Detection of potential arable land with remote sensing and GIS. A Case Study for Kjósarhreppur

• Main Author: Brynja Guðmundsdóttir 1955-
• Other Authors: Landbúnaðarháskóli Íslands
• Format: Master Thesis
• Language: English
• Published: 2014
• Subjects: Landupplýsingakerfi; Jarðrækt; GIS (geographic information systems); Iceland
• Online Access: http://hdl.handle.net/1946/19894

LUMA-GIS Thesis nr 29 Ritgerð frá Department of Physical Geography and Ecosystem Analysis Centre for Geographical Information Systems, Lund University, Sweden Arable land in Iceland is a valuable natural resource that should be preserved. Arable land is not an unlimited resource. According to the new Planning Act (No 123/2010) municipalities have to define arable and potential arable land, classify agricultural land with respect to the type of farming and cultivated and potential cultivated land for future use. The aim of the current study was to develop (digital) methods to define and locate potential arable land and make a feature set which is possible to use in strategy planning and planning work for land use. Different data sources were used for the analysis:, the Icelandic Farmland Database (Nytjaland), Icelandic Geographical Land Use Database, digital network of drainage ditches and cropland obtained from the Agricultural University of Iceland, aerial photographs, contour lines, lakes and rivers, roads from the municipality Kjósarhreppur and finally aerial photos, contour lines and elevation points from Samsýn (IT company, specialized in GIS). The project was divided into two parts. Firstly, an elevation model was constructed in order to delimit land below 200 m a.s.l. followed by an evaluation of how the land area changes with slope from 6° to 10°. For further analysis slope value of 10° was used. Secondly, an image analysis was carried out using SPOT-5 and Quickbird images to classify land into arable and potential arable land using both supervised and unsupervised classification. Subsequently it was examined whether it would be possible to use vegetation indices for this analysis. The resulting classification was verified by on-site analysis as well as the depth and stoniness of the potential arable land. The analysis shows that it is possible to identify arable and potential arable land from satellite data, with the aid from other data, especially aerial photographs for texture and forms and ...