Remotely sensed Land Surface Temperature (LST) represents a valuable source of data for a simple modelling of the dynamic of soil moisture (SM) over large areas. In this paper we evaluated the capability of LST monthly anomalies, derived from the MOD11C3 standard product, to capture the SM dynamic as modelled over Europe by means of an ensemble of three land surface models. The direct use of LST as proxy of SM outperformed other LST-derived quantities, such as surface-to-air temperature gradient and day-night temperature variations, returning significant correlation values over the whole domain. LST performed better over Southern Europe compared to the Northern part of the domain, with the best results over areas characterized by water-limited conditions and moderate stress. Additionally, the analysis of the contingency matrix shows that the LST model is skillful in capturing extreme dry SM events, and it also has a good overall capability to correctly detect the dry events in 66% of the cases, with an average probability of false alarm of about 30%. Overall, the use of LST anomalies seems a promising starting point for a reliable modelling of the SM dynamic with a minimum amount of information. Even if the adopted approach is simple, the results are encouraging for a practical use of LST in an operational drought monitoring system over the study area.
On the role of land surface temperature as proxy of soil moisture status for drought monitoring in Europe
Cammalleri C.;
2015-01-01
Abstract
Remotely sensed Land Surface Temperature (LST) represents a valuable source of data for a simple modelling of the dynamic of soil moisture (SM) over large areas. In this paper we evaluated the capability of LST monthly anomalies, derived from the MOD11C3 standard product, to capture the SM dynamic as modelled over Europe by means of an ensemble of three land surface models. The direct use of LST as proxy of SM outperformed other LST-derived quantities, such as surface-to-air temperature gradient and day-night temperature variations, returning significant correlation values over the whole domain. LST performed better over Southern Europe compared to the Northern part of the domain, with the best results over areas characterized by water-limited conditions and moderate stress. Additionally, the analysis of the contingency matrix shows that the LST model is skillful in capturing extreme dry SM events, and it also has a good overall capability to correctly detect the dry events in 66% of the cases, with an average probability of false alarm of about 30%. Overall, the use of LST anomalies seems a promising starting point for a reliable modelling of the SM dynamic with a minimum amount of information. Even if the adopted approach is simple, the results are encouraging for a practical use of LST in an operational drought monitoring system over the study area.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
