Salta al contenuto principale
Passa alla visualizzazione normale.

GIUSEPPE CIRAOLO

Thermal Inertia Modeling for Soil Surface Water Content Estimation: A Laboratory Experiment

  • Autori: Minacapilli, M; Cammalleri, C; Ciraolo, G; D'Asaro, F; Iovino, M; Maltese, A
  • Anno di pubblicazione: 2012
  • Tipologia: Articolo in rivista (Articolo in rivista)
  • OA Link: http://hdl.handle.net/10447/62037

Abstract

We are proposing a new method for estimating soil surface water content from thermal inertia distributions retrieved from visible–near infrared (VISNIR)and thermal infrared (TIR) images. A drying experiment was conducted on three fi ne-textured soils while acquiring multispectral VIS-NIR and TIR images. Simultaneous measurements of soil water content and thermal inertia were conducted by the thermogravimetric method and the heat pulse technique, respectively. Direct measurements were used to test the thermal inertia approach proposed by Murray and Verhoef that requires only knowledge of soil porosity and can be easily inverted to derive soil water content from thermal inertia. For the three considered soils, the performance of the Murray and Verhoef model was practically equal to that of the traditional approach based on the direct estimation of thermal conductivity and heat capacity, which requires more detailed information about soil properties. With the aim of simplifying the estimation of thermal inertia from remotely sensed images,a modified Kersten function was proposed in which the normalized thermal inertia is substituted by the normalized apparent thermal inertia. Comparison between the two modifi ed Kersten functions was satisfactory. The proposed approach allowed predictions of the surface soil moisture from apparent thermal inertia distributions with an acceptable level of accuracy for practical purposes (0.028 ≤ RMSE ≤ 0.043 m3 m−3) and therefore it can be considered a simple and effective tool for estimating the spatial and temporal distribution of surface soil moisture from VIS-NIR and TIR remotely sensed data.