No abstract

No abstract

No abstract

This study explored the capability of Support Vector Machines (SVMs) and regularised kernel Fisher's discriminant analysis (rkFDA) machine learning supervised classifiers in extracting flooded area from optical Landsat TM imagery. The ability of both techniques was evaluated using a case study of a riverine flood event in 2010 in a heterogeneous Mediterranean region, for which TM imagery acquired shortly after the flood event was available. For the two classifiers, both linear and non-linear (kernel) versions were utilised in their implementation. The ability of the different classifiers to map the flooded area extent was assessed on the basis of classification accuracy assessment metrics. Results showed that rkFDA outperformed SVMs in terms of accurate flooded pixels detection, also producing fewer missed detections of the flooded area. Yet, SVMs showed less false flooded area detections. Overall, the non-linear rkFDA classification method was the more accurate of the two techniques (OA = 96.23%, K = 0.877). Both methods outperformed the standard Normalized Difference Water Index (NDWI) thresholding (OA = 94.63, K = 0.818) by roughly 0.06 K points. Although overall accuracy results for the rkFDA and SVMs classifications only showed a somewhat minor improvement on the overall accuracy exhibited by the NDWI thresholding, notably both classifiers considerably outperformed the thresholding algorithm in other specific accuracy measures (e.g. producer accuracy for the "not flooded" class was ~10.5% less accurate for OPEN ACCESSRemote Sens. 2015, 7 3373 the NDWI thresholding algorithm in comparison to the classifiers, and average per-class accuracy was ~5% less accurate than the machine learning models). This study provides evidence of the successful application of supervised machine learning for classifying flooded areas in Landsat imagery, where few studies so far exist in this direction. Considering that Landsat data is open access and has global coverage, the results of this study offers important information towards exploring the possibilities of the use of such data to map other significant flood events from space in an economically viable way.

Acquiring information on the spatio-temporal variability of soil moisture is of key importance in extending our capability to understand the Earth system's physical processes, and is also required in many practical applications. Earth observation (EO) provides a promising avenue to observe the distribution of soil moisture at different observational scales, with a number of products distributed at present operationally. Validation of such products at a range of climate and environmental conditions across continents is a fundamental step related to their practical use. Various in situ soil moisture ground observational networks have been established globally providing suitable data for evaluating the accuracy of EO-based soil moisture products. This study aimed at evaluating the accuracy of soil moisture estimates provided from the Soil Moisture and Ocean Salinity Mission (SMOS) global operational product at test sites from the REMEDHUS International Soil Moisture Network (ISMN) in Spain. For this purpose, validated observations from in situ ground observations acquired nearly concurrent to SMOS overpass were utilized. Overall, results showed a generally reasonable agreement between the SMOS product and the in situ soil moisture measurements in the 0-5 cm soil moisture layer (root mean square error (RMSE) = 0.116 m 3 m −3 ). An improvement in product accuracy for the overall comparison was shown when days of high radio frequency interference were filtered out (RMSE = 0.110 m 3 m −3 ). Seasonal analysis showed highest agreement during autumn, followed by summer, winter, and spring seasons. A systematic soil moisture underestimation was also found for the overall comparison and during the four seasons. Overall, the result provides supportive evidence of the potential value of this operational product for meso-scale studies and practical applications.