Publications

Paruta, A; Ciraolo, G; Capodici, F; Manfreda, S; Dal Sasso, SF; Zhuang, RD; Romano, N; Nasta, P; Ben-Dor, E; Francos, N; Zeng, YJ; Maltese, A (2021). A Geostatistical Approach to Map Near-Surface Soil Moisture Through Hyperspatial Resolution Thermal Inertia. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 59(6), 5352-5369.

Abstract
Thermal inertia has been applied to map soil water content exploiting remote sensing data in the short and long wave regions of the electromagnetic spectrum. Over the last years, optical and thermal cameras were sufficiently miniaturized to be loaded onboard of unmanned aerial systems (UASs), which provide unprecedented potentials to derive hyperspatial resolution thermal inertia for soil water content mapping. In this study, we apply a simplification of thermal inertia, the apparent thermal inertia (ATI), over pixels where underlying thermal inertia hypotheses are fulfilled (unshaded bare soil). Then, a kriging algorithm is used to spatialize the ATI to get a soil water content map. The proposed method was applied to an experimental area of the Alento River catchment, in southern Italy. Daytime radiometric optical multispectral and day and nighttime radiometric thermal images were acquired via a UAS, while in situ soil water content was measured through the thermo-gravimetric and time domain reflectometry (TDR) methods. The determination coefficient between ATI and soil water content measured over unshaded bare soil was 0.67 for the gravimetric method and 0.73 for the TDR. After interpolation, the correlation slightly decreased due to the introduction of measurements on vegetated or shadowed positions ( r2 = 0.59 for gravimetric method; r2 = 0.65 for TDR). The proposed method shows promising results to map the soil water content even over vegetated or shadowed areas by exploiting hyperspatial resolution data and geostatistical analysis.

DOI:
10.1109/TGRS.2020.3019200

ISSN:
0196-2892