Data file: confusion matrices from pilot study of methodology for the development of farmland habitat reports for sustainability assessments
dc.contributor.author | Finn, John | |
dc.contributor.author | Moran, Patrick | |
dc.date.accessioned | 2020-03-09T14:36:41Z | |
dc.date.available | 2020-03-09T14:36:41Z | |
dc.date.issued | 2020-03-09 | |
dc.identifier.uri | http://hdl.handle.net/11019/1877 | |
dc.description | In each worksheet, data for the different habitat classes (coded as in Fossitt 2000), are provided as assessed by an ecologist reviewing satellite imagery, and as assessed by a different ecologist in a field survey. The row data indicate the habitat areas collected from field survey (ground truthed habitats confirmed to be present). The column data indicate the habitat areas from satellite imagery (habitat identiification based on interpretation of imagery). These data are from a pilot project with 187 farms, in which we developed and implemented a process for creating farm habitat reports and investigate the accuracy of visual interpretation of satellite imagery by an ecologist aiming to identify habitat types. Visual assessment of satellite imagery achieved an overall accuracy of 96% in its ability to discriminate between land parcels with habitats categorised by this study as being of either high or low nature conservation value. Assessment of satellite imagery achieved an overall accuracy of 90% in its ability to discriminate among Fossitt level II habitat classes, and an overall accuracy of 81% when using individual habitat classes (Fossitt level III). There was, however, considerable variation in the accuracy associated with individual habitat classes. | en_US |
dc.description.abstract | This Excel data file provides the confusion matrices associated with a publication in the Irish Journal of Agricultural and Food Research. This file contains four worksheets: 'High vs Low matrix', 'Level II matrix', 'Level III matrix' and 'Level III matrix (HH)'. Each worksheet presents the area of different habitat classes (as in Fossitt 2000) as determined by a desk-based study of remote sensing imagery, and compared with area of habitat classes as determined by a field-based survey (ground-truthing). The publication by John A. Finn and Patrick Moran is titled 'A pilot study of methodology for the development of farmland habitat reports for sustainability assessments'. | en_US |
dc.language.iso | en | en_US |
dc.rights | CC0 1.0 Universal | * |
dc.rights.uri | http://creativecommons.org/publicdomain/zero/1.0/ | * |
dc.subject | farmland wildlife | |
dc.subject | sustainability | |
dc.subject | habitat | |
dc.subject | mapping | |
dc.subject | biodiversity | |
dc.subject | GIS | |
dc.title | Data file: confusion matrices from pilot study of methodology for the development of farmland habitat reports for sustainability assessments | en_US |
dc.type | Dataset | en_US |
dc.identifier.rmis | 6793 | |
dc.contributor.sponsor | Teagasc | en_US |
dc.contributor.sponsorGrantNumber | 6793 | en_US |
refterms.dateFOA | 2020-03-09T14:36:42Z |