trajectories of the changes they are experiencing and their scale relationship for an effective management of these landscapes. This thesis synthesises four case studies that have documented spatiotemporal changes in plant assemblages driven by climate and land-use change over the last two decades, assessed the relationship between alpha diversity (plot-based species richness) and gamma diversity (regional species richness) with a focus on scale, and tested the performance of weighted averaging (WA) regression and calibration for quantifying the elevational changes of species assemblages. Studies were conducted in temperate, subalpine, and alpine vegetation at two locations in central Nepal. Scale sensitivity of the elevational species richness was assessed by treating the alpha diversity of different plant life-forms at different grain sizes and areas used for gamma estimation as response variables in a generalised linear model [Paper I]. Systematic changes in temporally resurveyed assemblages were analysed by ordinations and attributed to climate and land-use change using regression analyses [Papers II, III]. WA regression and calibration technique was evaluated by comparing the models for different types of temporal datasets [Paper IV]. Elevational gamma diversity can significantly predict alpha diversity and the relationship is largely scale invariant, although it is slightly less so for woody species. At a regional level, climate warming is a major driver of the demonstrated spatiotemporal changes, i.e. thermophilisation of plant assemblages. However, landuse change may confound or counteract the climatic effects at a local or landscape level. WA regression and calibration predicts fairly accurately the elevation of the plot-based vegetation assemblages. Use of species incidence data may improve the accuracy, but species data processing cannot guarantee more accurate calibrations. To conclude, the systematic spatiotemporal changes in plant assemblages over the last two decades in central Nepal are significant, are largely irrespective of spatial scale, and are most likely related to interactions and feedback mechanisms between climate change and land-use change at different spatiotemporal scales. Taxonomic, census, and sampling accuracy are crucial in the analyses of temporal changes, especially by environmental reconstructions.