Next Article in Journal
Assessing the Suitability of Future Multi- and Hyperspectral Satellite Systems for Mapping the Spatial Distribution of Norway Spruce Timber Volume
Previous Article in Journal
The Use of Multi-Temporal Landsat Imageries in Detecting Seasonal Crop Abandonment
Open AccessArticle

Remote Sensing Based Two-Stage Sampling for Accuracy Assessment and Area Estimation of Land Cover Changes

Remote Sensing and Geoinformation, Institute for Information and Communication Technologies, Joanneum Research, Steyrergasse 17, Graz A-8010, Austria
*
Author to whom correspondence should be addressed.
Academic Editors: Parth Sarathi Roy and Prasad S. Thenkabail
Remote Sens. 2015, 7(9), 11992-12008; https://doi.org/10.3390/rs70911992
Received: 20 May 2015 / Revised: 10 September 2015 / Accepted: 10 September 2015 / Published: 18 September 2015
Land cover change processes are accelerating at the regional to global level. The remote sensing community has developed reliable and robust methods for wall-to-wall mapping of land cover changes; however, land cover changes often occur at rates below the mapping errors. In the current publication, we propose a cost-effective approach to complement wall-to-wall land cover change maps with a sampling approach, which is used for accuracy assessment and accurate estimation of areas undergoing land cover changes, including provision of confidence intervals. We propose a two-stage sampling approach in order to keep accuracy, efficiency, and effort of the estimations in balance. Stratification is applied in both stages in order to gain control over the sample size allocated to rare land cover change classes on the one hand and the cost constraints for very high resolution reference imagery on the other. Bootstrapping is used to complement the accuracy measures and the area estimates with confidence intervals. The area estimates and verification estimations rely on a high quality visual interpretation of the sampling units based on time series of satellite imagery. To demonstrate the cost-effective operational applicability of the approach we applied it for assessment of deforestation in an area characterized by frequent cloud cover and very low change rate in the Republic of Congo, which makes accurate deforestation monitoring particularly challenging. View Full-Text
Keywords: land cover change; deforestation; REDD monitoring; accuracy assessment; area estimation; sampling; bootstrapping; confidence interval land cover change; deforestation; REDD monitoring; accuracy assessment; area estimation; sampling; bootstrapping; confidence interval
Show Figures

Figure 1

MDPI and ACS Style

Gallaun, H.; Steinegger, M.; Wack, R.; Schardt, M.; Kornberger, B.; Schmitt, U. Remote Sensing Based Two-Stage Sampling for Accuracy Assessment and Area Estimation of Land Cover Changes. Remote Sens. 2015, 7, 11992-12008.

Show more citation formats Show less citations formats

Article Access Map

1
Only visits after 24 November 2015 are recorded.
Back to TopTop