Browsing by Author "Mäkinen, Tero"

Periglacial zone can be defined as cold but non-glacial environment, where periglacial processes are active. Formation of permafrost and seasonal frost and the thawing of such phenomena are the essential elements of the development of the periglacial landscape. Conception is that periglacial geomorphology also consists of seasonal snow cover, fluvial processes and aeolian activity, which seem to have obvious impact to the landscape of cold non-glacial regions. According to observations and global climate models, climate is warming up and the heating is expected to be the most strongest at the high latitudes. Modern research methods and whole new approaches to understand the spatial phenomena and geomorphological analysis is an essential part of the global warming research. Spatial modelling is one of these modern research methods. The objective of this study is to collect the essential research concerning spatial modelling of the periglacial processes, and investigate through the already written literature, and to point out the strengths and weaknesses of the spatial modelling in the context of global warming. According to models, already a minor warming of the atmosphere rapidly declines the periglacial processes and leads to the degradation of permafrost. If te changes stay at the same level, this would lead to the almost total loss of these processes in certain periglacial domains by the end of this century. These processes could only stay active at locally restricted areas at favourable conditions. The better availability of the remote sensing and geographical information systems and also the recent development of quantitative modelling methods are making it possible to cost-effectively research the spatial occurrence of the periglacial processes and the factors relating to it. Spatial modelling could also make it possible to explore new and remote regions without expensive field surveys. Insufficient data could cause difficulties in these territories. Models are always imperfect acts of reality and some phenomena are hard to be simulated. Spatial geomorphological data often consists of complex process-environment relationships, non-linear responses and spatial autocorrelation. Many of these difficulties can be controlled by developed modelling methods, but the generalization and transferability (extrapolation) of the models still cause difficulties.