Skip to main content
SLU publication database (SLUpub)

Conference abstract2002

Using remote sensing as a tool within the reindeer industry: Scientific aspects of Renbruksplan

Leif, Hemberg; Lars, Göran Brandt; Karin, Baer; Thomas, Stenlund; Hans, Tömmervik; Sandström, Per; Granqvist, Pahlen Tina; Hagner, Olle; Olsson, Håkan; Edenius, Lars

Abstract

Because of the large tracts of lands used for reindeer grazing, the needs of the reindeer industry at times conflict with the needs of other land users. Remote sensing and GIS become valuable tools in attempting to resolve the complex issues of the different seasonal and temporal needs of the reindeer in relation to the needs and activities of other land users. Until recently, lack of overview and understanding - both by reindeer herders and other land users - of the land-use pattern of reindeer herding has lead to unnecessary conflict between different land users. In an attempt to resolve this complex issue a process has been developed to produce “land-use plans for reindeer husbandry” (in Swedish Renbruksplan (RBP)). The process included a number different and sometimes “unique” methods used in the digital capture of traditional ecological knowledge both through satellite image interpretation and field work, as well as different methods of digital image classification. The satellite image interpretation, on-screen digitizing, and field work leading to the identification of important grazing lands was carried out by the Sami village members with support from researchers. The combination of great knowledge about the local area and the familiarity interpreting satellite images was necessary for successful delineation of important grazing lands, a process which was followed by field visits to all identified important grazing lands. The process of developing RBP further included different methods of digital image classification. In the forested portion of the Sami villages we used the k-Nearest Neighbor method to produce estimates of occurrence for each reindeer grazing type. We then combined these estimates into one reindeer grazing type map with discrete classes. For the mountainous portion we used an unsupervised “hyperclustering” classification procedure in order to produce a vegetation maps for the area. Information from several parallel ongoing processes was combined with compiled information about activities of other land-users in the final RBP. The ultimate goal is that RBP will contribute to increased knowledge and understanding between different resource users and thereby reduce conflict. During 2000-2002, the project has been ongoing in Vilhelmina Norra and Malå Sami villages in the county of Västerbotten in northern Sweden. In addition to participants from the two Sami villages and SLU, cooperators include the Regional Forestry Board, the County Administrative Board of Västerbotten and Norwegian Institute if Nature Research (NINA). In the coming years we will use and refine the methods and produce RBP in additionally four Sami villages

Published in

Conference

RESE Scientific Conference